Nutrition in

Gastrointestinal and
Liver Diseases

Editors
Christos Dervenis, Athens
Herbert Lochs, Berlin

12 figures, 18 tables, 2003

Basel 폷 Freiburg 폷 Paris 폷 London 폷 New York 폷

Bangalore 폷 Bangkok 폷 Singapore 폷 Tokyo 폷 Sydney
S. Karger
Medical and Scientific Publishers
Basel 폷 Freiburg 폷 Paris 폷 London
New York 폷 Bangalore 폷 Bangkok
Singapore 폷 Tokyo 폷 Sydney
Fax + 41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Drug Dosage
The authors and the publisher have exerted every effort to en-
sure that drug selection and dosage set forth in this text are in
accord with current recommendations and practice at the time
of publication. However, in view of ongoing research, changes
in government regulations, and the constant flow of informa-
tion relating to drug therapy and drug reactions, the reader is
urged to check the package insert for each drug for any change
in indications and dosage and for added warnings and precau-
tions. This is particularly important when the recommended
agent is a new and/or infrequently employed drug.
All rights reserved.
No part of this publication may be translated into other
languages, reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying, recording,
microcopying, or by any information storage and retrieval
system, without permission in writing from the publisher or, in
the case of photocopying, direct payment of a specified fee to
the Copyright Clearance Center (see ‘General Information’).
© Copyright 2003 by S. Karger AG,
P.O. Box, CH–4009 Basel (Switzerland)
Printed in Switzerland on acid-free paper by
Reinhardt Druck, Basel
ISBN 3–8055–7657–9
 Vol. 21, No. 3, 2003
Contents
196 Editorial
Lochs, H. (Berlin); Dervenis, C. (Athens)
Review Articles
198 What We Have Learned about Cachexia in
Gastrointestinal Cancer
Palesty, J.A.; Dudrick, S.J. (Waterbury, Conn.)
214 Nutritional Support in Acute Pancreatitis
Avgerinos, C.; Delis, S.; Rizos, S.; Dervenis, C. (Athens)
220 Nutrition and Inflammatory Bowel Disease: Its
Relation to Pathophysiology, Outcome and Therapy
Gassull, M.A. (Badalona)
228 Factors Enhancing Intestinal Adaptation after
Bowel Compensation
Botsios, D.S.; Vasiliadis, K.D. (Thessaloniki)
237 Helicobacter pylori Infection, Vitamin B12 and
Homocysteine. A Review
Dierkes, J.; Ebert, M.; Malfertheiner, P.; Luley, C. (Magdeburg)
Original Papers
245 Prevalence of Malnutrition in Hospitalized Medical
Patients: Impact of Underlying Disease
Pirlich, M.; Schütz, T.; Kemps, M.; Luhman, N.;
Burmester, G.-R.; Baumann, G. (Berlin); Plauth, M. (Dessau);
Lübke, H.J.; Lochs, H. (Berlin)
© 2003 S. Karger AG, Basel
Fax + 41 61 306 12 34 Access to full text and tables of contents,
E-Mail karger@karger.ch including tentative ones for forthcoming issues:
www.karger.com www.karger.com/ddi_issues
252 Sugar Intake, Taste Changes and Dental Health
in Crohn’s Disease
Schütz, T.; Drude, C.; Paulisch, E.; Lange, K.-P.; Lochs, H.
(Berlin)
258 Serum Mineral Levels in Children with Intestinal
Parasitic Infection
Olivares, J.L.; Fernández, R.; Fleta, J.; Rodríguez, G.; Clavel, A.
(Zaragoza)
262 Impact of Body Mass Index on Fasting Blood Glucose
Concentration among Helicobacter pylori Carriers
Kyriazanos, I.D.; Sfiniadakis, I.; Dimakos, P.; Gizaris,V.;
Datsakis, K.; Dafnopoulou, A. (Athens)
266 Selenium Is Depleted in Crohn’s Disease on Enteral
Nutrition
Kuroki, F.; Matsumoto, T.; Iida, M. (Fukuoka City)
271 L-Carnitine in the Treatment of Mild or Moderate
Hepatic Encephalopathy
Malaguarnera, M.; Pistone, G.; Astuto, M.; Dell’Arte, S.;
Finocchiaro, G.; Lo Giudice, E.; Pennisi, G. (Catania)
276 Fructose Breath Hydrogen Test – Is It Really
a Harmless Diagnostic Procedure?
Müller, P. (Leisnig/Leipzig); Meier, C.; Böhme, H.J.; Richter, T.
(Leipzig)
279 Screening for Iron Overload in the Turkish Population
Barut, G.; Balci, H. (Istanbul); Bozdayi, M. (Ankara); Hatemi, I.;
Ozcelik, D.; Senturk, H. (Istanbul)
286 Author Index and Subject Index
 Editorial
Dig Dis 2003;21:196–197
DOI: 10.1159/000074105
Malnutrition – The Ignored Risk Factor
Herbert Lochs Christos Dervenis
Nutritional therapy, once a cornerstone of medical treatment, has
lost its attractivity in favor of drug treatment, molecular genetic
interventions or other high-tech therapies. As a consequence, data
about nutritional status or dietary habits are missing in many patient
charts. This development reflects the opinion of many doctors that
nutrition does not greatly affect the course of diseases, and that nutri-
tional intervention is laborious but much less effective than other
forms of therapy. In view of this trend, it seems interesting to analyze
the importance of nutrition in different dieseases.
Of course, it is well accepted that nutrition is an important factor
for the therapy and prognosis in several diseases, such as diabetes
mellitus, hyperlipidemia, obesity or hemochromatosis. However, the
most basic nutritional disturbance – malnutrition – is frequently
ignored since it is considered as a complication of the disease process,
with little bearing on the prognosis and little possibility for therapeut-
ic intervention. However, an analysis of the literature reveals that
malnutrition is an independent risk factor in many disease processes
and that treatment of malnutrition can indeed improve the patients’
prognosis.
Such an analysis has to address several questions, mainly the
prevalence and diagnosis of malnutrition and its impact on the
patients’ prognosis. It should set the stage for the papers in this issue
of Digestive Diseases, which deals with nutritional questions.
Prevalence and Diagnosis of Malnutrition in
Hospitalized Patients
An evaluation of the nutritional status of hospitalized patients in
different countries shows a prevalence of malnutrition between 20
and 60% of admitted patients (table 1). The prevalence is not differ-
ent among different countries; however, a subgroup analysis showed
that geriatric patients as well as patients with malignant diseases car-
ry the highest risk of becoming malnourished.
The parameters used do diagnose malnutrition varied consider-
ably in the different studies. In older studies, usually static parame-
ters of the nutritional status, like body mass index or anthropometric
measurements, were used. However, later on, it became clear that
body composition reflecting the percentage of lean body mass, water
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0196$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
and fat, or indices including dynamic parameters, like weight loss or
a stress factor, do correlate better with the patients’ prognosis than
plain body weight or body mass index. From these studies, several
nutritional indices have been developed which allow conclusions
about the prognosis and might therefore be used to define patients
with an indication for nutritional therapy.
The subjective Global Assessment (SGA) is one of the most wide-
ly used nutritional indices [1]. It consists of a patient history part
(recent weight loss, changes in eating habits, gastrointestinal symp-
toms, physical fitness and stress factor) and a physical examination
(body weight, subcutaneous fat mass, muscle atrophy, edema). The
subjective global assessment is easy to calculate and therefore conve-
nient to use in a clinical setting. A similar index (NRS 2002) has
recently been developed by the European Society of Parenteral and
Enteral Nutrition [2]. It also includes a patient history (weight loss,
reduced dietary intake) physical parameters (body mass index) and a
disease severity factor. Like the SGA, it can be calculated quickly and
is therefore useful in the clinical situation. It is important to mention
that malnutrition does not necessarily correlate with a low body
mass. Changes in body composition with an increased fat mass and
decreased body cell mass pose a similar risk to the patient than overt
malnutrition with reduced body mass, as has been shown in a recent
investigation [3].
Impact of Malnutrition on Prognosis
Malnutrition has a profound impact on a patient’s prognosis.
Malnourished patients have a higher risk of postoperative complica-
tions when compared to well-nourished patients and a longer hospi-
tal stay [3–7]. Furthermore the long-term mortality of malnourished
patients is much higher than that of comparable well-nourished
patients [8]. This is particularly important since physicians in the
hospital usually see the patient only for 7–10 days during his/her hos-
pital stay. As has been shown in several studies, posthospital mortali-
ty is significantly increased in malnourished patients. The fact that
this complication of malnutrition is developing so late explains why
hospital physicians are not more aware of malnutrition-associated
risks.
Prof. Dr. Herbert Lochs
Universitätsklinikum Charité
Medizinische Klinik mit Schwerpunkt Gastroenterologie, Hepatologie und Endokrinologie
Schumannstrasse 20/21, DE−10098 Berlin (Germany)
Tel. +49 30 450 514021, Fax +49 30 450 514923, E-Mail herbert.lochs@charite.de
Table 1. Prevalence of malnutrition in hospitalized patients prospectively calculated the expenses for 100 patients. Well-nour-
ished patients had a median stay of 10 days whereas malnourished
Authors n Country Disease Malnour- patients stayed for 16.6 days. The actual cost was USD 7,692 for the
ished, % well-nourished and USD 16,691 for the malnourished patients. On a
DRG basis, payment from well-nourished patients was USD 4,352
Bistrian et al. (1974) [12] 131 USA surgery 50 and USD 5,124 for malnourished patients. This clearly shows that
Bistrian et al. (1976) [13] 251 USA general medicine 44
malnutrition is not just a risk for the patient but also a financial risk
Hill et al. (1977) [14] 105 UK surgery 48
Weinsier et al. (1979) [15] 134 USA general medicine 48 for the hospital. A study by Braunschweig et al. [11] who followed the
Coats et al. (1993) [16] 228 USA general medicine 38 expenses for 404 patients who were hospitalized for more than 7 days
McWhirter and Pennington is especially interesting. Patients were divided into a reference group
(1994) [17] 500 UK multidisciplinary 40 which had not lost weight during their hospital stay and a group who
Cederholm et al. (1995) [6] 205 S geriatric 20 lost weight independently of their initial nutritional status. The con-
Naber et al. (1997) [5] 155 NL internal medicine 45–62
trol group generated an average expenditure of USD 28,631, whereas
Bruun et al. (1999) [18] 244 USA surgery 39
Edington et al. (2000) [19] 850 UK multidisciplinary 20
the group who lost weight generated an average expenditure of USD
Waitzberg et al. (2001) [20] 4,000 Brazil multidisciplinary 20 45,762.
Pirlich et al. (2003) [21] 803 Germany multidisciplinary 22 This issue of Digestive Diseases is devoted to nutritional prob-
lems. A number of papers deal with problems of malnutrition in dif-
ferent diseases. Besides the prevalence of malnutrition in hospital-
ized patients, questions concerning the impact of malnutrition in gas-
trointestinal cancer and acute pancreatitis as well as inflammatory
However, there are also economic impacts of malnutrition during bowel disease are discussed. Furthermore, nutritional therapy has
the hospital stay. Reilly et al. [9] calculated the patient costs of 771 proved to be effective not just as treatment of preexisting malnutri-
hospitalized patients in Internal Medicine and Surgery. They defined tion but also to influence disease processes, as has been shown in
malnourished patients on a history of weight loss, low albumin and inflammatory bowel disease, hepatic encephalopathy and other gas-
total lymphocyte count as well as low body mass index. The median trointestinal diseases. By raising their attention to nutritional prob-
expenditures for malnourished patients amounted to USD 11,217 lems, doctors will hopefully increasingly again include nutritional
versus USD 7,660 for well-nourished patients. Robinson et al. [10] therapy in patient care.

References

1 Detsky AS, McLaughlin JR, Baker JP, John-
ston N, Whittaker S, Mendelson RA, Jeejeeboy
KN: What is subjective global assessment of
nutritional status? JPEN J Parenter Enteral
Nutr 1987;11:8–13.
2 Kondrup J, Allison SP, Elia M, Vellas B, Plauth
M: ESPEN guidelines for nutrition screening
2002. Clin Nutr 2003;22:415–421.
3 Kyle UG, Pirlich M, Morabia A, Lochs H,
Schuets T, Pichard C: Abnormal body compo-
sition is associated with increased length of
hospital stay: A controlled population study in
Switzerland and Germany, submitted to Clini-
cal Nutrition.
4 Selberg O, Böttcher J, Tusch G, Pichlmayr R,
Henkel E, Müller MJ: Identification of high-
and low-risk patients before liver transplanta-
tion: A prospective cohort study of nutritional
and metabolic parameters in 150 patients. He-
patology 1997;25:652–657.
5 Naber TH, Schermer T, de Bree A, Nusteling
K, Eggink L, Kruimel JW, Bakkeren J, van
Heereveld H, Katan MB: Prevalence of malnu-
trition in nonsurgical hospitalized patients and
its association with disease complications. Am
J Clin Nutr 1987;66:1232–1239.
6 Cederholm T, Jagren C, Hellstrom K: Outcome
of protein-energy malnutrition in elderly medi-
cal patients. Am J Med 1995;98:67–74.
7 Detsky AS, Baker JP, O’Rourke K, Johnston N,
Whitwell J, Mendelson RA, Jeejeeboy KN:
Predicting nutrition-associated complications
for patients undergoing gastrointestinal sur-
gery. JPEN J Parenter Enteral Nutr 1987;11:
440–446.
8 Pirlich M, Schütz T, Gastell S, Lochs H: Mal-
nutrition affects long-term prognosis in hospi-
talized patients. Gastroenterology 2002;122:
363.
9 Reilly JJ Jr, Hull SF, Albert N, Waller A, Brin-
gardener S: Economic impact of malnutrition:
A model system for hospitalized patients.
JPEN J Parent Enteral Nutr 1988;12:371–376.
10 Robinson G, Goldstein M. Levine GM: Impact
of nutritional status on DRG length of stay.
JPEN J Parenter Enteral Nutr 1987;11:49–51.
11 Braunschweig C, Gomez S, Sheean PM: Im-
pact of declines in nutritional status on out-
comes in adult patients hospitalized for more
than 7 days. J Am Diet Assoc 2000;100:1316–
1322.
12 Bistrian BR, Blackburn GL, Hallowell E, Hed-
dle R: Protein status of general surgical pa-
tients. JAMA 1974;230:858–860.
13 Bistrian BR, Blackburn GL, Vitale J, Cochran
D, Maylor J: Prevalence of malnutrition in gen-
eral medical patients. JAMA 1976;235:1567–
1570.
14 Hill GL, Blackett RL, Pickford I, Burkinshaw
L, Young GA, Warren JV, Schorath CJ, Mor-
gan DB: Malnutrition in surgical patients. An
unrecognized problem. Lancet 1977;i:689–
692.
15 Weinsier RL, Hunker EM, Krumdieck CL,
Butterworth CE Jr: Hospital malnutrition. A
prospective evaluation of general medical pa-
tients during the course of hospitalization. Am
J Clin Nutr 1979;32:418–426.
16 Coats KG, Morgan SL, Bartolucci AA, Wein-
sier RL: Hospital-associated malnutrition: A
reevaluation 12 years later. J Am Diet Assoc
1993;93:27–33.
17 McWhirter JP, Pennington CR: Incidence and
recognition of malnutrition in hospital. BMJ
1994;308:945–948.
18 Bruun LI, Bosaeus I, Bergstad I, Nygaard K:
Prevalence of malnutrition in surgical patients:
Evaluation of nutritional support and docu-
mentation. Clin Nutr 1999;18:141–147.
19 Edinton J, Boorman J, Durrant ER, Perkins A,
Giffin CV, James R, Thomson JM, Oldroyd
JC, Smith JC, Torrance AD, Blackshaw V,
Green S, Hill CJ, Berry C, McKenzie C, Vicca
N, Ward JE, Coles SJ: Prevalance of malnutri-
tion on admission to four hospitals in England.
The Malnutrition Prevalance Group. Clin Nutr
2000;19:191–195.
20 Waitzberg DL, Caiaffa WT, Correia MI: Hos-
pital malnutrition: The Brazilian National Sur-
vey (IBRANUTRI): A study of 4,000 patients.
Nutrition 2001;17:573–580.
21 Pirlich M, Schütz T, Kemps M, Luhman N,
Burmester GR, Baumann G, Plauth M, Lübke
HJ, Lochs H: Prevalence of malnutrition in
hospitalized medical patients: Impact of un-
derlying disease. Dig Dis 2003;21:245–252.

Editorial Dig Dis 2003;21:196–197 197

 Review Article
Dig Dis 2003;21:198–213
DOI: 10.1159/000073337
What We Have Learned about Cachexia
in Gastrointestinal Cancer
J.A. Palesty S.J. Dudrick
St. Mary’s Hospital, Yale University Affiliate, Department of Surgery, Waterbury, Conn., USA
Key Words
Cancer W Cachexia W Cancer cachexia syndrome W
Malnutrition W Anorexia W Gastrointestinal cancer
Abstract
It is appreciated widely by clinicians that significant mal-
nutrition accompanies malignant processes in approxi-
mately 50% of patients and eventually leads to severe
wasting which accounts for approximately 30% of can-
cer-related deaths overall, 30–50% of deaths in patients
with gastrointestinal tract cancers, and up to 80% of
deaths in patients with advanced pancreatic cancer. The
body wasting known as cancer cachexia is a complex
syndrome characterized by progressive tissue depletion
and decreased nutrient intake that is manifested clinical-
ly as inexplicable, recalcitrant anorexia and inexorable
host weight loss. Decreased nutritional intake, increased
metabolic expenditure and dysfunctional metabolic pro-
cesses, including hormonal and cytokine-related abnor-
malities, all appear to play roles in the development of
cancer cachexia. Although this condition of advanced
protein-calorie malnutrition, sometimes described as the
cancer anorexia-cachexia syndrome, is not entirely un-
derstood, it appears to be multifactorial, is a major cause
of morbidity and mortality in cancer patients, and ulti-
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0198$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
mately leads to death. Therapeutic interventions have
met with little success, and, regardless of tremendous
efforts throughout the decades, the exact nature of the
mediators responsible for cancer cachexia remain elu-
sive. The pathogenesis of cancer cachexia appears to be
related to proinflammatory cytokines, alterations in the
neuroendocrine axis and tumor-derived catabolic fac-
tors. Despite trials of conventional and/or aggressive
nutritional support by a myriad of feeding techniques,
patients with cancer cachexia have failed to gain consis-
tent significant benefits in terms of weight gain, function-
al ability, quality of life or survival. Additionally, attempts
to ameliorate the abnormal clinical and metabolic fea-
tures of cancer cachexia with a variety of pharmacologic
agents have met with only limited success. Either until
cancer of the gastrointestinal tract can be cured or until it
is possible to identify the exact causes and mechanisms
of the cancer cachexia syndrome, the most realistic and
practical options currently are directed toward minimiz-
ing adverse gastrointestinal side effects or complications
of the malignant process and/or therapy, as well as
increasing appetite, food intake and nutrient utilization in
an effort to enhance quality of life and improve survival.
Copyright © 2003 S. Karger AG, Basel
Stanley J. Dudrick, MD, FACS
St. Mary’s Hospital, Yale University Affiliate
Department of Surgery, 56 Franklin Street
Waterbury, CT 06706 (USA)
Tel. +1 203 5746479, Fax +1 203 5975877, E-Mail sdudrick@stmh.org
 Introduction
Cachexia results from the inexorable and recalcitrant
progression of nutritional deterioration, which occurs in
approximately 50% of patients with malignant disorders,
and in about 80% of patients with cancers involving the
upper gastrointestinal tract. It is among the most debili-
tating and life-threatening aspects of cancer, and results in
body mass wasting that accounts for up to 30% of cancer-
related deaths [1]. Clinically, the cancer cachexia syn-
drome is characterized primarily by anorexia, early satie-
ty, wasting, weight loss, weakness, fatigue, poor mental
and physical performance, decreased capacity for wound
healing, impaired immunologic function, and a compro-
mised quality of life, none of which are resolved by forced
nutrient intake [1, 2]. Diminished nutritional intake,
increased metabolic expenditure and disordered or mal-
functioning metabolic processes, including hormonal and
cytokine-related abnormalities, all appear to play roles in
the development of cancer cachexia [1, 3]. Although this
condition of advanced protein-calorie malnutrition, fre-
quently referred to as the cancer anorexia-cachexia syn-
drome, is not entirely understood, it appears to be multi-
factorial, it is a major cause of morbidity and mortality in
cancer patients, and it ultimately leads to death [3, 4].
The origin of the word cachexia is both descriptive and
incriminating. It is derived from the Greek words kakos,
meaning bad, and hexis, meaning condition, habit or state
of being [2, 3].
Cancer cachexia was first reported as a commonly
occurring syndrome more than 70 years ago, when an
autopsy series of 500 cancer patients documented that the
immediate cause of death in cancer patients was second-
ary to inanition in 114 (22%) of the patients, and that up
to two-thirds of this cadre of patients exhibited some
degree of cachexia [5]. Since then, the scope of malnutri-
tion in the cancer patient has been studied in a wide vari-
ety of patient groups. In a series of 3,047 patients enrolled
in Eastern Cooperative Oncology Group (ECOG) chemo-
therapy protocols and who were all assessed for weight
loss before initiating chemotherapy, survival was signifi-
cantly lower in the patients who demonstrated weight loss
when compared with those who had not lost any weight
prior to starting chemotherapy [6, 7]. In this study group,
patients with breast cancer or sarcomas had the lowest fre-
quency of weight loss (31–40%), patients with colon can-
cer had an intermediate frequency of weight loss (48–
61%), and patients with pancreatic or gastric cancer had
the highest frequency of weight loss (83–87%), with about
one-third of these patients having presented initially with
What We Have Learned about Cachexia in
Gastrointestinal Cancer
greater than 10% weight loss [6, 7]. In a prospective study
of 280 cancer patients, malnutrition was related predomi-
nantly to tumor type and site, with stomach and esopha-
geal cancer patients demonstrating significantly higher
degrees of malnutrition compared with the other groups.
Moreover, as expected, malnutrition was shown to in-
crease in severity as the disease advanced [8]. In another
study of 365 patients with gastrointestinal cancer, almost
50% were shown to be malnourished. The incidence of
malnutrition was related to the site of the disease, and the
stage of the disease was identified as a predictor of weight
loss, with over 50% of stage III patients manifesting mal-
nutrition [9].
Although nutritional status is usually evaluated by a
combination of clinical assessment and anthropometric
tests, including body weight, skin-fold thickness and mid-
arm circumference, most clinicians rely on body weight as
the major measure of nutritional status, using usual adult
body weight as a point of reference [2].
Cachexia should be expected if an involuntary or unex-
pected weight loss of greater than 5% has occurred within
a previous 6-month period, especially when combined
with muscle wasting. A weight loss of 10% or more ordi-
narily indicates severe depletion, but can also be used as a
starting criterion for the anorexia-cachexia syndrome in
obese patients. Body compartment analysis has shown
that patients with cachexia lose approximately equal
amounts of fat and fat-free mass. Losses of fat-free mass
occur primarily from skeletal muscle and reflect decreases
both in cellular mass and intracellular potassium concen-
tration [2, 10]. Cancer patients with documented 5%
weight loss have a shorter median survival rate than
patients with stable weight, respond poorly to chemother-
apy, and experience increased toxicity to chemotherapy
[2, 7, 11].
How does cancer cachexia compare with other condi-
tions associated with weight loss, such as the anorexia
which may occur in uncomplicated starvation, or that fol-
lowing a major injury or accompanying sepsis? Usually,
the non-cancer patient with acute starvation initially de-
velops rapid breakdown of protein from lean body mass,
amino acid mobilization, gluconeogenesis and increased
excretion of nitrogen in the urine. These early responses
to the stressful stimulus are altered subsequently by the
body to conserve protein and energy. Addition of a major
injury or sepsis to simple starvation, or to the presence of
a tumor, compounds the situation by increasing the rate
and duration of the catabolic response, and the associated
high rate of tissue breakdown may persist for quite some
time [12]. The patient with cancer cachexia exhibits fea-
Dig Dis 2003;21:198–213 199
Table 1. Nutritional problems associated with the development of
cancer cachexia (adapted from Shils [14])
1 Anorexia and early satiety with progressive weight loss and
undernutrition
2 Altered taste or dysgeusia causing hypophagia, food aversion or
altered food intake
3 Alterations in protein, carbohydrate and fat metabolism
4 Increased energy expenditure despite decreased body mass
5 Impaired food intake secondary to:
A. Mechanical obstruction of the gastrointestinal tract at any
level
B. Intestinal dysmotility induced by various tumors
6 Malabsorption secondary to:
A. Deficiency or inactivation of pancreatic digestive enzymes
B. Deficiency or inactivation of bile salts
C. Failure of ingested food to mix and interact effectively with
digestive enzymes
D. Intestinal fistulas – internal and/or external
E. Infiltration of the small bowel wall or lymphatics and
mesentery by malignant cells
F. Blind loop syndrome with associated bacterial overgrowth
G. Malnutrition-induced villous hypoplasia in the small
intestine
7 Protein-losing enteropathy
8 Metabolic abnormalities induced by tumor-derived eutopic
hormones or peptides
A. Hypoglycemia induced by insulin-secreting tumors
B. Hyperglycemia induced by islet glucagonoma or somato-
statinoma
C. Hypercalcemia and hypophosphatemia with osteomalacia
induced by parathyroid hormone-like polypeptides secreted
by some tumors
D. Anemia secondary to chronic blood loss and/or bone
marrow suppression
9 Electrolyte and fluid derangements secondary to:
A. Persistent vomiting and intestinal obstruction or intra-
cranial tumors
B. Intestinal fluid losses through fistulas or diarrhea
C. Intestinal secretory abnormalities with hormone-secreting
tumors such as:
a) Carcinoid syndrome
b) Zollinger-Ellison syndrome (gastrinoma)
c) Verner-Morrison syndrome (VIPoma)
d) Villous adenoma
e) Increased calcitonin
D. Inappropriate antidiuretic hormone (ADH) secretion
associated with specific tumors such as lung carcinomas
E. Hyperadrenalism secondary to tumors producing cortico-
tropins or corticosteroids
10 Miscellaneous organ dysfunction with nutritional complica-
tions such as intractable gastric ulcers with gastrinomas, Fan-
coni’s syndrome with light-chain disease, or coma with brain
tumors
11 Tumor products stimulating monocyte production of various
interleukins
200 Dig Dis 2003;21:198–213
tures of both starvation and injury. However, in the can-
cer patient, not only is food intake usually inadequate, but
the presence of the malignant tumor may decrease the
ability of the patient to regulate the response to starvation.
Cancer cachexia is more likely to resemble the condition
produced by a major injury or sepsis than cachexia sec-
ondary to simple starvation [13].
In addition to the multiple effects of cancer on the
nutritional status of the patient, the antineoplastic thera-
pies currently in use can have adverse effects contributing
to the cachexia of the patient. Understanding the multiple
nutritional, metabolic and physiologic changes that can
occur provides the basis for evaluating the requirements
for nutritional support in various clinical situations and
for planning and providing support as indicated. Al-
though a localized tumor may exert various initial sys-
temic and generalized effects in a patient, as the tumor
grows and metastasizes, these effects often become even
more obvious as a result of the increased tumor burden
and its local and distant invasion. Additionally, some
malignant tumors can produce effects at a distance from
the original tumor or its metastases, further compounding
malnutrition [14]. The nutritional problems associated
with neoplastic disease which can cause or contribute to
the development of cancer cachexia are outlined in ta-
ble 1.
Metabolic Pathophysiology of Cancer Cachexia
[3]
Inability to gain and/or maintain weight despite seem-
ingly adequate nutrient intake could be due to at least
these obvious possibilities: (1) increased metabolic needs
generated by the accentuated nutritional demands of the
semiautonomous tumor; (2) generalized increased energy
expenditure in the cancer patient; (3) maladaptive metab-
olism, i.e., failure of the cancer patient’s host tissues to
utilize the nutrients available efficiently and effectively to
satisfy the energy requirements of the body [3]. Although
malignant cells do have their own specific nutrient re-
quirements and do expend energy, it has been shown that
this fact does not fully explain the cancer cachexia-
anorexia syndrome in human cancer patients [15]. There-
fore, it is important to differentiate data reported in ani-
mal tumors, which tend to represent a much larger pro-
portion of the total body weight of the organism, from
data reported with human tumors which generally repre-
sent a much smaller proportion of the weight of the host.
Palesty/Dudrick
 Many studies of resting energy expenditure (REE) and
basal energy expenditure have been performed to estab-
lish whether cancer patients are hypermetabolic [15].
These studies have demonstrated that approximately
60% of cancer patients have abnormal REE, of which
approximately 35% are hypometabolic, and 25% are hy-
permetabolic. A confounding aspect of these studies is
that lean body mass, rather than total body mass, corre-
lated with alterations found in measured REE. Thus, can-
cer patients, who ordinarily have early diminished lean
body mass, may be expected to have energy expenditure
levels which are low for their total weight, and what
appears to be hypometabolism, may really represent a
eumetabolic state. Despite this non-standard interpreta-
tion of REE data in cachectic cancer patients, it appears
that although hypermetabolism of the host and/or tumor
plays a role in cancer cachexia, hypermetabolism does not
suffice as a complete explanation of the cancer cachexia
syndrome.
For convenience and understanding, maladaptive me-
tabolism may be separated into metabolism of carbohy-
drate, and fat, which are obviously intimately interrelated
[16]. The most apparent precipitating event of maladap-
tive metabolism in the cachectic cancer patient is decreas-
ing blood glucose or a tendency toward hypoglycemia,
which results both from hypermetabolism of the tumor
and the host tissues, together with decreased nutrient
intake related to anorexia hypophagia or the anatomic or
physiologic changes produced by malignant cells of var-
ious origins. In normal starving human beings in whom
decreased nutrient intake is the principal factor, an imme-
diate increase in glucose production as well as in protein
synthesis and breakdown occurs as a compensatory mech-
anism. This self-preserving adjustment can occur because
the body responds with an increase in use of fat-derived
calories, primarily ketone bodies for energy production
[3]. However, in cancer patients, this conservation of glu-
coneogenesis does not occur, but rather both protein
breakdown and lipolysis occur at increasing rates in the
process of maintaining high rates of glucose synthesis.
Accordingly, the role of insulin in the cancer patient
appears to become paramount at this point as a state of
relative insulin resistance develops. The insulin resistance
state, in turn, is characterized by a decreased uptake and
use of glucose, predominantly in skeletal muscle, and a
tendency toward gluconeogenesis and lipolysis [3]. The
lipolytic component of insulin resistance is manifested by
excess fatty acid oxidation which does not decrease even
with increased fat or caloric intake. Additionally, as
decreased nutrient intake and absorption occur in the
What We Have Learned about Cachexia in
Gastrointestinal Cancer
Table 2. Abnormalities of carbohydrate, protein, and fat metabo-
lism associated with cancer cachexia (adapted from Shils [14])
Carbohydrate metabolism
Glucose intolerance
Insulin resistance
Abnormal insulin secretion
Impaired glucose clearance
Increased glucose production
Increased glucose turnover
Increased Cori cycle activity
Protein metabolism
Increased whole body protein turnover
Increased protein fractional synthesis rates in the liver
Decreased fractional synthesis rates in muscle
Increased hepatic protein synthesis
Recalcitrant muscle protein breakdown
Decreased plasma levels of branched-chain amino acids
Fat metabolism
Excess body fat depletion relative to body protein loss
Increased lipolysis
Increased free fatty acids
Increased glycerol turnover
Decreased lipogenesis
Hyperlipidemia
Failure of glucose to suppress oxidation of free fatty acids
Decreased serum lipoprotein lipase activity despite normal
insulin availability
cancer patient, increased rates of glycerol and free fatty
acid turnover occur which ordinarily does not occur dur-
ing adaptation in a non-tumor-bearing, otherwise healthy
but starving individual. Moreover, in healthy patients, a
glucose infusion suppresses lipolysis, but does not sup-
press lipolysis in cancer patients. Ultimately, the cancer
patient depletes fat stores, develops hypertriglyceridemia
and manifests decreased lipoprotein lipase levels. This
results in failure of production of free fatty acids and
monoacylglycerol, and subsequently, decreased adipocyte
triglyceride synthesis, which depends upon free fatty acids
[3].
A summary of the current status of the understanding
of the metabolic maladaptation of the cancer patient to
decreased nutrient intake and increased metabolic caloric
needs of both the tumor and the host include: increased
hepatic glucose production; failure of uptake of the gluco-
neogenic glucose by muscle, resulting in proteolysis and
production of glycogenic amino acids; uptake and use of
glucose by tumor cells with resulting lactate production
and occasional lactic acidosis; and increased lipolysis with
production of fatty acid and glycerol, ultimately used for
Dig Dis 2003;21:198–213 201
energy and further gluconeogenesis (table 2). Thus, the
nutritional requirements of the tumor are selectively
favored over those of the host by these aberrant metabolic
mandates in the cancer patient [3].
Pathogenic Mechanisms of Anorexia
The intake of energy substrates is significantly reduced
among cancer patients who lose weight, and although
anorexia is unlikely to be responsible solely for the wast-
ing seen in cancer patients, it may be a substantial con-
tributing factor [3, 16, 17]. Cancer patients, especially
those with gastrointestinal cancer, may frequently suffer
from physical obstruction of the alimentary tract, pain,
constipation, maldigestion, malabsorption, debility, or
the side effects of therapy, including opiates, radiothera-
py, or chemotherapy, any of which may lead to decreased
food intake [19, 20]. Additionally, hypercalcemia associ-
ated with cancer is a fairly common condition which can
cause nausea, vomiting and weight loss as well as other
untoward side effects. Moreover, anorexia is an extremely
distressing syndrome because appetite and the ability to
eat have been reported to be the most important factors in
the physical and psychologic aspects (especially depres-
sion) of a patient’s quality of life [12, 21]. However, no
clinical cause of reduced food intake is obvious in a large
number of patients with cancer and cancer cachexia. The
early satiety which accompanies reduced appetite in ano-
rectic cancer patients has been postulated to be caused by
the production of tumor cell factors that exert their effects
by acting on hypothalamic sensory cells [38]. Possible fac-
tors include the satietins which have been purified from
human plasma and found to consist of two extensively
glycosylated glycoproteins. The larger molecule, satie-
tin D, has been shown to produce a long-lasting anorectic
effect when injected into rats, although its role in the
development of anorexia has not been established. Anoth-
er possible cause of anorexia is the increased serotonergic
activity within the central nervous system of patients with
cancer cachexia, attributed to the enhanced availability of
tryptophan to the brain. A close relationship between ele-
vated plasma-free tryptophan and anorexia has been ob-
served in patients with cancer and reduced food intake
[22]. The fact that uptake of tryptophan into the brain is
competitive with uptake of branched-chain amino acids
has suggested the use of BCAA to decrease the incidence
of anorexia [22]. Because weight loss is such a potent stim-
ulus to food intake in healthy human beings, the persis-
tence of anorexia in cancer patients implies a failure of
202 Dig Dis 2003;21:198–213
this adaptive feeding response, which is so impressive in
normal human beings [23–26].
A hormone secreted by adipose tissue, leptin, is now
known to be an integral component of the mechanism for
body weight recognition [19, 27–34]. Because weight loss
causes leptin levels to fall in proportion to the loss of body
fat, it appears that leptin plays an important role in trigger-
ing the adaptive response to starvation [2]. The activity of
the hypothalamic orexigenic signals that stimulate feeding
and suppress energy expenditure are increased by low lep-
tin levels in the brain, which also decrease the activity of
anorexigenic signals that suppress appetite and increase
energy expenditure [2, 23–26]. In experimental animals
that are fasted, most of the orexigenic signals are known to
be up-regulated, suggesting that these signals play an
important role in facilitating the recovery of lost weight. It
appears that neuropeptide Y (NPY) produced in the hypo-
thalamus binds to a receptor (Y-5) in adjacent hypotha-
lamic cells, which in turn increases the response of addi-
tional NPY and stimulates the neuronal basis of appetite
[3]. It has further been shown that leptin also binds to Y-5,
inhibiting NPY activity and producing satiety [35]. Re-
cently, glucagon-like factor 1 and urocortin have also been
shown to be appetite suppressants [3, 36, 37].
The cancer anorexia syndrome may result from circu-
lating factors produced by the tumor or by the host in
response to the tumor, and several cytokines have been
proposed as possible mediators of the cachectic process
[19]. Elevated serum levels of tumor necrosis factor-·,
interleukin (IL)-1, and IL-6 have been found in some, but
not all cancer patients, and the serum levels of these cyto-
kines appear to correlate with the progression of the
tumor [38–40]. Interferon-·, interferon-Á and leukemia-
inhibitory factor are additional cytokines which have
been postulated to play a role in the etiology of cancer
cachexia, but which have not been proven to be responsi-
ble solely for the induction of cachexia [38]. However,
chronic administration of these cytokines, either alone or
in combination, is capable of reducing food intake and
reproducing the cancer anorexia-cachexia syndrome [19,
38–43]. These cytokines may produce long-term inhibi-
tion of appetite and feeding by stimulating the production
and release of leptin and/or by mimicking the hypotha-
lamic effect of excessive negative feedback signaling from
leptin, leading to the normal compensatory mechanisms
for decreases in food intake and body weight [18, 19, 23–
29, 34, 39, 44–46]. Thus, the weight loss which occurs in
cancer patients obviously differs considerably from that
which occurs in simple starvation in otherwise healthy
human beings.
Palesty/Dudrick
 Melanocortins, a family of regulatory peptides, which
include adrenocorticotropin (ACTH) and the melano-
cyte-stimulating hormones, have recently been reported
as potential contributory factors in anorexia and cachexia
[2, 47–49]. This group of peptides and their receptors are
important in memory, behavior and immunity, but also
help to regulate appetite and body temperature [19, 31–
33]. In animal studies, the melanocortin system remained
active during cancer-induced cachexia, whereas the nor-
mal response to marked loss of body weight would have
resulted in down-regulation of the anorexigenic melano-
cortin-signaling system in order to conserve energy stores.
Moreover, blockade of the melanocortin receptor re-
versed the anorexia and cachexia in the animals, suggest-
ing a pathogenetic role for this system [19, 47–49].
The specific roles and the relative importance of the
various factors in the cancer anorexia-cachexia syndrome
remain to be clarified. It is presumed that anorexia is
caused by the same mediator or mediators that also pro-
duce the metabolic derangements of cancer cachexia
itself. Use of antianorectic agents alone may improve
quality of life in the cancer patient, but may not solve the
problem of anorexia and its associated morbidity and
mortality in cancer patients [16].
The role of hormones in the cancer anorexia-cachexia
syndrome must be considered because of the obvious role
that hormones play in the intermediary metabolism of
carbohydrates [50]. Insulin, epinephrine, ACTH, human
growth hormone, and insulin-like growth factor have all
been suggested to have a role in the anorexia-cachexia
syndrome. During early starvation, decreased insulin lev-
els and increased glucagon and epinephrine result in acti-
vation of cyclic adenosine monophosphate and of a pro-
tein kinase that activates hormone-sensitive lipase. Fail-
ure of this normal mechanism may account for the weight
loss in cancer patients who have increased rates of glycer-
ol and free fatty acid turnover compared with starved
healthy patients [16].
Recent studies have indicated the presence of an acidic
peptide in animal tumors which appears to have lipolytic
properties that could lead to some of the characteristics of
the cancer cachexia syndrome [16, 51]. This factor ap-
pears to be a proteoglycan which has been found in the
urine of murine species with tumors as well as in human
patients with cancer anorexia-cachexia syndrome involv-
ing weight loss of 11.4 kg [16, 52]. In animals, this proteo-
glycan has been shown to mobilize free fatty acids from
adipose tissue, and amino acids from muscle, which
prompted the investigators to suggest that the role of these
tumor products is to mobilize nutrients for which the
What We Have Learned about Cachexia in
Gastrointestinal Cancer
tumor has the greatest affinity and which will result in
further increased tumor growth. Investigations are cur-
rently underway to determine whether this proteoglycan
is widely present in cachectic human cancer patients and
whether its pathophysiologic actions mimic those of the
cancer anorexia-cachexia syndrome [3]. A potentially
clinically useful feature of this research is the discovery of
an antagonist (eicosapentaenoic acid) of this proteoglycan
which could have a role in the treatment of the human
anorexia-cachexia syndrome [16, 53].
It is highly likely that multiple mediators rather than a
single mediator probably account for the metabolic and
pathophysiologic abnormalities associated with the an-
orexia-cachexia syndrome. It is possible, indeed likely,
that tumor cells produce multiple active peptides or glyco-
peptides; that they also stimulate the induction of addi-
tional cytokines or lymphokines from host immune cells;
and that they stimulate abnormal hormonal responses to
the metabolic stress induced by the malignancy. It is
unlikely that a single factor could account for the hyper-
metabolism, abnormal glucose metabolism, protein glu-
coneogenesis and insulin resistance, in addition to the
anorexia, anemia and fever that accompany the cancer
anorexia-cachexia syndrome.
Dysfunctional Gastrointestinal Manifestations
As a result of the malignant disease and/or its treat-
ment, abnormalities can occur in the mouth or elsewhere
in the gastrointestinal tract and may interfere with inges-
tion of food [19]. Aberrations of taste and smell have also
been documented in cancer patients [54, 55]. Altered
capacity to recognize and taste sweetness in foods occurs
in more than one-third of patients, whereas bitterness,
sourness, and saltiness are less frequently encountered in
cancer patients [56, 57]. Decreased threshold for recogni-
tion of bitter taste correlates well with meat aversion, and
other aversions to specific foods may be learned or devel-
oped secondary to unpleasant experiences that coincide
with exposure to a particular food [54]. This usually
occurs in cancer patients in association with chemothera-
py [58]. It appears that these changes in taste and smell
correlate with decreased nutrient intake, poor response to
therapy, and tumor progression including metastasis [57].
The roles of other factors in the etiology of cachexia
require clarification, including zinc deficiency, alterations
in brain neurotransmitters, and opioid peptides that af-
fect taste and nutrient selection [2, 39, 54, 58]. Direct
involvement in the gastrointestinal tract or its accessory
Dig Dis 2003;21:198–213 203
Table 3. Nutritional problems secondary to gastrointestinal cancer therapy (adapted from Shils [14])

I. Radiation therapy III. Pharmacologic therapy

A. Radiation of the oropharynx A. Corticosteroids
1. Impairment or destruction of taste; xerostomia and 1. Fluid and electrolyte imbalance
odynophagia; loss of teeth 2. Nitrogen and calcium losses
B. Radiation of lower neck and mediastinum 3. Hyperglycemia
1. Esophagitis with dysphagia B. Sex hormone analogue
2. Fibrosis with esophageal stricture 1. Fluid retention
C. Radiation of abdomen and pelvis 2. Nausea
1. Acute and chronic intestinal damage with diarrhea, 3. Glucocorticoid effects secondary to megestrol acetate
malabsorption, stenosis, obstruction, and fistula formation C. Immunotherapy
1. Tumor necrosis factor
II. Surgery therapy
a) Fluid retention
A. Radical resection of oropharyngeal structures
b) Hypotension
1. Chewing and swallowing difficulties
c) Nausea and vomiting
B. Esophagectomy
d) Diarrhea
1. Gastric stasis and hypochlorhydria secondary to vagotomy
2. Interleukin-2
2. Steatorrhea secondary to vagotomy
a) Hypotension
3. Diarrhea secondary to vagotomy
b) Fluid retention
4. Early satiety
c) Azotemia
5. Reflux and/or regurgitation
3. Interferons
C. Gastrectomy (total or near total)
a) Anorexia
1. Dumping syndrome
b) Nausea and vomiting
2. Malabsorption
c) Diarrhea
3. Achlorhydria and lack of intrinsic factor
d) Azotemia
4. Hypoglycemia
D. Cytotoxic chemotherapy
5. Early satiety
1. Nausea and vomiting
D. Intestinal resection
2. Mucositis and stomatitis
1. Jejunum
3. Diarrhea
a) Decreased efficiency of absorption of many nutrients
4. Abdominal pain or cramping
2. Ileum
5. Hepatotoxicity
a) Vitamin B12 deficiency
6. Gastrointestinal bleeding with anemia
b) Bile salt malabsorption with diarrhea or steatorrhea
7. Anorexia
c) Hyperoxaluria and renal stone formation
8. Weight loss
d) Calcium and magnesium depletion
9. Inhibition of protein synthesis
e) Fat and fat-soluble vitamin malabsorption
10. Inhibition of purine and pyrimidine synthesis
3. Massive bowel resection
11. Inhibition of DNA synthesis, replication and transcription
a) Life-threatening malabsorption (short bowel
12. Inhibition of various metabolic processes
syndrome)
b) Malnutrition
c) Metabolic acidosis
d) Dehydration
4. Ileostomy and/or colostomy
a) Complications of salt and water balance and
homeostasis digestive organs by tumors can cause problems with diges-
E. Blind loop syndrome tion and nutrient absorption, and subsequently lead to
1. Vitamin B12 malabsorption malnutrition and cachexia [2]. Odynophagia and/or dys-
F. Pancreatectomy
1. Malabsorption
phagia occur most commonly in patients with cancers of
2. Diabetes mellitus the head and neck and esophagus, and tumors elsewhere
in the gastrointestinal tract and the hepatobiliary tract are
often complicated by partial or total obstruction of the
digestive tract, causing nausea and/or vomiting [55]. Met-
astatic cancers from other systems to the gastrointestinal
tract can also cause, or contribute to, similar problems by
extrinsic pressure on the gastrointestinal tract structures.
Although satiety signals from the gastrointestinal tract
help to regulate appetite and food intake, early satiety is a

204 Dig Dis 2003;21:198–213 Palesty/Dudrick

characteristic of cachectic cancer patients, even without
direct involvement of the gastrointestinal tract, and may
be associated with increased activity of proinflammatory
cytokines including IL-1ß, and centrocorticotropic-releas-
ing factor (CRF), which is a potent anorexigenic factor
[19, 60, 61]. Changes in gastrointestinal motility observed
during stress suggest that CRF may be involved as the
triggering signal [2]. CRF may also induce the delayed
gastric emptying and gastric stasis observed not only in
cancer patients, but in non-neoplastic conditions such as
infection and anorexia nervosa [2, 54, 62, 63]. These
abnormal gastrointestinal effects may result in early satie-
ty and may have a negative influence on food intake.
Another major cause of malnutrition leading to an-
orexia-cachexia syndrome occurs concomitantly with an-
ticancer therapies [54, 56]. Nausea, vomiting, abdominal
cramping, bloating, mucositis and paralytic ileus can all
accompany chemotherapy. Fluorouracil, adriamycin,
methotrexate and cisplatin are among the many antineo-
plastic agents that may induce severe gastrointestinal
complications [64]. Because enterocytes are rapidly divid-
ing cells, they are more susceptible to the cytotoxic effects
of both chemotherapy and radiotherapy than most other
cells in the body. Erosive lesions that can occur at various
levels of the digestive tract, resulting in the impairment of
food intake, digestion and nutrient absorption, can be
caused by both chemotherapy and radiotherapy [2]. Con-
sequences of cancer treatment predisposing to nutritional
problems are listed in table 3.
Nutritional Support with Cancer Cachexia
Early studies promoted optimism that judicious enter-
al or parenteral nutritional support might overcome can-
cer anorexia, and modulate or obviate malnutrition and
cancer cachexia [19, 65–67]. However, the failure of
aggressive nutritional support to increase lean body mass,
especially skeletal muscle mass, in the patient with cancer
cachexia has been disappointing. In a meta-analysis of 28
randomized studies of cancer patients receiving total par-
enteral nutrition (TPN), the use of TPN preoperatively in
patients with gastrointestinal cancer helped to reduce
major surgical complications and operative mortality sig-
nificantly, but no significant benefit was shown on surviv-
al, tolerance of treatment, toxicity, or tumor response in
patients receiving chemotherapy and TPN [68]. A survey
analyzing results individually for patients receiving radia-
tion therapy and chemotherapy and for patients undergo-
ing major surgery, concluded that the routine use of pre-
What We Have Learned about Cachexia in
Gastrointestinal Cancer
operative intravenous feeding should be limited to pa-
tients unable to sustain lean body mass by oral or enteral
feeding [69]. The authors stated further that intravenous
feeding had not been documented to affect responses ei-
ther to therapy or survival favorably in patients receiving
radiation therapy or chemotherapy. However, they also
affirmed that selecting patients for support with TPN
remains a matter of clinical judgment, and that when
therapy response rates and nutritional morbidity are high,
intravenous feeding should be instituted until the host can
recover from the effects of antitumor therapy [69]. The
authors concluded additionally that there is clear evi-
dence that in individual patients with cancer, intravenous
feeding can prevent death from starvation and decrease
the morbidity of treatment, but only those patients un-
dergoing surgery for gastrointestinal tract neoplasia have
benefited significantly from the addition of adjuvant
intravenous feeding [69]. Another subsequent review con-
cluded that chronically malnourished patients given nu-
tritional support often have restoration of a sense of well-
being and become more physically functional; however,
whether this support results in long-term clinical benefits
is difficult to evaluate [70]. The authors stated also that it
appears that routine application of nutritional support to
all patients undergoing treatment for malignancy is not
justified, and that the primary indication for nutritional
support is for the malnourished patient undergoing a
major operation for upper gastrointestinal malignancy,
and for the chemotherapy patient with severe gastrointes-
tinal dysfunction [70]. In another analysis of 18 trials in
children with cancer, the author acknowledged the clear
benefits of nutritional support in patients undergoing a
bone marrow transplantation; however, there appeared to
be little support for the routine aggressive nutritional sup-
port in the non-surgical oncology patient [71]. Nonethe-
less, the author concluded further that circumstances exist
in which aggressive nutritional support by any and all
routes should be provided, especially during prolonged
inability to eat, when nutrition is secondary to poor
intake, when a nutrition support team is available to
decrease related complications, and when the tumor
present is deemed likely to respond to antineoplastic
treatment [71]. Additionally, parenteral nutrition may
facilitate administration of complete chemoradiation
therapy dosages in esophageal cancer patients and may
have beneficial effects for certain patients with decreased
food intake because of mechanical obstruction of the gas-
trointestinal tract [72–74]. Home parenteral nutrition can
also be efficacious and rewarding for such patients [2]. If
the gastrointestinal tract can be used for nutritional sup-
Dig Dis 2003;21:198–213 205
port, enteral nutrition has the advantage of maintaining
the integrity of the enterocytes, the mucosal barrier, and
immunologic function, as well as the advantage of having
lower cost and fewer adverse side effects [2, 54, 74].
Some surveys of the effectiveness of ineffectiveness of
nutritional support in cancer patients have emphasized
survival as an important criterion [14]. On this matter,
several points merit consideration. First, most surveys did
not include survival data, thus leaving only a relatively
small number of patients in this category. Second, if
malignancies such as metastatic esophageal, stomach, co-
lon and lung cancers do not respond well to any known
antitumor treatments, why should nutritional support be
expected to help such patients beyond delaying death and
perhaps improving the quality of life in some patients in
whom starvation existed to varying degrees? Finally, the
clinician or investigator interested in the subject of effica-
cy of nutritional support in cancer cachexia should review
the published meta-analysis and also the original trial
reports together with other reviews that provide evidence
of the value of many cases of improved nutrition for
appropriate patients with cancer cachexia [14, 75, 76].
There is much more to be learned about the metabolism
of cancer patients with progressive disease, and such
knowledge will likely lead to improvements in nutritional
support as more effective antineoplastic therapies are also
developed [14].
The effects of aggressive nutritional support on tumor
growth and development have been difficult to delineate
in cancer patients and are still being debated [2, 77]. How-
ever, a clear benefit of nutritional support may be derived
in cancer patients with severe malnutrition who may
require surgery or may have an obstructing, but potential-
ly responsive tumor to therapy [2, 55, 73, 78]. Experience
has also indicated that the cancer patient with a severely
dysfunctional alimentary tract resulting from radiation,
chemotherapy, surgery or combination thereof, but who is
free of residual malignant disease, is entitled to proper
management by oral, enteral, and/or parenteral feeding
support which can lead to a prolonged life of good quality
[14]. Finally, controversy remains regarding the influence
of aggressive nutritional support on the quality of life of
patients with advanced cancer, and it is very important to
consider the risks, benefits and ethical aspects involved
before beginning such a regimen. Physician attitudes,
patient age and prognosis, and family or patient’s percep-
tions often play important roles in the decision to admin-
ister nutritional support, and these variables will continue
as issues during the patient’s course [79].
206 Dig Dis 2003;21:198–213
The best treatment for cancer cachexia, especially that
related to cancer of the gastrointestinal tract, is to cure the
cancer. Unfortunately, this lofty goal remains an infre-
quent achievement currently among patients with solid
tumors. Since a cure for cancer is still under investigation
and forthcoming, the best palliative or therapeutic meta-
bolic options available are to stimulate increased nutri-
tional intake and ameliorate or prevent the catabolism of
muscle and fat. This can be accomplished to varying
degrees with a multitude of pharmacologic agents. How-
ever, a goal as monumental as this would be much easier
to achieve if it were possible to identify the exact causes
and mechanisms of the cancer anorexia-cachexia syn-
drome. Currently, only those elements of decreased food
intake directly related to the antineoplastic treatment,
such as nausea, vomiting, mucositis and dysfunctional
gastrointestinal motility, can be somewhat controlled or
modulated clinically. The most realistic and practical
options at this time are directed toward minimizing
adverse gastrointestinal side effects or complications and
increasing appetite and nutritional intake in an effort to
improve quality of life. Several types of drugs which have
been commonly employed in attempts to improve appe-
tite, food intake, sense of well-being and body weight gain,
include progestational drugs, corticosteroids and antisero-
tonergic drugs. Additionally, a newer group of drugs is
under investigation primarily because of their effects on
tumor necrosis factor-·, or their effects on muscle metab-
olism. These and other agents now present the possibility
of confronting this recalcitrant syndrome from a different
approach and an opportunity for combined therapeutic
endeavors to improve the quality of life of these patients
[80].
Progestational Agents
In many patients who were originally receiving proges-
tational drugs as part of a multimodal antineoplastic
treatment regimen, notable increases in appetite and sig-
nificant body weight gain occurred. These empirically
derived findings prompted controlled trials to study these
agents specifically regarding their ability to increase appe-
tite in patients with cancer anorexia-cachexia and perhaps
promote weight gain as well. Megestrol acetate (Megace)
has been the most widely used and studied drug in this
category, and its effects on increasing appetite, caloric
intake and weight gain has been shown to be dose related,
in doses ranging from 160 to 1,600 mg, with an optimal
dosage of about 800 mg/day [80, 81]. Generally, patients
Palesty/Dudrick
are started on the lowest effective dosage (160 mg/day),
and the dose is increased according to the clinical re-
sponse of the patient to the drug [80, 82].
Studies using medroxyprogesterone acetate have also
been associated with a significant increase in appetite
together with increases in serum thyroid binding prealbu-
min (transthyretin) and retinal binding protein without
any concomitant changes in body weight, functional per-
formance status, or anthropometric measurements [80,
83]. On the other hand, a controlled multicenter study
comparing medroxyprogesterone with a placebo demon-
strated significantly approved appetite and body weight
(but not quality of life) in the experimental group versus
the control group in advanced-stage, non-hormone sensi-
tive cancer [80, 84].
A combination of megestrol and prednisone was
shown to improve appetite significantly better, compared
with those patients given a placebo while receiving radia-
tion therapy [85]. Moreover, megestrol was significantly
better as an appetite stimulant than prednisone or fluoxy-
mesterone [80]. The data comparing megestrol with dexa-
methasone as appetite stimulant remain to be clarified
[80, 86].
The possible mechanisms of action of progestational
drugs is unclear, but may be related to their glucocorticoid
activity. Currently suggested explanations for their effec-
tiveness include stimulation of NPY, modulation of cal-
cium channels in the ventromedial hypothalamus, and
hindrance of the activity of cytokines, including TNF,
IL-1 and IL-6 [80, 87–89].
Concerns with the use of either megestrol or medroxy-
progesterone are centered on the associated side effects
including induction of thromboses and thromboemboli,
breakthrough uterine bleeding, hypertension, hypergly-
cemia, peripheral edema, alopecia, Cushing syndrome,
adrenal suppression and adrenal insufficiency [80, 90–
93]. However, these adverse effects have been encoun-
tered only rarely in most clinical trials [80, 94–97]. On the
other hand, caution must be taken in administering these
drugs to patients with known thromboembolic disease,
heart disease or high risk for fluid retention [2, 78].
Glucocorticoids
Although appetite stimulants, including corticoste-
roids, have been of some value in patients with low
nutrient intakes, they have not proven to be as useful in
patients with major metabolic abnormalities or cancer
anorexia-cachexia syndrome. Corticosteroids have been
What We Have Learned about Cachexia in
Gastrointestinal Cancer
shown to increase the sense of well-being and to alleviate
anorexia and asthenia in cancer patients; however, their
effects diminish appreciably after 3–4 weeks. Although
several randomized, controlled clinical studies have dem-
onstrated the symptomatic benefits of some of the differ-
ent corticosteroids, most reports have indicated a limited
beneficial effect on appetite, food intake, sense of well-
being and performance for up to 4 weeks [2, 82, 98–103].
Corticosteroids also have a significant antinausea effect
and can improve asthenia and pain control in some
patients; however, no lasting positive effect on body
weight has been observed. Moreover, prolonged cortico-
steroid administration has a negative effect on protein
synthesis, thus neutralizing potential benefits of increased
appetite. Additionally, prolonged corticosteroid treat-
ment may result in weakness, delirium, osteoporosis and
immunosuppression, compounding these adverse side ef-
fects, which are frequently already present in patients
with advanced cancer [2, 104]. Prednisolone and dexa-
methasone have been shown to improve appetite to a
greater extent than a placebo, and methylprednisolone
may improve quality of life; however, no compelling evi-
dence exists to indicate the superiority of any one gluco-
corticoid in stimulating appetite [2, 20, 78, 100].
Although the exact mechanism of action of these drugs
on appetite is unclear, some studies indicate that the corti-
costeroids inhibit prostaglandin metabolism and inhibit
the synthesis or release of the cytokines IL-1 and TNF-·,
which, in turn, theoretically decrease appetite through
their action on NPY, leptin, CRF and serotonin [2, 20, 78,
100]. At this time, no definitive evidence exists to support
an explanation for the exact mechanism of action of corti-
costeroids in stimulating appetite in cancer anorexia-
cachexia patients.
Anabolic Steroids
Although anabolic steroids have been shown to be
effective among athletes and other healthy people in gain-
ing muscle mass, they have not been shown to achieve this
benefit in gastrointestinal cancer patients [80, 106]. On
the other hand, nandrolone has been shown to increase
weight gain in a group of lung cancer patients by increas-
ing protein synthesis via the androgen receptor. However,
this is the only study that has demonstrated any useful-
ness of anabolic steroids in the treatment of cancer
patients [2, 107].
Dig Dis 2003;21:198–213 207
 Antiserotonergic Drugs
Cyproheptadine is an antiserotonergic drug with anti-
histamine actions which have been useful in the symp-
tomatic relief of rhinitis, conjunctivitis and pruritis asso-
ciated with allergic reactions, which has also been useful
as an appetite stimulant in anorexia nervosa [80]. Al-
though cyproheptadine did not prevent progressive
weight loss in patients with advanced cancer, it was suc-
cessful as a mild appetite stimulant in a randomized con-
trolled study [2, 82, 103, 108]. Because evidence suggests
that anorexia may be mediated by increased serotonergic
activity in the brain, its blockade by cyproheptadine
might mediate an antianorexia effect [2, 109, 110]. How-
ever, its primary side effect is sedation, which might limit
its usefulness in advanced cancer patients [80, 108].
Granisetron and odansetron, which are often used as
antiemetics during chemotherapy, are additional seroto-
nin receptor inhibitors that have been used in the cancer
anorexia-cachexia patient as antagonists to the satiating
effect of serotonin [2, 111]. Although these agents have
failed to promote weight gain in trials to date, they have
been noted to improve the patients’ ability to enjoy food
[2, 112]. Additional investigations in cancer anorexia-
cachexia patients are required to define the role of these
and other antiserotonergic mediators in stimulating appe-
tite.
Metabolic and Physiologic Modulators
Branched-chain amino acids may modulate the pe-
ripheral muscle proteolysis which occurs in cancer ca-
chexia by providing protein-sparing substrate for both
muscle metabolism and gluconeogenesis [2, 104]. When
used in TPN solutions, the branched-chain amino acids
have been shown to improve protein accretion and albu-
min synthesis while improving nitrogen balance [2, 113].
Moreover, oral supplementation of branched-chain ami-
no acids has been used successfully to decrease the severi-
ty of anorexia in cancer patients, theoretically, by compet-
ing with tryptophan, the precursor of brain serotonin,
thus blocking increased hypothalamic activity of seroto-
nin [2, 110, 111].
Following studies in animals, which demonstrated that
eicosapentaenoic acid inhibited lipolysis and muscle pro-
tein degradation in a cachexia model, patients with pan-
creatic cancer who received supplements of fish oil ex-
hibited decreased fatigue and a slight body weight gain
as well as a reduction of acute-phase proteins during a
208 Dig Dis 2003;21:198–213
3-month study [115]. In other studies in cachectic cancer
patients, the inclusion of eicosapentaenoic acid in the diet
significantly increased weight gain, lean body mass, per-
formance status and survival, the latter unexpected result
occurring in both the weight-losing and non-weight-losing
groups of patients [116, 117].
Gastroparesis and gastrointestinal dyskinesia are rec-
ognized complications of cancer cachexia which can cause
anorexia, nausea, early satiety and constipation, leading
to impaired food intake [2, 118]. Because many patients
with advanced cancer have delayed gastric emptying and
gastroparesis, prokinetic agents such as metoclopramide
have been used in cancer patients to prevent or ameliorate
the emesis associated with chemotherapy and may alle-
viate anorexia and early satiety with minimal side effects
[2, 54, 82, 119]. Other prokinetic agents such as domperi-
done and erythromycin require additional controlled ran-
domized studies to define their roles in treatment of can-
cer cachexia patients [2, 63, 104].
Other Potentially Useful Drugs
Because hydrazine sulfate inhibits phosphoenol-pyru-
vate carboxykinase, an integral enzyme in gluconeogene-
sis, it was hypothesized that by interrupting the Cori cycle
and normalizing carbohydrate metabolism, this agent
might be useful in improving appetite, caloric intake and
nutritional status in cachectic cancer patients [2, 20].
However, several large studies showed that this drug has
no significant impact on weight gain or any other benefit
in patients with advanced colorectal cancer [120–122].
Increased survival times, improved appetite and in-
creases in caloric intake and serum albumin levels have
been reported with the use of hydrazine sulfate in some
studies, but these favorable effects have not been con-
firmed by others [79]. Because of the lack of definitive
studies showing consistent efficacy, together with the sig-
nificant risk of neurotoxicity associated with its use, this
drug has not gained popularity with oncologists [2, 80].
Thalidomide has been shown to decrease the activity
of TNF-· through cytokine modulation in cancer pa-
tients, in addition to improving insomnia, restlessness
and nausea in advanced cancer patients, while improving
appetite and sense of well-being in the majority of patients
studied [2, 80, 123]. Side effects of thalidomide include
fever, skin rash and peripheral neuropathy in addition to
its well-known teratogenic effects [80, 124, 125].
Melatonin, a hormone produced by the pineal body,
has promoted increased weight gain and increased appe-
Palesty/Dudrick
tite in patients with advanced cancer in association with
its ability to decrease levels of circulating TNF-· [2, 80,
126]. A study of 100 patients with metastatic carcinoma
showed that weight loss of more than 10% was a much less
common finding among patients receiving melatonin
than among the patients receiving a placebo [2, 80, 126].
Pentoxifylline is a methylxanthine-derived phospho-
diesterase inhibitor which has been shown to decrease
TNF-· synthesis in cancer patients by decreasing gene
transcription [80, 127, 128]. However, its use in a con-
trolled trial of cancer patients with solid tumors elicited
no increase in appetite or body weight gain compared
with the controls [80, 129].
5-Deoxy-5-fluorouridine, a fluorinated pyrimidine nu-
cleoside, is a cytostatic agent which has been shown to
decrease the progression of cachexia in an animal model
by inhibiting IL-6 and proteolysis-inducing factor [2, 130,
131]. Although this agent decreased the progression of
tumor growth, it did not demonstrate preservation of
body weight even though it ameliorated progressive
weight loss and hyperglycemia while increasing the pro-
duction acute phase proteins [2, 130].
Cannabinoids are well-known appetite stimulants
which can promote weight gain. Marijuana and its deriva-
tives, dronabinol, marinol and nabilone have shown their
effectiveness as antiemetics in cancer patients especially
those receiving chemotherapy [2, 78, 102–104]. A num-
ber of studies have demonstrated improved appetite after
administration of cannabinoids; however, minimal im-
provement in overall body weight was achieved [2, 132,
133]. Limitations of these agents as potential appetite
stimulants include their adverse effects on the central ner-
vous system such as somnolence, confusion and percep-
tual disturbances which may be augmented further in
advanced cancer patients who often already exhibit some
cognitive impairment [2, 80, 134, 135].
ß2-Adrenergic receptor agonists have been shown to
prevent muscle protein wasting in tumor-bearing rats and
to increase muscle mass and function significantly in non-
tumor-bearing rats [80, 136–140]. Even without changing
food intake or rate of tumor growth, a recent study dem-
onstrated that salbutamol, salmeterol and clenbuterol had
a positive effect on muscle mass in tumor-bearing rats
[141]. Although studies in human beings are limited, this
class of drugs has been shown to improve muscle strength
significantly following orthopedic surgery, even without
the need for exercise. However, to date, no controlled
studies have been reported using these agents in cancer
patients [2, 142].
What We Have Learned about Cachexia in
Gastrointestinal Cancer
Non-steroidal anti-inflammatory drugs (NSAIDS) re-
duce circulating levels of IL-6, acute phase proteins and
cortisol, which renders them useful in treating fever and
pain in patients with advanced cancer [2]. In a series of
colorectal cancer patients, ibuprofen was shown to nor-
malize protein kinetics in cachectic patients and stabilize
weight, while reducing REE and increasing quality of life
in patients with pancreatic cancer [2, 143–146]. The main
limitation of this class of drugs is their adverse effect on
the gastrointestinal tract; however, recent studies into the
efficacy of COX-2 inhibitors such as celecoxib and rofe-
coxib, which appear to have fewer and less severe gas-
trointestinal side effects, have indicated promising early
results in their use as alternatives to the older NSAIDS.
Commentary
In this review of cachexia associated with gastrointesti-
nal cancer, we have attempted to summarize the more
common metabolic abnormalities in the cachectic cancer
patient, the theories proposed as explanations of these
phenomena and the nutritional and pharmacologic inter-
ventions currently in use for amelioration of the severe
host tissue depletion caused by this wasting syndrome.
The metabolic pathophysiology of cancer cachexia has
been reviewed, discussed and summarized, and the cur-
rent status of the understanding of the metabolic malad-
aptation of the cancer patient to decreased nutrient intake
and increased metabolic needs of both the tumor and the
host has been outlined. The nutritional problems associat-
ed with the development of cancer cachexia have also
been presented comprehensively with particular empha-
sis on the side effects and complications indigenous to
cancers of the gastrointestinal tract. Additionally, the
abnormalities of carbohydrate, protein and fat metabo-
lism commonly associated with cancer cachexia have
been outlined and discussed from historical, current and
future perspectives. Moreover, the nutritional problems
secondary to individual forms of antineoplastic therapy in
specific regions of the gastrointestinal tract have been
delineated extensively. The established and experimental
nutritional and pharmacologic support options in the
management of patients with cancer cachexia and ad-
vanced gastrointestinal cancer have been described brief-
ly in an attempt to update their status as therapeutic
adjuncts.
The perplexing problem of cancer cachexia and an-
orexia is compounded in patients with malignancies of
the gastrointestinal tract because not only are the usual
Dig Dis 2003;21:198–213 209
cancer-associated mediators of cachexia and anorexia ac-
tive in inducing weight loss, wasting and deterioration of
functional status, but, the normal means by which nu-
trient intake is attained by the patient is compromised by
direct involvement of the gastrointestinal tract anywhere
from the mouth to the anus by tumor, thus interfering
with the normal mechanisms favoring appetite, gustatory
satisfaction, food intake, digestion, nutrient absorption
and substrate assimilation. This ‘double jeopardy’ does
not ordinarily occur with malignancies of other organ sys-
tems unless or until metastases compromise the gastroin-
testinal tract and its appendages secondarily. Thus, the
increased metabolic expenditure and requirements im-
posed by the neoplastic process are less likely to be met by
the gastrointestinal tract alone. Furthermore, curative
and palliative therapies, including surgery, chemothera-
py, radiotherapy and immunotherapy alone or in combi-
nation, impose an additional burden on an already crip-
pled system.
Attempts to overcome this grave physical and func-
tional handicap initially included parenteral nutritional
supplementation and total parenteral nutrition; however,
lack of uniform success in achieving optimal nutritional
status, improvement of function, and increase in survival
has been disappointing and challenging. Currently, the
most promising therapeutic possibilities are dependent
upon the continuing development of combinations of
antineoplastic therapies, unique nutrient formulations,
and pharmacologic agents designed to ameliorate the met-
abolic derangements induced by the tumor and its prod-
ucts. Obviously, the ultimate answer to this vexing situa-
tion is to assault the biology of the neoplastic process at its
foundation and develop a cure for cancer.
Acknowledgment
The authors greatly appreciate the many significant contributions
of Joan Reeser in the preparation of the manuscript.

References

1 Toomey D, Redmond P, Bouchier-Hayes D:
Mechanisms mediating cancer cachexia. Can-
cer 1995;76:2418–2426.
2 Inui A: Cancer anorexia-cachexia syndrome.
CA Cancer J Clin 2002;52:72–91.
3 Puccio M, Nathanson: The cancer cachexia
syndrome. Semin Oncol 1997;24:277–287.
4 DeWys W: Pathophysiology of cancer cachex-
ia: Current understanding and areas of future
research. Cancer Res 1982;42(suppl):721–726.
5 Warren S: The immediate causes of death and
cancer. Am J Med Sci 1932;184:610–615.
6 Harrison LE, Fong Y: Enteral nutrition in the
cancer patient; in Rombeau JL, Rolandelli RH
(eds): Clinical Nutrition Enteral and Tube
Feeding, ed 3. Philadelphia, Saunders, 1997,
pp 300–323.
7 Dewys WD, Begg C, Lavin PT, Band PR, Ben-
nett JM, Bertino JR, Cohen MH, Douglass HO
Jr, Engstrom PF, Ezdinli EZ, Horton J, John-
son GJ, Moertel CG, Oken MM, Perlia C,
Rosenbaum C, Silverstein MN, Skeel RT,
Sponzo RW, Tormey DC: Prognostic effect of
weight loss prior to chemotherapy in cancer
patients. Eastern Cooperative Oncology
Group. Am J Med 1980;69:491–497.
8 Bozzetti F, Migliavacca S, Scotti A, Bonalumi
MG, Scarpa D, Baticci F, Ammatuna M, Pupa
A, Terno G, Sequeira C, Masserini C, Ema-
nuelli H: Impact of cancer, type, site, stage and
treatment on the nutritional status of patients.
Ann Surg 1982;196:170–179.
9 Meguid MM, Meguid V: Preoperative identifi-
cation of the surgical cancer patient in need of
postoperative supportive total parenteral nutri-
tion. Cancer 1985;55(suppl):258–262.
10 Kotler DP: Cachexia. Ann Intern Med 2000;
133:622–634.
11 Rosenbaum K, Wang J, Peirson RN Jr et al:
Time-dependent variation in weight and body
composition in healthy adults. J Parenter En-
teral Nutr 2000;24:52–55.
12 Brennan M: Uncomplicated starvation versus
cancer cachexia. Cancer Res 1977;37:2359–
2364.
13 Shamberger RJ: Nutrition and Cancer. New
York, Plenum Press, 1984, pp 353–366.
14 Shils ME: Nutrition and Diet in Cancer Man-
agement; in Shils ME, Olson JA, Shike M (eds):
Modern Nutrition in Health and Disease, ed 8.
Philadelphia, Lea & Febiger, 1994, pp 1317–
1348.
15 Heber D, Tchekmedyian NS: Mechanisms of
cancer cachexia. Contemp Oncol 1995, special
issue, pp 6–10.
16 Langstein HN, Norton JA: Mechanisms of can-
cer cachexia. Hematol Oncol Clin North Am
1991;5:103–123.
17 Staal-van den Brekel AJ, Schols AM, ten Velde
GP, Buurman WA, Wouters EF: Analysis of
the energy balance in lung cancer patients.
Cancer Res 1994;54:6430–6433.
18 Wigmore SJ, Plester CE, Ross JA, Fearon KC:
Contribution of anorexia and hypermetabo-
lism to weight loss in anicteric patients with
pancreatic cancer. Br J Surg 1997;84:196–197.
19 Bruera E: ABC of palliative care. Anorexia,
cachexia and nutrition. BMJ 1997;315:1219–
1222.
20 Barber MD, Ross JA, Fearon KC: Cancer ca-
chexia. Surg Oncol 1999;8:133–141.
21 Padilla GV: Psychological aspects of nutrition
and cancer. Surg Clin North Am 1986;66:
1121–1135.
22 Cangiano C, Laviano A, Meguid MM, Mulieri
M, Conversano L, Preziosa I, Rossi-Fanelli F:
Effects of administration of oral branched-
chain amino acids on anorexia and caloric
intake in cancer patients. J Natl Cancer Inst
1996;88:550–552.
23 Flier JS: Clinical review 94: What’s in a name?
In search of leptin’s physiologic role. J Clin
Endocrinol Metab 1998;83:1407–1413.
24 Schwartz MW, Dallman MF, Woods SC: Hy-
pothalamic response to starvation: Implica-
tions for the study of wasting disorders. Am J
Physiol 1995;269:949–957.
25 Schwartz MW, Seeley RJ: Seminars in medi-
cine of the Beth Israel Deaconess Medical Cen-
ter. Neuroendocrine responses to starvation
and weight loss. N Engl J Med 1997;336:1802–
1811.
26 Inui A: Feeding and body-weight regulation by
hypothalamic neuropeptides – Mediation of
the actions of leptin. Trends Neurosci 1999;22:
62–67.
27 Friedman JM, Halaas JL: Leptin and the regu-
lation of body weight in mammals. Nature
1998;395:763–770.
28 Elmquist JK, Maratos-Flier E, Saper CB, Flier
JS: Unraveling the central nervous system
pathways underlying responses to leptin. Nat
Neurosci 1998;1:445–450.
29 Woods SC, Seeley RJ, Porte D Jr, Schwartz
MW: Signals that regulate food intake and
energy homeostasis. Science 1998;280:1378–
1383.

210 Dig Dis 2003;21:198–213 Palesty/Dudrick

30 Bray GA, York DA: The Mona Lisa hypothesis
in the time of leptin. Recent Prog Horm Res
1998;53:95–117.
31 Kalra SP, Dube MG, Pu S, Xu B, Horvath TL,
Kalra PS: Interacting appetite-regulating path-
ways in the hypothalamic regulations of body
weight. Endocr Rev 1999;20:68–100.
32 Schwartz MW, Woods SC, Porte D Jr, Seeley
RJ, Baskin DG: Central nervous system con-
trol of food intake. Nature 2000;404:661–671.
33 Inui A: Transgenic approach to the study of
body weight regulation. Pharmacol Rev 2000;
52:35–61.
34 Inui A: Transgenic study of energy homeostasis
equation: Implications and confounding in-
fluences. FASEB J 2000;14:2158–2170.
35 Gerald C, Walker MW, Criscione L, Gustafson
EL, Batzl-Hartmann C, Smith KE, Vaysse P,
Durkin MM, Laz TM, Linemeyer DL, Schaff-
hauser AO, Whitebread S, Hofbauer KG, Ta-
ber RI, Branchek TA, Weinshank RL: A recep-
tor sub-type involved a neuropeptide-Y in-
duced food intake. Nature 1996;382:168–171.
36 Donaldson CJ, Sutton SW, Perrin, Corrigan
AZ, Lewis KA, Rivier JE, Vaughan JM, Vale
WW: Cloning and characterization of human
urocortin. Endocrinology 1996;137:2167–
2170.
37 Spina M, Merlo-Pich E, Chan RK, Basso AM,
Rivier J, Vale W, Koob GF: Appetite suppress-
ing effect of urocortin, a CRF-related neuro-
peptide. Science 1996;273:1561–1564.
38 Tisdale MJ: Biology of cachexia. J Natl Cancer
Inst 1997;89:1763–1773.
39 Inui A: Cancer anorexia-cachexia syndrome:
Are neuropeptides the key? Cancer Res 1999;
59:4493–4501.
40 Moldawer LL, Rogy MA, Lowry SF: The role of
cytokines in cancer cachexia. J Parenter Enter-
al Nutr 1992;16(suppl):43–49.
41 Noguchi Y, Yoshikawa T, Matsumoto A, Svan-
iger G, Gelin J: Are cytokines possible media-
tors of cancer cachexia? Surg Today 1996;26:
467–475.
42 Matthys P, Billiau A: Cytokines and cachexia.
Nutrition 1997;13:763–770.
43 Gelin J, Moldawer LL, Lonnroth C, Sherry B,
Chizzonite R, Lundholm K: Role of endoge-
nous tumor necrosis factor-· and interleukin-1
for experimental tumor growth and the devel-
opment of cancer cachexia. Cancer Res 1991;
51:415–421.
44 Moldawer LL, Copeland EM: Proinflammato-
ry cytokines, nutritional support and the ca-
chexia syndrome: Interactions and therapeutic
options. Cancer 1997;79:1828–1839.
45 Haslett PA: Anticytokine approaches to the
treatment of anorexia and cachexia. Semin On-
col 1998;25(suppl):53–57.
46 Mantovani G, Maccio A, Lai P, Massa E,
Ghiani M, Santona MC: Cytokine involvement
in cancer-related anorexia/cachexia: Role of
megestrol acetate and medroxyprogesterone.
Semin Oncol 1998;25(suppl 6):45–52.
47 Marks DL, Ling N, Cone RD: Role of the cen-
tral melanocortin system in cachexia. Cancer
Res 2001;61:1432–1438.
What We Have Learned about Cachexia in
Gastrointestinal Cancer
48 Wisse BE, Frayo RS, Schwartz MW, Cum-
mings DE: Reversal of cancer anorexia by
blockade of central melanocortin receptors in
rats. Endocrinology 2001;142:3292–3301.
49 Lechan RM, Tatro JB: Hypothalamic melano-
cortin signaling in cachexia. Endocrinology
2001;142:3288–3291.
50 Bartlett DL, Charland SL, Torosian MH: Re-
versal of tumor associated hyperglucagonemia
as treatment for cancer cachexia. Surgery 1995;
118:87–91.
51 Beck SA, Mulligan HD, Tisdale MJ: Lipolytic
factors associated with murine and human can-
cer cachexia. J Natl Cancer Inst 1990;82:1922–
1926.
52 Todorov P, Cariuk P, McDevitt T, Coles B,
Fearon K, Tisdale M: Characterization of a
cancer cachectic factor. Nature 1996;379:739–
742.
53 Beck SA, Smith KL, Tisdale MJ: Anti-cachec-
tic and anti-tumor effect of eicosapentaenoic
acid and its effect on protein turnover. Cancer
Res 1991;51:6089–6093.
54 Nelson KA, Walsh D, Sheehan FA: The cancer
anorexia-cachexia syndrome. J Clin Oncol
1994;12:213–225.
55 Nitenberg G, Raynard B: Nutritional support
of the cancer patient: issues and dilemmas. Crit
Rev Oncol Hematol 2000;34:137–168.
56 DeWys WD: Anorexia as a general effect of
cancer. Cancer 1979;43(suppl):2013–2019.
57 DeWys WD, Walters K: Abnormalities of taste
sensation in cancer patients. Cancer 1975;36:
1888–1896.
58 Bernstein IL: Learned taste aversions in chil-
dren receiving chemotherapy. Science 1978;
200:1302–1303.
59 Glass MJ, Billington CJ, Levine AS: Opioids
and food intake: Disturbed functional neural
pathways? Neuropeptides 1999;33:360–368.
60 Fujimiya M, Inui A: Peptidergic regulation of
gastrointestinal motility in rodents. Peptides
2000;21:11565–1582.
61 Tache Y, Garrick T, Raybould H: Central ner-
vous system action of peptides to influence gas-
trointestinal motor function. Gastroenterology
1990;98:517–528.
62 Inui A, Okano H, Miyamoto M, Aoyama N,
Uemoto M, Baba S, Kasuga MI: Delayed gas-
tric emptying in bulimic patients. Lancet 1995;
346:1240.
63 Okano H, Inui A, Ueno N, Morimoto S, Ohmo-
to A, Miyamoto M, Aoyama N, Nakajima Y,
Baba S, Kasuga M: EM523l, a nonpeptide mo-
tilin agonist, stimulated gastric emptying and
pancreatic polypeptide secretion. Peptides
1996;17:895–900.
64 Mitchell EP: Gastrointestinal toxicity of che-
motherapeutic agents. Semin Oncol 1992;19:
566–579.
65 Copeland EM, MacFadyen BV Jr, Lanzotti VJ,
Dudrick SJ: Intravenous hyperalimentation as
an adjunct to cancer chemotherapy. Am J Surg
1975;129:167–173.
66 Copeland EM, Souchon EA, MacFadyen BV
Jr, Rapp MA, Dudrick SJ: Intravenous hyper-
alimentation as an adjunct to radiation thera-
py. Cancer 1977;39:609–616.
Dig Dis 2003;21:198–213
67 Dudrick SJ, MacFadyen BV Jr, Souchon EA,
Englert DM, Copeland EM: Parenteral nutri-
tion techniques in cancer patients. Cancer Res
1977;37:2440–2450
68 Klein SM, Simes J, Blackburn GL: Total paren-
teral nutrition and cancer clinical trials. Cancer
1986;58:1378–1386.
69 Lowry SF, Brennan MF: Intravenous feeding
of the cancer patient; in Rombeau JL, Caldwell
MD (eds): Parenteral Nutrition. Philadelphia,
Saunders, 1986.
70 Shike M, Brennan MF: Supportive care of the
cancer patient; in DeVita VT, Hellman S, Ro-
senberg SA (eds): Principles and Practice of
Oncology, ed 3. Philadelphia, Lippincott, 1989.
71 Lipman TO: Clinical trials of nutritional sup-
port in cancer. Parenteral and enteral therapy.
Hematol Oncol Clin North Am 1991;5:91–
102.
72 Sikora SS, Ribeiro U, Kane JM 3rd, Landre-
neau RJ, Lembersky B, Posner MC: Role of
nutrition support during induction chemora-
diation therapy in esophageal cancer. J Parent-
er Enteral Nutr 1998;22:18–21.
73 Body JJ: Metabolic sequelae of cancers (exclud-
ing bone marrow transplantation). Curr Opin
Clin Nutr Metab Care 1999;2:339–344.
74 Body JJ: The syndrome of anorexia-cachexia.
Curr Opin Oncol 1999;11:255–260.
75 Bozetti F: Effects of artificial nutrition on the
nutritional status of cancer patients. J Parenter
Enteral Nutr 1989;13:406–420.
76 Chen MK, Souba WW, Copeland EM: Nutri-
tional support of the surgical oncology patient.
Hematol Oncol Clin North Am 1991;5:125–
145.
77 Miller M: Can reducing caloric intake also help
reduce cancer? J Natl Cancer Inst 1998;90:
1766–1767.
78 Nelson KA: The cancer anorexia-cachexia syn-
drome. Semin Oncol 2000;27:64–68.
79 Mercadante S: Parenteral versus enteral nutri-
tion in cancer patients: Indications and prac-
tice. Support Care Cancer 1998;6:85–93.
80 Gagnon B, Bruera E: A review of the drug treat-
ment of cachexia associated with cancer. Drugs
1998;55:675–688.
81 Loprinzi CL, Michalak JC, Schaid DJ: Phase
three evaluation of four doses of megestrol ace-
tate as therapy for patients with cancer anorex-
ia and/or cachexia. J Clin Oncol 1993;11:762–
767.
82 Mantovani G, Maccio A, Massa E, Madeddu
C: Managing cancer-related anorexia/cachexia.
Drugs 2001;61:499–514.
83 Downer S, Joel S, Allbright A, Plant H, Stubbs
L, Talbot D, Slevin M: A double-blind placebo-
controlled trial of medroxyprogesterone acetate
in cancer cachexia. Br J Cancer 1993;67:1102–
1105.
84 Simons JP, Aaronson NK, Vansteenkiste JF,
ten Velde GP, Muller MJ, Drenth BM, Erd-
kamp FL, Cobben EG, Schoon EJ, Smeets JB,
Schouten HC, Demedts M, Hillen HF, Blijham
GH, Wouters EF: Effects of medroxyprogeste-
rone acetate on appetite, weight and quality of
life in advanced stage non-hormone sensitive
cancer: A placebo-controlled multicenter
study. J Clin Oncol 1996;14:1077–1084.
211
85 Fietkau R, Riepl M, Kettner H: Supportive use
of megestrol acetate in patients with head and
neck cancer during radio(chemo)therapy. Eur J
Cancer 1997;33:75–79.
86 Loprinzi CL, Kugler J, Sloan J: Phase III ran-
domized comparison of megestrol acetate,
dexamethasone and fluoxymesterone for the
treatment of cancer anorexia/cachexia. Proc
ASCO 1997;16:48A.
87 McCarthy WD, Crowder RE, Dryden S, Wil-
liams G: Megestrol acetate stimulates food and
water intake in the rat: effects on regional hypo-
thalamic neuropeptide Y concentrations. Eur J
Pharmacol 1994;265:99–102.
88 Costa AM, Spence KT, Plata-Salaman CR,
French-Mullen JM: Residual Ca2+ channel cur-
rent modulation by megestrol acetate via a G-
protein-a s-subunit in rat hypothalamic neu-
rones. J Physiol 1995;487:291–303.
89 Mantovani G, Maccio A, Esu S, Lai P, Santona
MC, Massa E, Dessi D, Melis G, Del Giacco S:
Medroxyprogesterone acetate reduces the pro-
duction of cytokines and serotonin involved in
anorexia/cachexia and emesis by peripheral
blood mononuclear cells of cancer patient. Bio-
chem Soc Trans 1997;25:296S.
90 DeVita VT Jr, Hellman S, Rosenberg SA: Can-
cer Principles and Practice of Oncology, ed 5.
Philadelphia, Lippincott Raven, 1997.
91 Bruera E, Fainsinger RL: Clinical management
of cachexia and anorexia; in Doyle D, Hanks
G, MacDonald N (eds): Oxford Textbook of
Palliative Medicine. London, Oxford Medical,
1993, pp 330–337.
92 Steer KA, Kurtz AB, Honour JW: Megestrol-
induced Cushing’s syndrome. Clin Endocrinol
1995;42:91–93.
93 Leinung MC, Liporace R, Miller CH: Introduc-
tion of adrenal suppression by megestrol ace-
tate in patients with AIDS. Ann Intern Med
1995;122:843–845.
94 Bruera E, Macmillan K, Kuehn N, Hanson J,
MacDonald RN: A controlled trial of megestrol
acetate on appetite, caloric intake, nutritional
status, and other symptoms in patients with
advanced cancer. Cancer 1990;66:1279–1282.
95 Loprinzi CL, Ellison NM, Schaid DJ, Krook
JE, Athmann LM, Dose AM, Mailliard JA,
Johnson PS, Ebbert LP, Geeraerts LH: Con-
trolled trial of megestrol acetate for the treat-
ment of cancer, anorexia and cachexia. J Natl
Cancer Inst 1990;82:1127–1132.
96 Heckmayr M, Gatzeneir U: Treatment of can-
cer weight loss in patients with advanced lung
cancer. Oncology 1992;49(suppl 2):32–34.
97 Feliu J, Gonzalez-Baron M, Berracal A: Useful-
ness of megestrol acetate in cancer cachexia
and anorexia. J Clin Oncol 1992;15:436–440.
98 Moertel CG, Schutt AJ, Reitemeier RJ, Hahn
RG: Corticosteroid therapy of preterminal gas-
trointestinal cancer. Cancer 1974;33:1607–
1609.
99 Bruera E, Roca E, Cedaro L, Carraro S, Chacon
R: Action of oral methylprednisolone in termi-
nal cancer patients: A randomized, double-
blind study. Cancer Treat Rep 1985;69:751–
754.
212 Dig Dis 2003;21:198–213
100 Popiela T, Lucchi R, Giongo F: Methylpred-
nisolone as palliative therapy for female ter-
minal cancer patients. Eur J Cancer Clin On-
col 1989;25:1823–1829.
101 Willox JC, Corr J, Shaw J, Richardson M,
Calman KC, Drennan M: Prednisolone as an
appetite stimulant in patients with cancer. Br
Med J (Clin Res Ed) 1984;288:27.
102 Gagnon B, Bruera E: A review of the drug
treatment of cachexia associated with cancer.
Drugs 1998;55:675–688.
103 Fainsinger R: Pharmacological approach to
cancer anorexia and cachexia; in Bruera E,
Higginson I (eds): Cachexia-Anorexia in Can-
cer Patients. Oxford, Oxford University
Press, 1996, pp 128–140.
104 Argiles JM, Meijsing SH, Pallares-Trujillo J,
Guirao X, Lopez-Soriano FJ: Cancer cachex-
ia: A therapeutic approach. Med Res Rev
2001;21:83–101.
105 Daly JM, Redmond HP, Gallagher H: Peri-
operative nutrition in cancer patients. J Par-
enter Enteral Nutr 1992;16(suppl):100–105.
106 Forbes GB: The effect if anabolic steroids on
lean body mass: The dose-response curve.
Metabolism 1985;34:571–573.
107 Chlebowski RT, Herrold J, Ali I, Oktay E,
Chlebowski JS, Ponce AT, Heber D, Block
JB: Influence of nadrolone decanoate on
weight loss in advanced non-small cell lung
cancer. Cancer 1986;58:183–186.
108 Kardinal CG, Loprinzi CL, Schaid DJ, Hass
AC, Dose AM, Athmann LM, Mailliard JA,
McCormack GW, Gerstner JB, Schray MF: A
controlled trial of cyproheptadine in cancer
patients with anorexia and/or cachexia. Can-
cer 1990;65:2657–2662.
109 Meguid MM, Fetissov SO, Varma M, Sato T,
Zhang L, Laviano A, Rossi-Fanelli F: Hypo-
thalamic dopamine and serotonin in the regu-
lation of food intake. Nutrition 2000;16:843–
857.
110 Laviano A, Meguid MM, Yang ZJ, Gleason
JR, Cangiano C, Fanelli FR: Cracking the rid-
dle of cancer anorexia. Nutrition 1996;12:
706–710.
111 Moertel CG, Kvols LK, Rubin J: A study of
cyproheptadine in the treatment of metastatic
carcinoid tumor and the malignant carcinoid
syndrome. Cancer 1991;67:33–36.
112 Edelman MJ, Gandara DR, Meyers FJ, Ishii
R, O’Mahony M, Uhrich M, Lauder I, Hous-
ton J, Gietzen DW: Serotonergic blockade in
the treatment of the cancer anorexia-cachexia
syndrome. Cancer 1999;86:684–688.
113 Tayek JA, Bistrian BR, Hehir DJ, Martin R,
Moldawer LL, Blackburn GL: Improved pro-
tein kinetics and albumin syntheses by
branched chained amino acid-enriched total
parenteral nutrition in cancer cachexia. A
prospective randomized crossover trial. Can-
cer 1986;58:147–157.
114 Cangiano C, Laviano A, Meguid MM, Mu-
lieri M, Conversano L, Preziosa I, Rossi-
Fanelli F: Effects of administration of oral
branched-chain amino acids on anorexia and
caloric intake in cancer patients. J Natl Can-
cer Inst 1996;88:550–552.
115 Wigmore SJ, Ross JA, Falconer JS, Plester
CE, Tisdale MJ, Carter DC, Fearon KC: The
effect of polyunsaturated fatty acids on the
progress of cachexia in patients with pan-
creatic cancer. Nutrition 1996;12(suppl):27–
30.
116 Barber MD, Ross JA, Voss AC, Tisdale MJ,
Fearon KC: The effect of an oral nutritional
supplement enriched with fish oil on weight
loss in patients with pancreatic cancer. Br J
Cancer 1999;81:80–86.
117 Gogos CA, Ginopoulos P, Salsa B, Apostoli-
dou E, Zoumbos NC, Kalfarentzos F: Dietary
ˆ3 polyunsaturated fatty acids plus vitamin E
restore immunodeficiency and prolong sur-
vival for severely ill patients with generalized
malignancy: A randomized control trial. Can-
cer 1998;82:395–402.
118 Periera J, Bruera E: Chronic nausea; in
Bruera E, Higginson I (eds): Cachexia-An-
orexia in Cancer Patients. Oxford, Oxford
University Press, 1996, pp 23–27.
119 Davis MP, Dickerson D: Cachexia and an-
orexia: Cancer’s covert killer. Support Care
Cancer 2000;8:180–187.
120 Loprinzi CL, Kuross SA, O’Fallon JR, Gesme
DH Jr, Gerstner JB, Rospond RM, Cobau
CD, Goldberg RM: Randomized placebo-
controlled evaluation of hydrazine sulfate in
patients with advanced colorectal cancer. J
Clin Oncol 1994;12:1121–1125.
121 Loprinzi CL, Goldberg RM, Su JQ, Mailliard
JA, Kuross SA, Maksymiuk AW, Kugler JW,
Jett JR, Ghosh C, Pfeifle DM: Placebo-con-
trolled trial of hydrazine sulfate in patients
with newly diagnoses non-small-cell lung can-
cer. J Clin Oncol 1994;12:1126–1129.
122 Kosty MP, Fleishman SB, Herndon JE 2nd,
Coughlin K, Kornblith AB, Scalzo A, Morris
JC, Mortimer J, Green MR: Cisplatin, vin-
blastine and hydrazine sulfate in advanced,
non-small-cell lung cancer: A randomized,
placebo-controlled, double-blind phase III
study of the Cancer and Leukemia Group B. J
Clin Oncol 1994;12:1113–1120.
123 Bruera E, Neumann CM, Pituskin E, Calder
K, Ball G, Hanson J: Thalidomide in patients
with cachexia due to terminal cancer: Prelim-
inary report. Ann Oncol 1999;10:857–859.
124 Reyes-Teran G, Sierra-Madero JG, Martinez
del Cerro V, Arroyo-Figueroa H, Pasquetti A,
Calva JJ, Ruiz-Palacios GM: Effects of thali-
domide on HIV-associated wasting syn-
drome: A randomized, double-blind, placebo-
controlled clinical trial. AIDS 1996;10:1501–
1507.
125 Haslett P, Tramontana J, Burroughs M,
Hempstead M, Kaplan G: Adverse reactions
to thalidomide in patients infected with hu-
man immunodeficiency virus. Clin Infect Dis
1997;24:1223–1227.
126 McMillan DC, O’Gorman P, Fearon KC,
McArdle CS: A pilot study of megestrol ace-
tate and ibuprofen in the treatment of cachex-
ia in gastrointestinal cancer patients. Br J
Cancer 1997;76:788–790.
Palesty/Dudrick
127 Dezube BJ, Sherman ML, Fridovich-Keil JL,
Allen-Ryan J, Pardee AB: Down-regulation of
tumor necrosis factor expression by pentoxi-
fylline in cancer patients: A pilot study. Can-
cer Immunol Immunother 1993;36:57–60.
128 Lissoni P, Ardizzoia A, Barni S, Pittalis S,
Rossini F, Porta A, Tancini G: Characteriza-
tion of cancer-related disseminated intravas-
cular coagulation in relation to tumor necro-
sis factor-· blood concentrations: Possible
therapeutic role of pentoxifylline. Tumori
1996;82:78–80.
129 Goldberg RM, Loprinzi CL, Mailliard JA,
O’Fallon JR, Krook JE, Ghosh C, Hestorff
RD, Chong SF, Reuter NF, Shanahan TG:
Pentoxifylline for treatment of cancer anorex-
ia and cachexia? A randomized, double-blind,
placebo-controlled trial. J Clin Oncol 1995;
13:2856–2859.
130 Tanaka Y, Eda H, Fujimoto K, Tanaka T,
Ishikawa T, Ishitsuka H: Anticachectic activi-
ty of 5-deoxy-5-fluorouridine in a murine tu-
mor cachexia model, colon 26 adenocarcino-
ma. Cancer Res 1990;50:4528–4532.
131 Hussey HJ, Todorov PT, Field WN, Inagaki
N, Tanaka Y, Ishitsuka H, Tisdale MJ: Effect
of a fluorinated pyrimidine on cachexia and
tumor growth in murine cachexia models: Re-
lationship with a proteolysis inducing factor.
Br J Cancer 2000;83:56–62.
132 Nelson K, Walsh D, Deeter P, Sheehan F: A
phase II study of delta-9-tetrahydrocannabi-
nol for appetite stimulation in cancer-associ-
ated anorexia. J Palliat Care 1994;10:14–18.
What We Have Learned about Cachexia in
Gastrointestinal Cancer
133 Plasse TF, Gorter RW, Krasnow SH, Lane M,
Shepard KV, Wadleigh RG: Recent clinical
experience with dronabinol. Pharmacol Bio-
chem Behav 1991;40:695–700.
134 Beal JE, Olson R, Laubenstein L, Morales JO,
Bellman P, Yangco B, Lefkowitz L, Plasse TF,
Shepard KV: Dronabinol as a treatment for
anorexia associated with weight loss in pa-
tients with AIDS. J Pain Symptom Manage
1995;10:89–97.
135 Fainsinger RL, Tapper M, Bruera E: A per-
spective on the management of delirium in
the terminally ill. J Palliat Care 1993;9:4–8.
136 Costelli P, Garcia-Martinez C, Llovera M,
Carbo N, Lopez-Soriano FJ, Agell N, Tessi-
tore L, Baccino FM, Argiles JM: Muscle pro-
tein waste in tumor-bearing rats is effectively
antagonized by a ß2-adrenergic agonist (clen-
buterol): Role of the ATP-ubiquitin-depen-
dent proteolytic pathway. J Clin Invest 1995;
95:2367–2372.
137 Chance WT, Cao LQ, Zhang FS, Foley-Nel-
son T, Fischer JE: Clenbuterol treatment in-
creases muscle mass and protein content of
tumor-bearing rats maintained on total par-
enteral nutrition. J Parenter Enteral Nutr
1991;15:530–535.
138 Rehfeldt C, Weikard R, Reichel K: The effect
of the ß-adrenergic agonist clenbuterol on the
growth of skeletal muscles of rats. Arch Tier-
ernähr 1994;45:333–344.
139 Young OA, Watkins S, Oldham JM: The role
of insulin-like growth factor I in clenbuterol-
stimulated growth in growing lambs. J Anim
Sci 1995;73:3069–3077.
140 Zhou GH, Han ZK: Effects of dietary supple-
mentation of ß2-adrenergic agonist clenbute-
rol on carcass characteristics and some me-
tabolites in ducks. Br Poult Sci 1994;35:355–
361.
Dig Dis 2003;21:198–213
141 Carbo N, Lopez-Soriano J, Tarrago T, Gon-
zalez O, Llovera M, Lopez-Soriano FJ, Ar-
giles JM: Comparative effects of ß2-adrener-
gic agonists on muscle waste associated with
tumour growth. Cancer Lett 1997;115:113–
118.
142 Maltin CA, Delday MI, Watson JS, Heys SD,
Nevison IM, Ritchie IK, Gibson PH: Clenbu-
terol, a ß-adrenoreceptor agonist, increases
relative muscle strength in orthopaedic pa-
tients. Clin Sci 1993;84:651–654.
143 McMillan DC, Leen E, Smith J, Sturgeon C,
Preston T, Cooke TG, McArdle CS: Effect of
extended ibuprofen administration on the
acute phase protein response in colorectal
cancer patients. Eur J Surg Oncol 1995;21:
531–534.
144 Preston T, Fearon KC, McMillan DC, Win-
stanley FP, Slater C, Shenkin A, Carter DC:
Effect of ibuprofen on the acute-phase re-
sponse and protein metabolism in patients
with cancer and weight loss. Br J Surg 1995;
82:229–234.
145 Wigmore SJ, Falconer JS, Plester CE, Ross
JA, Maingay JP, Carter DC, Fearon KC: Ibu-
profen reduces energy expenditure and acute-
phase protein production in cancer patients
compared with placebo in pancreatic cancer
patients. Br J Cancer 1995;72:185–188.
146 McMillan DC, Wigmore SJ, Fearon KC,
O’Gorman P, Wright CE, McArdle CS: A pro-
spective randomized study of megesterol ace-
tate and ibuprofen in gastrointestinal cancer
patients with weight loss. Br J Cancer 1999;
79:495–500.
213
 Review Article
Dig Dis 2003;21:214–219
DOI: 10.1159/000073338
Nutritional Support in Acute
Pancreatitis
Costas Avgerinos Spiros Delis Spyros Rizos Christos Dervenis
1st Department of Surgery, Konstantopoulion, Agia Olga Hospital, Athens, Greece
Key Words
Acute pancreatitis W Gut barrier W Total parenteral
nutrition W Enteral nutrition
Abstract
Acute pancreatitis (AP), mainly the severe necrotizing
type, results in extreme energy demands which might
lead, if prolonged, to severe malnutrition. Besides that,
starving during AP contributes to gut barrier dysfunc-
tion, the main cause of bacterial translocation and sep-
sis. The aim of nutritional support in AP is to prevent mal-
nutrition and protect the gut by maintaining mucosal
integrity. Traditionally, nutritional support during the
acute phase of the disease has been provided through
total parenteral nutrition (TPN) solutions. However, re-
cent animal and human studies have identified new pat-
terns of pancreatic secretion and hormonal stimulation
during the course of AP, different from those assumed
for years. Thus it has become feasible to use the natural
enteral route for nutrition with potential benefits com-
pared with TPN.
Copyright © 2003 S. Karger AG, Basel
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0214$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
Acute pancreatitis (AP) is a disease with a wide spec-
trum of clinical courses, ranging from the mild form with
minimum morbidity and almost zero mortality, to the
severe form with a high percentage of complications and a
high risk for a lethal outcome [1, 2]. The latter form is
characterized by various degrees of necrosis of pancreatic
parenchyma as well as local and systemic complications
such as SIRS and MOF, representing a typical hypermeta-
bolic septic model with high-energy consumption and
protein breakdown that leads to severe malnutrition. On
the other hand, the mild form of the disease is much less
aggressive against body metabolism and is usually self-
limited. Patient’s oral intake starts within 3–7 days after
disease onset, resulting in minor and usually easily revers-
ible nutritional defects. As a result, nutritional support by
enteral feeding or total parenteral nutrition (TPN) should
be one of the main therapeutic aims in AP, and nutrition-
al management should depend on the underlying pan-
creatic disease form.
Malnutrition in Acute Pancreatitis
Malnutrition during the clinical course of AP is a result
of a combination of oral food intake restriction, increased
energy demands and often preexisting nutritional defects
(i.e. alcoholic malnutrition). All AP-pathogenetic factors
Christos Dervenis, MD
1st Department of Surgery, Konstantopoulion
Agia Olga Hospital, 3–5 Ag. Olgas str.
GR–14233 Athens (Greece)
Tel. +30 1 277 546, Fax +30 1 279 3969, E-Mail chrisder@otenet.gr
meet at a starting point where a common pattern leading
to the clinical presentation of the disease begins. This
pathway involves activation of trypsinogen to trypsin
after which an activation cascade of various enzymes is
started leading to pancreatic and peripancreatic tissue
autodigestion. At the same time, liberations of acute
phase reactants resulted in a hemodynamic response simi-
lar to that seen in sepsis syndrome (increased cardiac out-
put, decreased peripheral resistance and increased oxygen
consumption). This is accompanied by alterations in the
carbohydrates, proteins and lipids metabolism. The dis-
ease and its complications are also associated with the
release of cytokines (IL-1, IL-6, IL-8, TNF, PAF), activa-
tion of the compliment and creation of arachidonic acid
metabolic products. These energy-consuming biochemi-
cal alterations in parallel with reduction of food intake
result in a negative energy balance and a rapid loss of lean
body mass which adversely affects the host’s defenses and
immune competence [3]. Non-suppressive gluconeogene-
sis is observed in severe pancreatitis and net protein
catabolism can rise to 40 g nitrogen/day as opposed to
normal subjects where glucose supply cannot inhibit that
phenomenon completely [4, 5]. In 80% with severe AP
complicated by sepsis, a marked increase in resting energy
expenditure is observed and using the Harris-Benedict
equation total energy expenditure is 1.49 times the pre-
dicted resting energy expenditure [6, 7].
Pathophysiology of the Gut in Acute
Pancreatitis
Contamination of pancreatic necrosis and consequent
sepsis is the main cause of death in severe pancreatitis,
although in the early period of the disease, SIRS remains
the main fatal course [8]. Maintenance of gut integrity is
of high priority in severe AP in order to prevent pancreat-
ic necrosis contamination and preserve its immune func-
tion. Bacterial translocation through the gut wall is the
prominently proposed mechanism for pancreatic necrosis
contamination.
In normal subjects the gut which represents the largest
immune organ of the human body acts as a physiologic
barrier against bacteria and endotoxins. Secretory IgA
produced by gut lymphoid tissue prevents bacterial and
viral adherence to the mucosa [9]. Experimental work in
rats has shown that gut barrier function in the small intes-
tine is compromised very early after the onset of AP. This
observation has also been documented in humans [10,
11].
Nutritional Support in Acute Pancreatitis
Under physiologic conditions the entry of nutrients
into the duodenum results in a burst of pancreaticobiliary
digestive secretions that mix with the nutrients in order to
be exposed to the small intestine mucosa. It has been
shown that up to 70–80% of nutrients are absorbed from
the lumen as proximal as the duodenojejunal junction
[12, 13]. The rest remains unabsorbed within the small
intestine. This physiologic malabsorption is believed to
act as an energy and amino acid supply to the mucosa of
small and large intestine [14] with regard to fat, amino
acids and carbohydrates in different proportions [14–
17].
Enteral feeding support in AP aims to avert negative
nitrogen balance, to protect gut barrier function and
therefore prevent secondary pancreatic infection. The
rationale behind the concept of enteral feeding is that
there is at least some evidence that it is important in re-
storing and might prevent morphological changes in the
intestine associated with starvation. Lack of nutrients in
the gut lumen leads to loss of mucosal integrity as a result
of a decrease of mucosal thickness [18]. Enteral feeding
can also reverse the reduction in hilus height occurring
after TPN and theoretically can play an important role in
the management of severe AP as it probably prevents sep-
sis and immune failure.
Pathophysiology of Pancreas in Acute
Pancreatitis
For many years to put the pancreas at rest in AP was
considered as the best medical practice and the corner-
stone of therapeutic manipulations. It was stated that by
that way exacerbation of pancreatic inflammation and
autodigestion would be avoided as the absence of content
in stomach and duodenum diminish further production
of pancreatic proteolytic enzymes [19]. The true clinical
significance of that belief has been recently disputed. The
three phases of exocrine pancreatic secretion (cephalic,
gastric, intestinal) and the real pattern of pancreatic out-
put proved to be different in AP. Pancreatic exocrine
secretion is severely reduced early after the onset of AP
and the cholecystokinin-stimulated secretion is abolished
at the time of maximal histological damage to the pan-
creas [20].
Several experimental and clinical trials have shown
that delivery of nutrients into the jejunum does not
increase pancreatic secretion and is well tolerated with no
increase in complications. More specifically, although ad-
ministration of lipid in the duodenum is a strong stimula-
Dig Dis 2003;21:214–219 215
tory factor for pancreatic exocrine secretion, jejunal deliv-
ery of the same amount of lipids causes minimal pancreat-
ic reaction [21–24]. Intravenous lipid infusion has minor
effects as has been shown in human studies. Gastric or
duodenal protein or carbohydrate administration is also a
strong stimulus for pancreatic secretion, while as expected
jejunal delivery of the same nutrients is harmless to the
pancreas [25–31]. As there is no evidence that putting the
pancreas at rest has any clinical benefit, this theory is
strongly challenged nowadays [32] and enteral feeding has
gained increasing acceptance for nutritional support in
AP.
Energy Supply in Acute Pancreatitis
According to the ESPEN guidelines [33], a hypocaloric
energy supply of 15–20 kcal/kg/day is more suitable dur-
ing the early catabolic stage of non-surgical patients with
MOF in AP. In non-severely ill patients, nutritional
requirements can be resumed at approximately 25–
35 kcal/kg/day with 1.2–1.5 g protein/kg/day. Carbohy-
drates and lipids should be administered 3–6 and 2 g/kg/
day respectively, but care should be taken to avoid high
blood glucose and triglyceride concentrations.
Total Parenteral Nutrition in Acute Pancreatitis
Traditionally, TPN has for many years been the only
nutrient providing treatment in patients with AP and pro-
longed starvation. TPN achieves minimal exocrine pan-
creatic secretions as well as sufficient energy and protein
provision. Since 1974, when Feller et al. [34] showed in an
uncontrolled retrospective study a decrease in the mortal-
ity rate of patients who received intravenous hyperali-
mentation while suffering from AP, several other similar
retrospective uncontrolled clinical trials have failed to
reproduce the same results. On the contrary and despite
differences in the parameters of the trials, other authors
observed a higher incidence of catheter-related sepsis
among TPN groups [35, 36] but no difference in total
mortality [35–37]. Two prospective non-randomized
trials have been published on that subject.
In 1989, Sitzmann et al. [38] divided 73 patients with
AP of various severities into three groups depending on
their ability to tolerate glucose-free, lipid-based and lipid-
free nutrition. Within 15 days, most patients in all groups
achieved an improvement in nutritional status. A higher
mortality rate was observed in the fat-free group as well as
216 Dig Dis 2003;21:214–219
among patients with a persistent negative nitrogen bal-
ance. A high incidence of catheter sepsis was also docu-
mented.
In 1991, Kalfarentzos et al. [39] divided 67 patients
with severe AP (13 Ranson criteria) into two groups of
early (within 72 h after admission) and late (after 72 h)
onset of TPN. They noted a significantly lower incidence
of complications and mortality in the early group but a
high catheter sepsis incidence as well. Sax et al. [40] car-
ried out the only randomized controlled trial on the
effects of early parenteral nutrition versus no nutritional
support in patients with AP. The two groups of patients
(mean Ranson score 1) that received either conventional
therapy alone or conventional therapy with TPN had no
significant differences in length of hospital stay, number
of days of oral intake and number of complications as well
as a ninefold increase in the incidence of catheter sepsis in
the TPN group.
As a conclusion it can be stated that no strong level 1
information regarding the role of TPN in severe AP exists
and more trials are needed in order to show any benefit. A
trend for improvement of morbidity and mortality
though is delineated in patients with severe pancreatitis –
those with prolonged starvation and those who achieved a
positive nitrogen balance.
Enteral Nutrition in Acute Pancreatitis
More recently, attention has been given to the possible
role of the enteral route of delivering the necessary calo-
ries and nutrients. The rationale behind the concept of
enteral feeding is that there is at least some evidence that
is important in restoring and might help in preventing
morphological changes in the intestine associated with
starvation. As reported already, a lack of nutrients in the
gut lumen leads to loss of mucosal integrity as a result of a
decrease of mucosal thickness [32]. Enteral feeding also
can reverse the reduction in villus height occurring after
TPN. In an experimental model, rats with pancreatitis
were infused with Ringer’s lactate solution for 48 h fol-
lowed by parenteral or enteral nutrition until the 7th day.
Results showed lower endotoxin levels, a more prominent
villus height and T-cell levels in animals that received
enteral nutrition compared with those that received TPN
[41]. In theory, enteral nutrition could play an important
role in the treatment of severe AP, as it probably prevents
sepsis and immune failure.
Since the mid-1980s, some clinical evidence from criti-
cally ill patients, especially those with severe injuries, has
Avgerinos/Delis/Rizos/Dervenis
shown that giving enteral nutrition very early could favor-
ably alter the outcome. Moore et al. [42] studied 32
patients with severe trauma and compared early enteral
nutrition with no nutritional intervention. They found a
statistically significant difference in septic morbidity (9
vs. 29%).
The first retrospective study for the effects of the enter-
al nutrition in patients with AP was published in 1973 by
Voitk et al. [43], who demonstrated good tolerance and
even beneficial effects of an elemental diet delivered in
the stomach of 6 patients suffering from complicated pan-
creatitis. Kudsk et al. [44] and Parekh et al. [45] published
their experiences on enteral feeding in retrospective stud-
ies, the former through feeding jejunostomies and the lat-
er through gastric catheters. In general, they demonstrated
improvement of the nutritional status accompanied by
successful resolution of complications, but in both studies
enteral nutrition was commenced after the acute phase of
pancreatitis. Simpson et al. [46] in a similar study pub-
lished favorable results with nasoenteral feeding in 5
patients with acute alcoholic pancreatitis, while Nakad et
al. [47] in a prospective uncontrolled study described
their experience of early jejunal nutrition starting 36 h
after admission. Recently, Eatock et al. [48] in a prospec-
tive uncontrolled study revealed the positive effects of
gastric delivered nutrients in 26 patients with severe AP.
Overall, all of these papers have proved that enteral
nutrition can be started in patients with AP since it is a
safe method of nutritional support, it is well tolerated and
causes no aggravation of the disease. On the contrary,
Powell et al. [49] published in 2001 the only randomized
controlled study comparing 13 patients that received
enteral nutrition within 72 h of disease onset with 14 oth-
ers treated conventionally (nil by mouth). It was shown in
this study that early enteral nutrition did not ameliorate
the inflammatory response as demonstrated by a serum
concentration of inflammatory response markers (CRP,
IL-6, sTNFRI). Furthermore, abnormal intestinal perme-
ability occurred more frequently in patients receiving
enteral nutrition.
Recently, five randomized controlled studies were
published comparing enteral feeding (EN) with TPN. Kal-
farentzos et al. [50] randomized 38 patients, all with
severe AP, in two groups (EN vs. TPN). They found a
significant reduction in total, including septic complica-
tions in the enteral feeding group. The cost was three
times lower in EN than in TPN and they suggested the
preferable use of EN in all patients with severe disease.
The other study was from the UK [51], and 34 patients
were also randomized in two groups (EN vs. TPN). In that
Nutritional Support in Acute Pancreatitis
study, patients with moderate and severe disease were
included. Patients who received enteral feeding fared bet-
ter after 7 days with respect to Apache II score and CRP
levels compared with the TPN group. The authors also
reported an increase in serum IgM endotoxin core anti-
bodies in the TPN group which remained unchanged in
the EN group. The total antioxidant capacity was less in
the former group. They concluded that patients on enteral
nutrition were exposed to lower endotoxin levels, which
was probably related to preserved host defense.
More recently, Abou-Assi and O’Keefe [52] demon-
strated earlier recovery, shorter hospital stay and shorter
duration of nutritional support, better tolerance to restart-
ing oral feeding and much cheaper costs for nutrition in
the group of 17 enterally fed patients with AP compared
with 16 patients who received TPN. Catheter-related sep-
sis and hyperglycemia necessitating insulin were signifi-
cantly more common in the TPN group but overall mor-
tality was no different. Olah et al. [53] compared conven-
tional parenteral nutrition (48 patients) with early jejunal
nutrition (41 patients) in 89 patients admitted with AP.
The rate of septic complications, need for surgery, multi-
ple organ failure and death was higher in the TPN group
but differences were not statistically significant. Finally,
another randomized controlled trial was published by a
Scottish group [54] that studied the effect of early enteral
nutrition on markers of the inflammatory response in pre-
dicted severe AP. Serum IL-6, tumor necrosis factor
receptor I and CRP were used as inflammatory markers.
Despite previous findings, the authors documented that
early enteral nutrition did not ameliorate the inflammato-
ry response in patients with severe AP compared with no
nutritional intervention.
A randomized study is under way by our group to try
and identify the role of early enteral nutrition in severe
AP compared with standard TPN in reducing the need for
surgery in patients with predicted severe AP. We recently
reported preliminary results (23 patients) where we
showed that early enteral nutrition seems to reduce surgi-
cal interventions in the enteral nutrition group by reduc-
ing the incidence of sepsis (9 vs. 33%) [55].
Although the current practice is to use nasojejunal
tubes, placed endoscopically or under radiographic
screening, a recent study by the Glasgow group [48]
showed that nasogastric feeding is usually possible in
severe AP. They reported that this practice is safe, well
tolerated and without any sign of clinical or biochemical
deterioration.
Another question that needs to be answered is the pos-
sible role of immune-enhanced enteral diets in severe AP
Dig Dis 2003;21:214–219 217
[56]. There are a number of reports, mainly in severely
injured patients, dealing with this aspect [56–58]. A meta-
analysis [57] including 1,009 patients from 11 trials
showed that immune-modulated regimens resulted in a
significant reduction of infectious complications and
length of hospital stay, but with no effect on survival.
Only one study has dealt with the use of glutamine in AP,
as a supplement in the standard TPN. They found that
glutamine improves leukocyte activity and reduces proin-
flammatory cytokine release in AP. No conclusions can be
drawn from these studies, but as it seems possible that
immune-enriched diets could play a role, further trials are
needed to clarify this issue.
Conclusions
At present there is no definite evidence that artificial
nutrition support – either total parenteral or enteral –
alters the outcome in most patients with AP, unless mal-
nutrition is also a problem. Diagnosis of AP is not itself an
indication for instituting artificial nutrition, unless severi-
ty of the disease is the case. Enteral nutrition support is
safe, well tolerated and does not stimulate the pancreas
compared with parenteral feeding, and therefore should
be used preferably in the treatment or prevention of mal-
nutrition and probably immunosuppression and infec-
tion, in patients with AP. Finally, larger, well-conducted
trials, recruiting only patients with severe AP and strati-
fying them for disease severity, nutritional status and eti-
ology of pancreatitis before randomization, are needed
before any conclusive statement on the benefits of nutri-
tional support on outcome can be made.

References

1 Wycoll DL: The management of severe acute
necrotizing pancreatitis: An evidence-based re-
view of the literature. Intensive Care Med
1999;25:146–156.
2 Dervenis C, Johnson CD, Bassi C et al: Diagno-
sis, objective assessment of severity and man-
agement of acute pancreatitis: Santorini Con-
sensus Conference. Int J Pancreatol 1999;25:
195–210.
3 Scolapio JS, Mahli-Chowla N, Ukleja A: Nutri-
tion supplementation in patients with acute
and chronic pancreatitis. Gastroenterol Clin
North Am 1999;28:695–707.
4 Eckerwall G, Andersson R: Early enteral nutri-
tion in severe acute pancreatitis: A way of pro-
viding nutrients, gut barrier protection, immu-
nomodulation or all of them? Scand J Gas-
troenterol 2001;5:449–458.
5 Heys S, Walker L, Smith I, Eremin O: Enteral
nutritional supplementation with key nutrients
in patients with critical illness and cancer. Ann
Surg 1999;229:467–477.
6 Dickerson RN, Vehe KL, Mullen JL, Feurer
ID: Resting energy expenditure in patients with
pancreatitis. Crit Care Med 1991;19:484–490.
7 Bouffard YH, Delafosse BX, Annat GJ, Viale
JP, Bertrand OM, Motin JP: Energy expendi-
ture during severe acute pancreatitis. J Paren-
ter Enteral Nutr 1989;13:26–29.
8 Schwarz M, Büchler M, Thomsen J et al: Pan-
creatic infection in acute pancreatitis: A fre-
quent finding. Digestion 1993;54:A119.
9 Underdown BJ, Schiff JM: Immunoglobulin A:
Strategic defence initiative at the mucosal sur-
face. Annu Rev Immunol 1986;4:389–417.
10 Andersson R, Yang X, Sun Z, Deng X, Soltesz
V, Ihse I: Effect of a platelet-activating factor
antagonist on pancreatitis-associated gut bar-
rier dysfunction in rats. Pancreas 1998;17:107–
119.
11 Ammori BJ, Leeder PC, King RF, Barclay GR,
Martin IG, Larvin M, McMahon MJ: Early
increase in intestinal permeability in patients
with severe acute pancreatitis: Correlation with
endotoxaemia, organ failure and death. J Gas-
trointest Surg 1998;3:252–262.
12 Carriere F, Barrowman JA, Verger R, Lougier
R: Secretion and contribution to lipolysis of
gastric and pancreatic lipases during a test meal
in humans. Gastroenterology 1993;105:876–
888.
13 Holtmann G, Kelly DG, Sternby B, DiMagno
EP: Survival of human pancreatic enzymes
during small bowel transit: Effect of nutrients,
bile acids and enzymes. Am J Physiol 1997;
273:G553–G558.
14 Anderson IH, Levine AS, Levitt MD: Incom-
plete absorption of the carbohydrate in all-pur-
pose wheat flour. N Engl J Med 1981;304:891–
892.
15 Layer P, Go VL, Dimagno EP: Fate of pan-
creatic enzymes during small intestinal aboral
transit in humans. Am J Physiol 1986;251:
G475–G480.
16 Layer P, Zinsmeister AR, Dimagno EP: Effects
of decreasing intraluminal amylase activity on
starch digestion and postprandial gastrointesti-
nal function in humans. Gastroenterology
1986;91:41–48.
17 Layer P, von der One MR, Holst JI, Jansen MJ:
Altered postprandial motility in chronic pan-
creatitis: Role of malabsorption. Gastroenter-
ology 1997;112:1624–1634.
18 Johnson C, Kudsk K: Nutrition and intestinal
mucosal immunity. Clin Nutr 1999;18:337–
344.
19 Wyncoll D: The management of severe acute
necrotizing pancreatitis: An evidence-based re-
view of the literature. Intensive Care Med
1999;25:146–156.
20 Lobo D, Memon M, Allison S, Rowlands B:
Evolution of nutritional support in acute pan-
creatitis. Br J Surg 2000;87:695–707.
21 Burns GP, Stein TA: Pancreatic enzyme secre-
tion during intravenous fat infusion. J Parenter
Enteral Nutr 1987;11:60–62.
22 Fried GM, Ogden WD, Rhea A, Greeley G,
Thompson JC: Pancreatic protein secretion
and gastrointestinal hormone release in re-
sponse to parenteral amino acids and lipids in
dogs. Surgery 1982;92:902–905.
23 Stabile BE, Borzatta M, Stubbs RS, Debas HT:
Intravenous mixed amino acids and fats do not
stimulate exocrine pancreatic secretion. Am J
Physiol 1984;246:G274–G280.
24 Edelmann K, Valenzuela JE: Effect of intrave-
nous feeding on human pancreatic secretion.
Gastroenterology 1983;85:1063–1068.
25 Niederau C, Sonnenberg A, Erckenbrecht J:
Effects of intravenous infusion of amino acids,
fat, or glucose on unstimulated pancreatic se-
cretion in healthy humans. Dig Dis Sci 1985;
30:445–455.
26 Stabile BE, Debas HT: Intravenous vs. intradu-
odenal amino acids, fats and glucose as stimu-
lants of pancreatic secretion. Surg Forum 1981;
32:224–226.
27 Ragins H, Levenson SM, Singer R, Stamford
W, Seifner E: Intrajejunal administration of an
elemental diet at neutral pH avoids pancreatic
stimulation. Studies in dog and man. Am J
Surg 1973;126:606–614.

218 Dig Dis 2003;21:214–219 Avgerinos/Delis/Rizos/Dervenis

28 Johnson LR, Schanbacker LM, Dudrick SJ,
Copeland EM: Effect of long-term parenteral
feeding on pancreatic secretion and serum se-
cretin. Am J Physiol 1977;233:E524–E529.
29 Grant JP, Davey-McCrae J, Snyder PJ: Effect
of enteral nutrition on human pancreatic secre-
tions. J Parenter Enteral Nutr 1987;11:302–
304.
30 Vidon N, Hecketsveiler P, Butel J: Effect of
continuous jejunal perfusion of elemental and
complex nutritional solutions on pancreatic en-
zyme secretion in human subjects. Gut 1978;
19:194–198.
31 Variyam EP, Fuller RK, Brown FM, Quallich
LG: Effect of parenteral amino acids on human
pancreatic exocrine secretion. Dig Dis Sci
1985;30:445–455.
32 Saluja AK, Steer M: Pathophysiology of pan-
creatitis. Role of cytokines and other mediators
of inflammation. Digestion 1999;60(suppl
1):27–33.
33 Meier R, Beglinger C, Layer P, Gullo L, Keim
V, Laugier R, Friess H, Schweitzer M, Macfie
J, Espen Consensus Group: Espen guidelines
on nutrition in acute pancreatitis. Clin Nutr
2002;21:173–183.
34 Feller JH, Brown RA, Toussaint GPM,
Thompson AG: Changing methods in the treat-
ment of severe pancreatitis. Am J Surg 1974;
127:196–201.
35 Goodgame JT, Fischer JE: Parenteral nutrition
in the treatment of acute pancreatitis: Effects
on complications and mortality. Ann Surg
1977;186:651–658.
36 Grand JP, James S, Grabowski V, Trexler KM:
Total parenteral nutrition in pancreatic dis-
ease. Ann Surg 1984;200:627–631.
37 Robin AP, Campbell R, Palini CK, Liu K,
Donahue PE, Nyhus LM: Total parenteral nu-
trition during acute pancreatitis: Clinical expe-
rience with 156 patients. World J Surg 1990;
14:572–579.
38 Sitzmann JV, Steinborn PA, Zinner MJ, Came-
ron JN: Total parenteral nutrition and alter-
nate energy substrates in treatment of severe
acute pancreatitis. Surg Gynecol Obstet 1989;
168:311–317.
Nutritional Support in Acute Pancreatitis
39 Kalfarentzos FE, Karavias DD, Karatzas TM,
Alevizatos BA, Androulakis LA: Total paren-
teral nutrition in severe acute pancreatitis. J
Am Coll Nutr 1991;10:156–164.
40 Sax AC, Warner BW, Talamini MA, Hamilton
FN, Bell RH Jr, Fischer JE: Early total paren-
teral nutrition in acute pancreatitis: Lack of
beneficial effects. Am J Surg 1987;153:117–
124.
41 Kotani J, Usami M, Nomura H: Enteral nutri-
tion prevents bacterial translocation but does
not improve survival in acute pancreatitis.
Arch Surg 1999;134:287–292.
42 Moore F, Moore E, Kudsk K et al: Clinical ben-
efits of an immune enhancing diets for early
post-injury feeding. J Trauma 1994;37:607–
615.
43 Voitk A, Brown RA, Echave V, McArdle AH,
Gurd FN, Thompson AG: Use of an elemental
diet in the treatment of complicated pancreati-
tis. Am J Surg 1973;125:223–227.
44 Kudsk KA, Campbell SM, O’Brien T, Fuller R:
Postoperative jejunal feeding following compli-
cated pancreatitis. Nutr Clin Pract 1990;5:14–
17.
45 Parekh D, Lawson HH, Segal I: The role of
total enteral nutrition in pancreatic disease. S
Afr J Surg 1993;31:57–61.
46 Simpson WG, Marsano L, Gates L: Enteral
nutritional support in acute alcoholic pancre-
atitis. J Am Coll Nutr 1995;14:662–665.
47 Nakad A, Piessevaux H, Marot JC, Hoang P,
Geubel A, Van Steenberger W: Is early enteral
nutrition in acute pancreatitis dangerous?
About 20 patients fed by endoscopically placed
nasogastrojejunal tube. Pancreas 1998;17:187–
193.
48 Eatock F, Brombacher G, Steven A, Imrie C,
McKay C, Carter R: Nasogastric feeding in
severe acute pancreatitis may be practical and
safe. Int J Pancreatol 2000;28:25–31.
49 Powell JJ, Murchison JT, Fearon KC, Ross JA,
Siriwardena AK: Randomized controlled trial
of the effect of early enteral nutrition on mark-
ers of the inflammatory response in predicted
severe acute pancreatitis. Br J Surg 2000;87:
1357–1381.
Dig Dis 2003;21:214–219
50 Kalfarentzos F, Kehagias J, Mead N, Kokkinis
K, Gogos C: Enteral nutrition is superior to
parenteral nutrition in severe acute pancreati-
tis: Results of a randomised prospective trial.
Br J Surg 1997;83:349–353.
51 McClave SA, Greene LM, Snider HL, Makk
LJ, Cheadle WG, Owens NA et al: Comparison
of the safety of early enteral vs. parenteral
nutrition in mild acute pancreatitis. J Parenter
Enteral Nutr 1997;21:14–20.
52 Abou-Assi S, O’Keefe SJD: Nutrition support
during acute pancreatitis. Nutrition 2002;18:
938–943.
53 Olah A, Pardavi G, Belagyi T, Nagy A, Issekutz
A, Mohamed GE: Early nasojejunal feeding in
acute pancreatitis is associated with a lower
complication rate. Nutrition 2002;18:259–
262.
54 Powell JJ, Murchison JT, Fearon KCH, Ross
JA, Siriwardena AK: Randomized controlled
trial of the effect of early enteral nutrition on
markers of the inflammatory response in pre-
dicted severe acute pancreatitis. Br J Surg
2000;87:1375–1381.
55 Paraskeva C, Smailis D, Priovolos A, Sofianou
K, Lytras D, Avgerinos C, Rizos S, Karagiannis
J, Dervenis C: Early enteral nutrition reduces
the need for surgery in severe acute pancreati-
tis. Pancreatology 2001;1:372.
56 Eckerwall G, Andersson R: Early enteral nutri-
tion in severe acute pancreatitis: A way of pro-
viding nutrients, gut barrier protection, immu-
nomodulation or all of them? Scand J Gas-
troenterol 2001;5:449–458.
57 Heys S, Walker L, Smith I, Eremin O: Enteral
nutritional supplementation with key nutrients
in patients with critical illness and cancer. Ann
Surg 1999;229:467–477.
58 Olah A, Belagyi T, Issekutz A, Gamal ME,
Bengmark S: Randomized clinical trial of spe-
cific lactobacillus and fibre supplement to early
enteral nutrition in patients with acute pancre-
atitis. Br J Surg 2002;89:1103–1107.
219
 Review Article

Dig Dis 2003;21:220–227

DOI: 10.1159/000073339

Nutrition and Inflammatory Bowel

Disease: Its Relation to
Pathophysiology, Outcome and Therapy
Miquel A. Gassull
Department of Gastroenterology and Hepatology, Hospital Universitari Germans Trias i Pujol, Badalona,
Catalonia, Spain

Key Words Several facts relate nutrition and inflammatory bowel

Ulcerative colitis W Crohn’s disease W Inflammatory bowel
disease W Malnutrition W Enteral nutrition W Dietary fat W
Unabsorbable dietary carbohydrates W Colonic
fermentation
Abstract
Nutritional deficiencies are frequent in patients with
ulcerative colitis and Crohn’s disease, and negatively
influence the outcome of the disease. Growth retarda-
tion, osteopenia and thromboembolic phenomena are
some of the inflammatory bowel disease complications
in which nutritional deficits are involved. Moreover, nu-
trients can play a role in the pathogenesis of the disease
and, in some cases, can be a primary therapeutic tool.
Enteral nutrition has proven to play a therapeutic role in
Crohn’s disease. The nutrient(s) responsible for this ef-
fect are not well identified but dietary fat appears to be a
major factor. In ulcerative colitis, unabsorbable carbohy-
drates can modulate the intestinal microbial environ-
ment, thus contributing to improve colonic inflamma-
tion.
Copyright © 2003 S. Karger AG, Basel
disease (IBD) and many features even support the exis-
tence of a bidirectional relationship between them. In this
sense, a high prevalence of nutritional derangement
(macro- and micronutrients), both in ulcerative colitis
(UC) and Crohn’s disease (CD), has been reported [1–5].
In addition, qualitative differences in diet composition
have been suggested as being the cause of the low inci-
dence of IBD and other immune-based diseases in spe-
cific populations [6]. Finally, it has been shown that
changes in some components of the diet are able to down-
regulate the inflammatory response in vitro, in animals
and in humans [7–14].
In this paper the above-mentioned aspects will be
reviewed, with special emphasis on data relating to nutri-
tional deficiencies, IBD pathophysiology, clinical out-
come and the changes of incidence in some geographical
areas. Also the evidence about the role of nutritional
manipulation in inducing and maintaining clinical dis-
ease remission will be discussed.
Protein Energy Malnutrition and Micronutrient
Deficit in IBD: Their Effect on Disease Outcome

Various degrees of protein energy malnutrition, vita-

min, mineral and trace element deficiencies may develop
in UC and CD [1–5]. Anorexia has been blamed for many

© 2003 S. Karger AG, Basel Miquel A. Gassull MD, PhD, Assoc. Prof. Med.
ABC 0257–2753/03/0213–0220$19.50/0 Head, Department of Gastroenterology and Hepatology
Fax + 41 61 306 12 34 Hospital Universitari Germans Trias i Pujol, Carretera del Canyet s/n
E-Mail karger@karger.ch Accessible online at: ES–08916 Badalona, Catalonia (Spain)
www.karger.com www.karger.com/ddi Tel./Fax +34 93 4978951, E-Mail mgassull@ns.hugtip.scs.es
years as being one of the main causes of IBD-associated
malnutrition, growth failure and sexual development re-
tardation in children and adolescents. The cause of these
derangements is multifactorial and related to the exis-
tence of active inflammation. In animals with experimen-
tal colitis, this clinical situation has been reproduced and
found to be especially related to the release of IL-1 by the
inflamed mucosa [13–15].
Other causes of malnutrition in IBD include a hyper-
metabolic state, due to inflammation, extensive intestinal
involvement or surgical resection of the small intestine,
interfering with nutrient absorption, or an increase of
nutrient losses through the inflamed and ulcerated gut
(small and large bowel). In addition, the existence of fistu-
las and/or strictures in the small intestine facilitates bacte-
rial overgrowth. The excess of bacteria results in an
increased production of bacterial metabolites of nutrients
through fermentation and hydroxylation processes. Some
of the products such as folate may have a beneficial effect,
however an excess of short-chain fatty acids and especial-
ly fatty acid hydroxylation results in increased diarrhea
and hence nutrient wasting. Bacterial overgrowth also
induces vitamin B12 deficiency. Finally, medical treat-
ment (therapeutic fasting, steroids, cholestyramine) [16,
17] and self- or medically-recommended dietary restric-
tions may also contribute the development of nutritional
deficiencies in IBD patients [10, 17].
Nutritional derangement may influence adult IBD out-
come, since it may increase the need for surgical treat-
ment [1] and its complications [17]. However, it has been
in children and adolescents where malnutrition, in its
wider sense, together with steroid treatment (by IGF-1
suppression) have been blamed as inducing one of the
most severe IBD complications in this population – linear
growth and sexual maturation failure [18–20]. Experi-
ments in rats with colitis have shown the important
impact of the excessive release of proinflammatory cyto-
kines by the inflamed mucosa on growth retardation [21].
However, nutritional deterioration can also be blamed for
causing more than 50% of the delayed growth observed in
these experimental animals, because it is associated with a
low IGF-1 synthesis. Feeding properly colitic rats partial-
ly reversed the linear growth deficit, but it never reached
that of the control animals [21]. These studies strongly
support the fact that decreasing the inflammatory process
and feeding patients may adequately improve linear
growth in the infant and adolescent IBD population.
These objectives could be reached by using concomitantly
immunomodulatory therapy and nutritional support.
There is also the possibility that especially designed nutri-
Nutrition and Inflammatory Bowel Disease
tional formulas may achieve both goals; this aspect will be
dealt with later in this article.
Suboptimal levels of micronutrients are often found
both in children and adults with IBD, but except for iron
and folate, it is unusual to discover symptoms attributable
to these deficits [3–5]. In spite of this, subclinical deficits
may have a pathophysiological significance since they
may favor self-perpetuation of the disease (because of
defects in the mechanisms of tissue repair), defective
defense against damage produced by oxygen free radical
and can facilitate lipid peroxidation [22, 23], even in clin-
ically inactive or mildly active disease [24], as well as the
development of dysplasia in the intestinal mucosa [25].
However, the potential therapeutic role of antioxidant
micronutrients, both in inducing disease remission and in
the prevention of disease relapse, has scarcely been inves-
tigated. Recently, it was shown that the administration of
vitamins E and C for 4 weeks in mild to moderate cases of
CD induced a significant reduction of oxidative stress as
measured by breath pentane and ethane output, plasma
lipid peroxides and F-2 isoprostane, suggesting increased
requirements of such micronutrients [26].
One of the most risky situations in patients with acute
IBD is the appearance of thromboembolic complications,
related in part to the existence of nutritional imbalances,
resulting in excessive plasma homocysteine. Increased
homocysteine levels have proved to be thrombogenic and
an independent risk factor for cardiovascular disease [27,
28] and deep-vein thrombosis [29]. This sulfur-containing
amino acid is formed during the demethylation of methi-
onine. Hyperhomocysteinemia has been reported to be
prevalent in patients with IBD [28, 29] and related to both
low folate [30] and vitamin B12 [31] levels. However, only
a small amount of homocysteine is cleared via a remethy-
lation pathway dependent on vitamin B12 and folate, but
most of the homocysteine is converted to cystothione by
cystothione ß-synthase which is a vitamin B6-dependent
enzyme. Although low vitamin B6 levels were reported in
IBD patients many years ago [3], it is only recently that
they have been related to the high homocysteine levels in
IBD and the risk of thrombosis [32].
Osteopenia associated with IBD is an area of increas-
ing concern, especially in CD [33–37]. Nutritional defi-
ciencies, inflammatory cytokines and treatment with cor-
ticosteroids have been incriminated in its pathogenesis
[38, 39]. The disease itself seems to exert a direct effect in
inducing osteopenia in these patients. This was shown in
an in vitro experiment in which the serum of patients with
CD impaired osteoblastic function [40]. In 36 new cases
of pediatric CD, a dual X-ray absorptiometry study dem-
Dig Dis 2003;21:220–227 221
onstrated osteopenia and osteoporosis in 17 and 25%
respectively [41]. A bone fracture rate as high as 40% has
been reported in a population-based cohort study [42]. It
is noteworthy that low bone mineral density has been
described in adult CD patient populations as distant as
Puerto Rico, The Netherlands and Korea [43–45]. In
some cases, this abnormality is detected in newly diag-
nosed cases in both adults [45] and children [46]. Support-
ing the fact that inflammation is an important factor in
inducing osteopenia in IBD is that TNF-·, IL-1ß and IL-
11 (proinflammatory cytokines) promote the secretion of
a ligand able to stimulate receptors in osteoclast precur-
sors, inducing osteoclast maturation and bone resorption
[47].
Glucocorticoids, the gold standard treatment in IBD,
are associated with side effects including permanent dam-
age such as cataracts, growth failure, sexual development
retardation, and osteopenia. In fact, glucocorticoids are
an important factor aggravating bone loss in IBD, because
they decrease intestinal calcium absorption, increase re-
nal calcium excretion and induce parathyroid hormone
secretion. The result is decreased bone formation and
enhanced bone resorption [46, 47]. These data should be
enough to restrict their use in the treatment of IBD, both
as long-term therapy (even at low doses) and in patients
with frequent relapses (steroid-dependent) [48–51]. Bone
loss induced by steroid therapy is dose- and duration-
dependent. In addition, growth and sexual development
retardation in children and adolescents may be aggravat-
ed by glucocorticoid treatment due to its suppressive
effect on IGF-1, which mediates the effect of growth hor-
mone on the growth of plate bones [21]. Then it is highly
recommendable in this population to use steroids only
when strictly necessary, in short treatments, avoid re-
peated treatments (use other immunosuppressors) and
especially try nutrition as primary treatment.
Possible Role of Nutrition in the Incidence of
IBD in Some Geographical Areas
The low incidence of IBD and other immune-based
diseases in some communities like Greenland Eskimos, as
compared to western populations, has been attributed to
differences in lifestyle and the type of diet [52, 53]. It has
been hypothesized that components of the Eskimo’s diet
(abundant in n-3 fatty acid-rich marine oils) produce
changes in cell membrane composition (including that of
the immune-competent cells) [53]. This would result in a
downregulated expression of certain cytokines, adhesion
222 Dig Dis 2003;21:220–227
molecules, lymphocyte chemotactic activity and an in-
creased synthesis of eicosapentaenoic (n-3) acid-derived
eicosanoids with attenuated proinflammatory activity [8,
9, 54–57]. The hypothesis is supported both by in vitro
and in vivo studies [53–59]. Moreover, increases in the
annual incidence in both UC and CD have been observed
in countries where qualitative changes in dietary fat were
established because of public health reasons [60–62] or
when hit by western dietary influences as in Japan [63].
The major dietary changes in these countries were the
substitution of butter by margarine in Sweden [62] and
the decrease in intake of fish while increasing that of meat
and seed oils in Japan [63]. These changes, as mentioned,
were followed by increases in the incidence of IBD in
these countries.
Is Diet a Factor in the Etiopathogenesis of IBD?
Does It Explain the Therapeutic Effect of
Enteral Nutrition?
Nutrients in the form of chemically and physicochemi-
cally complex food products are one of the main compo-
nents of the intestinal environment. As such, the possibili-
ty that the food antigenic load may be one of the triggering
factors initiating and perpetuating the abnormal intesti-
nal inflammatory response has been frequently invocated.
This has been the basis for using therapeutic fasting in
acute UC and CD (bowel rest), amino acid-base elemental
diets (low antigenic power) for treating acute CD and
excluding the suspected offending food (exclusion diets)
in the treatment of active or maintenance of remission in
CD.
Various studies have shown that withholding nutrients
from the intestinal lumen does not influence disease out-
come [64–66], even in acute disease. Elemental diets have
shown to offer no advantage as compared to whole pro-
tein-based polymeric diets (higher antigenic load) in in-
ducing remission in active CD, at least in three meta-anal-
yses summarizing the findings of several trials [67–69].
The effect of exclusion diets in inducing or maintaining
remission is still doubtful, since the reintroduction of the
offensive food frequently is not accompanied by disease
relapse [70]. Recently, experimental work in vitro (human
peripheral blood lymphocytes) and humans (skin and rec-
tal mucosa) has tried to ascertain whether or not intestinal
epithelium and immune cells have an increased sensitivi-
ty to different components of the diet. Although peripher-
al blood lymphocytes and rectal mucosa are not, under
normal circumstances, exposed to intact food, the results
Gassull
showed increased sensitivity in CD compared to healthy
individuals [71, 72]. The true value in etiopathogenic and
therapeutic terms of these findings needs to be further
explored.
Several years ago, Gonzalez-Huix et al. [73] first sug-
gested that fat composition of the diet, rather than the
nitrogen source, may explain why enteral nutrition may
have an anti-inflammatory effect in active CD. This was
based on the fact that in a randomized clinical trial a poly-
meric enteral diet (higher antigenic load than elemental)
containing 30% of the calories as lipids (65% olive oil,
10% milk fat and 25% MCT) was as effective as 1 mg/kg
b.w. of prednisolone. These authors hypothesized that the
key factor for the therapeutic effectiveness of enteral
nutrition in CD was the amount (elemental low fat) or the
type of fat used in the design of the formula diet. This
hypothesis has been reinforced in a recent paper [74]. In
the literature a great deal of data supports the immuno-
modulatory role of dietary fat. The effect that fatty acids
have from different sources in modifying cell membrane
composition and hence the influence on the function of
the proteins embedded in it, has already been mentioned.
The consequences of this are relevant in terms not only on
the type of eicosanoids and other second messengers syn-
thesized [9], but also in the cell signaling system, influenc-
ing the activity of nuclear transcription factors and hence
the genes expressing proteins involved in the inflammato-
ry process (cytokines, adhesion molecules, iNOS, etc.)
[75–83]. It is worth emphasizing that usually there is a
tendency, especially by clinical nutritionists, in using the
words fat and fatty acids synonymously. Mammals do not
eat single fatty acids in the diet but different fat sources
with mixtures of fatty acids. The predominance of one or
some of them give to the fat source not only its physico-
chemical properties, but also its nutritional and metabolic
value. In this sense it is worth mentioning that different
fat sources with different predominant fatty acids have
different effects on the expression of different cytokines in
an elegant experimental setting in mice [84]. It may not be
the effect of a single fatty acid but the combination of sev-
eral that exerts the immunomodulatory action. An exam-
ple of this possibility is fish oil, which has been made syn-
onymous of a fat source rich in n-3 eicosapentanoic (EPA)
and docosahexaenoic acid (DHA). In fact, fish oil con-
tains no more than 25% of EPA and DHA together, the
larger component being saturated fatty acids (LCT +
MCT) 45%, with 16% of monounsaturated and 10% of
n-6 fatty acids [84]. It is then plausible that the anti-
inflammatory effect would be related more to an adequate
mixture of fatty acids (fat source) than to a single
Nutrition and Inflammatory Bowel Disease
one. In this sense, a recent paper showed that the use in an
enteral formula diet of an artificially made mixture with a
highly predominant (70%) synthetic fatty acid, did not
obtain any therapeutic effect as compared to one using a
natural source [74]. This fact also raises the question of
the role of the non-fatty acid component of the lipid
sources, since many compounds are found, at least in
some oil sources that may have, through their antioxidant
effect, immune regulatory properties [85, 86]. This possi-
bility has to be further explored.
An interesting but seldom mentioned aspect is the
potential anti-inflammatory role of MCT. These com-
pounds have always been regarded as energy providers in
malabsorptive states or easy sources of energy for mal-
nourished individuals. However, it has been shown exper-
imentally [84, 87–89] that MCT modulates the expression
of adhesion molecules and cytokines and the metabolism
of linoleic acid (precursor of potent proinflammatory
eicosanoids). Also, a polymeric enteral diet containing
high amounts of MCT is as effective as a very low fat ele-
mental diet in inducing remission in active CD [90]. In
fact, when the composition of the diets reported as highly
effective in inducing remission in active CD are carefully
analyzed, it can be observed that those with a higher
remission rate are those with very low fat content and
those where one third of the fat source was MCT [73,
91].
A further aspect related to dietary fat is its influence in
the cell cycle. A feature in CD is the resistance of T lym-
phocytes to the natural apoptotic process, leading to its
accumulation in the bowel wall and hence contributing to
perpetuate the inflammatory process [92]. Some lipid
sources (fish oil or olive oil, but not corn oil) induce apop-
tosis in intestinal cell lines cultures by, at least, inhibiting
the expression of the antiapoptotic mitochondrial protein
Bcl-2 [93]. This may be a further mechanism explaining
why some fat sources show a positive therapeutic effect in
active CD while others do not.
To summarize this part, it can be said that although T
lymphocytes and rectal mucosa have shown sensitization
to some food products, the therapeutic effect of exclusion
diets is still not well established and there is no difference
in the rate of remission achieved by whole protein or ami-
no acid formula diets when used as primary treatment in
CD.
Nutrients as components of cell structures: Changes in
dietary components can influence numerous cell func-
tions and the cell cycle. Fat is, up to now, the best studied
nutrient and has been shown to influence immune cell
response, although many other aspects of these com-
Dig Dis 2003;21:220–227 223
pounds need to be explored. So far, changes in the compo-
sition of dietary fat have been shown to be able to
influence disease outcome in CD and some data relate
major qualitative changes in fat intake in geographically
distant populations (Sweden and Japan) with an increase
in the incidence of CD, suggesting a possible role as trigger
in genetically susceptible individuals. Little is known
about the potential effect of changes in dietary fat on
intestinal bacterial populations.
Nutrition and Ulcerative Colitis
Up to now in this review, the relationship between
nutrition and UC has seldom been mentioned, except for
its capacity of inducing nutritional derangement. When
dealing with the role of enteral nutrition, the effect of dif-
ferent dietary fat composition in primary therapy, its
influence in disease etiopathogenesis and incidence,
mostly CD has been mentioned. In fact the enteral formu-
la diets used as primary treatment in CD have proven not
to be effective in UC, although improve patients’ nutri-
tional status and prevent complications related to surgery,
when compared to parenteral nutrition [94]. One of the
different possible explanations is that while CD is an
antiapoptotic disease, UC is not [92]. Moreover, there are
major differences in the etiopathogenesis of both diseases,
as far as the environmental factors are concerned. The
most obvious is the effect of smoking, being protective in
UC and inductor in CD. More is known about the rela-
tionship between the intestinal environment and the co-
lonic mucosa. Finally, the concentration of bacteria in the
colon is much higher than that in the small intestine and,
even though endogenous bacteria play a crucial role in the
pathogenesis of CD, little is known about the relationship
between small intestinal bacteria and nutrients present in
the intestinal lumen in such a condition.
The colon is highly dependent of the environment in
its lumen to maintain its structure and function. Almost
80% of the energy required for such a task is obtained
from nutrients reaching the large bowel. Several sub-
strates enter the colon every day, these include non-
absorbed carbohydrates, proteins and a few others. In
short, anaerobic bacteria deal with these substrates
through a fermentation process producing branched-
chain fatty acids from proteins and short-chain fatty acids
plus a large volume gas (H2 and CO2) from carbohydrates.
The gas volume is reduced by the action of methanogenic
and sulfate-reducing bacteria producing CH4 and SH2
respectively. Short-chain fatty acids (acetate, propionate
224 Dig Dis 2003;21:220–227
and butyrate) are involved in the maintenance of colonic
structure (butyrate is the preferred fuel for the colonocyte)
and immune homeostasis. The first consequence of short-
chain fatty acid synthesis is the decrease of intraluminal
pH in the colon [95]. This fact favors the development of
Lactobacilli, Bifidobacteria strains (probiotics) and a de-
crease of those of Clostridia, Bacteroides, Escherichia coli.
This is known as the ‘prebiotic effect’ effect of fiber.
As mentioned, short-chain fatty acid, especially buty-
rate, is oxidized by the colonic mucosa and used mainly as
an energy provider. In recent years it has been postulated
that butyrate oxidation is incomplete or defective in both
UC and pouchitis, suggesting that this defect may be
implicated in the pathogenesis of both conditions [96–
103]. Rectally administered 13C-labelled butyrate was sig-
nificantly less oxidized by active UC patients than by
healthy controls, whereas patients with inactive disease
showed a trend towards a lower oxidation than healthy
subjects [104]. These findings in non-active UC are sup-
ported by other studies [105].
Recent work suggests that in active UC there is an
excess of SH2 synthesis by sulfate-reducing bacteria in the
colon [103]. This excessive SH2 contributes to the ineffi-
cient butyrate oxidation in the colonic mucosa [100]. In
patients with proctitis refractory to treatment with topical
steroids and 5-aminosalicylic acid (5-ASA), the nutrition-
al approach, by intrarectally administering butyrate as
maintenance treatment for such a type of UC, has pro-
duced significantly positive results as compared to place-
bo [99, 106]. Moreover, the relapse rate of quiescent UC
treated with soluble dietary fiber (Plantago ovata seeds),
substrate precursor for the production of butyrate in the
colon (prebiotic), was not different than that of patients
given 5-ASA at the usual maintenance doses [107], al-
though there was a non-significant trend towards a lower
relapse rate in those patients on combined therapy. Inter-
estingly enough, patients on treatment with this fiber
preparation increased butyrate production in the colon
(as measured by stool analysis) as compared to those on
5-ASA alone [107]. This nutritional approach in the main-
tenance therapy of UC supports the role of butyrate in the
pathogenesis of the UC.
However, butyrate is not only an energy provider, but a
true anti-inflammatory substance, since it prevents the
activation of the nuclear transcription factor NFÎB and
through this mechanism prevents the expression of cyto-
kines and other molecules involved in the inflammatory
process [108–110]. This is achieved by the inhibition of
the phosphorylation or degradation of the inhibitory pro-
tein IÎB that holds NFÎB in the cell cytoplasm. Butyrate
Gassull
would be a dietary factor regulating the inflammatory
process from the intestinal lumen. Its therapeutic possibil-
ities as primary therapy in active UC are currently under
investigation. Its anti-inflammatory properties, together
with its prebiotic action, strongly suggest the possibility of
using a combined prebiotic-probiotic approach (symbiot-
ic) in the treatment of UC and pouchitis.
The UC nutritional approach both in terms of patho-
genesis and therapy seems to follow a different route than
that of CD, with a more clear interaction in the intestinal
ecosystem in the former. Thus a nutritional therapeutic
approach may also be possible in UC. The interest in the
knowledge of the possible role of nutritional imbalances
in IBD relies on its potential usefulness of nutritional
manipulation as a preventive measure in genetically sus-
ceptible individuals and as its use as a true biological ther-
apy, free of side effects, in certain groups or subgroups of
patients with IBD.

References

1 Gassull MA, Abad A, Cabré E, González-Huix
F, Giné JJ, Dolz C: Enteral tube feeding in
inflammatory bowel disease. Gut 1986;
27(suppl):76–80.
2 Cabré E, Fernández-Bañares F, Esteve M, Gas-
sull MA: Micronutrients in inflammatory bow-
el disease. J Clin Nutr Gastroenterol 1989;4:
100–102.
3 Fernández-Bañares F, Abad A, Xiol X, Giné
JJ, Dolz C, Cabré E, Esteve M, González-Huix
F, Gassull MA: Vitamin status in patients with
inflammatory bowel disease. Am J Gastroen-
terol 1989;84:744–748.
4 Fernández-Bañares F, Mingorance MD, Esteve
M, Cabré E, Lachica M, Abad A, Gil A, Hum-
bert P, Boix J, Gassull MA: Serum zinc, copper
and selenium levels in inflammatory bowel dis-
ease. Am J Gastroenterol 1990;85:1584–1589.
5 Abad A, Fernández-Bañares F, Cabré E, Gil A,
Esteve M, González-Huix F, Xiol X, Gassull
MA: The effect of total enteral tube feeding on
the vitamin status of malnourished patients
with inflammatory bowel disease. Int J Vitam
Nutr Res 1988;58:428–435.
6 Kromann N, Green A: Epidemiological studies
in the Upernavik district, Greenland. Acta
Med Scand 1980;208:401–406.
7 O’Morain C, Segal AW, Levi AJ: Elemental
diets as primary treatment of acute Crohn’s
disease. Br Med J 1984;228:1859–1862.
8 De Caterina, R, Cybulsky MI, Clinton SK,
Gimbrone M, Libby P: The ˆ-3 fatty acid
docosahexanoate reduces cytokine-induced ex-
pression of proatherogenic and proinflamma-
tory proteins in human endothelial cells. Arter-
ioscler Thromb 1994;14:1829–1836.
9 Lee T, Hoover R, Williams J, Sperling R, Rav-
alese J III, Spur B, Robinson D, Corey E, Lewis
R, Austen K: Effect of dietary enrichment with
eicosapentaenoic and docosahexaenoic acids
on in vivo neutrophil and monocyte leuco-
triene generation and neutrophil function. N
Engl J Med 1985;312:1217–1224.
10 Gassull MA, Fernández Bañares F, Cabré E,
Esteve M: Enteral nutrition in inflammatory
bowel disease; in Rombeau JL, Rolandelli RH
(eds): Clinical Nutrition. Enteral and Tube
Feeding, ed 3. Philadelphia, Saunders, 1997,
pp 403–416.
11 Morris GP, Beck PL, Herridge MS, Depew
WT, Szewczuk MR, Wallace JL: Hapten-in-
duced model of chronic inflammation and ul-
ceration of the rat colon. Gastroenterology
1989;96:795–803.
12 McHugh KJ, Weingarten HP, Keenan C, Wal-
lace J, Collins SM: On the suppression of food
intake in experimental models of colitis in the
rat. Am J Physiol 1993;264:R871–R876.
13 McHugh K, Castonguay TW, Collins SM,
Weingarten HP: Characterization of suppres-
sion of food intake following acute colonic
inflammation in the rat. Am J Physiol 1993;
265:R1001–R1005.
14 Otterness IG, Seymour PA, Golden HW, Rey-
nolds JA, Daumy GO: The effects of contin-
uous administration of interleukin-1 in the rat.
Physiol Behav 1989;43:797–804.
15 Ballinger A, El-Haj T, Perrett D, Turvill J,
Obeid O, Dryden S, Williams G, Farthing
MJG: The role of medial hypothalamic sero-
tonin in the suppression of feeding in a rat
model of colitis. Gastroenterology 2000;118:
544–553.
16 Gassull MA, Cabré E, Vilar L, Montserrat A:
Nivel de ingesta alimentaria y su posible papel
en el desarrollo de malnutrición calórico-pro-
teica en pacientes gastroenterológicos hospital-
izados. Med Clin (Barc) 1985;85:85–87.
17 Gassull MA, Fernández Bañares F: Nutrition
in inflammatory bowel disease; in Payne-
James J, Grimble G, Silk DBA (eds): Artificial
Nutrition Support in Clinical Practice, ed 2.
London, Greenwich Medical Media, 2001, pp
555–573.
18 González-Huix F, Fernández Bañares F, Es-
teve M, Abad A, Cabré E, Acero D, Figa M,
Guilera M, Humbert P, De León R, Gassull
MA: Enteral versus parenteral nutrition as ad-
junct therapy in acute ulcerative colitis. Am J
Gastroenterol 1993;88:227–232.
19 Winter HS: Special consideration in pediatric
inflammatory bowel disease; in Targan SR,
Shanahan F (eds): Inflammatory Bowel Dis-
ease. From Bench to Bedside. Baltimore, Wil-
liams & Wilkins, 1994, pp 701–710.
20 Griffiths AM, Nguyen P, Smith C, MacMillan
H, Sherman PM: Growth and clinical course of
children with Crohn’s disease. Gut 1993;34:
939–943.
21 Ballinger A, Azooz O, El-Haj Y, Poole S, Farth-
ing MJG: Growth failure occurs through a de-
crease in insulin-like growth factor, which is
independent of undernutrition in the rat model
of colitis. Gut 2000;46:1–6.
22 Miralles-Barrachina O, Savoye G, Belmonte-
Zalar L, Hochain P, Ducrotté P, Hecketsweiler
B, Lerebours E, Déchelotte P: Low levels of glu-
tathione in endoscopic biopsies of patients with
Crohn’s colitis: The role of malnutrition. Clin
Nutr 1999;18:313–317.
23 Fernández-Bañares F: Papel de los metabolitos
de oxı́geno reactivos en la patogenia de la enfer-
medad inflamatoria intestinal. Gastroenterol
Hepatol 1995;18:526–532.
24 Wendland B, Aghdassi E, Tam C et al: Lipid
peroxidation and plasma antioxidant micronu-
trients in Crohn’s disease. Am J Clin Nutr
2001;74:259–264.
25 Lashner BA, Provencher KS, Seidner DL, Kne-
sebeck A, Brzezinski A: The effect of folic acid
supplementation on the risk for cancer or dys-
plasia in ulcerative colitis. Gastroenterology
1997;112:29–32.
26 Aghdassi E, Wendland E, Steinhard H et al:
Antioxidant vitamin supplementation in
Crohn’s disease decreases oxidative stress: A
randomized controlled trial. Am J Gastroen-
terol 2003;98:348–353.
27 Graham IM, Daly LE, Refsum HM et al: Plas-
ma homocysteine as a risk factor for vascular
disease. The European Concerted Action Pro-
ject. JAMA 1997;277:1775–1781.
28 Nygard O, Nordrehaug JE, Refsum H et al:
Plasma homocysteine levels and mortality in
patients with coronary artery disease. N Engl J
Med 1997;337:230–236.
29 Den Heijer M, Koster T, Blom HJ et al: Hyper-
homocysteinemia as a risk factor for deep-vein
thrombosis. N Engl J Med 1996;334:759–762.
30 Chowers Y, Sela BA, Holland R et al: Increased
levels of homocysteine in patients with Crohn’s
disease are related to folate levels. Am J Gas-
troenterol 2000;95:3498–3502.
31 Romagnuolo J, Fedorak RN, Dias VC: Hyper-
homocysteinemia and inflammatory bowel dis-
ease: Prevalence and predictors in a cross-sec-
tional study. Am J Gastroenterol 2001;96:
2143–2149.

Nutrition and Inflammatory Bowel Disease Dig Dis 2003;21:220–227 225

32 Salveni S, Cattaneo M, Vecchi M et al: Low
vitamin B6 plasma levels, a risk factor for
thrombosis, in inflammatory bowel disease:
Role of inflammation and correlation with
acute phase reactants. Am J Gastroenterol
2003:98:112–117.
33 Silvennoinen JA, Karttunen TJ, Niemelae SE,
Manelius JJ, Lehtola JK: A controlled study of
bone mineral density in patients with inflam-
matory bowel disease. Gut 1995;37:71–76.
34 Bjarnason I, MacPherson A, Mackintosh C,
Buxton-Thomas M, Forgacs I, Moniz C: Re-
duced bone density in patients with inflamma-
tory bowel disease. Gut 1997;40:228–233.
35 Jahnsen J, Falch JA, Aadland E, Mowinckel P:
Bone mineral density is reduced in patients
with Crohn’s disease but not in patients with
ulcerative colitis: A population-based study.
Gut 1997;40:313–319.
36 Martin JP, Tonge KA, Bohnsle U, Jacyna MR,
Levi J: Bone mineral density in patients with
inflammatory bowel disease. Eur J Gastroen-
terol Hepatol 1999;11:537–541.
37 Cowan FJ, Warner JT, Dunstan FD, Evans
WD, Gregory JW, Jenkins HR: Inflammatory
bowel disease and predisposition to osteopenia
Arch Dis Child 1997;76:325–329.
38 Semeao E, Jawad AF, Stoufer NO, Babette SZ,
Piccoli DA, Stalling VA: Risk factors for low
bone mineral density in children and young
adults with Crohn’s disease. J Pediatr 1999;
135:593–600.
39 Issenman RM: Bone mineral metabolism in
pediatric inflammatory bowel disease. In-
flamm Bowel Dis 1999;5:192–199.
40 Hymans JS, Syzga N, Kreutzer DL, Justinich
CJ, Gronowicz GA: Alterations in bone metab-
olism in children with inflammatory bowel dis-
ease: An in vitro study. J Pediatr Gastroenterol
Nutr 1996;24:289–295.
41 Sylvester FA, Hymans JS, Justinich C, Zeiter
D, Davis P, Harris K, Hawker G, Griffiths
AM: Osteopenia is frequent in children with
Crohn’s disease at diagnosis (abstract). J Pe-
diatr Gastroenterol Nutr 1999;29:A524.
42 Bernstein CN, Blanchard JF, Leslie W, Wadja
A, Yu BN: The incidence of fracture among
patients with inflammatory bowel disease. A
population-based cohort study. Ann Intern
Med 2000;133:795–799.
43 Chinea B, Rosa A, Ohariz JJ, Ramirez M, Had-
dick L, Perez C, Torres EA: Osteopenia in
Puerto Ricans with Crohn’s disease. P R
Health Sci J 2000;19:329–333.
44 Scoon EJ, van Nunnen AB, Wouters RS, Stock-
brugger RW, Russel MG: Osteopenia and os-
teoporosis in Crohn’s disease: Prevalence in a
Dutch population-based cohort. Scand J Gas-
troenterol Suppl 2000;232:43–47.
45 Lee SH, Kim HJ, Yang SK, Kim WH, Joo YS,
Dong SH, Kim BH, Lee JI, Chang YW, Chang
R: Decreased trabecular bone mineral density
in newly diagnosed inflammatory bowel dis-
ease patients in Korea. J Gastroenterol Hepatol
2000;15:512–518.
226 Dig Dis 2003;21:220–227
46 Thearle M, Horlick M, Bilezikian JP, Levy J,
Gertner JM, Levine LS, Harbison M, Berdon
W, Oberfield SE: Osteoporosis: An unusual
presentation of childhood Crohn’s disease. J
Clin Endocrinol Metab 2000;85:2122–2126.
47 Roux S, Orcel P: Bone loss: Factors that regu-
late osteoclast differentiation. An update. Ar-
thritis Res 2000;2:451–456.
48 Canalis E: Mechanisms of glucocorticoid ac-
tion in bone: Implications to corticosteroid-
induced osteoporosis. J Clin Endocrinol Metab
1996;81:3441–3447.
49 Hodsman AB, Toogood JH, Jennings B, Fraher
LJ, Baskerville JC: Differential effect of in-
haled budesonide and oral prednisone in serum
osteocalcin. J Clin Endocrinol Metab 1991;72:
530–540.
50 Schoon EJ, Wolffenbuttel BHR, Stockbrügger
RW: Osteoporosis as a risk in inflammatory
bowel disease. Drugs Today 1999;35(suppl
A):17–28.
51 Adinoff AD, Hollister R: Steroid-induced frac-
tures and bone loss in patients with asthma. N
Engl J Med 1983;309:265–268.
52 Kromann N, Green A: Epidemiological studies
in the Upernavik district, Greenland. Acta
Med Scand 1980;208:401–406.
53 Bang H, Dyerberg J, Hjorne N: The composi-
tion of food consumed by Greenland Eskimos.
Acta Med Scand 1976;200:69–73.
54 Sanderson P, Calder PC: Dietary fish oil di-
minishes lymphocyte adhesion to macrophage
and endothelial dell monolayers. Immunology
1998;94:79–87.
55 De Caterina R, Libby P: Control of endothelial
leucocyte adhesion molecules by fatty acids.
Lipids 1996;31(suppl):57–63.
56 Yeh SL, Chang KY, Huang C, Chen WJ: Effect
of n-3 and n-6 fatty acids on plasma eicosanoid
and liver antioxidant enzymes in rats receiving
total parenteral nutrition. Nutrition 1997;13:
32–36.
57 Miles EA, Calder PC: Modulation of immune
function by dietary fatty acids. Proc Nutr Soc
1998;57:277–292.
58 Enders S: Messengers and mediators: Interac-
tions among lipids, eicosanoids and cytokines.
Am J Clin Nutr 1993;57(suppl):798–800.
59 Calder PC: n-3 polyunsaturated fatty acids and
immune cell function. Adv Enzyme Regul
1997;37:197–237.
60 Guthy E: Morbus Crohn und Nahrungsfette:
Hypothese zur Ätiologie der Enteritis regio-
nalis. Dtsch Med Wochenschr 1982;107:71–
73.
61 Guthy E: Ätiologie des Morbus Crohn: Was
spricht für Fette als mögliche Ursache. Dtsch
Med Wochenschr 1983;108:1729–1733.
62 Nordenval B, Brostrom O, Hellers G: Entzünd-
liche Darmkrankheiten und Nahrungsfette.
Dtsch Med Wochenschr 1982;107:1900.
63 Shoda R, Matsueda K, Yamato S, Umeda N:
Epidemiologic analysis of Crohn’s disease in
Japan: Increased dietary intake of n-6 polyun-
saturated fatty acids and animal protein relates
to the increase of Crohn’s disease in Japan. Am
J Clin Nutr 1996;63:741–745.
64 Lochs H, Meryn S, Marosi L, Ferenci P, Hort-
nag H: Has bowel rest a beneficial effect in the
treatment of Crohn’s disease? Clin Nutr 1983;
2:61–64.
65 Dickinson RJ, Ashton MG, Axon ATR, Smith
RC, Yeung CK, Hill GL: Controlled trial of
intravenous hyperalimentation and total bowel
rest as an adjunct to routine treatment of acute
colitis. Gastroenterology 1980;79:1199–1204.
66 McIntyre PB, Powel-Tuck J, Wood SR et al:
Controlled trial of bowel rest in the treatment
of severe acute colitis. Gut 1986;27:481–485.
67 Fernandez-Bañares F, Cabré E, Esteve-Comas
M, Gassull MA: How effective is enteral nutri-
tion in inducing remission in active Crohn’s
disease? A meta-analysis of the randomized
clinical trials. J Parenter Enteral Nutr 1995;19:
356–364.
68 Griffiths AM, Ohlsson A, Sherman PM, Su-
therland LR: Meta-analysis of enteral nutrition
as a primary treatment of active Crohn’s dis-
ease. Gastroenterology 1995;108:1056–1067.
69 Mesori A, Trallori G, D’Albasio G, Milla M,
Vannozzi G, Pacini F: Defined formula diets
versus steroids in the treatment of active
Crohn’s disease. A meta-analysis. Scand J Gas-
troenterol 1996;31:267–272.
70 Riordan AM, Hunter LJO, Cowman RE et al:
Treatment of active Crohn’s disease by exclu-
sion diet: East Anglian multicentre controlled
trial. Lancet 1993;342:1131–1134.
71 Van Den Bogaerde J, Kamm MA, Knight SC:
Immune sensitization to food, yeast and bacte-
ria in Crohn’s disease. Aliment Pharmacol
Ther 2001;15:1647–1653.
72 Van Den Bogaerde J, Cahill J, Emmanuel AV,
Vaizey CJ, Talbot IC, Knight SC, Kamm MA:
Gut mucosal response to food antigens in
Crohn’s disease. Aliment Pharmacol Ther
2002;11:1903–1915.
73 Gonzalez-Huix F, de Leon R, Fernandez-Ba-
ñares F, Esteve M, Cabré E, Acero D, Abad-
Lacruz A, Figa M, Guilera M, Planas R, Gas-
sull MA: Polymeric enteral diets as primary
treatment of active Crohn’s disease: A prospec-
tive steroid-controlled trial. Gut 1993;34:778–
782.
74 Gassull MA, Fernandez-Bañares F, Cabré E et
al: Fat composition may be the clue to explain
the primary therapeutic effect of enteral nutri-
tion in Crohn’s disease. Results of a double-
blind, randomised, multicenter European
study. Gut 2002;51:164–168.
75 Sanderson P, Calder PC: Dietary fish oil di-
minishes lymphocyte adhesion to macrophage
and endothelial dell monolayers. Immunology
1998;94:79–87.
76 De Caterina R, Libby P: Control of endothelial
leucocyte adhesion molecules by fatty acids.
Lipids 1996;31(suppl):57–63.
77 Yeh SL, Chang KY, Huang C, Chen WJ: Effect
of n-3 and n-6 fatty acids on plasma eicosanoid
and liver antioxidant enzymes in rats receiving
total parenteral nutrition. Nutrition 1997;13:
32–36.
78 Miles EA, Calder PC: Modulation of immune
function by dietary fatty acids. Proc Nutr Soc
1998;57:277–292.
Gassull
79 Enders S: Messengers and mediators: Interac-
tions among lipids, eicosanoids and cytokines.
Am J Clin Nutr 1993;57(suppl):798–800.
80 Calder PC: n-3 polyunsaturated fatty acids and
immune cell function. Adv Enzyme Regul
1997;37:197–237.
81 Chen LH, Zhao Y: Eicosapentaenoic acid sup-
presses lipopolysaccharide-induced tumor ne-
crosis factor-· expression by decreasing activa-
tion of activator protein-1. Am J Clin Nutr
2002;75(suppl):366.
82 Karsten S, Schäfer G, Schauder P: Cytokine
production and DNA synthesis by human pe-
ripheral lymphocytes in response to palmitic,
stearic, oleic and linoleic acid. J Cell Physiol
1994;161:15–22.
83 Fürst P, Kuhn KS: Fish oil emulsions: What
benefits can they bring? Clin Nutr 2000;19:7–
14.
84 Sadeghi S, Wallace FA, Calder PC: Dietary lip-
ids modify cytokine response to bacterial lipo-
polysaccharide in mice. Immunology 1999;96:
404–410.
85 Owen RW, Giacosa A, Hull WE, Haubner R,
Spiegelhalder B, Bartsch H: The antioxidant/
anticancer potential of phenolic compounds
isolated from olive oil. Eur J Cancer 2000;36:
1235–1237.
86 Mangas-Cruz MA, Fernandez A, Albi T, Guin-
da A, Relimpio F, Lanzon A, Pereira JL, Ser-
rera JL, Montilla C, Astorga R, Garcia-Luna
PP: Effects of minor constituents (non-glycer-
ide compounds) of virgin olive oil on plasma
lipid concentrations in male Wistar rats. Clin
Nutr 2001;20:211–215.
87 Wanten GJ, Geijtenbeek TBH, Raymakers
RAP, van Kooyk Y, Roos D, Jansen JBMJ,
Naber AHJ: Medium-chain triglyceride-con-
taining emulsions increase human neutrophil
ß2 integrin expression, adhesion and degranu-
lation. J Parenter Enteral Nutr 2000;24:228–
233.
88 Andoh A, Takaya H, Araki Y, Tsujikawa T,
Fujiyama Y, Bamba T: Medium- and long-
chain fatty acids differentially modulate inter-
leukin-8 secretion in human fetal intestinal epi-
thelial cells. J Nutr 2000;130:2636–2640.
89 Rodriguez-Palmero M, Kiss S, Fink M, Dem-
melmair H, Szabo I, Koletzko B: Effects of
medium-chain triglycerides on metabolism of
linoleic acid in preterm infants. Clin Nutr
1999;18(suppl 1):207A.
Nutrition and Inflammatory Bowel Disease
90 Sakurai T, Matsui T, Yal T et al: Short-term
efficacy of enteral nutrition in the treatment
of active Crohn’s disease: A randomized-con-
trolled trial comparing nutrient formulas. J
Parenter Enteral Nutr 2002,26:98–103.
91 Middleton SJ, Rucker JT, Kirby GA, Riordan
AM, Hunter JO: Long-chain triglycerides re-
duce the efficacy of enteral feeds in patients
with active Crohn’s disease. Clin Nutr 1995;
14:229–236.
92 Ina K, Itoh J, Fukushima K, Kusugami K et
al: Resistance of Crohn’s disease T cells to
multiple apoptotic signals is associated with
Bcl-2/Bax mucosal imbalance. J Immunol
1999;163:1081–1090.
93 Llor X, Pons E, Roca A, Alvarez M et al: The
effects of fish oil, olive oil, oleic acid, and lin-
oleic acid on coloreatal neoplastic processes.
Clin Nutr 2003;22:71–79.
94 González-Huix F, Fernández Bañares F, Es-
teve M, Abad A, Cabré E, Acero D, Figa M,
Guilera M, Humbert P, De León R, Gassull
MA: Enteral versus parenteral nutrition as
adjunct therapy in acute ulcerative colitis.
Am J Gastroenterol 1993;88:227–232.
95 Govers MJAP, Gannon NJ, Dunshea FR,
Gibson PR, Muir JG: Wheat bran affects the
site of fermentation of resistant starch and
luminal indexes related to colon cancer risk:
A study in pigs. Gut 1999;45:840–847.
96 Rombeau JL, Kripke SA: Metabolic effects of
short-chain fatty acids. J Parenter Enteral
Nutr 1990;14(suppl):181–185.
97 Royall D, Wolever TMS, Jeejeebhoy KN:
Clinical significance of colonic fermentation.
Am J Gastroenterol 1990;85:1307–1312.
98 Scheppach W: Effects of short-chain fatty
acids on gut morphology and function. Gut
1994;35:S35–S38.
99 Vernia P, Cittadini M, Caprilli R, Torsoli A:
Topical treatment of refractory distal ulcer-
ative colitis with 5-ASA and sodium butyrate.
Dig Dis Sci 1995;40:305–307.
100 Chapman MAS, Grahn MF, Boyle MA, Hut-
ton M, Rogers J, Williams NS: Butyrate oxi-
dation is impaired in the colonic mucosa of
sufferers of quiescent ulcerative colitis. Gut
1994;35:73–76.
101 Roediger WEW, Duncan A, Kapanidis O,
Millard S: Reducing sulfur compounds of the
colon impair colonocyte nutrition: Implica-
tions for ulcerative colitis. Gastroenterology
1993;104:802–809.
Dig Dis 2003;21:220–227
102 Roediger WEW, Duncan A, Kapanidis O,
Millard S: Sulphide impairment of butyrate
oxidation in rat colonocytes: A biochemical
basis for ulcerative colitis? Clin Sci 1993;85:
623–627.
103 Pitcher MCL, Beatty ER, Cummings JH: The
contribution of sulphate bacteria and 5-ami-
nosalycilic acid to faecal sulphide in patients
with ulcerative colitis. Gut 2000;46:64–72.
104 Den Hond E, Hiele M, Evenepoel P, Peeters
M, Ghoos Y, Rutgeerts P: In vivo butyrate
metabolism and colonic permeability in ex-
tensive ulcerative colitis. Gastroenterology
1998;115:584–590.
105 Simpson EJ, Chapman MAS, Dawson J, Ber-
ry D, Macdonald IA, Colen A: In vivo mea-
surement of colonic butyrate metabolism in
patients with quiescent ulcerative colitis. Gut
2000;46:73–77.
106 Scheppach W, Bartram HP, Richter F, Muell-
er JG, Greinwald R, Tauschel HD, Gierend
M, Weber A, Hegemann D, Kubetzko W,
Rabast U, Schuetz E, Raedsch R, Britsch R,
Rehmann IH, Otto P, Judmaier G, Press AG,
Woerdehoff D, Mlitz H, Stein J, Schmidt C:
Treatment of distal ulcerative colitis with
short-chain fatty acid enemas – A placebo-
controlled trial. Dig Dis Sci 1996;41:2254–
2259.
107 Fernández Bañares F, Hinojosa J, Gomollón
F, Sánchez-Lombraña JL, Rodrı́guez-Abascal
J, Gassull MA, Spanish Group for the Study
of Crohn’s Disease and Ulcerative Colitis:
Randomized clinical trial of Plantago ovata
(dietary fibre) as compared to mesalamine in
maintaining remission in ulcerative colitis.
Am J Gastroenterol 1999;94:427–433.
108 Inan MS, Rasoulpour RJ, Yin L, Hubbard
AK, Rosenberg DW, Giardina C: The lumi-
nal short-chain fatty acid butyrate modulates
NF-ÎB activity in a human colonic epithelial
cell line. Gastroenterology 2000;118:724–
734.
109 Wächterhaushäuser A, Loitsch SM, Stein J:
PPAR-Á is selectively upregulated in Caco-2
cells by butyrate. Biochem Biophys Res Com-
mun 2000;272:380–385.
110 Segain JP, Raingeard de la Blétière D, Bour-
reille A, Leray V, Gervois N, Rosales C, Fer-
rier L, Bonnet C, Blottiere HM, Galmiche JP:
Butyrate inhibits inflammatory responses
through NFÎB inhibition: Implications for
Crohn’s disease. Gut 2000;47:397–403.
227
 Review Article
Dig Dis 2003;21:228–236
DOI: 10.1159/000073340
Factors Enhancing Intestinal Adaptation
after Bowel Compensation
D.S. Botsios K.D. Vasiliadis
4th Surgical Clinic, Aristotle University of Thessaloniki, Thessaloniki, Greece
Key Words
Gut adaptation W Intestinal failure W Growth factors
Abstract
Intestinal failure (IF) refers to the condition in which cer-
tain causes lead to derangements in nutrient absorption
capacity. Gut adaptation occurs in response to IF and it is
both morphologic and physiologic in nature and can be
mediated by growth factors and nutrients. Our paper
reviews certain trophic growth factors that have impor-
tant interactions relevant for intestinal growth, function
and adaptation. Data Source: The literature was re-
viewed (data from both animal and human studies) and
certain trophic factors that modulate intestinal adapta-
tion are summarized. The factors reviewed are: epider-
mal growth factor, insulin-like growth factor I and II,
transforming growth factor · and ß, neurotensin, inter-
leukin-11, glucagon-like peptide-2, keratinocyte growth
factor, human growth hormone, short-chain fatty acids,
and glutamine. Conclusions: Growth factors augment
intestinal proliferation, diminish programmed apoptosis,
and modulate the adaptive process. They also have the
potential to improve nutrient absorption in some bowel
disease. The enhancement of gut adaptation may allow
patients to transition of parenteral/enteral to normal nu-
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0228$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
trition, in a shorter period of time, which reduce the rate
of adverse effects caused by artificial nutrition and im-
prove quality of life.
Copyright © 2003 S. Karger AG, Basel
Introduction
Intestinal failure (IF) refers to the condition in which
the functioning bowel is insufficient to allow for the
absorption of an adequate amount of nutrients. Consider-
able progress in the understanding and treatment of IF
has been made over the last years. Fleming and Reming-
ton [1] defined IF as ‘the reduction in functioning gut
mass below the amount necessary for adequate digestion
and absorption of food’. According to this definition, IF
can be present in patients with a normal, yet dysfunction-
al, length of bowel.
IF may be divided into two types: the first is character-
ized by an absolute reduction in normally functioning gut
mass (short bowel syndrome – SBS), and the second fea-
tured by an intestine with extensive lesions or functional
insufficiency (intestinal dysfunction). In addition, IF can
be complete (total small bowel enterectomy) or partial
(partial resection). The condition can be acute and tempo-
rary, as seen with recoverable motility disorders such as
D. Botsios, Assoc. Prof. Surg.
Kaukasou 64
GR–55133 Thessaloniki (Greece)
Tel. +30 2392051560, Fax +30 2310358000
E-Mail keva@med.auth.gr
ileus and obstruction, or chronic and permanent. Al-
though a wide spectrum of conditions can be associated
with IF, four principal underlying causes can be identified
– these are (a) the SBS, (b) total parenchymal bowel dis-
ease (e.g. Crohn’s disease), (c) motility disorders, such as
visceral myopathy and chronic intestinal obstruction, and
(d) small bowel fistulation which leads to premature loss
of enteric content. The major resulting nutritional disor-
ders are starvation and dehydration, but loss of body mass
is frequently made worse by catabolism from associated
sepsis [2].
SBS is the best-known form of IF. It usually results
from extensive small bowel resection for the conse-
quences of superior mesenteric artery or vein thrombosis
or volvulus. SBS can also result from repeated resection of
small bowel strictures due to Crohn’s disease. In addition,
SBS is associated with many aftereffects, such as gastric
hypersecretion, gallstone formation, renal calculi, liver
disease and metabolic bone disease. Chronic motility dis-
orders, such as those characterized, as intestinal pseudo-
obstruction or visceral myopathy, are rare causes of IF.
Regarding survival following massive bowel resection
in humans, a review made by Wilmore [3] documented
that in infants with an intact ileocecal valve, none with
small intestinal length !15 cm survived, while, in infants
receiving ileocecal resection, this length is extended to
40 cm. However, patients with an even shorter length of
remaining intestine can nowadays be successfully weaned
from parenteral nutrition (PN), if certain nutrients are
administered in combination. The lower limit of remain-
ing gut length compatible with successful weaning from
enteral feeding is estimated to be ^30 cm in children. In
adults, on the other hand, the extent of compensatory
hypertrophy is becoming very limited (the lower limit of
remaining intestinal length is said to be at least 50–70 cm)
[4].
Recovery of intestinal function can be expected from
efficient use of certain factors stimulating regeneration
and hypertrophy of the remnant intestinal mucosa, in
addition to prolonged nutritional management with en-
teral and parenteral nutritional regimens. It should be
noted that both SBS and IF are associated with the loss of
the gut barrier function.
Resolution of IF can occur spontaneously by the pro-
cess of intestinal adaptation, a process that is stimulated
by enteral nutrition, or in the case of intestinal fistula can
be resolved surgically. In those cases where resolution
does not take place, recourse has to be made to permanent
PN. The increased use of long-term total parenteral nutri-
tion (TPN) revealed numerous adverse effects that were
Factors Enhancing Intestinal Adaptation
after Bowel Compensation
not recognized previously. These are problems related to
(a) intravenous catheter care, (b) supply of micronutrients
and (c) the occurrence of irreversible hepatic failure.
Small Bowel Adaptation in Intestinal Failure
Where IF is the result of the SBS, it is the hope that
compensatory mechanisms may reduce dependence on
PN. Several mechanisms are available: (a) Spontaneous
adaptation, which results from cellular hyperplasia, vil-
lous enlargement, intestinal lengthening and dilatation,
altered motility, increased absorptive capacity and in-
creased activity of brush-border enzymes. Alimentary
secretions, ingested nutrients and hormones stimulate the
changes. (b) Bowel-lengthening procedures: over the past
16 years, Bianchi has pioneered bowel-lengthening proce-
dures in neonates and infants with SBS [5]. (c) Reserved
small bowel segments: segmental bowel reversal in pa-
tients with SBS aims to slow intestinal transit time, there-
by theoretically enhancing absorption. (d) Intestinal
transplantation, and (e) Pharmacological stimulation of
adaptation-growth factors and other modulators.
Growth Factors and Other Modulators
A number of specific factors that enhance the prolifera-
tion response of the enterocyte have been identified and
their effect was demonstrated mainly in animal studies.
Epidermal Growth Factor (EGF)
EGF, extracted from the mucous gland and Brunner’s
glands, is a low-molecular-weight polypeptide that is
known to enhance the growth of the intestinal mucosa. It
has been shown to be involved in the maturation of a
number of developing tissues. Several studies have dem-
onstrated the trophic effects of EGF on intestinal mucosa
in a variety of experimental models [6–11]. Some of the
effects of exogenous EGF on the gastrointestinal mucosa
are notably similar to the compensatory changes observed
after massive small bowel (SB) resection [12–14]. EGF
receptors are present throughout the intestinal tract and
located on both basolateral and brush-border mem-
branes.
In experimental studies continuous administration of
the EGF resulted in significant increases in body weight,
intestinal weight (SB and colon), SB length, and ileal
mucosal mass following massive SB resection [15]. In
addition, the mucosal DNA and protein contents in-
Dig Dis 2003;21:228–236 229
creased after EGF administration, suggesting an augmen-
tation of the usual mucosal hyperplasia observed after
massive SB resection.
Goodlad et al. documented that EGF administration
increased colon weight, which resulted from intestinal
epithelial proliferation. In addition, they observed an
increase in pancreatic weight after EGF administration,
which raised the possibility that EGF mediates some of its
trophic effects through increases in pancreatic growth or
secretion [16].
In normal mature intestinal mucosa, administration of
EGF has been reported to increase intestinal weight, lac-
tase-specific activity, net calcium transport [8], and galac-
tose and glycine absorption [17]. Similar trophic effects of
EGF on atrophic mucosa (caused by lack of enteral nutri-
tion while on TPN [7, 18, 19] or by administration of an
elemental diet) have also been reported [10]. Chaet et al.
showed in 1994 that administration of EGF after massive
SB loss can by of nutritional benefit through its enhance-
ment of the normal post-resection intestinal response
[15].
In a series of studies, Thomson et al. [9, 20–22] showed
that EGF stimulates neomucosal growth on serosal
patches created in the rabbit intestine. In their studies,
maximal stimulation was achieved through intravenous
[10] administration of EGF. EGF given enterally does not
stimulate cell proliferation in the colon, whereas intrave-
nous recombinant EGF reverses the marked intestinal
hypoplasia characteristically found in TPN-fed rats. In a
study of EGF administration after massive SB resection,
Read et al. [23] reported increased intestinal net weight
and sucrase activity after 7 days of oral EGF administra-
tion. Furthermore, EGF upregulates electrolytes and nu-
trient absorption in the small bowel and thus mediates the
absorptive capacities of various nutrients.
Read et al. [23], in an experimental study, adminis-
trated EGF orally and documented that there were no
positive synergistic effects of EGF and intestinal resec-
tion, nor did EGF fully compensate for the lack of luminal
contents. Because post-resection adaptation may begin
immediately after resection and continue for several
weeks [12], further studies are needed to evaluate EGF’s
effect on the residual SB after massive resection, when
administrated throughout the period of expected adapta-
tion. Although EGF had multiple trophic effects in experi-
mental models, EGF’s safety and utility for human thera-
py are presently under research.
230 Dig Dis 2003;21:228–236
Insulin-Like Growth Factors I and II (IGF-I and II)
IGF-I and II are growth factors similar in structure to
insulin, which is primarily involved in the regulation of
normal growth and development. The major source of cir-
culating IGF-I and II is the liver, and their effects are
more direct than that of growth hormone (GH). Several
studies have demonstrated the trophic effects of IGF-I
and II on intestinal mucosa in a variety of experimental
models [24].
IGF-I is produced in the liver in response to GH, and
leads to an increase in crypt cell production rate, with a
consequent increase in villus high and nutrient absorptive
capability [25, 26]. IGF-I enhances DNA and protein syn-
thesis in intestinal crypt cells in vitro, maintains intestinal
integrity, and enhances intestinal mucosal adaptation af-
ter intestinal resection. Localization of IGF receptors in
the intestinal epithelium suggests a functional role of
IGF-I in intestinal epithelial cell growth and differentia-
tion and thus a possible role of IGF-I in gut repair [27].
A recent experimental study demonstrated that admin-
istration of IGF-I enhanced epithelial restitution after
intestinal mucosal injury. In this model, increased thymi-
dine uptake, cell migration and increased TGF-ß mRNA
expression were noted in intestinal epithelial cells [27].
Sigalet and Martin [28] showed that IGF-I could im-
prove weight gain in a rat model of severe bowel syn-
drome (SBS). This improvement in weight gain was asso-
ciated with an increase in nutrient transport at the cellular
level and variable increases in villus size. IGF-I has the
greatest effect on weight gain, which should be the guide
to further studies regarding the mechanisms underlying
effects. IGF-I probably acts directly by improving nu-
trient absorption at the enterocyte level or by an effect on
adaptation. According to other similar studies [12, 29,
30], it is assumed that an increase in villus height accom-
panies any significant improvement in nutrient absorp-
tion characteristics. It is possible that IGF-I is acting
through endocrine pathways and only indirectly affecting
nutrient transport characteristics.
Although markedly beneficial effects of IGF-I and II
have been observed in animal studies, its clinical effects
on protein metabolism and gut integrity have not yet been
clearly demonstrated and further analysis is required.
Transforming Growth Factor (TGF)-· and ß
TGF-· is a 5.5-kDa protein containing 50-amino-acid
residues that share 35% structural homology with EGF, a
common membrane receptor with a similar spectrum of
biological activity [31]. TGF-· has been reported to stim-
ulate proliferation of a wide range of epidermal and epi-
Botsios/Vasiliadis
thelial cells and is important in intestinal cell proliferation
[32]. In a recent study, Falcone et al. [33] have shown that
in a mouse model of SBS, normal adaptation resulted in
30% increase in ileal TGF-· mRNA levels by post-opera-
tive day 3. Additionally, it has been shown that exogenous
TGF-· given within the first 3 days significantly en-
hanced enterocyte proliferation and stimulated intestinal
adaptation.
TGFs were isolated for the first time in 1978 from cul-
ture medium conditioned by Rous sarcoma virus-trans-
formed fibroblasts [34] and include two proteins, now
designated TGF-· and TGF-ß. Both proteins have a dif-
ferent structure and a different spectrum of biological
activity. Since its isolation from transformed cell lines,
TGF-· has been identified in a wide spectrum of epithe-
lial cells, including gastrointestinal epithelium [35].
TGF-· has recently been found to promote proliferation
in established cell lines and to exert a trophic effect on
intact gastric, intestinal and colonic mucosa. TGF-· is
considered important in epithelial cell proliferation in
digestive tissues during the developmental period [36], in
the promotion of cell migration and motivation of fluid
and electrolyte transport in the enterocyte [37], in the
inhibition of gastric acid secretion and promotion of gas-
tric ulcer healing [38] and gastric mucosal repair [39] and
in intestinal mucosal repair [40] following acute epithelial
injury.
Several lines of evidence have implicated TGF-· as a
potential modulator of intestinal adaptation in an animal
model of SBS.
In a recent study the potential effects of exogenous
TGF-· as a gut-trophic agent after small bowel resection
in the rat was investigated. It was shown that exogenous
TGF-· given at high doses (75 Ìg/kg/day) stimulated
intestinal adaptation in the remaining gut. This was evi-
dent from increased bowel and mucosal weights, mucosal
DNA and protein, villus height and crypt depth. Exoge-
nous TGF-· has been showed to accelerate the prolifera-
tion of intestinal cells. In addition, the administration of
TGF-· reduced cell death via apoptosis. Both increased
cell production and decreased cell death may indicate an
adaptive mechanism to increase enterocyte mass follow-
ing TGF-· administration [41]. These observations sug-
gest that TGF-· may have clinical utility for the patient
with SBS.
TGF-ß inhibits cell proliferation in vitro and in vivo.
No clinical studies have been performed to demonstrate
clinical efficacy.
Factors Enhancing Intestinal Adaptation
after Bowel Compensation
Neurotensin
Neurotensin is a tridecapeptide originally isolated
from bovine hypothalamus and subsequently found to be
widely distributed in the gastrointestinal tract. Neuroten-
sin-like immunoreactivity has been localized to a specific
mucosal endocrine cell type (N-cell) that is found in high-
est density in the ileum [42]. Neurotensin-like immunore-
activity in plasma increases two- to threefold after meals
[43]. Thus, potential physiological actions of neurotensin
on gastric and pancreatic secretion, gastrointestinal motil-
ity, and gastrointestinal growth may be mediated by a hor-
monal mechanism with release of neurotensin from N
cells of the intestinal mucosa, as well as paracrine and
neurocrine mechanisms involving local release from N
cells and nerve fibers.
Neurotensin increases contractile activity of the colon
[44] and disrupts the interdigestive myoelectric complex
of the small intestine [45]. Neurotensin also increases
intestinal blood flow and capillary permeability [42] and
stimulates net fluid secretion by small intestine in vitro
[47].
Wood et al. [48] showed that neurotensin had trophic
effects on the small intestine in rats. Neurotensin pro-
duced increases in weight, DNA and protein content of
the small intestine. There were also increases in total con-
tent of three brush-border enzymes measured in this
study, namely sucrase, maltase and leucine aminopepti-
dase. The potential action of neurotensin on small intesti-
nal growth is not due solely to an endocrine mechanism.
High doses of neurotensin may result in high concentra-
tions within the intestine, similar to those expected after
local release of neurotensin from endocrine cells or neu-
rons. By this reasoning, neurotensin could exert paracrine
or neurocrine effects on intestinal growth.
It is not yet established whether neurotensin acts
directly on the small intestine or indirectly through effects
on other growth factors. Neurotensin has been reported to
stimulate release of thyrotropin in rats [49] and insulin
and glucagon in dogs [50]. Other similar studies showed
that neurotensin prevents atrophy of the intestinal muco-
sa associated with TPN and promotes the release of secre-
tory IgA and IgM into the bile [51].
Interleukin-11 (IL-11)
IL-11, a 19-kDa bone marrow-derived growth factor, is
a multifunctional peptide that uses the gp130 receptor
common subunit for receptor function [52]. It was first
cloned in 1990 from an immortalized primate bone mar-
row stromal cell line, PU-34 [53]. Its hematopoietic
effects include increases in peripheral blood neutrophils
Dig Dis 2003;21:228–236 231
and platelets, and stimulation of erythroid elements, me-
gakaryocytes, and B-cell differentiation in vivo and in
vitro [54]. IL-11 also improves survival rate and decreases
bacterial translocation in burned mice [55]. Recent stud-
ies suggest that IL-11 enhances a small intestine mucosal
mass after massive small bowel resection [56, 57]. How-
ever, these studies did not determine if there were changes
in the function at the remnant small intestine. The current
findings show that exogenous administration of IL-11
increases mucosal mass, as noted by enhanced DNA con-
tent, in a dose-dependent manner. These data corroborate
the results of previously described in vivo studies and fur-
ther strengthen the assertion that IL-11 has a trophic
effect on intestinal epithelial cells. The systemic infusion
IL-11 had significantly enhanced galactose and glycine
absorption. [58].
Glucagon-Like Peptide-2 (GLP-2)
GLP-2 is a 33-amino-acid peptide known to enhance
mucosal integrity in normal intestine and in intestine that
has undergone adaptation after massive small bowel re-
section (MSBR) [59]. GLP-2 is liberated from the car-
boxy-terminus of proglucagon in small and large intestine
L cells by the action of tissue-specific proteases. GLP-2
belongs to a specific class of compounds referred to as pro-
glucagon-derived peptides (PGDPs), products of the post-
translational processing of proglucagon. A role for PGDPs
as intestinal hyperplasia and hypertrophy were noted in 2
patients with PGDP-secreting tumors [60, 61]. It was
shown subsequently that nude mice carrying subcuta-
neous PGDP-producing tumors had a significant increase
in small intestinal weight [62]. The PGDP responsible for
the intestinal epithelial proliferation in these mice was
identified as GLP-2. Subsequent studies confirmed that
the naturally occurring form of GLP-2 induces growth in
normal intestine [63–65].
Drucker et al. [66] further showed that a synthetic, pro-
tease-resistant form of GLP-2 was a potent agent for
increasing small and large intestine mass in normal rats.
Findings in other similar studies showed that GLP-2
increases mucosal mass and absorptive function in intes-
tine after MSBR [67]. In experiments on massive bowel
resection or ischemia/reperfusion injury, administration
of GLP-2 significantly increased DNA and protein con-
tent of the intestinal mucosa [59]. However, such effects
have not been confirmed in humans.
Keratinocyte Growth Factor (KGF)
KGF, synthesized and secreted by stromal fibroblasts,
is expressed predominantly in the dermis [68]. However,
232 Dig Dis 2003;21:228–236
expression of KGF receptor and KGF messenger RNA
(mRNA) has been detected throughout the gastrointesti-
nal tract [69]. Recent studies support a role for KGF in
tissue repair [70, 71]. In inflammatory bowel disease,
increased expression of KGF has been described in the
mucosa, suggesting a potential role for KGF in mediating
mucosal repair mechanisms [72, 73].
The role for KGF as an endogenous mediator of
growth and differentiation in the gastrointestinal tract is
supported by the isolation of mRNA for KGF and its
receptor in both the small and large intestine of rodents
[69, 74].
Recently it has been demonstrated that exogenous full-
length KGF significantly reduces the severity and extent
of mucosal injury in different experimental models of
colitis in rats and mice [75, 76]. KGF induces prolifera-
tion of colonic epithelium (or alternatively decreased
apoptosis) and quantitatively increases the mucin content
[75, 77]. KGF has also been demonstrated to increase the
number of goblet cells as well as mucin production
throughout the gastrointestinal tract of rats and mice [69,
75, 76]. Housley et al. [69] showed that KGF enhanced
the growth of the intestinal mucosa. An animal study by
Estivarez et al. [78] showed that KGF inhibits atrophy of
the duodenal and ileal mucosa induced by fasting. John-
son et al. [79] evaluated the effects of KGF on intestinal
adaptation after resection of 85% of the small intestine. It
was concluded that KGF enhances gut growth, differenti-
ation and gene regulation during adaptation in rat small
intestine after massive resection. The clinical relevance
was that KGF might be beneficial in the management of
patients undergoing massive intestinal resection. In a
recent experimental study in mice, Yang et al. [80]
showed that KGF administration improved epithelial ab-
sorptive function and stimulated intestinal proliferation
after SBS. They concluded that KGF improved intestinal
adaptation after small bowel syndrome and may have
clinical applicability.
Human Growth Hormone (hGH)
hGH belongs to anabolic hormones. In 1974, Wilmore
et al. [81] first reported the clinical effects of hGH. They
reported that administration of hGH to burn patients
resulted in an improvement of nitrogen and potassium
balance and enhanced wound healing. Based on initial
observations showing GH can increase nutrient and elec-
trolyte absorption in hypophysectemized animals [82],
Byrne et al. [83] reported in a nonrandomized, open study
that GH and dietary modification improved nutrient
absorption in patients with small bowel syndrome. Subse-
Botsios/Vasiliadis
quent studies using more rigorous techniques have shown
no effect on nutrient absorption with GH treatment in
human patients [84, 85], whereas a variety of animal stud-
ies have shown conflicting results [86–88].
Kissmeyer-Nielsen et al. [89] administrated GH to
TPN patients with gastrointestinal diseases and reported
its protein-sparing effect, a reduced incidence of infec-
tion, and a decrease in malaise. Concerning the proliferat-
ing effect of GH on intestinal mucosa, a number of animal
studies have demonstrated its beneficial effect on struc-
tural maintenance, adaptation and wound healing [90]. If
GH does affect the adaptive process, it is likely this effect
is mediated through IGF-I [24].
Sigalet and Martin [28], in their experimental study,
documented that GH administration improved weight
gain which resulted as a consequence in nutrient absorp-
tion characteristics [12, 29, 30]. Their experimental evi-
dence raises the hypothesis that GH is acting indirectly
through endocrine pathways, enhancing nutrient trans-
port. Gu et al. showed, in a rat model, that GH enhanced
small bowel adaptation. This result indicated that appli-
cation of GH could alleviate mucosal atrophy of the rem-
nant small intestine caused by PN and improve intestinal
adaptation significantly in short bowel rats [91].
Short-Chain Fatty Acids (SCFAs)
SCFAs are formed in the gastrointestinal tract of mam-
mals by microbial fermentation of carbohydrates, and
readily absorbed by intestinal and colonic mucosa, and
are trophic to intestinal mucosa. Acetate, propinate and
butyrate account for about 85% of SCFAs and are pro-
duced in a nearly constant molar ratio of 60:25:15 [92].
One week of SCFA supplementation has been shown to
retard TPN-associated atrophy in rats with intact bowels
[93] and after 80% intestinal resection [94]. Recent stud-
ies have reported that SCFA-supplemented TPN en-
hanced both structural [95] and functional [96] adapta-
tion to 80% intestinal resection as early as 3 days after
surgery.
SCFAs may influence small intestinal mucosal prolif-
eration by stimulating secretion of PGDPs. Intestinal
proglucagon mRNA abundance and plasma concentra-
tions of PGDPs are strongly correlated with cellular pro-
liferation during intestinal adaptation [97–100]. Precise
physiologic roles for each of the PGDPs continue to be
elucidated, however, it was shown recently that GLP-2
modulates basolateral membrane glucose transport in
rats [101]. Other studies showed that SCFA-supple-
mented TPN increases proglucagon gene expression 3
and 7 days after intestinal resection [95]. Physiologically,
Factors Enhancing Intestinal Adaptation
after Bowel Compensation
SCFAs are a logical mediator of trophic gastrointestinal
hormones (i.e., PGDPs) and ileal adaptation because
they are produced distally within the gastrointestinal
tract in response to malabsorbed substrate. Many studies
have established a strong relation between cellular prolif-
eration and elevated amounts of proglucagon mRNA and
PGDPs [97, 98, 102–104]. Taylor et al. [100] reported
that after intestinal resection, proglucagon expression in-
creased threefold, peaking 2 days after surgery and de-
clining thereafter. In a recent study it was shown that
SCFA-supplemented TPN increased proglucagon mRNA
abundance [95].
Tappenden et al. [105] showed that SCFA-supple-
mented TPN rapidly upregulated jejunal GLUT-2
mRNA (glucose transporter 2) and ileal GLUT-2 abun-
dance and ileal proglucagon mRNA. Within the intestinal
resection model, SCFAs may exert their trophic effect by
extending the duration of increased proglucagon expres-
sion.
Glutamine (Gln)
Gln is the most abundant nonessential amino acid in
the body. The most important function of Gln lies in its
metabolic role as a nitrogen carrier between major organs.
Gln prevents the degeneration of muscle protein, main-
tains body protein levels at the time of surgical stress and
is involved in the maintenance of intestinal epithelial cell
function as an energy source. A number of studies suggest
that Gln is conditionally essential for gut metabolism,
structure and function in the stressed animal [106, 107].
Gln has therefore been postulated to be of crucial impor-
tance for the integrity and growth of small intestinal
mucosa [108, 109]. Wiren et al. [110] were not able to
show better adaptation using Gln supplementation to the
diet compared to chow feeding, whereas several studies
confirmed that a dipeptive alanyl-Gln-enriched TPN so-
lution could increase the growth of the intestinal mucosa
in animals with intestinal atrophy due to TPN [111] and
massive bowel resection [112]. Dipeptive alanyl-Gln-
enriched TPN solution could also have a protective effect
on the mucus covering the surface of the intestinal muco-
sa [113].
For the clinical efficacy of Gln, several studies have
been performed demonstrating conflicting results [114–
116]. Specifically, concerning the effect of Gln on intesti-
nal function, Wilmore et al. [117] treated 300 short bowel
patients with a combination of GH, Gln and high carbo-
hydrate diet, and reported that 60% of the patients could
be weaned from TPN, in 30% TPN could be reduced, and
the remaining 10% still required the same quantity of
Dig Dis 2003;21:228–236 233
TPN. This result shows the importance of such a combi-
nation therapy in the adaptive process of SBS.
On the contrary, Scolapio et al. [84] reported no
improvements in small bowel morphology (villus height,
crypt depth), stool losses and macronutrient absorption in
response to active treatment with GH, Gln and high-car-
bohydrate, low-fat diet when compared with placebo car-
bohydrate in a randomized, crossover study in 8 patients
with well-established SBS. Additionally, Mandir and
Goodlad [118] also found there was no effect of Gln sup-
plementation on mitotic activity in the small intestine.
Alavi et al. [119] have also demonstrated that enteral
Gln does not promote adaptation either alone or in com-
bination with hepatocyte growth factor in resected rats.
However, further clinical studies need to be conducted to
confirm the efficacy of Gln-enriched TPN formulas in the
treatment of intestinal failure.
Conclusions
Growth factors augment intestinal proliferation, di-
minish programmed apoptosis, and modulate the adap-
tive process. They also have the potential to improve
nutrient absorption in some bowel diseases. The enhance-
ment of gut adaptation may allow the patient’s transition
from parenteral/enteral to normal nutrition in a shorter
period of time, which reduces the rate of adverse effects
caused by artificial nutrition and improves quality of life.
These data confirm that gut-trophic agent administration
may be of benefit to patients with IF during adaptation. In
addition, bowel adaptation enhancement could be useful
in the therapeutic management of patients with malab-
sorption syndromes. Although beneficial effects of these
trophic agents have been observed in human and animal
studies, further research and analysis is required for their
clinical applicability.

References

1 Fleming CR, Remington M: Intestinal failure.
Clin Surg Int 1981;2:219.
2 Irving M: Intestinal failure. J Gastroenterol
Hepatol 2000;15:G26–G29.
3 Wilmore DW: Factor correlating with a suc-
cessful outcome following extensive intestinal
resection in newborn infants. J Pediatr 1972;
80:88–95.
4 Okada A: Signaling factors for gut adaptation.
Nestle Nutr Workshop Ser Clin Perform Pro-
gramme 2002;7:169–178, discussion 178–181.
5 Jeppesen PB, Mortensen PB: Intestinal failure
defined by measurements of intestinal energy
and wet weight absorption. Gut 2000;46:701.
6 Al-Nafussi Al, Wright NA: The effect of epider-
mal growth factor on cell proliferation of the
gastrointestinal mucosa in rodents. Virchows
Arch B Cell Pathol 1982;40:63–69.
7 Goodlad RA, Wilson TJG, Lenton W et al:
Urogastrone-epidermal growth factor is trophic
to the epithelium of parenterally fed rats. Ex-
perientia 1985;41:1161–163.
8 Oka Y, Ghishan FK, Greene HL et al: Effect of
mouse epidermal growth factor/urogastrone on
the functional maturation of rat intestine. En-
docrinology 1983;112:940–944.
9 Thomson JS, Saxena SK, Greaton C et al: The
effect of the route of delivery of urogastrone on
intestinal regeneration. Surgery 1989;106:45–
51.
10 Tsujikawa T, Bamba T, Hosada S: The trophic
effect of epidermal growth factor on morpho-
logical changes and polyamine metabolism in
the small intestine of rats. Gastroenterol Jpn
1990;25:328–334.
11 Ulshen MH, Lyn-Cook LE, Raasch RH: Effects
of intraluminal epidermal growth factor on
mucosal proliferation in the small intestine of
adults rats. Gastroenterology 1986;91:1134–
1140.
12 Dowling RH, Booth CC: Structural and func-
tional changes following small intestinal resec-
tion in the rat. Clin Sci 1967;32:139–149.
13 Hanson WR: Proliferation and morphological
adaptation of the intestine to experimental re-
section. Scand J Gastroenterol 1982;17:11–20.
14 Booth CC, Evans KT, Menzies T et al: Intesti-
nal hypertrophy following partial resection of
the small bowel in the rat. Br J Surg 1959;46:
403–410.
15 Goodlad RA, Savage AP, Lenton W et al: Does
resection enhance the response of the intestine
to urogastrone-epidermal growth factor in the
rat? Clin Sci 1988;75:121–126.
16 Schwartz MZ, Storazuk RB: Influence of epi-
dermal growth factor on intestinal function in
the rat: Comparison of systemic infusion ver-
sus luminal perfusion. Am J Surg 1988;155:18–
22.
17 Goodlad RA, Raja KB, Peters TJ et al: Effects
of urogastrone-epidermal growth factor on in-
testinal brush border enzymes and mitotic ac-
tivity. Gut 1991;32:994–998.
18 Goodlad RA, Wilson TJG, Lenton W et al:
Intravenous but not intragastric urogastrone-
EGF to the intestine of parenterally fed rats.
Gut 1987;28:573–582.
19 Chaet MS, Arya G, Ziegler MM, Warner BW:
Epidermal growth factor enhances intestinal
adaptation after massive small bowel resection.
J Pediatr Surg 1994;29:1035–1039.
20 Thompson JS, Saxena SK, Sharp JG: Effect of
the duration of infusion of urogastrone on in-
testinal regeneration in rabbits. Cell Tissue Ki-
net 1989;22:303–309.
21 Thompson JS, Sharp JG, Saxena SK: Stimula-
tion of neomucosal growth by systemic urogas-
trone. J Surg Res 1987;42:402–410.
22 Thomson JS: Epidermal growth factor and the
short bowel syndrome. J Parenteral Enteral
Nutr 1999;23:S113–S116.
23 Read LC, Ford WDA, Filsell OH et al: Is orally-
derived epidermal growth factor beneficial fol-
lowing premature birth or intestinal resection?
Endocr Exp 1986;20:199–207.
24 Le Roith D: Insulin-like growth factors. Mass
Med Soc 1997;336:633–639.
25 Lemmey AB, Matine AA, Read LC et al: IGF-1
and the truncated analogue des-(1–3) IGF-1
enhance growth in rats after gut resection. Am J
Physiol 1991;260:E213–E219.
26 Steeb CB, Trahair JF, Tomas FM et al: Pro-
longed administration of IGF peptides en-
hances growth of gastrointestinal tissues in nor-
mal rats. Am J Physiol 1993;266:G1090–
G1098.
27 Chen K, Inoue M, Wasa M, Azuma T, Fukuza-
wa M, Kamata S, Okada A: Insulin-like growth
factor-I stimulates intestinal epithelial cell pro-
liferation and migration. Surg Forum 1997;48:
716–718.
28 Sigalet DL, Martin GR: Hormonal therapy for
short bowel syndrome. J Pediatr Surg 2000;35:
360–364.
29 Williamson R: Intestinal adaptation: Structur-
al, functional and cytokinetic changes. N Engl J
Med 1978;298:144, 1393–1402, 1450.
30 Thomson JS: Can the intestine adapt to a
changing environment? Gastroenterology
1997;113:1402–1412.

234 Dig Dis 2003;21:228–236 Botsios/Vasiliadis

31 Derynck R: Transforming growth factor-·. Cell
1988;54:593.
32 Alison MR, Sarraf CE: The role of growth fac-
tors in intestinal cell proliferation. Cell Biol Int
1994;18:1.
33 Falcone RA, Stern LE, Kemp CJ, Erwin CR,
Warner BW: Intestinal adaptation occurs inde-
pendent of transforming growth factor-·. J Pe-
diatr Surg 2000;35:365.
34 Delarco JE, Torado GJ: Growth factors from
murine sarcoma virus-transformed cells. Proc
Natl Acad Sci USA 1978;75:4001.
35 Malden LT, Novak U, Burgess AW: Expression
of transforming growth factor-· messenger in
the normal and neoplastic gastrointestinal
tract. Int J Cancer 1989;43:380.
36 Hormi K, Lehy T: Transforming growth factor-
· in vivo stimulates epithelial cell proliferation
in digestive tissues of suckling rats. Gut 1996;
39:532.
37 Barnard JA, Polk WH, Moses HL, Coffey RJ:
Production of transforming growth factor-· by
normal small intestine. Am J Physiol 1991;261:
C994.
38 Konturek SJ, Brzozowski T, Maka J, Dembins-
ki A, Slomiany A, Slomiany BL: Transforming
growth factor-· and epidermal growth factor in
protection and healing of gastric mucosal inju-
ry. Scand J Gastroenterol 1992;27:649.
39 Romano M, Polk WH, Awad JA, Arteaga CL,
Nanney LB, Wargovich MJ, Kraus ER, Boland
CR, Coffey RJ: Transforming growth factor-·
protection against drug-induced injury to the
rat gastric mucosa in vivo. J Clin Invest 1992;
90:2409.
40 Dignass AU, Stow JL, Babyatsky MW: Acute
epithelial injury in the rat small intestine in
vivo is associated with expanded expression of
transforming growth factor-· and ß. Gut 1996;
38:291.
41 Sukhotnik I, Yakirevich E, Coran AG, Siplo-
vich L, Krausz M, Hirsh M, Sabo E, Shiloni E:
Effect of transforming growth factor-· on intes-
tinal adaptation in a rat model of short bowel
syndrome. J Surg Res 2002;108:235–242.
42 Helmstaedter V, Taugner CH, Feurle GE,
Forssmann WG: Localization of neurotensin-
immunoreactive cells in the small intestine of
man and various mammals. Histochemistry
1977;53:35–41.
43 Holst-Pedersen J, Fahrenkrug J: Neurotensin-
like immunoreactivities in human plasma:
Feeding responses and metabolism. Peptides
1986;7:15–20.
44 Thor K, Rosell S: Neurotensin increases co-
lonic motility. Gastroenterology 1986;90:27–
31.
45 Andersson S, Rosell S, Hjelmquist V, Chang D,
Folkers K: Inhibition of gastric and intestinal
motor activity in dogs by [Gln4]-neurotensin.
Acta Physiol Scand 1977;100:231–235.
46 Harper SL, Barrowman JA, Kvietys PR, Gran-
ger DN: Effect of neurotensin on intestinal cap-
illary permeability and blood flow. Am J Physi-
ol 1984;247:G161–G166.
47 Mitchenere P, Adrian TE, Kirk RM, Bloom
SR: Effect of gut regulatory peptides on intesti-
nal luminal fluid in the rat. Life Sci 1981;29:
1563–1570.
Factors Enhancing Intestinal Adaptation
after Bowel Compensation
48 Wood JG, Hoang HD, Bussjaeger LJ, Solomon
TE: Neurotensin stimulates growth of small
intestine in rats. Am J Physiol 1988;255:813–
817.
49 Folkers K, Chang D, Humphries J, Carraway
R, Leeman SE, Bowers CY: Synthesis and ac-
tivities of neurotensin and its acid and amine
analogs: Possible natural occurrence of Gln4-
neurotensin. Proc Natl Acad Sci USA 1976;73:
3833–3837
50 Ukai M, Inoue I, Itatsu T: Effect of somatosta-
tin on neurotensin-induced glucagon release
and hyperglycemia. Endocrinology 1977;100:
1284–1286.
51 Burke D, Alverdy J, Aoys E, Agarwal A, Moss
G: The effect of neurotension and epidermal
growth factor on TPN induced IgA deficiency.
Surg Forum 1990;41:70–72.
52 Du XX, Williams DA: Interleukin-11: Review
of molecular, cell biology and clinical use.
Blood 1997;89:3897–3908.
53 Du XX, Williams DA: Interleukin-11: A multi-
functional growth factor derived from the he-
matopoietic microenvironment. Blood 1994;
83:2023–2030.
54 Du XX, Neben T, Goldman S: Effects of
recombinant human interleukin-11 on hemato-
poietic reconstitution in transplant mice: Ac-
celeration of recovery of peripheral blood neu-
trophils and platelets. Blood 1993;81:27–34.
55 Schindel D, Maze R, Liu, Q et al: Interleukin-
11 improves survival and reduces bacterial
translocation and bone marrow suppression in
burned mice. J Pediatr Surg 1997;32:312–315.
56 Liu Q, Du XX, Schindel DT et al: Trophic
effects of interleukin-11 in rats with experi-
mental short bowel syndrome. J Pediatr Surg
1996;31:1047–1051.
57 Fiore NF, Ledniczky G, Liu Q et al: Compari-
son of interleukin-11 and epidermal growth
factor on residual small intestine after massive
small bowel resection. J Pediatr Surg 1998;33:
24–29.
58 Alavi K, Prasad R, Lundgren K, Schwartz MZ:
Interleukin-11 enhances small intestine ab-
sorptive function and mucosal mass after intes-
tinal adaptation. J Pediatr Surg 2000;35:371–
374.
59 Prasad R, Alavi K, Schwartz MZ: Glucagon-
like peptide-2 analogue enhances intestinal
mucosal mass after ischemia and reperfusion. J
Pediatr Surg 2000;35:357–359.
60 Gleeson MH, Bloom SR, Polak JM et al: Endo-
crine tumour in kidney affecting small bowel
structure, motility and absorptive function.
Gut 1971;12:773–782.
61 Stevens FM, Flanagan RW, O’Gorman D et al:
Glucagonoma syndrome demonstrating giant
duodenal villi. Gut 1984;25:784–791.
62 Drucker DJ, Ehrlich P, Asa SL et al: Induction
of intestinal epithelial proliferation by gluca-
gon-like peptide 2. Proc Natl Acad Sci USA
1996;93:7911–7916.
63 Brubaker PL, Izzo A, Hill M et al: Intestinal
function in mice with small bowel growth in-
duced by glucagon-like peptide-2. Am J Physiol
1997;272:E1050–E1058.
Dig Dis 2003;21:228–236
64 Tsai C, Hill M, Asa SL et al: Intestinal growth
promoting properties of glucagon-like peptide-
2 in mice. Am J Physiol 1997;273:E77–E84.
65 Fischer KD, Dhanvantari S, Drucker DJ et al:
Intestinal growth is associated with elevated
levels of glucagon-like peptide-2 in diabetic
rats. Am J Physiol 1997;273:E815–E820.
66 Drucker DJ, DeForest L, Brubaker PL: Intesti-
nal response to growth factors administered
alone or in combination with human [Gly2]glu-
cagon-like peptide 2. Am J Physiol 1997;273:
G1252–G1262.
67 Kato Y, Alavi K, Yu D et al: Glucagon-like
peptide-2 enhances intestinal function follow-
ing massive small bowel resection. Gastroen-
terology 1998;114:A1152.
68 Aaronson SA, Bottaro DP, Miki T, Ron D,
Finch PW, Fleming TP et al: Keratinocyte
growth factor. A fibroblast growth factor family
member with unusual target cell specificity.
Ann NY Acad Sci 1991;638:62–77.
69 Housley RM, Morris CF, Boyle W, Ring B,
Biltz R, Tarpley JE et al: Keratinocyte growth
factor induces proliferation of hepatocytes and
epithelial cells throughout the rat gastrointesti-
nal tract. J Clin Invest 1994;94:1764–1777.
70 Werner S, Peters KG, Longaker MT, Fuller-
Pace F, Banda MJ, Williams LT: Large induc-
tion of keratinocyte growth factor expression in
the dermis during wound healing. Proc Natl
Acad Sci USA 1992;89:6896–6900.
71 Staiano-Coico L, Krueger JG, Rubin JS, D’limi
S, Vallat VP, Valentino L et al: Human kerati-
nocyte growth factor effects in a porcine model
of epidermal wound healing. J Exp Med 1993;
178:865–878.
72 Brauchle M, Madlener M, Wagner AD, Anger-
meyer K, Lauer U, Hofschneider PH et al: Ker-
atinocyte growth factor is highly over expressed
in inflammatory bowel disease. Am J Pathol
1996;149:521–529.
73 Finch PW, Pricolo V, Wu A, Finkelstein SD:
Increased expression of keratinocyte growth
factor messenger RNA associated with inflam-
matory bowel disease. Gastroenterology 1996;
110:441–451.
74 Finch PW, Rubin JS, Miki T, Ron D, Aaronson
SA: Human KGF is FGF-related with proper-
ties of a paracrine effector of epithelial cell
growth. Science 1989;245:752–755.
75 Zeeh JM, Proccacino F, Hoffmann P, Auker-
man SL, McRoberts JA, Soltani S et al: Kerati-
nocyte growth factor ameliorates mucosal inju-
ry in an experimental model of colitis in rats.
Gastroenterology 1996;110:1077–1083.
76 Egger B, Proccacino F, Sarosi I, Tolmos J,
Büchler MW, Eysselein VE: Keratinocyte
growth factor ameliorates dextran sodium sul-
fate colitis in mice. Dig Dis Sci 1999;44:836–
844.
77 Egger B, Tolmos J, Proccacino F, Sarosi I,
Friess H, Büchler MW et al: Keratinocyte
growth factor promotes healing of left sided
colon anastomoses. Am J Surg 1998;176:18–
24.
235
78 Estivarez CF, Jonas CR, Gu LH, Diaz EE, Wal-
lace TM, Pascal RR, Farrell CL, Ziegler TR:
Gut-trophic effects of keratinocyte growth fac-
tor in rat small intestine and colon during
enteral refeeding. J Parenter Enteral Nutr
1998;22:259–267.
79 Johnson WF, DiPalma CR, Ziegler TR, Scully
S, Farrell CL: Keratinocyte growth factor en-
hances early gut adaptation in a rat model of
short bowel syndrome. Vet Surg 2000;29:17–
27.
80 Yang H, Wildhaber BE, Teitelbaum DH: Har-
ry M. Vars Research Award: Keratinocyte
growth factor improves epithelial function af-
ter massive small bowel resection. J Parenter
Enteral Nutr 2003;27:198–206.
81 Wilmore DW, Moylan JA Jr, Bristow BF, Ma-
son AD Jr, Pruitt BA Jr: Anabolic effects of
human growth hormone and high caloric feed-
ings following thermal injury. Surg Gynecol
Obstet 1974;138:875–884.
82 Mainoya JR: Influence of ovine growth hor-
mone on water and NaCl absorption by the rat
proximal jejunum and distal ileum. Comp Bio-
chem Physiol 1982;71A:477–479.
83 Byrne TA, Persinger RL, Young LS et al: A new
treatment for patients with short bowel syn-
drome. Ann Surg 1995;222:243–255.
84 Scolapio JS, Calilleri M, Gleming CR et al:
Effect of growth hormone, glutamine and diet
on adaptation in short bowel syndrome: A ran-
domized, controlled study. Gastroenterology
1997;113:1074–1081.
85 Park JHY, Vanderhoof JA: Growth hormone
did not enhance mucosal hyperplasia after
small bowel resection. Scand J Gastroenterol
1996;31:349–354.
86 Vanderhoof JA, Kollmann KA, Griffin S et al:
Growth hormone and glutamine do not stimu-
late intestinal adaptation following massive
small bowel resection in the rat. J Pediatr Gas-
troenterol Nutr 1997;25:327–331.
87 Shulman DI, Hu CS, Duckett G et al: Effects of
short-term growth hormone therapy in rats un-
dergoing 75% small intestinal resection. J Pe-
diatr Gastroenterol Nutr 1992;14:3–11.
88 Benhamou PH, Canarelli JP, Richard C et al:
Human recombinant growth hormone in-
creases small bowel lengthening after massive
small bowel resection in piglets. J Pediatr Surg
1997;32:1332–1336.
89 Kissmeyer-Nielsen P, Jensen MB, Laurberg S:
Perioperative growth hormone treatment and
functional outcome after major abdominal sur-
gery. A randomized double-blind, controlled
study. Ann Surg 1999;229:298–302.
90 Gomez de Segura IA, Aguilera MJ, Codesal J,
De Miguel E: Administration of growth hor-
mone enhances the intestinal adaptive re-
sponse after resection of small intestine in rats.
Rev Esp Enferm Dig 1995;87:288–293.
91 Cummings MC, Winterford CM, Walker NI:
Apoptosis. Am J Surg Pathol 1997;21:88–101.
92 Cummings JH: Colonic absorption: The im-
portance of short chain fatty acids in man.
Scand J Gastroenterol 1984;19(suppl 93):89–
99.
236 Dig Dis 2003;21:228–236
93 Koruda MJ, Rolandeli RH, Bliss DZ, Hast-
ings J, Rombeau JL, Settle RG: Parenteral
nutrition supplemented with short-chain fat-
ty acids: Effects on the small-bowel mucosa in
normal rats. Am J Clin Nutr 1990;51:685–
689.
94 Koruda MJ, Rolandeli RH, Settle RG, Zim-
maro DM, Rombeau JL: Effect of parenteral
nutrition supplemented with short chain fatty
acids on adaptation to massive small bowel
resection. Gastroenterology 1988;95:715–
720.
95 Tappenden KA, Thomson ABR, Wild GE,
McBurney MI: Short-chain fatty acids in-
crease proglucagon and ornithine decarboxyl-
ase messenger RNAs following intestinal re-
section in rats. J Parenter Enteral Nutr 1996;
20:357–362.
96 Tappenden KA, Thomson ABR, Wild GE,
McBurney MI: Short-chain fatty acid supple-
mented total parenteral nutrition enhances
intestinal function following massive small
bowel resection. Gastroenterology 1997;112:
792–802.
97 Bloom SR, Polak JM: The hormonal pattern
of intestinal adaptation: A major role for ente-
roglucagon. Scand J Gastroenterol Suppl
1982;14:93–103.
98 Sagor GR, Ghatei MA, Al-Mukhtar MYT,
Wright NA, Bloom SR: Evidence for a hu-
moral mechanism after small intestinal resec-
tion: exclusion of gastrin but not enterogluca-
gon. Gastroenterology 1983;84:902–906.
99 Rountree DB, Ulshen MH, Selub S et al:
Nutrient independent increases in progluca-
gon and ornithine decarboxylase messenger
RNAs after jejunoileal resection. Gastroen-
terology 1992;103:462–468.
100 Taylor RG, Verity K, Fuller PJ: Ileal glucagon
gene expression: Ontogeny and response to
massive small bowel resection. Gastroenterol-
ogy 1990;99:724–729.
101 Cheeseman CI, Tsang R: The effect of gastric
inhibitory polypeptide and glucagons like
peptides on intestinal hexose transport. Am J
Physiol 1996;261:G477–G482.
102 Dowling RH: Small bowel resection and its
regulation. Scand J Gastroenterol Suppl
1982;17:53–74.
103 Gornacz GE, Al-Mukhtar MYT, Ghatei MA,
Sagor GR, Wright NA, Bloom SR: Pattern of
cell proliferation and enteroglucagon re-
sponse following small bowel resection in the
rat. Digestion 1984;29:65–72.
104 Al-Mukhtar MYT, Sagor GR, Ghati MA,
Bloom SR, Wright NA: The role of pancreati-
co-biliary secretions in intestinal adaptation
after resection, and its relationship to plasma
enteroglucagon. Br J Surg 1983;70:398–400.
105 Tappenden KA, Drozdowski LA, Thomson
AB, McBarney MI: Short-chain fatty acid-
supplemented total parenteral nutrition alters
intestinal structure, glucose transporter-2
mRNA and protein, and proglucagon mRNA
abundance in normal rats. Am J Clin Nutr
1998;68:118–125.
106 Souba WW: Intestinal glutamine metabolism
and nutrition. J Nutr Biochem 1993;4:2–9.
107 Winmueller HG, Spaeth AE: Identification of
ketone bodies and glutamine as the major
respiratory fuel in vivo for post-absorptive rat
small intestine. J Biol Chem 1978;1078:69–
76.
108 Wiren M, Permert J, Skullman S, Wang F,
Larsson J: Intraluminal glutamine refeeding
supports mucosal growth in rat jejunum.
Transplant Proc 1994;26:1460–1463.
109 Barber AE, Jones WG, Minei JP et al: Gluta-
mine or fiber supplementation of a defined
formula diet: Impact on bacterial transloca-
tion, tissue composition and response to en-
dotoxin. J Parenter Enteral Nutr 1990;14:
335–343.
110 Wiren M, Magnusson KE, Larsson J: Enteral
glutamine increases growth and absorptive
capacity of intestinal mucosa in the malnour-
ished rat. Scand J Gastroenterol 1995;30:
146–152.
111 Tamada H, Nezu R, Imamura I, Takagi Y,
Okada A: The dipeptide alanyl glutamine pre-
vents intestinal mucosal atrophy in parenter-
ally fed rats. J Parenter Enteral Nutr 1992;16:
110–116.
112 Tamada H, Nezu R, Matsuo Y, Imamura I,
Takagi Y, Okada A: Alanyl glutamine-en-
riched total parenteral nutrition restores in-
testinal adaptation after either proximal or
distal massive resection in rats. J Parenter
Enteral Nutr 1993;17:236–242.
113 Khan J, Liboshi Y, Cui L, Wasa M, Sando K,
Takagi Y, Okada A: Alanyl-glutamine-sup-
plemented parenteral nutrition increases lu-
minal mucus gel and decreases permeability
in the rat small intestine. J Parenter Enteral
Nutr 1999;23:24–31.
114 Ziegler TR, Young LS, Banfell K, Shelings M,
Hortos K, Bye R, Morrow FD, Jacobs DO,
Smith RJ, Antin JH, Wilmore DD: Clinical
and metabolic efficiency of glutamine-supple-
mented parenteral nutrition after bone mar-
row transplantation. A randomized double-
blind, control study. Ann Intern Med 1992;
116:821–828.
115 Mochizuki H, Saito H, Tashiro T, Tahara Y,
Okada A: A clinical study on the effect of ala-
nyl-glutamine supplemented TPN in gastrec-
tomized patients – Early II study (abstract). J
Parenter Enteral Nutr 1999;23(suppl):160.
116 Griffiths RD, Jones C, Palmer A: A six-
month outcome of critically ill patients given
glutamine-supplemented parenteral nutri-
tion. Nutrition 1997;13:295–302.
117 Wilmore DW, Lacey JM, Soultanakis RP,
Bosch RL, Burne TA: Factors predicting a
successful outcome after pharmacologic bow-
el compensation. Ann Surg 1997;226:288–
292.
118 Mandir N, Goodlad RA: The effects of gluta-
mine on intestinal cell proliferation in paren-
terally fed rats. Gut 1999;44:608–614.
119 Alavi K, Kato Y, Yu D, Schwartz MZ: Enter-
al glutamine does not enhance the effects of
hepatocyte growth factor in short bowel syn-
drome. J Pediatr Surg 1998;33:1666–1669.
Botsios/Vasiliadis
 Review Article

Dig Dis 2003;21:237–244

DOI: 10.1159/000073341

Helicobacter pylori Infection,

Vitamin B12 and Homocysteine
A Review

Jutta Dierkes a Matthias Ebert b Peter Malfertheiner b Claus Luley a

a Institute
of Clinical Chemistry and Biochemistry, and b Department of Gastroenterology and Hepatology,
University Hospital Magdeburg, Magdeburg, Germany

Key Words Introduction

Helicobacter pylori infection W Vitamin B12 deficiency W
Homocysteine W Pernicious anemia
Abstract
It has been suggested that there is an association be-
tween Helicobacter pylori infection, reduced cobalamin
absorption and cobalamin status and, consequently, ele-
vated homocysteine levels. This would offer an explana-
tion why H. pylori infection is associated with coronary
heart disease. To date, more than 25 studies have been
published that either deal with H. pylori infection and
homocysteine, H. pylori infection and cobalamin status,
or both. The design of these studies differs widely in
terms of definition of H. pylori status, measuring cobal-
amin status, selection of study cohorts and geographical
study areas. Therefore, results are fairly inconclusive at
present and do not suggest a major role of H. pylori
infection in the development of cobalamin deficiency
and elevated homocysteine levels.
Copyright © 2003 S. Karger AG, Basel
The aetiology of atrophic gastritis and gastric cancer
has been rewritten since the detection of Helicobacter
pylori during the 1980s. It is now known that the major
cause of atrophic gastritis is an infection with H. pylori,
which normally occurs in early childhood and persists
lifelong if left untreated. Prevalence rates of H. pylori
infection vary by age, country of origin, socioeconomic
status and are typically about 20–50% in healthy adults in
Western Europe and 80% in adults in economically less
developed countries [1]. Although the infection will cause
gastric inflammation in virtually all infected subjects, the
majority of infected subjects will remain asymptomatic,
while others develop atrophic gastritis and subsequently
gastric ulcer which can further develop to gastric cancer
[2].
One severe consequence of atrophic gastritis is the
malabsorption of cobalamin (vitamin B12), which is fre-
quent in the elderly due to hypo- or achlorhydria with sub-
sequent bacterial overgrowth, and reduced production
and secretion of intrinsic factor [3]. It has been suggested
that H. pylori infection may play an important role in the
reduction of acid production, reduced intrinsic factor
secretion and therefore the development of vitamin B12
deficiency. However, development of vitamin B12 defi-

© 2003 S. Karger AG, Basel Dr. Jutta Dierkes

ABC 0257–2753/03/0213–0237$19.50/0 Institute of Clinical Chemistry and Biochemistry
Fax + 41 61 306 12 34 Leipziger Strasse 44, DE–39120 Magdeburg (Germany)
E-Mail karger@karger.ch Accessible online at: Tel. +49 391 6713900, Fax +49 391 6713902
www.karger.com www.karger.com/ddi E-Mail jutta.dierkes@medizin.uni-magdeburg.de
Table 1. Studies on H. pylori and cobalamin, pernicious anemia and homocysteine

Group (first author) Patients Effect with H. pylori

Cobalamin
Van Asselt, 1998 [21]
Pilott, 1996 [22]
Cenerelli, 2002 [23]
Bloemenkamp, 2001 [24]
Shuval-Sudai, 2003 [25]
Tamura, 2002 [26]
Carmel, 2001 [28]
Serin, 2002 [29]
Pernicious anemia
Fong, 1991 [11]
Haruma, 1995 [12]
Marignani, 1999 [13]
Annibale, 2001 [14]
Andres, 2003 [15]
Annibale, 2000 [16]
Kaptan, 2000 [17]
Avcu, 2001 [18]
Ma, 1994 [19]
Homocysteine
Saxena, 1997 [30]
Leung, 2001 [31]
Yoshino, 2002 [32]
Bloemenkamp, 2001 [24]
Tamura, 2002 [26]
Whincup, 2000 [33]
Cenerelli, 2002 [23]
Effect of eradication therapy
Leung, 2001 [31]
Kaptan, 2000 [17]
Avcu, 2001 [18]
Serin, 2002 [29]
Mild cobalamin deficiency in healthy elderly No association
Serum cobalamin and folate in elderly No association
Serum cobalamin and folate in diabetes mellitus No association
type 2
Serum cobalamin and folate women with No association
peripheral artery disease
Serum cobalamin in outpatients Higher frequency of low serum cobalamin in H.pylori-
positive patients
Serum cobalamin and folate in patients with Lower serum cobalamin and folate in patient with
coronary artery disease H. pylori
Food cobalamin malabsorption in patients with Lower food cobalamin absorption in H.pylori-positive
low cobalamin patients
Gastric histology and serum cobalamin Serum cobalamin related to atrophy and
inflammation and H.pylori density
Pernicious anemia, case-control study No association
Pernicious anemia, case-control study No association
Patients with macrocytic anemia No association
Pernicious anemia No association
Cobalamin-deficient elderly No association
Pernicious anemia Frequency of H. pylori-positive patients: 60%
Vitamin B12-related anemia Frequency of H. pylori-positive patients: 56%
Vitamin B12 deficiency Frequency of H. pylori-positive patients: 57%
Pernicious anemia Frequency of H.pylori-positive patients: 83%
Patients No association
Dyspeptic patients No association
Case-control H. pylori-positive vs. -negative No association
Women with peripheral artery disease and controls No association
Patients with coronary artery disease Higher homocysteine in H. pylori-positive patients
Prospective study on CVD Higher homocysteine in H. pylori-positive patients
Patients with diabetes mellitus type 2 and controls Higher homocysteine in H. pylori-positive patients
Dyspeptic patients No change in homocysteine
Patients with low vitamin B12 and anemia Increase in serum cobalamin and improvement in
haematological variables
Vitamin B12 deficiency anemia Increase in serum cobalamin and improvement in
haematological variables
Low serum cobalamin, no anemia Increase in serum cobalamin

ciency occurs slowly due to the low requirement (2 Ìg/
day), the enterohepatic cycle of cobalamin and the liver
stores of the vitamin that have been built up during life
and that are about 2–3 mg of cobalamin by the age of 60
[3].
The classical sign of vitamin B12 deficiency is megalo-
blastic anaemia which, however, occurs in only 50% of
vitamin B12-deficient subjects. Other signs of vitamin B12
deficiency which are often overlooked are psychiatric and
neurodegenerative changes [4]. Diagnosis of vitamin B12
deficiency is usually made by low serum cobalamin con-
centrations which, however, has an only low diagnostic
sensitivity [5]. Other diagnostic tools to detect vitamin
B12 deficiency are the measurement of specific metabo-
lites such as homocysteine and methylmalonic acid [6, 7].
With the exception of homocysteine, these diagnostic

238 Dig Dis 2003;21:237–244 Dierkes/Ebert/Malfertheiner/Luley

tools require specialized equipment and are not offered
routinely.
Lack of intrinsic factor causing pernicious anaemia is
classically diagnosed by the Schilling test. This test, how-
ever, will not detect malabsorption of food-bound cobal-
amin, since free, crystalline vitamin B12 is used during the
Schilling test. Therefore, another test, the egg yolk cobal-
amin absorption test (EYCAT) has been introduced for
the diagnosis of malabsorption of food-bound cobalamin
[8]. However, this test is not widely used.
The interest in the association of H. pylori infection
with cobalamin status is further triggered by the associa-
tion of low cobalamin status with hyperhomocysteinaem-
ia which is a widely acknowledged risk factor of athero-
sclerosis [9]. The association between H. pylori – low
cobalamin – elevated homocysteine offers a metabolic
explanation why H. pylori infection is associated with
increased risk for cardiovascular disease, especially coro-
nary heart disease [10].
This review summarizes the present knowledge of the
association of H. pylori, cobalamin and homocysteine
concentrations. It considers studies published in English
until April 2003 (table 1).
Helicobacter and Pernicious Anaemia
If H. pylori indeed causes cobalamin deficiency, an
association of H. pylori with pernicious anaemia could
also be expected. This issue was investigated by a number
of studies in patients with pernicious anaemia which,
however, could not prove this hypothesis and even
showed a lower prevalence of H. pylori infection in
patients with pernicious anaemia.
Fong et al. [11] studied the frequency of H. pylori infec-
tion in 28 patients with pernicious anaemia and 28 age-
matched control subjects (mean age 59 B 14 years). Diag-
nosis of H. pylori infection was made by gastric biopsy
and serology. Positive biopsies were found in 3 of 28
patients with pernicious anaemia (11%) in contrast to 20
of 28 control subjects (71%). A similar result was reported
by Haruma et al. [12], who investigated 24 patients with
pernicious anaemia and 24 age-matched control subjects.
H. pylori status was diagnosed by biopsies and IgG anti-
bodies. No patient with pernicious anaemia but 67% of
control subjects were H. pylori-positive. A low frequency
of H. pylori infection in patients with macrocytic anaemia
in comparison to microcytic anaemia was also reported by
Marignani et al. [13]. H. pylori status was diagnosed by
biopsy and IgG antibodies. Patients with macrocytic
H. pylori Infection, Vitamin B12 and
Homocysteine
anaemia were older than those with microcytic anaemia,
had lower vitamin B12 levels and higher gastrin levels, but
only 2 of 44 patients were H. pylori-positive in contrast to
22 of 36 patients with microcytic anaemia.
Annibale et al. [14] investigated 150 patients with atro-
phic body gastritis. Patients were grouped according to
their H. pylori status and the prevalence of pernicious
anaemia. 37 patients were H. pylori-negative by both
serology and histology; 79 patients were negative by his-
tology but positive by serology and 34 patients were posi-
tive by both serology and histology. The lowest frequency
of pernicious anaemia was found in patients being posi-
tive by serology and histology (12%), an intermediate fre-
quency was found in the group with negative histology but
positive serology (46%) and the highest frequency of per-
nicious anaemia was found in the group negative for H.
pylori (76%).
Andres et al. [15] investigated the H. pylori infection
status in 60 elderly patients who had been selected by
cobalamin deficiency due to food-cobalamin malabsorp-
tion and who presented with polyneuropathy (35%), de-
mentia and confusion (30%) and anaemia (27%). Gastric
atrophy was documented in 19 out of 32 investigated
patients, but infection with H. pylori was documented in
only 1 of the patients.
The above cited studies suggest that the hypothesis
that H. pylori is a causative factor in pernicious anaemia
does not hold true and that H. pylori infection is even pro-
tective in the development of cobalamin-associated
anaemia. However, some other studies report that pa-
tients with pernicious anaemia have H. pylori infection
rates comparable to the normal population.
Annibale et al. [16] reported that 49 of 81 patients with
pernicious anaemia were positive for H. pylori infection
(60%), although only 8 showed a positive histological H.
pylori status and 41 were positive by serological investiga-
tion. A Turkish study revealed comparable results [17].
They investigated 138 patients with vitamin B12 deficien-
cy and anaemia which was not due to intrinsic factor defi-
ciency or gastrectomy. Diagnosis of H. pylori infection
was confirmed by rapid urease testing and gastric biopsies
in 77 of 138 patients (56%). Another study from Turkey
[18] reported that among 421 patients with vitamin B12
deficiency, H. pylori infection was detected in 241 of them
(57%). A small Swedish study [19] involving 30 patients
with pernicious anaemia reported that 25 of them had
antibodies against H. pylori (83%).
Thus, so far, studies revealed heterogeneous data on
the association of H. pylori infection and cobalamin-relat-
ed anaemia. However, it has been suggested that the low
Dig Dis 2003;21:237–244 239
frequency of H. pylori infection in total gastric atrophy
can be explained by the fact that infected subjects become
Helicobacter-negative as a result of the virtual absence of
acid production which inhibits survival of H. pylori [20].
This would mean that the low frequency of H. pylori
infection reported by some of the studies reflects a false-
negative result. As all studies on this issue had a cross-
sectional design, a definitive answer cannot be given at
present. Longitudinal studies, however, are difficult to
perform and are ethically questionable.
Helicobacter and Subclinical Cobalamin
Deficiency
Vitamin B12 deficiency is rare in young adults and has
an increasing prevalence in the elderly. Development of
vitamin B12 requires years and depends mainly on pres-
ence of achlorhydria and on whether intrinsic factor is
present or absent, since the intrinsic factor is necessary to
maintain the enterohepatic circle of cobalamin. Thus,
hypo- and achlorhydria which compromise food-cobal-
amin absorption must be present for long time before
vitamin B12 deficiency occurs. Furthermore, subclinical
vitamin B12 deficiency is difficult to diagnose as haemato-
logical changes are rare in this stage. Therefore, it is main-
ly diagnosed by low serum cobalamin or metabolic
changes as elevated concentrations of the metabolites
homocysteine or methylmalonic acid. Studies have also
investigated whether food-cobalamin malabsorption is as-
sociated with H. pylori infection and whether the degree
of gastric atrophy can be linked to cobalamin deficiency.
Van Asselt et al. [21] investigated the determinants of
cobalamin deficiency in healthy elderly, but did not find a
statistically significant association of H. pylori infection
with mild cobalamin deficiency. The study included 105
healthy elderly Dutch subjects (age range 74–80 years)
who were classified according to cobalamin and methyl-
malonic acid concentrations as ‘cobalamin-deficient’ (n =
25), ‘possibly cobalamin-deficient’ (n = 53) or ‘not cobal-
amin-deficient’ (n = 27). H. pylori IgG antibodies were
found in 54, 68 and 44% of the respective groups. Thus, it
was concluded that both H. pylori infection and mild
cobalamin deficiency was common in this cohort, but H.
pylori infection did not contribute significantly to cobal-
amin deficiency.
Also, other studies did not find a difference in cobal-
amin levels between H pylori-positive and -negative pa-
tients. Pilott et al. [22] reported no effect of H. pylori
infection on nutritional indices including serum cobala-
240 Dig Dis 2003;21:237–244
min and folate in 102 elderly. H. pylori antibodies were
found in 84% of the elderly.
Cenerelli et al. [23] investigated patients with diabetes
mellitus type 2 and controls and found no effect of H.
pylori infection on the level of cobalamin. Even higher
cobalamin levels were reported by Bloemenkamp et al.
[24] in H. pylori-positive subjects in a study including
women with peripheral artery disease and controls. Since
the latter two studies also investigated homocysteine, they
are described in detail below.
On the other hand, a significant association of H. pylori
infection and prevalence of low cobalamin concentrations
was reported by Shuval-Sudai and Granot [25]. 133
patients were investigated in Israel for H. pylori infection
and cobalamin and folate status. Patients were enrolled in
a community primary care clinic, were between 18 and 90
years of age, and were eligible for the study if they had no
history of H. pylori eradication, previous gastrointestinal
disease or surgery, vegetarian diet or multivitamin use. H.
pylori status was assessed by IgG antibodies. A signifi-
cantly higher frequency of low cobalamin levels (cut-off
!250 pg/ml) was found in H. pylori-positive patients (n =
96). Similar results were obtained with the lower cut-off
level of 180 or 145 pg/ml. Interestingly, lower folate levels
were observed in the H. pylori-positive patients.
A similar result was reported by Tamura et al. [26] who
investigated 93 patients with coronary artery disease and
observed both lower cobalamin and folate levels in pa-
tients positive for H. pylori. H. pylori infection was diag-
nosed by biopsies. In this study, homocysteine was also
investigated (see below).
Reasons for differences in cobalamin status due to H.
pylori infection may be that H. pylori infection affects the
absorption of either food-cobalamin or free cobalamin
which can be studied employing the EYCAT test or the
Schilling test, respectively. In particular, food-cobalamin
malabsorption has been investigated with respect to H.
pylori infection.
The contribution of different factors to food-cobala-
min malabsorption was studied in a small group of 19 vol-
unteers by Cohen et al. [27]. Volunteers had been selected
on the basis of age and their cobalamin levels to ensure a
wide range of deficient and replete subjects. Food-cobal-
amin absorption was tested using the EYCAT. H. pylori
status was determined by biopsies and 13C-urea breath
tests. Gastric function was investigated by basal and pen-
tagastrin-stimulated gastric aspirates, and gastric inflam-
mation and atrophy were scored on scales ranging from 0
to 3 and 1 to 4, respectively. Six of the 19 patients had
severe food cobalamin malabsorption. Three of them also
Dierkes/Ebert/Malfertheiner/Luley
had H. pylori infection and antral gastritis while the other
3 patients did not have H. pylori infection but corpus atro-
phy with achlorhydria.
The issue whether H. pylori infection affects food-cobal-
amin malabsorption was further studied by Carmel et al.
[28]. Data of 202 subjects who underwent the EYCAT test
in their clinic, among them 43 healthy volunteers and 159
patients with low cobalamin levels, were analysed retro-
spectively. H. pylori infection was confirmed either by the
presence of IgG antibodies or the 13C-urea breath test.
Determination of H. pylori status was done in 167 subjects
and in 133 subjects, serum pepsinogen I and II were mea-
sured, while serum gastrin and parietal cell antibodies were
assayed in 158 subjects. Severe food-cobalamin malabsorp-
tion was found in 24% of the patients with low cobalamin
levels and in 9% of the healthy volunteers. The mean excre-
tion rate of radiolabelled cobalamin during the EYCAT
was significantly lower in H. pylori-positive subjects than in
H. pylori-negative subjects (2.26 B 1.90% vs. 3.43 B
2.22%). H. pylori infection was associated with presence of
severe food-cobalamin malabsorption (78% H. pylori-posi-
tive subjects vs. 50% H. pylori-positive subjects in mild
food-cobalamin malabsorption and 44% H. pylori-positive
subjects with normal absorption). In a multivariate regres-
sion analysis using data of 136 subjects, ethnic origin, age,
H. pylori status, and serum gastrin contributed indepen-
dently to the EYCAT result.
Thus, present studies support the view that H. pylori
infection interferes with food-cobalamin absorption, al-
though results by Cohen et al. [27] suggest that different
mechanisms may be involved. Further studies investi-
gated whether cobalamin status is related to gastric atro-
phy or inflammation. Serin et al. [29] selected 145 pa-
tients from a gastroenterological clinic who did not show
signs of atrophy, erosions, or ulcers. Biopsy specimen of
the gastric antrum and corpus were taken and analysed
for signs of inflammation, gastritis and H. pylori density.
Vitamin B12 was measured in serum. Histopathological
scores for both antral and corpus H. pylori density and
inflammation were significantly inversely associated with
serum vitamin B12 levels. In multivariate analyses, only
H. pylori density was associated with vitamin B12 levels.
Is Helicobacter Infection Related to
Hyperhomocysteinaemia?
Although the role for H. pylori in the development of
cobalamin deficiency is not clear, it offers an attractive
explanation why H. pylori infection could be associated
H. pylori Infection, Vitamin B12 and
Homocysteine
with hyperhomocysteinaemia. A further link may exist
due to the observed effect of H. pylori on folate status.
Both vitamin B12 and folate status are important for
maintenance of low homocysteine concentrations. Pres-
ent evidence suggests that elevated homocysteine concen-
trations are independently associated with increased risk
for cardiovascular disease. Thus, hyperhomocysteinaem-
ia in patients with H. pylori infection may be one mecha-
nism by which H. pylori is associated with coronary heart
disease.
However, an association of H. pylori infection and
homocysteine concentrations was not reported in a num-
ber of studies. Saxena et al. [30] investigated homocys-
teine in 220 subjects (mean age 66 B 10 years). They
reported no differences in homocysteine levels in patients
with positive serology for H. pylori (n = 122) compared to
those with no antibodies against H. pylori. The result
remained the same after analyzing men and women sepa-
rately. Interestingly, the values reported by the authors
are remarkably high with a mean homocysteine value of
22.7 B 11 Ìmol/l.
A similar result was reported by Leung et al. [31] who
investigated the association between H. pylori and homo-
cysteine in 49 patients with dyspepsia referred for endos-
copy. Patients were relatively young (median age 43
years), and 37 were found to be H. pylori-positive by gas-
tric biopsies. There was no difference in homocysteine in
infected compared to non-infected subjects and no asso-
ciation of homocysteine to gastric histology or H. pylori
density.
A Japanese study [32] included 45 healthy middle-
aged subjects infected with H. pylori and 45 age-matched
controls without infection. The mean age was 39 years in
both groups. H. pylori infection was diagnosed by 13C-
urea breath test and serology. No difference was found in
plasma homocysteine between patients and controls.
None of these studies investigated cobalamin or folate sta-
tus.
No difference in homocysteine according to H. pylori
status was found in 150 women with peripheral artery dis-
ease and 412 control women [24]. H. pylori was diagnosed
by positive IgG antibody test. The frequency of H. pylori
infection in PAD patients was 42% in comparison to 27%
in control subjects, resulting in an odds ratio for PAD of
2.0 (1.3–2.9). Homocysteine was significantly higher in
patients than in controls, however, adjustment for homo-
cysteine did not alter substantially the association be-
tween H. pylori and PAD. In this study, folate and vita-
min B12 levels were also measured. No difference was
found in homocysteine and folate levels between H. pylo-
Dig Dis 2003;21:237–244 241
ri-positive and -negative patients and even higher cobal-
amin levels were measured in H. pylori-positive subjects.
The authors conclude that homocysteine cannot explain
the association between H. pylori infection and athero-
sclerosis. It has to be noted that the blood sample for anal-
yses was taken between 1998 and 2000, while the diagno-
sis of PAD was made between 1990 and 1999.
In contrast to the above-mentioned reports on lack of
association between H. pylori and homocysteine concen-
trations, other investigators reported significantly higher
levels of homocysteine in H. pylori-positive patients. Ta-
mura et al. [26] studied homocysteine, related vitamins
and H. pylori infection in 93 Japanese patients who
underwent coronary arteriography. The mean age of the
patients was 64 B 9 years, and 57 of them were found to
be H. pylori-positive as diagnosed by stomach biopsies.
Significantly lower cobalamin and folate levels and higher
levels of homocysteine were found in H. pylori-positive
patients than in H. pylori-negative patients. Homocys-
teine was weakly, but significantly related to the atrophic
score of the stomach of the patients.
Whincup et al. [33] reported higher homocysteine lev-
els both in H. pylori-positive patients with coronary heart
disease and in healthy controls which have been prospec-
tively investigated in the British Regional Heart Study.
However, the effect of H. pylori infection was, although
significantly, not very strong and increased the homocys-
teine modestly. No vitamin levels were reported in this
study.
Cenerelli et al. [23] investigated homocysteine in pa-
tients with diabetes type 2 and healthy controls (n = 30
and 43, respectively) and found no difference between
patients and controls. However, subjects with H. pylori
infection showed significantly higher homocysteine con-
centrations. The diagnosis of H. pylori was made by 13C-
urea breath test. There was no difference in folate or
cobalamin between H. pylori-positive and -negative sub-
jects.
Does H. pylori Eradication Affect Cobalamin
Status and Homocysteine?
There are a few studies that investigate the effect of H.
pylori eradication therapy on homocysteine or cobalamin
levels. The only study that investigated homocysteine was
reported by Leung et al. [31] who treated 37 dyspeptic
patients. Eradication was confirmed by a negative 13C-
urea breath test. After 24 weeks, homocysteine was mea-
sured again in these patients. There was no significant
242 Dig Dis 2003;21:237–244
change in homocysteine concentration after eradication
(10.5 vs. 10.2 Ìmol/l).
A number of studies reported a spontaneous improve-
ment of vitamin B12 status after H. pylori eradication,
without specific treatment with cobalamin. Serin et al.
[29] investigated 65 of 145 patients after eradication ther-
apy. At baseline, 145 patients with no gastric atrophy, ero-
sion or ulcers were included. Patients were then given an
eradication therapy and 65 of the patients were re-ana-
lyzed 2–3 months later. Baseline cobalamin levels were
indicative for cobalamin deficiency although no anaemia
was present. In 33 of these 65 patients, eradication thera-
py was successful. The authors observed a significant
increase in vitamin B12 after eradication therapy with a
more pronounced increase in those with complete eradi-
cation.
Avcu et al. [18] investigated 108 H. pylori-infected
patients where H. pylori was also present in dental plaques
and who suffered from vitamin B12 deficiency and from
signs of anaemia. Mean patient age was 44 years. After
eradication therapy, gastric biopsies were repeated in 61
patients and confirmed the successful eradication thera-
py. Haematological parameters (red blood cell count, hae-
moglobin, haematocrit and MCV) improved significantly
as well as vitamin B12 levels which increased from 71 B
11 to 293 B 11 pg/ml. Mean follow-up time was 15 B 1
months and no treatment for vitamin B12 deficiency was
given.
A similar effect of eradication therapy was observed by
Kaptan et al. [17]. Eradication therapy was given to 77
patients with a positive H. pylori biopsy who were also
anaemic and had low cobalamin levels. In 31 patients, H.
pylori eradication therapy was successful and the cobal-
amin level and the blood count were re-investigated after
a mean follow-up time of 42 B 19 months. There was a
significant improvement in haematocrit, and MCV, and a
significant increase in serum cobalamin at re-investiga-
tion (63 B 30 to 223 B 38 pmol/l). Patients with unsuc-
cessful eradication therapy received cyanocobalamin
therapy, but results were not reported.
Discussion
Although the hypothesis on the association of H. pylori
infection, cobalamin and homocysteine status is clear and
intriguing, the results of the clinical studies on this asso-
ciation revealed highly disparate results and do not clearly
support the hypothesis. On the other hand, there is evi-
dence that cobalamin status improves after H. pylori erad-
Dierkes/Ebert/Malfertheiner/Luley
ication suggesting a causal effect of H. pylori in the devel- teine concentrations. Lower folate concentrations in H.
opment of cobalamin deficiency. pylori-positive patients have been shown in two of four
The different results on H. pylori and cobalamin status studies that also reported folate concentrations. Dimin-
may depend on study design, selection of patients and ished folate absorption has been shown in patients with
variables measured. Cobalamin status in the studies var- atrophic gastritis [35], but the effect of H. pylori infection
ies widely with pernicious anaemia as the ‘tip of the ice- has not been investigated in particular.
berg’ in cobalamin deficiency, and milder forms of cobal- In conclusion, clinical studies do not show a uniform
amin deficiency that are not always clearly defined. It has association of H. pylori infection with cobalamin or folate
to be considered that many patients with cobalamin mal- deficiency or elevated homocysteine concentrations.
absorption do not develop pernicious anaemia at all or Therefore, routine determination of cobalamin or homo-
only after decades. H. pylori infection obviously is associ- cysteine in H. pylori-infected patients as a means of pre-
ated with food-cobalamin malabsorption, but the associa- vention of deficiency or with respect to atherosclerotic
tion with cobalamin deficiency due to intrinsic factor risk cannot be recommended at present. On the other
deficiency was not demonstrated. Furthermore, the low hand, the studies do suggest that H. pylori infection has a
prevalence of H. pylori infection in pernicious anaemia is role in atrophic gastritis and food-cobalamin malabsorp-
a striking phenomenon that may be explained by the pos- tion. Subgroups of infected patients are obviously at risk
sibility that infected subjects become Helicobacter-nega- of developing cobalamin deficiency and hyperhomocys-
tive as a result of the virtual absence of acid production teinaemia, who are not identified at present. Before a con-
which inhibits survival of H. pylori [20]. cise recommendation can be given, more data in patient
Most studies investigated the effect of H. pylori on groups with a well-defined cobalamin status are needed.
serum cobalamin levels. However, serum cobalamin has
been shown to have limited sensitivity to detect deficien-
cy since many patients still have low-to-normal serum Acknowledgement
concentrations [6]. Instead, diagnosis of subclinical cobal-
M. Ebert is supported by the Heisenberg-Programme of the DFG
amin deficiency should be based on metabolic markers
(Eb 187/5-1).
like methylmalonic acid or homocysteine [5]. Most stud-
ies, however, did not use these additional diagnostic
tools.
How can the increase of vitamin B12 after eradication
therapy be explained? The only study that also measured
homocysteine after eradication did not find an effect of
eradication on homocysteine [31]. Three studies reported
on vitamin B12 response to eradication therapy. None of
these studies reported individual data and in each case, a
substantial portion of patients was missed during follow-
up. Thus, it can be questioned whether the effect may be
influenced by regression to the mean or selection bias.
However, it is possible that there is a real effect since
reversal of food-cobalamin malabsorption in atrophic gas-
tritis after tetracycline therapy was reported earlier [34].
A likely mechanism is the reduction of bacteria in the
stomach and small intestine in hypo- or achlorhydria.
This finding supports the view that H. pylori infection
may also play a role in food-cobalamin malabsorption.
In addition to the association of H. pylori to cobala-
min, studies on the association of H. pylori to homocys-
teine revealed also inconsistent results. A significant ef-
fect of H. pylori infection on plasma homocysteine was
found in three of six studies. Besides cobalamin deficien-
cy, folate deficiency is also associated to higher homocys-

H. pylori Infection, Vitamin B12 and Dig Dis 2003;21:237–244 243

Homocysteine
 References
1 Suerbaum S, Michetti P: Helicobacter pylori
infection. N Engl J Med 2002;347:1175–1186.
2 Parsonnet J, Friedman GD, Vandersteen DP,
Chang Y, Vogelman JH, Orentreich N, Sibley
RK: Helicobacter pylori infection and the risk
of gastric carcinoma. N Engl J Med. 1991;
325(16):1127–1131.
3 Carmel R: Cobalamin, the stomach and aging.
Am J Clin Nutr 1997;66:750–759.
4 Nexo E, Hansen M, Rasmussen K, Lindgren A,
Grasbeck R: How to diagnose cobalamin defi-
ciency. Scand J Clin Lab Invest Suppl 1994;
219:61–76.
5 Stabler SP, Lindenbaum J, Allen RH: The use
of homocysteine and other metabolites in the
specific diagnosis of vitamin B12 deficiency. J
Nutr 1996;126(suppl):1266–1272.
6 Bolann BJ, Solli JD, Schneede J, Grottum KA,
Loraas A, Stokkeland M, Stallemo A, Schjoth
A, Bie RB, Refsum H, Ueland PM: Evaluation
of indicators of cobalamin deficiency defined
as cobalamin-induced reduction in increased
serum methylmalonic acid. Clin Chem 2000;
46:1744–1750.
7 Holleland G, Schneede J, Ueland PM, Lund
PK, Refsum H, Sandberg S: Cobalamin defi-
ciency in general practice. Assessment of the
diagnostic utility and cost-benefit analysis of
methylmalonic acid determination in relation
to current diagnostic strategies. Clin Chem
1999;45:189–198.
8 Doscherholmen A, Ripley D, Chang S, Silvis
SE: Influence of age and stomach function on
serum vitamin B12 concentration. Scand J Gas-
troenterol 1977;12:313–319.
9 Graham IM, O’Callaghan P: Vitamins, homo-
cysteine and cardiovascular risk. Cardiovasc
Drugs Ther 2002;16:383–389.
10 Pellicano R, Oliaro E, Gandolfo N, Aruta E,
Mangiardi L, Orzan F, Bergerone S, Rizzetto
M, Ponzetto A: Ischemic cardiovascular dis-
ease and Helicobacter pylori. Where is the link?
J Cardiovasc Surg (Torino) 2000;41:829–833.
11 Fong TL, Dooley CP, Dehesa M, Cohen H,
Carmel R, Fitzgibbons PL, Perez-Perez GI,
Blaser MJ: Helicobacter pylori infection in per-
nicious anemia: A prospective controlled
study. Gastroenterology 1991;100:328–332.
12 Haruma K, Komoto K, Kawaguchi H, Okamo-
to S, Yoshihara M, Sumii K, Kajiyama G: Per-
nicious anaemia and Helicobacter pylori infec-
tion in Japan: Evaluation in a country with a
high prevalence of infection. Am J Gastroen-
terol 1995;90:1107–1110.
13 Marignani M, Delle Fave G, Mecarocci S, Bor-
di C, Angeletti S, D’Ambra G, Aprile MR, Cor-
leto VD, Monarca B, Annibale B: High preva-
lence of atrophic body gastritis in patients with
unexplained microcytic and macrocytic
anaemia: a prospective screening study. Am J
Gastroenterol 1999;94:766–772.
244 Dig Dis 2003;21:237–244
14 Annibale B, Negrini R, Caruana P, Lahner E,
Grossi C, Bordi C, Delle Fave G: Two-thirds of
atrophic body gastritis patients have evidence
of Helicobacter pylori infection. Helicobacter
2001;6:225–233.
15 Andres E, Noel E, Kaltenbach G, Perrin AE,
Vinzio S, Goichot B, Schlienger JL, Blickle JF:
Vitamin B12 deficiency with normal Schilling
test or non-dissociation of vitamin B12 and its
carrier proteins in elderly patients. A study of
60 patients (in French). Rev Méd Interne 2003;
24:218–223.
16 Annibale B, Lahner E, Bordi C, Martino G,
Caruana P, Grossi C, Negrini R, Delle Fave G:
Role of Helicobacter pylori infection in perni-
cious anaemia. Dig Liver Dis 2000;32:756–
762.
17 Kaptan K, Beyan C, Ural AU, Cetin T, Avcu F,
Gulsen M, Finci R, Yalcin A: Helicobacter
pylori – Is it a novel causative agent in vitamin
B12 deficiency? Arch Intern Med 2000;160:
1349–1353.
18 Avcu N, Avcu F, Beyan C, Ural AU, Kaptan K,
Ozyurt M, Nevruz O, Yalcin A: The relation-
ship between gastric-oral Helicobacter pylori
and oral hygiene in patients with vitamin B12-
deficiency anemia. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 2001;92:166–169.
19 Ma JY, Borch K, Sjostrand SE, Janzon L,
Mardh S: Positive correlation between H,K-
adenosine triphosphatase autoantibodies and
Helicobacter pylori antibodies in patients with
pernicious anaemia. Scand J Gastroenterol
1994;29:961–965.
20 Kuipers EJ, Klinkenberg-Knol EC, Vanden-
broucke-Grauls CM, Appelmelk BJ, Schenk
BE, Meuwissen SG: Role of Helicobacter pylori
in the pathogenesis of atrophic gastritis. Scand
J Gastroenterol Suppl 1997;223:28–34.
21 Van Asselt DZ, de Groot LC, van Staveren
WA, Blom HJ, Wevers RA, Biemond I, Hoef-
nagels WH: Role of cobalamin intake and atro-
phic gastritis in mild cobalamin deficiency in
older Dutch subjects. Am J Clin Nutr 1998;68:
328–334.
22 Pilott A, Fabrello R, Franceschi M, Scagnelli
M, Soffiati F, Di Mario F, Fortunato A, Valerio
G: Helicobacter pylori infection in asymptom-
atic elderly subjects living at home or in a nurs-
ing home: Effects on gastric function and nutri-
tional status. Age Aging 1996;25:245–249.
23 Cenerelli S, Bonazzi P, Galeazzi R, Testa I,
Bonfigli AR, Sirolla C, Giunta S, Galeazzi L,
Fumelli D, Testa R: Helicobacter pylori masks
differences in homocysteine plasma levels be-
tween controls and type 2 diabetic patients. Eur
J Clin Invest 2002;32:158–162.
24 Bloemenkamp DG, Mali WP, Tanis BC, Ro-
sendaal FR, van den Bosch MA, Kemmeren
JM, Algra A, Visseren FL, van der Graaf Y:
The relation between Helicobacter pylori and
atherosclerosis cannot be explained by a high
homocysteine concentration. Eur J Clin Invest
2002;32:549–555.
25 Shuval-Sudai O, Granot E: An association be-
tween Helicobacter pylori infection and serum
vitamin B12 levels in healthy adults. J Clin Gas-
troenterol 2003;36:130–133.
26 Tamura A, Fujioka T, Nasu M: Relation of
Helicobacter pylori infection to plasma vitamin
B12, folic acid and homocysteine levels in pa-
tients who underwent diagnostic coronary arte-
riography. Am J Gastroenterol 2002;97:861–
866.
27 Cohen H, Weinstein WM, Carmel R: Hetero-
geneity of gastric histology and function in food
cobalamin malabsorption: Absence of atrophic
gastritis and achlorhydria in some patients
with severe malabsorption. Gut 2000;47:638–
645.
28 Carmel R, Aurangzeb I, Qian D: Associations
of food-cobalamin malabsorption with ethnic
origin, age, Helicobacter pylori infection, and
serum markers of gastritis. Am J Gastroenterol
2001;96:63–70.
29 Serin E, Gumurdulu Y, Ozer B, Kayaselcuk F,
Yilmaz U, Kocak R: Impact of Helicobacter
pylori on the development of vitamin B12 defi-
ciency in the absence of gastric atrophy. Helico-
bacter 2002;7:337–341.
30 Saxena V, Markus H, Swaminathan S, Mendall
ME: Hyperhomocysteinaemia, Helicobacter
pylori and coronary heart disease. Heart 1997;
78:524.
31 Leung WK, Ma PK, Choi PC, Ching JY, Ng
AC, Poon P, Woo KS, Sung JJ: Correlation
between Helicobacter pylori infection, gastric
inflammation and serum homocysteine con-
centration. Helicobacter 2001;6:146–150.
32 Yoshino N, Adachi K, Takashima T, Miyaoka
Y, Yuki T, Ishihara S, Kinoshita Y: Helicobac-
ter pylori infection does not affect the serum
level of homocysteine. Am J Gastroenterol
2002;97:2927–2928.
33 Whincup P, Danesh J, Walker M, Lennon L,
Thomson A, Appleby P, Hawkey C, Atherton J:
Prospective study of potentially virulent strains
of Helicobacter pylori and coronary heart dis-
ease in middle-aged men. Circulation 2000;
101:1647–1652.
34 Suter PM, Golner BB, Goldin BR, Morrow FD,
Russell RM: Reversal of protein-bound vita-
min B12 malabsorption with antibiotics in atro-
phic gastritis. Gastroenterology 1991;101:
1039–1045.
35 Russell RM, Krasinski SD, Samloff IM, Jacob
RA, Hartz SC, Brovender SR: Folic acid mal-
absorption in atrophic gastritis. Possible com-
pensation by bacterial folate synthesis. Gastro-
enterology 1986;91:1476–1482.
Dierkes/Ebert/Malfertheiner/Luley
 Original Paper

Dig Dis 2003;21:245–251

DOI: 10.1159/000073342

Prevalence of Malnutrition in
Hospitalized Medical Patients: Impact of
Underlying Disease
Matthias Pirlich a Tatjana Schütz a Martin Kemps a Niklas Luhman a
Gerd-Rüdiger Burmester b Gert Baumann c Mathias Plauth d
Heinrich Josef Lübke e Herbert Lochs a
a Medizinische Klinik mit Schwerpunkt Gastroenterologie, Hepatologie und Endokrinologie, b Rheumatologie und

klinische Immunologie, c Kardiologie, Pulmonologie und Angiologie, Universitätsklinikum Charité,

Humboldt-Universität zu Berlin; d Klinik für Innere Medizin, Städtisches Klinikum Dessau und e Medizinische Klinik I,
Krankenhaus Zehlendorf, Berlin, Deutschland

Key Words diseases. Patients with gastrointestinal diseases, how-

Malnutrition, prevalence W Malnourished hospitalized
patients W Subjective global assessment W
Gastrointestinal diseases W Malnutrition, diagnoses
Abstract
Background/Aims: Malnutrition is common among hos-
pitalized patients. We investigated whether certain dis-
eases predispose more frequently for malnutrition than
others. Methods: Nutritional state was assessed by clini-
ever, were not more frequently malnourished than other
medical patients (28.8 vs. 22.0%). Malnourished patients
were significantly older (70.0 B 13.6 vs. 58.3 B 15.6
years, p ! 0.0001) and had a 40% longer hospital stay
(13.1 B 8.1 vs. 9.3 B 6.8 days, p ! 0.0001) than well-
nourished patients. Conclusions: Patients with malig-
nancies, inflammatory bowel disease, chronic heart fail-
ure and benign lung diseases need special attention due
to the high prevalence of malnutrition.
Copyright © 2003 S. Karger AG, Basel

cal scores, anthropometry and bioimpedance analysis in

502 consecutively admitted patients in the departments
of internal medicine in two hospitals in Berlin (n = 300,
university hospital; n = 202, district hospital). The preva-
lence of malnutrition was compared in patient groups
with a different diagnosis. Results: Malnutrition was
present in 24.2% of all patients. A clear association
between diagnoses and malnutrition was found: the
prevalence of malnutrition was significantly higher in
malignant than in non-malignant diseases (50.9 vs.
21.0%, p ! 0.0001). High prevalence rates 1 30% were
observed in subgroups of patients with inflammatory
bowel diseases, chronic heart failure and benign lung
Introduction
Malnutrition is a frequent finding in hospitalized pa-
tients [1–7]. Data on the prevalence of malnutrition in
these patients, however, differ considerably according to
the population investigated and the definitions used.
From a number of studies performed during the last 30
years, predominantly in the US and UK, it can be esti-
mated that about 20–50% of all medical and surgical
patients admitted to hospitals show signs of malnutrition
[8]. Malnutrition has been found to be associated with

© 2003 S. Karger AG, Basel Dr. med. Matthias Pirlich

ABC 0257–2753/03/0213–0245$19.50/0 Medizinische Klinik und Poliklinik
Fax + 41 61 306 12 34 Universitätsklinikum Charité, Humboldt-Universität zu Berlin
E-Mail karger@karger.ch Accessible online at: DE–10098 Berlin (Germany)
www.karger.com www.karger.com/ddi Tel. +49 30 450 514006, Fax +49 30 450 514923, E-Mail matthias.pirlich@charite.de
Table 1. Participating specialities Table 2. Distribution of diagnoses

n Gender, m/f Age (mean B SD) Diagnoses n %

University hospital Gastrointestinal diseases

Gastroenterology 100 48/52 56.9B16.3 Chronic liver disease 45 9.0
Cardiology 100 74/26 62.0B12.0 Malignancies of liver, pancreas or bile ducts 22 4.4
Rheumatology 100 37/63 52.2B15.5 Gastrointestinal malignancies 17 3.4
Community hospital Gastrointestinal bleeding 16 3.2
Gastroenterology 101 38/63 65.0B15.7 Benign pancreatic or biliary diseases 13 2.6
Cardiology 101 39/62 69.6B14.4 Inflammatory bowel disease 10 2.0
All patients 502 236/266 61.2B16.0 Reflux disease 10 2.0
Other gastroenterology 8 1.6
Gastroenteritis (without salmonella infection) 7 1.4
Other benign intestinal diseases (colonic polyps,
IBS, diverticulosis) 8 1.6
increased morbidity [2, 4], with prolonged hospital stay at All gastrointestinal diseases 156 31.1
substantial extra cost of health care [7, 9, 10], and with
Other internal diseases
increased mortality especially in elderly patients [3, 5,
Coronary heart disease 85 16.9
11]. The association between malnutrition and poor sur- Arrhythmia 40 8.0
vival is especially established in cancer patients [7, 8, 12]. Rheumatoid arthritis 25 5.0
However, despite an increasing number of studies in oth- Heart failure 24 4.8
er countries, the influence of diagnoses on the develop- Systemic infection 20 4.0
Valvular defect 16 3.2
ment of malnutrition in medical patients is unclear. It has
Vasculitis/myositis 15 3.0
been suggested that patients with digestive diseases are at Extraintestinal malignancies 15 3.0
higher risk of developing malnutrition than patients with Systemic lupus erythematodes 13 2.6
other internal diseases [4, 7]. Cerebral ischemia 13 2.6
The primary objectives of this study were (1) to esti- All other diseases 13 2.6
Benign lung diseases 13 2.6
mate the overall prevalence of malnutrition among medi-
Other rheumatology 12 2.4
cal patients and (2) to evaluate the impact of diagnoses on Other cardiology 11 2.2
the prevalence of malnutrition. The study was performed Dermatosclerosis 9 1.8
in two hospitals with different settings (one university Diabetes mellitus 9 1.8
hospital and one community hospital) in the city of Ber- Urology/kidney diseases 7 1.4
Other collagenoses 6 1.2
lin, Germany.
All other internal diseases 346 68.9

Patients and Methods

Patients
The study protocol was approved by the Ethics Committee of the
Universitätsklinikum Charité. The nutritional state of 502 patients
consecutively admitted to different specialities of internal medicine
was studied in two different hospitals: 300 patients were included in
the University Hospital Charité (specialities: gastroenterology (n =
100), cardiology (n = 100), and rheumatology (n = 100)), and 202
patients were included in the District Hospital of Zehlendorf (spe-
cialities: gastroenterology (n = 101) and cardiology (n = 101)).
Patients were considered eligible for entry if they were over the
age of 18, were assumed to stay longer than 2 days, and were willing
and able to give written informed consent. Patients admitted to day
care units or for observation after endoscopic or other invasive treat-
ment and those admitted to intensive care units were excluded. The
number of patients in each speciality, the distribution of gender and
the mean age are given in table 1. Patients in the community hospital
were slightly older than patients in the university hospital.
For further classification of patients according to the different
diseases, the main diagnoses determined at the time of discharge or
referral from the speciality were used. Since there was an overlap of
diagnoses between the different specialities, the data analysis of the
nutritional state was not performed according to the speciality but to
the main diagnoses of the patients. The distribution of diagnoses is
given in table 2. 156 patients (31.1%) had diseases of the gastrointes-
tinal tract, and 346 patients (68.9% of all patients) had other internal
diseases.
Assessment of Nutritional State
Two investigators (N.L. and M.K) were trained by the principle
investigator (M.P.) in performing the nutritional assessment. The
nutritional state of the patients was assessed on the day of hospital
admission according to clinical scores, anthropometric measure-
ments, and to the results of bioelectrical impedance analysis. The
subjective global assessment was used as the main criterion for the
classification of malnutrition.

246 Dig Dis 2003;21:245–251 Pirlich/Schütz/Kemps/Luhman/Burmester/

Baumann/Plauth/Lübke/Lochs
 Subjective Global Assessment (SGA)
The SGA was established by Detsky et al. [13] and relies primari-
ly on physical signs of malnutrition (loss of subcutaneous fat or mus-
cle mass, edema, ascites) and the patient’s history regarding weight
loss, dietary intake, gastrointestinal symptoms, functional capacity,
and the disease and its relation to nutritional requirements. Each
patient was classified as either well nourished (SGA A), moderately
or suspected of being malnourished (SGA B) and severely malnour-
ished (SGA C). The SGA requires only a few minutes by a trained
clinician. Its validity to indicate malnutrition-associated risks of
poor outcome has been proven in a number of studies [4, 5, 7, 12, 14].
Since subgroups of patients classified as SGA C were too small, com-
parative analyses were performed between malnourished patients
classified as either SGA B or C and well-nourished patients classified
as SGA A.
Nutritional Risk Index (NRI)
The NRI was developed by Buzby et al. [15] and is based on the
serum albumin concentration and the ratio of actual to usual weight:
NRI = 1.489 ! serum albumin (g/l) ! 41.7 ! (present weight/usual
weight). Calculated values 1 100 indicate good nutritional state,
97.5–100 mild malnourishment, 83.5 to ! 97.5 moderate malnour-
ishment, and ! 83.5 severe malnourishment. In our analysis, we used
values ! 97.5 for classification of malnutrition.
Bioelectrical Impedance Analysis (BIA)
BIA was performed by the whole-body tetrapolar contact elec-
trode approach applying an alternating electric current of 800 ÌA at
50 kHz (BIA 2000-M, Data Input GmbH, Frankfurt am Main, Ger-
many). Two pairs of current-introducing and voltage-sensing elec-
trodes were attached to the dorsum of hand and foot of the dominant
side of the body [16]. Resistance (R), reactance (Xc) and the phase
angle (·) were measured. All impedance measurements were taken
under standardized conditions [17]. Patients with implanted cardiac
pacemaker or defibrillator were excluded from BIA measurements
because of possible interactions with even small electrical currents.
Total body water was calculated as 0.69 ! H2/R + 0.8 [16] and fat
free mass = total body water/0.732. Body cell mass (BCM) was calcu-
lated [18] as BCM = fat free mass ! 0.29 ! ln (·). According to
Süttmann [19], a BCM ! 30% of body weight was considered as an
indicator of protein malnutrition.
Anthropometry
Body height was measured without shoes to the nearest 0.5 cm
with a stadiometer. Weight was measured using calibrated Seca chair
scales and compared with the body weight 6 months prior to admis-
sion to calculate weight loss. Midarm circumference was measured
with a tape measure. Triceps skinfold thickness was measured using a
Lange caliper (Holtain Ltd, Crymych, Dyfed, UK). The average of
three measurements at each site was used for the calculation of arm
muscle area and arm fat area according to Gurney and Jelliffe [20].
Measured arm muscle and arm fat area were compared with refer-
ence data [21], and values below the 10th percentile were considered
as an indicator of malnutrition.
Albumin
Serum albumin was measured in the clinical chemistry laboratory
by an automated analyzer using standard procedures, and values
! 3.5 g/dl were considered as indicators of impaired protein synthe-
sis.
Hospital Malnutrition
Table 3. Clinical characterization of the study population
Gastrointestinal Other internal
diseases (n = 156) diseases (n = 346)
Gender, m/f (%) 71/85 (45.5/54.5) 165/181 (47.7/52.3)
Age, years 60.3B16.3 61.6B15.9
Height, m 1.68B0.09 1.69B0.09
Weight, kg 69.5B15.8** 74.4B15.0
BMI, kg/m2 24.6B4.5* 25.9B4.6
Length of stay, days1 9.7B7.6 (n = 139) 9.8B6.7 (n = 299)
* p ! 0.01; ** p ! 0.001.
1 64/502 (12.7%) patients were transferred and are therefore not
included.
Statistical Analysis
All data are given as mean B SD. Comparison of mean values
between two groups was performed by Mann-Whitney U-test, and
differences in frequencies were compared by ¯2 test. p ! 0.05 was
considered to be significant.
Results
Age and gender distribution as well as body height
were not different between patients with digestive dis-
eases and patients with other internal diseases (table 3).
Body weight and, thus, body mass index (BMI) were sig-
nificantly lower in patients with digestive than in patients
with other internal diseases (p ! 0.01). The average length
of hospital stay was not different between both patient
groups.
Assessment of malnutrition by the SGA demonstrated
that almost every fourth patient was malnourished (ta-
ble 4). A similar result was obtained by using the NRI or the
body cell mass (!30% of body weight) as criteria for malnu-
trition. Other parameters frequently used for the diagnosis
of malnutrition such as BMI, upper arm anthropometry or
albumin resulted in lower incidence values of malnutrition
ranging from 3.8% for BMI to 14.5% for albumin.
The incidence of malnutrition as defined by SGA was
not significantly different between patients with gastroin-
testinal diseases and other medical patients. In contrast,
the NRI indicated a significantly higher prevalence of
malnutrition among patients with digestive diseases when
compared with the other patients. Since the NRI includes
serum albumin, it is not surprising that low serum albu-
min levels were also more frequently observed in gastroin-
testinal patients. Regarding BMI and weight loss 110%
during the last 6 months prior to admission, there were no
Dig Dis 2003;21:245–251 247
 Table 4. Prevalence of malnutrition in all patients and in subgroups of patients with gastrointestinal vs. other internal
diseases (% values in parentheses)

All patients Gastrointestinal Other internal

(n = 502) diseases (n = 156) diseases (n = 346)

Nutritional scores
SGA B+C 121/501 (24.2) 45/156 (28.8) 76/345 (22.0) ns
Nutritional risk index ! 97.5 107/435 (24.6) 52/138 (37.7) 55/297 (18.5) p ! 0.001
Albumin ! 3.5 g/dl 63/434 (14.5) 34/138 (24.6) 29/296 (9.8) p ! 0.001
Anthropometry
BMI ! 18.5 kg/m2 19/501 (3.8) 8/156 (5.1) 11/345 (3.2) ns
Weight loss 1 10% body weight 48/501 (9.6) 17/156 (10.9) 31/345 (9.0) ns
AMA ! 10th percentile 51/496 (10.3) 23/153 (15.0) 28/343 (8.2) p = 0.020
AFA ! 10th percentile 55/495 (11.1) 22/153 (14.4) 33/342 (9.6) ns
Impedance analysis
BCM ! 30% body weight 123/453 (27.2) 43/147 (29.3) 80/306 (26.1) ns

SGA = Subjective global assessment; AMA = arm muscle area; AFA = arm fat area; BCM = body cell mass.
Statistics: ¯2 test, gastrointestinal diseases compared to other diseases.

were classified as SGA B or C than patients with non-

***
60 malignant diseases (fig. 1). Patients with malignancies
Prevalence of malnutrition (%)

were then divided into three subgroups: malignancies of

50
liver, pancreas or bile ducts (n = 22), malignancies of the
40 gastrointestinal tract (n = 17), and extraintestinal malig-
nancies (n = 15). According to the SGA, the highest preva-
30
lence of malnutrition of 63.6% was observed in patients
20 with malignancies of the liver, pancreas or bile ducts,
while patients with gastrointestinal or extraintestinal ma-
10
lignancies had a similar prevalence of 41.2 and 40.0%,
0 respectively (fig. 2).
nt nt al rs Among patients with benign diseases (table 5) the high-
na es na es tin es he
l g
i as ig te
s s ot est prevalence of malnutrition was observed in patients
a
m ise al as in sea
- m ise tro di
d n d s with chronic inflammatory bowel disease (40%), chronic
no ga
heart failure (37.5%), benign lung diseases (30.8%),
chronic liver diseases (28.9%) and rheumatoid arthritis
Fig. 1. Prevalence of malnutrition as diagnosed by SGA in malignant (28%). In contrast, a rather low percentage of patients
(n = 54) vs. non-malignant (n = 448) diseases, and in gastrointestinal with benign digestive diseases was classified as malnour-
diseases vs. other internal diseases; *** p ! 0.0001.
ished (ranging from 10 to 15.4%).
Finally, all patients were classified into two groups
with either good nutritional state or malnutrition accord-
differences observed between gastrointestinal patients ing to SGA (table 6). There were no differences between
and other medical patients. Regarding the results of upper both groups regarding the gender distribution. However,
arm anthropometry, we found that significantly more malnourished patients were significantly older than well-
patients with gastrointestinal disease had a low arm mus- nourished patients. Malnutrition was also associated with
cle area than the other patients, but frequencies for arm a significantly higher length of hospital stay. Further anal-
fat area were not significantly different. ysis demonstrated that NRI, albumin, BMI and parame-
When all patients were classified according to malig- ters of body composition obtained either by anthropome-
nant (n = 54) and non-malignant disease (n = 448), we try or BIA all yielded significantly lower values in patients
found that significantly more patients with malignancies with malnutrition than in well-nourished subjects.

248 Dig Dis 2003;21:245–251 Pirlich/Schütz/Kemps/Luhman/Burmester/

Baumann/Plauth/Lübke/Lochs
Table 5. Frequency of malnutrition in
selected benign diseases Total SGA B+C SGA B+C, %

Inflammatory bowel disease 10 4 40.0

Heart failure 24 9 37.5
Benign lung diseases 13 4 30.8
Chronic liver disease 45 13 28.9
Rheumatoid arthritis 25 7 28.0
Valvular defect 16 4 25.0
Systemic lupus erythematodes 13 3 23.1
Cerebral ischemia 13 3 23.1
Arrhythmia 40 9 22.5
Systemic infection 20 4 20.0
Vasculitis/myositis 15 3 20.0
Coronary heart disease 85 14 16.5
Benign pancreatic or biliary diseases 13 2 15.4
Gastrointestinal bleeding 16 2 12.5
Other benign gastrointestinal diseases 16 2 12.5
Reflux disease 10 1 10.0
Other cardiology 11 1 9.1
All other diseases 13 1 7.7
Other rheumatology 12 0 0

Note: This table includes only disease groups with n 6 10.

Discussion Table 6. Clinical characteristics and body composition in patients

with SGA A compared to SGA B+C
In this study the nutritional state of 502 hospitalized
SGA A SGA B+C p value
medical patients with a wide variety of disease states was (n = 380) (n = 121)
prospectively assessed. Using the SGA as the main criteri-
on, we found that almost every fourth patient admitted to Age, years 58.3B15.6 70.0B13.6 0.000
hospital was malnourished. This prevalence rate appears Gender, m/f 184/196 52/69 ns
Nutritional score
to be high, especially if one considers the fact that over-
Nutritional risk index 105.8B7.6 93.6B9.9 0.000
but not undernutrition is the main nutrition-related Albumin, g/d l4.3B0.5 3.7B0.6 0.000
health problem in Germany, indicated by a high and Anthropometry
obviously increasing average BMI in the healthy popula- BMI, kg/m2 26.6B4.3 22.1B3.6 0.000
tion, especially in the age group over 50 years [22]. Body weight, kg 76.2B14.8 62.21B12.0 0.000
AMA, mm2 5,584B1,499 4,421B1,121 0.000
Prevalence rates for malnutrition from other countries
AFA, mm2 2,829B1,320 1,650B854 0.000
reported during the last 15 years ranged from 20% [3, 6] Impedance analysis
up to 62% [4]. The majority of studies on medical BCM, kg 26.1B6.4 19.3B4.7 0.000
patients, however, reported malnutrition rates around Outcome
40% [1, 2, 5, 7, 9]. However, there are no representative LOS, days 9.3B6.8 13.1B8.1 0.000
LOS, median (range) 7.0 (2–55) 12.0 (3–49)
data from Germany. Therefore, we cannot answer the
question whether or not the nutritional state of hospital- SGA = Subjective global assessment; AMA = arm muscle area;
ized patients in Germany might have changed along with AFA = arm fat area; BCM = body cell mass; LOS = length of hospital
changes of medical care or the demographic transition stay.
during the last decades.
Differences in prevalence data on malnutrition among
different studies not only depend on the population select-
ed or on the institutional setting, but also on the different
diagnostic criteria used for the definition [11]. This was
highlighted in a very detailed study on 155 non-surgical

Hospital Malnutrition Dig Dis 2003;21:245–251 249

 Malignancies: liver, pancreas, bile ducts n=22 63.6
Extraintestinal malignancies n=15 42.9
Gastrointestinal malignancies n=17 41.2
Inflammatory bowel disease n=10 40.0
Heart failure n=24 37.5
Benign lung diseases n=13 30.8
Chronic liver disease n=45 28.9
Rheumatoid arthritis n=25 28.0
Cerebral ischemia n=13 23.1
Systemic infection n=20 20.0

Benign pancreatic or biliary disease n=13 15.4

0 10 20 30 40 50 60 70
Prevalence of malnutrition: SGA B+C (%)
Fig. 2. Prevalence of malnutrition in select-
ed diagnoses.

patients applying four different established clinical scores
to the same patients [4]. The authors found that depend-
ing on the instrument used for diagnosis, between 40 and
62% of their patients were classified as malnourished.
The question, however, which score or which diagnostic
instrument might provide the best description of the
nutritional state, is still open [23].
We decided to choose the SGA as the primary diagnos-
tic criterion in our study because this score is simple, inex-
pensive, non-invasive and can be performed at the bed-
side in a very short period of time. The SGA has also been
proven to be of prognostic relevance in a number of differ-
ent clinical settings, especially indicating an increased
mortality of patients with internal diseases [4, 5, 12, 14,
24]. In our study, the SGA yielded results similar to the
NRI, and patients who were classified as malnourished
according to SGA also had significantly impaired body
composition and plasma albumin compared to patients
classified as well nourished. We also found that malnour-
ished patients were on average 12 years older and had a
40% longer hospital stay than well-nourished patients.
This is in accordance with studies in other countries [7, 9]
and further supports the assumptions that increasing age
is a risk factor for the development of malnutrition [11],
and that malnutrition, on the other hand, is associated
with an impaired clinical course and extra costs of health
care [4, 7–9].
Two more recent studies from the Netherlands [4] and
from Brazil [7] also using the SGA for diagnosis suggested
that malnutrition might show a stronger association to
diseases of the digestive system than to some other medi-
cal conditions. This assumption appears plausible, since
impaired digestion per se can be expected to influence the
nutritional state. In fact, Naber et al. [4] found that 61%
of their gastrointestinal patients but only 30% of other
medical patients were malnourished. ln the study of
Waitzberg et al. [7], 61% of all patients with gastrointesti-
nal disorders were also found to be malnourished while on
average 48% of the patients were found to be malnour-
ished.
In contrast, we found a much lower frequency of mal-
nutrition of 29% in gastrointestinal patients. Further-
more, our gastrointestinal patients as a whole group were
not different from patients with other internal diseases, so
that our data obviously do not support the assumption of
patients with digestive diseases being at higher risk for
malnutrition. To better understand possible reasons for
these contradictory results, we performed more detailed
analyses in subgroups of patients. Not unexpectedly, we
found that the malnutrition rate in patients with malig-
nancies was more than twice as high as in benign medical
conditions. The highest prevalence of 64% was seen
among patients with malignancies of the liver, pancreas or
bile ducts. A subgroup analysis of benign diseases demon-
strated high prevalence rates of 130% in patients with
inflammatory bowel disease, heart failure and benign lung
disease (fig. 2). In contrast, benign pancreatic or biliary
diseases, gastrointestinal bleeding, reflux disease or other
benign digestive diseases were not associated with higher
malnutrition rates, but – compared with other medical

250 Dig Dis 2003;21:245–251 Pirlich/Schütz/Kemps/Luhman/Burmester/

Baumann/Plauth/Lübke/Lochs
conditions – demonstrated rather under average preva-
lence values.
In the cited studies from the Netherlands [4] and Brazil
[7], the gastrointestinal patients were not subclassified.
Therefore, we can only speculate that the higher malnutri-
tion rates of gastrointestinal patients in these studies com-
pared with our results might be caused by the inclusion of
more severely sick patients with inflammatory bowel dis-
eases, more patients with maldigestion or malabsorption,
or by inclusion of a higher percentage of malignancies of
the digestive system.
From previous studies we know that malnutrition in
hospitalized patients is frequently underestimated by the
medical staff [2], and even simple diagnostic procedures
such as the body weight are not performed in a relevant
number if not in the majority of patients [2, 7]. Since the
health system in Germany is currently under high eco-
nomic pressure it can be expected that in the future more
sick patients will be hospitalized and a number of patients
included in the present study will no longer be subjected
to hospitalization but more frequently directed to ambu-
latory treatment. Therefore, one can expect that in the
next years the prevalence of malnourished hospitalized
patients will increase in our country. Moreover, our data
show that some patient groups like malignancies, inflam-
matory bowel disease, chronic heart failure, or benign pul-
monary diseases have a specifically high prevalence of
malnutrition. In these patient groups, special care has to
be taken in the evaluation of the nutritional state and
nutritional support. It is, therefore, important to improve
the awareness of physicians, nurses and other profession-
als for this problem by better education in nutritional
assessment.
Acknowledgement
Dr. M. Pirlich was supported by the Else Kröner-Fresenius-Stif-
tung, Bad Homburg, Germany.

References

1 Coats KG, Morgan SL, Bartolucci AA, Wein-
sier RL: Hospital-associated malnutrition: A
reevaluation 12 years later. J Am Diet Assoc
1993;93:27–33.
2 McWhirter JP, Pennington CR: Incidence and
recognition of malnutrition in hospital. BMJ
1994;308:945–948.
3 Cederholm T, Jägrén C, Hellström K: Outcome
of protein-energy malnutrition in elderly medi-
cal patients. Am J Med 1995;98:67–74.
4 Naber THJ, Schermer T, de Bree A, Nusteling
K, Eggink L, Kruimel JW, Bakkeren J, van
Heereveld H, Katan MB: Prevalence of malnu-
trition in nonsurgical hospitalized patients and
its association with disease complications. Am
J Clin Nutr 1997;66:1232–1239.
5 Sacks GS, Dearman K, Replogle WH, Cora
VL, Meeks M, Canada T: Use of subjective
global assessment to identify nutrition-associ-
ated complications and death in geriatric long-
term care facility residents. J Am Coll Nutr
2000;19:570–577.
6 Malnutrition Prevalence Group: Prevalence of
malnutrition on admission to four hospitals in
England. Clin Nutr 2000;19:191–195.
7 Waitzberg DL, Caiaffa WT, Correia MITD:
Hospital malnutrition: the Brazilian National
Survey (IBRANUTRI): A study of 4,000 pa-
tients. Nutrition 2001;17:573–580.
8 Pirlich M, Luhmann N, Schütz T, Plauth M,
Lochs H: Mangelernährung bei Klinikpatien-
ten: Diagnostik und klinische Bedeutung. Akt
Ernähr Med 1999;24:260–266.
9 Robinson G, Goldstein M, Levine GM: Impact
of nutritional status on DRG length of stay. J
Parenter Enteral Nutr 1987;11:49–51.
10 Tucker HN, Miguel SG: Cost containment
through nutrition intervention. Nutr Rev 1996;
54:111–121.
11 Pirlich M, Lochs H: Nutrition in the elderly.
Best Pract Res Clin Gastroenterol 2001;15:
869–884.
12 Persson C, Sjöden PO, Glimelius B: The Swed-
ish version of the patient-generated global as-
sessment of nutritional status: Gastrointestinal
vs. urological cancers. Clin Nutr 1999;18:71–
77.
13 Detsky AS, McLaughlin JR, Baker JP, John-
ston N, Whittaker S, Mendelson RA, Jeejeeb-
hoy KN: What is subjective global assessment
of nutritional status? J Parenter Enteral Nutr
1987;11:8–13.
14 Lawson JA, Lazarus R, Kelly JJ: Prevalence
and prognostic significance of malnutrition in
chronic renal insufficiency. J Ren Nutr 2001;
11:16–22.
15 Buzby GP, Knox LS, Crosby LO, Eisenber JM,
Haakenson C, McNeal GE, Page CP, Peterson
OL, Reinhardt GF, Williford W: Study proto-
col: A randomized clinical trial of total paren-
teral nutrition in malnourished surgical pa-
tients. Am J Clin Nutr 1988;47:357–365.
16 Kushner RF, Schoeller DA: Estimation of total
body water by bioelectrical impedance analy-
sis. Am J Clin Nutr 1986;44:417–424.
17 Pirlich M, Biering H, Gerl H, Ventz M,
Schmidt B, Ertl S, Lochs H: Loss of body cell
mass in Cushing’s syndrome: Effect of treat-
ment. J Clin Endocrinol Metab 2002;87:1078–
1084.
18 Pirlich M, Schütz T, Spachos T, Ertl S, Weiss
M-L, Lochs H, Plauth M: Bioelectrical impe-
dance analysis is a useful bedside technique to
assess malnutrition in cirrhotic patients with
and without ascites. Hepatology 2000;32:
1208–1215.
19 Süttmann U, Ockenga J, Selberg O, Hooge-
straat L, Deicher H, Müller MJ: Incidence and
prognostic value of malnutrition and wasting in
human immunodeficiency virus-infected out-
patients. J Acquir Immune Defic Syndr Hum
Retrovirol 1995;8:239–246.
20 Gurney JM, Jelliffe DB: Arm anthropometry in
nutritional assessment: nomogram for rapid
calculation of muscle circumference and cross-
sectional muscle and fat mass. Am J Nutr 1973;
26:912–915.
21 Frisancho AR: New norms of upper limb fat
and muscle areas for assessment of nutritional
status. Am J Clin Nutr 1981;34:2540–2545.
22 Pirlich M, Schwenk A, Müller M, Ockenga J,
Schmidt S, Schütz T, Selberg O, Volkert D:
Leitlinie enterale Ernährung, Ernährungssta-
tus. Akt Ernähr Med 2003;28(suppl 1):10–25.
23 Klein S, Kinney J, Jeejeebhoy K, Alpers D,
Hellerstein M, Murray M, Twomey P: Nutri-
tion support in clinical practice: Review of
published data and recommendations for fu-
ture research directions. Am J Clin Nutr 1997;
66:683–706.
24 Pirlich M, Schütz T, Gastell S, Lochs H: Mal-
nutrition affects long-term prognosis in hospi-
talized patients. Gastroenterology 2002;122
(suppl 1):A363.

Hospital Malnutrition Dig Dis 2003;21:245–251 251

 Original Paper

Dig Dis 2003;21:252–257

DOI: 10.1159/000073343

Sugar Intake, Taste Changes and Dental Health

in Crohn’s Disease
Tatjana Schütz a Clemens Drude b Erika Paulisch b Klaus-Peter Lange b
Herbert Lochs a
a Medizinische Klinik mit Schwerpunkt Gastroenterologie, Hepatologie und Endokrinologie, Universitätsklinikum

Charité Campus Mitte, Berlin und b Zentrum für Zahnmedizin, Abteilung für zahnärztliche Prothetik und
Alterszahnmedizin, Universitätsklinikum Charité Campus-Virchow, Berlin, Deutschland

Key Words ar consumption and insufficient oral hygiene seem to

Crohn’s disease W Taste changes W Sugar intake W Zinc W cause the higher caries prevalence. Obviously, patients
Dental caries with CD belong to a high-risk group, and preventive mea-
sures should be taken early in the course of the disease.
Copyright © 2003 S. Karger AG, Basel

Abstract
Background: An increased intake of sucrose has been
reported in patients with Crohn’s disease (CD). Since
subclinical zinc deficiency reduces taste perception for
sweet, we investigated taste perception, sucrose intake
and plasma zinc levels as well as dental status in CD
patients. Methods: Carbohydrate intake and plasma zinc
levels were assessed in 24 CD patients and 24 age-
matched controls (Con). Taste threshold for sucrose, oral
hygiene and caries prevalence were evaluated. Results:
In CD a higher sucrose intake (CD 107.1 B 27.7 vs. Con
71.9 B 13.7 g/day; p ! 0.001), a higher taste threshold for
sweet (CD 7.31 vs. Con 2.91 g/l; p ! 0.001) and lower plas-
ma zinc levels (CD 11.5 B 1.5 vs. Con 13.5 B 2.0 Ìmol/l;
p ! 0.001) were found. API was poor (CD 85.4 B 23.6, Con
31.8 B 24.1, p ! 0.001) and correlated with sucrose intake
(p ! 0.01). Caries prevalence was increased in patients
with longer disease (1 3 years) (DMFT index: 13 years
15.6 B 5.7 vs. ! 3 years 9.5 B 4.3; p ! 0.05). Conclusion:
Dental status in CD patients is poor. Both increased sug-
Introduction
Malnutrition is frequently observed in patients with
Crohn’s disease (CD) and manifests itself in weight loss
and nutrient deficiencies [1]. Special importance is at-
tached to zinc and its role in taste perception. In this con-
text, Henkin et al. [2] and Solomons et al. [3] described
that reducing zinc triggers a reversible hypogeusia.
In the 1970s, studies on the dietary habits of CD
patients drew attention to higher intakes of carbohy-
drates, sugar or added sugar before or after the onset of
the disease compared to patients with ulcerative colitis [4]
or healthy controls [5–11]. As one possible cause for the
increased sugar consumption, an impaired gustatory
function for the sweet taste has been discussed. However,
the hypothesis that an excessive intake of refined carbohy-
drates may be the result of a higher taste threshold for
sweet was refuted by three studies [12–14]. A higher
threshold for sweet was only observed in patients with

© 2003 S. Karger AG, Basel Dr. Tatjana Schütz

ABC 0257–2753/03/0213–0252$19.50/0 Universitätsklinikum Charité
Fax + 41 61 306 12 34 Medizinische Klinik mit Schwerpunkt Gastroenterologie, Hepatologie und Endokrinologie
E-Mail karger@karger.ch Accessible online at: Schumannstrasse 20/21, DE–10098 Berlin (Germany)
www.karger.com www.karger.com/ddi Tel. +49 30 450 514059, Fax +49 30 450 514923, E-Mail elke-tatjana.schuetz@charite.de
active disease (Crohn’s disease activity index – CDAI Table 1. Characteristics of CD patients and healthy controls (mean
1200) and patients with CD for more than 2 years [15]. B SD)
Studies on dental status imply that CD patients have a
CD patients Controls
high caries frequency and activity [16, 17]. The reason for
the increased caries prevalence is unclear but as the most Number 24 24
likely link the dietary regimen can be suggested. Gender, m/f 10/14 12/12
In this overlapping field of gastroenterology, otorhino- Age, years 36.2B15.0 35.4B10.8
Smoker/non-smoker 13/11 12/12
laryngology and dentistry this interdisciplinary study was
Crohn’s disease activity index 153B98 –
conducted to examine the association between serum zinc Disease duration, years 5.0B3.9 –
concentrations and the taste threshold for sweet, as well as
the intake of sucrose and carbohydrates and its effect on
the risk of caries and caries prevalence in CD patients
compared to healthy controls.
Table 2. Mono-, di- and polysaccharide intake (mean B SD) in CD
patients and healthy controls
Methods
CD patients Controls
Patients
Twenty-four CD diagnosed by generally accepted criteria [18] Monosaccharides, g/day 43B15 46B14
and 24 age-matched healthy subjects participated in the study (ta- Glucose, g/day 20B7 20B5
ble 1). Disease activity was assessed according to the CDAI [19]. Dis- Disaccharides, g/day 122B32*** 87B16
ease was active (CDAI 1 150) in 50% of the patients. The duration of Sucrose, g/day 107B28*** 72B74
disease ranged between 1 and 18 years with a median of 4 years. Polysaccharides, g/day 195B27* 202B21
Eleven patients were current smokers. Patients received standard Total carbohydrates, g/day 367B64 332B70
medical treatment including corticosteroids (n = 11), mesalazine (n =
17), azathioprine (n = 2), and metronidazole (n = 1) as mono- or * p ! 0.05; *** p ! 0.001.
combination therapy. Two patients were irregularly supplemented
with vitamin A (n = 1) or zinc (n = 1). Patients with severe secondary
diseases, otorhinolaryngologic and neurologic disorders, or diseases
of the oral cavity as well as patients taking medication that influences
taste perception (diuretics, antidiabetics, antihypertensives, anti-
rheumatics, psychoactive drugs) were excluded from the study. The
study protocol was approved by the local ethical committee and all typical site for the perception of the sweet taste, and the subjects were
patients and healthy subjects gave their written informed consent. asked to identify the test stimulus. After each tasting the subjects
expectorated the samples and rinsed their mouth with water. Taste
Procedure threshold was defined as the lowest concentration which the subject
Sucrose and carbohydrate intake of the present daily diet were correctly identified.
assessed by a food-frequency questionnaire in which the intake of 64 Plasma zinc concentration: Metal-free plastic syringes and con-
food items and 9 beverages was documented in three categories: regu- tainers (Sarstedt, Nümbrecht, Germany) were used to collect blood
larly (once to several times a day), rarely (once a week to once a for the zinc measurement. Zinc and other laboratory parameters (al-
month), and never. Standard portion sizes were applied to the food bumin, hematocrit) were determined by standard methods.
categories and glucose and carbohydrate content was then calculated Dental and oral health: The prevalence of caries was assessed by
using the EBIS software (Forschungszentrum für Ernährung in Prä- determining the percentage of decayed, missing and filled teeth
vention und Therapie, Hohenheim GmbH, Stuttgart, Germany), (DMFT index) [22]. Low values indicate good dental health. Oral
which is based on the German Food and Nutrition Database (BLS, hygiene was assessed by using the API index (approximal plaque
version II.1). index), which is calculated as the percentage of contaminated ap-
Taste threshold: Olfactometric and gustometric testing was per- proximal sites [23]. Analogous to the DMFT index, low values point
formed in the morning at least 1 h after breakfast by a standard pro- to sufficient oral hygiene.
cedure [20]. Shortly, subjects were asked to restrain from eating,
smoking and using toothpaste during that period. Prior to the gus- Statistics
tometric measurements, anosmia was excluded by testing the smell Results are given as mean B SD. Means were compared by
stimuli camphor, vanillin, dichlorethane, menthol and formic acid in Mann-Whitney U-test and frequencies by the ¯2 test with p ! 0.05 as
different concentration ranges using the sniff-bottle technique [20]. level of significance. All calculations were performed by the comput-
The taste threshold for sweet was quantitatively determined with 8 er software SPSS (version 10.0, SPSS Inc., Chicago, Ill., USA).
sucrose solutions in graded concentrations (1–256 g/l in deionized
water) using the pipette method [21]. 0.5 ml of the solutions in
ascending order was given drop by drop on the tip of the tongue, the

Sugar Intake in Crohn’s Disease Dig Dis 2003;21:252–257 253


200 **
Sucrose intake (g/day)
100
0
Controls CD CD
(n = 24) (n = 6) (n = 12)
<3 years 3-6 years
Fig. 1. Sucrose intake in healthy controls and CD patients in groups
according to length of disease. Box plots with horizontal bars indicat-
ing median values, boxes indicating the 25th centiles, error bars indi-
cating the 95% confidence interval and [ indicating values outside
the 95th centile. ** p ! 0.01.
Results
Eating Habits, Sucrose and Carbohydrate Intake. Eat-
ing habits differed between patients and controls. Signifi-
cantly more CD patients abstained from alcoholic bever-
ages (wine: 50.0 vs. 12.5%; beer: 62.5 vs. 25.0%; p ! 0.01)
and dried fruits (83.3 vs. 50%, p ! 0.05) compared to
healthy controls. Total carbohydrate intake was not dif-
ferent between CD patients and healthy controls (table 2).
CD patients, however, consumed 40% more disaccha-
rides than healthy controls, which was due to a signifi-
cantly higher sucrose intake (p ! 0.001) (fig. 1). This dif-
ference remained significant when sucrose intake was cal-
culated per kilogram body weight (CD 1.72 B 0.62 vs.
control 1.03 B 0.26 g/kg b.w.; p ! 0.001). Surprisingly,
patients with shorter disease duration had a significantly
higher sucrose intake than patients with a longer lasting
disease (fig. 1). The higher sucrose intake was due to a
more frequent consumption of sweet-tasting food prod-
ucts, such as pudding (25 vs. 0%, p ! 0.01), honey and jam
(70.8 vs. 37.5%, p ! 0.05), chocolate (58.3 vs. 16.7%, p !
0.01), and cake (50.0 vs. 12.5%, p ! 0.01).
Olfactometry and Gustometry. All subjects could smell
at least one concentration level of each odorant so that
anosmia, which would negatively influence taste percep-
254 Dig Dis 2003;21:252–257
60 CD Controls
50
***
Number of subjects (%)
40
***
30
***
20
CD 10
(n = 6)
>6 years
0
1 2 4 8 16 32
Sucrose (g/l)
Fig. 2. Proportion of CD patients compared to healthy controls, who
detected the tested sucrose concentration. *** p ! 0.001.
tion, was excluded in the study population. Taste thresh-
old for sweet was significantly higher in CD patients com-
pared to healthy controls (7.31 vs. 2.91 g/l, p ! 0.001). In
contrast to 58.3% of the control subjects, none of the CD
patients were able to recognize the two lowest sucrose con-
centrations (fig. 2). There was no difference in taste detec-
tion comparing patients with active (CDAI 1150) to inac-
tive (CDAI !150) disease. There was no correlation
between the taste threshold for sweet and sucrose intake
in either group.
Plasma Zinc Concentration. Since plasma zinc is bound
to albumin, we also measured albumin concentrations.
Plasma zinc concentrations significantly correlated with
albumin concentrations in both CD patients and controls
(CD: r2 = 0.150, control: r2 = 0.052, both p ! 0.001). The
plasma zinc values of 5 CD patients (20.8%) and 2 con-
trol subjects (8.3%) were below the normal range (10.6–
17.9 Ìmol/l). The mean value for the group of CD patients
was significantly lower than that of the healthy controls
(11.5 B 1.5 vs. 13.5 B 2.0 Ìmol/l; p ! 0.001) (fig. 3). Nei-
ther in CD patients nor in healthy controls were plasma
zinc levels correlated with sucrose intake.
Dental and Oral Health. Dental health was impaired in
CD patients compared to controls. However, this became
apparent only with longer lasting disease with an in-
Schütz/Drude/Paulisch/Lange/Lochs

20 ***
Plasma zinc concentration (µmol/l)
18
16
14
12
10
6
Controls
(n = 24)
Fig. 3. Plasma zinc concentration in healthy controls and CD
patients. Box plots as explained in figure 1. *** p ! 0.001.
creased DMFT index indicating higher caries prevalence
(fig. 4). Oral hygiene was insufficient in CD patients with
a significantly higher API index compared to controls
(CD: 85.5 B 23.6% vs. control: 31.8 B 24.1%, p ! 0.001).
The API index was significantly correlated to sucrose
intake in both groups (CD: r2 = 0.0104; control:
r2 = 0.0059; p ! 0.01). 25% of CD patients vs. 4% of con-
trols did not regularly go to the dentist’s.
Discussion
Several studies have shown a higher sugar intake in CD
patients when compared to the healthy population even
before the onset of the disease [4–11]. The reason for this
dietary habit is unclear. However, a correlation with the
decreased zinc status was suspected since zinc is an
important trace element for the taste perception sweet [2,
3]. Furthermore, a higher caries prevalence would be
expected in CD patients as a consequence of the higher
sugar consumption.
Zinc deficiency has frequently been described in CD
patients [3, 24, 25], and alterations in taste acuity have
been observed in these patients [13, 14, 24]. The mecha-
nism of chemosensory dysfunction is not well understood
but the key role in the influence of zinc on taste perception
is attributed to the zinc-dependent parotid saliva enzyme
Sugar Intake in Crohn’s Disease
30 *
*
DMFT-index (%)
20
10
0
Controls CD CD CD
CD
( = 24) (n = 6) (n = 12) (n = 6)
(n = 24)
<3 years 3-6 years >6 years
Fig. 4. Prevalence of caries (DMFT index) in healthy controls and in
CD patients in groups according to length of disease. Box plots as
explained in figure 1. * p ! 0.05.
gustin/carbonic anhydrase IV, which is decreased in pa-
tients with hypogeusia and dysgeusia [2]. It seems to exert
its action as a trophic factor on the taste bud stem cells
and so promotes growth and development of taste buds.
So far there have been no data on gustin/carboanhydrase
IV concentrations in saliva of CD patients with impaired
taste acuity. Moreover, a possible influence of medication
on gustatory function must be assumed.
Our data do in fact demonstrate higher sugar intake,
higher caries prevalence and lower zinc plasma levels in
CD patients. However, these changes appear not to be
related to each other. The taste threshold for sweet was
also significantly higher in CD patients than healthy con-
trols. Only Lederer et al. [15] could also measure a signifi-
cantly higher taste threshold for sweet in patients with
active disease (CDAI 1200), whereas other studies only
reported a significantly higher threshold for salt taste [13]
and for acid taste [14]. Solomons et al. [3] found a signifi-
cantly lower overall taste detection score, which com-
prised all four basic taste modalities, whereas Kasper et al.
[12] could not detect any differences.
The question has to be raised why a correlation be-
tween low plasma zinc levels and increased taste thresh-
old for sweet could not have been found. This might be
due to the fact that plasma zinc levels do not necessarily
reflect the zinc status but could rather be dependent on
inflammatory processes. This might, however, not ex-
Dig Dis 2003;21:252–257 255
plain our results since there was no difference in the taste
threshold between patients with active disease versus
patients with quiescent disease. It would be interesting to
investigate the effect of zinc supplementation on the taste
threshold in CD patients and consequently on sugar
intake. To our knowledge this has not yet been adequately
investigated.
The higher sugar intake was not accompanied by a
higher total carbohydrate intake in our study but was rath-
er due to the consumption of sweets. This is very well in
line with former studies, using different methods to evalu-
ate dietary habits of CD patients [4–11]. This specific eat-
ing of sweets makes it unlikely that CD patients substitute
carbohydrates for a possible fat intolerance since it would
be expected that the intake of all carbohydrates would be
increased.
One consequence of high sugar intake is an increase in
the incidence of caries. Dental caries is a multifactorial
disease in which the effects of the two main initiating fac-
tors dental plaque and fermentable carbohydrates are
modified by various endogenic and exogenic factors. CD
patients are reported to have a higher incidence of dental
caries than healthy controls [16, 17]. In our study the
extent of caries increased with the duration of disease.
This could be due to long-lasting high sugar intake but
also to nutritional deficiencies like calcium and fluoride
deficiency. A further explanation could be high levels of
cariogenic bacteria in saliva, such as streptococci and lac-
tobacilli [26], which may result from high sucrose intake
in combination with insufficient oral hygiene. Factors of
the development of dental caries resulting from CD per se
such as altered tooth structure [17], salivary flow rate [26]
and salivary constituency in regard to antimicrobial pro-
teins [27] have not been proven.
The poor dental status was correlated to insufficient
oral hygiene and the lack of regular visits to the dentist’s
in CD patients. This clearly indicates that CD patients are
a risk group for poor dental health, and therefore mea-
sures should be taken early in the course of the disease to
avoid deterioration. Since eating habits are difficult to
influence, training for dental hygiene and specific atten-
tion to regular dentist visits should be taken in CD
patients.
Conclusion
The dental status in CD patients was found to be poor.
The taste threshold for sweet was increased in CD pa-
tients; an association to plasma zinc concentrations and
sucrose intake, however, could not be proven. Both in-
creased sugar consumption and insufficient oral hygiene
seem to cause the higher caries prevalence. Obviously,
CD patients belong to a high-risk group, and preventive
measures should be taken early in the course of the dis-
ease.

References

1 Geerling BJ, Badart-Smook A, Stockbrügger
RW, Brummer RJ: Comprehensive nutritional
status in patients with long-standing Crohn’s
disease currently in remission. Am J Clin Nutr
1998;67:919–926.
2 Henkin RI, Martin BM, Agarwal RP: Efficacy
of exogenous oral zinc treatment of patients
with carbonic anhydrase IV deficiency. Am J
Med Sci 1999;318:392–405.
3 Solomons NW, Rosenber IH, Sandstead HH
Vo-Khactu KP: Zinc deficiency in Crohn’s dis-
ease. Digestion 1977;16:87–95.
4 Mayberry JF, Rhodes J, Newcombe RG: In-
creased sugar consumption in Crohn’s disease.
Digestion 1980;20:323–326.
5 Martini GA, Brandes JW: Increased consump-
tion of refined carbohydrates in patients with
Crohn’s disease. Klin Wochenschr 1976;54:
367–371.
6 Kasper H, Sommer H: Dietary fiber and nu-
trient intake in Crohn’s disease. Am J Clin
Nutr 1979;32:1898–1901.
7 Silkoff K, Hallak A, Yegena L, Rozen P, May-
berry JF, Rhodes J, Newcombe RG: Consump-
tion of refined carbohydrate by patients with
Crohn’s disease in Tel-Aviv-Yafo. Postgrad
Med J 1980;56:842–846.
8 Mayberry JF, Rhodes J, Allan R, Newcombe
RG, Regan GM, Chamberlain LM, Wragg KG:
Diet in Crohn’s disease. Two studies of current
and previous habits in newly diagnosed pa-
tients. Dig Dis Sci 1981;26:444–448.
9 Järnerot G, Järnmark I, Nilsson K: Consump-
tion of refined sugar by patients with Crohn’s
disease, ulcerative colitis, or irritable bowel
syndrome. Scand J Gastroenterol 1983;18:
999–1002.
10 Tragnone A, Valpiani D, Miglio F, Elmi G,
Bazzocchi G, Pipitone E, Lanfranchi GA: Di-
etary habits as risk factors for inflammatory
bowel disease. Eur J Gastroenterol Hepatol
1995;7:47–51.
11 Reif S, Klein I, Lubin F, Farbstein M, Hallak
A, Gilat T: Pre-illness dietary factors in inflam-
matory bowel disease. Gut 1997;40:754–760.
12 Kasper H, Sommer H, Wild M: Ge-
schmacksschwellen bei Kranken mit Enteritis
regionalis (Morbus Crohn). Akt Ernähr Med
1980;5:196–198.
13 Tiomny E, Horwitz C, Graff E, Rozen P, Gilat
T: Serum zinc and taste acuity in Tel-Aviv
patients with inflammatory bowel disease. Am
J Gastroenterol 1982;77:101–104.
14 Penny WJ, Mayberry JF, Aggett PJ, Gilbert JO,
Newcombe RG, Rhodes J: Relationship be-
tween trace elements, sugar consumption and
taste in Crohn’s disease. Gut 1983;24:288–292.
15 Lederer PC, Cidlinsky K, Kobal G, Lux G:
Geschmacksstörungen bei M. Crohn Patien-
ten. Elektrogustometrie und chemische Ge-
schmacksprüfung. Z Gastroenterol 1985;23:
470.
16 Sundh B, Hultén L: Oral status in patients with
Crohn’s disease. Acta Chir Scand 1982;148:
531–534.

256 Dig Dis 2003;21:252–257 Schütz/Drude/Paulisch/Lange/Lochs

17 Rooney T: Dental caries prevalence in patients
with Crohn’s disease. Oral Surg 1984;57:623–
624.
18 Malchow H, EWE K, Brandes JW, Goebell
HM, Ehms S, Sommer H, Jesdinsky H: Euro-
pean Cooperative Crohn’s Disease Study
(ECCDS): Results of drug treatment. Gastroen-
terology 1984;86:249–266.
19 Best WR, Becktel JM, Singleton W: Develop-
ment of a Crohn’s disease activity index. Na-
tional Cooperative Crohn’s Disease Study.
Gastroenterology 1976;70:439–444.
Sugar Intake in Crohn’s Disease
20 Anonymous: Mitteilungen der Gesellschaft:
Empfehlungen zur Untersuchung des Riech-
und Schmeckvermögens. HNO Prax 1980;5:
62–67.
21 Rollin H: Funktionsprüfung und Störungen des
Geschmackssinnes. Arch Otorhinolaryngol
1975;210:165–207.
22 WHO Oral Health Survey, ed 2. Geneva,
WHO, 1977.
23 Lange DE: Die gezielte Vorbehandlung vor der
systemischen Paradontolbehandlung. Zahn-
ärztl Welt/Reform 1975;8:366.
24 McClain C, Soutor C, Zieve L: Zinc deficiency:
A complication of Crohn’s disease. Gastroen-
terology 1980;78:272–279.
Dig Dis 2003;21:252–257
25 Ainley CC, Cason J, Carlsson LK, Slavin BM,
Thompson RP: Zinc status in inflammatory
bowel disease. Clin Sci 1988;75:277–283.
26 Sundh B, Emilson CG: Salivary and microbial
conditions and dental health in patients with
Crohn’s disease: A 3-year study. Oral Surg Oral
Med Oral Pathol 1989;67:286–290.
27 Sundh B, Johansson I, Emilson CG, Nordgren
S, Birkhed D: Salivary antimicrobial proteins
in patients with Crohn’s disease. Oral Surg Oral
Med Oral Pathol 1993;76:564–569.
257
 Original Paper

Dig Dis 2003;21:258–261

DOI: 10.1159/000073344

Serum Mineral Levels in Children with

Intestinal Parasitic Infection
José L. Olivares a Ramona Fernández a Jesús Fleta a Gerardo Rodrı́guez a
Antonio Clavel b
a Departmentof Paediatrics and b Microbiology and Parasitology ‘Lozano Blesa’ Hospital, University of Zaragoza,
School of Medicine, Zaragoza, Spain

Key Words Introduction

Copper W Zinc W Magnesium W Enterobius vermicularis W
Giardia lamblia
Abstract
Parasitic infections are highly prevalent in the general
population. A relation between a parasitic infection and
absorption of minerals is not an easy task. Serum levels
of copper, zinc and magnesium were prospectively mea-
sured in 64 children with intestinal parasitic infection.
Thirty-nine children with Enterobius vermicularis were
treated with pyrantel pamoate and 25 children with Giar-
dia lamblia with tinidazole and metronidazole. Three
months after treatment, significant differences in serum
copper, zinc and magnesium were seen in patients with
E. vermicularis infection, and in serum magnesium lev-
els in patients with G. lamblia. Although the pathogenic
mechanism is not clear, these findings could reflect a
deficiency related to malabsorption due to mucous affec-
tion. Early detection and treatment of intestinal parasito-
sis could avoid these serum mineral deficiencies.
Copyright © 2003 S. Karger AG, Basel
Childhood is probably the most demanding period of
life for meeting the body’s nutritional demands. Minerals
perform important functions in children’s growth and
development. The recognition that trace minerals regu-
late key metabolic pathways, modulate the immune re-
sponse, and suppress the incidence of various disease
states serves to emphasize their direct importance in
health maintenance [1]. Intestinal parasitosis remains an
important public health concern because of the high fre-
quency reached in several countries as well as its nutri-
tional consequences. Although childhood copper, zinc
and magnesium deficiencies are rather unusual, recent
studies suggest that several diseases (malnutrition, intesti-
nal malabsorption syndromes and intestinal parasitosis)
may produce them. Early diagnosis and treatment of these
disorders are very important to obtain optimal levels of
growth, development, immune response and intellectual
capacity [2–4].
The aim of this study was to compare serum copper,
zinc and magnesium levels in children with Enterobius
vermicularis and Giardia lamblia intestinal infection, be-
fore and after treatment without infection.

© 2003 S. Karger AG, Basel Prof. José L. Olivares López

ABC 0257–2753/03/0213–0258$19.50/0 Departamento de Pediatrı́a
Fax + 41 61 306 12 34 Facultad de Medicina, Universidad de Zaragoza
E-Mail karger@karger.ch Accessible online at: C/. Domingo Miral s/n, ES–50009 Zaragoza (Spain)
www.karger.com www.karger.com/ddi Tel. +34 976 761724/761725, Fax +34 976 761726, E-Mail olivares@posta.unizar.es
Table 1. Anthropometric measures and serum mineral levels of
patients with E. vermicularis at diagnosis and after treatment (mean
B SD)
Diagnosis After treatment
Weight-for-age Z-score – 0.65B1.37 – 0.32B1.16*
Body mass index, kg/m2 17.28B2.31 17.60B2.41**
Copper, Ìg/dl 116.34B18.62 123.08B24.29*
Zinc, Ìg/dl 91.32B10.90 97.07B13.12**
Magnesium, mg/dl 1.65B0.11 1.70B0.16*
Differences between paired sets of time, diagnosis – after treat-
ment, using Student’s test.
* ! 0.05; ** ! 0.01.
Patients and Methods
We studied 64 Spanish children (34 boys, 30 girls) from Aragón,
Northeast Spain, who were diagnosed as having one unique parasite
infection; 39 of these 64 cases were infected by E. vermicularis and 25
by G. lamblia. Fifty-one patients came from an urban area (Zarago-
za) and 13 from rural areas. Most of these children were of a medium
socioeconomic status. Ages at diagnosis ranged from 10 months to 15
years (mean 9.71 B 3.63 years) for E. vermicularis and 5.82 B 3.31
years for G. lamblia. Abdominal pain, acute diarrhea, anorexia, anal
itch and fever were the most frequent symptoms.
Identification of E. vermicularis was carried out by the Graham
technique [5]. For G. lamblia identification, concentration of fecal
stool was performed by the method described by Ritchie [6], using
ether instead of acetyl acetate [7, 8]. Copper, zinc and magnesium
levels were assessed by atomic absorption spectrophotometer (model
Video 11E, Thermo Jarrel Ash). Serum copper and zinc levels were
previously diluted to a 1/5 proportion and magnesium to a 1/100.
The method obtains directly measurements of mineral concentration
[9]. Children were evaluated twice, first at diagnosis and on a second
follow-up visit 3 months after the treatment and without active infec-
tion.
Pyrantel pamoate (10 mg/kg/day, two doses separated by 2 weeks)
was the treatment for E. vermicularis. Tinidazole (50 mg/kg/day, two
doses, separated by 2 weeks) was used to treat G. lamblia-infected
children and, when G. lamblia parasitation persisted after this treat-
ment, metronidazole (25 mg/kg/ day, 7 days) was employed [10].
Anthropometric measures and serum copper, zinc and magne-
sium concentrations were determined 3 months after treatment,
when patients were asymptomatic and stools were not infected.
Anthropometric data were compared with international standards
(Anthro database, WHO). Fecal stool samples were collected after
the completion of treatment to verify that there was no intestinal
parasitic infection.
Kolmogorov-Smirnov (Lilliefors modification) was applied to
assess normality of each variable. All variables were found to be nor-
mally distributed (Kolmogorov-Smirnov: Z 1 0.05) and then a para-
metric test was performed. Differences between variables at diagno-
sis and after treatment were examined using Student’s t test. Statisti-
cal programs SPSS for Windows 11.5 (SPSS Inc.) were employed.
Statistical significance was defined as a p ! 0.05.
Serum Minerals in Intestinal Parasitic
Infection
Table 2. Anthropometric measures and serum mineral levels of
patients with G. lamblia at diagnosis and after treatment (mean B
SD)
Diagnosis After treatment
Weight-for-age Z-score –0.11B1.25 0.34B1.38
Body mass index, kg/m2 16.88B2.50 17.42B2.62**
Copper, Ìg/dl 132.17B29.23 132.34B26.69
Zinc, Ìg/dl 92.03B13.42 92.12B12.63
Magnesium, mg/dl 1.64B0.12 1.69B0.13*
Differences between paired sets of time, diagnosis – after treat-
ment, using Student’s test.
* p ! 0.05; ** p ! 0.01.
Parents were fully informed about the aims of the study and
signed a consent form for participation. The study protocol was
reviewed and approved by the Ethical Research Committee of the
‘Lozano Blesa’ Zaragoza University Hospital (Spain).
Results
Table 1 shows measurements of patients with E. vermi-
cularis intestinal infection at diagnosis and 3 months after
treatment, without infection. After treatment we ob-
served a significant improvement in weight for age Z-
score (p ! 0.05), body mass index (BMI) (p ! 0.01), serum
copper (p ! 0.05), zinc (p ! 0.01) and magnesium (p !
0.05) concentrations. Table 2 shows measurements of pa-
tients with G. lamblia infection at diagnosis and 3 months
after treatment. We also observed a significant improve-
ment in BMI (p ! 0.01) and serum magnesium (p ! 0.05)
levels after treatment but not in weight-for-age Z-score
(p = 0.083), copper (p = 0.963) and zinc (p = 0.996) lev-
els.
Differences found in patients with E. vermicularis
intestinal infection compared with the G. lamblia group
were only significant for copper levels at diagnosis (116.3
B 18.6 vs. 132.1 B 29.2 Ìg/dl respectively; p = 0.01) but
not after treatment 3 months later (123.1 B 24.3 vs. 132.3
B 26.7 Ìg/dl respectively; p = 0.157). The rest of the vari-
ables did not show significant differences.
Discussion
The physiologic role and dietary needs for minerals in
the nutrition of the children are known with varying
degrees of certainty, and only limited information exists
Dig Dis 2003;21:258–261 259
on the bioavailability of these elements in different foods.
Copper, zinc and magnesium are components of some cel-
lular enzymes, participate in several immune processes
and they play an important role in the resistance to free
radical damage by stabilizing the cellular membrane [1].
Absorption of copper in the small intestine has both
active and passive components that do not appear to be
dramatically influenced by the form in which the copper
is presented. Inside the mucosal cells, copper can asso-
ciate with metallothionein. Fructose and sucrose accen-
tuate the deficiency signs induced by a diet low in copper,
compared with the effect of starch and glucose. Although
the mechanisms and control of zinc are no completely
understood, intestinal absorption involves uptake by the
intestinal cell, movement through the mucosal cell, trans-
fer to the portal circulation, and secretion of endogenous
zinc back into the intestinal cell. Zinc supply and tissue
reserves are major factors in the homeostatic control of
zinc absorption by regulation of mucosal cell absorption.
Magnesium absorption occurs mainly in the ileum and
colon by passive diffusion, pulling of solvent and active
transport. Vitamin D increases magnesium absorption
but the presence of phytates decreases it [1, 2].
The presence of gastrointestinal parasites in children is
negatively associated with anthropometric characteris-
tics, physical work capacity, blood hemoglobin levels and
nutritional status [2, 11]. Pegelow et al. [4], in a study
developed in 8- to 10-year-old children with E. vermicu-
laris infection and other parasitic infestations, from ten
schools of the rural districts of Sukaraja, West Java,
Indonesia, concluded that their nutritional status was
characterized by anemia (13%) and a stunting prevalence
of 51%. In our study, after 3 months of treatment, we have
observed a significant increase in weight for age Z-score
and BMI in E. vermicularis group, and in BMI of children
with G. lamblia infection.
Koltas et al. [12], in children with enterobiosis, showed
that mean levels of serum copper, zinc and magnesium
were significantly lower in the infected group than in the
control group. In our study we have observed a significant
increase in serum concentrations of copper, magnesium
and, especially of zinc, 3 months after treatment with tini-
dazole or metronidazole.
Karakas et al. [13], in children with giardiasis or ame-
biasis, showed that serum zinc levels were significantly
decreased when compared to controls. After metronida-
zole therapy, a significant increase in zinc levels was
observed. There was no significant difference in serum
copper levels between patients and controls before thera-
py. Among G. lamblia-infected children, Ertan et al. [14]
260 Dig Dis 2003;21:258–261
revealed that giardiasis increased the serum copper levels
and decreased the zinc and iron levels. In our study in
children with parasitic G. lamblia intestinal infection, we
have observed a significant increase in serum magnesium
levels 3 months after treatment. There are no significant
differences in serum copper or zinc levels after therapy.
The pathogenic mechanism of these serum mineral
changes observed in intestinal parasitic infections is not
clear. Abel et al. [15] pointed out the possible roles of pro-
tein kinase A (PKA) in cell motility and excystation of the
early diverging eukaryote G. lamblia. Kinetic analysis of
the recombinant PKA showed that ATP and magnesium
are preferred in this enzymatic process. Otherwise, Cragg
et al. [16] showed that a novel zinc-regulated human zinc
transporter, hZTL1 (human ZnT-like transporter 1), is
located in the enterocyte apical membrane. Localization,
regulatory properties and function of hZTL1 indicate a
role in regulating the absorption of nutrients.
In conclusion, this study shows significant differences
in serum copper, zinc and magnesium levels in patients
with E. vermicularis intestinal infection, and in serum
magnesium concentrations in patients with G. lamblia, 3
months after treatment. Although the pathogenic mecha-
nism is not clear, persistent small intestinal damage is a
matter of importance in such children [17]. There is clear
evidence that zinc and other minerals are involved in the
recovery of small intestinal mucosa after injury. Early
detection and treatment for each case of intestinal parasit-
ic infection could avoid these mineral deficiencies.
Acknowledgment
The authors thank Prof. Tomás Martı́nez, Department of Statis-
tics, School of Medicine, University of Zaragoza.
Olivares/Fernández/Fleta/Rodrı́guez/Clavel
 References
1 Milner JA: Trace minerals in the nutrition of
children. J Pediatr 1990;117:147–155.
2 Stephenson LS, Lathan MC, Ottesen EA: Mal-
nutrition and parasitic helminth infections. Pa-
rasitology 2000;121(suppl):23–38.
3 Bahader S, Ali G, Shaalan A, Khalil H, Khalil
N: Effects of Enterobius vermicularis infection
on intelligence quotient and anthropometric
measurements of Egyptian rural children. J
Egypt Soc Parasitol 1995;25:183–194.
4 Pegelow K, Gross R, Pietrzik K, Lukito W,
Richards AL, Fryauff D: Parasitological and
nutritional situation of school children in the
Sukaraja district, West Java, Indonesia. South-
east Asian J Trop Med Public Health 1997;28:
173–190.
5 Graham CF: A device for the diagnosis of Ente-
robius vermicularis. Am J Trop Med 1941;21:
159–161.
6 Ritchie LS: An ether sedimentation technique
for routine stool examinations. Bull US Army
Med Dept 1948;8:326.
Serum Minerals in Intestinal Parasitic
Infection
7 Young K, Bullock S, Melvin D: Ethyl acetate as
a substitute for diethyl ether in the formalin-
ether sedimentation techniques. J Clin Micro-
biol 1979;10:852–853.
8 Erdman DD: Clinical comparison of ethyl ace-
tate and diethyl in the formalin-ether sedimen-
tation techniques. J Clin Microbiol 1981;14:
483–485.
9 Dawson JB, Walker BE: Direct determination
of zinc in wholes blood plasma and urine by
atomic absorption spectrometry. Clin Chim
Acta 1969;26:465–475.
10 Hill DR: Giardiasis. Issues in diagnosis and
management. Infect Dis Clin North Am 1993;
7:503–525.
11 Wilson WM, Dufour DL, Staten LK, Barac-
Nieto M, Reina JC Spurr GB: Gastrointestinal
parasitic infection, anthropometrics, nutrition-
al status and physical work capacity in Colom-
bian boys. Am J Human Biol 1999;11:763–
771.
12 Koltas IS, Ozcan K, Tanner L, Aksungur P:
Serum copper, zinc and magnesium levels in
children with enterobiosis. J Trace Elem Med
Biol 1997;11:49–52.
Dig Dis 2003;21:258–261
13 Karakas Z, Demirel N, Tarakcioglu M, Mete
N: Serum zinc and copper levels in southeast-
ern Turkish children with giardiasis or amebia-
sis. Biol Trace Elem Res 2001;82:11–18.
14 Ertan P, Yereli K, Kurt O, Balcioglu IC, Onag
A: Serological levels of zinc, copper and iron
elements among Giardia lamblia-infected chil-
dren in Turkey. Pediatr Int 2002;44:286–288.
15 Abel ES, Davids BJ, Robles LD, Loflin CE,
Gillin FD, Chakrabarti R: Possible roles of pro-
tein kinase A in cell motility and excystation of
the early diverging eukaryote Giardia lamblia.
J Biol Chem 2001;276:10320–10329.
16 Cragg RA, Christie GR, Phillips SR, Russi
RM, Kury S, Mathers C, Taylor PM, Ford D: A
novel zinc-regulated human zinc transporter,
hZTL1, is localized to the enterocyte apical
membrane. J Biol Chem 2002;277:22789–
22797.
17 Walker-Smitth JA: Post-infection diarrhea.
Curr Opin Infect Dis 2001:14;567–571.
261
 Original Paper
Dig Dis 2003;21:262–265
DOI: 10.1159/000073345
Impact of Body Mass Index on Fasting
Blood Glucose Concentration
among Helicobacter pylori Carriers
Ioannis D. Kyriazanos a Ioannis Sfiniadakis b Panagiotis Dimakos c
Vasilios Gizaris d Konstantinos Datsakis e Aggeliki Dafnopoulou f
a 2nd
Department of Surgery and Departments of b Pathology, c Endocrinology, d Immunology and e Gastroenterology,
Naval Hospital of Athens, and f Department of Radiology, P. Faliron Medical Center, Naval Hospital of Athens,
Athens, Greece
Key Words
Helicobacter pylori W Fasting blood glucose levels W
Obesity W Body mass index
Abstract
Background/Aims: Despite the fact that Helicobacter py-
lori (Hp) is regarded as a major gastroduodenal patho-
gen, it has recently been suggested to be an important
factor for non-gastroenterologic conditions such as dia-
betes mellitus. Accordingly, it seems that Hp infection
may have implications in glycemic control and in fasting
plasma glucose concentrations. As overnutrition and
obesity are directly related to impaired glucose toler-
ance, the aim of the present study was to determine
whether Hp infection leads to alterations in fasting plas-
ma glucose concentrations of Hp carriers and especially
in relation to their body mass index. Methods: Serum
was obtained from 224 young, male navy recruits. An
enzyme-linked immunosorbent assay to detect Hp-spe-
cific IgG serum antibodies as well as gastroscopy along
with biopsy was used to identify the infected individuals.
Serum levels of glucose, urea, creatinine and uric ac-
id were also determined. Non-fasting subjects and per-
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0262$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
sons with abnormal oral glucose tolerance curve test
were excluded. Results: Among Hp-positive individuals,
obese persons presented with a significantly lower mean
blood glucose level than non-obese persons. Obese Hp-
contaminated participants had significantly lower mean
fasting blood glucose concentrations as well as a signifi-
cantly smaller percentage of participants with abnormal
elevated blood glucose levels than obese participants
negative to Hp infection. Conclusions: Our data suggest
that obesity in combination with Hp infection may induce
an enhanced response to insulin leading to reduced fast-
ing blood glucose levels, among Hp-positive obese per-
sons in comparison to Hp-positive lean persons.
Copyright © 2003 S. Karger AG, Basel
Introduction
Despite the fact that Helicobacter pylori (Hp) is re-
garded as a major gastroduodenal pathogen etiologically
linked with duodenal and gastric disease, it has recently
been suggested to be an important factor for non-gas-
troenterologic conditions such as coronary heart disease
and diabetes mellitus [1]. Although the relationship be-
Ioannis D. Kyriazanos, MD
22, Thetidos str. P. Faliro
GR–17561 Athens (Greece)
Tel. +30 2109845327, Fax +30 2107710671
E-Mail medkyri@yahoo.com
tween diabetes mellitus and Hp remains controversial, it
seems that Hp infection may have implications in gly-
cemic control [2]. Furthermore, Hp infection raises basal
and meal-stimulated serum gastrin concentrations and
lowers iron stores, which may in turn reduce fasting plas-
ma glucose concentrations [3].
Overnutrition and obesity, due to their well-known
complications including heart disease and diabetes melli-
tus, are directly related to impaired glucose tolerance.
Although we have already reported that obesity and
increased BMI cannot be considered as predisposing fac-
tors for Hp contamination [4], it could be possible that the
combination of Hp infection and obesity can facilitate
Hp-related alterations of plasma glucose concentrations.
Accordingly, the aim of the present study was to deter-
mine whether Hp infection leads to alterations in fasting
plasma glucose concentrations of Hp carriers and in rela-
tion to their body mass index (BMI).
Materials and Methods
Subjects
The study population consisted of 224 healthy male Hellenic
navy recruits (median age 22.84 years old, range 20–30). All the sub-
jects were Caucasians and came from different regions of the coun-
try. They were randomly selected out of 600 males inducted into the
Hellenic Navy at the naval base of Salamis. A blood sample was col-
lected in February 2000, during their induction. After centrifugation,
the serum sample was split into three aliquots, which were frozen at
–70 ° C and thawed once, before analysis.
Determination of Biochemical Parameter Concentrations and
Hp Status
Titers of Hp-specific IgG antibodies, as well as serum levels of
glucose, urea, creatinine and uric acid were determined in the
Department of Immunology of Naval Hospital of Athens. Individu-
als with abnormally increased blood glucose concentrations under-
went an oral glucose tolerance curve test. Persons with abnormal test
results as well as non-fasting subjects were excluded from the study.
For the serological determination of the specific antibody against Hp,
an enzyme-linked immunosorbent assay was used (Hp IgG Elisa Kit,
Hycor Biomedical GmbH, Kassel, Germany; inter-assay coefficient
of variation: 7%, intra-assay coefficient of variation: 4.3%, sensitivi-
ty: 0.1 IU/ml). Levels of IgG were categorized as seropositive for Hp
according to an OD titer of IgG antibodies to Hp of 1 0.5. The normal
range for blood glucose was 60–110 mg/dl, for serum urea 10–40 mg/
dl, for creatinine ! 1.2 mg/dl and for uric acid up to 5.3 mg/dl. All
seropositive subjects underwent endoscopy in the Department of
Gastroenterology at the Naval Hospital of Athens. Along with the
examination, two biopsy samples of the antrum and one of the cor-
pus were obtained from each of the individuals, for the verification of
Hp infection. Subjects with both sero-surveillance- and biopsy-posi-
tive results were finally considered as Hp-contaminated individuals.
Fasting Blood Glucose Concentrations and
H. pylori Infection
Obesity Definition
More specifically by calculating the BMI, we created two distinct
groups: the group of overweight individuals with BMI 625 kg/m2,
and the lean group with BMI ! 25 kg/m2. The cut-off point of ! 25 or
625 kg/m2 was proposed in the first federal obesity clinical guide-
lines released by NHLBI and NIDDK in 1998 as the most acceptable
classification for obesity [5].
Statistical Analyses
To examine whether the Hp-affected members differed from the
non-affected members for each study variable, ¯2, Fischer’s exact and
Student’s t tests were used from the SPSS statistical package version
8 (SPSS, Chicago, Ill., USA, 1997).
Results
Among several biochemical parameters, fasting blood
glucose revealed interesting results. Blood glucose levels
were not significantly different between either obese or
non-obese participants (104.67 B 11.6 vs. 104.4 B 11.7,
p = 0.86), nor between Hp-positive and Hp-negative
members (103.5 B 8.3 vs. 104.9 B 12.6, p = 0.79).
Abnormal elevated fasting blood glucose levels (nor-
mal range 70–110 mg/dl) were detected in 52 (23.2%) out
of 224 recruits and the number of abnormal blood glucose
carriers was comparable between obese and non-obese
(28/137, 20.4% vs. 24/87, 27.6%, p = 0.25) as well as
between Hp-positive and Hp-negative (17/61, 27.8% vs.
35/163, 21.4%, p = 0.72) groups.
Subsequently and by stratifying the participants in two
groups according to their Hp status (positive or negative),
we further evaluated the relation between fasting blood
glucose levels and BMI for each group separately. Among
163 uninfected participants (Hp-negative group) there
was no significant difference between obese and non-
obese persons either in the mean blood glucose levels
(105.8 B 12.7 vs. 103.3 B 12.7, 95% CI = –1.5 to 6, p =
0.22) or in the number of individuals with abnormal
blood glucose levels (23/102, 22.5% vs. 12/61, 19.7%, p =
0.85). On the contrary, among 61 Hp-positive individu-
als, obese persons presented with a significantly lower
mean blood glucose level than non-obese persons (101.2
B 6.4 vs. 106.9 B 8.6, 95% CI = – 9.5 to –1.8, p = 0.005).
Furthermore, a significantly higher number of non-obese
individuals had blood glucose levels above the normal
range (1110 mg/dl), in comparison to obese members in
the Hp-positive group of participants (12/26, 46.1% vs.
5/35, 14.2%, p = 0.009).
Among obese participants, Hp-positive individuals
had a significantly lower mean blood glucose level (101.2
B 6.4 vs. 105.8 B 12.7, 95% CI = 1.2–7.9, p = 0.007) as
Dig Dis 2003;21:262–265 263
well as a significantly smaller percentage of members with
abnormal elevated fasting blood glucose concentrations
(10.7% vs. 29.4%, p = 0.032) than obese Hp-negative per-
sons. The same was not true among lean individuals.
Discussion
Hp colonization of the human stomach produces an
innate host immune response, which finally stimulates G
cells in production of gastrin and gastric acid [6, 7].
Hypergastrinemia becomes more enhanced by the ago-
nistic action of the microorganism to the H3 histamine
receptors and its inhibitory effect to CCK and somatosta-
tin secretion that normally diminish gastrin release [8].
The increased levels of gastrin and the diminished soma-
tostatin release, in combination with others hormones col-
lectively known as the ‘enteroinsular axis’, could poten-
tate insulin secretion, increase plasma insulin levels and
as a result diminish fasting blood glucose levels [9], as
somatostatin normally has an inhibitory effect on insulin
release. Additionally, insulin can directly stimulate epi-
nephrine release, which may induce hypoglycemia [10].
Accordingly, Hp infection can affect glucose metabolism
and Hp-positive individuals might have higher insulin
and lower glucose plasma levels in comparison to Hp-neg-
ative persons. It is already identified that the inflammato-
ry reaction, which Hp infection produces, could be delete-
rious for the control of glucemia, especially for patients
with diabetes, as well as that Hp gastritis may contribute
to postprandial hypoglycemia due to hyperinsulinemia
[11–13]. Peach and Barnett [3] reported that Hp infection
may lead to a lower fasting plasma glucose concentration
among women.
In the present study we could not identify any signifi-
cant difference in fasting blood glucose levels between
Hp-positive and Hp-negative participants. Hp infection
demonstrated its role in controlling fasting blood glucose
concentrations only when it was combined with obesity.
Among Hp-infected individuals, obese persons presented
with a significantly lower mean blood glucose level than
non-obese persons, further supporting the finding that
obese Hp-positive individuals had a significantly lower
mean blood glucose level than obese Hp-negative persons.
Among Hp non-infected individuals, the mean serum glu-
cose level was higher (even not statistically significant) for
obese than lean persons as it is well known that obesity
can be accompanied by alterations in glucose metabolism
[14].
264 Dig Dis 2003;21:262–265
The identification and sequencing of the mouse obese
(ob) gene by Friedman’s group in 1994 opened important
new avenues in obesity research. The ob gene encodes the
protein leptin, which is produced in the adipocytes,
secreted into the blood and informs the brain about the
size of the fat mass. It is known that plasma leptin highly
correlates to body fat content and BMI, with obese per-
sons having increased leptin levels [15]. Accumulating
evidence suggests that the role of leptin is much broader
than that of an anti-obesity hormone; leptin exerts several
other metabolic effects on peripheral tissue including
modification of the insulin action. Supraphysiologic lev-
els of insulin markedly increase circulating leptin levels,
while administration of leptin improves insulin resistance
in mice and insulin-resistance has been associated with
increased leptin levels [16, 17]. Although the mechanism
underlying the changes of leptin induced by insulin and
vice versa remains to be studied in more detail, these data
demonstrate a significant relationship between leptin and
insulin.
Insulin and corticosteroids are obvious candidates sig-
naling fat cells for leptin production. Thus, the increased
leptin expression observed in obesity could result from
chronic hyperinsulinemia and increased cortisol turnover
[18] as insulin plasma levels in obese individuals are ele-
vated compared to those in normal subjects [19, 20]. The
fact that in normoglycemic obese persons, leptin levels are
higher than the leptin levels in obese persons with dia-
betes, can support the option that normally, leptin levels
are negatively correlated with fasting glucose levels [14,
21, 22].
Taking all these points into consideration and in accor-
dance with our results, it could be possible that in obese
persons, stimulated sympathetic activity, hyperleptinem-
ia and its related hyperinsulinemia, can be combined with
the increment of gastrin, reduction of somatostatin and
increased histamine resulting from Hp infection, with
finally further hyperinsulinemia. This combined incre-
ment of insulin might be a reason for the lower fasting
serum glucose values that we recognized in obese than in
lean Hp-positive persons. Alterations of glucose metabo-
lism in Hp colonization need further investigation.
Kyriazanos/Sfiniadakis/Dimakos/Gizaris/
Datsakis/Dafnopoulou
 References
1 Senturk O, Canturk Z, Cetinarslan B, Ercin C,
Hulagu S, Canturk NZ: Prevalence and com-
parison of five different diagnostic methods for
Helicobacter pylori in diabetic patients. Endocr
Res 2001;27:179–189.
2 De Luis DA, Cordero JM, Caballero C, Boixe-
da D, Aller C, Canton R, De la Calle H: Effect
of the treatment of Helicobacter pylori infec-
tion on gastric emptying and its influence on
the glycaemic control in type 1 diabetes melli-
tus. Diabetes Res Clin Pract 2001;52:1–9.
3 Peach HG, Barnett NE: Helicobacter pylori in-
fection and fasting plasma glucose concentra-
tion. J Clin Pathol 2001;54:446–449.
4 Kyriazanos ID, Sfiniadakis I, Gizaris V, Houn-
tis P, Hatziveis K, Dafnopoulou A, Datsakis K:
The incidence of Helicobacter pylori infection
is not increased among obese young individu-
als in Greece. J Clin Gastroenterol 2002;34:
541–546.
5 NIH: Clinical Guidelines on the Identification,
Evaluation and Treatment of Overweight and
Obesity in Adults. Bethesda, NIH, 1998.
www.nhlbi.nih.gov/guidelines
6 Van Den Brink GR, ten Kate FJ, Ponsioen CY,
Rive MM, Tytgat GN, van Deventer SJ, Pep-
pelenbosch MP: Expression and activation of
the NF-ÎB in the antrum of the human stom-
ach. J Immunol 2000;164:3353–3359.
Fasting Blood Glucose Concentrations and
H. pylori Infection
7 Rose S: Gastrointestinal and Hepatobilliary
Pathophysiology. Philadelphia, Fence Creek
Publishing, 1998, pp 161–172.
8 Calam J, Gibbons A, Haley Z, Bliss P, Arebi N:
How does Helicobacter pylori cause mucosal
damage? Its effect on acid and gastrin physiolo-
gy. Gastroenterology 1997;113:S43–S49.
9 Acbay O, Celik AF, Gundogdu S: Does Helico-
bacter pylori-induced gastritis enhance food-
stimulated insulin release? Dig Dis Sci 1996;
41:1327–1331.
10 Landsberg L, Krieger DR: Obesity, metabolism
and the sympathetic nervous system. Am J
Hypertens 1989;2:125S–132S.
11 Acbay O, Celik AF, Kadioglu P, Goksel S,
Gundogdu S: Helicobacter pylori-induced gas-
tritis may contribute to occurrence of postpran-
dial symptomatic hypoglycemia. Dig Dis Sci
1999;44:1837–1842.
12 Perdichizzi G, Bottari M, Pallio S, Fera MT,
Carbone M, Barresi G: Gastric infection by
Helicobacter pylori and antral gastritis in hy-
perglycemic obese and in diabetic subjects.
New Microbiol 1996;19:149–154.
13 Begue RE, Mirza A, Compton T, Gomez R,
Vargas A: Helicobacter pylori infection and in-
sulin requirement among children with type 1
diabetes mellitus. Pediatrics 1999;103:e83.
14 Zimmet P, Hodge A, Nicolson M, Staten M, de
Courten M, Moore J, Morawiecki A, Lubina J,
Collier G, Alberti G, Dowse G: Serum leptin
concentration, obesity and insulin resistance in
Western Samoans: Cross-sectional study. BMJ
1996;313:965–969.
Dig Dis 2003;21:262–265
15 Janeckowa R: The role of leptin in human
physiology and pathophysiology. Physiol Res
2001;50:443–459.
16 Mantzoros CS: The role of leptin in human
obesity and disease: A review of current evi-
dence. Ann Intern Med 1999;130:671–680.
17 Wolf G, Chen S, Han DC, Ziyadeh FN: Leptin
and renal disease. Am J Kidney Dis 2002;39:
1–11.
18 Fried SK, Ricci MR, Russell CD, Laferrere B:
Regulation of leptin production in humans. J
Nutr 2000;130:3127S–3131S.
19 Rasmussen MH, Hvidberg A, Juul A, Main
KM, Gotfredsen A, Skakkebaek NE, Hilsted J:
Massive weight loss restores 24-hour growth
hormone release profiles and serum insulin-
like growth factor-I levels in obese subjects. J
Clin Endocrinol Metab 1995;80:2446.
20 Larsson H, Elmstahl S, Ahren B: Plasma leptin
levels correlate to islet function independently
of body fat in postmenopausal women. Dia-
betes 1996;45:1580–1584.
21 Sorensen IT, Echwald MS, Holm JC: Leptin in
obesity. BMJ 1996;313:953–954.
22 Clement K, Lahlou N, Ruiz J, Hager J, Boug-
neres P, Basdevant A, Guy-Grand B, Froguel
P: Association of poorly controlled diabetes
with low serum leptin in morbid obesity. Int J
Obes Relat Metab Disord 1997;21:556–561.
265
 Original Paper
Dig Dis 2003;21:266–270
DOI: 10.1159/000073346
Selenium Is Depleted in Crohn’s Disease
on Enteral Nutrition
Fumitoshi Kuroki Takayuki Matsumoto Mitsuo Iida
Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University,
Fukuoka City, Japan
Key Words
Selenium W Crohn’s disease W Enteral nutrition
Abstract
Backgroud/Aims: Selenium is an important trace ele-
ment and its deficiency has been reported to be associat-
ed with cardiomyopathy or gastrointestinal cancer. The
aim of this study is to clarify the selenium status in
Crohn’s disease (CD) on enteral nutrition. Methods: We
measured serum selenium concentrations in 53 patients
with CD and compared them with those in 21 healthy
controls. Twenty-nine patients were under the treatment
by enteral nutrition (EN group), and the remaining 24
patients were free from formulated enteral nutrition
(non-EN group). Results: While the serum selenium con-
centration in the non-EN group was not decreased when
compared to controls, the value in the EN group was sig-
nificantly lower than those in the non-EN group and in
controls. Clinical manifestations of selenium deficiency
were found in a patient on exclusive enteral nutrition. In
the EN group, the serum selenium concentration
showed an inverse correlation with the duration and the
daily dose of enteral nutrition. In the non-EN group, the
serum selenium concentrations were inversely corre-
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0266$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
lated with the Crohn’s disease activity index. Conclusion:
These findings suggest that patients with CD on enteral
nutrition are at risk for selenium deficiency and that even
patients without enteral nutrition may develop selenium
deficiency at the active phase of the disease.
Copyright © 2003 S. Karger AG, Basel
Introduction
Selenium is an essential trace element which is a com-
ponent of enzyme glutathione peroxidase. The enzyme
protects membrane structures by detoxifying oxygen spe-
cies. Because selenium is contained in various foods, defi-
ciency of the trace mineral rarely occurs, except in sub-
jects who reside in areas where the soil contains a low
amount of selenium and for patients on long-term paren-
teral nutrition [1, 2].
Because patients with Crohn’s disease (CD) usually
have various nutritional disturbances, parenteral nutri-
tion or enteral nutrition (EN) are applied for the manage-
ment of malnutrition. In addition, EN has been reported
to be one of the primary therapies for active CD [3, 4] and
widely accepted for prolongation of remission in children
and adolescents [5]. On such occasions, subnormal seleni-
Fumitoshi Kuroki
Department of Medicine and Clinical Science
Graduate School of Medical Sciences, Kyushu University
3-1-1 Maidashi, Higashiku, Fukuoka City 812-8582 (Japan)
Tel. +81 982 550505, Fax +81 982 550500, E-Mail fkuro60@f3.dion.ne.jp
Table 1. Details of the patients’ clinical characteristics at the time of investigation

Controls Patients
(n = 21)
EN group (n = 29) non-EN group (n = 24)

Age, years 22–46 (30.0B6.3) 16–77 (29.8B11.7) 16–53 (30.6B10.4) NSa

Sex, m/f 14/7 25/4 18/6 NSa
Body mass index 15.8–22.7 (20.0B2.0) 14.8–24.2 (19.2B2.2) 15.1–24.3 (19.2B2.5) NSa
Time intervalc 0.2–18 (7.1B5.6) 0.1–28 (6.6B7.1) NSa
CD activity index 13–377 (127B98) 15–414 (129B106) NSa
Bowel involvement
Small bowel only 10 11
Large bowel only 2 5 NSb
Small and large bowel 17 8
History of intestinal resection 9 5 NSb
Current medical treatments
Salazosulfapyridine 10 2
Prednisolone 1 2
None 18 22

Values are expressed as the range (mean B SD).

a Comparison among the EN group, non-EN group, and controls.
b Comparison between the EN group and non-EN group.
c Time interval from onset of symptoms to diagnosis, in years.

um status in CD patients on total parenteral nutrition has
already been reported [6, 7]. However, there have been a
few reports concerning the selenium status in CD patients
on EN [8, 9]. The aim of this study is to investigate the
selenium status in CD patients with a reference to EN.
Subjects and Methods
Subjects
Fifty-three patients with an established diagnosis of CD and 21
healthy control subjects were enrolled for the present study. The diag-
nosis of CD was based on confirmation of clinical, radiographic, or
endoscopic findings and histologic features characteristic of this dis-
ease. None of the patients had steatorrhea, which is a major clinical
manifestation of malabsorption.
CD patients were divided into two groups – EN and non-EN. EN
was used for primary therapy for active disease and for prolongation
of remission. The EN group was composed of 29 patients who had
been treated by EN. The non-EN group consisted of 24 patients who
had been ingesting ordinary Japanese foods without any formulated
diet. Ten of them had not been treated by any medical and nutrition-
al treatments, and the others had rejected EN. Details of the patients’
clinical characteristics at the time of investigation are shown in
table 1. In the EN group, 7 patients had been ingesting elemental diet
(Elental, Roussel Morishita, Osaka, Japan), 7 had been ingesting
polymeric diet (Ensure Liquid, Dainabot, Osaka, Japan), and 15 had
been ingesting both. Among the EN group, 22 had been ingesting
ordinary foods, but 3 had been on exclusive EN without any food
intake. The duration of EN ranged from 0.1 to 5.3 years (1.5 B 1.3
years, mean B SD), and the daily dose of EN ranged from 500 to
2,300 kcal (1,384 B 483 kcal).
Twenty-one normal healthy volunteers participated in this study.
They were randomly selected from our laboratory staff, according to
the following criteria: (1) completely free from previous or present
illness, (2) normal physical examination and blood chemistry, and
(3) consumption of ordinary Japanese food. Neither the CD patients
nor controls had been supplied with trace minerals at the time of
selenium measurement. Informed consent was obtained from each
patient and control prior to blood sampling.
Disease Activity
Activity of CD at the time of investigation was calculated, accord-
ing to the formula described as Crohn’s disease activity index (CDAI)
[10]. A cut-off value of CDAI 150 was used to divide each patient
into active (CDAI 1 150) or inactive (CDAI ^ 150) disease.
Measurement of Serum Selenium Concentration
After an overnight fast, blood samples were obtained from each
patient and control. Blood was centrifuged for separation of serum.
After the initial reduction of the serum sample and treatment with
the palladium modifier, the concentrations of selenium were deter-
mined by graphite-furnace atomic absorption spectrometry using
Model AA-40G atomic absorption spectrophotometer (Varian Cana-
da Inc., Georgetown, Ont., Canada) [11]. In addition, hematocrit val-
ue, erythrocyte sedimentation rate, serum albumin and C-reactive

Selenium and Crohn’s Disease Dig Dis 2003;21:266–270 267


Fig. 1. Comparison of serum selenium concentrations among the EN
group, non-EN group, and controls.
protein were measured and body mass index (BMI) was calculated as
follows: weight (kg)/height (m)2.
Statistical Analysis
Results are expressed as mean B SD. Comparisons among the
EN group, non-EN group and the controls were performed by one-
way analysis of variance with ‘a posterior’ Scheffé test. Correlation
coefficients (r) between serum selenium concentrations and various
parameters in each CD group were calculated by regression analysis.
Probabilities ! 0.05 were considered significant.
Results
Age, gender and BMI values were not different among
the two CD groups and the controls (table 1). Serum albu-
min concentration in the CD groups showed lower values
than the controls (p ! 0.0001), but the value was not dif-
ferent between the two CD groups (EN group, 3.9 B
0.6 mg/dl; non-EN group, 3.8 B 0.6 mg/dl; controls, 4.8
B 0.2 mg/dl).
Comparison of Selenium Status
The selenium status of the CD patients and controls
are shown in figure 1. The serum selenium concentration
in the EN group (7.1 B 3.5 Ìg/dl) was significantly lower
than in the non-EN group (10.2 B 2.0 Ìg/dl, p ! 0.0001)
and in controls (11.8 B 1.4 Ìg/dl, p ! 0.0001). In particu-
268 Dig Dis 2003;21:266–270
Fig. 2. a Thumbnails of the patient prior to selenite supplementa-
tion. The nails are rough and wavy. b The rough and wavy changes of
the nails have improved after selenite supplementation.
lar, the serum selenium concentration in 3 patients with
exclusive EN showed extremely low values (0.6, 0.7 and
2.9 Ìg/dl, respectively). One of them had been clinically
diagnosed as having selenium deficiency because of im-
paired serum glutathione peroxidase activity, cardiac dys-
function, fingernail deformities, and subsequent im-
provement of clinical symptoms by selenite supplement
(fig. 2). In each CD group, the serum selenium concentra-
tion was not different according to the site of the disease
and to prior history of the intestinal resection.
The CDAI ranged from 13 to 414 (128 B 101). As
indicated in table 1, the CDAI was not different between
the EN and non-EN group. The serum selenium concen-
tration was not different between active CD and inactive
CD in each group.
When possible, the correlation between serum seleni-
um concentration and CDAI was further investigated, the
selenium concentration was inversely correlated with
CDAI in the non-EN group (r = –0.513, p ! 0.01) (fig. 3),
but such a significant correlation was not found in the EN
group. In contrast, the serum selenium concentration in
the EN group correlated with daily dose (r = –0.374, p !
0.05) and duration of EN (r = –0.433, p ! 0.05) with sta-
tistical significance (fig. 4). In both EN and non-EN
groups, the serum selenium concentration did not show
any significant correlation with serum albumin levels,
BMI, C-reactive protein or erythrocyte sedimentation
rate values.
Kuroki/Matsumoto/Iida
Fig. 3. Correlation between serum selenium concentration and
CDAI in the non-EN group (r = –0.513, p ! 0.01).

Discussion

Cardiomyopathy, weakness in proximal muscle, eleva-

tion of liver transaminase and creatine kinase activities,
and nail changes are the major clinical manifestations of
selenium deficiency [12]. In addition, selenium depletion
has been known to be associated with an increased risk for
gastrointestinal and prostatic cancer [13]. Because cardio-
myopathy in selenium deficiency is irreversible and fatal
[2], early recognition of the condition seems to be inevita-
ble. In clinical situations, selenium status has generally Fig. 4. Correlation between serum selenium concentration and daily
been assessed by serum or erythrocyte glutathione peroxi- dose of EN (a) (r = –0.374, p ! 0.05) or duration of EN (b) (r =
–0.433, p ! 0.05) in the EN group.
dase activities. Because a significant correlation between
serum selenium concentration and glutathione peroxi-
dase activity has been established, the former applied for
the present investigation has been regarded to be appro-
priate for the assessment of selenium status [14].
CD has been known to be accompanied by various patients with severe disease may develop profound seleni-
nutritional disturbances. Low selenium concentrations in um deficiency.
active CD have been shown in previous investigations Parenteral nutrition and EN are established strategies
[14–19], and insufficient oral intake and/or malabsorp- for the management of CD [1–9, 14]. These nutritional
tion have been considered to be the main cause of low therapies contribute to the improvement of nutritional
selenium levels. In this study, the selenium concentration status, intestinal lesions and disease activity [3, 4]. In par-
in the non-EN group was not different from controls. The ticular, EN has been accepted to prolong the remission of
insignificant difference in selenium concentration can be CD in children and in adolescents [5]. Because most of the
explained by the inclusion of inactive or mild disease. enteral formulas prevailing in Japan or in the USA do not
However, since the selenium concentration showed in- contain sufficient selenium [20–22], patients on EN are at
verse correlation with CDAI in the non-EN group, CD high risk of selenium deficiency. In this study, the serum

Selenium and Crohn’s Disease Dig Dis 2003;21:266–270 269

selenium concentration in the EN group was significantly
lower than in the non-EN group and controls. Because
there were no differences in clinical or nutritional param-
eters other than the diet between the two CD groups,
insufficient selenium content in the enteral formulas
seems to be the major cause for decreased selenium con-
centration in the EN group.
It has been reported that Elental and Ensure Liquid
used as the formula of EN for our patients contain 1.6 and
7.3 Ìg/1,000 kcal of selenium respectively [22]. Because
the recommended dietary allowance of selenium has been
established to be 0.87 Ìg/kg/day by the US Research
Council [23], selenium intake with the formula of EN
alone would be much lower than the recommended value.
In fact, selenium concentrations in 3 cases on exclusive
EN showed an extraordinarily low value, and a patient
had actually developed clinical manifestations of seleni-
um deficiency. In addition, the selenium concentration
was even low in patients who ingested both EN and ordi-
nary food. Because the selenium levels were inversely cor-
related with the dose and the duration of EN, it would be
necessary to take account of selenium deficiency in CD
patients under a prolonged period of EN even if ordinary
food is not prohibited.
In conclusion, this study indicated that selenium defi-
ciency possibly occurs in patients with CD, especially in
those on EN. While selenium supplement seems to be nec-
essary for a certain proportion of CD patients, the optimal
dose in consideration of possible toxic effects should fur-
ther be elucidated.

References

1 Van Riji AM, Thomson CD, Mckenzie JM,
Robinson MF: Selenium deficiency in total
parenteral nutrition. Am J Clin Nutr 1979;32:
2076–2085.
2 Fleming CR, Lie JT, McCall JT, OBrien JF,
Baillie EE, Thistle JL: Selenium deficiency and
fatal cardiomyopathy in a patients on home
parenteral nutrition. Gastroenterology 1982;
83:689–693.
3 González-Huix F, de León R, Fernández-Ba-
ñares F, Esteve M, Cabré E, Acero D, Abad-
Lacruz A, Figa M, Guilera M, Planas R, Gas-
sull MA: Polymeric enteral diet as primary
treatment of active Crohn’s disease: A prospec-
tive steroid controlled trial. Gut 1993;34:778–
782.
4 Hiwatashi N: Enteral nutrition for Crohn’s dis-
ease in Japan. Dis Colon Rectum 1997;40:
S48–S53.
5 Wilschanski M, Sherman P, Pencharz P, Davis
L, Corey M, Griffiths A: Supplementary enter-
al nutrition maintains remission in paediatric
Crohn’s disease. Gut 1996;38:543–548.
6 Fleming CR, McCall JT, O’Brien JF, Forsman
RW, Ilstrup DM, Petz J: Selenium status in
patients receiving home parenteral nutrition. J
Parenter Enteral Nutr 1984;8:258–262.
7 Jacobson S, Plantin LO: Concentration of sele-
nium in plasma and erythrocytes during paren-
teral nutrition in Crohn’s disease. Gut 1985;26:
50–54.
8 Fernández-Bañares F, Mingorance MD, Esteve
M, Cabré E, Lachica M, Abad-Lacrus A, Gil A,
Humbert P, Boix J, Gassull MA: Serum zinc,
copper, and selenium levels in inflammatory
bowel disease: Effect of total enteral nutrition
on trace element status. Am J Gastroenterol
1990;85:1584–1589.
9 Thomas AG, Miller V, Shenkin A, Fell GS,
Taylor F: Selenium and glutathione peroxidase
status in paediatric health and gastrointestinal
disease. J Pediatr Gastroenterol Nutr 1994;19:
213–219.
10 Best WR, Becktel JM, Singleton JW, Kern F:
Development of a Crohn’s disease activity in-
dex. Gastroenterology 1976;70:439–444.
11 Jacobson BE, Lockitch G: Direct determina-
tion of selenium in serum by graphite-furnace
atomic absorption spectrometry with deuter-
ium background correction and a reduced pal-
ladium modifier: Age-specific reference ranges.
Clin Chem 1988;34:709–714.
12 Lockitch G: Selenium: Clinical significance
and analytical concepts. Crit Rev Clin Lab Sci
1989;27:483–541.
13 Willett WC, Morris JS, Pressel S: Prediagnostic
serum selenium and risk of cancer. Lancet
1983;ii:130–134.
14 Rannem T, Ladefoged K, Hylander E, Hegnhøj
J, Staun M: Selenium depletion in patients with
gastrointestinal disease: Are there any predic-
tive factors? Scand J Gastroenterol 1998;33:
1057–1061.
15 Penny WJ, Mayberry JF, Aggett PJ, Gilbert JO,
Newcombe RG, Rhodes J: Relationship be-
tween trace elements, sugar, consumption and
taste in Crohn’s disease. Gut 1983;24:288–
292.
16 Loeschke K, König A, Haeberlin SH, Lux F:
Low blood selenium concentration in Crohn
disease. Ann Intern Med 1987;106:908.
17 Hinks LJ, Inwards KD, Lloyd B, Clayton B:
Reduced concentrations of selenium in mild
Crohn’s disease. J Clin Pathol 1988;41:198–
201.
18 Rannen T, Ladefoged K, Hylender E, Hegnhøt
J, Jarnum S: Selenium status in patients with
Crohn’s disease. Am J Clin Nutr 1992;56:933–
937.
19 Geerling BJ, Badart-Smook A, Stockbrügger
RW, Brummer RJM: Comprehensive nutri-
tional status in patients with long-standing
Crohn disease currently in remission. Am J
Clin Nutr 1998;67:919–926.
20 Zabel NL, Harland J, Gormican A, Ganther
HE: Selenium content of commercial formula
diets. Am J Clin Nutr 1978;31:850–858.
21 Martin RF, Young VR, Janghorbani M: Seleni-
um content of enteral formulas. J Parenter En-
teral Nutr 1986;10:213–215.
22 Tanaka T, Higashi A, Matsuda I, Suzuki I, Asa-
kawa M: Selenium content of Japanese enteral
formulas (Engl abstract). J Jpn Soc Nutr Food
Sci 1995;48:147–50.
23 Food and Nutritional Board, National Re-
search Council: Trace Elements. Recom-
mended Dietary Allowances, ed 10. Washing-
ton, National Academy Press, 1989, pp 217–
224.

270 Dig Dis 2003;21:266–270 Kuroki/Matsumoto/Iida

 Original Paper
Dig Dis 2003;21:271–275
DOI: 10.1159/000073347
L-Carnitine in the Treatment of Mild or
Moderate Hepatic Encephalopathy
Mariano Malaguarnera a Giovanni Pistone a Marinella Astuto c
Simona Dell’Arte a Giovanna Finocchiaro a Emilia Lo Giudice a
Giovanni Pennisi b
a Dipartimento di Scienze della Senescenza, Urologiche e Neurologiche, Università degli Studi di Catania, e
b Dipartimenti di Neuroscienze e c Specialità Medico-chirurgiche Sez. Anestesiologia, Università di Catania, Italia
Key Words
Hepatic encephalopathy W L-Carnitine treatment
Abstract
Hepatic encephalopathy (HE) is one of the major compli-
cations of cirrhosis. Experimental and clinical findings
observed in liver, muscle and brain have provided new
insights into the ammonia mechanism of action. L-Carni-
tine (LC), inducing ureagenesis, may decrease blood and
brain ammonia levels. 120 patients meeting inclusion
criteria were randomized either to a treatment for 60
days with LC or placebo (2 g twice a day). Previous stud-
ies have reported a significant protective effect of LC in
mice and rats, which is associated with a significant
reduction of blood and brain ammonia concentration,
suggesting an action of LC either at peripheral or central
sites. Results of our study show a protective effect of LC
in ammonia-precipitated encephalopathy in cirrhotic pa-
tients. Either in subjects with HE 1 or 2 we observed a
significant reduction at day 30 and more markedly at day
60 of treatment. A significant therapeutic effect of LC was
also observed in the NCT-A, which is an accepted and
reliable psychometric test for the assessment of mental
function in cirrhotic patients with HE.
Copyright © 2003 S. Karger AG, Basel
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0271$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
Introduction
Hepatic encephalopathy (HE) is one of the major com-
plications of cirrhosis. Clinical grading is based on im-
paired mental function, neurological disorders and al-
tered states of consciousness, to stupor and coma. Patho-
genesis of this common syndrome remains partially un-
known. Hyperammoniemia, due to the several mecha-
nisms proposed, is one of the main pathogenetic factors in
the development of HE, as shown by the association of
increased serum ammonia levels with impairment of
mental function in patients with liver failure. Experimen-
tal and clinical findings observed in liver, muscle and
brain have provided new insights into the ammonia
mechanism of action. In fact, these organs play a funda-
mental role in the removal of ammonia in cases of hyper-
ammoniemia [1, 2].
A decrease in the muscle mass is a jeopardizing factor
for the development of encephalopathy [3]. Muscle mass
loss is also associated with a decreased ability to change
ammonia in glutamine. All morphological and functional
characteristics of the brain in experimental models and
humans rule out an initial involvement of neurons in HE,
the astrocytes being the first target of the disease [4]. As
showed by Ullian et al. [5], these cells play a key role in the
correct development and function of neurons. In HE, no
Mariano Malaguarnera, AP
Viale Andrea Doria, 69
IT–95126 Catania (Italy)
Tel. +39 095 7262008, Fax +39 095 7262011
E-Mail malaguar@mbox.unict.it
morphological neuronal alterations are detected, changes
in astrocyte characteristics being present (Alzheimer type
II cells) [6]. The majority of therapeutic measures current-
ly in use are therefore aimed at reduction of serum ammo-
nia levels, by decreasing enteric ammonia production [7].
Carnitine is a natural substance involved in regulating
substrate flux and energy balance across cell membranes.
It is a cofactor for the shuttle mechanism whereby long-
chain fatty acids are transformed into L-acetylcarnitine
metabolites and may be carried out into mitochondria in
order to be part of ß-oxidation [8]. Furthermore, carnitine
and the carnitine acetyltransferase are involved in the fol-
lowing reaction: acetyl-coenzyme A + carnitine = acetyl-
carnitine + coenzyme A. This reaction modulates the
intracellular concentration of coenzyme A and acetyl-
CoA. By this reaction, carnitine produces free CoA for
other metabolic reactions and reduces the ratio of acetyl-
CoA to CoA. This reduction stimulates the activity of
pyruvate dehydrogenase and, thus, enhances the oxida-
tive use of glucose.
O’Connor et al. [9] showed a protective effect of L-
carnitine (LC) administration to mice 1 h before a lethal
injection of ammonium acetate. Therrien et al. [10]
reported that LC in the portocaval shunted rat prevented
not only the development of severe encephalopathy, but
also reduced mortality. It was suggested that LC, inducing
ureagenesis, may decrease blood and brain ammonia lev-
els. In order to assess the clinical efficacy of LC in the
treatment of HE, a randomized, double-blind, placebo-
controlled study with oral administration in cirrhotic
patients with hyperammoniemia and chronic symptoms
has been carried out.
The aim of our study was to evaluate in a practice-
adapted design the influence of LC using the number con-
nection test A (NCT-A), mental conditions and ammonia
effects.
Patients and Methods
120 randomly selected patients (9 alcoholics, 33 hepatitis B virus
infected, 63 hepatitis C virus infected, 15 cryptogenetic cirrhosis)
met the following inclusion criteria and were enrolled in the study:
(1) chronic hepatitis with spontaneous manifest HE (mental state
grade 1 or 2 according to the West Haven criteria) and an NCT-A
performance time 1 30 s; (2) hyperammoniemia (venous ammonia
concentration 1 50 mmol/l), and (3) cooperative, hospitalized adult
patients with liver cirrhosis diagnosed by clinical, histological and
ultrasonographic findings.
Exclusion criteria were the following: (1) major complications of
portal hypertension, such as gastrointestinal blood loss, hepatorenal
syndrome or bacterial peritonitis; (2) acute superimposed liver inju-
272 Dig Dis 2003;21:271–275
ry; (3) patient with other neurological disease and metabolic disor-
ders such as alcoholism, diabetes mellitus, unbalanced heart failure
and/or respiratory failure or end-stage renal disease; (4) severe HE
(mental state grade 3–4); (5) administration of anti-HE medications
such as neomycin, lactulose, lactitol, branched-chain amino acids;
(6) any additional precipitating factors such as high protein intake
(additional high-protein meals), constipation or intake of psycho-
stimulants, sedatives, antidepressants, benzodiazepines, or benzodi-
azepine antagonists (flumazenil), and (7) patients with fever, sepsis
or shock were also excluded to avoid variations caused by body tem-
perature.
Study Design
Patients meeting inclusion criteria were randomized in a double-
blind fashion to either a treatment for 60 days with LC (2 g twice a
day) (group A) or placebo (which were identical in appearance and
packaging) (group B). Randomization was based on a computer-gen-
erated list. The consumption of oral LC at a dose of 4 g/day divided
into two doses (2 g in the morning and 2 g in the evening) for 60 days
met the 100% of compliance. A minimum of 210 pills (each pill 1 g)
was considered as good compliance. Concomitant medications
(whose administration continued throughout the study) included
diuretics (36 of group A and 30 of group B) and ß-blockers (27 of
group A and 25 of group B).
All study series were subdivided into groups belonging to HE 1 or
HE 2 according to the initial HE grade (West Haven criteria). Group
A was composed of patients with initial HE 1 (LC 35 patients, place-
bo 33 patients); group B with initial HE 2 (LC 25 patients, placebo 27
patients). The effectiveness of therapy was compared and evaluated
separately in the different subgroups.
The groups were homogeneous with regard to anamnestic and
diagnostic criteria. Differences in the composition of the two groups
with respect to precipitant factors may be minimized, because the
patient population was well defined by inclusion and exclusion crite-
ria. Any pathological, clinical or laboratory findings observed during
the study were monitored and documented until their normaliza-
tion.
NCT-A
The NCT-A measures cognitive and motor ability, and consists of
a combination of number series on a sheet with best celerity possible.
A decrease in the time employed to complete the NCT-A reflects an
improvement of the neurological function.
Venous Ammonia Concentration
The blood ammonia levels were evaluated by the enzymatic
determination using glutamate dehydrogenase in a rapid and inter-
ference-free photometric determination (340 nm) of NH +4 in native
blood plasma according to the De Fonseca-Wollheim method [12].
The blood sample was immediately taken after withdrawal to the lab-
oratory for immediate (within 15 min) determination of NH4 +4.
Hepatic Encephalopathy
Encephalopathy grade was diagnosed on the basis of the evalua-
tion of consciousness, intellectual functions, behavior and neuromus-
cular functions according to the West Haven criteria introduced by
Conn and Liebenthal [13, 14].
Malaguarnera/Pistone/Astuto/Dell’Arte/
Finocchiaro/Lo Giudice/Pennisi
Table 1. Baseline characteristics of patients
Parameter Carnitine group (A) Placebo group (B)

Male/female 40/20 38/22

Age 51.7B11.8 52.4B10.4
Cirrhosis etiology
Alcohol 4 5
Post-hepatitis B 16 17
Post-hepatitis C 32 31
Cryptogenetic 8 7
Child-Pugh class
A 29 30
B 32 31
C 9 9
Prothrombin time, % 60.4B6.7 61.2B6.9
Serum albumin level, g/dl 2.7B0.6 2.8B0.5
Serum bilirubin level, mg/dl 3.8B1.4 3.6B1.5
Serum alanine aminotransferase level, IU/l 128B61 120B63
Blood urea nitrogen, mg/dl 44B10 43B12
Serum creatinine level, mg/dl 1B0.15 0.98B0.14
Natriemia, mEq/l 133B3.4 132B4.7

Safety Parameters differences in serum NH +4 levels. In group HE 1, after 30

Safety parameters included blood tests (hemoglobin, hematocrit,
white and red blood cell count, platelet count) and liver function tests
(alanine aminotransferase, aspartate aminotransferase, Á-glutamyl-
transpeptidase, cholinesterase activity, serum bilirubin, serum albu-
min concentrations, prothrombin time and partial thromboplastin
time) on days 0, 30 and 60.
Statistical Analysis
Descriptive statistics were prepared from the study sample and
results were expressed as means B SD. Statistical significance in con-
tingency tables was evaluated using ¯2 test and Fisher’s exact test.
Student’s test for unpaired data, one-way ANOVA and Mann-Whit-
ney rank sum test were used for comparisons of continuous variables.
Statistical analyses were performed using appropriate tests for re-
peated measures as well as by controlling for multiple comparisons
with correction of the Duncan procedure. The study protocol was
received and approved by the Institutional Review Board of the Hos-
pital following the guidelines of the 1975 Declaration of Helsinki.
days of treatment, p was 0.003 (CI 6.37–30.43), being
0.000 (CI 27.82–48.98) after 60 days compared to base-
line values and 0.000 (CI 11.14–28.86) compared to day
30. In group HE 2, p was 0.000 (CI 15.76–40.04) after 30
days, being 0.000 (CI 32.96–53.24) after 60 days com-
pared to baseline values and 0.001 (CI 6.01–24.39) com-
pared to day 30 of treatment respectively. With reference
to NCT-A, in group HE 1 we observed a p of 0.000 (CI
6.23–19.17) after 30 days and 0.000 (CI 17.78–30.62)
after 60 days compared to baseline values as well as 0.000
(CI 5.15–17.85) with respect to day 30 of treatment
respectively. In group HE 2 we found a p of 0.000 (CI
7.91–21.89) after 30 days and 0.000 (CI 20.20–34.20)
after 60 days compared to baseline values and 0.000 (CI
5.63–18.97) with respect to day 30 of treatment.

Results
Baseline Values
The two groups were homogeneous for demographic
characteristic, etiology, casting of disease and Child-Pugh
grade. Serum NH +4 fasting concentrations were not signifi-
cantly different before the treatment. NCT-A did not
show significant differences at baseline.
L-Carnitine Treatment
The patients treated with carnitine either in group HE
1 or in group HE 2 showed fasting statistical significant
Placebo Treatment
No statistical difference was found in these patients at
days 30 and 60 with respect to baseline and day 30 values
respectively in both groups. In fact, for fasting serum NH +4
levels, in group HE 1 we found a p of 0.477 after 30 days
(CI –7.63 to 16.23), after 60 days a p of 0.419 (CI –7.21 to
17.21) with respect to baseline values and 0.905 com-
pared to values of day 30 of treatment (CI –10.87 to
12.27).
In group HE 2 we found a p of 0.396 (CI –7.01 to
17.61) after 30 days and of 0.303 (CI –5.95 to 18.95) after
60 days with respect to baseline and 0.832 (CI –10.01 to
12.41) with respect to values of day 30 respectively. For

L-Carnitine in the Treatment of Mild or Dig Dis 2003;21:271–275 273

Moderate HE
Table 2. Comparison between evaluated parameters of the two groups (days of treatment)
Carnitine group
0 days 30 days
NH +4 fasting HE 1 80.2B36.9 61.8B29.2
HE 2 88.7B35.6 60.8B31.4
NCT-A HE 1 62.8B18.1 50.1B17.7
HE 2 66.4B20.2 51.5B18.4
NCT-A, in group HE 1 we found a p of 0.582 after 30 days
(CI –5.18 to 9.18) and of 0.287 (CI –3.41 to 11.41) after
60 days compared to baseline values and 0.586 (CI –5.25
to 9.25) with respect to values of day 30. Following the
same sequence of the previously listed parameters, in
group HE 2 we found a p of 0.852 (CI –6.74 to 8.14) after
30 days, 0.524 (CI –5.45 to 10.65) after 60 days with
respect to basal conditions and 0.629 (CI –5.86 to 9.66)
compared to day 30 of administration.
Comparison between Treatments
In group HE 1 for NH +4 at day 30 of administration, the
patients treated with carnitine showed a response statisti-
cally significant with respect to placebo (p = 0.009 CI
–25.252 to –3.68); in group HE 2, at day 30 of administra-
tion p was 0.000 (CI –31.81 to –9.39). For NCT-A, in
group HE 1 at day 30 of administration a statistical differ-
ence was found with p 0.012 (CI –15.31 to –1.89), while
in group HE 2 at day 30 of administration p was 0.000 (CI
–20.49 to –6.71) in favor of carnitine patients. At day 60
for NH +4 in group HE 1, carnitine-treated patients showed
a better significant response than placebo patients, p was
0.000 (CI –43.55 to –24.25); in group HE 2, p was 0.000
(CI –43.73 to –25.41) in favor of carnitine. About NCT-
A, at day 60 the response was strikingly in favor of carni-
tine-treated patients: in group HE 1, p was 0.000 (CI
–25.01 to –11.19), while in group HE 2, p was 0.000 (CI
–31.57 to –16.43).
Adverse Events
Both LC and placebo were well tolerated in 100% of
patients. In the group treated with LC, 1 patient com-
plained of nausea, 2 of slight headache and 2 of abdomi-
nal pain. In placebo group, 2 patients complained of diar-
rhea and 1 of moderate headache. Nobody withdrew from
the planned treatment.
274 Dig Dis 2003;21:271–275
Placebo group
60 days 0 days 30 days 60 days
41.8B18.7 80.7B34.7 76.4B31.2 75.7B32.8
45.6B17.5 86.7B37.2 81.4B30.6 80.2B31.4
38.6B17.4 60.7B20.3 58.7B19.4 56.7B20.7
39.2B18.5 65.8B21.4 65.1B19.7 63.2B23.1
Discussion
Several pathogenetic mechanisms have been suggested
to explain the onset and progression of HE. The most
ancient hypothesis is linked to the neurotoxicity of ammo-
nia. The majority of blood NH +4 amount results from mus-
cle protein catabolism at intestinal level; the remnant is
produced by the action of colic bacteria on the natrium
present in digested foods. Ammonia is vehicled to the liv-
er throughout the portal flux and is normally eliminated
as urea. The liver damage or the presence of portosys-
temic shunts increase its serum levels [15, 16]. The
exceeding ammonia is eliminated from the blood by
transforming of glutamate into glutamine in skeletal mus-
cle as well as central nervous system. Neurotoxicity of
ammonia is probably due to a direct action on neurons,
because the reduction of glutamate and the increase of
glutamine may induce a swelling of the astrocytes [16].
Previous studies have reported a significant protective
effect of LC in mice and rats, which is associated with a
significant reduction of blood and brain ammonia con-
centration [9, 17], suggesting an action of LC either at
peripheral or central sites. In the study by O’Connor et al.
[9], blood urea concentrations were significantly in-
creased and inversely related to the lowering of blood
ammonia after one carnitine administration to normal
mice to whom toxic doses of ammonia were adminis-
tered. This fact suggests that LC’s effect in these animals
was caused, at least partially, by an action on hepatic urea
synthesis. Results of our study show a protective effect of
LC in ammonia-precipitated encephalopathy in cirrhotic
patients. Either in subjects with HE 1 or HE 2, we
observed a significant reduction at day 30 and more
markedly at day 60 of treatment. A significant therapeutic
effect of carnitine was also observed in the NCT-A, which
is an accepted and reliable psychometric test for the
assessment of mental function in cirrhotic patients with
HE [13, 18]. LC crosses the hematoencephalic barrier
Malaguarnera/Pistone/Astuto/Dell’Arte/
Finocchiaro/Lo Giudice/Pennisi
slowly (brain uptake index 5.5%) but in spite of this fact,
its amount in the brain is relatively large [19]. In the study
of Therrien et al. [20], the protective effect of LC was
accompanied by a significant attenuation of the increased
cerebrospinal fluid and brain alanine as well as cerebro-
spinal fluid lactate content, caused by ammonium acetate
administration. This fact suggests that mitochondria res-
piration is at least partially restored in LC-treated ani-
mals. The possible beneficial effect of carnitine may be
related to an improved pyruvate oxidation, Krebs cycle
and flux through glutamate dehydrogenase. The latter
could then explain the lowering of blood ammonia levels
that follows LC administration. In conclusion, this study
demonstrates a clinically significant effect of LC on men-
tal conditions and ammonia levels in patients with mild
or moderate HE.

References

1 Basile AS, Jones EA: Ammonia and GABAer-
gic neurotransmission: Interrelated factors in
the pathogenesis of hepatic encephalopathy.
Hepatology 1997;25:103–105.
2 Lockwood AH, Yap EWH, Wong WH: Cere-
bral ammonia metabolism in patients with se-
vere liver disease and minimal hepatic enceph-
alopathy. J Cereb Blood Flow Metab 1991;11:
337–341.
3 Tarter RE, Panzak G, Switala J, Lu S, Simke-
vitz H, Van Thiel D: Isokinetic muscle strength
and its association with neuropsychological ca-
pacity in cirrhotic alcoholics. Alcohol Clin Exp
Res 1997;21:191–196.
4 Haussinger D, Kircheis G, Fischer R, Schiess
F, Von Dahl S: Hepatic encephalopathy in
chronic liver disease: A clinical manifestation
of astrocyte swelling and low-grade cerebral
edema? J Hepatol 2000;32:1035–1038.
5 Ullian EM, Sepperstein SK, Christopherson
KS, Barres BA: Control of synapse number by
glia. Science 2001;291:657–661.
6 Noremberg MD: Astrocyte-ammonia interac-
tions in hepatic encephalopathy. Semin Liver
Dis 1996;16:245–253.
7 Conn HO, Leevy CM, Vlahovic ZR, et al:
Comparison of lactulose and neomycin in the
treatment of chronic portal systemic encepha-
lopathy. Gastroenterology 1997;72:573–583.
8 Bremer J: The role of carnitine in cell metabo-
lism; in De Simone C, Famularo G (eds): Car-
nitine Today. Heidelberg, Springer, 1997, pp
1–37.
9 O’Connor JE, Costell M, Grisolia S: Pre-
vention of ammonia toxicity by L-carnitine:
Metabolic changes in brain. Neurochem Res
1984;9:563–570.
10 Therrien G, Rose C, Butterworth J, Butter-
worth RF: Protective effect of L-carnitine in
ammonia-precipitated encephalopathy in the
portocaval shunted rat. Hepatology 1997;25:
551–556.
11 Reitan RM, Woegson D: The Halstead-Reitan.
Neuropsycological test battery: theory and clin-
ical interpretation. Tucson, Neuropsychology
Press, 1993.
12 De Fonseca-Wollheim F: Direkte Plasmaam-
moniakbestimmung ohne Enteiweissung. J
Clin Chem Clin Biochem 1973;11:421–431.
13 Conn HO, Liebertahl MM: The Hepatic Coma
Syndrome and Lactulose. Baltimore, Williams
& Wilkins, 1979, pp 1–121.
14 Conn HO: Trail-making and number connec-
tion tests in the assessment of mental state in
portal systemic encephalopathy. Am J Dig Dis
1977;22:541–550.
15 Rikkers L, Jenko P, Rudman D, Freides D:
Subclinical hepatic encephalopathy: Detection
prevalence and relationship to nitrogen metab-
olism. Gastroenterology 1978;75:462–469.
16 Gilberstadt SJ, Gilberstadt H, Zieve L, Buegel
B, Collier RO, McClain CJ: Psychomotor per-
formance defects in cirrhotic patients without
overt encephalopathy. Arch Intern Med 1980;
140:519–521.
17 Matsouka M, Igsu H, Kohriyama K, Inoue N:
Suppression of neurotoxicity of ammonia by L-
carnitine. Brain Res 1991;567:328–331.
18 Atterbury CE, Maddrey WC, Conn HO: Neo-
mycin-sorbitol and lactulose in the treatment
of acute portal-systemic encephalopathy. Dig
Dis 1978;23:398–408.
19 Hearn TJ, Coleman AE, Lai JCK, Griffith OW,
Cooper AJL: Effect of orally administered L-
carnitine on blood ammonia and L-carnitine
concentrations in portocaval-shunted rats. He-
patology 1989;10:822–828.
20 Therrien G, Giguere JF, Butterworth RF: In-
creased cerebrospinal fluid lactate reflects dete-
rioration of neurological status in experimental
portal-systemic encephalopathy. Metab Brain
Dis 1991;6:231–238.

L-Carnitine in the Treatment of Mild or Dig Dis 2003;21:271–275 275

Moderate HE
 Original Paper
Dig Dis 2003;21:276–278
DOI: 10.1159/000073348
Fructose Breath Hydrogen Test –
Is It Really a Harmless Diagnostic
Procedure?
P. Müller a, b C. Meier b H.J. Böhme c T. Richter b, d
a HELIOS Hospital, Leisnig; b Children’s Hospital, University of Leipzig; c Institute of Biochemistry,
University of Leipzig, and d St. Georg Hospital, Leipzig, Germany
Key Words
Fructose malabsorption W Hereditary fructose
intolerance W Fructose breath hydrogen test
Abstract
Usage of hydrogen breath tests has become one of the
standard procedures in diagnosing chronic unspecific
abdominal pain. These tests are said to be of sufficient
specificity and sensitivity, are easily done, non-invasive
and are more often practiced in outpatients. A 13-year-
old boy is reported with chronic unspecific abdominal
pain and growth retardation and so far misdiagnosed
hereditary fructose intolerance (HFI), who developed life-
threatening adverse effects during the fructose breath
hydrogen test. It is concluded that the possibility of HFI
should be excluded first by a carefully explored dietary
history before the fructose breath test is performed
under medical supervision. If there is any suspicion of
HFI, a molecular genetic analysis should be preferred.
Copyright © 2003 S. Karger AG, Basel
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0276$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
Introduction
Patients with chronic abdominal pain, enteritis and
meteorism are frequent clients in clinical and ambulatory
practice. One possible cause of these complaints are indi-
vidual distinctions of the intestinal capacity of absorption
for fructose, which is estimated in adults between 0.07
and 0.7 g/kg body weight (b.w.) [1]. If this individual
intestinal maximum of absorption for fructose is ex-
ceeded, fructose will be metabolized by bacterial fermen-
tation and typical signs and symptoms of an isolated fruc-
tose malabsorption develop [2]. The prevalence of this
clinical entity increases with increasing use of high fruc-
tose corn syrup as sweetener in processed food. The cur-
rent view in the literature is that fructose is transported
across the brush-border membrane by GLUT-5, a mem-
ber of the facilitative fructose transporter family [3].
However, screening for mutations in the coding region of
the GLUT-5 gene was without success and the conclusion
of the study was that isolated fructose malabsorption does
not result from the expression of mutant GLUT-5 protein
[4]. Moreover, it has recently been shown that in addition
to GLUT-5, GLUT-2 also contributes to the fructose
transport across the brush-border membrane – at least in
rats [5]. Thus, isolated fructose malabsorption still awaits
Peter Müller, MD
Department of Pediatrics, HELIOS Hospital Leisnig
Colditzer Strasse 48
DE–04703 Leisnig (Germany)
Tel. +49 34321 8310, Fax +49 34321 8111, E-Mail pmueller@leisnig.helios-kliniken.de
an explanation. Hydrogen breath tests are known as suffi-
cient methods to detect intestinal carbohydrate malab-
sorptions. An increase of exhaled hydrogen of 120 ppm
after oral application of fructose (1 g/kg b.w., maximum
25 g) points to fructose malabsorption [6, 7]. Although
this test seems to be simple and safe, evaluations in
infants and children showed problems when measure-
ment results had to be interpreted and when the predic-
tive value and consecutive indications of this breath test
were proven [8–10].
We report on an adverse effect during a hydrogen
breath test with fructose in a schoolboy who suffered from
unknown inherited fructose intolerance. Such a case has
not been described so far and therefore, it is not advisable
to perform a hydrogen-exhaled breath test after oral fruc-
tose loading without relevant indications.
Case Report
A growth-retarded 13-year-old boy who had suffered from ab-
dominal pain for several years visited for gastroenterological consul-
tation. At the time, a partial growth hormone deficiency was treated
with growth hormone. The family history was unremarkable regard-
ing gastrointestinal diseases. Parents were of normal heights, but a
sister who was 2 years older was also growth-retarded. The patient’s
dietary history was suspicious with aversion to fruits, small amounts
of sugar, but a considerable amount of milk products. In contrast, his
sister tolerated fructose-containing foods very well. The boy’s appe-
tite was good, stool was normal and reduction of weight or efficiency
was not observed. Clinical investigation, especially of the abdomen,
showed normal findings. The liver was palpable 1 cm below the ribs.
A fructose breath hydrogen test was performed to confirm the sus-
pected fructose malabsorption. 60 min after oral fructose load with
1.5 g/kg b.w., the patient became indisposed with tachycardia (110/
min), hypotonia (105/70 mm Hg), cold-sweaty skin, disturbed con-
sciousness and vomiting. Measurement of blood glucose revealed a
hypoglycemia with 2.2 mmol/l and a metabolic acidosis (pH 7.32,
base excess –7.5 mmol/l). Signs and symptoms immediately normal-
ized after intravenous injection of 10 g glucose. Further investiga-
tions revealed elevated liver enzyme activities of GPT (1.10 Ìmol/
s-l) and GOT (1.68 Ìmol/s-l) and an increased lactate concentration
(3.3 mmol/l), while anorganic phosphate was slightly reduced to
1.41 mmol/l. Uric acid in serum was found to be in the normal range.
These findings pointed to hereditary fructose intolerance. Mutation-
al analysis revealed compound heterozygosity of the patient for the
two common mutations A149P in exon 5 and the mutation N334K
in exon 9 of the aldolase B gene and therefore, no liver biopsy for the
determination of the enzyme activity of aldolase B was needed. The
patient was treated with a diet low in fructose and sucrose (maximum
15 g fructose/day) with the results of physical wellness and catch-up
growth without hormone therapy.
Fructose Breath Hydrogen Test – Really a
Harmless Diagnostic Procedure?
Discussion
At present, the content of fructose as the naturally
sweetest occurring sugar in foodstuffs is enormous in
developed industrial countries. The estimated daily con-
sumption of sugar was 4.5 g per capita in England in 1700,
in 1800 already 37 g was consumed and in 1968 this had
risen to 140 g per capita daily [11]. These remarkable
changes in nutritional habits are one of the common
causes of chronic unspecific abdominal pain in children
(toddler’s diarrhea) due to fructose malabsorption. Beside
other influences of environment and stress, this dysfunc-
tion is also a frequent cause of chronic enteritis, meteo-
rism and irritable bowel syndrome in adults, especially in
cases where sorbitol-containing food for diabetics is con-
sumed [12]. If the fructose malabsorption is recognized as
the cause of complaints, the dietary management with
supplementation of equimolar amounts of glucose will
immediately lead to recovery [13]. Affected children have
no growth retardation. In contrast, poor growth and fail-
ure to thrive are typical signs of patients with autosomal
recessively inherited fructose intolerance (HFI) after
chronic fructose intoxication, which is pathogenetically
distinguished from intestinal absorption disturbances
[14]. HFI was first described as a clinical entity in 1956
[15]. The molecular basis of this disease is a deficiency of
the enzyme aldolase B (EC 4.1.2.13) in liver, intestine and
kidney [16]. After exposition to fructose, dose-dependent
cytosolic accumulations of fructose-1-phosphate occur
which inhibit liver phosphorylase A. Glycogenolysis and
glycolysis are interrupted and the resulting hypoglycemia
is resistant to glucagon. Hypophosphatemia, which inten-
sifies the inhibition of liver phosphorylase A, is another
consequence of sequestration of phosphate as fructose-1-
phosphate. Intracellular deficiency of ATP and phosphate
lead to hyperuricemia and hypermagnesemia, respective-
ly [14]. At present, 21 mutations within the human aldo-
lase B gene have been published, of these, 15 are single-
base substitutions. In Germany, 93% of HFI alleles are
one of three common mutations: A149P, A174D and
N334K [17, 18]. Hypoglycemia, metabolic acidosis, hypo-
phosphatemia, hyperuricemia and hypermagnesemia are
the laboratory hallmarks. Several of these findings
pointed to the diagnosis in our case. To the best of our
knowledge, this is the first report with casual diagnosis of
HFI by means of the fructose breath hydrogen test. This
possibility has to be taken into account when a hydrogen
breath test is intended to be performed because an oral
fructose load with 1 up to 2 g/kg b.w. (we give a maximum
of 25 g) is considerably high. This amount of fructose is
Dig Dis 2003;21:276–278 277
even higher than the dose used for an oral fructose loading
test, which is meanwhile considered as an obsolete proce-
dure to diagnose HFI [19]. Patients with HFI who are not
recognized within the first year of life very often develop
aversions against all sweet foods. This habit becomes
gradually accepted by the parents, and the diet will be
adapted by preparations that avoid symptoms like vomit-
ing and/or abdominal pain. For these reasons, it was
reported several times that patients with unrecognized
HFI were diagnosed during childhood or even during
adulthood [12, 20]. Moreover, tragedies of iatrogenic
deaths related to medicinal use of fructose-based intrave-
nous solutions are known [21]. As a consequence, before
an oral fructose load test is performed, it is necessary to
get exact information about the subject’s nutritional hab-
its. HFI is probable when an avoidance to fruits or sweets
is reported. The mutational analysis is then a feasible
first-line investigation because in 93% of the German HFI
patients there is a prevalence of three mutations, which
can easily be detected [18]. Considering the propagan-
dized practicability of the fructose hydrogen breath test,
especially in outpatients, we think that the fructose load
under domestic conditions is risky and therefore, we rec-
ommend to perform this procedure under medical super-
vision and blood glucose monitoring (0, 30 and 60 min),
for instance at a pediatric day-ward. In order to prevent
such incidents as in our case, we recommend the follow-
ing diagnostic work-up: If there are any doubts regarding
fructose tolerance after exploring the dietary history, a
PCR-based mutational analysis of the aldolase B gene
should be performed because the prevalence of HFI car-
riers in Caucasians is about 1:50 [17]. If no relevant muta-
tion can be detected, the fructose breath hydrogen test is
indicated to exclude fructose malabsorption. But it should
be pointed out that the absence of hydrogen-producing
bacteria in intestine can lead to false-negative results.
Therefore, it is advisable to perform a following breath
test with lactulose which is not absorbed by the small bow-
el. In dependence of hydrogen production, we investigate
an intestine biopsy and measure the enzyme activity of
sucrase. In cases with normal enzyme activities, other
causes of chronic abdominal pain than fructose-induced
disturbances have to be examined.

References

1 Ramussen JJ, Gudmand-Hoyer E: Absorption
capacity of fructose in healthy adults. Compari-
son with sucrose and its constituent monosac-
charides. Gut 1986;27:1161–1168.
2 Ledochowski M, Widner B, Bair H, Probst T,
Fuchs D: Fructose- and sorbitol-reduced diet
improves mood and gastrointestinal distur-
bances in fructose malabsorbers. Scand J Gas-
troenterol 2002;35:1113–1126.
3 Burant CF, Takeda J, Brot-Laroche E, Bell GI,
Davidson NO: Fructose transporter in human
spermatozoa and small intestine is GLUT-5. J
Biol Chem 1992;267:14523–14526.
4 Wassermann D, Hoekstra JH, Tolia V, Taylor
CJ, Kirschner S, Takeda J, Bell GI, Taub R,
Rand EB: Molecular analysis of the fructose
transporter gene (GLUT-5) in isolated fructose
malabsorption. J Clin Invest 1996;98:2398–
2402.
5 Helliwell, PA, Richardson M, Affleck J, Kel-
lett, GL: Regulation of GLUT-5, GLUT-2 and
intestinal brush-border fructose absorption by
the extracellular signal-regulated kinase, p38
mitogen-activated kinase and phosphatidylino-
sitol 3-kinase intracellular signalling pathways:
Implications for adaptation to diabetes. Bio-
chem J 2000;50:163–169.
6 Barnes G, McKellan W, Lawrence S: Detection
of fructose malabsorption by breath hydrogen
test in a child with diarrhea. J Pediatr 1983;
103:575–577.
7 Bond, JH, Levitt MD: Investigation of small
bowel transit time in man utilizing pulmonary
hydrogen (H2) measurements. J Lab Clin Med
1975;85:546–555.
8 Romagnuolu J, Schiller D, Bailey RJ: Using
breath tests wisely in a gastroenterology prac-
tice: An evidence-based review of indications
and pitfalls in interpretation. Am J Gastroen-
terol 2002;97:1113–1126.
9 Hoekstra JH: Fructose breath hydrogen test in
infants with chronic non-specific diarrhoea.
Eur J Pediatr 1995;154:362–364.
10 Götze H, Mahdi A: Role of fructose malabsorp-
tion in dysfunctional bowel disorders. Mo-
natsschr Kinderheilkd 1992;140:814–817.
11 Cox TM: Fructose intolerance: Diet and inheri-
tance. Proc Nutr Soc 1991;50:305–309.
12 Wössmann W, Wiemann J, Körber F, Görtner
L: Hereditäre Fruktoseintoleranz (HFI) als Ur-
sache einer isolierten Á-GT-Erhöhung bei ei-
nem 5-jährigen Jungen mit Hepatomegalie.
Klin Pädiatr 2000;212:108–109.
13 Kneepkens CMF, Vonk RJ, Fernandes J: In-
complete intestinal absorption of fructose.
Arch Dis Child 1984;59:735–738.
14 Steinmann B, Gitzelmann R, van den Berghe
G: Disorders of fructose metabolism; in Scriver
CR, Beaudet AL, Sly WS, Valle D (eds): The
Metabolic and Molecular Bases of Inherited
Disease, ed 8. New York, McGraw-Hill, 2001,
pp 1489–1520.
15 Chambers RA, Pratt RTC: Idiosyncrasy to
fructose. Lancet 1956;271:340.
16 Hers HG, Joassin G: An anomaly in hepatic
aldolase during fructose intolerance. Enzymol
Biol Clin 1961;1:4–14.
17 Tolan DR: Molecular basis of hereditary fruc-
tose intolerance: Mutations and polymor-
phisms in the human aldolase B gene. Hum
Mutat 1995;6:210–218.
18 Müller P, Koppelt B, Böhme HJ, Bührdel P:
Detection of DNA mutations in German pa-
tients with hereditary fructose intolerance.
Amino Acids 1997;12:389.
19 Koppelt B, Pöge AP, Müller P, Bührdel P,
Böhme HJ: Molecular biological investigation
in a family with hereditary fructose intolerance
and diagnostic consequences. Monatsschr Kin-
derheilkd 1996;144:383–386.
20 Burmeister LA, Valdivia T, Nuttall FQ: Adult
hereditary fructose intolerance. Arch Intern
Med 1991;151:773–776.
21 Cox TM: Iatrogenic deaths in hereditary fruc-
tose intolerance. Arch Dis Child 1993;69:413–
415.

278 Dig Dis 2003;21:276–278 Müller/Meier/Böhme/Richter

 Dig Dis 2003;21:279–285
DOI: 10.1159/000073985
Screening for Iron Overload in the
Turkish Population
Gultekin Barut a Huriye Balci b Mithat Bozdayi d Ibrahim Hatemi a
Dervis Ozcelik c Hakan Senturk a
a Department of Internal Medicine, b Central Research Laboratory, c Department of Biophysics,
Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, and d Institute of Hepatology, Ankara Medical Faculty,
University of Ankara, Ankara, Turkey
Key Words
Hereditary hemochromatosis W HFE gene mutation W
Epidemiology W Iron overload, prevalence
Abstract
Background/Aims: Hereditary hemochromatosis (HH),
the most common autosomal recessive disease in the
white population, is characterized by excessive gastroin-
testinal absorption of iron and loading of parenchymal
organs. HFE mutations of C282Y and H63D are largely
responsible for HH in populations of Celtic ancestry.
Although many screening studies related to HH have
been done in Northern Europe, the USA and Australia, as
yet, no such study has been published on Turkey. In this
study we aimed to screen the Turkish population for iron
overload. Methods: Random samples were obtained
from 4,633 healthy adults (3,827 male, 806 female, mean
age B SD 35 B 8 years, range 14–76) for the measure-
ment of transferrin saturation (TS). Measurements were
repeated after an overnight fast in the subjects whose
initial TS was 650%. Serum ferritin levels and C282Y
and H63D gene mutations were studied in cases when
fasting TS was 650%. In cases where the serum ferritin
level was 1 200 ng/ml with or without HFE mutations, liv-
er biopsy was performed for histological evaluation and
determination of iron content. Results: In 158 subjects,
TS was 650% in the non-fasting state. A second deter-
© 2003 S. Karger AG, Basel
ABC 0257–2753/03/0213–0279$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi
mination of TS after an overnight fast was performed in
135 subjects. In 26 subjects, the TS was 650% in the fast-
ing state. HFE mutation and serum ferritin levels were
measured in these 26 subjects. Eleven subject (10 male, 1
female) were heterozygote and 1 male subject was ho-
mozygote in reference to H63D. C282Y mutation was not
found. Four of these 26 subjects (all males, aged 23, 24,
40, 49) had increased serum ferritin levels and liver biop-
sy was performed. In 1 male (aged 49) who was hetero-
zygote for H63D genotype with a serum ferritin level of
645 ng/ml, iron overload in liver tissue was shown by
histology as well as atomic absorption spectrophotome-
try. Conclusion: The prevalence of hemochromatosis in
the Turkish population is much lower in comparison to
populations of Celtic ancestry and C282Y mutation is
non-existent.
Copyright © 2003 S. Karger AG, Basel
Introduction
Hereditary hemochromatosis (HH) is the most com-
mon autosomal recessive disease in the white population
[1–3]. Typical clinical symptoms do not develop until
very late stages of the disease at which time treatment is
not rewarding. The value of screening studies largely by
means of measurement of transferrin saturation (TS) was
evaluated in several populations [1–10]. Mutation of HFE
Hakan Senturk, MD, Professor in Medicine
Cihangir, Akyol cad. No:18-5 Makbul apt
TR–80060 Istanbul (Turkey)
Tel./Fax +90 212 5299565
E-Mail drhakansenturk@yahoo.com
Fig. 1. Frequency distribution of transferrin
saturation in 4,633 participants.
gene, is largely responsible for this disease in populations
of Celtic origin [6, 11]. To date, 37 allelic variants of the
HFE gene have been reported – C282Y and H63D are the
most common. The gene frequency is 5% in the Anglo-
Saxon population. Homozygote HH prevalence is be-
tween 0.3 and 0.5%, and the heterozygote carrier rate is
approximately 10% in European populations [1–5, 12,
13]. As yet, no such screening study has been published on
Turkey. HH is known as an inherently rare disease in this
country. In this study we aimed to screen the Turkish pop-
ulation for iron overload.
Methods
We included 4,633 healthy subjects (3,827 men, 806 women;
mean age B SD 35 B 8, range 14–76) who were employees of Istan-
bul’s gas distribution, and water and sewage companies. Ten millili-
ters of venous blood was obtained from non-fasting volunteers.
Serum was separated by centrifugation and frozen at –70 ° C until
analysis. Serum iron levels and total iron binding capacity (TIBC)
were measured by a standard colorimetric method on an automated
analyzer (model Au-800, Olympus). Serum TS was calculated as
serum iron/TIBC ! 100. In cases where the TS value was 650%, the
participant was re-contacted to obtain a second sample in the fasting
state and measurement of TS was repeated.
Fifteen-milliliter whole-blood samples were obtained from 26
subjects whose fasting TS was 650%. These blood samples were
immediately transported to the laboratory within 24 h at +4 ° C for
analysis of HFE gene mutations. DNA was prepared from these
whole-blood samples by a method reported elsewhere [14]. The two
HFE mutations were screened with PCR assay followed by restric-
tion-enzyme digestion with Rsa1 for the C282Y mutation and Bcl1
for H63D mutation. PCR amplification of the regions containing
these two HFE mutations was performed with the primer sequences
280 Dig Dis 2003;21:279–285
of Feder et al. Serum ferritin levels were measured by an immunoas-
say system by chemiluminescent immunoassay method in 26 sub-
jects who had 650% fasting TS and were analyzed for HFE gene
mutations.
Percutaneous needle biopsy of the liver were performed in 4 sub-
jects who had a fasting TS of 650% with a serum ferritin level
1 200 ng/ml. The tissue specimens were sent for routine histological
examination with Prussian blue staining. Liver iron content was
measured by atomic absorption spectrophotometry and hepatic iron
index (HII) was calculated by dividing liver iron concentration in
Ìmol Fe/g dry liver weight with subject’s age in years.
Results
The mean TS was 29% for males and 25% for females
(fig. 1). In 158 subjects (142 men and 16 women; mean
age B SD 36 B 8 years, range 17–52) who constituted
3.4% of total screened population, TS was 650% in the
non-fasting state. A second determination of TS after an
overnight fast was performed in 135 of 158 subjects.
Twenty-three subjects either could not be re-contacted or
denied further examination. In 26 subjects (24 men and 2
women; mean age B SD 34 B 8 years, range 22–52), who
constituted 0.6% of total screened population and 19% of
re-evaluated population, the TS was 650% in the fasting
state. Analysis of HFE mutations was done in these 26
subjects (fig. 2). C282Y mutation was not detected. Ho-
mozygote H63D mutation was found in only 1 male sub-
ject who had a normal serum ferritin level. Heterozygote
H63D mutation was found in 10 males and 1 female.
Serum ferritin levels were also measured in 26 subjects. In
4, the serum ferritin level was 1200 ng/ml, while in 1 of
Barut/Balci/Bozdayi/Hatemi/Ozcelik/
Senturk
 Fig. 2. The algorithm of screening.
these 4 subjects, the serum ferritin level was 645 ng/ml; in
3 other subjects, it was !300 ng/ml (table 1) – all 4 were
males. In the subject with a serum ferritin level of
645 ng/ml, an increased liver iron content (133.05 Ìmol/g
and HII = 2.71) and intensive iron staining were found.
Hereditary Hemochromatosis in Turkey
This subject had heterozygote H63D genotype. Remain-
ing three subjects did not have either increased liver iron
content or staining. The HII of these 3 subjects was !1.9
(table 1). Only 1 of these 3 subjects had heterozygote
H63D genotype (HII = 0.67) (table 1).
Dig Dis 2003;21:279–285 281
Table 1. Laboratory features of participants whose phenotypes and since it is inherited autosomal recessively. However, the
genotypes were analyzed prevalence of HH and iron overload was higher in males
Hepatic iron Hepatic Serum Transferrin Mutation Mutation Age
than females in previous screening studies which had
concentration iron ferritin saturation C282Y H63D years been based on the phenotyping expression of the disease
Ìmol/g index ng/ml % [4, 6]. This can be explained by decreased clinical expres-
85.0 51.2 –/– –/– 34 sion resulting from menstrual blood loss and pregnancy in
37.4 51.4 –/– –/– 28 women. Many epidemiologic studies about HH and iron
37.27 1.55 238.5 90.6 –/– –/– 24 overload have been reported from different countries and
57.2 66.8 –/– –/+ 39 populations in the last 15 years. The prevalence of iron
39.9 52.0 –/– –/+ 27
overload resulting from HH was found to be 0.36% in
100.2 65.3 –/– –/– 40
36.3 59.8 –/– –/– 34 Australia where the population is predominantly com-
33.6 55.8 –/– –/+ 37 posed of whites of European descent [2]. The prevalence
67.2 64.9 –/– –/+ 29 of iron overload attributable to HH was reported as 0.40,
47.8 51.2 –/– –/+ 29 0.42 and 0.45% from different ethnic populations of the
25.20 1.09 200.8 52.3 –/– –/– 23
USA [6–8]. The prevalence was lower in blacks [3, 6, 15].
116.6 51.0 –/– –/– 27
53.1 65.2 –/– –/+ 31 Different approaches have been proposed for screening
42.9 51.4 –/– +/+ 25 HH and iron overload. The choice of screening method de-
27.8 52.4 –/– –/– 29 pends on whether one is seeking to detect all homozygotes
17.1 52.9 –/– –/– 42 or only those with iron overload [13]. As screening tests, TS,
69.6 60.1 –/– –/– 28
unsaturated iron binding capacity, serum ferritin level and
190.0 55.1 –/– –/+ 52
161.3 71.9 –/– –/– 42 HFE gene analysis were used in combination or alone.
86.6 67.9 –/– –/+ 40 Because of its ease of application, low cost, high sensi-
26.81 0.67 285.4 58.8 –/– –/+ 40 tivity and acceptable specificity, TS was the initial screen-
70.9 63.8 –/– –/+ 36 ing test in most of the previous studies [1–10]. The prob-
146.3 55.5 –/– –/– 29
lems with TS are related to its ill-defined threshold value
133.05 2.71 645.2 52.5 –/– –/+ 49
68.7 56.5 –/– –/+ 34 as well as low positive predictivity. The threshold value in
103.0 64.8 –/– –/– 30 previous studies has ranged from 45 to 70% [1, 5–8, 13,
15–17]. The rate of false positivity was intended to be
reduced with repeated measurements of TS after an over-
night fast in the subjects who had higher initial non-fast-
ing TS. Lower threshold TS values have been used for
Discussion females compared to males in previous studies. It has
been reported that a TS of 60% or more, identifies nearly
Turkey bridges the Middle East, Middle Asia and all homozygotes with iron loading, whereas a TS of 50%
Europe and has a population of over 60 million. The cur- identifies nearly all homozygotes regardless of sex or iron
rent Turkish population is a mixture of people who loading [13]. We set the threshold value to 50% to detect
migrated from Middle Asia (now Mongolia) around 1071 all homozygotes regardless of sex.
AD and the indigenous population of Anatolia. Over 95% A combination of determination of iron status with
of the population is Muslim and the rate of alcohol con- phenotyping methods associated with DNA testing have
sumption is much lower in comparison to European coun- the best sensitivity and specificity to detect risk from iron
tries and the USA. overload [7]. We used a strategy which was reported as
This study was performed in Istanbul and consisted of worthy of consideration in a previous study and started
4,633 workers from sewage and gas distribution compa- with determination of TS followed by an HFE gene analy-
nies. Istanbul is Turkey’s melting pot and most of the sis in subjects with elevated fasting TS [7].
workers have come from different parts of Anatolia in the In the four screening programs which were performed
last decade. Therefore, this population represents not only in the USA, the TS was elevated on initial testing in 2.5–
Istanbul but Turkey in general. 5.8% of the several screened populations [8]. The ratio
HH is the most common autosomal recessively inher- was 6.2% in the Edwards’ study [4]. The prevalence of an
ited disease in the white population [1–3]. The prevalence elevated initial TS rate of 3.4% in our screened popula-
of HH is supposed to be equal in both males and females, tion was similar to these studies.

282 Dig Dis 2003;21:279–285 Barut/Balci/Bozdayi/Hatemi/Ozcelik/

Senturk
 A second determination of TS after an overnight fast
was performed in 135 of 158 subjects with elevated initial
TS. All drop-outs appeared at this stage of screening. The
compliance rate for a second determination of TS after an
overnight fast was calculated as 85.4% (135/158). The
same rate in Phatak’s study, Edwards’ study and in the
other two United States’ studies were 80.2% (747/932),
67.6% (465/688), 85.0% (1,147/1,349) and 87% (47/53)
respectively [4, 6, 8]. In 26 subjects (24 men and 2 wom-
en), who constitute 19.2% (26/135) of our re-evaluated
population, the TS was 650% in the fasting state in our
study. The rates of elevated fasting TS in Edwards’ study,
in Kaiser Permanente Medical Group’s study and in Pha-
tak’s study were 24.5% (114/465), 23% (73/318) and
10.9% (82/747) (the threshold value for TS was 55% in
this study), respectively [4, 6, 8]. These rates are also simi-
lar to rates of the present study.
In our study, 15.4% (4/26) of the subjects with elevated
fasting TS had 6200 ng/ml serum ferritin levels. The val-
ue was 50.2% (128/255) in Phatak’s study [6]. In our pop-
ulation, most of the subjects with elevated fasting TS did
not have increased serum ferritin levels which is an
important marker for liver iron content [6].
Liver biopsy was performed in 4 subjects who had
increased serum ferritin. Iron overload was detected in
only 1 subject in our study. In conclusion, the prevalence
of iron overload was 0.021% in our population. The prev-
alence of iron overload was reported as 0.36% from Aus-
tralia and 0.4, 0.42 and 0.45% from different ethnic popu-
lations of the USA [2, 6–8].
The candidate gene leading to hemochromatosis was
first discovered in 1996 by Feder et al. on the short arm of
chromosome 6, and was termed the ‘HFE gene’ [18]. This
HFE gene mutation is largely responsible for HH and
related iron overload in populations of Celtic origin [6,
11]. The interest was focused on C282Y and H63D muta-
tions. The C282Y mutation results from G to A transition
at nucleotide 845 of the HFE gene (845G→A) that pro-
duces a substitution of cysteine for a tyrosine at amino
acid position 282 in the protein product [16]. In the
H63D mutation, a G replaces C at nucleotide 187 of the
gene (187C→G), causing aspartate to substitute for histi-
dine at amino acid position 63 in the HFE protein [16]. In
the last 10 years, studies performed in populations of
largely Celtic descent showed that 60–100% of patients
with typical HH were homozygotes for C282Y mutation
[3, 7, 9, 12, 15–17, 19]. In the general population, com-
pound heterozygote (C282Y/H63D) and homozygote
H63D genotypes occur more frequently than homozygote
C282Y, but these genotypes are responsible for only a
Hereditary Hemochromatosis in Turkey
small proportion of patients with iron loading [16]. The
average prevalences of homozygote and heterozygote
C282Y mutations were reported to be 0.4 and 9.2%
respectively from different European countries [16]. Al-
though C282Y homozygosity has not been reported,
C282Y heterozygosity has been found to be 1–3% in
Southern or Eastern Europe populations [16]. The highest
prevalence of heterozygote C282Y mutation has been
reported as 24.8% from Ireland [16]. The prevalence of
homozygosity for C282Y mutation is 0.5% and the preva-
lence of heterozygosity is 9% in North America [16].
C282Y homozygosity was not found in the Asian-Indian
subcontinent, African/Middle Eastern and Australasian
populations, and the frequency of C282Y heterozygosity
was very low (0–5%) [16]. The prevalences of compound
heterozygote (C282Y/H63D) and homozygote H63D
have been found to be approximately 2% in European
populations [16]. The prevalences of compound heterozy-
gotes (C282Y/H63D), homozygote H63D and heterozy-
gote H63D were reported 2.5, 2.1 and 23% respectively in
North American populations [16]. The prevalence of het-
erozygote H63D was 15–22% in European populations
[9, 16]. Another study showed that only 64% of patients
with hemochromatosis in Italy were homozygotes for the
C282Y mutation [19]. The study performed in Hong
Kong showed that the C282Y mutation was not found in
Chinese patients with hepatic iron overload disease [20].
In a review by Lucotte [11] concerning 20 European pop-
ulations, the highest C282Y allele frequency (6.88%) was
observed in residual Celtic populations in the UK and
France. This study showed that C282Y allele frequencies
were distributed among a decreasing line from north to
south of Europe [11]. The lowest C282Y allele frequencies
were observed in Italians (1.00%), Ashkenazi Jews
(1.31%) and French Basques (1.63%) [11]. These data
suggest that HFE gene mutation is probably of Celtic ori-
gin.
In our study, analysis of HFE mutations was done in
26 subjects whose fasting TS values were 650%. C282Y
mutation was not detected. Homozygote H63D mutation
was found in only 1 subject, and heterozygote H63D
mutation was found in 11 subjects. The subject who had
homozygote H63D mutation did not have any clinical
symptom for hemochromatosis. In this subject, there was
no clinical expression of HH except fasting TS (fasting
TS: 51.3%) and the serum ferritin level was within normal
ranges (serum ferritin: 42.9 ng/ml). Our data are insuffi-
cient to comment on the prevalence of HFE mutations in
the general population of this country.
Dig Dis 2003;21:279–285 283
 In previous studies, different populations were chosen
for screening hemochromatosis – blood donors [1, 4, 17],
hospital inpatients [3, 5], primary care patients [6], bank
and insurance company employees [2], employees of
health maintenance organizations [7]. The advantages of
using blood donors for screening were expounded in the
screening study of Adams’ group [1, 17]: (1) exclusion of
persons with conditions such as hepatitis B and C and
sometimes elevated liver enzyme values reduce the num-
ber of false-positive screening tests; (2) blood has already
been obtained for other reasons; (3) systems are well
established for collection and distribution of samples, and
(4) donors represent a young asymptomatic population
that would benefit from the early detection of hemochro-
matosis. In spite of all these advantages, there are some
disadvantages of using blood donors, for example; the
false negative results for measurement of TS and serum
ferritin may occur because regular donation may cause
false low iron stores [6], subjects with elevated liver
enzyme concentrations and sideroblastic anemia who
may have HFE gene mutation are excluded from the
blood donor pool. Hospital inpatients are, by nature,
selected groups, therefore they do not represent the gener-
al population. The other screening populations are less
selective, but the actual prevalence of disease in the gener-
al population cannot be detected with these groups be-
cause they may not contain a sufficient enough number of
subjects. Increasing the number of screened subjects
makes the investigation difficult due of logistic and eco-
nomical reasons. However, we believe that our screened
population was sufficiently large enough to represent a
general population.
The serologic tests of hepatitis B and C were done, and
none of the HFE-gene-analyzed subjects with 650% fast-
ing TS had HBsAg or anti-HCV. Most of the subjects in
our study are teetotalers and none consumed 180 g/day
alcohol.
The age range in this study was 14–76 years. The mini-
mum age in Edwards’, Phatak’s and Olynyk’s studies were
17, 18 and 20 years, respectively [4, 6, 9]. The maximum
age in this study was similar to the maximum age of pre-
vious ones. The mean age was 35.7 in males and 32.2 in
females in our study. It was 37.5 in males and 34.7 in
females in Edwards’ study [4]. The major limitation of
our study was that males predominated (83%).
Life-threatening clinical manifestations, such as liver
failure, heart failure and diabetes mellitus, usually occur
after a latent period of 40–60 years in subjects with iron
overload [1]. The risk of death as a direct consequence of
liver cirrhosis is significantly increased in patients with
284 Dig Dis 2003;21:279–285
hemochromatosis as compared to a matched normal popu-
lation [12]. The risk of death from liver cancer in a patient
with hemochromatosis is more than 100 times that of the
matched healthy subjects [4]. There is a long asymptomatic
phase in HH. The fact that hemochromatosis has a high
prevalence, a potential to cause substantial morbidity and
mortality, a long asymptomatic phase and prevention of
disease-associated complications by means of early diag-
nosis and treatment such as phlebotomy, make screening
studies a cost-effective strategy [1, 13].
Recent studies have reported that hemochromatosis
screening based on TS testing is cost-effective [7]. The
cost analysis study of Adams and Valberg [17], which was
one of the largest comparing genotyping and phenotyping
models, showed that an optimal strategy for screening
includes initial testing for iron overload (phenotyping)
with confirmatory genetic testing, or initial genetic testing
if the test is less than USD 28 for each person screened. In
this study, it was reported that the cost of detecting one
homozygote with potential life-threatening illness by ini-
tial testing for iron overload (phenotyping) with confirma-
tory genetic testing was USD 2,711 [17]. For cost analysis,
10,000 blood donors were used and the prevalence of HH
was 0.3% [17]. Adams and Valberg [17] concluded that
the cost of detection of one homozygote with HH by ini-
tial testing for iron overload (phenotyping) with confirma-
tory genetic testing can be higher in a population with a
lower prevalence of the disease.
The prevalence of iron overload in our study was
0.021%. Homozygote C282Y mutation was not found
and the prevalence of homozygote H63D mutation was
0.021%. The cost of HFE gene mutation analysis for each
person was USD 75 in our study. The total cost of the
study was calculated approximately USD 50,000 by using
cost analysis. This total cost was equal to the cost of detec-
tion in the only subject with iron overload. However, it
seems that such a screening is not cost-effective in the
Turkish population; the rationality of prevention of one
case of cirrhosis with eventual decompensation necessi-
tating liver transplantation or death at a cost of USD
50,000 should be further scrutinized.
In conclusion, the prevalence of HH and iron overload
in the Turkish population is much lower in comparison to
North European populations of Celtic ancestry and
C282Y mutation as a cause of HH is non-existent.
Acknowledgement
This work was supported by the Research Fund of The University
of Istanbul, Project No. T-825/07032000.
Barut/Balci/Bozdayi/Hatemi/Ozcelik/
Senturk
 References
1 Adams PC, Gregor JC, Kertesz AE, Valberg
LS: Screening blood donors for hereditary he-
mochromatosis: Decision analysis model based
on a 30-year database. Gastroenterology 1995;
109:177–188.
2 McLaren CE, McLachlan GJ, Halliday JW,
Webb SI, Leggett BA, Jazwinska EC, Crawford
DHG, Gordeug VR, McLaren GD, Powell LW:
Distribution of transferrin saturation in an
Australian population: Relevance to the early
diagnosis of hemochromatosis. Gastroenterol-
ogy 1998;114:543–549.
3 Beutler E, Felitti V, Gelbart T, Ho N: The
effect of HFE genotypes on measurements of
iron overload in patients attending a health
appraisal clinic. Ann Intern Med 2000;133:
329–337.
4 Edwards CQ, Griffen LM, Goldgar D, Drum-
mond C, Skolnick MH, Kushner JP: Preva-
lence of hemochromatosis among 11,065 pre-
sumably healthy blood donors. N Engl J Med
1988;21:1355–1362.
5 Balan V, Baldus W, Fairbanks V, Michels V,
Burrit M, Klee G: Screening for hemochroma-
tosis: A cost-effectiveness study based on
12,258 patients. Gastroenterology 1994;107:
453–459.
6 Phatak PD, Sham RL, Raubertas RF, Dunni-
gan K, O’Leary MT, Braggins C, Cappuccio
JD: Prevalence of hereditary hemochromatosis
in 16,031 primary care patients. Ann Intern
Med 1998;129:954–961.
Hereditary Hemochromatosis in Turkey
7 McDonnell SM, Hover A, Gloe D, Ou CY,
Cougswell ME, Grummer L: Population-based
screening for hemochromatosis using pheno-
typic and DNA testing among employees of
health maintenance organizations in Spring-
field, Missouri. Am J Med 1999;107:30–37.
8 McDonnell SM, Phatak PD, Felitti V, Hover
A, McLaren GD: Screening for hemochroma-
tosis in primary care settings. Ann Intern
Med1998;129:962–970.
9 Olynyk JK, Cullen DJ, Aquilia S, Rossi E,
Summerville L, Powell LW: A population-
based study of the clinical expression of the
hemochromatosis gene. N Engl J Med 1999;
341:718–724.
10 McLaren CE, Gordeug VR, Looker AC, Has-
selblad V, Edwards CQ, Griffen LM, Kushner
JP, Brittenham GM: Prevalence of heterozy-
gotes for hemochromatosis in the white popula-
tion of the United States. Blood 1995;86:2021–
2027.
11 Lucotte G: Celtic origin of the C282Y mutation
of hemochromatosis. Blood Cells Mol Dis
1998;24:433–438.
12 Niederau C, Erhard A, Haussinger D, Stroh-
meyer G: Haemochromatosis and liver. J He-
patol 1999;30:6–11.
13 Edwards CQ, Kushner JP: Screening for hemo-
chromatosis. N Engl J Med 1993;328:1616–
1620.
Dig Dis 2003;21:279–285
14 Boretto J, Jouanolle AM, Yaouaq J, et al: Ano-
nymous markers located on chromosome 6 in
the HLA-A class I region: Allelic distribution in
genetic haemochromatosis. Hum Genet 1992;
89:33–36.
15 Cogswell ME, McDonnell SM, Khoury MJ,
Franks AL, Burke W, Brittenham G: Iron over-
load, public health and genetics: Evaluating the
evidence for hemochromatosis screening. Ann
Intern Med 1998;129:971–979.
16 Hanson HE, Imperatore G, Burke W: HFE
gene and hereditary hemochromatosis. HuGE
Rev IPDAS #549 (March 12, 2001).
17 Adams PC, Valberg LS: Screening blood do-
nors for hereditary hemochromatosis: Decision
analysis model comparing genotyping to phe-
notyping. Am J Gastroenterol 1999;94:1593–
1600.
18 Feder JN, Gnirke A, Thomas W, et al: A novel
MHC class-I like gene is mutated in patients
with hereditary hemochromatosis. Nat Genet
1996;13:399–408.
19 Pietrangelo A, Montosi G, Totaro A, Garuti C,
Conte D, Cassanelli S, Fraquelli M, Sardini C,
Vasta F, Gasparini P: Hereditary hemochro-
matosis in adults without pathogenic muta-
tions in the hemochromatosis gene. N Engl J
Med 1999;341:725–732.
20 Tsui WM, Lam PW, Lee KC, Ma KF, Chan
YK, Wong MW, Yip SP, Wong CS, Chow AS,
Lo ST: The C282Y mutation of the HFE gene
is not found in Chinese haemochromatotic pa-
tients: Multicentre retrospective study. Hong
Kong Med J 2000;6:153–158.
285
 Author Index Vol. 21, No. 3, 2003

Astuto, M. 271 Ebert, M. 237 Matsumoto, T. 266

Avgerinos, C. 214 Fernández, R. 258 Meier, C. 276
Balci, H. 279 Finocchiaro, G. 271 Müller, P. 276
Barut, G. 279 Fleta, J. 258 Olivares, J.L. 258
Baumann, G. 245 Gassull, M.A. 220 Ozcelik, D. 279
Böhme, H.J. 276 Gizaris, V. 262 Palesty, J.A. 198
Bozdayi, M. 279 Hatemi, I. 279 Paulisch, E. 252
Botsios, D.S. 228 Iida, M. 266 Pennisi, G. 271
Burmester, G.-R. 245 Kemps, M. 245 Pirlich, M. 245
Clavel, A. 258 Kuroki, F. 266 Pistone, G. 271
Dafnopoulou, A. 262 Kyriazanos, I.D. 262 Plauth, M. 245
Datsakis, K. 262 Lange, K.-P. 252 Richter, T. 276
Delis, S. 214 Lo Giudice, E. 271 Rizos, S. 214
Dell’Arte, S. 271 Lochs, H. 196, 245, 252 Rodrı́guez, G. 258
Dervenis, C. 196, 214 Lübke, H.J. 245 Schütz, T. 245, 252
Dierkes, J. 237 Luhman, N. 245 Senturk, H. 279
Dimakos, P. 262 Luley, C. 237 Sfiniadakis, I. 262
Drude, C. 252 Malaguarnera, M. 271 Vasiliadis, K.D. 228
Dudrick, S.J. 198 Malfertheiner, P. 237

Subject Index Vol. 21, No. 3, 2003

Acute pancreatitis 214 Gastrointestinal cancer 198 Malnourished hospitalized patients 245
Anorexia 198 – diseases 245 Malnutrition 198
Body mass index 262 Giardia lamblia 258 –, diagnoses 245
Cachexia 198 Growth factors 228 –, prevalence 245
Cancer 198 Gut adaptation 228 Obesity 262
– cachexia syndrome 198 – barrier 214 Pernicious anemia 237
L-Carnitine treatment 271 Helicobacter pylori 262 Selenium 266
Copper 258 – infection 237 Subjective global assessment 245
Crohn’s disease 252, 266 Hepatic encephalopathy 271 Sugar intake 252
Dental caries 252 Hereditary fructose intolerance 276 Taste changes 252
Enteral nutrition 214, 266 – hemochromatosis 279 Total parenteral nutrition 214
Enterobius vermicularis 258 HFE gene mutation 279 Vitamin B12 deficiency 237
Epidemiology 279 Homocysteine 237 Zinc 252, 258
Fasting blood glucose levels 262 Intestinal failure 228
Fructose breath hydrogen test 276 Iron overload, prevalence 279
– malabsorption 276 Magnesium 258

© 2003 S. Karger AG, Basel

ABC
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ddi