Effect of micronutrient supplementation on treatment outcomes in children with intrathoracic tuberculosis: a randomized controlled trial1–4 Rakesh Lodha, Aparna Mukherjee, Varinder Singh, Sarman Singh, Henrik Friis, Daniel Faurholt-Jepsen, Shinjini Bhatnagar, Savita Saini, Sushil K Kabra, Harleen MS Grewal, and the Delhi Pediatric TB Study Group

INTRODUCTION

Micronutrient deficiencies impair the immune system with subsequent higher risk of infections as well as poorer prognosis (1, 2). Micronutrient deficiencies are widespread in the developing world and are thought to contribute to the high burden and poor outcomes of many infectious diseases including tuberculosis (3–6). Globally, 1.5 million new cases and 130,000 deaths as a result of tuberculosis are reported annually (7) in children, which make tuberculosis an important cause of childhood morbidity and mortality.

The nutritional rehabilitation of tuberculosis patients may enhance the proliferation of T-lymphocyte subpopulations in response to specific antigens and influence key cytokines involved in the formation of organized granulomas and macrophage activation (8). Micronutrients, especially zinc, are crucial components of both intracellular and intercellular signaling systems of most immunocompetent cells with subsets of T lymphocytes and monocytes being most affected in zinc deficiency. In addition, the macrophage, which is a pivotal cell in the host defense against tuberculosis, is adversely affected by zinc deficiency (9). A better understanding of the relations between isolated or combined micronutrient deficiencies and tuberculosis can lead to interventions to reduce the burden and improve outcomes. Although many studies have documented a high prevalence of vitamin D deficiency in patients with tuberculosis (10–14), there have been few studies with contrary results (15), and vitamin D supplementation trials have not shown consistent improvements in outcomes (16–20). Vitamin A and zinc supplementation showed improved sputum conversion in adult Indonesian sputum positive tuberculosis patients (21). Although multimicronutrient supplements with zinc have shown improved outcomes, supplements without zinc did not (22). Although micronutrient deficiencies may influence the immune system in both adults and children, the pertinent issues of growth and development are unique to the pediatric population. Because of the high growth rate and developing immune system, 1

From the Department of Pediatrics (RL, AM, SB, S Saini, and SKK) and Division of Clinical Microbiology & Molecular Medicine, Laboratory Medicine (S Singh), All India Institute of Medical Sciences, New Delhi, India; the Department of Pediatrics, Kalawati Saran Children Hospital, New Delhi, India (VS); the Department of Nutrition, Exercise and Sports, Faculty of Science, University of Copenhagen, Copenhagen, Denmark (HF and DF-J); and the Department of Clinical Science, Infection, University of Bergen, Bergen, Norway (HMSG). 2 RL and AM contributed equally to the study, and SKK and HMSG contributed equally to the study. 3 Supported by the Norwegian Programme for Development, Research and Education (NUFUPRO-2007/10183) and the Research Council of Norway Global Health and Vaccination Research (GLOBVAC). 4 Address correspondence to SK Kabra, Department of Pediatrics, All India Institute of Medical Sciences, New Delhi 110029, India. E-mail: skkabra@ hotmail.com; or HMS Grewal, Department of Clinical Science, Infection, University of Bergen, Bergen, Norway. E-mail: [email protected]. Received December 19, 2013. Accepted for publication August 22, 2014. First published online September 10, 2014; doi: 10.3945/ajcn.113.082255.

Am J Clin Nutr 2014;100:1287–97. Printed in USA. Ó 2014 American Society for Nutrition

1287

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

ABSTRACT Background: Micronutrients play an important role in immune function. To our knowledge, there have been no comprehensive studies on the role of micronutrient supplementation in children with tuberculosis. Objective: We assessed the effect of micronutrient supplementation in children treated with antituberculosis therapy (ATT). Design: A randomized, double-blind, placebo-controlled trial that used a 2 3 2 factorial design was undertaken at 2 teaching hospitals in Delhi. Children with newly diagnosed intrathoracic tuberculosis were enrolled, and they received ATT together with daily supplementation for 6 mo with either zinc alone, micronutrients without zinc, micronutrients in combination with zinc, or a placebo. Main outcomes were weight gain and an improvement in a chest X-ray (CXR) lesion assessed at 6 mo of treatment. Results: A total of 403 children were enrolled and randomly assigned. A microbiological diagnosis of tuberculosis was confirmed in 179 children (44.4%). The median (95% CI) increase in weight-for-age z score at 6 mo was not significantly different between subjects who received micronutrients [0.75 (0.66, 0.84)] and those who did not receive micronutrients [0.76 (0.67, 0.85)] and between subjects who received zinc [0.76 (0.68, 0.85)] and those who did not receive zinc [0.75 (0.66, 0.83)]. An improvement in CXR was observed in 285 children, but there was no difference between those receiving zinc and no zinc or between those receiving micronutrients and no micronutrients after 6 mo of ATT. However, children who received micronutrients had a faster gain in height over 6 mo than did those who did not receive micronutrients (height-for-age z score D = 0.08; P = 0.014). Conclusions: Micronutrient supplementation did not modify the weight gain or clearance of lesions on CXR in children with intrathoracic tuberculosis. However, micronutrient supplementation during treatment may improve height gain in children with intrathoracic tuberculosis. This trial was registered at clinicaltrials.gov as NCT00801606. Am J Clin Nutr 2014;100:1287–97.

1288

LODHA ET AL

SUBJECTS AND METHODS

This randomized, double-blind, placebo-controlled trial, with the use of a 2 3 2 factorial design, was carried out from January 2008 to June 2012 at the All India Institute of Medical Sciences and Kalawati Saran Children Hospital associated with Lady Hardinge Medical College in Delhi, India. The study protocol was approved by the respective institutional ethics committees. This trial was registered at clinicaltrials.gov as NCT00801606.

4) signs of an upper-airway obstruction; 5) oxygen saturation ,92%; 6) signs of renal, hepatic, or cardiovascular disease; 7) inability to attend follow-up session for reading of the TST; 8) documented intake of zinc continuously .2 wk in the 4 wk preceding enrollment; 9) central nervous system, osteoarticular, pericardial, or renal tuberculosis; 10) history of contact with a documented case of drug-resistant tuberculosis; or 11) were nonresidents of Delhi. Children were not routinely tested for HIV infection. However, stored serum samples of children enrolled in the randomized controlled trial were anonymously tested for HIV, and the prevalence was shown to be 1%. Written informed consent was obtained from guardians of children. For children .7 y of age, written assent was also taken. Antituberculosis therapy regimen All patients received the drug regimen as per the recommendation of the Revised National Tuberculosis Control Program of India; this regimen had a 2-mo intensive phase that used 3 or 4 drugs (isoniazid, rifampicin, pyrazinamide and ethambutol) followed by 4 mo of isoniazid and rifampicin (25, 26). Category 1 treatment was the use of 4 drugs (isoniazid, rifampicin, pyrazinamide, and ethambutol) in the intensive phase followed by 4 mo of 2 drugs (isoniazid and rifampicin), whereas category 3 treatment was the use of 3 drugs (isoniazid, rifampicin, and pyrazinamide) in the intensive phase followed by 4 mo of maintenance phase of 2 drugs. Antituberculosis therapy (ATT) was administered in daily doses on the basis of weight (ie, 5–7 mg isoniazid/kg, 10–13 mg rifampicin/kg, 35–40 mg pyrazinamide/kg, and 20–25 mg ethambutol/kg.

Patient eligibility and assessment Children from 6 mo to 15 y of age who presented with any of the following symptoms were considered tuberculosis suspects and screened for tuberculosis (24): 1) cough .2 wk with no improvement after a 7–10-d course of amoxicillin; 2) fever .2 wk with no improvement during a 7–10-d course of amoxicillin; 3) recent unexplained weight loss or failure to thrive; 4) unusual or unexplained fatigue (reduced playfulness) or lethargy; or 5) subtle clinical symptoms and history of close contact with adult patient with tuberculosis. All tuberculosis suspects underwent a CXR (lateral and posterior-anterior) and a tuberculin skin test (TST). If the CXR, as read by the site investigator, was consistent with intrathoracic tuberculosis (ie, hilar or mediastinal adenopathy, consolidation, cavity, miliary shadows, or pleural effusion), the child was classified as having probable intrathoracic tuberculosis. This diagnostic algorithm was similar to what is recommended by the Revised National Tuberculosis Control Program for childhood tuberculosis (25). Children with newly diagnosed probable intrathoracic tuberculosis, with or without an extrapulmonary lesion, were considered for enrollment. Children were excluded if they had any of the following: 1) bilateral pedal edema; 2) known HIV infection; 3) history of antituberculosis treatment or isoniazid prophylaxis .4 wk; 5 Abbreviations used: AFB, acid fast bacillus; ATT, antituberculosis therapy; CXR, chest X-ray; DSMB, data safety and monitoring board; DST, drug-sensitivity testing; HAZ, height-for-age z score; MGIT, mycobacterial growth indicator tube; MTB, Mycobacterium tuberculosis; MUAC, midupper arm circumference; NCHS, National Center for Health Statistics; TST, tuberculin skin test; WAZ, weight-for-age z score.

Random assignment, allocation concealment, and intervention All patients received the standard of care including ATT. In addition, for 6 mo, participants received a daily dose of 5 mL syrup preparation that contained 1 of 4 regimens as follows: 1) zinc (20 mg elemental zinc) only; 2) micronutrients (vitamin A, thiamine, riboflavin, vitamins B-6 and B-12, folic acid, niacin, vitamins C, E, and D, selenium, and copper) without zinc; 3) micronutrients in combination with zinc (vitamin A, thiamine, riboflavin, vitamin B-6 and B-12, folic acid, niacin, vitamins C, E, and D, selenium, copper, and 20 mg elemental zinc); or 4) placebo. Micronutrient preparations and the placebo were purchased from Messers Inda Medica. Doses of the micronutrients other than zinc were based on age (ie, 6 mo to 3 y, 4–6 y, 7–9 y, and 10–15 y) (Table 1). Doses given correspond to 1–2 times the RDA according to age (27), except for zinc, for which 20 mg was given to all children because of the high requirements during catch-up growth. Children were randomly assigned to 1 of 4 interventions. A scientist, who was not involved in the data collection and analysis, generated random-allocation sequences in permuted blocks of variable sizes separately for the 2 sites. Separate sequences were generated for the age groups 6 mo to 3 y, 4–6 y, 7–9 y, and 10–15 y. Bottles that contained the micronutrient supplements were serially numbered for each stratum for the 2 sites. The 4 study syrup preparations had a similar packaging, appearance, and smell. The patient, physician, and laboratory personnel were blinded to the intervention. We maintained the masking during the data

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

micronutrient deficiencies are likely to be more widespread in children and as a consequence supplementation is important. Little is known about the effect of micronutrient supplementation on childhood tuberculosis outcomes; a study in children with tuberculosis from Tanzania reported no effect on weight gain after 8 wk of multivitamin supplementation (23). We carried out a randomized, double-blind, placebo-controlled trial in children with newly diagnosed intrathoracic tuberculosis in a high tuberculosis burden setting. The aim of the study was to assess the effect of micronutrient supplementation on treatment outcomes in children with intrathoracic tuberculosis. Primary outcomes were the change in weight-for-age z score (WAZ)5 at 6 mo and resolution of pulmonary lesions as assessed by using a chest X-ray (CXR) at 6 mo. Secondary outcomes were 1) heightfor-age z score (HAZ), midupper arm circumference, triceps skinfold thickness, and BMI z score at 2 and 6 mo; 2) a resolution of symptoms as reported by parents at 2 mo; 3) the proportion of children requiring extension of intensive phase of therapy at 2 mo; and 4) the improvement in the CXR at 2 mo.

1289

MICRONUTRIENTS IN CHILDHOOD TUBERCULOSIS TABLE 1 Composition of micronutrient supplements (study syrup) according to age group1 Vitamin Age

Mineral

Vitamin A Thiamin Riboflavin Vitamin B-6 Vitamin B-12 Folic acid Niacin Vitamin C Vitamin D Vitamin E Selenium Copper

6 mo to 3 y 4–6 y 7–9 y 10–15 y 1

mg RE 800 900 1000 1200

mg 1 1.2 1.8 2.4

mg 1 1.2 1.8 2.6

mg 1 1.2 2 2.6

mg 3.8 2.4 3.6 4.8

mg DFE 300 400 600 800

mg NE 12 16 24 32

mg 60 60 70 80

mg 10 10 10 10

mg 10 10 14 20

mg 10 10 10 10

mg 2 3 4 5

DFE, dietary folate equivalent; NE, niacin equivalent; RE, retinol equivalent.

Assessment of adherence Adherence to medications and supplementation was assessed by providing a diary to the guardian who was asked to maintain a daily record that indicated the administration of the study intervention and ATT. In addition, adherence was assessed at each visit by counting remaining pills (ATT) and measuring the volume of the remaining study syrup. Patients’ guardians were reminded of the visit to the clinic 1 d in advance by telephone. If a patient did not attend the clinic on the assigned date, the parent or guardian was contacted by telephone. If telephone contact was not feasible, a fieldworker visited the child’s home. Clinical and laboratory evaluation All patients enrolled in the study were assessed for clinical symptoms, history of tuberculosis contact, history of Bacillus Calmette–Gue´rin vaccination, demographic information, detailed anthropometric measurements, and physical examination. Patients were followed every 2 wk until 2 mo of treatment and, thereafter, every 4 wk until they completed the assigned ATT. On each follow-up visit, a detailed clinical history and examination including anthropometric measurements were recorded. Dietary history was recorded by the 24-h dietary recall method. The nutritive value of the diet was calculated by using Dietsoft software (Invincible IDeAS). Anthropometric measures Child’s weight was measured with minimal clothing to the nearest 0.1 kg by using an electronic scale. Children ,2 y old were weighed by using an infant weighing scale, and their lengths were measured by using an infantometer to the nearest 0.1 cm. In children able to stand and .2 y of age, height was measured by using a stadiometer to the nearest 0.1 cm. The midupper arm circumference was measured to the nearest 0.1 mm by using nonstretchable tape. Triceps skinfold thickness was measured by using a Harpenden skinfold-thickness caliper (Baty British Indicators) to the nearest 0.1 mm. All the measurements were taken on each visit by the same observer as far as possible to avoid interobserver variation. Periodic quality-control exercises were carried out at both centers by senior study investigators. The WAZ, HAZ, and weight-for-height z score were calculated by using the nutritional anthropometry module of Epi Info version 5 software (CDC) on the basis of the National Center for Health Statistics–WHO reference curves 2000 (28).

Laboratory workup Gastric lavage and sputum induction on 2 consecutive days were performed on an ambulatory basis [baseline and at 2 mo if the baseline sample was positive for acid fast bacillus (AFB) by smear or mycobacterial growth indicator tube (MGIT) culture and at 6 mo if the AFB smear or MGIT culture was positive at 2 mo). Samples were processed for smear examination and culture for Mycobacterium tuberculosis (MTB) as reported earlier (29). For the purpose of study, the drug-sensitivity testing (DST) was done on isolates that were stored, and therefore, the same could be done in only a proportion of children. However, in specific scenarios of nonresponse or worsening, DST was requested for and obtained earlier. A QuantiFERON Gold in-Tube test (Cellestis) was performed at baseline and 2 and 6 mo as per the manufacturer’s guidelines (30). Serum zinc and copper (baseline and 2 and 6 mo) were determined by using a flame furnace atomic absorption spectrophotometer (GBC Avanta) by using standard techniques and with Seronorm Trace Elements Serum (Sero AS). Limits of detection were 10–160 mg/dL for serum zinc and 40–320 mg/dL for copper. For outliers above limits of detection, the sample was further diluted and reassayed, and the concentration was calculated accordingly. For quality control, pooled serum with a known value was used. Interassay variability remained ,10%. Serum zinc concentrations ,65 mg/dL were considered low (31). CXR was performed at baseline and 2 and 6 mo of enrollment. All CXRs were read initially by site investigators who undertook patient enrollment into the study. CXRs shown to be of poor quality were repeated. The reading of CXRs for the study purpose was done subsequently by 2 pediatricians (VS, SKK, or RL) who were blinded for the clinical diagnosis of patient, and a standardized format was used for reporting. If findings of the 2 pediatricians differed, the third person was asked to review the CXR and final findings that matched for 2 of the pediatricians were recorded for the purpose of analysis. CXRs taken at 2 and 6 mo were classified as follows: significant improvement (more than twothird clearance from baseline), some improvement (improvement but less than two-third clearance from baseline), no improvement (no change in lesion size), and worsening (increase in the lesion size from baseline) (32). A TST was performed by using 5 tuberculin units purified protein derivative (Span Diagnostics), which was administered intradermally by a trained study nurse on the flexor aspect of the left forearm. Induration (transverse diameter) was measured between 48–72 h after the test was applied. Any induration $10 mm was considered as positive. Wherever indicated, additional laboratory investigations including fine-needle aspiration cytology, biopsy of lymph node,

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

analysis by coding the treatment allocation with 4 letters. All patients were provided with ATT and the study supplement in sufficient quantity until the next visit with 5 additional doses.

1290

LODHA ET AL

and a pleural fluid examination were performed, and culture on the MGIT was done. Trial deviates A participant was registered as a trial deviate for the following reasons: 1) if the duration of interruption of ATT was .2 wk in the intensive phase and .4 wk in the maintenance phase; 2) the patient received a zinc preparation .2 wk during enrollment in the study; or 3) the patient had an HIV infection diagnosed after the start of study. Change of therapy

Quality assurance and monitoring of adverse events Before study initiation, study procedures were standardized, and standardized operating procedure manuals were prepared. All study staff were trained and monitored periodically at both sites. A data safety and monitoring board (DSMB) was constituted, and DSMB members met before the study initiation and every 3 mo after the study start. All adverse events were communicated to the DSMB. Good Clinical Practice guidelines for the monitoring of clinical trials were adhered to. Sample-size calculation There are limited data on the radiological response to ATT in children with pulmonary tuberculosis. On the basis of clinical experience from the 2 participating centers, we assumed that, after 6 mo of ATT, 60% of children in the placebo arm would have a significant resolution of CXR tuberculosis lesion. We anticipated that, in the intervention arm, 80% of children would have a significant resolution of tuberculosis lesion on the CXR, which would be a clinically significant effect. With a power of 80% and a error of 0.05, a sample size of 91 children/group was required to detect an increase in the proportion with resolution $80% in micronutrient- or zinc-supplemented children. With the allowance for 10% attrition, we required a sample size of 100 children/group (ie, a total trial size of 400 children). This sample size would enable us to detect a 10% change in WAZs between groups. Data management and statistical analysis Site supervisors checked all completed case-report forms before they were sent for interactive double data entry by using MS Access software (version 2007; Microsoft Corp) with built-in logic, consistency, and range checks. Data were analyzed according to

RESULTS

Of 1572 children screened, 52.2% of subjects (820 children) were diagnosed as having probable intrathoracic tuberculosis. A total of 417 children were excluded (Figure 1), and 49.1% of subjects (403 of 820 children) were enrolled in the study and randomly assigned to 1 of 4 interventions (Figure 1). Data on any 1 of 2 primary outcomes were available at 2 mo for 97.5% (393 children) and at 6 mo for 94.5% (381 children) of 403 subjects. For most demographic and clinical profiles, intervention arms were comparable at baseline; however, some differences were identified with regard to sex distribution, parental educational level, diagnosis, and energy and zinc intakes before enrollment (Table 2). Nearly one-half of children (54.3%) had serum zinc concentrations ,65 mg/dL. The median (IQR) monocyte: lymphocyte ratio at baseline was 0.054 (0–0.29) and was significantly different in 3 types of disease categories as classified by the CXR (P = 0.003), with the lowest in progressive disease. DST was carried out in 63.9% of children (94 of 147 subjects) at baseline. The mean (6SD) time to positivity for MGIT culture

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

ATT was changed to category 2 in the case of a nonresponse either clinical or radiological. Category 2 ATT includes 2 mo of streptomycin, isoniazid, rifampicin, pyrazinamide, and ethambutol followed by 1 mo of isoniazid, rifampicin, pyrazinamide, and ethambutol and 5 mo of isoniazid, rifampicin, and ethambutol (26). Any child who did not improve with category 2, and for whom an alternative diagnosis had been ruled out, was started on second-line ATT, which included an injection of kanamycin, ofloxacin, and ethionamide and an additional drug (cycloserine or coamoxyclav). Any child who completed the total duration of therapy, irrespective of the regimen administered, was considered as having completed therapy.

the intention-to-treat principle, which, because of few missing data at follow-up, resulted in available case analyses. All analyses were done with Stata 11.0 software (StataCorp). Treatment completion, change of therapy, lost to follow-up, deaths, and withdrawals by intervention arm were assessed by using either chi-square tests or Fisher’s exact tests when appropriate. All children who completed their ATT, irrespective of the regimen they were assigned to, were considered under the treatment-completion category. Main-effect analyses were based on a 2 3 2 factorial design (33), and because of repeated measures, we used linear mixedeffects models for continuous outcomes (WAZ, HAZ, triceps skinfold thickness, midupper arm circumference, and BMI z score) with study identification numbers included as random effects. Primarily, effects of the interventions (micronutrients or zinc) were evaluated by using unadjusted analyses with the follow-up visit variable as a fixed effect. Second, adjusted models were obtained by additionally including a number of fixed effects (age, sex, mother’s education, father’s education, diagnosis, and energy and zinc intakes) and random effects (study site). We assessed interactions between zinc and micronutrient groups in the adjusted analysis. Results from adjusted models were only shown when estimates and interpretations of results changed. In addition, the CXR variable at follow-up was considered to be recorded on an ordinal scale (worsening, no change, some improvement, and significant clearance). Thus, with respect to zinc and micronutrients, the intervention data analysis was based on Wilcoxon’s rank-sum test at 2 and 6 mo of follow-up, respectively. Serum zinc and copper were measured at baseline and follow-up with changes over time as well as differences between micronutrients compared with no micronutrients and zinc compared with no zinc (2 3 2 factorial design) assessed by using linear mixed-effects models with identification numbers as random effects. In unadjusted analyses, the intervention (micronutrients or zinc) and follow-up visit were added as fixed effects. Adjusted analyses also included the following fixed-effects variables: age, sex, mother’s education, father’s education, diagnosis, and energy and zinc intakes as well as random effects for study sites.

MICRONUTRIENTS IN CHILDHOOD TUBERCULOSIS

was 24 6 5 d, which was not significantly different for the 4 intervention arms. Adherence to the study supplement was high (96.3%) in all study groups as was adherence to ATT (95%) (Table 3). Sixty children (14.9%) required an extension of the 4-drug intensive phase of treatment because of a nonresolution of symptoms or persistence of AFB in the smear from gastric lavage or induced sputum. The treatment regimen had to be changed to category 2 or second-line ATT in 24 children (5.9%). The distribution of

1291

these patients between the 4 arms of the study was not significantly different. Drug sensitivity could be done in only 94 of 147 culture-positive children (63.9%). Five children (5.3%) had multidrug-resistant tuberculosis; 4 children had received a category 2 treatment, whereas one child was started directly on second-line treatment. Of 152 children who tested positive for AFB in the smear or MTB in MGIT culture at baseline, AFB was positive in the smear from gastric lavage and/or induced sputum at the end of 2 mo of ATT in 13 children (8.5%), whereas MTB

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014 FIGURE 1. Flow of study. 1A participant was registered as a trial deviate for the following reasons: 1) the duration of interruption of ATT was .2 wk in the intensive phase and .4 wk in the maintenance phase; 2) the participant was receiving a zinc preparation for .2 wk during enrollment in the study; or 3) the participant had HIV infection diagnosed after the start of the study. ATT, antituberculosis therapy.

1292

LODHA ET AL

TABLE 2 Baseline characteristics in 403 children with intrathoracic tuberculosis randomly assigned to micronutrient and zinc supplementation1 Micronutrient + zinc (n = 102) 105.2 12 21 23 46 42 77 33

6 45.32 (11.76) (20.59) (22.55) (45.10) (41.2) (75.5) (32.3)

105.09 11 17 23 49 46 75 39

6 46.3 (11.0) (17.0) (23.0) (49.0) (46.0) (75.0) (39.0)

Zinc (n = 101) 103.05 10 19 21 51 50 76 37

6 43.01 (9.90) (18.81) (20.79) (50.50) (49.5) (75.2) (36.6)

Placebo (n = 100) 105.88 12 15 25 48 37 72 41

6 45.7 (12.0) (15.0) (25.0) (48.0) (37.0) (72.0) (41.0)

14 (13.7) 71 (69.6) 17 (16.7)

18 (18.0) 74 (74.0) 8 (8.0)

17 (16.8) 69 (68.3) 15 (14.8)

24 (24.0) 63 (63.0) 13 (13.0)

5 (4.9) 73 (71.6) 24 (23.5)

6 (6.0) 74 (74.0) 20 (20.0)

10 (9.9) 59 (58.4) 32 (31.7)

6 (6.0) 71 (73.0) 23 (21.0)

11 91 95 75/89

(10.8) (89.2) (93.1) (84.3)

11 89 91 69/90

26 (25.5) 67 (65.7) 9 (8.8) 87 15 28 36 36 23.9 1/18 38.2 1064.0 2.98 60 0.05

(11.0) (89.0) (91.0) (76.7)

31 (31.0) 55 (55.0) 14 (14.0)

(85.3) (14.7) (27.4) (35.3) (35.3) 6 4.6 (5.6) 6 21.3 6 409.0 (2.38–4.61)4 (58.8) (0–0.36)

77 23 28 34 34 22.6 1/29 34.1 1004.1 2.97 51 0

(77.0) (23.0) (28.0) (34.0) (34.0) 6 4.5 (3.4) 6 14.3 6 430.2 (1.66–3.74) (51.0) (0–0.30)

7 94 90 78/90

(6.9) (93.1) (89.1) (86.7)

36 (35.6) 49 (49.6) 16 (15.8) 78 23 23 38 35 24.0 1/21 32.1 901.9 2.35 53 0.08

(77.2) (22.8) (22.8) (37.6) (34.7) 6 5.9 (4.8) 6 16.4 6 456.9 (1.25–3.52) (52.5) (0–0.28)

10 90 95 75/93

(10.0) (90.0) (95.0) (80.6)

27 (27.0) 58 (58.0) 15 (15.0) 84 16 28 44 42 24.9 2/26 40.6 1075.6 3.73 55 0.01

(84.0) (16.0) (28.0) (44.0) (42.0) 6 5.8 (7.9) 6 18.4 6 383.2 (2.37–4.75) (55) (0–0.26)

1 Micronutrient + zinc: receiving both zinc and micronutrient supplementation. Micronutrient: receiving only micronutrient supplementation. Zinc: receiving only zinc supplementation. Placebo: receiving placebo. AFB, acid fast bacillus; BCG, bacille Calmette-Gue´rin; GL, gastric lavage; IS, induced sputum; MTB, Mycobacterium tuberculosis; QFT, Quantiferon Gold In-tube test (Cellestis); TST, tuberculin skin test. 2 Mean 6 SD (all such values). 3 Dietary intakes were estimated by using a 24-h dietary recall method. 4 Median; IQR in parentheses (all such values).

culture was positive in 41 children (26.9%). There was no difference in the distribution of children who did not undergo smear and culture conversion at the end of 2 mo of therapy between the 4 groups.

subjects receiving micronutrients (0.75; 95% CI: 0.66, 0.84) compared with no micronutrients (0.76; 95% CI: 0.67, 0.85) and zinc (0.76; 95% CI: 0.68, 0.85) compared with no zinc (0.75; 95% CI: 0.66, 0.83). Additional adjustment did not change results.

Primary outcomes

Improvement in CXR

Change in WAZ

A total of 285 children (74.8%) showed significant improvement on the CXR at 6 mo of treatment. There was no difference in CXR changes between subjects receiving zinc compared with no zinc or micronutrients compared with no micronutrients at 6 mo of ATT (Table 4).The agreement between 2 readers was 63.3% (k = 0.28) for CXRs done at 2 mo and 65% (k = 0.29) for CXRs done at 6 mo.

At 6 mo, the median weight gain was 4.0 kg (95% CI: 3.4, 4.7 kg), and the WAZ was 0.75 (95% CI: 0.69, 0.82). The change in the WAZ was assessed at 2 and 6 mo (Table 4). There was no interaction between micronutrients and zinc groups at both follow-up time points (P-interaction . 0.48). The median increase in WAZ at 6 mo was not significantly different between

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

Age (mo) 6 mo to 3 y [n (%)] 4–6 y [n (%)] 7–9 y [n (%)] 10–15 y [n (%)] Boys [n (%)] Received BCG [n (%)] History of contact present [n (%)] Mother’s education [n (%)] Illiterate Up to or less than class X More than class X Father’s education [n (%)] Illiterate Up to or less than class X More than class X Residence [n (%)] Urban Rural TST positivity [n (%)] QFT positivity [n/N tested (%)] Diagnosis [n (%)] Primary pulmonary complex Progressive pulmonary disease Pleural effusion Antituberculosis treatment category [n (%)] 1 3 Associated extrapulmonary tuberculosis [n (%)] AFB/MTB positive in GL and/or IS [n (%)] Culture positive in GL and/or IS [n (%)] Time to culture positivity (d) Multidrug resistance [n/N tested (%)] Protein intake (g)3 Energy intake (kcal)3 Zinc intake (mg)3 Children with serum zinc concentration ,65 mg/dL [n (%)] Monocyte:lymphocyte ratio

Micronutrient (n = 100)

1293

MICRONUTRIENTS IN CHILDHOOD TUBERCULOSIS TABLE 3 Treatment completion, change of therapy, lost to follow-up, and deaths in 403 children enrolled1 Micronutrient + zinc (n = 102)

Micronutrient (n = 100)

Zinc (n = 101)

97 (95.1) 3 (2.9) 0 2 (1.9) 2 (1.9) 1 (0.9)

97(97) 5 (5) 2 (2) 1(1) 2 (2) 0

98 (97.03) 8 (7.9) 0 2 (1.9) 1 (0.9) 0

2

Treatment completion [n (%)] Change of therapy to category 2 [n (%)] Change to second line ATT3 [n (%)] Lost to follow-up [n (%)] Deaths [n (%)] Withdrawal from study [n (%)]

Placebo (n = 100)

P

97 8 3 2

(97) (8) (3) (2) 0 1 (1)

0.90 0.34 0.08 1.00 0.70 0.87

Secondary outcomes Micronutrient supplementation increased the HAZ at 6 mo (difference: 0.08; 95% CI: 0.02, 0.14); full adjustment did not change the effect estimate (difference: 0.08; 95% CI: 20.02, 0.18). There were no effects of micronutrients supplementation on other anthropometric outcome, and no effects of zinc supplementation were observed on any of the anthropometric outcomes (Table 5). Additional adjustment did not change estimates. There were no interactions between micronutrient and zinc interventions.

At the end of the intensive phase of ATT (2 mo), 80% of children showed an improvement in cough, 94.6% of children showed an improvement in fever, 95.4% of children showed an improvement in appetite, and 94.2% of children, showed an improvement in lymphadenopathy. Except for subjects who were not receiving zinc, serum zinc increased over 6 mo of treatment (Table 6). Serum copper decreased over 6 mo in all treatment arms. Dietary intakes of energy and zinc were comparable in the 4 groups at 2 and 6 mo (data not shown).

TABLE 4 Effect of micronutrient supplementation on weight-for-age z score and resolution of lesions on a chest X-ray at 2 and 6 mo in children with intrathoracic tuberculosis randomly allocated to micronutrient or placebo and zinc or placebo supplementation1 Micronutrient supplementation No-micronutrient group (n = 201)

Micronutrient group (n = 202)

Zinc supplementation

P

No-zinc group (n = 200)

Zinc group (n = 203)

P

2

Weight-for-age z score Month 0 Month 2 Change Difference Month 6 Change Difference Chest X-ray [% (n)]3 Month 2 Worsening No change Some improvement Significant clearance Month 6 Worsening No change Some improvement Significant clearance

22.71 22.25 0.46 0.03 21.94 0.76 20.01

(22.92, 22.50) (22.46, 22.04) (0.37, 0.54) (20.09, 0.15) (22.16, 21.74) (0.67, 0.85) (20.13, 0.11)

22.79 (23.00, 22.57) 22.30 (22.51, 22.09) 0.49 (0.40, 0.57) 22.04 (22.25, 21.82) 0.75 (0.66, 0.84) —

— — — 0.63 — — 0.88

22.87 22.40 0.46 0.02 22.12 0.74 0.02

(23.08, 22.65) (22.62, 22.19) (0.38, 0.55) (20.10, 0.14) (22.33, 21.91) (0.66, 0.83) (20.10, 0.14)

22.63 (22.84, 22.42) 22.15 (22.36, 21.94) 0.48 (0.40, 0.57) 21.87 (22.08, 21.66) 0.76 (0.68, 0.85) —

0.81 3.1 27.6 36.7 32.7

(6) (54) (72) (64)

3.1 25.4 39.1 32.5

(6) (50) (77) (64)

1.0 3.7 20.4 74.9

(2) (7) (39) (143)

1.0 5.3 19.0 74.7

(2) (10) (36) (142)

— — — 0.71 — — 0.77 0.54

2.1 26.7 37.4 33.9

(4) (52) (73) (66)

4.0 26.3 38.4 31.3

(8) (52) (76) (62)

1.0 5.7 17.1 76.2

(2) (11) (33) (147)

1.1 3.2 22.3 73.4

(2) (6) (42) (138)

0.78

0.70

1 No-micronutrient group: groups receiving only zinc and placebo. Micronutrient group: group receiving micronutrient supplementation alone 1 group receiving micronutrient and zinc combined. No-zinc group: groups receiving micronutrient alone and placebo. Zinc group: group receiving zinc alone and group receiving micronutrient and zinc combined. 2 All values are bs; 95% CIs in parentheses. Values were determined by using an unadjusted linear mixed-effects model with follow-up time as a fixed effect and the participant as a random effect. Data were analyzed as intention to treat on the basis of the available case analysis. The interaction between micronutrient and zinc supplementation was tested in an unadjusted model (P-interaction = 0.68) and a model with additional fixed (age, sex, maternal and paternal educational level, baseline diagnosis, and energy and zinc intakes) and random (study site) effects (P-interaction = 0.48). 3 Differences were tested by using Wilcoxon’s rank-sum test at 2 and 6 mo, respectively.

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

1 Data are based on the chi-square or Fisher’s exact test. Micronutrient + zinc: receiving both zinc and micronutrient supplementation. Micronutrient: receiving only micronutrient supplementation. Zinc: receiving only zinc supplementation. Placebo: receiving placebo. 2 All children who completed their antituberculosis therapy, irrespective of the regimen they were assigned to, were considered under treatment completion. 3 ATT, antituberculosis therapy.

1294

LODHA ET AL

In 313 children, 690 episodes of adverse events were documented. Adverse events included upper respiratory tract infection, megaloblastic anemia, hepatitis, and vomiting and were unrelated to study interventions. There were 5 deaths (2 deaths each in the micronutrient-zinc and micronutrients groups, one death in the zinc group, and no deaths in the placebo group). None of the deaths were related to study interventions. The final diagnosis in these 5 cases were miliary tuberculosis with grade 3 protein-energy malnutrition with secondary infection, pulmonary tuberculosis with secondary sepsis, septic shock with intestinal perforation, pulmonary tuberculosis with septic shock, and pulmonary tuberculosis with super infection with rickets and respiratory failure.

We did not observe any effect of micronutrient or zinc supplementation on weight gain or resolution in CXR lesions in

TABLE 5 Height-for-age z score, triceps skinfold thickness, arm circumference, and BMI-for-age z score results at 0, 2, and 6 mo in 403 children infected with tuberculosis randomly allocated to micronutrient or placebo and zinc or placebo supplementation1 Micronutrient supplementation No-micronutrient group (n = 201) Height-for-age z score Month 0 Month 2 Change Difference Month 6 Change Difference Triceps skinfold thickness Month 0 Month 2 Change Difference Month 6 Change Difference Arm circumference Month 0 Month 2 Change Difference Month 6 Change Difference BMI-for-age z score Month 0 Month 2 Change Difference Month 6 Change Difference

Zinc supplementation

Micronutrient group (n = 202)

P

No-zinc group (n = 200)

21.63 (21.82, 21.44) 21.59 (21.78, 21.40) 0.04 (20.001, 0.08) — 21.54 (21.73, 21.35) 0.09 (0.05, 0.13) —

— — — 0.28 — — 0.01

21.70 21.66 0.03 20.02 21.63 0.07 20.04

(21.89, 21.51) (21.85, 21.47) (20.01, 0.08) (20.08, 0.04) (21.82, 21.44) (0.03, 0.11) (20.10, 0.02)

Zinc group (n = 203)

P

21.49 (21.68, 1.30) 21.47 (21.66, 21.29) 0.01 (20.03, 0.06) — 21.46 (21.65, 21.27) 0.03 (20.02, 0.07) —

— — — 0.46 — — 0.17

21.55 21.54 0.01 0.03 21.54 0.01 0.08

(21.74, 21.36) (21.73, 21.36) (20.03, 0.05) (20.03, 0.09) (21.73, 21.35) (20.03, 0.19) (0.02, 0.14)

5.15 5.54 0.38 20.19 6.50 1.35 20.31

(4.75, 5.56) (5.13, 5.95) (0.03, 0.74) (20.70, 0.31) (6.09, 6.92) (0.99, 1.71) (20.82, 0.20)

5.15 (4.75, 5.56) 5.35 (4.94, 5.76) 0.19 (20.17, 0.55) — 6.19 (5.78, 6.61) 1.04 (0.67, 1.40) —

— — — 0.46 — — 0.23

4.99 5.38 0.39 20.20 6.15 1.15 0.09

(4.59, 5.40) (4.97, 5.80) (0.03, 0.75) (20.71, 0.31) (5.73, 6.56) (0.79, 1.51) (20.43, 0.60)

5.31 (4.91, 5.72) 5.51 (5.10, 5.91) 0.19 (20.17, 0.55) — 6.56 (6.14, 6.97) 1.24 (0.87, 1.61) —

— — — 0.44 — — 0.74

15.71 16.51 0.79 20.08 17.33 1.61 20.10

(15.34, 16.08) (16.14, 16.88) (0.63, 0.96) (20.32, 0.15) (16.95, 17.70) (1.44, 1.78) (20.34, 0.14)

15.70 (15.33, 16.07) 16.41 (16.04, 16.78) 0.71 (0.55, 0.88) — 17.21 (16.84, 17.58) 1.51 (1.34, 1.68) —

— — — 0.49 — — 0.42

15.71 16.47 0.76 20.01 17.20 1.49 0.15

(15.34, 16.09) (16.10, 16.84) (0.59, 0.92) (20.24, 0.23) (16.83, 17.58) (1.32, 1.66) (20.09, 0.39)

15.70 (15.33, 16.07) 16.45 (16.08, 16.82) 0.75 (0.59, 0.92) — 17.34 (16.97, 17.71) 1.64 (1.47, 1.81) —

— — — 0.96 — — 0.22

22.40 21.63 0.78 0.06 21.17 1.23 20.08

(22.59, 22.21) (21.82, 21.43) (0.63, 0.92) (20.14, 0.26) (21.37, 20.98) (1.08, 1.37) (20.28, 0.12)

22.48 (22.67, 22.29) 21.65 (21.84, 21.45) 0.84 (0.70, 0.98) — 21.33 (21.53, 21.14) 1.15 (1.00, 1.29) —

— — — 0.54 — — 0.44

22.49 21.72 0.76 0.08 21.37 1.12 0.14

(22.68, 22.30) (21.92, 21.53) (0.62, 0.91) (20.12, 0.28) (21.57, 21.18) (0.97, 1.26) (20.06, 0.35)

22.40 (22.59, 22.21) 21.55 (21.74, 21.36) 0.84 (0.71, 0.99) — 21.14 (21.33, 20.94) 1.26 (1.12, 1.40) —

— — — 0.42 — — 0.17

1 All values are bs; 95% CIs in parentheses. No-micronutrient group: groups receiving only zinc and placebo. Micronutrient group: group receiving micronutrient supplementation alone 1 group receiving micronutrient and zinc combined. No-zinc group: groups receiving micronutrient alone and placebo. Zinc group: group receiving zinc alone and group receiving micronutrient and zinc combined. Values were determined by using unadjusted linear mixedeffects model with follow-up time as a fixed effect and the participant as a random effect. Data were analyzed as intention to treat on the basis of the available case analysis. The interaction between micronutrient and zinc supplementation was tested in an unadjusted model and a model with additional fixed (age, sex, maternal and paternal educational level, baseline diagnosis, and energy and zinc intakes) and random (study site) effects. In all instances, P-interaction . 0.70.

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

DISCUSSION

children with newly diagnosed intrathoracic tuberculosis treated with antituberculosis drugs. However, children who were receiving micronutrients supplements had greater linear growth. We did not observe any differences in the proportion of children who showed sputum conversion or treatment failures with regard to the intervention. In the past decade, there has been considerable interest in studying the effect of micronutrient supplementation on outcomes of pulmonary tuberculosis; most of these studies have been carried out in adult patients. To our knowledge, there has only been one recently published randomized controlled trial that compared multivitamin supplementation (vitamins B complex, C, and E) for 8 wk with placebo in childhood tuberculosis (23). No differences in the clearance of CXR, weight gain, or mortality rates were observed in 255 children who participated in the trial (23). The study was limited to children ,5 y of age, and only vitamins were supplemented for a shorter period of 8 wk (23). A small

1295

MICRONUTRIENTS IN CHILDHOOD TUBERCULOSIS

TABLE 6 Serum zinc and copper concentrations at 2 and 6 mo in children with intrathoracic tuberculosis randomly allocated to micronutrient or placebo and zinc or placebo supplementation1 Micronutrient supplementation No-micronutrient group (n = 201) 65.9 73.7 7.8 22.8 72.5 6.6 22.5 148.4 125.1 223.3 3.0 109.3 239.0 5.2

(62.1, 69.7) (69.8, 77.5) (2.9, 12.7) (29.7, 4.2) (68.6, 76.4) (1.7, 11.5) (29.5, 4.5) (143.9, 152.9) (120.5, 129.7) (228.7, 217.9) (24.8, 10.7) (104.7, 114.0) (244.5, 233.6) (22.5, 13.0)

67.1 (63.3, 70.8) 72.1 (68.2, 75.9) 5.0 (0.1, 9.9) — 71.2 (67.3, 75.1) 4.1 (20.8, 9.1) — 145.1 (140.5, 149.6) 124.7 (120.1, 129.3) 220.3 (225.8, 214.9) — 111.3 (106.6, 115.9) 233.8 (239.3, 228.2) —

P

No-zinc group (n = 200)

— — — 0.44 — — 0.49

66.4 64.6 21.9 16.6 63.0 23.5 18.0

— — — 0.45 — — 0.19

146.5 126.0 220.5 22.7 112.3 234.2 24.5

(62.7, 70.1) (60.8, 68.3) (26.7, 2.9) (9.8, 23.4) (59.2, 66.7) (28.3, 1.4) (11.2, 24.8) (141.9, 151.0) (121.4, 130.6) (225.9, 215.0) (210.5, 5.0) (107.6, 116.9) (239.7, 228.8) (212.3, 3.3)

Zinc group (n = 203) 66.5 (62.9, 70.2) 81.3 (77.5, 85.0) 14.7 (9.9, 19.5) — 81.1 (77.3, 84.9) 14.5 (9.7, 19.4) — 147.0 (142.5, 151.5) 123.7 (119.1, 128.4) 223.2 (228.7, 217.8) — 108.3 (103.6, 113.0) 238.7 (244.3, 233.2) —

P — — — ,0.001 — — ,0.001 — — — 0.48 — — 0.26

All values are bs; 95% CIs in parentheses. No-micronutrient group: groups receiving only zinc and placebo. Micronutrient group: group receiving micronutrient supplementation alone 1 group receiving micronutrient and zinc combined. No-zinc group: groups receiving micronutrient alone and placebo. Zinc group: group receiving zinc alone and group receiving micronutrient and zinc combined. Values were determined by using an unadjusted linear mixedeffects model with follow-up time as a fixed effect and the participant as a random effect. Data were analyzed as intention to treat on the basis of the available case analysis. The interaction between micronutrient and zinc supplementation was tested in an unadjusted model and a model with additional fixed (age, sex, maternal and paternal educational level, baseline diagnosis, and energy and zinc intakes) and random (study site) effects. In the fully adjusted model, there was no interaction between micronutrient and zinc for serum zinc (P = 0.21) and serum copper (0.08). In the unadjusted analysis, there was an interaction between micronutrient and zinc for serum zinc (P = 0.02) but not serum copper (P = 0.09). 1

study in 22 children compared zinc, vitamin A, and fish-oil supplementation along with ATT with only ATT and could not find any difference in BMI (34). Effects of micronutrients used in various combinations have been reported in adult tuberculosis patients by using different outcome measures. Initial reports from such studies were encouraging; however, the more-recent ones have not been able to support the beneficial role of micronutrient supplementation in the tuberculosis outcome (18–20). Although vitamin D deficiency is common in patients with tuberculosis, the effect of supplementation on plasma concentrations has been inconsistent. Vitamin D supplements may have a beneficial effect on early sputum conversion, but larger trials are needed to confirm this effect (16, 35).

Effects on growth We did not observe any effects of micronutrients or zinc supplementation on weight gain in children who were suffering from intrathoracic tuberculosis. In adults, supplementation with micronutrients and zinc increased weight gain in Tanzanian tuberculosis patients (22). A later, larger study in the same population also showed increased weight gain but only in HIV-negative tuberculosis patients (36). In contrast, there were no effects of vitamin A and zinc supplementation in adult tuberculosis patients neither in Indonesia (21, 37) nor South Africa (38). A recent guideline on nutritional supplementation did not recommend any additional micronutrient supplementation for patients with active tuberculosis (39). Micronutrient supplements do not provide energy, and thus, an effect on weight is only conceivable if supplements increase

appetite and, hence, energy intake, if supplements increase the synthesis of lean rather than fat mass, or if supplements prevent or lead to a faster recovery from diseases. Effects of supplementation with zinc and other micronutrients are likely to differ between populations, because the background intake and status are likely to be different, and effects are only likely if deficiencies are widespread (40). In contrast, we showed that micronutrient but not zinc supplementation was associated with a higher gain in height. However, height gain was not a predefined outcome, and this result could have been an incidental finding. Nevertheless, in the trial in children with tuberculosis in Tanzania, an effect of multivitamins on height was also seen but only in a subgroup of HIV-positive children (23). Meta-analyses of studies in healthy children in lowincome countries showed that micronutrient supplementation improved linear growth (41), and zinc supplementation improved both linear and ponderal growths (42). We did not see any effect of zinc on linear growth, which could have been due to the lack of other growth nutrients. A systematic review reported no significant improvement in death rates and smear positivity at 1 and 2 mo with the use of multiple micronutrient supplements in adult patients (35, 36).

Effects on resolution of CXR findings The radiological improvement in our cohort of children was good across all intervention arms. These results were obtained in the setting of a clinical trial, which may have been different from in a routine setting. Specific provisions were made to ensure

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

Serum zinc (mg/dL) Month 0 Month 2 Change Difference Month 6 Change Difference Serum copper (mg/dL) Month 0 Month 2 Change Difference Month 6 Change Difference

Micronutrient group (n = 202)

Zinc supplementation

1296

LODHA ET AL

adherence and follow-ups, which were likely to have positively influenced overall outcomes. We did not show any influence of zinc or micronutrients supplementation on CXR resolution in our cohort of children. A study in Tanzanian children could not show any improvement in CXR resolution in vitamin-supplemented compared with placebo groups (23). Studies in adult populations have documented inconsistent effects of micronutrient supplementation on CXR resolution. Karyadi et al (21) documented an earlier resolution of CXR lesions with vitamin A and zinc supplementation in Indonesian patients, but a later study by the same groups refuted this observation (37). A study on South African patients too could not show any beneficial effect of vitamin A and zinc supplementation on CXR resolution rates (38).

We did not find any difference in the sputum conversion across intervention arms of this study. An earlier sputum conversion was shown in adult Indonesian tuberculosis patients who were supplemented with vitamin A and zinc (21). A later trial, in severely malnourished Indonesian tuberculosis patients, did not confirm that a single or combined supplementation of zinc and vitamin A significantly reduced sputum conversion time or had other significant benefits (37). A systematic review reported no significant improvement in death rates and smear positivity at 1 and 2 mo with the use of micronutrient supplements in adult patients (43, 44). Serum zinc was observed to be increasing with supplementation as was expected, but this result did not translate into a greater improvement in anthropometric outcomes in these children. Serum copper showed declining trends with ATT irrespective of supplementation. This result might have been because copper and ceruloplasmin are positive acute-phase reactants (45). There have been some attempts to look into the effect of micronutrient supplementation on the cell-mediated immunity against tuberculosis. Kawai et al (46) observed that there was no beneficial effect of vitamins (A, B complex, C, and E) and selenium supplementation on the lymphocyte proliferation in response to mycobacterial proteins. Doses used in the current trial and previously reported trials may have been insufficient for a therapeutic effect; additional studies may consider the use of higher doses of micronutrients. Strengths of our study included the double-blind, randomized, placebo-controlled, 2 3 2 factorial design with a sufficient sample size and low loss to follow-up. Limitations included that the confirmatory diagnosis was based on clinicoradiologic features and not only on a microbiologic basis; however, we had a microbiologic confirmation in 44.4% of subjects. Analyses of data in subjects with microbiologically confirmed tuberculosis were not significantly different (data not shown). Primary outcomes were weight gain and radiologic clearance; radiologic findings may have been subjective. We tried to minimize the subjectivity; CXRs were read by 2 pediatricians who were unaware of clinical features and outcomes. A standardized format was used to report findings of X-ray films; the agreement between 2 readers was fair. The interobserver variability in readings of CXRs in pediatric patients has been often reported (47, 48). To counteract this variability, a third pediatrician was asked to read CXRs in cases of disagreement between the first 2 readers, and the majority decision was accepted. No adjustment for multiple

We acknowledge the contributions of the research study staff and statistician Christian Ritz, University of Copenhagen, for advice on the statistical analyses. The Delhi Pediatric TB study group (in alphabetical order) is as follows: S Aneja, Tina Arya, S Bhatnagar, J Chandra, AK Dutta, TM Doherty (Denmark), H Friis (Denmark), Harleen MS Grewal (Norway), AC Hesseling (South Africa), SK Kabra, Rakesh Lodha, B Marais (Australia), Aparna Mukherjee, Deepak Parashar, Suneel Prajapati, Kamna Purohit, Deepak Saini, Savita Saini, Ravi Raj Singh, Sarman Singh, and Varinder Singh. Members of the DSMB were as follows: RM Pandey, Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India; GR Sethi, Department of Pediatrics, Maulana Azad Medical College, New Delhi, India; and Anand Jaiswal, Department of Tuberculosis & Respiratory Diseases, National Institute of Tuberculosis and Respiratory Diseases, New Delhi, India. The authors’ responsibilities were as follows—RL, VS, S Singh, SB, HF, SKK, and HMSG: designed the research; RL, AM, VS, S Saini, S Singh, and SKK: conducted the research; RL, AM, and DF-J: analyzed data; RL, AM, HF, SKK, and HMSG: wrote the manuscript; SKK and HMSG: had primary responsibility for the final content of the manuscript; and all authors: read and approved the final manuscript. None of the authors had a conflict of interest.

REFERENCES 1. Stro¨hle A, Wolters M, Hahn A. Micronutrients at the interface between inflammation and infection-ascorbic acid and calciferol. Part 2: calciferol and the significance of nutrient supplements. Inflamm Allergy Drug Targets 2011;10:64–74. 2. Stro¨hle A, Wolters M, Hahn A. Micronutrients at the interface between inflammation and infection–ascorbic acid and calciferol: part 1, general overview with a focus on ascorbic acid. Inflamm Allergy Drug Targets 2011;10:54–63. 3. Taylor CE, Camargo CA Jr. Impact of micronutrients on respiratory infections. Nutr Rev 2011;69:259–69. 4. Gunville CF, Mourani PM, Ginde AA. The role of vitamin D in prevention and treatment of infection. Inflamm Allergy Drug Targets 2013; 12:239–45. 5. Yakoob MY, Theodoratou E, Jabeen A, Imdad A, Eisele TP, Ferguson J, Jhass A, Rudan I, Campbell H, Black RE, et al. Preventive zinc supplementation in developing countries: impact on mortality and morbidity due to diarrhea, pneumonia and malaria. BMC Public Health 2011;11(suppl 3):S23. 6. Ross AC. Vitamin A and retinoic acid in T cell-related immunity. Am J Clin Nutr 2012;96:1166S–72S. 7. WHO. TB Report, 1996. Geneva, Switzerland: World Health Organization, 1996. 8. Cegielski JP, McMurray DN. The relationship between malnutrition and tuberculosis: evidence from studies in humans and experimental animals. Int J Tuberc Lung Dis 2004;8:286–98. 9. Byrd RP Jr, Mehta JB, Roy TM. Malnutrition and pulmonary tuberculosis. Clin Infect Dis 2002;35:634–5. 10. Koh GC, Hawthorne G, Turner AM, Kunst H, Dedicoat M. Tuberculosis incidence correlates with sunshine: an ecological 28-year time series study. PLoS ONE 2013;8:e57752. 11. Gray K, Wood N, Gunasekera H, Sheikh M, Hazelton B, Barzi F, Isaacs D. Vitamin d and tuberculosis status in refugee children. Pediatr Infect Dis J 2012;31:521–23. 12. Talat N, Perry S, Parsonnet J, Dawood G, Hussain R. Vitamin D deficiency and tuberculosis progression. Emerg Infect Dis 2010;16:853–5. 13. Williams B, Williams AJ, Anderson ST. Vitamin D deficiency and insufficiency in children with tuberculosis. Pediatr Infect Dis J 2008; 27:941–2.

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

Effects on sputum conversion rates

comparisons was made because primary outcomes were nonsignificant. In conclusion, micronutrient supplementation does not modify the outcome of childhood intrathoracic tuberculosis in the form of weight gain or X-ray clearance. However, micronutrient supplementation with or without zinc may improve linear growth in children with intrathoracic tuberculosis. Additional trials that use different formulations and doses may be performed to evaluate the role of micronutrient supplementation in childhood tuberculosis.

MICRONUTRIENTS IN CHILDHOOD TUBERCULOSIS

30. 31. 32. 33. 34. 35. 36.

37.

38.

39. 40. 41. 42.

43. 44. 45. 46.

47.

48.

duced sputum in children with intra-thoracic tuberculosis. Pediatr Infect Dis J 2013;32:1313–7. Cellestis. QuantiFERON - TB Gold In-Tube. Available from: http:// www.cellestis.com/irm/content/aust/qtfproducts_tbgoldintube.html (cited 16 October 2013). de Benoist B, Darnton-Hill I, Davidsson L, Fontaine O, Hotz C. Conclusions of the Joint WHO/UNICEF/IAEA/IZiNCG Interagency Meeting on Zinc Status Indicators. Food Nutr Bull 2007;28:S480–4. Seth V, Kabra SK. Pulmonary tuberculosis. In: Seth V, Kabra SK, eds. Essentials of Tuberculosis in Children. New Delhi: Jaypee, 2010: 101–21. McAlister FA, Straus SE, Sackett DL, Altman DG. Analysis and reporting of factorial trials: a systematic review. JAMA 2003;289:2545–53. Nenni V, Nataprawira HM, Yuniati T. Role of combined zinc, vitamin A, and fish oil supplementation in childhood tuberculosis. Southeast Asian J Trop Med Public Health 2013;44:854–61. Paton NI, Chua YK, Earnest A, Chee CB. Randomized controlled trial of nutritional supplementation in patients with newly diagnosed tuberculosis and wasting. Am J Clin Nutr 2004;80:460–5. PrayGod G, Range N, Faurholt-Jepsen D, Jeremiah K, Faurholt-Jepsen M, Aabye MG, Jensen L, Jensen AV, Grewal HM, Magnussen P, et al. Daily multi-micronutrient supplementation during tuberculosis treatment increases weight and grip strength among HIV-uninfected but not HIV-infected patients in Mwanza, Tanzania. J Nutr 2011;141:685–91. Pakasi TA, Karyadi E, Suratih NM, Salean M, Darmawidjaja N, Bor H, van der Velden K, Dolmans WM, van der Meer JW. Zinc and vitamin A supplementation fails to reduce sputum conversion time in severely malnourished pulmonary tuberculosis patients in Indonesia. Nutr J 2010;9:41. Visser ME, Grewal HM, Swart EC, Dhansay MA, Walzl G, Swanevelder S, Lombard C, Maartens G. The effect of vitamin A and zinc supplementation on treatment outcomes in pulmonary tuberculosis: a randomized controlled trial. Am J Clin Nutr 2011;93:93–100. Guideline WHO. Nutritional care and support for patients with tuberculosis. Geneva, Switzerland: World Health Organization, 2013. Friis H. Micronutrient interventions and HIV infection: a review of current evidence. Trop Med Int Health 2006;11:1849–57. Ramakrishnan U, Nguyen P, Martorell R. Effects of micronutrients on growth of children under 5 y of age: meta-analyses of single and multiple nutrient interventions. Am J Clin Nutr 2009;89:191–203. Brown KH, Peerson JM, Rivera J, Allen LH. Effect of supplemental zinc on the growth and serum zinc concentrations of prepubertal children: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2002;75:1062–71. Abba K, Sudarsanam TD, Grobler L, Volmink J. Nutritional supplements for people being treated for active tuberculosis. Cochrane Database Syst Rev 2008;4:CD006086. Sinclair D, Abba K, Grobler L, Sudarsanam TD. Nutritional supplements for people being treated for active tuberculosis. Cochrane Database Syst Rev 2011;11:CD006086. Cernat RI, Mihaescu T, Vornicu M, Vione D, Olariu RI, Arsene C. Serum trace metal and ceruloplasmin variability in individuals treated for pulmonary tuberculosis. Int J Tuberc Lung Dis 2011;15:1239–45. Kawai K, Meydani SN, Urassa W, Wu D, Mugusi FM, Saathoff E, Bosch RJ, Villamor E, Spiegelman D, Fawzi WW. Micronutrient supplementation and T cell-mediated immune responses in patients with tuberculosis in Tanzania. Epidemiol Infect 2014;142:1505–9. Bada C, Carreazo NY, Chalco JP, Huicho L. Inter-observer agreement in interpreting chest X-rays on children with acute lower respiratory tract infections and concurrent wheezing. Sao Paulo Med J 2007;125: 150–4. Patel AB, Amin A, Sortey SZ, Athawale A, Kulkarni H. Impact of training on observer variation in chest radiographs of children with severe pneumonia. Indian Pediatr 2007;44:675–81.

Downloaded from ajcn.nutrition.org at GEORGE WASHINGTON UNIVERSITY on October 29, 2014

14. Sita-Lumsden A, Lapthorn G, Swaminathan R, Milburn HJ. Reactivation of tuberculosis and vitamin D deficiency: the contribution of diet and exposure to sunlight. Thorax 2007;62:1003–7. 15. Friis H, Range N, Changalucha J, Praygod G, Jeremiah K, FaurholtJepsen D, Krarup H, Mølgaard C, Andersen AB, Vitamin D. Status among pulmonary TB patients and non-TB controls: a cross-sectional study from Mwanza, Tanzania. PLoS ONE 2013;8:e81142. 16. Salahuddin N, Ali F, Hasan Z, Rao N, Aqeel M, Mahmood F. Vitamin D accelerates clinical recovery from tuberculosis: results of the SUCCINCT Study [Supplementary Cholecalciferol in recovery from tuberculosis]. A randomized, placebo-controlled, clinical trial of vitamin D supplementation in patients with pulmonary tuberculosis’. BMC Infect Dis 2013;13:22. 17. Ganmaa D, Giovannucci E, Bloom BR, Fawzi W, Burr W, Batbaatar D, Sumberzul N, Holick MF, Willett WC. Vitamin D, tuberculin skin test conversion, and latent tuberculosis in Mongolian school-age children: a randomized, double-blind, placebo-controlled feasibility trial. Am J Clin Nutr 2012;96:391–6. 18. Martineau AR, Timms PM, Bothamley GH, Hanifa Y, Islam K, Claxton AP, Packe GE, Moore-Gillon JC, Darmalingam M, Davidson RN, et al. High-dose vitamin D(3) during intensive-phase antimicrobial treatment of pulmonary tuberculosis: a double-blind randomised controlled trial. Lancet 2011;377:242–50. 19. Wejse C, Gomes VF, Rabna P, Gustafson P, Aaby P, Lisse IM, Andersen PL, Glerup H, Sodemann M. Vitamin D as supplementary treatment for tuberculosis: a double-blind, randomized, placebo-controlled trial. Am J Respir Crit Care Med 2009;179:843–50. 20. Ralph AP, Waramori G, Pontororing GJ, Kenangalem E, Wiguna A, Tjitra E, Sandjaja, Lolong DB, Yeo TW, Chatfield MD, et al. L-arginine and vitamin D adjunctive therapies in pulmonary tuberculosis: a randomised, double-blind, placebo-controlled trial. PLoS ONE 2013;8:e70032. 21. Karyadi E, West CE, Schultink W, Nelwan RH, Gross R, Amin Z, Dolmans WM, Schlebusch H, van der Meer JW. A double-blind, placebocontrolled study of vitamin A and zinc supplementation in persons with tuberculosis in Indonesia: effects on clinical response and nutritional status. Am J Clin Nutr 2002;75:720–7. 22. Range N, Changalucha J, Krarup H, Magnussen P, Andersen AB, Friis H. The effect of multi-vitamin/mineral supplementation on mortality during treatment of pulmonary tuberculosis: a randomized two-by-two factorial trial in Mwanza, Tanzania. Br J Nutr 2006;95:762–70. 23. Mehta S, Mugusi FM, Bosch RJ, Aboud S, Chatterjee A, Finkelstein JL, Fataki M, Kisenge R, Fawzi WW. A randomized trial of multivitamin supplementation in children with tuberculosis in Tanzania. Nutr J 2011;10:120. 24. Migliori GB, Borghesi A, Rossanigo P, Adriko C, Neri M, Santini S, Bartoloni A, Paradisi F, Acocella Gl. Proposal of an improved score method for the diagnosis of pulmonary tuberculosis in childhood in developing countries. Tuber Lung Dis 1992;73:145–9. 25. Chauhan LS, Arora VK; Central TB Division, Directorate General of Health Services, Ministry of Health and Family Welfare; Indian Academy of Pediatrics. Management of pediatric tuberculosis under the Revised National Tuberculosis Control Program (RNTCP). Indian Pediatr 2004;41:901–5. 26. Kabra SK, Lodha R, Seth V. Category based treatment of tuberculosis in children. Indian Pediatr 2004;41:927–37. 27. Expert group of Indian Council of Medical Research. Nutrient requirements and Recommended Dietary Allowances for Indians. Hyderabad, India: Indian Council of Medical Research, 1990. 28. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). NHANES - National Health and Nutrition Examination Survey homepage. 2005. Available from: http://www.cdc. gov/NCHS/nhanes.htm (cited 4 March 2013). 29. Mukherjee A, Singh S, Lodha R, Singh V, Hesseling A, Grewal HM, Kabra S. for the Delhi Pediatric TB Study Group. Ambulatory gastric lavages provide better yields of Mycobacterium tuberculosis than in-

1297

Effect of micronutrient supplementation on treatment outcomes in children with intrathoracic tuberculosis: a randomized controlled trial.

Micronutrients play an important role in immune function. To our knowledge, there have been no comprehensive studies on the role of micronutrient supp...
197KB Sizes 0 Downloads 8 Views