SEARCH

SEARCH BY CITATION

Keywords:

  • micronutrient;
  • infection;
  • tuberculosis;
  • adult;
  • Nigeria;
  • clinical trial

Summary

  1. Top of page
  2. SummaryEssai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaireEnsayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Objective  To assess the efficacy of weekly zinc or zinc plus retinol as adjuncts for the treatment of pulmonary tuberculosis.

Methods  Double-blind, randomized, placebo-controlled trial in 350 patients >15 years old with smear-positive tuberculosis in Nigeria (ISRCTN36636609). In addition to antituberculous treatment, patients were randomly allocated to weekly supplements of zinc (90 mg), zinc plus retinol (5000 IU) or placebos for 6 months. Primary outcomes were time to sputum smear conversion and resolution of radiographic abnormalities.

Results  After 8 weeks of treatment, 68% had achieved sputum smear conversion, and the median conversion time was 6.5 weeks. Hazard ratios (HR, 95%CI) for sputum conversion relative to the placebo group were not significant for zinc (1.07, 0.92–1.29) or zinc plus retinol (0.89, 0.76–1.07). Significant predictors of time to sputum conversion were lung abnormality score, sputum smear grade, age and serum C-reactive protein. HIV co-infection and gender were not independent predictors of time to sputum conversion. There were no significant differences between supplement groups in clinical, radiological or laboratory outcomes at 2 months or 6 months. There were 9, 9 and 2 deaths in patients receiving zinc, zinc plus retinol or placebos, respectively. Mortality in those who received zinc (HR 1.71, 0.88–3.58) or zinc plus retinol (HR 1.54, 0.78–3.26) did not differ significantly from those who received placebos. Most deaths occurred in patients co-infected with HIV.

Conclusions  Supplementation with zinc or zinc plus retinol did not lead to better outcomes than placebos, and caution is warranted regarding routine micronutrient supplementation, particularly in patients co-infected with HIV.

Essai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaire

Objectif:  Evaluer l’efficacité de la prise hebdomadaire de zinc ou de zinc plus du rétinol comme compléments dans le traitement de la tuberculose (TB) pulmonaire.

Méthodes:  Essai randomisé, placebo contrôlée en double-aveugle chez 350 patients de plus de 15 ans atteints de TB à frottis positif au Nigeria (ISRCTN36636609). En plus du traitement antituberculeux, les patients ont été randomisés pour des suppléments hebdomadaires de zinc (90 mg), zinc plus rétinol (5000 UI) ou de placebo durant 6 mois. Les objectifs primaires consistaient au temps pour la conversion des frottis de crachats et la résolution des anomalies radiologiques.

Résultats:  Après 8 semaines de traitement, 68% des patients avaient atteint une conversion des frottis d’expectoration et le temps médian de conversion était de 6,5 semaines. Les rapports de risque (HR; IC95%) pour la conversion des expectorations par rapport au groupe placebo n’étaient pas significatifs pour le zinc (1,07; 0,92 à 1,29) ou le zinc plus rétinol (0,89; 0,76 à 1,07). Les facteurs prédictifs significatifs pour le temps à la conversion des expectorations étaient les scores d’anomalie des poumons, le degré de positivité du frottis, l’âge et la protéine C réactive. La coinfection avec le VIH et le sexe n’étaient pas des prédicteurs indépendants du temps de conversion des crachats. Il n’y avait pas de différences significatives entre les groupes recevant des suppléments pour les résultats cliniques, radiologiques ou de laboratoire à 2 mois ou 6 mois. Il y avait 9, 9 et 2 décès chez les patients recevant le zinc, le zinc plus rétinol et le placebo, respectivement. La mortalité chez ceux ayant reçu du zinc (HR: 1,71; 0,88 à 3,58) ou du zinc plus rétinol (HR: 1,54; 0,78 à 3,26) ne différait pas significativement de celle chez qui ont reçu du placebo. La plupart des décès sont survenus chez des patients coinfectés par le VIH.

Conclusions:  La supplémentation en zinc ou zinc plus rétinol n’a pas conduit à de meilleurs résultats que le placebo, la prudence est justifiée pour ce qui concerne la supplémentation en micronutriments en routine, en particulier chez les patients coinfectés par le VIH.

Mots-clés:  micronutriments, infection, tuberculose, adultes, Nigéria, essai clinique

Ensayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar

Objetivo:  Evaluar la eficacia del zinc y del zinc más retinol como adyuvantes en el tratamiento de la tuberculosis pulmonar.

Métodos:  Ensayo doble ciego, aleatorizado y controlado con placebo en 350 pacientes >15 de años con frotis positivo para tuberculosis en Nigeria (ISRCTN36636609). Además del tratamiento anti-tuberculoso, las pacientes fueron aleatorizados para recibir suplementos semanales de zinc (90 mg), zinc más retinol (5000 IU), o placebo durante 6 meses. Los resultados primarios fueron el tiempo hasta la conversión del frotis de esputo y la resolución de las anormalidades radiográficas.

Resultados:  Trás 8 semanas de tratamiento, un 68% habían alcanzado la conversión del esputo, y el tiempo medio hasta la conversión era de 6.5 semanas. La tasa de riesgo relativo (HR, 95% IC) para la conversión del frotis de esputo con relación al grupo placebo no eran significativas para el zinc (1.07, 0.92-1.29) o el zinc más retinol (0.89, 0.76-1.07). Los vaticinadores significativos para el tiempo hasta la conversión del esputo eran el grado de anormalidad pulmonar, la graduación de los resultados de la baciloscopia, la edad, y la proteína C-reactiva en suero. La co-infección con VIH y el género no eran vaticinadores independientes del tiempo hasta la conversión del esputo. No había diferencias significativas entre los grupos de suplementos en cuanto a resultados clínicos, radiológicos o de laboratorio a los 2 o 6 meses. Hubo 9, 9, y 2 muertes entre pacientes recibiendo zinc, zinc más retinol, o placebo, respectivamente. La mortalidad entre aquellos que recibieron zinc (HR 1.71, 0.88-3.58) o zinc más retinol (HR 1.54, 0.78-3.26) no difirió significativamente de aquellos que recibieron placebo. La mayoría de las muertes ocurrieron en pacientes co-infectados con VIH.

Conclusiones:  La suplementación con zinc o zinc más retinol no llevó a mejores resultados que los placebos, y se debe asegurar cautela en la suplementación rutinaria con micronutrientes, especialmente en pacientes coinfectados con VIH.

Palabras clave:  micronutrientes; infección; tuberculosis; adulto; Nigeria; ensayo clínico


Introduction

  1. Top of page
  2. SummaryEssai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaireEnsayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

The integrity of the immune system is influenced by the nutritional status of an individual, and nutritional status is affected by infection (Chandra & Kumari 1994; Scrimshaw 1995; Wellinghausen et al. 1997). In addition to protein and energy malnutrition, micronutrient deficiencies contribute to the morbidity and mortality of infectious diseases. Several studies have described lower micronutrient concentrations in infected than in non-infected persons (Guerrant et al. 2000; Karyadi et al. 2000). Patients with chronic infections like tuberculosis (TB) often have low micronutrient concentrations (Onwubalili et al. 1988; Saha & Rao 1989; Karyadi et al. 2000). These include vitamin A and zinc deficiencies (Evans & Attock 1971; van Lettow et al. 2004), which are more severe in patients co-infected with HIV (Rwangabwoba et al. 1998; van Lettow et al. 2003; Mugusi et al. 2003).

The observation of micronutrient deficiencies in adults with TB led to intervention studies to explore whether replacing these micronutrients would improve patient recovery. The results of these trials, however, have been inconclusive. For example, a trial in Indonesia reported that patients with pulmonary TB (PTB) supplemented with zinc and vitamin A had earlier sputum conversion and faster X-ray lesion resolution than controls (Karyadi et al. 2002). However, a similar trial in Tanzania failed to reproduce these findings (Range et al. 2005). In addition, whether the potential benefits of zinc and/or vitamin A supplementation would be similar in patients with and without HIV co-infection is unknown. A study conducted in Malawi found that a widely used micronutrient supplement did not reduce mortality among antiretroviral-naïve HIV-infected adults with PTB (Semba et al. 2007). Additionally, a Cochrane review concluded that micronutrient supplementation in patients infected with HIV does not reduce morbidity or mortality (Irlam et al. 2005). However, another Cochrane review found limited evidence that some combinations of zinc with other micronutrients in tuberculosis may reduce mortality in those co-infected with HIV and improve weight gain, and further trials were recommended (Abba et al. 2008).

There have been no previous zinc or vitamin A supplementation studies in patients with TB or HIV in Nigeria, and limited data regarding the prevalence of micronutrient deficiencies in the general population are available. The mean dietary intake and mean serum retinol concentrations of pregnant women receiving antenatal care in a south-eastern Nigerian city were higher than the FAO/WHO recommended values (Williams et al. 2008). However, a nationwide survey found that 30% of children under age 5 years were vitamin A deficient, as determined by serum retinol concentrations (Maziya-Dixon et al. 2006). The national prevalence of zinc deficiency was 44% and 28% in pregnant women and mothers, respectively (Maziya-Dixon et al. 2004).

We conducted a randomized, double-blind, placebo-controlled clinical trial to test the hypothesis that zinc and vitamin A supplementation would improve the response to antituberculous drug treatment by earlier sputum conversion. We compared whether weekly supplementation with zinc or zinc plus retinol led to more rapid clearance of infectious bacilli from sputum and more rapid clinical improvement than placebo supplementation in persons treated for PTB in Nigeria, an area with a relatively high prevalence of HIV.

Materials and methods

  1. Top of page
  2. SummaryEssai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaireEnsayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Subjects

Patients ≥15 years of age with a history of chronic cough for >2 weeks’ duration were prospectively recruited from eight district hospitals in Abuja, Nigeria, between September 2003 and June 2005. Potential patients were excluded if they had had major surgery during the previous month and had a history of diabetes mellitus or severe cardiovascular, hepatic or renal disease, or previous treatment of TB. Patients were also excluded if they were taking corticosteroids or zinc supplements during the previous month, were pregnant, lactating, or taking oral contraceptives, or were unable to return for regular follow-up visits.

All patients underwent three sputum examinations, collected on the spot, early morning and second spot, as recommended by the National TB Control Programme (NTP) of Nigeria. Sputum samples were stained with Ziehl-Neelsen and examined using direct light microscopy. Patients with at least one smear with ≥1 acid fast bacillus (AFB) were invited to participate. Sputum smears were graded as negative, scanty (1–9 bacilli per 100 fields), + (10–99 bacilli per 100 fields), ++ (1–9 bacilli per field) or +++ (10 or more bacilli per field) (Lawson et al. 2006). Sputum grades were rank ordered with values of 0–4, respectively, for the purposes of analysis. The highest sputum grade of the three sputum examinations was used for analysis. Patients were enrolled after obtaining informed oral consent. Ethical approval for the study was obtained from the Liverpool School of Tropical Medicine Research Ethics Committee and Zankli Medical Centre Institutional Review Board. The clinical trial was registered as ISRCTN36636609.

Intervention

All patients were provided the standard NTP treatment for TB. One of the investigators who was not at the study site prepared the allocation sequence using random numbers generated with Minitab (Seattle, WA), and the randomization code was kept at the Liverpool School of Tropical Medicine. Treatment allocation into three supplement groups was performed at the Liverpool School of Tropical Medicine using permuted block randomization with four different block sizes. Treatment group allocation was designated by a lettered code and concealed from the investigators and subjects until data analysis was completed. One group received 90 mg of elemental zinc by mouth weekly as two tablets of zinc sulphate in a lactose matrix (Pharmadass Ltd., Middlesex, U.K.) and a placebo resembling the vitamin A supplement. The second group received weekly zinc tablets and vitamin A capsules as 1500 mcg retinol equivalents (5000 IU) (G&G Food Supplies Ltd., West Sussex, U.K.). The third group received placebo tablets resembling both the zinc and vitamin A supplements. We chose to provide weekly doses of supplements to facilitate monitoring and assure compliance, as patients were given weekly supplies of antituberculous drugs. However, because of safety concerns that high doses of zinc or vitamin A could potentially increase mortality, we selected a weekly dose approximately five times the daily dose of zinc and the daily dose of vitamin A used in an Indonesian trial (Karyadi et al. 2002).

Treatment assignments were prepared in serially numbered sealed envelopes that were opened only at enrolment. Full courses of supplements were packaged in sequentially numbered packets assigned consecutively to individual subjects at enrolment, according to the allocation sequence. Study subjects, the clinicians providing treatment and assessing outcomes, and the laboratory staff were unaware of the treatment group assignments. We did not explicitly assess the success of blinding. Supplements were given weekly under direct observation for the first 2 months. Following the first 2 months, a monthly supply of supplements was provided to complete 6 months of supplementation. Enrolled patients were instructed not to take any other vitamin supplements during the study. Patients were asked by the supervising nurse to come with a responsible family member who was entrusted to supervise the patient taking the drugs. Apart from recording in a patient’s card, the family member was asked to accompany the patient with the empty drug foils for the previous month at the monthly visits. Patients who returned for follow-up consistently returned with empty foils.

All patients received rifampin, isoniazid, ethambutol and pyrazinamide as a fixed-dose combination for 2 months followed by isoniazid and ethambutol for 6 months. Patients were monitored through the TB clinics of the hospitals, following directly observed treatment short course (DOTS) program guidelines. Patients were offered HIV testing on enrolment with pre- and post-test counselling. Patients who did not return for scheduled follow-up appointments were contacted in their homes to ascertain any mortality and arrange continued follow-up. None of the patients with HIV received any antiretroviral drugs during their antituberculous drug treatment.

Clinical variables

Blood samples were collected at enrolment, and a complete blood count, erythrocyte sedimentation rate (ESR), and biochemical tests were performed. Blood samples were tested for HIV using ImmunoComb HIV 1 & 2 BiSpot kit (Orgenics Ltd., Yavne, Israel). A chest X-ray was obtained in each patient at enrolment and at 2 and 6 month after initiation of therapy. Sputum specimens were collected weekly for the first 8 weeks of therapy and at 12, 16, 20 and 24 weeks. On each occasion, patients were asked to provide two sputum specimens, collected as morning and on-the-spot, except at enrolment when three sputum samples were collected. Patients were declared sputum negative for a given follow-up if both smears were available and were negative.

Chest X-rays were read independently by a chest physician (PDOD) and a radiologist. X-rays were scored as 0–6 depending on the number of abnormalities in the lung zones. Each lung was divided into three zones, following the anatomical distribution of the right lung, and each lung zone was graded as 0 (normal) or 1 (affected). A separate score was used for cavities, which were scored from 0 to 3 based on the total (cumulative) diameter of the cavities in the lung fields. No cavity present was graded as zero; cavities with a cumulative diameter <2 cm were graded as 1; cavities with a cumulative diameter between 2 and 4 cm were graded as 2 and those with a cumulative diameter >4 cm were graded as 3 (Simon 1966; Lawson et al. 2008).

Outcomes

The primary outcomes for the study were the time for sputum smear conversion and the change in X-ray scores. The time of sputum conversion was defined as the first of two consecutive sputum smears without bacilli. If a patient later became smear positive after two consecutive negative smears, the time to sputum conversion was considered the last two consecutive smears that reconverted to negative. If the sputum remained positive at the end of the study, the patient was not considered to have sputum smear conversion. Secondary outcomes included improvement in clinical symptoms, weight gain expressed as body mass index (BMI), and improvement in the Karnofsky performance score, ESR and haemoglobin values.

Statistical analysis

The analysis was conducted blindly using the treatment codes, and the codes were broken only after completing the initial analysis. Comparisons were made across treatment groups and between patients lost to follow-up and those remaining in the study. Data were entered and analysed using Epi Info 3.0 (CDC, Atlanta, Georgia). Outcomes were analysed on an intention-to-treat basis. Pearson’s chi-square tests and analysis of variance (anova) were used to compare categorical and continuous variables, respectively. Non-parametric tests were used for continuous variables with skewed distributions.

The sample size calculation was based on a previous trial (Karyadi et al. 2002) that reported an absolute difference of 20% in sputum conversion at 8 weeks between the supplement and placebo groups. Consequently, we calculated that 100 subjects in each group would provide 80% power to demonstrate a 20% difference in sputum conversion between the supplement and placebo groups at 8 weeks. We increased the enrolment total to 350 to allow for loss to follow-up.

We had initially intended to compare and report the mean time to sputum conversion between the three groups by anova. Because anova demonstrated a significant difference in sputum conversion time between the three groups and the conversion time was not normally distributed, we elected to explore this difference further in an unblinded manner with a Kaplan–Meier analysis (JMP 8.0, SAS Institute Inc. Cary, NC, USA). A Cox proportional hazards analysis was used to control for the independent contribution of variables on the time to sputum conversion. The linearity of continuous variables was confirmed by adding a squared term for each continuous variable to the proportional hazards model and assessing for significance of the squared term. None of the continuous variables violated the assumption of linearity. Mortality differences between groups were assessed with a Cox proportional hazards analysis. P-values <0.05 were considered significant. Three interim analyses were performed at the Liverpool School of Tropical Medicine for safety surveillance prior to conclusion of the study. The supplement codes of two patients with major adverse events were broken, and the patients were withdrawn from the study. Both patients were in the dual placebo group.

Results

  1. Top of page
  2. SummaryEssai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaireEnsayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

A total of 1321 patients with chronic cough were screened and 399 were smear positive. Of these, 49 were not eligible for enrolment, because they had relocated to another address, did not return for their results, were pregnant or had surgery. A total of 350 smear-positive patients were enrolled and randomized into the supplementation groups (Figure 1). Of these, 117, 117, and 116 were randomized to receive zinc, zinc plus retinol and dual placebos, respectively. Baseline characteristics of the three groups are shown in Table 1. The mean (±SD) ages were 31 ± 9, 34 ± 12 and 29 ± 8 years for subjects in the zinc, zinc plus retinol, and dual placebo groups, respectively. The mean baseline BMI was greater in the zinc group than in the other two groups at baseline. Overall, 112 (32%) were undernourished with a BMI <18.5 kg/m2. There were no differences in the proportion of men, educational characteristics, working status or clinical characteristics between groups. The overall prevalence of HIV-1 seropositivity in enrolled patients was 45% and did not differ significantly between groups. One patient each from the zinc and placebo groups was HIV-2 seropositive. A total of 89 (76%), 81 (69%) and 91 (78%) completed the trial in the zinc, zinc plus retinol and placebo groups, respectively. Over the course of 6 months, 19 (16%), 27 (23%), and 24 (21%) in the zinc, zinc plus retinol and placebo groups, respectively, were lost to follow-up and could not be located (= 0.41).

image

Figure 1.  Flow chart of study.

Download figure to PowerPoint

Table 1.   Base-line characteristics of study subjects. Data are shown as mean ± SD (range)
 PlacebosZincZinc + Retinol
(= 116)(= 117)(= 117)
  1. ESR, erythrocyte sedimentation rate; PTB, pulmonary TB.

Age (year) 29 ± 8 (15–58) 31 ± 9 (15–70) 34 ± 12 (18–92)
 Male 30 ± 8 (20–58) 32 ± 9 (19–65) 35 ± 13 (19–92)
 Female 27 ± 8 (15–50) 29 ± 11 (15–70) 30 ± 9 (18–16)
Male: Female [n (% male)] 73:43 (63%) 76:41 (65%) 83:34 (71%)
Work status [n (%)]
 Working 66 (29%) 78 (34%) 83 (37%)
 Not working 10 (32%) 11 (36%) 10 (32%)
 Student 31 (51%) 15 (25%) 15 (25%)
 Housewife 9 (29%) 13 (42%) 9 (29%)
Education [n (%)]
 None 10 (29%) 9 (26%) 16 (46%)
 Primary 24 (32%) 29 (38%) 23 (30%)
 Secondary 54 (32%) 57 (34%) 58 (34%)
 Tertiary 28 (40%) 22 (31%) 20 (29%)
Smokers [n (%)] 30 (33%) 31 (34%) 29 (32%)
 Duration of smoking (year) 9 ± 11 (1–46) 10 ± 7 (1–27) 12 ± 12 (1–43)
 Cigarettes/day 8 ± 6 (2–30) 8 ± 8 (1–40) 8 ± 8 (1–30)
Persons sleeping in same room 3 ± 2 (1–10) 3 ± 2 (1–19) 3 ± 2 (1–10)
Contact with person with cough ≥3 week [n (%)] 19 (36%) 16 (30%) 18 (34%)
 Duration of contact (week) 5 ± 5 (1–20) 3 ± 5 (1–20) 7 ± 11 (1–44)
Contact with person with fever ≥3 wk [n (%)] 10 (35%) 5 (17%) 14 (48%)
Contact with known PTB [n (%)] 30 (43%) 20 (29%) 20 (29%)
 Duration of contact (year) 6 ± 6 (1–25) 4 ± 3 (1–16) 4 ± 3 (1–12)
HIV positive 52 (45%) 56 (49%) 47 (42%)
Presenting symptoms
 Cough116 (100%)117 (100%)116 (99%)
 Fever 82 (71%) 72 (62%) 87 (74%)
 Night Sweats 86 (74%) 83 (71%) 91 (78%)
 Hemoptysis 28 (24%) 34 (29%) 28 (31%)
 Dyspnoea 80 (69%) 81 (69%) 84 (72%)
 Chest Pain107 (92%)106 (91%)100 (86%)
Abnormal Chest Exam 58 (50%) 65 (56%) 53 (45%)
Karnofsky performance score 59 ± 11 (20–100) 60 ± 12 (20–100) 59 ± 11 (40–100)
BMI (kg/m2) 19.8 ± 3.3 (12.4–33.1) 21.3 ± 4.7 (11.7–35.5) 19.6 ± 3.5 (13.0–31.9)
Weight (kg) 51.6 ± 9.0 (28–78) 54.7 ± 10.2 (28–75) 53.1 ± 8.5 (35–73)
Haemoglobin (g/l)111 ± 22 (60–170)113 ± 22 (60–160)110 ± 23 (50–160)
Albumin (g/l) 32 ± 8 (16–70) 32 ± 8 (17–60) 32 ± 9 (14–70)
ESR (mm/h) 86 ± 39 (7–160) 82 ± 41 (2–150) 87 ± 37 5–152)
Total Bilirubin (μm) 7.6 ± 6.8 (0–51) 8.4 ± 8.0 (1.7–68) 8.0 ± 5.8 (1.7–51)
Alkaline phosphate (U/l)245 ± 171 (74–911)238 ± 122 (92–675)260 ± 169 (4–940)
C-reactive protein (nm)383 ± 169 (64–974)396 ± 164 (118–910)352 ± 102 (155–645)
Lung abnormality X-ray score 3.2 ± 1.5 (0–6) 3.1 ± 1.4 (0–6) 3.0 ± 1.3 (0–6)
Cavity X-ray score 1.6 ± 1.0 (0–3) 1.6 ± 1.0 (0–3) 1.7 ± 1.0 (0–3)
Sputum smear grade
 Scanty 6 7 5
 + 27 29 34
 ++ 29 33 37
 +++ 54 48 41

Overall, by the end of 8 weeks of treatment, 68% had cleared their sputum of bacilli, and the median sputum conversion time was 6.5 weeks (Figure 2). Hazard ratios (HR, 95%CI) for sputum conversion relative to the placebo group were not significant for zinc (1.07, 0.92–1.29) or zinc plus retinol (0.89, 0.76–1.07). Limiting the analysis to the 112 undernourished subjects with BMI <18.5 kg/m2 did not demonstrate significant results for zinc (0.73, 0.59–1.15) or zinc plus retinol (0.92, 0.77–1.40).

image

Figure 2.  Rates of sputum bacilli conversion in three treatment groups.

Download figure to PowerPoint

Losses to follow-up were similar across the groups. At the end of 6 months of follow-up, one receiving zinc, nine receiving zinc plus retinol and five receiving placebos had a positive sputum smear. Of the patients who converted back to positive after having two consecutive negative sputum smears in the placebo group, five remained positive and four reconverted to negative by the end of the study. In the zinc group, one remained positive and three reconverted to negative. In the zinc plus retinol group, two remained positive and four reconverted to negative.

Increasing baseline lung abnormality score, sputum smear grade, age and CRP were significantly associated with a longer time to sputum conversion in a Cox proportional hazards analysis (Table 2). Supplement group, gender and HIV status were not associated with the time to sputum conversion. Those with HIV co-infection did not respond differently to zinc (HR 1.01, 0.85–1.20) or zinc plus retinol (HR 0.89, 0.74–1.05) than those without HIV co-infection, indicating that there were no significant interactions of treatment group with HIV status on the time to sputum conversion. The addition of BMI to this analysis did not substantially alter the results.

Table 2.   Proportional hazards analysis of factors related to sputum conversion
FactorHazard ratio*95% confidence limits
  1. *Likelihood of sputum clearance per unit change in predictor variable.

  2. †Placebo group was the referent group.

Supplement group†
 Zinc1.280.95–1.72
 Zinc + Retinol0.820.59–1.10
 C-reactive protein (nm)0.9980.996–0.999
 Enrolment age (y)0.9800.958–0.999
 Lung abnormality score0.750.64–0.89
 Sputum grade0.700.56–0.89
 HIV co-infection1.020.82–1.27
 Female gender0.970.76–1.23

Clinical, radiological and laboratory measures improved with treatment in all three groups (Figure 3). There were no significant differences between groups in the outcomes at 2 months or 6 months, when controlling for baseline values.

image

Figure 3.  Change in outcomes over 6 months by supplement group.

Download figure to PowerPoint

A total of 9 (7.7%), 9 (7.8%) and 2 (1.7%) patients died in the zinc, zinc plus retinol and placebo groups, respectively. Of these 20 patients, 16 died during the first 2 month of treatment and 15 had HIV co-infection. When the three groups were compared simultaneously, mortality of those who received zinc (HR 1.71, 0.88–3.58) or zinc plus retinol (HR 1.54, 0.78–3.26) did not differ significantly from those who received placebos (= 0.10 by log-rank test). However, if the zinc and zinc plus retinol groups were combined, those receiving any supplement had a significantly greater likelihood of death (HR 2.07, 1.11–5.20) than those in the placebo group (= 0.03 by log-rank test), even after adjusting for HIV co-infection.

As all patients received their supplement under direct observation for the first 2 months, compliance was assured for those who remained in the trial. Of those in the placebo group, 91 (79%) completed 2 months and 80 (69%) completed 6 months. Of those in the zinc group, 92 (79%) completed 2 months and 80 (68%) completed 6 months. Of those in the retinol + zinc group, 87 (74%) completed 2 months and 75 (64%) completed 6 months. During the study, there was massive demolition of the shanty towns bordering Abuja from where most of the patients came. This led to displacement and movement of some patients away from the study area. Another factor contributing to loss to follow-up was that the DOTS centres were in the city, and some patients found it difficult to travel to the centres.

Discussion

  1. Top of page
  2. SummaryEssai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaireEnsayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

We found that patients with active PTB receiving zinc or zinc plus retinol supplements had no more rapid sputum conversion or radiographic improvement than patients receiving placebos. A smaller Indonesian trial in 80 patients reported earlier sputum smear conversion accompanied by a greater improvement in lesion area on chest X-ray in patients supplemented with zinc and vitamin A compared with whose who received no supplementation in addition to their anti-TB therapy (Karyadi et al. 2002). These investigators reported nearly 100% sputum conversion by 8 weeks in the group that received zinc plus retinol, whereas we observed 63% sputum conversion by 8 weeks in the zinc plus retinol group. However, Karyadi et al. found improvement in only one of three radiologic signs at a single 2 months time point. Most subjects in that trial had a low sputum grade (<1 AFB per 10 fields), whereas the majority of patients in our trial had abundant bacilli (at least 1 AFB per field) in their sputum smear. This accounts for our observation of a longer median sputum clearance time (6.5 weeks) than that of the Indonesian trial (3.9 weeks). The Indonesian trial used higher doses of vitamin A (5000 IU daily) and zinc (15 mg daily) than we did, but serum zinc levels did not differ from the placebo group during supplementation. Our sample size had 80% power to demonstrate a 20% difference in sputum conversion at 8 weeks or a difference of 2.5 weeks in mean sputum clearance time between the supplement and placebo groups.

A small study in India described more rapid improvement of the general condition, greater weight gain and more rapid sputum conversion in patients whose anti-TB therapy was supplemented with zinc than in those receiving anti-TB therapy alone (Pant et al. 1987). In contrast, a Tanzanian study reported that zinc or multi-micronutrient supplementation did not have an effect on sputum conversion or X-ray resolution (Range et al. 2005). Vitamin A supplementation was reported to have an adverse effect on clinical response in South African children with PTB (Hanekom et al. 1997). The lack of response to micronutrient supplementation observed in African studies, including ours, may be related to the higher prevalence of HIV in patients with PTB in Africa compared with other parts of the world. The prevalence of HIV in Indonesia is relatively low (1%) (Dye et al. 1999, 2005) compared to the much higher prevalence observed in sub-Saharan Africa (Asamoah-Odei et al. 2004; World Health Organization 2009). However, we failed to identify a significant modification of the effect of micronutrient supplementation by HIV status.

An unexpected finding in our study was a higher mortality in patients receiving micronutrient supplementation, although the difference was not statistically significant. Our study was not powered to assess differences in mortality, but this finding warrants that future studies of micronutrient supplementation in TB monitor mortality as an outcome. Most deaths occurred during the early stages of treatment in patients co-infected with HIV. In our study, patients with HIV co-infection did not receive antiretroviral therapy. Zinc or zinc plus retinol may result in a greater mortality by provoking enhanced HIV viremia, immune reconstitution or side effects to the anti-TB treatment (Kupka & Fawzi 2002; Goebel 2005). Zinc and vitamin A have a prominent role in cell-mediated immunity and other immunological mechanisms (Driessen et al. 1994; Semba 1999; Thurnham et al. 2000). Patients with HIV who receive TB treatment often experience a paradoxical deterioration in clinical symptoms, caused by increased immunological response. This immunological reconstitution may be related to improvement in micronutrient status but is incompletely understood (Conesa-Botella et al. 2009).

Side effects of zinc and vitamin A supplementation have rarely been reported in other clinical trials, and supplementation is considered safe at standard nutritional doses. A study in Bangladesh reported that malnourished children receiving zinc supplementation in doses several times the recommended daily requirements had a higher mortality (Doherty et al. 1998). Observational studies examining the relationship between individual micronutrients and pregnancy outcome in HIV-infected women have described an association between low plasma concentrations of vitamin A and a higher risk of mother to child transmission of HIV (Fawzi & Msamanga 2004). Although most prospective randomized trials have failed to demonstrate a beneficial effect of vitamin A supplementation in the prevention of transmission of HIV (Fawzi et al. 2005), mothers receiving vitamin A did not have more adverse effects. One notable exception was a study in Tanzania, where supplemented mothers were more likely to transmit HIV to their babies (Fawzi et al. 1999; Wiysonge et al. 2005).

Zinc pivotally influences the actions of hundreds of enzymes, stabilizes cell membranes and modulates humoral and cell-mediated immunity (Thurnham et al. 2000), potentially influencing the host response to tubercle bacilli. Observational studies have provided conflicting results on the role of zinc and HIV progression (Kupka & Fawzi 2002), and the data are difficult to interpret. An observational study of individuals infected with HIV in the USA reported that a high dietary zinc intake was associated with an increased risk of progression to AIDS and mortality (Kupka & Fawzi 2002), although other studies have also reported a positive association with CD4 counts (Hodkinson et al. 2007; Prasad 2007).

A large study of patients with PTB in Tanzania reported that micronutrient supplementation (vitamins A, B complex, C, E and selenium) reduced the rate of recurrence of TB but had no effect on mortality in patients with or without HIV infection (Villamor et al. 2008). A 64% reduction in mortality was noted in patients without HIV infection, but the effect did not reach statistical significance. Another study in Tanzania reported that the same micronutrients but with additional zinc, vitamin D and copper reduced the mortality rate in TB patients co-infected with HIV, but not in the absence of HIV co-infection (Range et al. 2006). A study in Malawi found no effect of a multivitamin and mineral supplement on mortality in patients with or without HIV infection that were treated for TB (Semba et al. 2007). These conflicting results may in part be related to differences in micronutrient composition and dosages or because of differences in the prevalence of micronutrient deficiencies in the populations studied (Benn et al. 2008).

One limitation of our study is that we did not attempt to estimate the dietary intake of vitamin A or zinc in our subjects. We also did not measure plasma zinc or retinol concentrations to determine the baseline prevalence of micronutrient deficiencies or the relationship of micronutrient levels with clinical response. However, compliance with micronutrient supplementation was high, because supplements were given under direct observation. Our intention to treat design was meant to evaluate the effectiveness of micronutrient supplementation in clinical practice. It is possible that few of our patients were deficient in vitamin A or zinc, and for this reason, additional supplementation did not improve clinical response. Approximately one-third of our subjects were undernourished (BMI <18.5 kg/m2), which represents a lower proportion of undernourished subjects compared with other studies reporting a benefit of micronutrient supplementation (Karyadi et al. 2002; Semba et al. 2007). However, limiting our analysis to undernourished subjects, who are more likely to have micronutrient deficiencies did not alter our results.

The doses of vitamin A or zinc that we used may have been too small to observe an effect on outcomes. However, these doses would be expected to be adequate to correct deficiency states in these nutrients.

Another limitation of this study was the unknown status of those who failed to return for follow-up and who could not be located. However, as there were no significant differences in the proportions of subjects lost to follow-up between the groups, it is unlikely that the outcomes of those lost to follow-up differed significantly. Our observed 20% overall loss to follow-up for a 6-month tuberculosis treatment trial in Nigeria is relatively small.

Although studies have reported improvement in the clinical status of patients supplemented with zinc and vitamin A (Smith 1980; Karyadi et al. 2002), human data on the interaction between zinc and vitamin A are limited and inconclusive. We conclude that there was no additional benefit of supplementation with the combination of vitamin A and zinc compared with zinc alone. We found no evidence of benefit in patients with TB supplemented with zinc or zinc plus retinol compared with placebo supplements. The non-significantly greater mortality in supplemented patients warrants caution regarding routine micronutrient supplementation, particularly in African patients co-infected with HIV.

References

  1. Top of page
  2. SummaryEssai randomisé contrôlé de l’utilisation du zinc et de vitamine A comme co-adjuvants dans le traitement de la tuberculose pulmonaireEnsayo aleatorizado y controlado de zinc y vitamina A como coadyuvantes en el tratamiento de la tuberculosis pulmonar
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References
  • Abba K, Sudarsanam TD, Grobler L & Volmink J (2008) Nutritional supplements for people being treated for active tuberculosis. Cochrane Database of Systematic Reviews 4, CD006086.
  • Asamoah-Odei E, Garcia Calleja JM & Boerma JT (2004) HIV prevalence and trends in sub-Saharan Africa: no decline and large subregional differences. Lancet 364, 3540.
  • Benn CS, Friis H & Wejse C (2008) Should micronutrient supplementation be integrated into the case management of tuberculosis? Journal of Infectious Diseases 197, 14871489.
  • Chandra RK & Kumari S (1994) Nutrition and immunity: an overview. Journal of Nutrition 124, 1433S1435S.
  • Conesa-Botella A, Mathieu C, Colebunders R et al. (2009) Is vitamin D deficiency involved in the immune reconstitution inflammatory syndrome? AIDS Research and Therapy 6, 4.
  • Doherty CP, Sarkar MA, Shakur MS et al. (1998) Zinc and rehabilitation from severe protein-energy malnutrition: higher-dose regimens are associated with increased mortality. American Journal of Clinical Nutrition 68, 742748.
  • Driessen C, Hirv K, Rink L & Kirchner H (1994) Induction of cytokines by zinc ions in human peripheral blood mononuclear cells and separated monocytes. Lymphokine and Cytokine Research 13, 1520.
  • Dye C, Scheele S, Dolin P, Pathania V & Raviglione MC (1999) Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. JAMA 282, 677686.
  • Dye C, Watt CJ, Bleed DM, Hosseini SM & Raviglione MC (2005) Evolution of tuberculosis control and prospects for reducing tuberculosis incidence, prevalence, and deaths globally. JAMA 293, 27672775.
  • Evans DI & Attock B (1971) Folate deficiency in pulmonary tuberculosis: relationship to treatment and to serum vitamin A and beta-carotene. Tubercle 52, 288294.
  • Fawzi W & Msamanga G (2004) Micronutrients and adverse pregnancy outcomes in the context of HIV infection. Nutrition Reviews 62, 269275.
  • Fawzi WW, Mbise RL, Hertzmark E et al. (1999) A randomized trial of vitamin A supplements in relation to mortality among human immunodeficiency virus-infected and uninfected children in Tanzania. Pediatric Infectious Disease Journal 18, 127133.
  • Fawzi W, Msamanga G, Spiegelman D & Hunter DJ (2005) Studies of vitamins and minerals and HIV transmission and disease progression. Journal of Nutrition 135, 938944.
  • Goebel FD (2005) Immune reconstitution inflammatory syndrome (IRIS) – another new disease entity following treatment initiation of HIV infection. Infection 33, 4345.
  • Guerrant RL, Lima AA & Davidson F (2000) Micronutrients and infection: interactions and implications with enteric and other infections and future priorities. Journal of Infectious Diseases 182(Suppl. 1), S134S138.
  • Hanekom WA, Potgieter S, Hughes EJ et al. (1997) Vitamin A status and therapy in childhood pulmonary tuberculosis. Journal of Pediatrics 131, 925927.
  • Hodkinson CF, Kelly M, Alexander HD et al. (2007) Effect of zinc supplementation on the immune status of healthy older individuals aged 55-70 years: the ZENITH Study. Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 62, 598608.
  • Irlam JH, Visser ME, Rollins N & Siegfried N (2005) Micronutrient supplementation in children and adults with HIV infection. Cochrane Database of Systematic Reviews 4, CD003650.
  • Karyadi E, Schultink W, Nelwan RH et al. (2000) Poor micronutrient status of active pulmonary tuberculosis patients in Indonesia. Journal of Nutrition 130, 29532958.
  • Karyadi E, West CE, Schultink W et al. (2002) A double-blind, placebo-controlled study of vitamin A and zinc supplementation in persons with tuberculosis in Indonesia: effects on clinical response and nutritional status. American Journal of Clinical Nutrition 75, 720727.
  • Kupka R & Fawzi W (2002) Zinc nutrition and HIV infection. Nutrition Reviews 60, 6979.
  • Lawson L, Yassin MA, Ramsay A et al. (2006) Microbiological validation of smear microscopy after sputum digestion with bleach; a step closer to a one-stop diagnosis of pulmonary tuberculosis. Tuberculosis (Edinb) 86, 3440.
  • Lawson L, Yassin MA, Thacher TD et al. (2008) Clinical presentation of adults with pulmonary tuberculosis with and without HIV infection in Nigeria. Scandinavian Journal of Infectious Diseases 40, 3035.
  • Van Lettow M, Fawzi WW & Semba RD (2003) Triple trouble: the role of malnutrition in tuberculosis and human immunodeficiency virus co-infection. Nutrition Reviews 61, 8190.
  • Van Lettow M, Kumwenda JJ, Harries AD et al. (2004) Malnutrition and the severity of lung disease in adults with pulmonary tuberculosis in Malawi. The International Journal of Tuberculosis and Lung Disease 8, 211217.
  • Maziya-Dixon B, Akinyele IO, Oguntona EB et al. (2004) Nigeria Food Consumption and Nutrition Survey 2001–2003 Summary. International Institute of Tropical Agriculture, Ibadan.
  • Maziya-Dixon BB, Akinyele IO, Sanusi RA et al. (2006) Vitamin A deficiency is prevalent in children less than 5 y of age in Nigeria. Journal of Nutrition 136, 22552261.
  • Mugusi FM, Rusizoka O, Habib N & Fawzi W (2003) Vitamin A status of patients presenting with pulmonary tuberculosis and asymptomatic HIV-infected individuals, Dar es Salaam, Tanzania. The International Journal of Tuberculosis and Lung Disease 7, 804807.
  • Onwubalili JK, Nwosu CM & Onuigbo WI (1988) Deaths from AIDS-like illnesses in west Africans. East African Medical Journal 65, 867873.
  • Pant K, Biswas SK, Chawla R, Shah A & Singh MM (1987) Zinc in active pulmonary tuberculosis. Indian Journal of Chest Diseases and Allied Sciences 29, 144149.
  • Prasad AS (2007) Zinc: mechanisms of host defense. Journal of Nutrition 137, 13451349.
  • Range N, Andersen AB, Magnussen P, Mugomela A & Friis H (2005) The effect of micronutrient supplementation on treatment outcome in patients with pulmonary tuberculosis: a randomized controlled trial in Mwanza, Tanzania. Tropical Medicine and International Health 10, 826832.
  • Range N, Changalucha J, Krarup H et al. (2006) The effect of multi-vitamin/mineral supplementation on mortality during treatment of pulmonary tuberculosis: a randomised two-by-two factorial trial in Mwanza, Tanzania. British Journal of Nutrition 95, 762770.
  • Rwangabwoba JM, Fischman H & Semba RD (1998) Serum vitamin A levels during tuberculosis and human immunodeficiency virus infection. The International Journal of Tuberculosis and Lung Disease 2, 771773.
  • Saha K & Rao KN (1989) Undernutrition in lepromatous leprosy. V. Severe nutritional deficit in lepromatous patients co-infected with pulmonary tuberculosis. European Journal of Clinical Nutrition 43, 117128.
  • Scrimshaw NS (1995) The new paradigm of public health nutrition. American Journal of Public Health 85, 622624.
  • Semba RD (1999) Vitamin A and immunity to viral, bacterial and protozoan infections. Proceedings of the Nutrition Society 58, 719727.
  • Semba RD, Kumwenda J, Zijlstra E et al. (2007) Micronutrient supplements and mortality of HIV-infected adults with pulmonary TB: a controlled clinical trial. The International Journal of Tuberculosis and Lung Disease 11, 854859.
  • Simon G (1966) Radiology in epidemiological studies and some therapeutic trials. British Medical Journal 2, 491494.
  • Smith JC Jr (1980) The vitamin A-zinc connection: a review. Annals of the New York Academy of Sciences 355, 6275.
  • Thurnham DI, Northrop-Clewes CA, McCullough FS, Das BS & Lunn PG (2000) Innate immunity, gut integrity, and vitamin A in Gambian and Indian infants. Journal of Infectious Diseases 182(Suppl. 1), S23S28.
  • Villamor E, Mugusi F, Urassa W et al. (2008) A trial of the effect of micronutrient supplementation on treatment outcome, T cell counts, morbidity, and mortality in adults with pulmonary tuberculosis. Journal of Infectious Diseases 197, 14991505.
  • Wellinghausen N, Martin M & Rink L (1997) Zinc inhibits interleukin-1-dependent T cell stimulation. European Journal of Immunology 27, 25292535.
  • Williams IO, Eka OU & Essien EU (2008) Vitamin A status of pregnant women in Calabar metropolis, Nigeria. Pakistan Journal of Biological Sciences 11, 17021707.
  • Wiysonge C, Shey M, Sterne J, Brocklehurst P & Wiysonge CS (2005) Vitamin A supplementation for reducing the risk of mother-to-child transmission of HIV infection. Cochrane Database of Systematic Reviews 4, CD003648.
  • World Health Organization (2009) Global Tuberculosis Control: A Short Update to the 2009 Report. WHO, Geneva.