SEARCH

SEARCH BY CITATION

Keywords:

  • tuberculosis;
  • randomization;
  • micronutrients;
  • human immunodeficiency virus;
  • Tanzania

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Objective  The aim of the study was to assess the effects of micronutrient supplementation on culture conversion in tuberculosis (TB) patients.

Design  The study was a randomized, double-blind placebo-controlled 2 × 2 trial of zinc and multi-micronutrient (MMN) supplementation in pulmonary TB patients in Tanzania.

Results  A total of 499 pulmonary TB patients were included in the trial after being confirmed sputum-positive by microscopy or culture. At 8 weeks, 25% were sputum-smear positive but only 11% were culture-positive (P < 0.0001). No significant differences were observed in culture conversion rate among those allocated to MMN or placebo (89.5 vs. 86.2%, P = 0.29) at 8 weeks, although at week 4 those allocated to MMN had a slightly reduced culture conversion rate (42.8 vs. 52.8%, P = 0.058). Zinc had no effects on culture conversion. MMN increased weight gain by 0.78 kg [95% confidence interval (CI): 0.12–1.43] at week 8, while zinc supplementation had no effect. The effects of MMN and zinc did not interact and neither MMN nor zinc interacted with human immunodeficiency virus status, sex and culture-intensity at baseline.

Conclusion  Neither zinc nor MMN supplementation had significant effects on culture conversion, but MMN supplementation increased weight gain in TB patients.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Despite effective tuberculosis (TB) chemotherapy, a treatment period of 6–8 months is required for cure. Morbidity and mortality are still high, particularly in those co-infected with human immunodeficiency virus (HIV) (Small et al. 1991; Kassim et al. 1995; Kang'ombe et al. 2000). In addition, treatment outcome depends on a well-functioning cell-mediated immune system. Several studies have reported a number of specific micronutrient deficiencies in TB patients (Ray et al. 1998; Rwangabwoba et al. 1998; Karyadi et al. 2000, 2002; van Lettow et al. 2003; Mugusi et al. 2003), which may impair the immune system and hence affect response to chemotherapy.

We conducted a randomized trial among TB patients found pulmonary smear or culture-positive (PTB+), to assess the effects of multi-micronutrient (MMN) and zinc supplementation on culture conversion, i.e. changing from being sputum-smear- or culture-positive to culture-negative.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Study area and population

Local residents aged 15 years or older with newly diagnosed TB were enrolled from five health facility centres in Mwanza and Magu districts in Tanzania between August 2001 and July 2002. TB patients returning to treatment after defaulting or smear-positive failure cases (patients who remain smear positive after 5 months treatment), those with clinical evidence of renal, hepatic or cardiovascular disease, pregnant women and patients with advanced respiratory or other diseases considered unlikely to survive were excluded. TB Clinical Officers collected demographic, clinical and anthropometrical data. Diagnosis and treatment of TB followed recommended standard procedures (Rieder et al. 1998; WHO 2003) and the study was conducted within the framework of the National TB Programme.

Baseline examinations

Each patient suspected for TB submitted three sputum specimens (Bwire et al. 1999) for microscopic examination at the centre, using Ziehl–Neelsen (ZN) staining technique (Githui et al. 1993; Heifets & Good 1994). Those found smear-positive (PTB+), before start of treatment, produced a morning sputum specimen in a sterile universal bottle for Auramine O staining and culture at the Zonal TB Reference Laboratory (ZTRL). Culture was performed on Lowenstein–Jensen (LJ) solid media (Githui et al. 1993; Heifets & Good 1994). At baseline, helminth infections were determined and questionnaires were administered to elicit demographic information and smoking habits. Anthropometrical measurements were taken at baseline and at 8 weeks.

Randomization and study intervention

Patients diagnosed at the study centres as PTB+ were randomized. The MMN trial included only those with either smear and/or culture-positive results at the ZTRL. Those diagnosed at the study centres as PTB+ but found to be both smear- and culture-negative at the ZTRL were excluded. Enrolment, randomization and trial participants’ profile are shown in Figure 1.

image

Figure 1. Trial profile. Sputum positive pulmonary tuberculosis patients were randomized to daily supplementation with zinc or placebo and, independently, to multi-micronutrients (MMN) or placebo. Those found sputum negative at reference laboratory after randomization were excluded.

Download figure to PowerPoint

Two allocation sequences, using blocks of four, were created using computer-generated random numbers. Study numbers were then assigned consecutively to the generated sequence numbers.

Multi-micronutrient and placebo, and zinc and placebo tablets were manufactured by Almega (Ringsted, Denmark). Placebo tablets were identical in colour, shape and size. Zinc tablets contained 45 mg of elementary zinc. MMN tablets contained: vitamin A 5000 IU, vitamin B1 20 mg, vitamin B2 20 mg, vitamin B6 25 mg, vitamin B12 50 μg, folic acid 0.8 mg, niacin 40 mg, vitamin C 200 mg, vitamin E 60 mg, vitamin D3 200 IU, selenium 0.2 mg and copper 5 mg. The composition of the MMN supplement was based on the high micronutrient requirements of patients with TB and similar to the supplement given to pregnant HIV-infected women in Dar es Salaam (Fawzi et al. 2004), except that it contained more vitamin E (60 instead of 30 mg), less vitamin C (200 instead of 1000 mg) and was added vitamin D and preformed vitamin A, as well as selenium and copper.

Tablets were packed in identical containers and labelled with four different alphabetical letters from the factory, which were then replaced with study serial numbers from 1 to 550 based on the computer-generated random sequence. Each PTB+ patient was assigned two containers containing equal quantities of tablets, labelled with the same study identification number and kept at the health facilities. Codes for micronutrient and zinc remained sealed until the completion of data analysis.

Treatment and micronutrient supplementation

All PTB+ patients were treated with anti-TB short-course regimen according to the WHO TB treatment guidelines (WHO 2003). During the first 2 months, four different types of anti-TB drugs were taken daily at the health facility under supervision of a health staff, i.e. directly observed short-course treatment. During the 6-month continuation phase, patients collected drugs on monthly basis for self-administering at home. Anti-TB drugs were provided by the Ministry of Health and only first-line drugs were used.

Along with anti-TB treatment, PTB+ patients took two tablets containing MMN or placebo and zinc or placebo every day. Patients with haemoglobin <100 g/l or helminthic infections were treated.

Outcome measures

The primary outcome was culture conversion at weeks 2, 4 and 8. As mycobacteria detected by microscopy during treatment are often dead, sputum smear conversion was not a primary outcome.

Sample size requirements

It was assumed that half of the TB patients would become culture-negative after 4 weeks. In order to detect a 40% increase in the culture conversion rate, with 80% power and 95% confidence, while allowing for 20% loss to follow-up, 125 patients were needed in each group. By enrolling 500 patients, using 2 × 2 factorial design and assuming no interaction between zinc and MMN, it would be possible to detect a 30% increase in the culture conversion rate. Similarly, given a mean of 5 kg (SD 1.5) increase in weight, we would be able to detect a 0.4 kg difference in weight gain.

Statistical analysis

Data were analysed in SPSS for windows version 10.0. Two-sample t-tests and chi-square test were used to compare means and proportions between groups, respectively. Multiple logistic regression analyses were used to estimate the effects of the intervention, while controlling for potential confounders. Intention-to-treat analysis was used.

Ethical consideration

The study was approved by the Ethics Committee of the National Institute for Medical Research in Tanzania and recommended by the Danish Central Medical Ethics Committee. Oral consent was obtained from all study patients before inclusion. Pre-test HIV counselling was mandatory; post-test counselling was given to those who wanted to know their HIV results.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Of the 530 patients diagnosed as PTB+ at the enrolment centres and randomized, only 499 (94.2%) were confirmed positive by microscopy or culture at the ZTRL and included in the trial (Figure 1). Of these 499, 454 (91%) were alive at the end of 2-month follow-up and continued with treatment, 13 (2.6%) had died and 32 (6.4%) were lost to follow-up. Of the 13 deaths, seven occurred among those allocated to zinc and six to those allocated to placebo and five were among those allocated to MMN and eight to placebo (P > 0.10). Of the 32 patients lost to follow-up, 10 were in the zinc and 22 in the placebo group (P = 0.04), 19 were in the MMN and 13 in the placebo group (P > 0.10).

Baseline characteristics

Of the 499, mean age was 35.4 (SD 12.3) years and 205 (41.1%) were females. Wasukuma were the dominant ethnic group (51.3%) and 43% were predominantly farmers. HIV, Schistosoma mansoni and hookworm infections were relatively common, whereas malaria was less common. Over 95% of the trial participants were sputum-smear-positive with 50% having heavy acid-fast bacilli intensity and 96.9% were culture-positive with 64% having heavy growth intensity. As seen in Table 1, randomization resulted in baseline equivalence with respect to baseline intensity of TB infection, as well as other potential risk factors of conversion, although the prevalence of present smokers was higher among those allocated to zinc compared with placebo.

Table 1.  Baseline characteristics of the 499 PTB+ patients randomized to MMN or placebo, and zinc or placebo, supplementation
 nMMN interventionZinc intervention
Placebo (n = 248)MMN (n = 251)Placebo (n = 253)Zinc (n = 246)
  1. PTB+, pulmonary tuberculosis-positive; MMN, multi-micronutrient.

  2. * Negative: no acid–fast bacilli (AFB); low: 1–100 (AFB)/100 fields; moderate: 1–10 AFB/field; heavy: >10 AFB/field.

  3. † Negative: no growth; low: 1–100 colonies; moderate: >100 non-confluent colonies; heavy: confluent growth.

Age [mean (SD)]49935.3 (12.3)35.5 (12.2)35.1 (11.7)35.7 (12.8)
Female sex [n (%)]205108 (43.5)97 (38.6)112 (44.3)93 (37.8)
Weight (kg) [mean (SD)]49950.5 (8.7)50.4 (7.6)50.8 (8.6)50.0 (7.7)
Ethnicity [n (%)]
 Wasukuma256125 (50.4)131 (52.2)133 (52.6)123 (50.0)
 Other tribes243123 (49.6)120 (47.8)120 (47.4)123 (50.0)
Occupation [n (%)]
 Farmer214105 (42.3)109 (43.4)111 (43.9)103 (41.9)
 Business11046 (18.5)64 (25.5)50 (19.8)60 (24.42)
 Employed9343 (17.3)50 (19.9)44 (17.4)49 (19.9)
 Others8254 (21.8)28 (11.2)48 (19.0)34 (13.8)
Infections [n (%)]
 HIV-infected213106 (42.7)107 (42.6)107 (42.3)106 (43.1)
 Malaria239 (3.6)14 (5.6)12 (4.7)11 (4.5)
 Schistosoma mansoni17390 (36.4)83 (33.1)79 (31.3)94 (38.2)
 Hookworm10151 (20.6)50 (19.9)49 (19.4)52 (21.1)
Smoking [n (%)]
 Past9648 (19.4)48 (19.1)50 (19.8)46 (18.7)
 Present5726 (10.5)31 (12.4)21 (8.0)36 (14.6)
Smear intensity [n (%)]*
 Negative2713 (5.3)14 (5.6)15 (6.0)12 (4.9)
 Low13559 (24.0)76 (30.3)67 (26.6)68 (27.8)
 Moderate9852 (21.1)46 (18.3)57 (22.6)41 (16.7)
 Heavy237122 (49.6)115 (45.8)113 (44.8)124 (50.6)
Culture intensity [n (%)]†
 Negative75 (2.1)2 (0.8)2 (0.8)5 (2.1)
 Low10946 (19.5)63 (25.4)25 (21.0)44 (18.3)
 Moderate6226 (11.0)36 (14.5)19 (16.0)31 (12.9)
 Heavy306159 (67.4)147 (59.3)75 (63.0)160 (66.7)

Smear and culture conversion

Comparison of smear- and culture-positive cases from week 0 to 8 is shown in Figure 2. As seen, the proportion found positive was slightly higher with respect to culture compared with smear after 2 weeks (71.6%vs. 63.7%), but not after 4 weeks. At week 8, 25% of the baseline-positive smears, but only 11% of the positive cultures were still positive (P < 0.0001), suggesting that the majority of bacilli observed by smear were dead.

image

Figure 2. Comparison of sputum-smear and culture-positive patients at 0, 2, 4 and 8 weeks.

Download figure to PowerPoint

At week 4, the proportion converted to culture-negative was slightly higher in the placebo than in the MMN group (52.4 vs. 42.8%), but the difference was not statistically significant (P = 0.058) (Table 2). There were no differences in culture conversion between the zinc and the placebo groups.

Table 2.  Culture results at 2, 4 and 8 weeks among 499 PTB+ patients randomly allocated to MMN or placebo, and zinc or placebo, supplementation
ResultsnMMN interventionZinc intervention
PlaceboMNP-valuePlaceboZincP-value
  1. MMN, multi-micronutrient; PTB+, pulmonary tuberculosis-positive.

  2. All values refer to numbers (%).

  3. Those found culture-negative were considered to have culture converted.

Week 2 [n (%)]4111192120.712102010.21
 Negative6932 (16.1)37 (17.5)40 (19.0)29 (14.4)
 Positive342167 (83.9)175 (82.5)170 (81.0)172 (85.6)
Week 4 [n (%)]3921912010.0582021900.86
 Negative186100 (52.4)86 (42.8)95 (47.0)91 (47.9)
 Positive20691 (47.6)115 (57.2)107 (53.0)99 (52.1)
Week 8 [n (%)]4292102190.292182110.64
 Negative377181 (86.2)196 (89.5)190 (87.2)187 (88.6)
 Positive5229 (13.8)23 (10.5)28 (12.8)24 (11.4)

Logistic regression analyses with culture conversion as a dependent variable were performed to test for interaction and adjust for confounding. There was no interaction between MMN and zinc supplementation (P > 0.10). Similarly, the effects of the interventions on culture conversion did not differ by HIV status, sex, occupation, ethnicity and study centres (interaction, P > 0.10). As seen in Table 3, adjusting for baseline differences in HIV status, sex, culture intensity and smoking had little effect on the estimated effects on culture conversion. The adjusted odds ratio of MMN supplementation was 0.51 (95% CI: 0.33–0.79) at week 4, which corresponds to a reduction in the probability of culture conversion among those receiving MMN compared with placebo. Nevertheless, at 8 weeks, there was no difference.

Table 3.  Adjusted odds ratios (OR) with 95% confidence intervals (95% CI) for the effect of zinc and multi-micronutrient supplementation on culture conversion at 2, 4 and 8 weeks after start of TB treatment*
VariablesWeek 2Week 4Week 8
nOR95% CIP-valuenOR95% CIP-valuenOR95% CIP-value
  1. * Adjusted for HIV status, sex, smoking and baseline culture intensity.

  2. Week 2: n = 416; week 4: n = 392; week 8: n = 429.

Supplements
 Zinc2060.890.57–1.390.612350.850.55–1.320.622200.780.43–1.430.43
 Multi-micronutrients2160.840.54–1.300.432460.510.33–0.790.0032231.280.70–2.320.42

Weight gain

Those allocated to MMN gained 0.78 kg (95% CI: 0.12–1.43, P = 0.02) more weight than those allocated to placebo. The weight gain did not differ between those allocated to zinc and placebo (0.26  kg, 95% CI: −0.40 to 0.92, P = 0.44). The effect of MMN and zinc on weight gain did not differ between HIV infected and uninfected (interaction, P > 0.10). Furthermore, the effect was independent on baseline culture intensity and culture conversion at 4 and 8 weeks (interaction, P > 0.10).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Culture conversion is the major surrogate marker of the efficacy of anti-TB chemotherapy (Liu et al. 1999). The results of our study were not similar to those of an Indonesian study (Karyadi et al. 2002), where vitamin A and zinc supplements were given daily to 80 newly diagnosed TB patients. Sputum smear conversion and resolution of X-ray lesion area were reported to have occurred earlier in the zinc and vitamin A group compared with placebo. In contrast, we did not find significant effects of MMN or zinc supplementation on culture conversion. The different findings in the Indonesian study and our study cannot be attributed to HIV infection, as we found no effect neither among HIV infected nor uninfected. Furthermore, in contrast to the Indonesian study, we found evidence of micronutrient deficiencies in the study population, in that MMN resulted in increased weight gain. We also used culture rather than sputum conversion as our primary outcome. Culture conversion is a more reliable outcome, because those with no sputum conversion did not excrete viable bacilli. Our study had considerable power and loss to follow-up at week 8 was modest and similar between groups.

Most of our patients excreted live bacilli after 2 weeks, but not after 8 weeks. Similarly, a study in Madagascar reported that >50% of the bacilli observed on smear microscopy at 2 months were not viable (Ramarokoto et al. 2002). However, at week 4, although most patients had become smear negative, over 90% of those detected by smear were still culture-positive. While culture is not routinely performed under national TB programmes, it is generally assumed that TB treatment leads to early non-infectious states. Thus our findings provide important information on the level of infectivity for PTB+ patients during treatment period.

Faster culture conversion among HIV-infected patients has been reported from Kenya (Brindle et al. 1993), but not from studies in Uganda and in the USA (Telzak et al. 1997). The fact that HIV-infected individuals tend to harbour fewer mycobacteria compared with the uninfected, because of fewer cavities (Brindle et al. 1993; Ellner 1997) may explain these findings.

Weight gain

Despite lack of effect on culture conversion, MMN confers beneficial effect in terms of nutritional improvement. The 0.8 kg increase in weight gain because of MMN supplementation is likely to be important to TB patients, as they usually have considerable weight loss at the time of diagnosis. Importantly, the effect of MMN was independent of HIV status and culture conversion. Thus, the slight and transiently lower culture conversion among those allocated to MMN was not accompanied by a lower weight gain.

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Neither MMN nor zinc supplementation was associated with increased sputum-smear or culture conversion, but MMN supplementation was associated with increase in weight gain. Whether the weight gain is of importance to treatment response and survival of TB patients remains to be seen.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

We thank the patients for consenting to participate in the trial, the TB district coordinators and nurses for enrolling patients and laboratory technicians for performing smear and culture analyses. We thank the National Institute for Medical Research for granting study approval. The study was funded by DANIDA through the DBL-Institute for Health Research and Development.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  • Brindle RJ, Nunn PP, Githui W et al. (1993) Quantitative bacillary response to treatment in HIV-associated pulmonary tuberculosis. American Review of Respiratory Disease 147, 958961.
  • Bwire R, Borgdorff MW, Sticht-Groh V et al. (1999) Tuberculosis chemotherapy and sputum conversion among HIV-seropositive and HIV-seronegative patients in south-eastern Uganda. East African Medical Journal 76, 307313.
  • Ellner JJ (1997) Review: the immune response in human tuberculosis – implications for tuberculosis control. Journal of Infectious Diseases 176, 13511359.
  • Fawzi WW, Msamanga GI, Spiegelman D et al. (2004) A randomized trial of multivitamin supplements and HIV disease progression and mortality. New England Journal of Medicine 351, 2332.
  • Githui W, Kitui F, Juma ES et al. (1993) A comparative study on the reliability of the fluorescence microscopy and Ziehl–Neelsen method in the diagnosis of pulmonary tuberculosis. East African Medical Journal 70, 263266.
  • Heifets LB & Good RC (1994) Current laboratory methods for the diagnosis of tuberculosis. In: Tuberculosis: Pathogenesis, Protection and Control (ed. BloomBR) American Society for Microbiology, Washington, DC, pp. 85110.
  • Kang'ombe C, Harries AD, Banda H et al. (2000) High mortality rates in tuberculosis patients in Zomba Hospital, Malawi, during 32 months of follow-up. Transactions of the Royal Society of Tropical Medicine and Hygiene 94, 305309.
  • 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.
  • Kassim S, Sassan-Morokro M, Ackah A et al. (1995) Two-year follow-up of persons with HIV-1- and HIV-2-associated pulmonary tuberculosis treated with short-course chemotherapy in West Africa. AIDS 9, 11851191.
  • 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.
  • Liu Z, Shilkret KL & Ellis HM (1999) Predictors of sputum culture conversion among patients with tuberculosis in the era of tuberculosis resurgence. Archives of Internal Medicine 159, 11101116.
  • 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. International Journal of Tuberculosis and Lung Disease 7, 804807.
  • Ramarokoto H, Randriamiharisoa H, Rakotoarisaonina A et al. (2002) Bacteriological follow-up of tuberculosis treatment: a comparative study of smear microscopy and culture results at the second month of treatment. International Journal of Tuberculosis and Lung Disease 6, 909912.
  • Ray M, Kumar L & Prasad R (1998) Plasma zinc status in Indian childhood tuberculosis: impact of antituberculosis therapy. International Journal of Tuberculosis and Lung Disease 2, 719725.
  • Rieder HL, Chonde TM, Myking H et al. (1998) The public health service national tuberculosis reference laboratory and the national laboratory network. Minimum requirements, role and operation in a low-income country, International Union Against Tuberculosis and Lung Disease. http://www.iuatld.org/pdf/en/guides_publications/public_health_service.pdf .
  • Rwangabwoba JM, Fischman H & Semba RD (1998) Serum vitamin A levels during tuberculosis and human immunodeficiency virus infection. International Journal of Tuberculosis and Lung Disease 2, 771773.
  • Small PM, Schecter GF, Goodman PC et al. (1991) Treatment of tuberculosis in patients with advanced human immunodeficiency virus infection. New England Journal of Medicine 324, 289294.
  • Telzak EE, Fazal BA, Pollard CL et al. (1997) Factors influencing time to sputum conversion among patients with smear-positive pulmonary tuberculosis. Clinical Infectious Diseases 25, 666670.
  • World Health Organization (2003) Treatment of Tuberculosis: Guidelines for National Programmes, Edn. 3. WHO, Geneva.