Вы находитесь на странице: 1из 8

INT J TUBERC LUNG DIS 19(11):13691375

Q 2015 The Union


http://dx.doi.org/10.5588/ijtld.14.0963

The impact of nutritional state on the duration of sputum


positivity of Mycobacterium tuberculosis
K. Hatsuda,* M. Takeuchi, K. Ogata, Y. Sasaki, T. Kagawa, H. Nakatsuji, M. Ibaraki,
M. Sakaguchi, M. Kurata, S. Hayashi
*Department of Clinical Laboratory, Kinki-Chuo Chest Medical Center, National Hospital Organization, Osaka,

Graduate School of Human Environmental Sciences, Mukogawa Womens University, Nishinomiya, Departments
of Food and Nutrition, and Internal Medicine, Kinki-Chuo Chest Medical Center, National Hospital Organization,
Osaka, Japan
SUMMARY
B A C K G R O U N D : The outcome of anti-tuberculosis
treatment varies according to patient factors.
O B J E C T I V E : To retrospectively identify risks related to
the extension of time to negative sputum culture (Tn)
and to determine their clinical significance.
D E S I G N : Patients with bacilli susceptible to isoniazid
and rifampicin who received initial standard treatment
without cessation were recruited into the study. A total
of 630 consecutive in-patients were included in the risk
development analysis (development cohort) and another
611 consecutive in-patients in the risk validation
analysis (validation cohort).
R E S U L T S : Univariate analysis showed that Tn was
related to sex, body mass index (BMI), white blood cell
count (WBC), serum albumin, fasting blood sugar,

haemoglobin A1c, C-reactive protein and total cholesterol levels and sputum smear positivity (SSP). Multivariate analysis showed that BMI, WBC and SSP were
significant risk factors related to extended Tn. Optimal
cut-offs of BMI and WBC for predicting good (Tn , 46
days) and poor responders (Tn 7 46 days) according to
each risk were determined by receiver operating
characteristics analysis. Risks were verified with the
validation cohort. Tn increased according to the number
of risks; the median Tn for patients with three risks was
21 days longer than that of patients with none.
C O N C L U S I O N : The nutritional state of a TB patient
can be used to predict Tn.
K E Y W O R D S : body mass index; glucose intolerance;
white blood cells

TUBERCULOSIS (TB) remains a major cause of


morbidity and mortality worldwide.1 In Japan, TB
incidence has been decreasing steadily in the past 10
years,2 but it remains categorised as an intermediate
TB burden country, at 19.4 per 100 000 population.3
Hospitalisation and hospital discharge of infectious
TB patients in Japan are controlled by laws designed
to minimise the spread of infection;4 95.7% of
sputum smear-positive patients are hospitalised.5
The provision of sufficient numbers of isolation
facilities with an acceptable number of empty beds
is therefore critical for the efficient use of health
resources and the elimination of public health risks.
To meet this goal, sputum culture conversion needs to
be confirmed in a timely manner.
One of the significant concerns of health administration in terms of anti-tuberculosis treatment is the
delay with which sputum cultures convert to negativity after standard treatment of patients.2 We and
other investigators found that low body mass index
(BMI) and malnutrition are critical to the develop-

ment of TB,68 and that mortality due to TB is


significantly high in populations with low BMI.9
However, few studies conducted on the relationship
between nutritional markers and TB outcomes have
assessed conversion to negative sputum cultures as an
outcome. Identifying TB patients who may produce
persistently positive sputum cultures and directing
interventions at these patients may comprise a
rational and promising approach for improving
responsiveness to anti-tuberculosis treatment and
for reducing the period of hospitalisation.
The present study aimed to determine nutritional
risk factors for delays in negative conversion of
sputum TB cultures.

STUDY POPULATION AND METHODS


Patients
To assess risks that delay negative conversion of TB
sputum culture, this study included 1204 consecutive
patients with microbiologically or clinically diag-

Correspondence to: Seiji Hayashi, Kinki-Chuo Chest Medical Center, National Hospital Organization, 1180 Nagasone-cho,
Kita-ku, Sakai-city, Osaka 591-8555, Japan. Tel: (81) 722 52 3021. Fax: (81) 722 51 1372. e-mail: shayashi@kch.hosp.
go.jp
Article submitted 18 December 2014. Final version accepted 11 June 2015.

1370

The International Journal of Tuberculosis and Lung Disease

nosed TB admitted to the Kinki-Chuo Chest Medical


Center, Osaka, Japan (herein referred to as the
Center) between April 2005 and March 2007.
Relapsed cases or those with isoniazid (INH) or
rifampicin (RMP) resistance were excluded. Of the
1204 participants, 630 (431 males and 199 females)
had sputum cultures positive for Mycobacterium
tuberculosis and had completed .2 weeks of
standard treatment without interruption. Participants
who met the eligibility criteria comprised the
development cohort.
To determine the importance of the above risks, a
further group of patients was recruited independently
of the development cohort. Between April 2007 and
March 2013, of 1234 consecutive microbiologically
or clinically diagnosed TB patients admitted to the
Center, 611 (415 males and 196 females) TB sputum
culture-positive patients who met the same eligibility
criteria as those of the development cohort were
recruited for investigation; this was the validation
cohort.
This investigation did not routinely test enrolled
patients for human immunodeficiency virus status.
Disseminated TB or other diseases indicating acquired immune-deficiency syndrome were not observed in the study population.
The Institutional Review Board of the Kinki-Chuo
Chest Medical Center, Osaka, Japan, approved the
study.
Diagnosis and assessment of the effects of antituberculosis treatment
Sputum samples obtained from patients were placed
on glass slides and stained using the Ziehl-Neelsen
procedure. Acid-fast bacilli (AFB) were detected
using bright-field microscopy at 10003 magnification, i.e., under a high-power field (HPF). TB bacilli
count was classified into the following grades:
negative (no AFB in 100 HPF), scanty (19 AFB/
100 HPF), 1 (1099 AFB/100 HPF), 2 (110 AFB/1
HPF on average) and 3 (.10 AFB/1 HPF on
average).10 The amount of M. tuberculosis in the
sputum before starting anti-tuberculosis treatment
(i.e., sputum smear positivity [SSP]) was defined as
the highest sputum positive grade among three
consecutive sputum samples before starting the
administration of anti-tuberculosis drugs.
Sputum cultures of M. tuberculosis were confirmed
using polymerase chain reaction (Spoligotyping Kit,
Ocimum Biosolutions, Hyderabad, India), immunochromatography (Capilia TB-Neo, Tauns Laboratories Inc, Shizuoka, Japan) or both. Time to negative
sputum culture (Tn) was defined as the period from
detection of the first sputum culture positive for M.
tuberculosis to that of the first of three consecutive
sputum cultures negative for M. tuberculosis. In this
study, the first sputum sample was tested on the first
day of admission, and sputum samples were subse-

quently tested every 2 weeks up to the end of


treatment and every month thereafter up to the end
of follow-up. Standard anti-tuberculosis treatment
(RMP, 10 mg/kg bodyweight (BW)/day; INH, 5 mg/
kg BW/day; pyrazinamide, 25 mg/kg BW/day, and
ethambutol, 15 mg/kg BW/day) was started within 1
week after admission and continued for 4 months,
followed by INH and RMP for 2 or 5 months.11
Physical examination and laboratory tests
At the time of admission, height and BW were
measured and blood was taken for sampling. TB
patients with fasting blood sugar (FBS) levels of
7126 mg/dl, haemoglobin A1c levels (HbA1c) of
76.5%12 or both were diagnosed as glucose intolerance. Mean voluntary dietary intake during the first
week of admission was recorded by nurses. Energy
intake level (EIL) was defined as the mean voluntary
dietary intake divided by the dietary reference intake
among Japanese subjects.13
Statistical analysis
Spearmans rank correlations were performed to
determine the association among laboratory tests.
Mann-Whitney tests were used to compare the EIL
and haematological results of each subgroup. Tn was
calculated using the Kaplan-Meier method, and the
log-rank test was used to evaluate the difference in
Tn. Cox proportional hazard regression analysis with
a step-up procedure was used in multivariate analysis
of Tn, and the likelihood ratio was used as the
criterion for adding significant variables. v2 tests were
performed to examine categorical data. SPSS, version
15.0J (Statistical Package for the Social Sciences,
Chicago, IL, USA) was used for statistical calculations. Statistical significance was assumed at P ,
0.05.

RESULTS
Background characteristics of both development and
validation cohorts are shown in Figure 1. In both
cohorts, height, BMI and C-reactive protein (CRP)
titre of male patients (B, C and F) were significantly
higher than those of female patients. In the validation
cohort, white blood cell count (WBC) was higher
among males (D) than females. In the development
cohort, the FBS levels among males (H) were
significantly higher than among females. In both
cohorts, the FBS level was higher among males than
females.
In the development cohort, the numbers of male
and female subjects with glucose intolerance were
respectively 137 (32.6%) and 42 (21.5%), compared
to respectively 165 (40.0%) and 64 (32.8%) in the
validation cohort. In both cohorts, rates of glucose
intolerance were higher in males than in females. The

Nutrition and anti-tuberculosis treatment outcomes

1371

Figure 1 Background characteristics of tuberculosis patients. Box-and-whisker diagrams denote


ranges of various parameters of the development cohort (open box) and the validation cohort (dot
meshing box). The first and second numbers in parentheses denote numbers of subjects tested in
the development cohort and in the validation cohort, respectively. A) Age; B) haemoglobin; C) BMI;
D) WBC count; E) lymphocyte count; F) CRP level; G) albumin level; H) fasting blood sugar level; I)
Hb A1c level. The Mann-Whitney test was used to compare data. * P , 0.05. P , 0.01. BMI
body mass index; WBC white blood cell; CRP C-reactive protein; HbAIC haemoglobin A1c.

percentage of glucose intolerance in both males and


females was significantly higher in the validation
cohort than in the development cohort. The various
comorbidities present and the percentage of patients
affected in the development cohort were as follows:
cerebrovascular disease (6.0%), cardiovascular disease (4.9%), gastrointestinal disease (27.8%), pulmonary disease other than TB (20.9%), hepatic
disease (9.8%), urological or gynaecological disease
(10.1%), collagen vascular disease (1.8%), malignancy in any organ (7.8%) and history of systemic
corticosteroid administration (3.4%).
Chronological change in sputum culture positivity
and distribution of culture-positive periods in the
development cohort were examined. The Tn curve in
Figure 2 closely fits an exponential equation, y
0.9806e0.03x. Histograms indicating the distribution
of Tn values are shown in the inset (vertical bars and
left axis). Tn values of the development cohort are
arranged in ascending order, with a summation of
culture-positive periods (person x day) shown in the
inset (polyline and right axis). The period resulting in
50% of total Tn values for all subjects (20 199 person
days) was a Tn of 46 days. Patients with Tn values of
,46 days and 746 days were thus designated good
and poor responders, respectively.
Background characteristics of good and poor
responders are shown in Table 1. Among male
patients, EIL during the first week of admission and

BMI at the time of admission were significantly lower


in poor responders than in good responders, while
CRP levels were significantly higher in poor responders than in good responders. Among female patients,
BMI was significantly lower in poor responders than
in good responders, while CRP levels were significantly higher in poor responders than in good
responders. Regardless of the response to standard
treatment, EIL among male patients was lower than
that of female patients in the same cohort.
M. tuberculosis amounts in sputum samples were
compared: sputum smear grades negative, scanty and
1 indicated a low SSP score, whereas 2 and 3
indicated a high SSP score. The incidence of high SSP
was significantly higher in good than poor male
responders. However, among females in the cohort,
there was no significant difference between good and
poor responders in the incidence of high SSP scores.
Of the 630 patients in the development cohort, 21
died during the observation period. No significant
difference was observed in death rates between good
and poor responders.
Testing the relationship of clinical parameters and
Tn using univariate analysis (Table 2), we found that
Tn was significantly associated with sex, BMI, WBC,
serum albumin concentration, FBS, HbA1c, CRP,
total cholesterol levels (T-chol) and SSP score. Sexsegregated analysis using the log-rank test showed
that the Tn of male patients was significantly longer

1372

The International Journal of Tuberculosis and Lung Disease

Figure 2 Time to negative sputum culture for participants in the development cohort.
Chronological change in sputum culture positivity of tuberculosis subjects from the entire
development cohort (closed dots) was approximated with the curve, y 0.9806e0.03x (grey line).
Vertical bars (left axis) a histogram of the distribution of Tn values of the entire development
cohort. Tn values of the development cohort are arranged in ascending order; the polyline (right
axis) a summation curve of culture-positive periods (person-days). The Tn value that
corresponded to 50% of total Tn values for all subjects (20 199 person days) was 46 days. Tn
time to negative.

than that of female patients (P , 0.05) (data not


shown); median Tn values for males and females were
respectively 28 and 24 days. Comorbidities and Tn
were analysed using univariate analysis for the
presence of any correlations, but none of the
comorbidities were significantly associated with Tn
prolongation (data not shown).
Predictors of extended Tn were then extracted.
Multicollinearity among the covariates was tested
before performing multivariate analysis. Albumin
showed a negative correlation with CRP (r
0.643, P , 0.01) and a positive correlation with
T-chol (r 0.426, P , 0.01). Positive but weak
correlations were observed between WBC and the
markers CRP (r 0.232, P , 0.01), Tn (r 0.151, P ,
0.01) and SSP (r 0.175, P , 0.01). As FBS and
HbA1c levels showed a statistically significantly
negative correlation (r 0.635, P , 0.01), we
defined glucose intolerance by combining these two

parameters, as shown in the Methods. On univariate


analysis, glucose intolerance was significantly associated with elongation of Tn (hazard ratio [HR] 1.233,
95% confidence interval 1.0011.519, P 0.049).
On the basis of the results of univariate analysis and
of multicollinearity analysis, we included age, sex,
BMI, WBC, albumin, glucose intolerance and SSP
score as covariates for multivariate analysis. The
results imply that low BMI, high WBC and high SSP
score were all significantly associated with the
extension of Tn (Table 3).
To estimate a cut-off value that distinguishes
between good and poor responders, an analysis of
receiver operating characteristics was conducted
(Figure 3). The optimal cut-offs for distinguishing
between good and poor responders were 20.0 kg/m2
for BMI and 7650/mm3 for WBC. We tested the
reliability of the three predictorslow BMI (,20.0
kg/m2), high WBC (.7650/mm3) and high SSP score

Table 1 Background characteristics of good and poor responders in the development cohort*
Male
Good responders
Age, years, median (range)
EIL, %, median (range)
BMI, kg/m2, median (range)
CRP, mg/dl, median (range)
High SSP, n/N

64.5
85.7
20.5
2.1

(17.1103.1)
(0.0117.8)
(13.828.2)
(0.028.6)
180/324

Poor responders
64
84.5
19.1
5.3

(22.088.0)
(0.0105.7)
(12.126.9)
(0.135.5)
90/105

Female
P value

Good responders

NS
,0.05
,0.01
,0.01
,0.01

59.1
98.5
19.8
1.1

(16.097.1)
(0.0134.8)
(11.735.8)
(0.018.9)
94/165

Poor responders
68
91.4
18.2
4

(18.088.9)
(10.2134.8)
(14.830.4)
(0.019.1)
24/32

P value
NS
NS
,0.05
,0.05
NS

* The Mann-Whitney test was used to compare age, EIL and BMI data between male and female patients and between good and poor responders. The v2 test was
used to compare SSP data between male and female patients and between good and poor responders.

The difference between EIL of good male responders and good female responders was statistically significant (P , 0.01).

The difference between EIL of poor male responders and poor female responders was not statistically significant.
NS not significant; EIL energy intake level at the first week of admission; BMI body mass index; CRP C-reactive protein; SSP sputum smear positivity.

Nutrition and anti-tuberculosis treatment outcomes

1373

Table 2 Univariate analysis for the association of Tn with


clinical parameters*
HR (95%CI)
Sex
Age
Body mass index
White blood cells
Haemoglobin
Lymphocytes
Albumin
Fasting blood sugar
Haemoglobin A1c
C-reactive protein
Total cholesterol
Sputum smear positivity

0.844
0.999
1.069
0.994
1.026
1.010
1.255
0.999
0.864
0.958
1.005
0.527

P value

(0.7130.999)
(0.9951.003)
(1.0421.096)
(0.9910.997)
(0.9871.066)
(0.9971.022)
(1.1091.420)
(0.9971.000)
(0.8010.932)
(0.9420.974)
(1.0031.007)
(0.3940.706)

,0.05
NS
,0.01
,0.01
NS
NS
,0.01
,0.05
,0.01
,0.01
,0.01
,0.01

* Univariate Cox proportional hazard regression analysis.


Tn time to negative cultures of TB bacilli; HR hazard ratio; CI confidence
interval; NS not significant.

(smear grades 2 or 3)by using data from the


validation cohort, whose constituents were independent of the development cohort (Figure 4). The
validation cohort was divided into four groups
according to the number of risks: G0 (patients with
no risk), G1 (patients with one of three risks), G2
(patients with two of three risks), and G3 (patients
with all three risks). The Tn increased in accordance
with the number of risks, and Tn values for G2 and
G3 were significantly longer than for G0. The median
Tn for G3 was 21 days longer than that for G0. Of the
611 patients in the validation cohort, eight died
during the observation period. No significant difference was observed in death rates between G0, G1, G2
and G3.

DISCUSSION
We found that low BMI, high WBC and high SSP
score at the time of admission were significantly
associated with extended Tn. By applying these risk
factors to patients who were independent of the
development cohort, we validated their reliability as
predictors of Tn extension. BMI may reflect nutritional status, while WBC may reflect TB severity,
because positive (albeit weak) correlations were
observed between WBC and other markers such as
CRP, Tn and SSP. Because the risk factors identified in
our study are representative and widely used clinical
parameters, the predictors advocated in our study are
highly versatile.
Our study demonstrated that BMI, which is a
Table 3

Figure 3 Receiver operating characteristics curve analysis used


to predict responses to standard treatment for tuberculosis. A)
Analysis of body mass index showed on optimal cut-off value
distinguishing between good responders (Tn , 46 days) and
poor responders (Tn 746 days) of 20.0 kg/m2. AUC, sensitivity
and 1-specificity were 0.620, 0.537 and 0.350, respectively. B)
Analysis of white blood cell count showed an optimal cut-off
value that distinguishes between good and poor responders of
7650/mm3. AUC, sensitivity and 1-specificity were 0.614, 0.489
and 0.299, respectively. Tn time to negative; AUC area under
the curve.

representative nutritional marker, was significantly


associated with Tn extension. According to the
WHO, BMI ,18.5 kg/m2 is regarded as chronic
energy deficiency.14 However, in this investigation,
not only emaciated but also underweight patients
(i.e., those with a BMI 18.520.0 kg/m2) were at
higher risk of Tn prolongation. Albumin was not
significantly associated with Tn using multivariate
analysis; however, albumin was significantly associated with Tn in univariate analysis (HR 1.255). This
result can be partly explained by altered immune
system responses against TB caused by malnutrition.15 Albumin is a regularly assessed and useful
nutritional marker.16 A high score for nutritional risk,
such as in cases with low BMI and low albumin, is a
good predictor of mortality among pulmonary and
miliary TB patients.17,18 Similarly, Pednekar et al.
reported that extremely thin patients were at a high
risk of dying due to TB.9
The issue of the most profound clinical significance
is the improvement in cure and mortality rates.9,1820
As mentioned above, a Tn of 46 days distinguishes
between good and poor responders to standard antituberculosis treatment. As medical resources invested
in anti-tuberculosis treatment correspond to the
cumulative number of sick beds occupied by patients,
the grouping we propose is useful for identifying
patients who are at risk and who therefore require

Multivariate analysis for the association of Tn with clinical parameters*

Parameter
Body mass index
White blood cells
Sputum smear positivity

Partial regression coefficient

HR (95%CI)

P value

0.050
0.003
0.664

1.052 (1.0251.079)
0.997 (0.9930.999)
1.943 (1.6192.332)

,0.01
,0.05
,0.01

* Cox proportional hazard regression analysis with a step-up procedure was performed. The likelihood ratio was used
as the criterion for adding significant variables.
Tn time to negative cultures of TB bacilli; HR hazard ratio; CI confidence interval.

1374

The International Journal of Tuberculosis and Lung Disease

Figure 4 Chronological changes in sputum culture positivity in


the validation cohort. Patients were divided into four groups
according to number of risk factors: G0 (thick solid line: patients
with no risks, n 66); G1 (thin solid line: patients with 1/3 risks,
n 207), G2 (dotted line: patients with 2/3 risks, n 213) and
G3 (dashed line: patients with all 3 risks, n 92). Between-group
comparisons were performed using the log-rank test. Tn time
to negative; NS not significant.

more medical resources. We would therefore like to


highlight the significance of assessing Tn from a
financial viewpoint.
Because TB is an airborne infection,21 isolation is
sometimes recommended to minimise the spread of
infection.11 However, with the continuous reduction
in the number of TB patients, the rate of dedicated
isolation facilities for TB is decreasing in Japan.3 To
use medical resources effectively, reliable predictors
for the length of the infectious period urgently need to
be developed; predictors should be able to distinguish
patients with risk factors for extended Tn. Nutritional
interventions could then be used to improve treatment outcomes and quality of life. In communities
with decreasing numbers of TB patients, our approach may contribute to the more effective use of
medical resources.
Univariate analysis indicated that FBS and HbA1c
were significantly associated with extension of Tn.
Diabetes mellitus (DM), another metabolic and
nutritional issue associated with TB, induces delay
in conversion to negativity of sputum cultures.22,23 In
the present study, multivariate analysis showed that
glucose intolerance was not a significant predictor of
extension of Tn. DM has been reported to increase the
likelihood of treatment failure,24 and to extend the
time to culture conversion.19 These reports appear to
contradict our observations. The reason for this
discrepancy, however, is unclear. Possible reasons
may be differences in the incidence of glucose
intolerance between the development and validation
cohorts, as well as differences between males and
females. Recruitment for the validation cohort, which
followed the recruitment of the development cohort,
took 6 years due to the curtailment of the Centers
isolation facilities following the decreasing trend in

TB morbidity in Japan.3 The difference in incidence


between the two cohorts might be affected by the
increasing prevalence of DM.2528 Furthermore, male
patients fared worse, as demonstrated by the lower
caloric intake and higher levels of CRP (Table 1 and
Figure 1). These factors may explain the extension of
Tn in male patients. Poorer nutritional status might be
a reflection of lifestyle, as total energy uptake at the
time of admission might reflect eating habits immediately before admission. Although average caloric
and animal protein intake in Japan have been
sufficient in recent years,29 the EIL of TB patients
may be low. Investigations are required into lifestyle
and dietary habits before admission that contribute to
improved therapeutic outcomes.

CONCLUSION
We found that some nutritional markers precisely
predict the delay of conversion of sputum cultures to
negativity. The assessment of nutritional markers at
the time of admission will contribute to the efficient
use of sick beds and isolation facilities dedicated to
TB patients, particularly during periods of continuous reduction in TB morbidity. Nutritional interventions may improve therapeutic outcomes for TB
patients, and the establishment of protocols to
improve such outcomes for TB patients is a critical
problem that may be addressed in the future.
Acknowledgements
This study was supported by a Grant-in-Aid from the Osaka
Tuberculosis Foundation, Osaka, Japan. The authors are grateful to
R Hayashi for her linguistic assistance.
Conflicts of interest: none declared.

References
1 World Health Organization. Global tuberculosis report, 2013.
WHO/HTM/TB/2013.11. Geneva, Switzerland: WHO, 2013.
2 Ministry of Health, Labor and Welfare, Japan. Changes in
number of newly registered tuberculosis patients, prevalence
rate, and the number of deaths. Tokyo, Japan: Ministry of
Health, Labor and Welfare, 2012. http://www.mhlw.go.jp/
english/wp/wp-hw5/dl/23010226e.pdf Accessed July 2015.
3 Ministry of Health, Labor and Welfare, Japan. Number and
incidence rate of newly registered tuberculosis patients by
patient classification and age group. Tokyo, Japan: Ministry of
Health, Labor and Welfare, 2010.
4 Ministry of Health, Labor and Welfare, Japan. Summary of the
act regarding infectious disease prevention and medical care for
the patients. Tokyo, Japan: Ministry of Health, Labor and
Welfare, 1999. http://www.mhlw.go.jp/english/wp/wp-hw3/dl/
2-083.pdf Accessed July 2015.
5 Omori M. [The present status of diagnosis and treatment of
tuberculosis patients in Japan]. Tokyo, Japan: The Research
Institute of Tuberculosis, Japan Anti-tuberculosis Association,
1998. www.jata.or.jp/rit/rj/07kinkyu.html Accessed July 2015.
[Japanese]
6 Lonnroth
K, Williams B G, Cegielski P, Dye C. A consistent log
linear relationship between tuberculosis incidence and body
mass index. Int J Epidemiol 2009; 39: 149155.

Nutrition and anti-tuberculosis treatment outcomes

7 Dye C, Bourdin Trunz B, Lonnroth


K, Roglic G, Williams B G.

Nutrition, diabetes and tuberculosis in the epidemiological


transition. PLOS ONE 2011; 6: e21161.
8 Hayashi S, Takeuchi M, Hatsuda K, et al. The impact of
nutrition and glucose intolerance on the development of
tuberculosis in Japan. Int J Tuberc Lung Dis 2014; 18: 8489.
9 Pednekar M S, Hakama M, Hebert J R, Gupta P C. Association
of body mass index with all-cause and cause-specific mortality:
findings from a prospective cohort study in Mumbai (Bombay),
India. Int J Epidemiol 2008; 37: 524535.
10 Akhtar M, Bretzel G, Boulahbal F. Technical guide. Sputum
examination for tuberculosis by direct microscopy in lowincome countries. In: Management of tuberculosis: a guide for
low-income countries. Paris, France: International Union
Against Tuberculosis and Lung Disease, 2000: pp 1415.
11 Taylor Z, Nolan C M, Blumberg H M. Controlling tuberculosis
in the United States. Recommendations from the American
Thoracic Society, CDC, and the Infectious Diseases Society of
America. MMWR Recomm Rep. 2005; 54: 181.
12 The International Expert Committee. International Expert
Committee report on the role of the A1C assay in the diagnosis
of diabetes. Diabetes Care 2009; 32: 13271334.
13 Ministry of Health Law, Japan. [Dietary reference intakes in
Japan]. Tokyo, Japan: Ministry of Health Law, 2005. http://
www.mhlw.go.jp/houdou/2004/11/h1122-2.html. Accessed
July 2015. [Japanese].
14 World Health Organization Expert Committee. WHO
Technical Report Series 854. Physical status: the use and
interpretation of anthropometry. Geneva, Switzerland: WHO,
1995: pp 362364.
15 Hernandez-Pando R, Orozco H, Aguilar D. Factors that
deregulate the protective immune response in tuberculosis.
Arch Immunol Ther Exp 2009; 57: 355367.
16 Weinsier R L, Hunker E M, Krumdieck C L, Butterworth C E,
Jr. Hospital malnutrition. A prospective evaluation of general
medical patients during the course of hospitalization. Am J Clin
Nutr 1979; 32: 418426.
17 Kim D K, Kim H J, Kwon S Y, et al. Nutritional deficit as a
negative prognostic factor in patients with miliary tuberculosis.
Eur Respir J 2008; 32: 10311036.

1375

18 Kim H J, Lee C H, Shin S, et al. The impact of nutritional deficit


on mortality of in-patients with pulmonary tuberculosis. Int J
Tuberc Lung Dis 2010; 14: 7985.
19 Dooley K E, Tang T, Golub J E, Dorman S E, Cronin W. Impact
of diabetes mellitus on treatment outcomes of patients with
active tuberculosis. Am J Trop Med Hyg 2009; 80: 634639.
20 Isanaka S, Mugusi F, Urassa W, et al. Iron deficiency and
anemia predict mortality in patients with tuberculosis. J Nutr
2012; 142: 350357.
21 Riley R L, Mills C C, Nyka W, et al. Aerial dissemination of
pulmonary tuberculosis. A two-year study of contagion in a
tuberculosis ward, 1959. Am J Epidemiol 1995; 142: 314.
22 Alisjahbana B, Sahiratmadja E, Nelwan E J, et al. The effect of
type 2 diabetes mellitus on the presentation and treatment
response of pulmonary tuberculosis. Clin Infect Dis 2007; 45:
428435.
23 Guler M, Unsal E, Dursun B, Aydln O, Capan N. Factors
influencing sputum smear and culture conversion time among
patients with new case pulmonary tuberculosis. Int J Clin Pract
2007; 61: 231235.
24 Baker M A, Harries A D, Jeon C Y, et al. The impact of diabetes
on tuberculosis treatment outcomes: a systematic review. BMC
Med 2011; 9: 81.
25 Akazawa Y. Prevalence and incidence of diabetes mellitus by
WHO criteria. Diabetes Res Clin Pract 1994; 24 (Suppl): S23
S27.
26 Ito C. Trends in the prevalence of diabetes mellitus among
Hiroshima atomic bomb survivors. Diabetes Res Clin Pract
1994; 24 (Suppl): S29S35.
27 Kuzuya T. Prevalence of diabetes mellitus in Japan compiled
from literature. Diabetes Res Clin Pract 1994; 24 (Suppl): S15
S21.
28 Islam M M, Horibe H, Kobayashi F. Current trend in
prevalence of diabetes mellitus in Japan, 19641992. J
Epidemiol 1999; 9: 155162.
29 Ministry of Health, Labor and Welfare, Japan. National Health
and Nutrition Survey. Tokyo, Japan: Ministry of Health, Labor
and Welfare, 2012. http: //www.mhlw.go.jp/bunya/kenkou/
kenkou_eiyou_chousa.html Accessed August 2015.

Nutrition and anti-tuberculosis treatment outcomes

RESUME
C O N T E X T E : Les r e sultats du traitement de la
tuberculose (TB) varient en fonction de plusieurs
facteurs.
O B J E C T I F : Identifier retrospectivement les risques lies
au delai de negativation de la culture de crachats (Tn) et
determiner leur signification clinique.
S C H E M A : Les patients dont les bacilles sont sensibles a`
lisoniazide et a` la rifampicine et qui ont initialement
recu un traitement standard sans interruption ont e te
recrutes. Un total de 630 patients consecutifs ont e te
inclus dans lanalyse de de veloppement du risque
(cohorte de developpent) et 611 patients consecutifs
dans lanalyse de validation du risque (cohorte de
validation).
R E S U LT A T S : Lanalyse univariee a montre que Tn e tait
lie au sexe, a` lindice de masse corporelle (BMI), a` la

numeration des leucocytes (WBC), au niveau de serum


albumine (Alb), a` la glycemie a` jeun, a` lhemoglobine
A1c, a` la proteine C-reactive, au cholesterol total et a` la
positivit e du frottis de crachats (SSP). Lanalyse
multivariee a montre que le BMI, les WBC et le frottis
de crachats positif e taient des risques significatifs
dextension du Tn. Les seuils optimaux du BMI et des
WBC permettant de predire quels patients seraient bon
repondeurs (Tn , 46 jours) et mauvais repondeurs (Tn
7 46 jours) selon chaque risque ont e te determines par
analyse de la fonction defficacite du recepteur. Les
risques ont e te verifies avec la cohorte de validation. Le
Tn augmentait en fonction du nombre de risques et le Tn
median des patients ayant trois risques e tait plus long de
21 jours que celui des patients sans risque.
C O N C L U S I O N : Letat nutritionnel peut predire le Tn.
RESUMEN

M A R C O D E R E F E R E N C I A: El desenlace del tratamiento


antituberculoso es variable, en funcion
de factores que
dependen del paciente.
O B J E T I V O: Determinar de manera retrospectiva los
factores de riesgo asociados con un lapso ma s
prolongado hasta la conversion
del cultivo de esputo
clnica.
(Tn) y definir su significacion
M E T O D O: Se incluyeron en el estudio pacientes con
tuberculosis que albergaban bacilos sensibles a
isoniazida y rifampicina, que haban recibido un
tratamiento inicial corriente sin interrupcion.

En el
analisis de determinacion
de riesgos se incluyeron 630
pacientes consecutivos (cohorte de determinacion)
y en
analisis de validacion
del riesgo se incluyeron otros 611
pacientes consecutivos (cohorte de validacion).

R E S U LT A D O S: El analisis monofactorial revelo


que el
Tn del cultivo de esputo se relaciono con el sexo, el ndice
de masa corporal (BMI), el recuento de leucocitos
(WBC), la concentraci on
s e rica de albumina,

la
glucemia en ayunas, la concentracion
de hemoglobina,

hemoglobina A1c, protena C reactiva (CRP), colesterol


total y la positividad de la baciloscopia del esputo. El
analisis multifactorial puso en evidencia que el BMI, el
WBC y la positividad de la baciloscopia del esputo (SSP)
eran factores de riesgo que se asociaban de manera
significativa con una prolongacion
del Tn del cultivo de
esputo. Mediante un analisis de eficacia diagnostica

se
definieron los umbrales optimos

del BMI y del WBC que


se deben usar segun
cada riesgo, a fin de pronosticar los
pacientes que responderan bien (Tn , 46 das) y los que
responderan mal (Tn 7 46 das). Los riesgos se
verificaron con la cohorte de validacion.

El Tn del
cultivo de esputo aumento en funcion
del numero

de
factores de riesgo y la mediana del Tn en los pacientes
con tres factores de riesgo fue 21 das superior a la
mediana de los pacientes que no presentaban ningun

riesgo.
N: El
CONCLUSIO
estado nutricional permite
pronosticar el Tn del cultivo de esputo.