Clinical Nephrology, Vol. 64 – No.

2/2005 (103-112)

Cardiovascular risk factors in severe chronic

renal failure: the role of dietary treatment
F. Bergesio1, G. Monzani2, A. Guasparini3, R. Ciuti4, M. Gallucci5, C. Cristofano6,
E. Castrignano7, A. Cupisti8, G. Barsotti8, R. Marcucci9, R. Abbate9, S. Bandini1,
Original M. Gallo1, P.L. Tosi2 and M. Salvadori1
©2005 Dustri-Verlag Dr. K. Feistle
ISSN 0301-0430
1Department of Nephrology, Dialysis and Transplantation, Azienda Ospedale

Careggi, Florence, 2Renal Unit Ospedale di Torregalli, Azienda Sanitaria Firenze,

Florence, 3Dietary Service, 4Clinical Chemistry, Azienda Ospedale Careggi,
Florence, 5Renal Unit Ospedale di Galatina, Azienda Ospedale Galatina,
6Renal Unit Ospedale di Manduria, Azienda Ospedale Manduria,
7Renal Unit Ospedale di Campi salentina, Azienda Ospedale di Campi Salentina,
8Nephrology Unit, Department of Internal Medicine, University of Pisa,
9Thrombosis Center, University of Florence, Florence, Italy

Cardiovascular risk factors in patients with end-stage renal disease

Key words
cardiovascular risk fac-
tors – chronic renal fail-
ure – C-reactive protein
(CRP) – low-protein diet
– vegan diet – oxidative
stress – Lp(a) – homo-
cysteine – LDL choles-
terol – oxidized LDL
Received
June 1, 2004;
accepted in revised form
December 6, 2004
Correspondence to
Dr. F. Bergesio
Via del Pino, 50137
Firenze, Italy
fraberge@tin.it
Abstract. Background: Lipoprotein ab-
normalities and increased oxidized LDL
(OxLDL) are often observed in uremia and
are reported to play a central role in the devel-
opment of cardiovascular disease (CVD).
Vegan diet, known for its better lipoprotein
profile and antioxidant vitamins content,
could protect against CVD. Aim of this study
was to investigate the influence of vegan diet
supplemented with essential amino acids
(EAA) and ketoanalogues (VSD) on both tra-
ditional and non-traditional cardiovascular
risk factors (CVRF). Methods: Twenty-nine
patients (18 M, 11 F) aged 55 years (range
29 – 79 years) with advanced chronic renal
failure (median sCr: 5.6 mg/dl) on very low
protein vegetarian diet (0.3 g/kg/day) supple-
mented with a mixture of EAA and ketoacids
(VSD) and 31 patients (20 M, 11 F) aged 65
years (range 29 – 82 years) on conventional
low-protein diet (CD: 0.6 g/kg/day) with a
similar renal function (median sCr: 5.2
mg/dl), were investigated for lipids and apo-
lipoprotein parameters (traditional CVRF) as
well as for oxidative stress (oxidized LDL,
antibodies against OxLDL and thiobarbituric
acid-reactive substances (TBARS)), total
homocysteine (tHcy), lipoprotein(a) (Lp(a)),
albumin and c-reactive protein (CRP) (non-
traditional CVRF) including vitamins A, E,
B12 and folic acid. Results: Compared to pa-
tients on CD, those on VSD showed increased
HDL cholesterol levels (p < 0.005) with a re-
duction of LDL cholesterol (p < 0.01) and an
increase of apoA1/ apoB ratio (p < 0.02).
Among non-traditional CVRF, a mild but sig-
nificant reduction of OxLDL (p < 0.05) with
lower TBARS concentrations (p < 0.01) and a
significant reduction of total homocysteine (p
< 0.002), Lp(a) (p < 0.002) and CRP levels (p
< 0.05) were also observed in these patients.
Concentrations of vitamin E and A were not
different between the two groups while vita-
min B12 and folic acid resulted markedly in-
creased in patients on VSD. OxLDL signifi-
cantly correlated with total and LDL choles-
terol, triglycerides and Apo B in CD but not in
VSD patients. Patients on CD also showed a
significant correlation between urea and CRP.
After a multivariate analysis, only urea (p <
0.001) and OxLDL (p < 0.006) were associ-
ated to a risk of CRP > 0.3 mg/dl. Conclu-
sions: These results indicate a better lipopro-
tein profile in patients on vegan diet including
non-traditional CVRF. In particular, these pa-
tients show a reduced oxidative stress with a
reduced acute-phase response (CRP) as com-
pared to patients on conventional diet. We hy-
pothesize that urea, significantly lower in pa-
tients on VSD, may account, possibly to-
gether with the reduction of other protein
breakdown products, for the decreased
acute-phase response observed in these pa-
tients. Our findings suggest that low-protein
diets, and vegan in particular, may exert a
beneficial effect on the development of car-
diovascular disease in patients with end-stage
renal disease (ESRD).
Introduction
Patients with chronic renal disease show a
high prevalence of morbidity and mortality
for atherosclerotic cardiovascular disease
(CVD] [Foley et al. 1998]. Among risk fac-
tors for CVD, hyperlipidemia has been indi-
cated to play a major role in the development
Bergesio, Monzani, Guasparini et al.
of atherosclerotic disease [Chan et al. 1981,
Attman and Alaupovich 1991]. Lipid per-
oxidation, namely oxidized low-density lipo-
protein (OxLDL), would represent the trig-
gering event in this process [Witztum 1994].
Indeed, OxLDL would act as a chemo-
attractant for monocytes inducing an inflam-
matory reaction within the arterial wall whose
final event is the atherosclerotic plaque for-
mation [Kaplan and Aviram 1999]. To date,
much evidence was produced for the occur-
rence of LDL oxidation in vivo like, for in-
stance, the demonstration of oxidized LDL
[Avogaro et al. 1988] and anti-OxLDL auto-
antibodies (OxLDLAb) in the arterial wall as
well as in the circulation of both rabbit and
man [Salonen et al. 1992, Shojiet et al. 2000,
Yla-Herttuala et al. 1989]. In addition, auto-
antibodies against malondialdehydelysine-
modified LDL (MDA-LDL) were found in a
cross-sectional study by Salonen and co-
workers to be an independent predictor of the
progression of carotid atherosclerosis in
Finnish men [Salonen et al. 1992]. Con-
versely, they were suggested to exert a protec-
tive role by Shoji et al. in a recent prospective
study where they found an inverse associa-
tion between autoantibodies against OxLDL
and mortality from CVD [Shoji et al. 2002].
Accordingly, circulating OxLDL were found
to be elevated in patients affected by CVD
suggesting they might represent a possible
risk marker for the disease [Toshima et al.
2000]. Up to now, few and conflicting results
have been available concerning the role of ei-
ther levels of OxLDL or OxLDLAb in pa-
tients with renal failure [Bergesio et al. 2001,
Maggi et al. 1994]. Besides lipids peroxida-
tion, total homocysteine, lipoprotein a and
C-reactive protein would all act as independ-
ent cardiovascular risk factors both in general
population and in uremics [Moustapha et al.
1998].
According to the risk of development
CVD, vegan diet was reported to exert a bene-
ficial effect in the general population as indi-
cated by a more favorable BMI, a better lipo-
protein profile and a lower blood pressure
[McCarthy 1999, Ritter and Richter 1995,
Rouse et al. 1984]. In addition, it would also
contribute to maintain a higher antioxidant vi-
tamin status (vitamins C, E, and b-carotene),
thus preventing lipid peroxidation [Hostmark
et al. 1993, Rauma and Mykkanen 2000].
104
Accordingly very low protein vegan diet
supplemented with EAA and ketoanalogues
(vegan-supplemented diet (VSD)) was found
to improve serum total cholesterol and glucose
tolerance in diabetic patients with chronic re-
nal failure and to reduce triglyceride levels in
male patients with chronic uremia [Barsotti et
al. 1988, Ciardella et al. 1986].
More recently, a better lipid and apolipo-
protein profile including Lp(a), OxLDLAb ti-
tre and a reduced lipid peroxidation of eryth-
rocyte membranes has been pointed out in
patients on such a dietary regimen [Bergesio
et al. 2001, Peuchant et al. 1997].
Concerning homocysteine levels, patients
on vegan diet are reported to show increased
levels unless the diet is supplemented with vi-
tamin B12 and B6 [Bissoli et al. 2002,
Mezzano et al. 2000].
In the past years, much evidence has been
gathered about the role of acute-phase pro-
teins, namely C-reactive protein (CRP), and
reduced albumin levels in the development of
the atherosclerotic process and in predicting
cardiovascular mortality both in general pop-
ulation and in uremia [Ridker and Haughie
1998, Yeun et al. 2000]. However, no data are
available about the influence of diet on the
above factors in uremia.
The aim of our study is to investigate the
influence of vegan-supplemented diet (VSD)
versus conventional low-protein diet (CD) on
traditional and non-traditional cardiovascular
risk factors (CVRF) in patients with ad-
vanced chronic renal failure.
Materials and methods
Patients
Stable chronic renal failure outpatients
with moderate to severe renal failure (serum
creatinine > 3.5 mg/dl) either on vegan-sup-
plemented diet (VSD) or conventional diet
(CD) were considered for this observational
cross-sectional study.
Patients with nephrotic syndrome, diabe-
tes mellitus, obesity (BMI > 30), endocrine or
liver diseases, alcohol abuse, malignancy or
on immunosuppressive or steroid therapy
were excluded. Patients on lipid-lowering
treatment entered the study after a wash-out
period of at least eight weeks. Patients with
 Cardiovascular risk factors in patients with end-stage renal disease 105

acid according to the need. Patients followed

Table 1. Characteristics of patients.
the diet for a median time of 30 months (range
Vegan- Conventional 6 – 120 months).
supplemented low protein p All centers participating in the study fol-
diet diet lowed a similar dietetic approach consisting
at first in prescribing patients a conventional
Patient No. 29 31
protein restriction (0.6 g protein/kg/day). On
Age (years) 55 65
(29 – 79) (29 – 82) deterioration of renal function (sCr > 3.5
Gender (M/F) 18/11 20/11 mg/dl), compliant and motivated patients
BMI (kg/m2) 25 24 were offered, after a nutritional interview
(20 – 29) (21 – 29) with a dietician and a clinical evaluation by a
nPi* (g/kg/day) 0.35 0.6 p < 0.001 nephrologist, the possibility to switch to the
(0.3 – 0.6) (0.4 – 0.9)
vegan diet in order to obtain a better meta-
sCr (mg/dl) 6.5 5.2 p < 0.02
(3.5 – 10.5) (3.4 – 10.5) bolic control of the uremic syndrome.
Urea (mg/dl) 74 103 p < 0.001 Dietary protein intake was estimated by
(32 – 150) (56 – 147) calculation of protein catabolic rate (PCR)
using Maroni’s formula [Maroni et al. 1985].
* normalized protein intake, according to Maroni’s formula. A normalized protein catabolic rate (nPCR)
of 0.6 g/kg/day and of 0.9 g/kg/day was as-
sumed as the upper limits of tolerance for pa-
overt metabolic acidosis or weight loss in the tients respectively on VSD and CD (Table 1).
last six months were excluded as well. Patients with bad compliance to dietary
Sixty patients on either of the two dietary prescriptions (mean value of the last three-
treatments for at least six months were finally monthly determinations of nPCR over afore-
selected to participate in the study. mentioned limits) were excluded from the
One group included 29 patients: 18 males study.
and 11 females aged 55 years (range 29 – 79 All selected patients gave their informed
years) on very low protein vegetarian diet consent. The study was approved by the local
(0.3 g/kg/day) supplemented with a mixture Ethics Committee of the Florence Hospital.
of essential amino acids and ketoacids (ALFA
KAPPA, Shire, Firenze, Italy) at the dosage of
one tablet per 5 – 8 kg/day VSD. Energy pre- Methods
scription was 30 – 35 Kcal/kg/day supplied
by carbohydrates (63 – 67%), proteins (3 – Blood samples were collected after an
5%), fats (28 – 32% with an unsaturated/satu- overnight fast and then stored at –20 °C until
rated fatty acid ratio of 2.5/1). Calcium car- assayed for determination of both traditional
bonate and iron were administered when nec- and non-traditional CVRF; 10 ml of blood
essary. All patients were supplemented with a were allowed to coagulate and the serum ana-
mixture of B group vitamins in which vitamin lyzed for creatinine, albumin, lipids, apolipo-
B12 content ranged between 4 – 500 mg per proteins A1 and B, lipoprotein(a), C-reactive
day and vitamin B6 between 2 – 250 mg ac- protein, vitamins A and E and antibodies
cording to different commercially available against OxLDL; 5 ml of blood containing
tablets. Folic acid was not usually supplied. EDTA (1 mg/ml) were centrifuged at 3,000 g
Patients were on this diet for a median time of for 5 min and the plasma analyzed for thio-
15 months (range 6 – 93 months). barbituric acid-reactive substances, LDL-
The other group included 31 patients: 20 oxidized antigen and total homocysteine.
males and 11 females aged 65 years (range
29 – 82 years) on conventional low-protein
(0.6 g/kg/day) diet (CD). Energy supply was Traditional cardiovascular risk
30 – 35 Kcal/kg/day, supplied by carbohy- factors
drates (55 – 60%), proteins (8 – 10%), fats
(30 – 35% with unsaturated/saturated fatty Total cholesterol (TC), triglycerides (TG),
acid ratio of 3.5 : 1). The diet was supple- high-density lipoprotein cholesterol (HDL-C),
mented with calcium carbonate, iron and folic creatinine and albumin were analyzed using a
Bergesio, Monzani, Guasparini et al.
Figure 1. Traditional cardiovascular risk factors.
standard technique and commercial kits from
Abbott (Abbott Park, IL, USA) performed on
Aeroset instrument. LDL cholesterol (LDL-
C) was calculated using the Friedewald for-
mula.
Apolipoproteins A1 and B (Apo A1 and
Apo B) were assayed by immunonephelo-
metry (Dade-Behring, Newark, DE, USA).
Non-traditional cardiovascular
risk factors
Albumin and C-reactive protein (CRP)
were determined by immunonephelometry
using two high sensitivity kits (Dade Behring,
Newark, DE, USA) on Behring nephelometer
considering value of 0.3 mg/dl as a cut-off for
cardiovascular risk.
Total homocysteine (tHcy) was evaluated
in plasma using an FPIA technique on Imx
Analyzer (Abbott, Abbott Park, IL, USA).
OxLDL antigen (OxLDL) was measured
by a competitive ELISA method utilizing a
specific murine monoclonal antibody, mAb-
4E6 with the Mercodia oxidized LDL Elisa
Kit (Mercodia, Uppsala, Sweden) [Holvoet et
al. 1998].
Autoantibodies against oxidized LDL
(OxLDLAb) were determined by an ELISA
method using Cu(II)-oxidized LDL coated
onto microtiter strip as antigen (oLAB kit,
Biomedica-Gruppe, Vienna, Austria) [Eber
1994].
Thiobarbituric acid-reactive substances
(TBARS) were determined evaluating spec-
106
trophotometrically the reaction of serum with
thiobarbituric acid mixed with 1% of tri-
chloroacetic acid after heating for 30 min at
100 °C. The chromogen was extracted with
n-butanol; the absorbance of the organic
phase was determined at 535 nm against
blank [Richards 1992].
Vitamins A and E were determined by
HPLC-UV method.
Vitamin B12 and folates were analyzed si-
multaneously using a radioassay kit (ICN
Pharmaceuticals Simul TRAC-SNB RIA,
New York, NY, USA). The competitive
immunoassay utilizes two tracers 57 °C for
vitamin B12 and 125I for folates.
Lipoprotein (a) was measured by an
ELISA kit (Imubind Lp(a) ELISA Kit, Amer-
ican Diagnostica Inc, Greenwick, CT, USA)
utilizing affinity purified polyclonal antibod-
ies raised against Lp(a).
Statistics
All data are reported as median values
within range. The non-parametric Mann-
Whitney U-test for unpaired data was used to
compare each parameter between the two
groups of patients.
Spearman’s rank correlation coefficient
was employed for analysis of the relationship
between two variables.
A multiple linear regression model was
employed to evaluate the influence of several
parameters on CRP levels used as outcome
variable. Odds ratios for CRP > 0.3 mg/dl
were given with their 95% confidence inter-
val and were computed by multivariate logis-
tic regression analysis; a p < 0.05 was consid-
ered statistically significant.
Results
Age, sex and BMI distribution were simi-
lar in the two groups. On the contrary, serum
creatinine levels were slightly but signifi-
cantly higher in VSD group (p = 0.02).
Concerning the traditional CVRF, pa-
tients on VSD showed a better lipid and
apolipoprotein profile compared to patients
on CD. In particularly higher HDL-C levels
with a significant reduction of LDL-C were
found in the group on vegan diet who also
 Cardiovascular risk factors in patients with end-stage renal disease 107

showed an increased apoA1/B ratio (Figure
1). Among non-traditional CVRF, vegan pa-
tients showed a mild but significant reduction
of LDL oxidation with lower TBARS con-
centrations.
Figure 2. Nontraditional cardiovascular risk factors
and oxidative stress.
In spite of that, vegan patients did not dis-
play higher levels of vitamin E and A.
No significant differences of antibody
titers against OxLDL were found between the
two groups (Figure 2). Patients on CD
showed a significant correlation between
OxLDL and lipids or apolipoproteins (TC,
TG, LDL-C and Apo B) which was not ob-
served in vegan group (Table 2). Figure 4
shows the different relationship between
OxLDL and LDL-C, Apo-B in both groups of
patients.
In addition, patients on VSD showed sig-
nificantly lower levels of tHcy, Lp(a) and
CRP together with significantly increased
concentrations of vitamin B12 and folic acid
(Figure 3).
Patients on CD exhibited a positive corre-
lation between CRP and urea (p < 0.05) (Fig-
ure 5) and an inverse correlation between
Lp(a) and albumin both absent in vegan pa-
tients.
In Table 3, the differences of CRP, Lp(a)
and nutritional parameters such as urea, pro-
tein intake and albumin levels are reported.
A multiple linear regression model for
CRP levels was employed in the overall popu-
lation of patients. In this model, only urea re-
sulted significantly correlated to CRP levels
(Table 4). At the multivariate analysis, both
urea and OxLDL were found significantly as-
sociated to CRP levels > 0.3 mg/dl (Table 5).

Discussion

Our results point out to an overall benefi-

cial effect of vegan diet on traditional and
Figure 3. Nontraditional cardiovascular risk fac- non-traditional CVRF including, for the first
tors and related vitamins. time, a reduction of CRP levels.

Table 2. Spearman rank correlation analysis between OxLDL and oxidative stress and traditional cardiovascular risk factors.

LDL-C HDL-C TC TG A1 B A/B TBARS VIT E OxLDL-Ab

Vegan- r –0.20 0.11 – 0.20 0.11 0.06 – 0.25 0.14 0.18 0.08 – 0.09
supplemented
diet p NS NS NS NS NS NS NS NS NS NS

Conventional r 0.52 0.01 0.56 0.50 0.11 0.54 – 0.28 0.04 0.22 – 0.12
low-protein
diet p 0.002 NS 0.001 0.003 NS 0.001 NS NS NS NS
 Bergesio, Monzani, Guasparini et al. 108

Figure 4. Spearman rank correlation analysis between OxLDL and ApoB and LDL-C.

Figure 5. Spearman rank correlation analysis between Urea and CRP.

In particular, VSD patients showed not

Table 3. Protein intake (Pi), Urea, CRP, Lp(a) and albumin in patients on only a reduced lipid peroxidation, as evi-
vegan-supplemented (VSD) and conventional low-protein (CD) diet.
denced by the significant reduction of OxLDL
VSD CLPD p and TBARS levels, but also a reduced lipid
susceptibility to oxidation, as suggested by
Urea 74 103 < 0.001 the failure of a correlation between lipids and
(mg/dl) (40 – 150) (56 – 147) LDL oxidation.
Pi 26.5 42.5 < 0.001
The lower concentrations of OxLDLAb
(g/day) (15 – 37) (31 – 66)
Albumin 3 547 3 937 < 0.05 titre found in vegan patients could reflect the
(mg/dl) (2 364 – 4 751) (2 797 – 5 636) reduced LDL oxidation and are in keeping
CRP 0.2 0.4 < 0.05 with previous reports indicating OxLDLAb
(mg/dl) (0 – 0.7) (0 – 1.7) as a marker of the atherosclerotic process
Lp(a) 124 222 < 0.002
[Bergesio et al. 2001, Maggi et al. 1994]. By
(mg/dl) (11 – 800) (35 – 840)
contrast, Shoij and coworkers [2000] have re-
cently suggested a possible anti-atherogenic
role of OxLDLAb by maintaining a low
plasma concentration of OxLDL. Recently, a
Cardiovascular risk factors in patients with end-stage renal disease
Table 4. Multiple linear regression analysis of
different risk factors on CRP (n = 60).
r value p value
Pi (Maroni) 0.09 NS
Urea 0.52 < 0.001
tHcy 0.08 NS
Lp(a) –0.02 NS
OxLDL 0.17 NS
prospective observational study demon-
strated OxLDLAb titre to be a factor inde-
pendently associated with lower cardiovascu-
lar mortality in patients with end-stage renal
disease (ESRD).
Further investigations are needed to as-
certain the actual role of OxLDL and
OxLDLAb in the atherogenic process.
Several studies have reported an antioxidant
effect of vegan diet due to its higher content of
vitamins C, A and in particular of a-tocopherol
with a decrease of lipid peroxidation of both
plasma [Krajcovicova-Kudlackova et al. 1995,
Pronczuk et al. 1992] and red blood cell mem-
branes [Peuchant et al. 1997].
In our study the reduced lipid peroxida-
tion observed in vegan patients cannot be ex-
plained by the different vitamins A and E sta-
tus, as similar serum levels were detected in
both groups. However, it is possible that a
higher vitamin C content as well as an in-
creased catalase activity [Pronczuk et al.
1992] may contribute to such a reduction.
Another possible mechanism could be di-
rectly linked to the antioxidative action of
a-ketoacids which could scavenge hydrogen
peroxide thus reducing lipid peroxidation, as
reported by Nath and coworkers [1994] and
observed on RBC membranes by Peuchant
and coworkers [1997].
The inhibition of LDL oxidation could fi-
nally be influenced also by the higher HDL-C
levels linked, at least in part, to the presence of
several enzymes especially paraoxonase, PAF
acetylhydrolase and lecithin-cholesterol acyl-
transferase recognised to have an antioxidant
activity [Bonnefort-Rousselot et al. 1999].
Lp(a) and tHcy, both regarded as non-tra-
ditional factors which may affect cardiovas-
cular morbidity and mortality in the general
population [Boushey et al. 1995, Genest et al.
1991] as well as in uremia [Cressman et al.
109
Table 5. Multivariate analysis on relative risk of
CRP > 0.3 mg/dl (n = 60).
OR CI 95% p
Age 0.96 0.91 – 1.02 NS
Gender 0.66 0.1 – 3.9 NS
Urea 1.046 1.019 – 1.074 0.001
OxLDL 1.044 1.012 – 1.077 0.006
1992, Moustapha et al. 1998], were found to
be markedly reduced in vegan patients.
In particular, the lower tHcy levels found
in our patients on VSD could be explained
by the significantly higher concentrations of
folic acid, vitamin B12 and possibly vitamin
B6 (not investigated in our study) present in
these patients. Indeed vegan diet is rich in fo-
lic acid which, like vitamin E, would afford
antioxidant protection through a both direct
and indirect mechanism [Bayes et al. 2001].
On the other hand VSD is deficient of vitamin
B12 and B6 unless it is regularly supplemented
as we used to do in our patients. When this
supplementation is not provided, tHcy levels
are increased [Bissoli et al. 2002, Mezzano et
al. 2000].
Both Lp(a) and tHcy are reported to in-
duce, during their oxidative process, super-
oxide anion radical and hydrogen peroxide
generation [Loscalzo 1996], through which
they are believed to exert their major mecha-
nism for development of atherosclerotic
vascular disease.
Besides the possible role of vitamin C, re-
duced tHcy and Lp(a) levels, increased HDL-C
and folic acid concentrations together with a
direct antioxidant activity of a-ketoacids may
all contribute to the reduced LDL oxidation
observed in vegan patients.
Most important, patients on vegan diet
showed a moderate but significant reduction
of CRP levels, recently regarded as a strong
and independent predictor for CVD in both
predialysis and dialysis patients [Stenvinkel
et al. 1999, Zimmermann et al. 1999].
To date, many studies have been published
on the relationship between inflammation,
malnutrition and atherosclerosis in uremia
[Arici and Walls 2001, Bergstrom and Lind-
strom 1998, Stenvinkel et al. 1998, 1999, Zim-
mermann et al. 1999]. In dialysis patients,
Bergesio, Monzani, Guasparini et al.
Zimmerman and coworkers reported that the
presence of an inflammatory condition, indi-
cated by an elevated CRP and serum amyloid
A (SAA) concentration, could be responsible
for the decrease of serum albumin levels as
well as for the increased concentrations of
Lp(a), fibrinogen and even for many abnor-
malities of lipoprotein metabolism commonly
observed in chronic renal failure, like the re-
duced concentrations of HDL-C and Apo A1
including the general reduction of lipolytic en-
zymes activity [Zimmermann et al. 1999].
At the same time, the positive correlation
reported in recent investigations between
CRP and TBARS and OxLDL confirmed the
pathophysiologic link existing between the
acute-phase inflammatory response and
biomarkers of oxidant stress [Himmelfarb et
al. 2002, Stenvinkel et al. 1998].
Malnutrition and hypoalbuminemia have
been associated to increased cardiovascular
mortality in dialysis patients [Bergstrom and
Lindholm 1998]. However, both conditions
are often associated to an activated acute-
phase response which actually would be the
crucial factor for the development of CVD.
Despite their normal serum albumin levels,
patients on conventional diet showed both
Lp(a) and CRP concentrations significantly
higher than patients on vegan diet. The result
is even more interesting if we think that albu-
min levels of vegan patients are significantly
lower than those observed in patients on CD.
As a matter of fact, the reduced albumin con-
centrations cannot be ascribed either to any
evident inflammatory condition, as it is mani-
fested by normal C reactive protein level, or
to a real malnutrition as indicated by stable
clinical and biochemical parameters.
On the other hand, it is unlikely that latent
inflammatory conditions may have occurred
exclusively in patients on CD.
Indeed, both groups were just selected to
be in a good metabolic control and no appar-
ent differences could be elicited between
them. Consequently, we suggest that the dif-
ferences observed in CRP and Lp(a) levels
are likely to occur independently from nutri-
tional status and/or inflammatory condi-
tions.
The great difference in urea levels found
between the two groups may well represent a
possible explanation. Urea in fact was the
only factor to show a significant relative risk
110
for increased CRP levels. Multivariate linear
regression analysis further confirmed and
strengthened the role of urea as a major factor
affecting CRP concentrations in our study.
However, urea would not likely play a role by
itself, but rather as a biomarker of other re-
tained uremic solutes, also known as “uremic
toxins” like guanidines and others, which ac-
tually would be involved in inflammation and
atherogenesis [Glorieux et al. 2004].
To date, no direct data have been reported
about a possible causal relationship between
urea and/or uremic toxins and the acute-phase
proteins in ESRD. Irrespectively of whether
they act either as a surrogate, more likely, or
as a direct toxin, elevated urea levels would
reflect an overall increase of protein catabolic
products capable to maintain/activate an un-
derlying chronic inflammatory process as
demonstrated by the accumulation of CRP,
SAA and fibrinogen [Loscalzo 1996] found
in uremic patients either on conservative or
on dialysis treatment [Stenvinkel et al. 1999,
Pereira et al. 1994].
In the presence of increased concentrations
of urea, many molecules can be carbamylated
including LDL, which can be taken up by
scavenger receptors on monocytes and trig-
ger the atherosclerotic process [Horkko et al.
1992]. Accumulation of advanced glycation
end products (AGEs) in plasma and vas-
culature of uremic patients is another possible
mechanism through which the uremic state
leads to the production of proinflammatory
cytokines (IL-1, IL-6, TNF-a] [Schwedler et
al. 2001].
Moreover, Moeslinger et al. reported that
elevated urea concentrations in vitro, well
above indeed those currently observed in our
patients as well as in most patients on protein
restriction diets, might induce macrophage
proliferation and inhibition of inducible nitric
oxide synthesis thus contributing to develop-
ment of atherosclerotic lesions [Moeslinger et
al. 1999].
To date, although the biochemical path-
ways through which malnutrition, and more
often, inflammation may cause increased
urea levels in uremic patients are well-known
[Mitch 2002], little if no evidence at all exists
concerning the role of urea as a possible me-
diator of the acute-phase response with sub-
sequent increased oxidative stress and devel-
opment of lipoprotein abnormalities.
Cardiovascular risk factors in patients with end-stage renal disease
It is noteworthy that just moderately ele-
vated urea levels, as those found in our pa-
tients on CD, may be associated to an acti-
vated acute-phase response.
These findings, although derived from a
cross-sectional and not from a randomized
prospective or cross-over study, indicate that
a proper protein restriction in ESRD patients
may prevent and/or downregulate the devel-
opment of the inflammatory atherosclerotic
process, thus contributing to reduce cardio-
vascular morbidity and mortality.
In conclusion, we suggest that the more
favorable profile of traditional and non-tra-
ditional cardiovascular risk factors includ-
ing the better antioxidant status observed in
patients on vegan-supplemented diet is
likely the consequence of an overall reduced
activation of the acute-phase proteins sec-
ondary, at least in part, to a reduced urea and
uremic solutes generation. Our results stim-
ulate a new interest in the role of low-protein
diets in preventing the development of car-
diovascular disease, at present the leading
cause of mortality in uremic population.
References
Arici M, Walls J 2001 End-stage renal disease, cardiovas-
cular, and cardiovascular mortality: is C-reactive pro-
tein the missing link? Kidney Int 59: 407-414
Attman PO, Alaupovich P 1991 Lipid and apolipoprotein
profile of uremic dislypoproteinemia: relation to renal
function and dialysis. Nephron 54: 401-410
Avogaro P, Bittolo BG, Cazzolato G 1998 Presence of a
modified low density lipoprotein in humans. Arthero-
sclerosis 8: 79-97
Barsotti G, Navalesi R, Giampietro O, Ciardella F,
Morelli E, Cupisti A, Mantovanelli A, Giovannetti S
1988 Effects of a Vegetarian Supplemented Diet on re-
nal function, proteinuria, and glucose metabolism in
patients with “overt” diabetic nephropathy and renal
insufficiency. Contrib Nephrol 65: 87-94
Bayes B, Pastor MC, Bonal J, Junca J, Romero R 2001
Homocysteine and lipid peroxidation in hemodialysis:
role of folinic acid and vitamine E. Nephrol Dial
Transplant 16: 2172-2175
Bergesio F, Monzani G, Ciuti R, Cirami C, Martinelli F,
Salvadori M, Tosi P 2001 Autoantibodies against oxi-
dized LDL in chronic renal failure: role of renal func-
tion, diet and lipids. Nephron 87: 127-133
Bergstrom J, Lindholm B 1998 Malnutrition, cardiac dis-
ease and mortality: an integrated point of view. Am J
Kidney Dis 32: 834-841
Bissoli L, Di Francesco V, Ballarin A, Mandragona R,
Trespidi R, Brocco G, Caruso B, Bosello O, Zamboni
M 2002 Effect of vegetarian diet on homocysteine lev-
els. Ann Nutr Metab 46: 73-79
111
Bonnefont-Rousselot D, Therond P, Beaudeux JL, Peynet
J, Legrand A, Delattre J 1995 High density lipo-
proteins (HDL) and the oxidative hypothesis. Clin
Chem Lab Med 37: 939-948
Boushey CJ, Beresford SAA, Omenn GS, Motulsky AG
1995 A quantitative assessment of plasma homo-
cysteine as a risk factor for vascular disease. JAMA
270: 1049-1057
Chan MK, Varghese Z, Moorhead JF 1981 Lipid abnor-
malities in uremia, dialysis and transplantation. Kid-
ney Int 19: 625-637
Ciardella F, Morelli E, Niosi F, Caprioli R, Baldi R,
Cupisti A, Petronio G, Carbone C, Barsotti G 1986 Ef-
fects of a low phosforus, low nitrogen diet supple-
mented with essential aminoacids and ketoanalogues
on serum triglycerides of chronic uremic patients.
Nephron 42: 196-199
Cressman MD, Heyka RJ, Paganini EP, O’Neil J,
Skibinski CI, Hoff HF 1992 Lipoprotein(a) is an inde-
pendent risk factor for cardiovascular disease in
hemodialysis patients. Circulation 86: 475-482
Eber B, Schumacher M, Tatzber F, Kaufmann P, Luha O,
Esterbauer H, Klein W 1994 Autoantibodies to oxi-
dized low density lipoproteins in restenosis following
coronary angioplasty. Cardiology 84: 310-315
Ellis FR 1977 Angina and vegan diet. Am Heart J 93:
803-805
Foley RN, Parfrey PS, Sarnak MJ 1998 Cardiovascular
disease in chronic renal disease. Am J Kidney Dis 32
(Suppl 3): 112-119
Genest J Jr, Jenner JL, McNamara JR,Ordovas JM,
Silberman SR, Wilson PWF, Schaefer E 1991 Preva-
lence of lipoprotein(a) excess in coronary artery dis-
ease. Am J Cardiol 67: 1039-1045
Glorieux GL, Dhondt AW, Jacobs P, Van Langeraert J,
Lameire NH, De Deyn PP, Vanholder RC 2004 In vitro
study of the potential role of guanidines in leukocyte
functions related to atherogenesis and infection. Kid-
ney Int 65: 2184-2192
Himmelfarb J, Stenvinkel P, Ikizler TA, Hakim RM 2002
The elephant in uremia: oxidant stress as a unifying
concept of cardiovascular disease in uremia. Kidney
Int 62: 1524-1538
Holvoet P, Vanhaecke J, Janssens S, Van de Werf F,
Collen D 1998 Oxidized LDL and malondialdehyde-
modified LDL in patients with acute coronary syn-
dromes and stable coronary artery disease. Circulation
98: 1487-1494
Horkko S, Savolainen MJ, Kervinen K, Kesaniemi YA
1992 Carbamylation-induced alterations in low-density
lipoprotein metabolism. Kidney Int 41: 1175-1181
Hostmark AT, Lystad E, Vellar OD, Hovi K, Berg JE 1993
Reduced plasma fibrinogen, serum peroxides, lipids,
and apolipoproteins after a three-week vegetarian diet.
Plant Foods Hum Nutr 43: 55-61
Kaplan M, Aviram M 1999 Oxidized low-density lipo-
protein: atherogenic and role proinflammatory charac-
teristics during macrophage foam cell formation. An
inhibitory role for nutritional antioxidants and serum
paraoxonase. Clin Chem Lab Med 37: 777-787
Krajcovicova-Kudlackova N, Simoncic R, Babinska K,
Bederova A 1995 Levels of lipid peroxidation and an-
tioxidants in vegetarians. Eur J Epidemiol 11: 207-
211
Loscalzo J 1996 The oxidant stress of hyperhomo-
cysteinemia. J Clin Invest 98: 5-7
Maggi E, Bellazzi L, Falaschi F, Frattoni A, Perani G,
Finardi G, Gazo A, Nai M, Romanini D, Bellomo G
Bergesio, Monzani, Guasparini et al. 112

1994 Enhanced LDL oxidation in uremic patients: an 2000 Inverse relationship between circulating oxi-
additional mechanism for accelerated atherosclerosis? dized low density lipoprotein (oxLDL and anti-
Kidney Int 45: 876-883 oxLDL) antibody levels in healthy subjects. Athero-
Maroni BJ, Steinman T, Mitch WE 1985 A method for es- sclerosis 148: 171-217
timating nitrogen intake of patients with chronic renal Shoji T, Fukumoto M, Kimoto E, Shinohara K, Emoto M,
failure. Kidney Int 27: 58-65 Tahara H, Koyama H, Ishimura E, Nakatani T, Miki K,
McCarty MF 1999 Vegan proteins may reduce risk of Tsujimoto Y, Tabata T, Nishizawa Y 2002 Antibody to
cancer, obesity and cardiovascular disease by promot- oxidized low-density lipoprotein and cardiovascular
ing increased glucagon activity. Med Hypotheses 53: mortality in end-stage renal disease. Kidney Int 62:
459-485 2230-2237
Mezzano D, Kosiel K, Martinez C, Cuevas A, Panes O, Stenvinkel P, Heimburger O, Paultre F, Diczfalusy U,
Aranda E, Strobel P, Perez DD, Pereira J, Rozowski J, Wang T, Berglund L, Jogestrand T 1999 Strong associ-
Leigton F 2000 Cardiovascular risk factors in vegetar- ation between malnutrition, inflammation, and athero-
ians. Normalization of hyperhomocysteinemia with sclerosis in chronic renal failure Kidney Int 55:
vitamin B (12) and reduction of a platelet aggregation 1899-1911
with n-3 fatty acids. Thromb Res 100: 153-160 Stenvinkel P, Heinburger O, Tuck CH, Berglund L 1998
Mitch W 2002 Malnutrition: a frequent misdiagnosis for Apo(a) isoform size, nutritional status and inflamma-
hemodialysis patients. J Clin Invest 110: 437-439 tory markers in chronic renal failure. Kidney Int 53:
Moeslinger T, Friedl R, Monika IV, Brunner M, Baran H, 1336-1342
Koller E, Spieckermann PG 1999 Urea induces Toshima S, Hasegawa A, Kurabayashi M, Itabe H,
macrophage proliferation by inhibition of inducible Takano T, Sugano J, Shimamura K, Kimura J,
nitric oxide synthesis. Kidney Int 56: 581-588 Machishita I, Suzuki T, Nagai R 2000 Circulating oxi-
Moustapha A, Maso A, Nahlawi M, Gupta A, Aheart KL, dized low density lipoprotein levels. A biochemical
Jacobsen DW, Robinson K, Dennis VW 1994 Prospec- risk marker for coronary heart disease. Arterioscler
tive study of hyperhomocysteinemia as an adverse Thromb Vasc Biol 20: 2243-2247
cardiovascular risk factor in end-stage renal disease. Witztum JL 1994 The oxidation hypothesis of atheroscle-
Circulation 97: 138-41 rosis. Lancet 344: 793-795
Nath KA, Fischereder M, Hostetter TH 1994 The role of Yeun JY, Levine RA, Mantadilok V, Kaysen GA 2000 C-re-
oxidant in progressive renal injury. Kidney Int 45: active protein predicts all-cause and cardiovascular
S111-S115 mortality in hemodialysis patients. Am J Kidney Dis
Pereira BJG, Shapiro L, King AJ, Falagas ME, Strom JA, 35: 469-476
Dinarello CA 1994 Plasma levels of IL-1b, TNF-a and Yla-Herttuala S, Palinski W, Rosenfeld ME, Partha-
their specific inhibitors in undialyzed chronic renal sarathy S, Careu TE,Butler S, Witztum JL, Steinberg D
failure, CAPD and hemodialysis patients. Kidney Int 1998 Evidence for the presence of oxidatively modi-
45: 890-896 fied low density lipoprotein in atherosclerotic lesions
Peuchant E, Del Mas-Beauvieux MC, Dubourg L, of rabbit and man. J Clin Invest 84: 1086-1095
Thomas MJ, Perromat A, Aparicio M, Clerc M, Zimmermann J, Herrlinger S, Pruy A, Metzger T, Wanner
Combe C 1997 Antioxidant effects of a supplemented C 1999 Inflammation enhances cardiovascular risk
very low protein diet in chronic renal failure. Free and mortality in hemodialysis patients. Kidney Int 55:
Radic Biol Med 22: 313-320 648-658
Pronczuk A, Kipervarg Y, Hayes KC 1992 Vegetarians
have higher plasma alpha-tocopherol relative to choles-
terol than do nonvegetarians. J Am Coll Nutr 11: 50-55
Rauma AL, Mykkanen H 2000 Antioxidant status in
vegetarians versus omnivores. Nutrition 16: 111-119
Richard MJ, Portal B, Meo J, Coudray C, Hadjian A,
Favier A 1992 Malondialdehyde kit evaluated for
determining plasma and lipoprotein fractions that
react with thiobarbituric acid. Clin Chem 38: 704-709
Ridker PM, Haughie P 1998 Prospective studies of C-re-
active protein as a risk factor for cardiovascular dis-
ease. J Invest Med 46: 391-395
Ritter MM, Richter WO 1995 Effects of a vegetarian life
style on health. Fortschr Med 113: 239-242
Rouse IL, Beilin LJ, Armstrong BK, Vandongen R 1984
Vegetarian diet, blood pressure and cardiovascular
risk. Aust NZ J Med 14: 439-443
Salonen JT, Yla-Herttuala S, Yamamoto R, Butler S,
Korpela H, Salonen R, Nyyssonen K, Palinsky W,
Witztum JL 1992 Autoantibody against oxidized LDL
and progression of carotid atherosclerosis. Lancet
339: 883-887
Schwedler S, Schinzel R, Vaith P, Wanner C 2001 Inflam-
mation and advanced glycation end products in ure-
mia: simple coexistence, potentiation or causal rela-
tionship? Kidney Int Suppl 78: S-32-S36
Shoji T, Nishizawa Y, Fukumoto M, Shimamura K,
Kimura J, Kanda H, Emoto M, Kawagishi T, Morii H