Micronutrient supplementation in children and adults with HIV infection

  • Review
  • Intervention

Authors


Abstract

Background

Micronutrient deficiencies are widespread and compound the effects of HIV disease; micronutrient supplements may be effective and safe in reducing this burden.

Objectives

To assess whether micronutrient supplements are effective and safe in reducing mortality and morbidity in adults and children with HIV infection.

Search methods

The CENTRAL, EMBASE, PubMed, and GATEWAY databases were searched for randomised controlled trials of micronutrient supplements using the search methods of the Cochrane HIV/AIDS Group.

Selection criteria

Randomised controlled trials were selected that compared the effects of micronutrient supplements (vitamins, trace elements, and combinations of these) with other supplements, placebo or no treatment on mortality, morbidity, pregnancy outcomes, immunologic indicators, and anthropometric measures in HIV-infected adults and children. Any adverse effects of supplementation were recorded.

Data collection and analysis

Two reviewers independently selected trials, appraised trial quality for risk of bias using standardised criteria, and extracted data using standardised forms.

Main results

Sixteen additional trials are included in this update to the original Cochrane review (Irlam 2005). Overall, 30 trials involving 22 120 participants are reviewed: 20 trials of single supplements (vitamin A, vitamin D, zinc, selenium) and 10 of multiple micronutrients. Eight trials were undertaken in child populations.

None of the six trials of vitamin A or beta-carotene supplementation in adults demonstrated any significant reduction in HIV disease progression. Vitamin A halved all-cause mortality in a meta-analysis of three trials in African children, had inconsistent impacts on diarrhoeal and respiratory morbidity, and improved short-term growth in one trial. No significant adverse effects of vitamin A in adults or children have been reported.

Zinc supplements reduced diarrhoeal morbidity and had no adverse effects on disease progression in a single safety trial in South African children. No significant clinical benefits were found from zinc supplementation of pregnant Tanzanian women or Peruvian adults with persistent diarrhoea.

Selenium reduced diarrhoeal morbidity in pregnant women in Tanzania, and reduced viral load in two separate small trials in American adults.

Single trials of vitamin D supplements in adults, and in adolescents and children, demonstrated safety but no clinical benefits.

Multiple micronutrient supplements conferred multiple clinical benefits to pregnant women and their offspring in a large Tanzanian trial. Supplementation in another Tanzanian trial reduced the recurrence of pulmonary TB and increased weight gain in co-infected patients. No significant adverse effects were reported.

Authors' conclusions

Multiple micronutrient supplements reduced morbidity and mortality in HIV-infected pregnant women and their offspring and also improved early child growth in one large randomised controlled trial in Africa. Additional research is needed to determine if these are generalisable findings. Vitamin A supplementation is beneficial and safe in HIV-infected children, but further evidence is needed to establish if supplementation confers similar benefits in HIV-infected adults. Zinc is safe in HIV-infected adults and children. It may have similar benefits in HIV-infected children and adults, and uninfected children with diarrhoea, as it does in HIV-uninfected children.

Further trials of single supplements (vitamin D, zinc, and selenium) are required to build the evidence base. The long-term clinical benefits, adverse effects, and optimal formulation of multiple micronutrient supplements require further investigation in individuals with diverse disease status.

 

Resumen

Antecedentes

Suplementos de micronutrientes para niños y adultos con infección por VIH

Las carencias de micronutrientes son generalizadas y acrecientan los efectos de la enfermedad por VIH; los suplementos de micronutrientes podrían ser efectivos y seguros para aliviar dichas complicaciones.

Objetivos

Evaluar si los suplementos de micronutrientes son efectivos y seguros en la reducción de la morbimortalidad en niños y adultos con infección por VIH.

Estrategia de búsqueda

Se realizaron búsquedas en las bases de datos de CENTRAL, EMBASE, PubMed, y GATEWAY para hallar ensayos controlados con asignación aleatoria de suplementos de micronutrientes con los métodos de búsqueda del Grupo Cochrane de SIDA/VIH (Cochrane HIV/AIDS Group).

Criterios de selección

Se seleccionaron los ensayos controlados con asignación aleatoria que compararon los efectos de los suplementos de micronutrientes (vitaminas, oligoelementos, y sus combinaciones) con otros suplementos, placebo o ningún tratamiento en la mortalidad, la morbilidad, los desenlaces del embarazo, los indicadores inmunológicos y las medidas antropométricas en niños y adultos infectados por VIH. Se registraron todos los efectos adversos de los suplementos.

Obtención y análisis de los datos

Dos revisores, de forma independiente, seleccionaron y estimaron la calidad de los ensayos con respecto al riesgo de sesgo según los criterios estandarizados, y extrajeron los datos con formularios estandarizados.

Resultados principales

Se incluyen 16 ensayos adicionales en esta actualización de la revisión Cochrane original (Irlam 2005). En total se analizaron 30 ensayos con 22 120 participantes: veinte ensayos de suplementos simples (vitamina A, vitamina D, cinc, selenio) y diez de múltiples micronutrientes. Se realizaron ocho ensayos en poblaciones de niños.

Ninguno de los seis ensayos de suplementos de vitamina A o betacarotenos en adultos demostró una reducción significativa en la progresión de la enfermedad por VIH. La vitamina A redujo a la mitad la mortalidad por todas las causas en un metanálisis de tres ensayos en niños africanos, tuvo consecuencias variadas en la diarrea y en la morbilidad respiratoria; y, en un ensayo, mejoró el crecimiento a corto plazo. No se informaron efectos adversos significativos de la vitamina A en adultos ni en niños.

En un único ensayo de seguridad en niños sudafricanos, los suplementos de zinc redujeron la morbilidad por diarrea y no tuvieron efectos adversos en la progresión de la enfermedad. No se hallaron beneficios clínicos con los suplementos de zinc en pacientes de Tanzania embarazadas ni en peruanas adultas con diarrea persistente.

El selenio redujo la morbilidad por diarrea en embarazadas de Tanzania y disminuyó la carga vírica en dos ensayos pequeños de estadounidenses adultos.

Los ensayos simples de suplementos de vitamina D en adultos, y en adolescentes y niños, demostraron la seguridad pero no hallaron beneficios clínicos.

En un ensayo amplio de Tanzania, los suplementos con múltiples micronutrientes proporcionaron numerosos beneficios clínicos a las embarazadas y sus hijos. En otro ensayo tanzano, los suplementos redujeron la recurrencia de tuberculosis pulmonar e incrementaron el aumento de peso en pacientes con coinfección. No se informaron efectos adversos significativos.

Conclusiones de los autores

En un ensayo controlado con asignación aleatoria amplio africano, los suplementos de múltiples micronutrientes redujeron la morbimortalidad en embarazadas infectadas por VIH y sus hijos, además mejoraron el crecimiento infantil precoz. Se necesita investigación adicional para determinar si estos hallazgos pueden generalizarse. Los suplementos de vitamina A son beneficiosos y seguros en niños infectos por VIH, aunque se necesitan más pruebas para establecer si los suplementos otorgan beneficios similares en adultos infectados por VIH. El cinc es seguro en niños y adultos infectados por VIH. Tal vez posea beneficios similares en niños y adultos infectados por VIH, y en niños con diarrea no infectados, como sucede con los niños no infectados por VIH.

Se necesitan más ensayos de suplementos simples (vitamina D, cinc y selenio) para construir la base de pruebas. Se necesita mayor investigación en individuos con diversos estados de la enfermedad acerca de los beneficios clínicos a largo plazo, los efectos adversos y la fórmula óptima de los suplementos con múltiples micronutrientes.

Traducción

Traducción realizada por el Centro Cochrane Iberoamericano

Résumé scientifique

Supplémentation en micronutriments chez les enfants et les adultes atteints du VIH

Contexte

Les déficiences en micronutriments sont largement répandues et font partie des effets de la maladie du VIH ; les suppléments en micronutriments pourraient être efficaces et sûrs pour réduire cette comorbidité.

Objectifs

Évaluer si les suppléments en micronutriments sont efficaces et sûrs dans la réduction de la mortalité et la morbidité chez les adultes et les enfants atteints du VIH.

Stratégie de recherche documentaire

Nous avons examiné les bases de données CENTRAL, EMBASE, PubMed et GATEWAY à la recherche d'essais contrôlés randomisés sur les suppléments de micronutriments à l'aide des méthodes de recherche du groupe VIH/SIDA Cochrane.

Critères de sélection

Des essais contrôlés randomisés ont été sélectionnés qui comparaient les effets des suppléments en micronutriments (vitamines, oligoéléments et leurs combinaisons) à d'autres suppléments, un placebo ou l'absence de traitement, sur la mortalité, la morbidité, la grossesse, les indicateurs immunologiques et les mesures anthropométriques chez les adultes et les enfants atteints du VIH. Tout effet indésirable de la supplémentation était répertorié.

Recueil et analyse des données

Deux évaluateurs ont indépendamment sélectionné les essais, évalué leur qualité méthodologique et leurs risques de biais à l'aide de critères standardisés et extrait des données à l'aide de formulaires standardisés.

Résultats principaux

Seize essais additionnels ont été inclus dans cette mise à jour de la revue systématique Cochrane originale (Irlam 2005). En tout, 30 essais impliquant 22 120 participants ont été évalués : 20 essais de suppléments simples (vitamine A, vitamine D, zinc, sélénium) et 10 essais de plusieurs micronutriments. Huit essais ont été réalisés chez des populations d'enfants.

Aucun de ces six essais de supplémentation en vitamine A ou bêta-carotène chez les adultes n'a montré de réduction significative de la progression du VIH. La vitamine A a réduit de moitié la mortalité toutes causes confondues dans une méta-analyse de trois essais sur des enfants africains, a eu des effets incohérents sur la morbidité respiratoire et diarrhéique, et a amélioré la croissance à court terme dans un essai. Aucun effet indésirable significatif de la vitamine A chez les adultes ou les enfants n'a été reporté.

Les suppléments en zinc ont réduit la morbidité diarrhéique et n'ont eu aucun effet indésirable sur la progression de la maladie lors d'un essai d’innocuité réalisé sur des enfants sud-africains. Aucun effet bénéfique clinique significatif n'a été trouvé suite à la supplémentation en zinc de femmes tanzaniennes enceintes ou d'adultes péruviens avec une diarrhée persistante.

Le sélénium a réduit la morbidité diarrhéique chez les femmes tanzaniennes enceintes et a réduit la charge virale au cours de deux petits essais séparés réalisés sur des adultes américains.

Des essais uniques de supplémentation en vitamine D chez des adultes, adolescents et enfants, a démontré la sécurité mais aucun bénéfice clinique.

Plusieurs suppléments en micronutriments conféraient différents effets bénéfiques cliniques aux femmes enceintes et à leur progéniture lors d'un large essai conduit en Tanzanie. La supplémentation, lors d'un autre essai tanzanien, a réduit la récurrence de TB pulmonaire et augmenté le gain de poids chez des patients co-infectés. Aucun effet indésirable notable n'a été signalé.

Conclusions des auteurs

Plusieurs suppléments en micronutriments ont réduit la morbidité et la mortalité des femmes enceintes atteintes du VIH et leur progéniture, et ont également amélioré la croissance précoce de l'enfant lors d'un large essai contrôlé randomisé conduit en Afrique. Des recherches supplémentaires doivent être effectuées afin de déterminer si ces résultats peuvent être généralisés. La supplémentation en vitamine A est bénéfique et inoffensive pour les enfants atteints du VIH, mais d'autres preuves sont nécessaires afin de déterminer si la supplémentation peut conférer des bénéfices similaires chez les adultes infectés par le VIH. Le zinc est sûr pour les adultes et enfants atteints du VIH. Il pourrait offrir des bénéfices similaires pour les enfants et adultes atteints du VIH, ainsi que pour les enfants non-atteints mais présentant une diarrhée, comme il l'offre aux enfants non-atteints du VIH.

D'autres essais de suppléments simples (vitamine D, zinc et sélénium) devront être réalisés pour enrichir la base de preuves. Les effets bénéfiques cliniques à long terme, les effets indésirables et la préparation optimale de plusieurs suppléments en micronutriments nécessitent la conduite d'études complémentaires auprès d'individus présentant divers états de la maladie.

 

Resumo

Suplementação de micronutrientes em crianças e adultos infectados pelo HIV

Introdução

As deficiências de micronutrientes são comuns e pioram os efeitos da doença causada pelo HIV; a suplementação de micronutrientes pode ser efetiva e segura na redução desse problema.

Objetivos

Avaliar se a suplementação de micronutrientes é efetiva e segura na redução da mortalidade e morbidade em adultos e crianças infectadas pelo HIV.

Métodos de busca

Foram pesquisados ensaios clínicos randomizados sobre suplementação de micronutrintes nas seguintes bases de dados: CENTRAL, EMBASE, PubMed, e GATEWAY. Os métodos de pesquisa do Grupo Cochrane HIV/AIDS foram utilizados.

Critério de seleção

Foram selecionados os ensaios clínicos randomizados que compararam os efeitos da suplementação de micronutrientes (vitaminas, oligoelementos e combinações destes) com outros suplementos, placebo ou nenhum tratamento sobre a mortalidade, a morbidade, desfechos da gestação, indicadores imunológicos e medidas antropométricas, em adultos e crianças infectados pelo HIV. Qualquer efeito adverso da suplementação foi registrado.

Coleta dos dados e análises

Dois revisores, de modo independente, selecionaram e avaliaram a qualidade dos estudos para o risco de viés utilizando critérios padronizados, e extraíram os dados através de formulários padronizados.

Principais resultados

Dezesseis estudos adicionais foram incluídos nesta atualização da revisão original da Cochrane (Irlam 2005)Ao todo, 30 estudos envolvendo 22.120 participantes foram revisados: 20 estudos sobre a utilização de suplementos isolados (vitamina A, vitamina D, zinco, selênio) e 10 estudos sobre a utilização de micronutrientes combinados. Oito estudos foram realizados em crianças.

Nenhum dos seis estudos sobre suplementação com vitamina A ou betacaroteno em adultos demonstrou redução significativa na progressão da doença causada pelo HIV. A vitamina A diminuiu pela metade todas as causas de mortalidade em uma metanálise, que incluiu três estudos realizados com crianças africanas. A intervenção produziu um efeito inconsistente sobre os desfechos diarreia e morbidade respiratória (1 estudo cada), e melhora no crescimento em curto prazo (1 estudo). Não foram reportados efeitos adversos significativos da vitamina A em adultos ou em crianças.

A suplementação com zinco reduziu a morbidade por diarreia e não apresentou efeitos adversos sobre a progressão da doença em um único estudo sobre segurança clínica, em crianças sul-africanas. Não houve nenhum benefício clínico significativo da suplementação com zinco em grávidas da Tanzânia ou em adultos peruanos com diarreia persistente.

A suplementação com selênio reduziu a morbidade por diarreia em grávidas, na Tanzânia, e reduziu a carga viral em dois estudos diferentes, de pequeno porte, em adultos norte-americanos.

Estudos individuais sobre a suplementação com vitamina D em adultos, adolescentes e em crianças demonstraram segurança, mas nenhum benefício clínico.

A suplementação com micronutrientes combinados produziu múltiplos benefícios clínicos para as grávidas e seus filhos em um grande estudo da Tanzânia. Em outro estudo, também na Tanzânia, a suplementação reduziu a recorrência de tuberculose pulmonar e aumentou o ganho de peso em pacientes coinfectados. Nenhum efeito adverso significativo foi reportado.

Conclusão dos autores

A suplementação com múltiplos micronutrientes reduziu a morbidade e a mortalidade em grávidas HIV-infectadas e em seus filhos, além de melhorar o crescimento precoce da criança em um grande ensaio clínico randomizado realizado na África. Pesquisas adicionais são necessárias para determinar se esses resultados podem ser generalizados. A suplementação com vitamina A é benéfica e segura em crianças infectadas pelo HIV, porém, mais evidências são necessárias para estabelecer se a suplementação também confere benefícios semelhantes em adultos infectados pelo HIV. A suplementação com zinco é segura em adultos e em crianças infectados pelo HIV. Ela pode trazer benefícios semelhantes para crianças e adultos infectados com o HIV, e para crianças não infectadas, mas com diarreia, da mesma maneira que as crianças não infectadas pelo HIV se beneficiam do zinco.

Mais estudos sobre suplementação isolada de micronutrientes (vitamina D, zinco e selênio) são necessários para construir uma base de evidências. Além disso, é necessária uma investigação mais aprofundada sobre os benefícios clínicos no longo prazo, os efeitos adversos e a formulação ideal dos suplementos com múltiplos micronutrientes em pessoas em diversas fases da doença.

Notas de tradução

Tradução do Centro Cochrane do Brasil (Isnard Elman Litvin)

Plain language summary

Micronutrient supplementation for children and adults with HIV infection

Multiple micronutrient supplements offer some benefits and are safe in HIV-infected pregnant women and their offspring. Vitamin A and zinc supplements are beneficial and safe in HIV-exposed and HIV-infected children. Further research is needed to build the evidence base for single supplements in adults and children in diverse settings.

Résumé simplifié

Supplémentation en micronutriments pour les enfants et les adultes atteints du VIH

De multiples suppléments en micronutriments ont des effets bénéfiques et sont sûrs pour les femmes enceintes atteintes du VIH et leur progéniture. Les suppléments en vitamine A et en zinc sont bénéfiques et sûrs pour les enfants exposés et atteints du VIH. Des recherches plus approfondies sont nécessaires afin de construire les bases de preuve quant aux effets des suppléments simples chez les adultes et les enfants dans divers contextes.

Notes de traduction

Traduit par: French Cochrane Centre 1st March, 2013
Traduction financée par: Instituts de Recherche en Sant� du Canada, Minist�re de la Sant� et des Services Sociaux du Qu�bec, Fonds de recherche du Qu�bec-Sant� et Institut National d'Excellence en Sant� et en Services Sociaux

Resumo para leigos

Suplementação de micronutrientes em crianças e adultos infectados pelo HIV

A suplementação com múltiplos micronutrientes oferece alguns benefícios e é segura em gestantes infectadas pelo HIV e em seus filhos. A suplementação com vitamina A e zinco é benéfica e segura tanto em crianças expostas como nas infectadas pelo HIV. Mais pesquisas são necessárias para construir uma base de evidências para a utilização de suplementos únicos em adultos e em crianças, em diversos contextos.

Notas de tradução

Tradução do Centro Cochrane do Brasil (Isnard Elman Litvin)

Background

Description of the condition

The HIV/AIDS pandemic has severely affected sub-Saharan Africa, more than any other part of the world. With about a tenth of the world's population, the region is home to two thirds of all people living with HIV worldwide, an estimated 22.4 million in 2008, 1.8 million of whom are children under 15 years of age.  The estimated prevalence of adult HIV in the region is 5.2% [4.9%; 5.4%] (UNAIDS 2009). It is estimated that HIV-related deaths account for about 14% of all child deaths in southern Africa (Black 2010).

Malnutrition takes many forms, but in sub-Saharan Africa it most commonly refers to inadequate protein and energy intake (protein energy malnutrition or PEM), usually with associated multiple micronutrient insufficiency. Micronutrient deficiencies are common in HIV-infected children and adults, particularly in developing communities where diets are frequently inadequate to meet the recommended daily requirements. They are also more pronounced in individuals with advanced disease, as a consequence of reduced nutrient intake due to AIDS and opportunistic infections, and excessive losses due to diarrhoea, malabsorption, and parasitic infections (ASSAf 2007).

Observational studies have suggested that both PEM and micronutrient deficiencies may hasten the progression of HIV infection, and that HIV worsens malnutrition. HIV infection and malnutrition therefore form a "vicious cycle" of immune dysfunction, infectious disease, and malnutrition (Piwoz 2000; Semba 1999).

Micronutrient supplements are either single or multiple formulations of vitamins and trace elements that have multiple functions including immune regulation and facilitating the body’s utilisation of the macronutrients (carbohydrates, fats, and proteins) for energy and growth. It has been shown that supplementation can correct micronutrient deficiency states in malnourished HIV-infected individuals (Baeten 2002b; Fawzi 1998). Widespread supplementation may lessen the effects of concurrent micronutrient deficiency and help to reduce the morbidity and mortality due to HIV (Semba 1999), which is particularly significant for developing countries where nutritional deficiencies are common (ASSAf 2007; Micronutrient Initiative 2009).

Why it is important to do this review

A previous version of this Cochrane review included 16 trials in 7 countries based on a July 2004 search (Irlam 2005), and found no conclusive evidence that micronutrient supplementation effectively reduces or increases morbidity and mortality in HIV-infected adults. Vitamin A was found to be beneficial for HIV-infected children.

The HIV/AIDS pandemic has had a major impact on global health, nutrition, and overall socio-economic development. An update of the review based on recent, valid research is therefore important. Micronutrient supplements have potential benefit for people living with HIV infection. However, in order to understand the magnitude of this benefit and how supplements should be positioned alongside the proven advantages of antiretroviral drugs, a robust evidence-base to guide policy and practice is required.

Additional note, December 2011: At the request of the World Health Organization, we have prepared three separate reviews examining the question of micronutrient supplementation in three separate populations with HIV. While the present review includes ‘adults and children,’ the WHO reviews separately examine children, pregnant women, and non-pregnant adults. We are in the process of preparing these reviews for publication in the Cochrane Library. A new version of the existing review (that you are presently reading) will focus only on non-pregnant adults. There will be a separate review focusing on children. Another review will focus on pregnant women. There may for a short time be some overlap among the three reviews, but we anticipate publishing the three separate reviews by early 2012.

Objectives

To assess whether micronutrient supplements are effective and safe in reducing mortality and morbidity in adults and children with HIV infection.

Methods

Criteria for considering studies for this review

Types of studies

All randomised controlled trials (RCTs) of micronutrient supplements compared with other supplements, placebo, or no treatment.

Types of participants

Adults and children with confirmed HIV infection (as reported in the trials) were included. No differentiation was made between HIV-exposed and HIV-infected infants.

HIV-infected pregnant women receiving vitamin A supplements were excluded as others have reviewed these trials (Kongnyuy 2009; Wiysonge 2005).

Types of interventions

Micronutrient supplements include vitamins (A, D, E, C, B1, B2, niacin, B6, B12, K, folate, beta-carotene), trace elements (zinc, selenium, magnesium, iron, iodine, copper, manganese, chromium, cobalt, molybdenum), and combinations of the above only.

Types of outcome measures

Primary outcomes

The primary outcomes considered were mortality, morbidity, hospitalisations, and pregnancy outcomes.

Secondary outcomes

The secondary outcomes were indicators of HIV disease progression (viral load, T cell counts) and anthropometric measures.

All adverse effects of supplementation were also considered.

Search methods for identification of studies

Periodic search updates to identify studies additional to the original review (Irlam 2005) were undertaken in December 2005, February 2007, April 2008, June 2009, and January 2010 of the CENTRAL, EMBASE, PUBMED, and GATEWAY databases, using the search methods of the Cochrane HIV/AIDS Group. Conference abstracts were not searched and no unpublished data were included. There were no language restrictions to the search.

The full search strategies for each of the databases above are presented in Appendices 1 to 4, and the flowchart of search results is in Figure 1.

Figure 1.

Data collection and analysis

Selection of studies

Three of the authors (JI, MV and NR) independently selected and appraised the papers for inclusion in the original review (Irlam 2005). JI and MV did the selection of studies for the review update.

Data extraction and management

Two of the authors (JI and MV) independently extracted data from the included studies using a data extraction form in the original review (Irlam 2005). Two of the authors (NS and NR) helped to resolve any queries about the papers. The extraction form was modified for the review update and completed independently by JI and CN for all studies in the original and updated review. Data were entered into the Review Manager 5 software.

Assessment of risk of bias in included studies

The risk of bias in the included studies was assessed independently by two authors (JI and NS) as high, low, or unclear using the six domains recommended by the Cochrane Collaboration 'Risk of bias' tool. These domains are sequence generation, allocation concealment (both described under Allocation below), blinding, incomplete outcome data, selective outcome reporting, and other potential sources of bias. The assessments are presented in the 'Risk of bias' tables and are summarised in Figure 2 and Figure 3.

Figure 2.

Methodological quality summary: review authors' judgements about each methodological quality item for each included study.

Figure 3.

Methodological quality graph: review authors' judgements about each methodological quality item presented as percentages across all included studies.

Measures of treatment effect

The measures of treatment effect used were the risk ratio (RR) for dichotomous data, the weighted mean difference (WMD) for continuous data measured on the same scale, and the standardised mean difference (SMD) for continuous data measured on different scales, presented with 95% confidence intervals. Review Manager 5 and GradePro (GradePro 2008) software were used to produce Summary of Findings and Evidence Profile tables.

Dealing with missing data

Authors were not contacted for data that were missing from the included studies, with the exception of the ZVITAMBO study where clarity about the statistical analyses was sought.

Assessment of heterogeneity

Studies were first assessed for clinical heterogeneity by examining variability in the participants, interventions and outcomes. Statistical heterogeneity was assessed visually and by means of the chi-squared test for heterogeneity. Inconsistency across the studies in the meta-analysis was quantified by means of the I-squared statistic included in the Review Manager forest plots.

Data synthesis

Random effects meta-analyses were performed by JI and checked by a second author if it was agreed that the studies were sufficiently homogeneous.

Subgroup analysis and investigation of heterogeneity

Subgroup analyses were planned according to the intervention of interest (single supplements of vitamin A, D, zinc, or selenium, or multiple supplements) and whether it was performed in adults or children.

Results

Description of studies

Results of the search

The PRISMA flow diagram (Figure 1) summarises the results of the search.

Sixteen additional trials were included in this update to the original review (Irlam 2005). Overall, 30 trials involving 22 120 participants are included: 20 trials of single supplements (vitamin A, vitamin D, zinc, selenium) and ten of multiple micronutrients.

Eight of the 30 trials were undertaken in child populations.

Included studies

Details of the included studies are reported in the table Characteristics of included studies.

They are grouped as follows by intervention in the description of the Effects of the interventions below:

  • Vitamin D (1 trial in adults; 365 participants: Wejse 2009; 1 trial in children; 59 participants: Arpadi 2009)

Excluded studies

Five studies were excluded due to inadequate methodological quality, use of interventions that were not exclusively micronutrients, or study outcomes that were not eligible for this review. See Characteristics of excluded studies.

Risk of bias in included studies

The Risk of Bias (ROB) tables (see Characteristics of included studies) provide the authors' judgements and comment on the risk of bias in each study (high, low, or unclear) with respect to each of the six domains described above. Figure 2 and Figure 3 present an overall graphical summary of the ROB assessments.

Allocation

Allocation refers to the generation of the random allocation sequence as well as concealment of the allocation code.

The methods of generating the random allocation sequence were sufficiently well described in three vitamin A trials in adults (Baeten 2002; Semba 1998; ZVITAMBO 2006) and judged to be at low risk of bias. The methods were too unclear in the remaining three studies (Coodley 1993; Coodley 1996; Humphrey 1999) to permit other than a judgement of "unclear" risk of bias. Allocation concealment was judged adequate in two studies (Semba 1998; ZVITAMBO 2006), and unclear in the remaining four (Baeten 2002; Coodley 1993; Coodley 1996; Humphrey 1999).

Three placebo-controlled vitamin A trials in children (Coutsoudis 1995; Hanekom 2000; Semba 2005) adequately described methods of randomisation. Two reported that allocation was adequately concealed (Fawzi 1999; Semba 2005).

One trial of vitamin D in adults and one of vitamin D in children were included. Both vitamin D trials used adequate random allocation to vitamin D or placebo; and allocation concealment was adequately described in both trials (Arpadi 2009; Wejse 2009).

Of the two trials of zinc in adults, random sequence generation was judged adequate in one (Carcamo 2006) and unclear in the other (Fawzi 2005). Allocation concealment was unclear in the Fawzi trial (Fawzi 2005) and adequate in the other trial (Carcamo 2006). Randomisation was adequately described in the two zinc trials in children; allocation concealment was judged unclear in one (Bobat 2005).

Of the three selenium studies, one (Burbano 2002) was unclear about the method of randomisation and allocation concealment, and another (Hurwitz 2007) was clear about randomisation but unclear about allocation concealment. The third trial was well reported and judged to have a low risk of bias in this domain (Kupka 2008).

Blinding

Blinding of patients, treatment providers and outcome assessors was judged adequate in all studies with the exception of one trial in adults with persistent diarrhoea (Kelly 1999). A multiple supplement was compared to a non-identical placebo, and it was unclear whether providers and assessors were blinded.

Incomplete outcome data

Four (Coodley 1993; Humphrey 1999; Semba 1998; ZVITAMBO 2006) of the six vitamin A trials in adults were judged to have adequately addressed incomplete outcome data. Incomplete outcome data was adequately addressed in one vitamin A trial (Hanekom 2000) in children, judged inadequate (i.e. at high risk of bias) in another (Coutsoudis 1995), and unclear in the remaining three trials (Fawzi 1999; Hussey 1996; Semba 2005).

One vitamin D trial (Arpadi 2009) adequately addressed incomplete outcome data while the other (Wejse 2009) did not.

Of the selenium studies, one trial (Kupka 2008) adequately addressed incomplete outcome data, one did not (Burbano 2002), and the judgment was "unclear" for the third (Hurwitz 2007).

Incomplete outcome data were adequately addressed in both zinc trials in children (Bobat 2005; Luabeya 2007), was unclear in one zinc trial in adults (Fawzi 2005), and not adequately addressed in the second zinc trial in adults (Carcamo 2006).

Incomplete outcome data was judged to be adequately addressed in two of the multi-nutrient supplement trials (Fawzi 1998; Kaiser 2006), judged "unclear" in another four (Austin 2006; Range 2006; Semba 2007b; Villamor 2008), and considered inadequately addressed in the remaining four (Allard 1998; Jiamton 2003; Kelly 1999; Semba 2007a).

Selective reporting

Insufficient information was provided to permit judgment of the extent of bias due to selective reporting of outcomes in all but 5 (Fawzi 1998; Luabeya 2007; Semba 2005; Semba 2007b; Wejse 2009) of the included studies.

Other potential sources of bias

Twelve trials (Bobat 2005; Fawzi 1998;Fawzi 2005; Hurwitz 2007; Jiamton 2003; Kaiser 2006; Kupka 2008; Range 2006; Semba 2005; Villamor 2008; Wejse 2009; ZVITAMBO 2006) were judged to be free of other biases, one trial (Austin 2006) was judged as not adequately free of other bias (stopped early due to interruption in the supply of medicine).

The judgement was "unclear" for the remaining seventeen trials; sixteen did not declare on potential conflicts of interest (Allard 1998; Arpadi 2009; Baeten 2002; Burbano 2002; Carcamo 2006; Coodley 1993; Coodley 1996; Coutsoudis 1995; Fawzi 1999; Hanekom 2000; Humphrey 1999; Hussey 1996; Kelly 1999; Semba 1998; Semba 2007a; Semba 2007b) and one (Luabeya 2007) experienced a delay in shipment which prevented 243 children from receiving their supplements for eleven weeks.

All but eleven trials (Allard 1998; Austin 2006; Burbano 2002; Coodley 1993; Coodley 1996; Hanekom 2000; Humphrey 1999; Hussey 1996; Jiamton 2003; Kaiser 2006; Kelly 1999) were funded either fully or partly from government sources; four were fully or partly funded by pharmaceutical companies (Coodley 1993; Kaiser 2006; Kelly 1999; Jiamton 2003); and the source of funding was not provided in two trials (Hanekom 2000; Hussey 1996).

Effects of interventions

Vitamin A in adults

Six trials of vitamin A in adults were included with a total of 14 763 participants, including both HIV-infected and uninfected subjects.

The majority of these came from a large Zimbabwean trial by the ZVITAMBO Study Group (ZVITAMBO 2006), which used a factorial design to compare the effect of four postpartum regimens of single large-dose maternal (400 000 IU) and infant (50 000 IU) vitamin A supplements versus placebo among 14 110 mother-infant pairs, which included 4495 HIV-infected mothers. Thirty percent of these mothers were vitamin A deficient, as indicated by serum retinol concentrations less than 1.05 micromol/L at 6 weeks.

This study found no effect on overall child mortality between baseline and 24 months. However, in HIV-exposed infants who were HIV-negative at baseline and HIV-infected at 6 weeks there was a 28% reduction in mortality (p=0.01); in HIV-exposed infants who were HIV-negative at 6 weeks of age there was approximately double the mortality in supplemented infants (p<=0.05).

There was no effect of vitamin A supplementation on morbidity among the HIV-positive women as measured by total number of sick clinic visits. Cause-specific visits for malaria, pelvic inflammatory disease, cracked and bleeding nipples, and vaginal infection were significantly reduced by between 16% and 40% depending on cause, compared to placebo.

High doses (180mg/ day or 600 000 IU) of beta-carotene for one month raised CD4 counts in a cross-over trial in 21 American outpatients (Coodley 1993), but not when given with multivitamins in an extended evaluation at 3 months (n=72); viral load was also unaffected (Coodley 1996). Vitamin A (10 000 IU) given daily for six weeks to 400 Kenyan women with a high prevalence of vitamin A deficiency (59% had serum retinol concentrations <30 microgram/dL) showed no effect on vaginal shedding of HIV-1, viral load, or CD4 counts (Baeten 2002). 

Two small trials of short duration to test the safety of single high doses of vitamin A, 200 000  IU in 120 intravenous drug users for 4 weeks (Semba 1998), and 300 000 IU in 40 women of reproductive age for 8 weeks (Humphrey 1999), found no effect on CD4 counts or viral load. 

Vitamin A in children

Five small trials of vitamin A in children (n=1120) were included.

In Ugandan children (n=181), quarterly vitamin A supplementation (60mg RE or 200 000 IU) from 15 to 36 months reduced all-cause mortality by 46% (adjusted RR = 0.54; 95% CI: 0.30-0.98) after follow-up for a median of 17.8 months, a shorter period than planned due to early stopping of the trial . Morbidity effects were a halving of persistent cough (OR = 0.47; 95% CI: 0.23-0.96) , and reduction in the duration of ear discharge (p=0.03) (Semba 2005).

Periodic doses (100 000 IU to infants; 200 000 IU to 1-5 year olds) for 24 months to Tanzanian children (n=687) admitted with pneumonia halved all-cause mortality (RR = 0.51; 95% CI: 0.29-0.90) and significantly reduced the risk of severe watery diarrhoea (adjusted OR = 0.56; 95% CI: 0.32-0.99) (Fawzi 1999). In a subgroup of 58 HIV-infected children, all-cause mortality was reduced by 63% (RR = 0.37; 95% CI: 0.17-0.84) and AIDS-related deaths by two-thirds (RR = 0.32; 95% CI: 0.1-0.99) (Fawzi 1999). There was a non-significant reduction in cough and rapid respiratory rate (RR = 0.54; 95% CI: 0.24-1.20) in this subgroup, and a non-significant increase in acute diarrhoea (RR = 1.55; 95% CI: 0.75-3.17) (Fawzi 1999). After four months of supplementation, there was a mean increase in height of 2.8 cm (95% CI: 1.0-4.6) of HIV-infected children under 18 months of age (Fawzi 1999; Villamor 2002a).

In South Africa, periodic vitamin A supplementation (50 000 IU at 1 and 3 months; 100 000 IU quarterly thereafter) of all children born to 118 HIV-infected women reduced all-cause morbidity by a third (OR = 0.69; 95% CI: 0.48-0.99) during 18 months of follow-up. In a subgroup of 28 children with HIV infection at birth, there was no effect on mortality but episodes of diarrhoea were halved (OR= 0.51; 95% CI: 0.27-0.99) (Coutsoudis 1995).

A meta-analysis of the effect of vitamin A on mortality in 267 HIV+ children, based on the three trials above, showed an overall significant reduction of 50% (RR = 0.50; 95% CI: 0.31-0.79) (Figure 4).

Figure 4.

Forest plot of comparison: 2 Vitamin A in children, outcome: 2.1 All-cause mortality.

A small trial (n=59) of 200 000 IU of vitamin A for two days in North American children receiving influenza vaccine decreased HIV viral load at 14 days from the time of vaccination, compared to an increase in the placebo group (p = 0.02) (Hanekom 2000). An identical regimen raised CD4 counts at four weeks in 75 South African children (p = 0.03) (Hussey 1996).

Vitamin D

Wejse et al (Wejse 2009) examined the effect of periodic vitamin D (100 000 IU cholecalciferol at baseline, and at 5 months and 8 months after inception) given supplementary to anti-TB treatment in 365 adults in Guinea-Bissau, 131 of whom were HIV-infected. They found no reduction in a clinical severity score (TB score) and no effect on 12-month mortality (HR = 1.8; 95% CI: 0.8–4.1) overall, or in the HIV-infected subgroup, compared to placebo.

A small trial (n=59) of vitamin D (100 000 IU cholecalciferol bimonthly) and calcium versus double placebo to evaluate the effect on monthly serum vitamin D concentrations over 12 months in HIV-infected American children and adolescents, found significant increases in vitamin D levels (p< 0.0001) in the supplemented group, and no adverse effects on HIV disease progression, as measured by CD4 counts (p=0.18) and viral load (p=0.66) (Arpadi 2009).

Zinc

Daily zinc supplements (25mg) given to 400 pregnant Tanzanian women until 6 weeks after delivery had no effect versus placebo on foetal and neonatal mortality, duration of pregnancy, birth weight, maternal T cell counts, or viral load. The zinc supplements did appear to inhibit increases from baseline in haematological indicators compared to placebo, namely haemoglobin (p=0.03), packed cell volume (p=0.01), and red blood cell count (p<0.01). Zinc supplementation was also associated over 22 weeks’ follow-up with a threefold increase in the risk of maternal wasting (risk of MUAC < 22 cm; p=0.03), and a 4 mm decline in MUAC during the second semester (p=0.02) (Fawzi 2005).

A placebo-controlled trial of 50 mg oral zinc twice daily for 14 days to 159 Peruvian adults with diarrhoea lasting for a week or more found no effect on the persistence and severity of diarrhoea at day 14 of treatment (Carcamo 2006). The participants were not on ART; those in the zinc group had median CD4 counts of 65 (0-909) versus 55 (2-2021) in the placebo group.

A placebo-controlled equivalence trial to determine the safety of zinc supplementation in HIV- infected children was conducted in 96 South African children aged 6 to 60 months (Bobat 2005).  A daily dose of 10mg zinc sulphate for up to 6 months did not increase viral load. There was a significant reduction in the secondary outcome of diarrhoeal morbidity, as measured by the proportion of scheduled and illness clinic visits where a diagnosis of watery diarrhoea was made (p=0.001).

Diarrhoeal and respiratory morbidity in 341 HIV-uninfected and 32 HIV-infected rural South African children, aged 4 to 6 months, was measured by maternal report during home visits (pneumonia also confirmed by measurement of rapid respiratory rate) in a trial that compared supplementation for prophylaxis for a median of 14.9 months with 10 mg zinc plus vitamin A, or with zinc plus vitamin A and multiple supplements, to vitamin A alone. There were no differences between the treatment arms in either the HIV-uninfected or HIV-infected children on prevalent days or incidence density of diarrhoea (Luabeya 2007).

Selenium

Daily selenium supplements of 200 mg for 12 months in 186 American drug users versus placebo reduced the number of participants hospitalised for opportunistic infections and HIV-related conditions (RR = 0.40; 95% CI: 0.21-0.75), and lowered the risk of a CD4 decline of greater than 50 cells/mm3 (Burbano 2002). Hurwitz et al evaluated an identical regimen for 9 months in a placebo-controlled trial of 262 HIV-infected adults, and found that supplements elevated serum levels (p<0.001), suppressed viral load (p<0.02), and increased CD4 counts (p<0.04) (Hurwitz 2007).

Kupka et al (Kupka 2008; Kupka 2009) randomised pregnant women in Tanzania (n=913) to 200µg daily selenium vs. placebo until 6 months post-delivery. The supplements did not improve maternal HIV disease progression or haemoglobin concentrations, but they did reduce maternal diarrhoeal morbidity by 40% (RR=0.60; 95% CI: 0.42; 0.84). There was no significant effect on neonatal or overall child mortality, but a reduction in child mortality at 6 weeks was observed (RR=0.43; 95% CI: 0.19; 0.99).

Multiple supplements

A large trial was conducted in 1078 pregnant Tanzanian women, which employed a factorial design of daily multivitamins (mainly vitamins B, C and E in doses up to 22 times RDA) with and without vitamin A versus vitamin A only or placebo (Fawzi 1998). The supplementation commenced at enrollment (12-27 weeks gestation) and lasted throughout pregnancy and lactation. The median follow up with respect to survival was 71 months (IQR: 46-80 months). Significant delays in the progression to stage 4 disease or AIDS-related mortality (HR = 0.71; 95% CI: 0.51-0.98) and in progression to stage 4 disease alone (HR = 0.50; 95% CI: 0.28-0.90) were observed in women receiving multivitamin supplements, compared to placebo. Data were censored at the time of death when the cause of death could not be ascertained by means of standardised interviews and/or review of medical records, or was not deemed to be AIDS-related. Viral load and all signs of HIV-related complications in multivitamin-supplemented women were reduced relative to placebo, and CD4 and CD8 counts were raised, compared to vitamin A or placebo. Adding vitamin A to the multivitamin regimen reduced the beneficial effects on almost all maternal and infant outcomes to be generally comparable with the placebo.

Multivitamins improved mean maternal weight gain during the third trimester compared to no multivitamins (mean difference (MD) = 304 g; 95% CI: 17- 590), and reduced the risk of low weight gain (<= 100g per week; RR = 0.73; 95% CI: 0.58- 0.93) (Fawzi 1998; Villamor 2002b) and the risk of wasting (first episode of mid-upper arm circumference (MUAC) <22 cm; RR = 0.66; 95% CI: 0.47- 0.94) (Fawzi 1998; Villamor 2005b). Compared to vitamin A or placebo, multivitamins reduced the risk of hypertension during pregnancy (RR = 0.62; 95% CI: 0.40- 0.94) (Fawzi 1998; Merchant 2005), protected against depression (RR = 0.78; 95% CI: 0.66- 0.92) and improved quality of life (RR = 0.72; 95% CI: 0.59- 0.88 for social functioning) (Fawzi 1998; Smith Fawzi 2007).

Multivitamins reduced the risk of adverse pregnancy outcomes, including foetal deaths (RR = 0.61; 95% CI: 0.39-0.94), low birth weight (RR = 0.55; 95% CI: 0.38-0.81), severe preterm births (<34 weeks) (RR = 0.61; 95% CI: 0.38-0.95), and babies small for gestational age (RR = 0.57; 95% CI: 0.40-0.83) (Fawzi 1998).The mean birthweight of infants born to HIV-negative mothers in the supplemented group was significantly higher (MD = 94 g, p = 0.02) (Fawzi 2000). Child growth during the first two years of life was improved by maternal multivitamin supplementation, namely attained weight (MD = 459g; 95% CI: 35, 882), weight-for-age (MD = 0.42; 95% CI: 0.07, 0.77), and weight-for length (MD = 0.38; 95% CI: 0.07, 0.68) (Fawzi 1998; Villamor 2005a).

Maternal multivitamin supplementation also reduced the risk of child diarrhoea (RR = 0.83; 95% CI: 0.71-0.98 at 24 months), and improved children’s CD4 counts (MD = 153 cells/µl; 95% CI: 67.6-238.4) (Fawzi 1998; Fawzi 2003) and micronutrient status during the first six months of age (Baylin 2005; Fawzi 1998 ). Vitamin A supplements reduced the risk in children of cough with rapid respiratory rate (RR = 0.69; 95% CI: 0.49-0.96) (Fawzi 1998; Fawzi 2003) and the incidence of clinical malaria by 71% (RR=0.29; 95% CI: 0.09-0.89) in children born to HIV-infected women (Fawzi 1998; Villamor 2007).

In multivitamin-supplemented women with low immunological or nutritional status, significant reductions were reported in the risk of child mortality by 24 months of age: 70% (RR = 0.30; 95% CI: 0.1-0.92) in women in the lowest quartile of lymphocyte counts (<1340/ mm3); 60% (RR = 0.40; 95% CI: 0.17-0.98) in women with low baseline vitamin A (<20 µg/dl); and 69% (RR = 0.31; 95% CI: 0.13-0.73) in women with low baseline vitamin E (<9.7 µmol/l) (Fawzi 1998; Fawzi 2002).

In a trial of 481 Thai adults, twice-daily supplementation for 48 weeks with a commercial mix of 18 vitamins and minerals, some at doses up to 20 times higher than the recommended daily allowance (RDA), had no significant effect on overall mortality (p=0.1), CD4 counts (p>0.3), or plasma viral load (p=0.4). Mortality was significantly reduced in those with low CD4 counts at baseline (CD4 <100: hazard ratio (HR) = 0.26; 95%CI: 0.07-0.97). A trend towards a reduced death rate in those with CD4 counts <200 at baseline was also reported (HR=0.37; 95% CI: 0.13, 1.06) (Jiamton 2003).

Daily supplementation with micronutrients and natural mixed carotenoids in a community randomised controlled trial of 331 Canadian adults with advanced AIDS had no effect on the primary outcome of time to new or recurrent AIDS-defining event or death (HR = 1.81; 95%CI: 0.95-3.42, p=0.07). A reduction in the secondary outcome of time to all-cause mortality was apparent in multivariate analysis (HR= 3.15; 95%CI: 1.10-8.98, p=0.03) and there was improved survival in those with higher baseline concentrations of serum carotene (p=0.04) or CD4 counts (p=0.005) (Austin 2006).

A small (n=40) randomised placebo-controlled trial of micronutrients given twice daily for 12 weeks in American adults on highly active antiretroviral therapy (HAART) reported a significant increase in the primary outcome of CD4 counts in the treatment group (p=0.029) (Kaiser 2006). There was no significant change in any of the secondary endpoints, namely virologic and metabolic parameters, neuropathy scores, and general health status.

The effects of multivitamin (MVM)/ zinc (Zn) supplementation during treatment of pulmonary tuberculosis (PTB) was tested by means of a two-by-two factorial trial in 499 PTB patients in Tanzania, 213 of whom were co-infected with HIV. There was a non-significant effect of MVM (RR 0·60; 95% CI 0·34, 1·05) and Zn (RR 0·63, 95% CI 0·37, 1·08) on mortality in co-infected patients when given separately but when MVM and Zn were given in combination there was a significant reduction in mortality of 71% (RR= 0.29; 95%CI: 0.10-0.80, p=0.016). The combined interventions did not however yield significant weight gain (2·37 kg; 95% CI 0·91, 3·83), irrespective of HIV status (Range 2006).

A trial in 1402 Malawian patients with PTB, 829 of them HIV-infected, of daily micronutrients, including zinc versus placebo for 24 months, did not reduce mortality in the HIV-infected subgroup (Semba 2007b). A placebo-controlled trial of daily micronutrients (retinol; vitamins B1, B2, B6, B12; niacin; vitamin C; vitamin E; folic acid; selenium) was conducted for 8 months among 887 Tanzanian adults with PTB, 471 of them with HIV infection (Villamor 2008). It found no effect on overall mortality, HIV disease progression, or nutritional indicators, but the risk of TB recurrence in HIV-infected patients was reduced (RR=0.37; 95%CI: 0.15;0.92).

In the USA, the effects of daily micronutrient and iron supplementation on iron status and anaemia were tested in 458 hepatitis C-positive female injection drug users, 138 of whom were HIV-infected. The daily supplement reduced anaemia and improved iron status in all women at 12 months versus those not receiving iron, without an adverse effect on viral load or CD4 counts (Semba 2007a).

Large daily doses of vitamins E and C for three months reduced measures of oxidative stress in a small (n = 49) Canadian trial, and showed a non-significant trend towards a reduction in viral load, but had no effect on HIV-associated morbidity. Vitamin C and E serum levels at three months were both significantly higher (p<0.005) in the treatment group than placebo, but levels of vitamin A, carotenoids, zinc and selenium were unchanged (Allard 1998).

Daily micronutrient (vitamins A, C, and E; selenium; zinc) supplementation in a trial of 135 Zambian adults with persistent diarrhoea, more than half of whom were classified vitamin A and vitamin E deficient,  had no effects on mortality, diarrhoeal morbidity, CD4 counts, body mass index, quality of life scores, or vitamin A and E serum levels (Kelly 1999). 

A trial of prophylactic zinc or multiple micronutrient supplementation to reduce diarrhoea and respiratory disease in young South African children is described above (Luabeya 2007). Neither intervention, alone or in combination, had an effect in either HIV-infected or uninfected children to support the prophylactic use of zinc or multiple micronutrients to reduce morbidity, although the sample size of HIV-infected children was very small.

Adverse effects of supplementation

No significant adverse effects of vitamin A in adults were reported. Adverse effects of vitamin A in children, irrespective of HIV status, included vomiting and bulging fontanelle (Coutsoudis 1995); and a small risk of acute diarrhoea in normally nourished (RR = 1.37; 95% CI: 1.06-1.79) and in growth-stunted children (RR = 1.84; 95% CI: 1.16-2.90) (Fawzi 1999) hospitalised with pneumonia. HIV-negative children who received vitamin A had a non-significant increased risk of cough and rapid respiratory rate compared to those on placebo (p = 0.07).

Vitamin A reduced the benefits of multivitamin supplementation in pregnant women in Tanzania (Fawzi 1998), and increased mortality in a sub-sample of infants who were HIV-negative at 6 weeks in a large trial of supplemented Zimbabwean mother-infant pairs (ZVITAMBO 2006).

Daily zinc supplements (25mg) to pregnant women in Tanzania inhibited increases in haematological indicators and increased the short-term risk of maternal wasting (Fawzi 2005).

Discussion

Summary of main results

Vitamin A in adults

Despite strong evidence from observational studies (Tang 1993; Semba 1993; Semba 1995) of an association between low vitamin A levels and HIV disease progression and mortality in adults, none of the six trials reviewed demonstrated any significant reduction in HIV disease progression, or any significant adverse effects, associated with vitamin A or beta-carotene supplementation.

The ZVITAMBO trial of a single large dose of vitamin A given to Zimbabwean women and their infants during the postpartum period found a two-year mortality reduction in infants infected during the late intrauterine/ intrapartum/ early postnatal period, which supports existing evidence that vitamin A supplementation of HIV-infected infants and children is likely to prolong survival. Mortality was higher, however, in a subgroup of infants who were not HIV-infected at 6 weeks of age. The authors suggest the possibility that priming with vitamin A may have increased viral load in babies subsequently infected during breastfeeding. This finding, together with the lack of effect on maternal mortality and the limited effect on maternal morbidity, leads the authors to raise concern about universal maternal postpartum vitamin A supplementation in HIV-endemic areas (ZVITAMBO 2006). This subgroup mortality effect is unique to this trial however, and further research is required to fully inform policy decisions.

Research is needed into the effect of smaller doses on maternal mortality in an HIV-prevalent population with a poorer vitamin A status than the women who participated in the ZVITAMBO trial. This issue is suggested by a large trial in a rural, undernourished, and presumably HIV-negative population in Nepal, which found that weekly vitamin A supplements halved maternal mortality up to 12 weeks postpartum (West 1999). Further research into other potential benefits of vitamin A supplementation is needed, such as reduced night-blindness in pregnant women where deficiency is common, irrespective of HIV status (van den Broek 2002). A systematic review of five trials has found no evidence however to support prenatal and postnatal vitamin A supplementation for reducing the risk of mother-to-child transmission (Kongnyuy 2009; Wiysonge 2005 ).

Vitamin A in children

Vitamin A is standard care in children without HIV infection who present with persistent diarrhoea and severe acute malnutrition (UNICEF 2007; WHO/UNICEF 2004 ). Six monthly regular supplements of vitamin A are also recommended for all children between 6 months and five years to support growth and development and to reduce all-cause mortality, irrespective of HIV status (Baeten 1993; WHO 2009).

There is strong evidence from a few African trials of clinical benefits of vitamin A on mortality in HIV-infected children, and moderate evidence for morbidity and growth benefits (Fawzi 1999; Villamor 2002a; Coutsoudis 1995; Semba 2005). A meta-analysis of three trials (Coutsoudis 1995; Fawzi 1999; Semba 2005) (Figure 4) shows that vitamin A halved all-cause mortality, although the HIV-infected children in two of the trials were small subgroups of the study populations. There is inconsistent evidence on the effects of vitamin A on morbidity, with some benefits and adverse effects on diarrhoeal and respiratory morbidity being reported, which may be due to the small sample sizes. The clinical benefits may be consequent to an improvement in immune function and to the rehabilitation of mucosal integrity, leading to a reduction in the severity and incidence of diarrhoeal infection, and an improvement in short-term growth.

Vitamin D

Vitamin D supplementation in HIV-infected individuals has not been substantially studied, but evidence from the two reviewed studies indicates that it does not have a deleterious effect on HIV disease progression. Wejse et al speculate that the lack of benefit in patients with TB in their study may have been due to suboptimal dosage, and that trials investigating prevention of latent TB infection with larger, perhaps daily, vitamin D doses may therefore be warranted (Wejse 2009). Arpadi et al recommend further research to determine what level of supplementation with vitamin D, together with daily calcium supplementation, will lead to clinically significant gains in bone mass accrual in children and adolescents with HIV, as well as other potential benefits (Arpadi 2009).

Zinc

In observational studies of HIV-infected adults, low levels of serum zinc have been related to HIV disease progression (Graham 1991), decreased CD4 cell counts (Baum 2003) and increased mortality (Baum 1997; Baum 2003), and low dietary intake has been associated with decreased survival (Baum 2003). An association has also been reported between self-prescribed zinc consumption above the recommended dietary allowance (RDA ) and rapid progression to AIDS (Tang 1993) and mortality (Tang 1996), although this may have been due to reverse causality if adults with more advanced disease chose to take more supplements.

The authors of a trial in Tanzania concluded that there is no compelling evidence for the addition of zinc to antenatal supplements for pregnant HIV-infected women, as there were no reductions in adverse pregnancy outcomes, and increases in haemoglobin levels were inhibited (Fawzi 2005). This reported lack of benefit is consistent with trials in HIV-uninfected pregnant women in Nepal (Christian 2003) and Mexico (Ramakrishnan 2004) in which zinc was found to inhibit the absorption of iron and mitigate increases in haemoglobin levels.

A two-week course of 100 mg of elemental zinc daily to Peruvian adults presenting with diarrhoea had no effect on the persistence of diarrhoea (Carcamo 2006). Treatment benefit may have been concealed due to the short duration and high attrition rates however.

In children without HIV infection, zinc supplementation has been shown to reduce the risk and severity of diarrhoea and pneumonia in several studies (Baqui 2002; Bhandari 2002; Bhutta 1999; Bhutta 2000; Strand 2002; Zinc Investigators Collaborative Group 2000). The reported benefits in HIV-infected South African children on the secondary outcome of diarrhoeal morbidity (Bobat 2005) are consistent with the evidence from these earlier studies. The primary finding in the South African trial that HIV-1 viral load was not raised, suggests that zinc supplementation may be considered safe as a specific therapy in children with HIV infection who present with diarrhoea. A subsequent study by Luabeya et al (Luabeya 2007) did not provide support however, for the prophylactic use of zinc or multiple supplements to reduce diarrhoeal or respiratory morbidity in rural South African children, including 32 HIV-infected children. Further research in larger and diverse populations of HIV-infected children is therefore needed.

Selenium

Multiple observational studies have reported on the associations between selenium and HIV disease. There have been few randomised trials directly assessing the relationship. Two trials that provided identical regimens of daily selenium supplements to North American adults reported a reduction in hospitalisations (Burbano 2002) and suppression of viral load and improved CD4 counts at 9 months, without any adverse effects (Hurwitz 2007). Mechanisms for the effect on viral load include an indirect effect of selenium via diminished oxidative stress, and a direct effect of the selenoproteins on viral replication. Both trials had small sample sizes however, and the significance of their findings need to be confirmed in other populations, especially in African settings where diets may be very different.

In the setting of Dar es Salaam, Tanzania, selenium supplements given during and after pregnancy did not delay HIV disease progression or improve pregnancy outcomes, but may improve child survival and decrease diarrhoeal morbidity (Kupka 2008) . Explanations offered for the lack of effect of selenium on maternal and pregnancy outcomes include a limiting effect of co-supplements; an effect restricted to patients with advanced HIV disease or those receiving HAART, which was not the case in this study; and the fact that selenium deficiency was probably uncommon. There is therefore good evidence from this large and rigorously designed trial of little to no benefit of providing selenium supplements to HIV-infected populations who are naı¨ve to HAART, who receive high-dose multivitamin supplements, and who live in areas where selenium deficiency is uncommon.

Multiple supplements

Multivitamin supplements in pregnant and lactating women in a large Tanzanian trial with long-term follow-up were associated with a number of clinical benefits for mothers and their offspring (Fawzi 1998). The authors suggested that the supplements enhanced immunity, namely improved T-cell counts, reduced HIV viral replication and somehow protected oral and gastrointestinal epithelia and hence reduced HIV-related complications. Multivitamins reduced the risks of adverse pregnancy outcomes, improved child growth during the first two years of life, and reduced child mortality in the first two years of life among those born to immunologically- and nutritionally-compromised women. These benefits may have been due to improved quality of the mothers' breast milk, with consequent enhancement of the infants' micronutrient intake and immune function causing a reduction in mortality and morbidity from diarrhoeal disease and upper respiratory tract infections. It is also likely that the improvement in the mothers' health and nutritional status enabled them to provide better care for their children.

In evaluating the promising benefits of this trial, it is worth noting that the doses of some of the vitamins provided were far in excess of recommended daily requirements, and were based not on preliminary pharmacokinetic studies, but on the prevalence of deficiencies in relation to self-reported and self-prescribed intakes of HIV-infected American men (Baum 1994). It is possible that a micronutrient regimen with much lower doses could be equally effective.

In one trial the addition of vitamin A (both preformed retinol and high dose (30mg) beta carotene) reduced the benefits of the micronutrient supplement. This may have been due to beta carotene functioning as a pro-oxidant, rather than as an antioxidant as would be expected, on other micronutrients that were given as part of the supplement. The potential pro-oxidative effect of beta carotene is recognised and this reason has been speculated as the cause of adverse effects in certain cancer studies (ATBC 1994; ATBC 1996; Omenn 1996). In this trial the combination of high doses of beta carotene and iron may have accounted for some of the adverse outcomes in the Vitamin A groups. Finally, the reduced morbidity reported in this trial may not be specific to HIV-infected women, but may also apply to HIV-free women in other populations at risk of undernutrition and high maternal mortality. This needs to be determined in order to inform general recommendations.

Evidence of clinical benefit of multiple micronutrients has also been provided by a large trial in Thailand (n=481), although the reduction in mortality was restricted to those with the lowest CD4 count (CD4<100) at baseline. In contrast to the Tanzanian studies, there was no effect on CD4 counts, viral load or overall mortality, suggesting that the supplement reduced the risks of other infections or mediated its benefit by helping to maintain lean body mass (Jiamton 2003).

A trial of supplementation with micronutrients and natural mixed carotenoids in Canadian adults with advanced AIDS on antiretroviral therapy, observed a benefit in adjusted analyses of the secondary outcome of overall mortality (Austin 2006). The authors postulated that the treatment effect may be due the direct antioxidant action of the carotenoids, or to interactions with the ARVs that improve the drug response. Important limitations of the study were early termination due to interruptions in the supply of study medication, and a lack of uniformity in the formulated dose during the study.

In a trial of forty American adults on highly active antiretroviral therapy (HAART) who received multiple micronutrients an increase in CD4 count was also reported. The authors suggested that certain nutrients may enhance cellular immunity, and thereby decrease the rate of CD4 cell death (Kaiser 2006).

Two African trials of supplementation in pulmonary TB patients, some of whom were co-infected with HIV, evaluated the effect on mortality. Range et al found a significant reduction in deaths of co-infected HIV and TB patients when supplemented with MVM, but only when given in combination with zinc during PTB treatment; only marginal effects were observed when either MVM or Zn were given alone. Some of the benefit may have been secondary to the considerable increases in weight gain in the supplemented group (Range 2006).

Villamor also reported that daily supplementation with micronutrients reduced the recurrence of TB in HIV-infected adults not on antiretroviral treatment, and decreased complications among HIV-negative adults (Villamor 2008). However, in a Malawian placebo-controlled trial that provided similar supplements but with lower levels of vitamins and minerals than those given in the Tanzanian study, there were no beneficial effects reported (Semba 2007a). This may have been because HIV-infected individuals require higher doses of micronutrients, either because of malabsorption states or increased utilization, to achieve clinical benefit. A meta-analysis of these two trials showed a non-significant reduction in mortality. Given the major contribution of TB to HIV-related mortality, it is important to conduct comparable studies in different settings to determine if these supplements may be effective adjuncts to antiretroviral drugs in preventing mortality.

Villamor et al reported that daily supplementation with micronutrients reduced the recurrence of TB in HIV-infected adults with TB not on antiretroviral treatment, and decreased complications among HIV-negative adults. Investigation of similar benefits is therefore required in HIV-infected patients receiving antiretroviral therapy in other settings, which could provide evidence for micronutrient supplementation as a useful and relatively inexpensive element of TB treatment regimens (Villamor 2008).

Semba et al have also studied the effects of daily micronutrient and iron supplementation in American female injection drug users with hepatitis C infection, a third of whom were HIV-infected. Although they found a reduction in anaemia and improved iron status compared to those not receiving iron, they caution against generalising these findings to other populations who were excluded from their study. The trial was also of short duration, and was underpowered to detect small but nevertheless significant adverse effects on HIV viral load in the long term (Semba 2007a).

A Zambian trial of multi-micronutrient supplements (vitamins A, C, and E, selenium, zinc) that was designed to test whether supplementation improves the clinical response to albendazole in the treatment of persistent diarrhoea in adults with HIV-wasting showed no effect on morbidity or mortality or serum vitamin levels (Kelly 1999). The absence of effect may have been real, or may have been due to inadequate duration of supplementation or to malabsorption of micronutrients.

Quality of the evidence

The quality of the evidence reviewed to date has some important limitations, as revealed in the summary assessment of the methodological quality of all included studies (Figure 3). Methods of random sequence generation and blinding were judged to be of low risk of bias in the majority of studies, but insufficient information about allocation concealment in many of the studies meant that the risk of bias is unclear. Fewer than half of the studies were judged to be of low risk of bias with respect to selective outcome reporting, or incomplete outcome data. The quality of evidence with respect to each of the interventions is presented below and summarised by study in Figure 2.

Vitamin A in adults

The method of randomisation was clearly described for three of six trials assessing vitamin A and not clearly described in the remaining three trials (Coodley 1993; Coodley 1996; Humphrey 1999). Allocation concealment was judged adequate in two (Semba 1998; ZVITAMBO 2006), and unclear in the remaining four (Baeten 2002; Coodley 1993; Coodley 1996; Humphrey 1999)

Placebo was given to participants, and treatment providers and assessors were blinded to the treatment assignments in all trials. Incomplete outcome data was deemed to be adequately addressed in four of the six trials, with the remaining two not clearly described (Baeten 2002; Coodley 1996). All six trials were judged to be unclear with respect to selective reporting. Only one trial (ZVITAMBO 2006) was judged to be free of other biases, while it was unclear for the remaining five.

Vitamin A in children
Five placebo-controlled trials in children (n= 1 120) were included (Coutsoudis 1995; Fawzi 1999; Hanekom 2000; Hussey 1996; Semba 2005), three with adequately described methods of randomisation and two with allocation concealment clearly described. All were blinded studies and incomplete outcome data was deemed to be adequately addressed in only one trial (Hanekom 2000), not described in another (Coutsoudis 1995) and inadequately addressed in the remaining three. Only one trial (Semba 2005) was judged to be free of selective reporting as well as free of other biases.

Vitamin D

One trial of vitamin D in adults (n= 365) (Wejse 2009) and one in children (n= 59) (Arpadi 2009) were included. Both were blinded, and used adequate random allocation to vitamin D or placebo. Allocation concealment was clearly described in both trials. Incomplete outcome data was adequately addressed in one trial (Arpadi 2009) and the other trial (Wejse 2009) was judged to be free of selective reporting as well as other biases.

Zinc

In the two trials of zinc in adults (Carcamo 2006; Fawzi 2005) (n=559), random generation was deemed to be adequate in one (Carcamo 2006). Allocation concealment was unclear in the Fawzi trial but clearly reported by Carcamo et al. Both trials were blinded placebo-controlled studies. Incomplete outcome data was inadequately addressed in one trial and not addressed in the other. It was unclear as to whether the trials were free of selective reporting but one trial was judged to be free of other biases (Fawzi 2005).

The two trials in children (Bobat 2005; Luabeya 2007) (n = 128) met most of the criteria for methodological quality with the exception of allocation concealment, which were not clearly described in the Bobat study. In addition, the Bobat trial was not judged free of selective reporting while the Luabeya trial was not deemed to be free of other biases.

Selenium

Three blinded placebo-controlled trials in adults (n= 1 361) were included (Burbano 2002; Hurwitz 2007; Kupka 2008). The Burbano trial was unclear about or failed to meet all but one of the methodological quality criteria, whereas the Kupka trial met all but one of them. The Hurwitz trial was unclear about allocation concealment, did not adequately address incomplete outcome data and was not judged free of selective reporting.

Multiple supplements

Ten trials of multiple heterogeneous supplements in adults (n= 3 765) include a single large multi-factorial trial (n = 1 078) in pregnant and lactating women (Fawzi 1998) that has yielded multiple papers examining diverse outcomes at different time points. It meets all the criteria for methodological quality..

The remaining nine trials in adults (n= 2 687) (Allard 1998; Austin 2006; Jiamton 2003; Kaiser 2006; Kelly 1999; Range 2006 (pulmonary tuberculosis (PTB) patients); Semba 2007a (women injectable drug users or IDUs); Semba 2007b (PTB patients); Villamor 2008 (PTB patients) were generally well-reported trials, with the exception of the study by Kelly et al. The Kaiser trial did not adequately report on the method of randomisation and allocation concealment, and was not deemed to be free of selective reporting. Allocation concealment was not clearly described in the study by Villamor et al. Only two studies adequately addressed incomplete outcome data (Austin 2006; Kaiser 2006) while it was deemed to be unclear in three trials and inadequately addressed in the remaining four. Only one trial (Semba 2007b) was judged to be free of selective reporting, the remaining eight unclear. Four trials were judged free of other biases, one trial (Austin 2006) not free of other biases, and in the remaining four it was judged to be unclear.

Potential biases in the review process

Biases in the review process were minimised by performing a comprehensive search of the literature, independently selecting and appraising the studies, and extracting the data. Assessment of the risk of bias (RoB) of all studies was repeated for the studies in the original review (Irlam 2005) and for the newer studies using the updated RoB tool by JI and NS.

For the purpose of the review, the HIV status of children determined by each study was accepted as sufficient. In past years laboratory methods have improved and become more sensitive and specific, thereby reducing the likelihood of false positive or false negative attribution of HIV status. However, further analysis or validation of children's status was beyond the scope of this review. 

Several studies have also reported on the increased morbidity and mortality of infants and children born to HIV-infected mothers compared to those born to HIV-uninfected mothers (Kuhn 2005; Newell 2004; Shapiro 2007, ZVITAMBO 2006). There are several possible mechanisms to explain this. First, the health and survival of HIV-infected mothers exert a major effect on child survival. Low maternal CD4 count or death increases child mortality by factors of 3.5 and 4 respectively, demonstrating the importance of basic care practices provided by a healthy mother. HIV-infected mothers also tend to breastfeed their infants for shorter intervals, which reduces the immune protection provided by breast milk and increases the risk of diarrhoea and malnutrition associated with other feeding practices. Lastly, infants born to HIV-infected mothers may have transient impairment of their immune systems that may increase susceptibility to non-HIV infectious diseases such as pneumonia. However there is insufficient basic evidence to determine which of these, or which combination, is most significant.

The effect of micronutrient supplements on HIV exposed but not infected infants and children was outside the scope of this review and was not formally evaluated. The few studies that did include these children in their reports (Luabeya 2007) did not suggest a differential response to micronutrient interventions of HIV exposed but uninfected children compared with children born to HIV-uninfected mothers.

Agreements and disagreements with other studies or reviews

In July 2007 the Academy of Science of South Africa (ASSAf) published HIV/AIDS, TB and Nutrition, a scientific inquiry into the nutritional influences on human immunity with special reference to HIV infection and active TB in South Africa (ASSAf 2007). The inquiry found a dearth of reliable and informative studies, and recommended improved nutritional policy and practice informed by high-quality research. It was recognised that nutritional interventions should be part of a comprehensive, integrated approach to HIV and TB, but are no substitute for anti-retroviral drugs in preventing transmission; that nutritional care should focus on diversified diets of available, affordable and culturally acceptable foods, as well as safe levels of macro- and micronutrients; and that priority should be accorded to HIV-infected pregnant women, lactating mothers and their babies.

Key recommendations of the ASSAf Scientific Panel with respect to micronutrient supplementation that were derived from the earlier version of this review (Irlam 2005) included:

  • promoting adequate dietary intake of micronutrients at recommended INL98 levels

  • providing elevated levels of micronutrients (at least 1-2 INL98s) through food fortification or supplements in settings where micronutrient deficiencies are endemic,

  • offering multivitamin supplementation at INL98 levels to HIV-infected women

  • better definition of the indicators of vitamin and mineral micronutrient levels in individuals and populations

A 2009 review (Investing in the Future: A United Call to Action on Vitamin and Mineral Deficiencies) prepared by the Micronutrient Initiative (an international not-for-profit organization) in partnership with UNICEF, WHO and others (Micronutrient Initiative 2009), reported that the 2008 Copenhagen Consensus, a group of world-renowned economists, ranked micronutrient supplements (high-dose vitamin A, and zinc supplements for children with diarrhoea) as the top development priority out of more than 40 interventions considered (Horton 2008). The benefit to cost ratio, as well as the feasibility and sustainability of the interventions, were considered. Vitamin A supplementation every 4 to 6 months for children from age 6 months to 5 years has been shown to reduce all-cause mortality by 23% (Baeten 1993), and 10 to14 days of therapeutic zinc supplementation for diarrhoea up to the age of 5 can halve diarrhoeal mortality (Baqui 2002).

Large scale vitamin A supplementation began in the late 1990’s with mass polio immunization campaigns following WHO recommendations, and coverage has been adopted as an indicator of progress toward the Millennium Development Goal (MDG) of reducing child mortality by 2015 (Wagstaff 2004). There has been less progress in achieving universal post-partum vitamin A supplementation for breastfeeding mothers, which is recommended to boost the immune system of infants in the first months of life (UNICEF 2007). A review of recent research has suggested that neonatal mortality can be reduced by supplementing newborns within the first few days of life (Haider 2008), but there is as yet no international WHO recommendation on this. Low-dose supplementation of pregnant women with xerophthalmia, which may be due to systemic vitamin A deficiency, has also been recommended but not widely adopted (Horton 2008).

Strong evidence about the benefits of therapeutic zinc together with low osmolarity oral rehydration salts for reducing childhood diarrhoea (Baqui 2002; Bhutta 2000; Robberstad 2004; Zinc Investigators Collaborative Group 2000) resulted in a joint WHO/ UNICEF recommendation in 2004 of 10 to 14 days of therapeutic zinc for children under 5 years of age (WHO/UNICEF 2004).

The WHO/UNICEF report concludes with a number of priority interventions by national governments, industry and international organizations for achieving the MDG on child mortality by 2015. These interventions include:

  • six monthly vitamin A supplementation for children aged between 6 months and five years, to achieve at least 80% coverage on a recurrent basis

  • zinc supplementation as part of national diarrhoea management policy

  • multiple micronutrient supplements for children in non-malaria endemic regions

  • improved iron intake by young children in malarial areas

  • iron and folic acid supplementation for all women of childbearing age, with special focus on pregnant women

  • testing the feasibility of providing women with multiple vitamin and mineral supplement

Although the international reviews above primarily refer to HIV-uninfected populations, their recommendations also apply to populations with HIV infection unless there is evidence of adverse effects.

Authors' conclusions

Implications for practice

The evidence-base for the specific effect of micronutrient supplements in children and adults with HIV is limited, but is sufficient to make some recommendations for practice. In the absence of population-specific adverse effects, there is no reason to decline similar recommendations for HIV-infected populations.

  • Periodic vitamin A supplementation of HIV-infected children over six months of age in resource-limited settings is supported by three African trials in this review, which is consistent with evidence of benefit that supports supplementation of HIV-uninfected children.

  • Zinc supplements reduced diarrhoeal morbidity and had no adverse effects on disease progression in a single safety trial in South African children. Children with HIV should receive zinc supplements in the management of diarrhoea and severe acute malnutrition in the same way as HIV-uninfected children with the same conditions.

  • Providing daily multivitamin supplements to HIV-infected pregnant and lactating women is supported by a large Tanzanian trial, although the optimal composition and dosage still needs to be established.

In keeping with WHO recommendations (WHO 2003), everything possible should be done to promote and support adequate dietary intake of micronutrients at Individual Nutrient Intake Level (INL98) levels, while recognising that this may not be sufficient to correct specific micronutrient deficiencies in all HIV-infected individuals.

In situations where micronutrient deficiencies are endemic, these nutrients should be provided through food fortification or micronutrient supplements where available that contain at least one to two INL98s. Importantly however, micronutrient deficiencies and immune dysfunction in HIV-infected adults and children may only be restored when there is effective suppression of viral replication of HIV.

Implications for research

Adequately powered studies are still required to determine the efficacy and safety of some single and multiple micronutrient supplements in people with HIV infection to determine their short-term and long-term benefits. In view of the potential significance of preliminary results in HIV-infected populations or the proven benefits in HIV-uninfected populations some specific micronutrients warrant particular investigation, namely selenium, vitamin D and zinc. The optimal composition and dosage of various supplements requires investigation, as these can vary considerably among commercial supplements and therefore may not have equivalent effects. The cost-benefit or cost-effectiveness of nutritional interventions also requires evaluation.

Research participants should be diverse with respect to stage of disease, use of antiretroviral therapy, immune status, and nutritional status. The special needs of children, of HIV-exposed infants, and of pregnant and lactating women should be taken into account. 

Future research should also determine the effect of lifelong antiretroviral therapy on micronutrient concentrations, independent of inflammatory markers, and whether micronutrient supplements affect HIV-related outcomes in HIV-infected persons receiving HAART.

Nutritional interventions to improve the health and well-being of persons living with HIV/AIDS need to be optimised and research into identifying optimal interventions and operational strategies is therefore required.  Such research should not be to the detriment of antiretroviral treatment, as this remains the one intervention to date that has consistently been shown to reduce morbidity and mortality, and improve the nutritional status of adults and children infected with HIV/AIDS.

Acknowledgements

We are grateful to the South African Cochrane Centre, the editorial base of the HIV/AIDS Cochrane Review Group and the Cochrane HIV/AIDS Review Mentoring Programme for assistance and support in preparing this review.

We thank Dr Lize van der Merwe of the Biostatistics Unit of the Medical Research Council for providing statistical support.

Ms Claudia Naidu (CN), a research assistant to Mr Irlam, was supported by funding from the SA Cochrane Centre for preparing this update.

Data and analyses

Download statistical data

Comparison 1. Vitamin A in adults
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Maternal mortality1 Hazard Ratio (Random, 95% CI)Totals not selected
1.1 Twelve months1 Hazard Ratio (Random, 95% CI)0.0 [0.0, 0.0]
1.2 Twenty four months1 Hazard Ratio (Random, 95% CI)0.0 [0.0, 0.0]
2 Hospitalised at least once by 12 months post partum14495Risk Ratio (M-H, Random, 95% CI)0.96 [0.76, 1.22]
3 Infant death or HIV infection at 24 months14495Risk Ratio (M-H, Random, 95% CI)1.02 [0.94, 1.10]
4 Infant HIV infection at 24 months14495Risk Ratio (M-H, Random, 95% CI)0.99 [0.91, 1.09]
Analysis 1.1.

Comparison 1 Vitamin A in adults, Outcome 1 Maternal mortality.

Analysis 1.2.

Comparison 1 Vitamin A in adults, Outcome 2 Hospitalised at least once by 12 months post partum.

Analysis 1.3.

Comparison 1 Vitamin A in adults, Outcome 3 Infant death or HIV infection at 24 months.

Analysis 1.4.

Comparison 1 Vitamin A in adults, Outcome 4 Infant HIV infection at 24 months.

Comparison 2. Vitamin A in children
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 All-cause mortality3262Risk Ratio (M-H, Random, 95% CI)0.55 [0.37, 0.82]
2 Morbidity rates1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
2.1 All-cause morbidity1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 Diarrhoea1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2.3 Diarrhoea lasting a week or more1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2.4 Lower respiratory tract infection1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3 Child growth at 12 months1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
3.1 Wasting1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.2 Stunting1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 2.1.

Comparison 2 Vitamin A in children, Outcome 1 All-cause mortality.

Analysis 2.2.

Comparison 2 Vitamin A in children, Outcome 2 Morbidity rates.

Analysis 2.3.

Comparison 2 Vitamin A in children, Outcome 3 Child growth at 12 months.

Comparison 3. Vitamin D in adults
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Clinical severity score (TBscore) at 8 months171Mean Difference (IV, Random, 95% CI)0.10 [-0.53, 0.73]
2 All cause mortality at 12 months1131Risk Ratio (M-H, Random, 95% CI)1.15 [0.65, 2.02]
Analysis 3.1.

Comparison 3 Vitamin D in adults, Outcome 1 Clinical severity score (TBscore) at 8 months.

Analysis 3.2.

Comparison 3 Vitamin D in adults, Outcome 2 All cause mortality at 12 months.

Comparison 4. Vitamin D in children
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CD4 counts at 12 months156Mean Difference (IV, Random, 95% CI)115.0 [-74.26, 304.26]
2 Viral load at 12 months156Mean Difference (IV, Random, 95% CI)-0.10 [-0.63, 0.43]
Analysis 4.1.

Comparison 4 Vitamin D in children, Outcome 1 CD4 counts at 12 months.

Analysis 4.2.

Comparison 4 Vitamin D in children, Outcome 2 Viral load at 12 months.

Comparison 5. Zinc in adults
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Newborn outcomes1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
1.1 Duration of pregnancy1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
1.2 Birth weight1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
1.3 Birth length1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
1.4 Head circumference1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
1.5 Placental weight1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2 Low birth weight and prematurity1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
2.1 Preterm < 37 wks1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.2 Low birth weight < 2500 g1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.3 Small for gestational age1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3 Fetal loss and early child mortality1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
3.1 Miscarriage1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.2 Stillbirth1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.3 Fetal death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.4 Perinatal death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.5 Neonatal death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
4 Immune cell counts1 Mean Difference (IV, Random, 95% CI)Totals not selected
4.1 CD4 counts at 6 weeks post partum1 Mean Difference (IV, Random, 95% CI)0.0 [0.0, 0.0]
4.2 CD8 counts at 6 weeks postpartum1 Mean Difference (IV, Random, 95% CI)0.0 [0.0, 0.0]
4.3 CD3 counts at 6 weeks postpartum1 Mean Difference (IV, Random, 95% CI)0.0 [0.0, 0.0]
5 Viral load at 6 weeks postpartum1100Mean Difference (IV, Random, 95% CI)-0.32 [-0.69, 0.05]
6 Anthropometric outcomes1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
6.1 Overall weight gain during pregancy1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
6.2 Rate of weight gain during pregancy1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
6.3 Total change in MUAC durng pregnancy1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
7 Persistent diarrhoea at 2 weeks1104Risk Ratio (M-H, Random, 95% CI)1.05 [0.76, 1.44]
Analysis 5.1.

Comparison 5 Zinc in adults, Outcome 1 Newborn outcomes.

Analysis 5.2.

Comparison 5 Zinc in adults, Outcome 2 Low birth weight and prematurity.

Analysis 5.3.

Comparison 5 Zinc in adults, Outcome 3 Fetal loss and early child mortality.

Analysis 5.4.

Comparison 5 Zinc in adults, Outcome 4 Immune cell counts.

Analysis 5.5.

Comparison 5 Zinc in adults, Outcome 5 Viral load at 6 weeks postpartum.

Analysis 5.6.

Comparison 5 Zinc in adults, Outcome 6 Anthropometric outcomes.

Analysis 5.7.

Comparison 5 Zinc in adults, Outcome 7 Persistent diarrhoea at 2 weeks.

Comparison 6. Zinc in children
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Mortality at 9 months196Risk Ratio (M-H, Random, 95% CI)0.31 [0.07, 1.42]
2 Scheduled and illness visits1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
2.1 Scheduled visits; watery diarrhoea1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.2 Scheduled visits; pneumonia1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.3 Scheduled visits; URI1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.4 Scheduled visits; ear infection1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.5 All visits; watery diarrhoea1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.6 All visits; URI1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.7 All visits; pneumonia1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.8 All visits; ear infection1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3 Viral load at 9 months185Mean Difference (IV, Random, 95% CI)-0.10 [-0.40, 0.20]
4 CD4 % at 9 months185Mean Difference (IV, Random, 95% CI)1.0 [-2.87, 4.87]
Analysis 6.1.

Comparison 6 Zinc in children, Outcome 1 Mortality at 9 months.

Analysis 6.2.

Comparison 6 Zinc in children, Outcome 2 Scheduled and illness visits.

Analysis 6.3.

Comparison 6 Zinc in children, Outcome 3 Viral load at 9 months.

Analysis 6.4.

Comparison 6 Zinc in children, Outcome 4 CD4 % at 9 months.

Comparison 7. Selenium in adults
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Adverse pregnancy outcomes1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
1.1 Low birth weight < 2500 g1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.2 Very low birth weight < 2000 g1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.3 Preterm birth < 37 weeks1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.4 Preterm birth < 34 weeks1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.5 Low birth weight and preterm birth1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.6 Low birth weight and term birth1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.7 Small for gestational age1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2 Adverse pregnancy outcomes1 Mean Difference (IV, Random, 95% CI)Totals not selected
2.1 Mean birth weight (g)1 Mean Difference (IV, Random, 95% CI)0.0 [0.0, 0.0]
3 Perinatal mortality1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
3.1 Miscarriage before 28 wks.1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.2 Stillbirth between 28 wk and delivery1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.3 Fetal loss1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.4 Perinatal death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
4 Infant mortality1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
4.1 Neonatal death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
4.2 Infant death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
4.3 Neonatal or infant death1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
5 Adult mortality1913Risk Ratio (M-H, Random, 95% CI)1.00 [0.51, 1.98]
6 Viral load change 0-9 months1130Mean Difference (IV, Random, 95% CI)-0.33 [-0.51, -0.15]
7 CD4 count change from 0- 9 months1130Mean Difference (IV, Random, 95% CI)65.0 [56.80, 73.20]
8 Hospitalised for all conditions1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
9 Hospitalised for OIs and HIV-related conditions1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
Analysis 7.1.

Comparison 7 Selenium in adults, Outcome 1 Adverse pregnancy outcomes.

Analysis 7.2.

Comparison 7 Selenium in adults, Outcome 2 Adverse pregnancy outcomes.

Analysis 7.3.

Comparison 7 Selenium in adults, Outcome 3 Perinatal mortality.

Analysis 7.4.

Comparison 7 Selenium in adults, Outcome 4 Infant mortality.

Analysis 7.5.

Comparison 7 Selenium in adults, Outcome 5 Adult mortality.

Analysis 7.6.

Comparison 7 Selenium in adults, Outcome 6 Viral load change 0-9 months.

Analysis 7.7.

Comparison 7 Selenium in adults, Outcome 7 CD4 count change from 0- 9 months.

Analysis 7.8.

Comparison 7 Selenium in adults, Outcome 8 Hospitalised for all conditions.

Analysis 7.9.

Comparison 7 Selenium in adults, Outcome 9 Hospitalised for OIs and HIV-related conditions.

Comparison 8. Multiple supplements in non-pregnant adults
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Mortality3 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2 Mortality (all cause) by 48 weeks1 Hazard ratio (Random, 95% CI)Totals not selected
2.1 New Subgroup1 Hazard ratio (Random, 95% CI)0.0 [0.0, 0.0]
3 Mortality by 48 weeks (baseline CD4 < 200)1 Hazard ratio (Random, 95% CI)Subtotals only
4 Mortality by 48 weeks (baseline CD4 >= 200)1 Hazard ratio (Random, 95% CI)Totals not selected
5 Mortality by 48 weeks (baseline CD4 < 100)1 Hazard ratio (Random, 95% CI)Totals not selected
6 Mortality by 48 weeks (baseline CD4 >=100)1 Hazard ratio (Random, 95% CI)Totals not selected
7 New AIDS -defining infections116Risk Ratio (M-H, Random, 95% CI)3.11 [0.44, 22.00]
8 New HIV-associated infections1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
9 Recurrent HIV-associated infections2 Risk Ratio (M-H, Random, 95% CI)Totals not selected
10 New other infections1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
11 Recurrent other infections1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
12 Viral load at 12 months5401Mean Difference (IV, Fixed, 95% CI)-1.07 [-1.25, -0.90]
13 CD 4 counts at 12 months4358Mean Difference (IV, Fixed, 95% CI)1.31 [-86.14, 88.76]
14 CD 8 counts at 12 months2222Mean Difference (IV, Fixed, 95% CI)18.0 [-219.65, 255.65]
15 CD 3 counts (entire period)1172Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
16 Weight gain at 7 months1192Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
17 Viral load change : Baseline to 2 months196Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
18 CD4 cell count change: Baseline to 2 months2136Mean Difference (IV, Fixed, 95% CI)24.0 [5.75, 42.25]
19 Change in weight: Baseline to 7 months1192Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
20 Treatment failure after 1 month1322Odds Ratio (M-H, Fixed, 95% CI)0.73 [0.39, 1.37]
Analysis 8.1.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 1 Mortality.

Analysis 8.2.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 2 Mortality (all cause) by 48 weeks.

Analysis 8.3.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 3 Mortality by 48 weeks (baseline CD4 < 200).

Analysis 8.4.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 4 Mortality by 48 weeks (baseline CD4 >= 200).

Analysis 8.5.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 5 Mortality by 48 weeks (baseline CD4 < 100).

Analysis 8.6.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 6 Mortality by 48 weeks (baseline CD4 >=100).

Analysis 8.7.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 7 New AIDS -defining infections.

Analysis 8.8.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 8 New HIV-associated infections.

Analysis 8.9.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 9 Recurrent HIV-associated infections.

Analysis 8.10.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 10 New other infections.

Analysis 8.11.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 11 Recurrent other infections.

Analysis 8.12.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 12 Viral load at 12 months.

Analysis 8.13.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 13 CD 4 counts at 12 months.

Analysis 8.14.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 14 CD 8 counts at 12 months.

Analysis 8.15.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 15 CD 3 counts (entire period).

Analysis 8.16.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 16 Weight gain at 7 months.

Analysis 8.17.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 17 Viral load change : Baseline to 2 months.

Analysis 8.18.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 18 CD4 cell count change: Baseline to 2 months.

Analysis 8.19.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 19 Change in weight: Baseline to 7 months.

Analysis 8.20.

Comparison 8 Multiple supplements in non-pregnant adults, Outcome 20 Treatment failure after 1 month.

Comparison 9. Multiple supplements in pregnant and lactating women [maternal outcomes]
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Progression to stage 4 disease / death from AIDS-related causes1 Hazard ratio (Random, 95% CI)Totals not selected
2 Death from AIDS-related causes1 Hazard ratio (Random, 95% CI)Totals not selected
3 Progression to stage 4 disease1 Hazard ratio (Random, 95% CI)Totals not selected
4 Progression to stage 3 disease or higher1 Hazard ratio (Random, 95% CI)Totals not selected
5 HIV-related complications: thrush1 Hazard ratio (Random, 95% CI)Totals not selected
6 HIV-related complications: gingival erythema1 Hazard ratio (Random, 95% CI)Totals not selected
7 HIV-related complications: angular cheilitis1 Hazard ratio (Random, 95% CI)Totals not selected
8 HIV-related complications: oral ulcer1 Hazard ratio (Random, 95% CI)Totals not selected
9 HIV-related complications: reported mouth and throat ulcers1 Hazard ratio (Random, 95% CI)Totals not selected
10 HIV-related complications: painful tongue or mouth1 Hazard ratio (Random, 95% CI)Totals not selected
11 HIV-related complications: difficult or painful swallowing1 Hazard ratio (Random, 95% CI)Totals not selected
12 HIV-related complications: nausea and vomiting1 Hazard ratio (Random, 95% CI)Totals not selected
13 HIV-related complications: diarrhoea1 Hazard ratio (Random, 95% CI)Totals not selected
14 HIV-related complications: dysentery1 Hazard ratio (Random, 95% CI)Totals not selected
15 HIV-related complications: fatigue1 Hazard ratio (Random, 95% CI)Totals not selected
16 HIV-related complications: rash1 Hazard ratio (Random, 95% CI)Totals not selected
17 HIV-related complications: acute upper respiratory tract infection1 Hazard ratio (Random, 95% CI)Totals not selected
18 CD4 difference (baseline up to 3 months)1 Mean Difference (IV, Random, 95% CI)Totals not selected
19 CD4 difference (baseline to over 3 months)1 Mean Difference (IV, Random, 95% CI)Totals not selected
20 CD8 difference (baseline up to 3 months)1 Mean Difference (IV, Random, 95% CI)Totals not selected
21 CD8 difference (baseline to over 3 months)1 Mean Difference (IV, Random, 95% CI)Totals not selected
Analysis 9.1.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 1 Progression to stage 4 disease / death from AIDS-related causes.

Analysis 9.2.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 2 Death from AIDS-related causes.

Analysis 9.3.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 3 Progression to stage 4 disease.

Analysis 9.4.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 4 Progression to stage 3 disease or higher.

Analysis 9.5.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 5 HIV-related complications: thrush.

Analysis 9.6.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 6 HIV-related complications: gingival erythema.

Analysis 9.7.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 7 HIV-related complications: angular cheilitis.

Analysis 9.8.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 8 HIV-related complications: oral ulcer.

Analysis 9.9.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 9 HIV-related complications: reported mouth and throat ulcers.

Analysis 9.10.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 10 HIV-related complications: painful tongue or mouth.

Analysis 9.11.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 11 HIV-related complications: difficult or painful swallowing.

Analysis 9.12.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 12 HIV-related complications: nausea and vomiting.

Analysis 9.13.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 13 HIV-related complications: diarrhoea.

Analysis 9.14.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 14 HIV-related complications: dysentery.

Analysis 9.15.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 15 HIV-related complications: fatigue.

Analysis 9.16.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 16 HIV-related complications: rash.

Analysis 9.17.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 17 HIV-related complications: acute upper respiratory tract infection.

Analysis 9.18.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 18 CD4 difference (baseline up to 3 months).

Analysis 9.19.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 19 CD4 difference (baseline to over 3 months).

Analysis 9.20.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 20 CD8 difference (baseline up to 3 months).

Analysis 9.21.

Comparison 9 Multiple supplements in pregnant and lactating women [maternal outcomes], Outcome 21 CD8 difference (baseline to over 3 months).

Comparison 10. Multiple supplements in pregnant and lactating women [child outcomes]
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Foetal death (miscarriage+stillbirth)1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2 Total mortality by 24 months including foetal deaths1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
3 Mortality by 24 months among all live births1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
4 Mortality by 24 months among HIV-infected live births1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
5 Mortality by 24 months among HIV-infected infants at 6 weeks of age1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
6 Mean birthweight1 Mean Difference (IV, Random, 95% CI)Subtotals only
7 Birthweight < 2000 g1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
8 Birthweight < 2500 g1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
9 Preterm birth (<37 weeks)1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
10 Severe preterm birth (<34 weeks)1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
11 Small for gestational age1 Risk Ratio (M-H, Random, 95% CI)Subtotals only
12 CD4 count >= 3 months1 Mean Difference (IV, Random, 95% CI)Subtotals only
Analysis 10.1.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 1 Foetal death (miscarriage+stillbirth).

Analysis 10.2.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 2 Total mortality by 24 months including foetal deaths.

Analysis 10.3.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 3 Mortality by 24 months among all live births.

Analysis 10.4.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 4 Mortality by 24 months among HIV-infected live births.

Analysis 10.5.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 5 Mortality by 24 months among HIV-infected infants at 6 weeks of age.

Analysis 10.6.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 6 Mean birthweight.

Analysis 10.7.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 7 Birthweight < 2000 g.

Analysis 10.8.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 8 Birthweight < 2500 g.

Analysis 10.9.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 9 Preterm birth (<37 weeks).

Analysis 10.10.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 10 Severe preterm birth (<34 weeks).

Analysis 10.11.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 11 Small for gestational age.

Analysis 10.12.

Comparison 10 Multiple supplements in pregnant and lactating women [child outcomes], Outcome 12 CD4 count >= 3 months.

Appendices

Appendix 1. Clib search strategy

Appendix 2. PubMed search strategy

SearchMost Recent QueriesTimeResult
#5Search ("2009/06/01"[Publication Date] : "2010/01/29"[Publication Date]) AND (#1 AND #2 AND #3)04:12:5115
#4Search #1 AND #2 AND #304:10:19802
#3Search trace elements OR carotenoids OR vitamins04:09:59469770
#2Search (randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized [tiab] OR placebo [tiab] OR drug therapy [sh] OR randomly [tiab] OR trial [tiab] OR groups [tiab]) NOT (animals [mh] NOT humans [mh])04:09:222198940
#1Search HIV Infections[MeSH] OR HIV[MeSH] OR hiv[tw] OR hiv-1*[tw] OR hiv-2*[tw] OR hiv1[tw] OR hiv2[tw] OR hiv infect*[tw] OR human immunodeficiency virus[tw] OR human immunedeficiency virus[tw] OR human immuno-deficiency virus[tw] OR human immune-deficiency virus[tw] OR ((human immun*) AND (deficiency virus[tw])) OR acquired immunodeficiency syndrome[tw] OR acquired immunedeficiency syndrome[tw] OR acquired immuno-deficiency syndrome[tw] OR acquired immune-deficiency syndrome[tw] OR ((acquired immun*) AND (deficiency syndrome[tw])) OR "sexually transmitted diseases, viral"[MESH:NoExp]04:09:00253650

Appendix 3. EMBASE search strategy

No.QueryResultsDate
#5 #1 AND #2 AND #3 AND [humans]/lim AND [embase]/lim AND [2009-2010]/py229 Jan 2010
#4 #1 AND #2 AND #36829 Jan 2010
#3 trace AND ('elements'/de OR 'elements') OR 'carotenoids'/de OR 'carotenoids' OR 'vitamins'/de OR 'vitamins'6526329 Jan 2010
#2 random*:ti OR random*:ab OR factorial*:ti OR factorial*:ab OR cross?over*:ti OR cross?over:ab OR crossover*:ti OR crossover*:ab OR placebo*:ti OR placebo*:ab OR (doubl*:ti AND blind*:ti) OR (doubl*:ab AND blind*:ab) OR (singl*:ti AND blind*:ti) OR (singl*:ab AND blind*:ab) OR assign*:ti OR assign*:ab OR volunteer*:ti OR volunteer*:ab OR 'crossover procedure'/de OR 'crossover procedure' OR 'double-blind procedure'/de OR 'double-blind procedure' OR 'single-blind procedure'/de OR 'single-blind procedure' OR 'randomized controlled trial'/de OR 'randomized controlled trial' OR allocat*:ti OR allocat*:ab95489229 Jan 2010
#1 'human immunodeficiency virus infection'/exp OR 'human immunodeficiency virus'/exp OR hiv:ti OR hiv:ab OR 'hiv-1':ti OR 'hiv-1':ab OR 'hiv-2':ti OR 'hiv-2':ab OR 'human immunodeficiency virus':ti OR 'human immunodeficiency virus':ab OR 'human immuno-deficiency virus':ti OR 'human immuno-deficiency virus':ab OR 'human immunedeficiency virus':ti OR 'human immunedeficiency virus':ab OR 'human immune-deficiency virus':ti OR 'human immune-deficiency virus':ab OR 'acquired immune-deficiency syndrome':ti OR 'acquired immune-deficiency syndrome':ab OR 'acquired immunedeficiency syndrome':ti OR 'acquired immunedeficiency syndrome':ab OR 'acquired immunodeficiency syndrome':ti OR 'acquired immunodeficiency syndrome':ab OR 'acquired immuno-deficiency syndrome':ti OR 'acquired immuno-deficiency syndrome':ab30031929 Jan 2010

Appendix 4. Gateway search strategy

Search
Number
SearchItems
Found
#4 Search: (("HIV Infections"[MeSH] OR "HIV"[MeSH] OR hiv [tw] OR hiv-1*[tw] OR hiv-2*[tw] OR hiv1[tw] OR hiv2[tw] OR hiv infect*[tw] OR human immunodeficiency virus[tw] OR human immunedeficiency virus[tw] OR human immuno-deficiency virus[tw] OR human immune-deficiency virus[tw]) OR (((human immun*) AND (deficiency virus[tw])) OR acquired immunodeficiency syndrome[tw] OR acquired immunedeficiency syndrome[tw] OR acquired immuno-deficiency syndrome[tw] OR acquired immune-deficiency syndrome[tw] OR ((acquired immun*) AND (deficiency syndrome[tw])) OR "Sexually Transmitted Diseases, Viral"[MeSH:NoExp])) AND ((randomized controlled trial OR controlled clinical trial OR randomized OR placebo OR drug therapy OR randomly OR trial OR groups) NOT (animals [mh] NOT humans [mh])) AND (trace elements OR carotenoids OR vitamins) Limit: 2009/06/01:2010/01/29250 
#3 Search: trace elements OR carotenoids OR vitamins505930 
#2 Search: (randomized controlled trial OR controlled clinical trial OR randomized OR placebo OR drug therapy OR randomly OR trial OR groups) NOT (animals [mh] NOT humans [mh])3008540 
#1 Search: ("HIV Infections"[MeSH] OR "HIV"[MeSH] OR hiv [tw] OR hiv-1*[tw] OR hiv-2*[tw] OR hiv1[tw] OR hiv2[tw] OR hiv infect*[tw] OR human immunodeficiency virus[tw] OR human immunedeficiency virus[tw] OR human immuno-deficiency virus[tw] OR human immune-deficiency virus[tw]) OR (((human immun*) AND (deficiency virus[tw])) OR acquired immunodeficiency syndrome[tw] OR acquired immunedeficiency syndrome[tw] OR acquired immuno-deficiency syndrome[tw] OR acquired immune-deficiency syndrome[tw] OR ((acquired immun*) AND (deficiency syndrome[tw])) OR "Sexually Transmitted Diseases, Viral"[MeSH:NoExp])367979 

What's new

Last assessed as up-to-date: 25 February 2010.

DateEventDescription
29 November 2011AmendedParagraph added about the review being split into three reviews.

History

Protocol first published: Issue 4, 2003
Review first published: Issue 4, 2005

DateEventDescription
19 January 2011AmendedExternal source of support added.
9 November 2010Feedback has been incorporatedExternal reviewers' feedback incorporated into update
9 November 2010New citation required and conclusions have changedSubstantial update of the review.
9 November 2010New search has been performedSubstantial update.
30 September 2010New search has been performedInclusion of 16 additional trials, assessment of Risk of Bias using new ROB tool, and extensive updating of text.

Contributions of authors

J. Irlam (JI) initiated the review and contributed to all stages of the initial and updated review.
M. Visser (MV) assisted with all stages of the initial review.
N. Rollins (NR) assisted with study selection and commented on the report of the initial and updated review.
N. Siegfried (NS) mentored JI and MV and assisted with study selection, assessment of risk of bias, and commented on the report of the initial and updated review.

Declarations of interest

None.

Sources of support

Internal sources

  • SACC HIV/AIDS Mentoring Programme, South Africa.

  • South African Cochrane Centre, South Africa.

  • Medical Research Council, South Africa.

  • UCT Primary Health Care Directorate, South Africa.

  • UCT Child Health Unit, South Africa.

External sources

  • Cochrane Child Health Field, Canada.

  • World Health Organization, Switzerland.

    The World Health Organization's Department of Nutrition for Health and Development provided funding for the preparation of this updated review.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Allard 1998

Methods

Country: Canada

Setting: primary care physicians

Duration of recruitment: Apr 1995 - Aug 1996

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

Patients of participating physicians with stable HIV-infection.

EXCLUSION CRITERIA:
Active opportunistic infection, smoking, prior antioxidant therapy, hyperlipidaemia, kidney/liver dysfunction, intractable diarrhoea ( >= 6 liquid stools/d), vomiting, GI bleeding

Participants randomised: 49

- 47 M and 2 F

- mean age = 39 yrs

Participants analysed: 49

Loss to follow-up/ withdrawal: 0

Exclusions post-randomisation: 0

Interventions

INTERVENTION: 800 IU vitamin E, and 1000 mg vitamin C

CONTROL: placebo

DURATION: daily for 3 months.

Outcomes

PRIMARY OUTCOMES:
Viral load, oxidative stress (lipid peroxides, malondialdehyde, breath pentane)

SECONDARY OUTCOMES:
Plasma micronutrients (vitamin E, C, A carotenoids, zinc, selenium)

New and recurrent infections (AIDS-defining, HIV-associated and other)

Adverse eventsEpigastric discomfort
Notes

Number of patients on anti-retroviral therapy:
Supplement group: 22/ 23 (85%)
Control group: 18/ 26(78%)
Controlled diet 2 weeks prior to randomisation and throughout study period, and dietary counselling.

Source of funding: Canadian Foundation for AIDS Research

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table
Allocation concealment (selection bias)Low riskAllocation code not broken until post-analysis
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskHigh and unequal proportions of missing outcomes
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Arpadi 2009

Methods

Country: US

Setting: 4 hospital-based HIV treatment programmes

Duration of recruitment: 2004-2005

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

Perinatally-infected children and adolescents, aged 6-16 years.

EXCLUSION CRITERIA:
severe vitamin D deficiency

Participants randomised: 59

- 26 M and 33 F

- mean age = 10.4 yrs

Participants analysed: 56

Loss to follow-up/ withdrawal: 6

Exclusions post-randomisation: 0

Interventions

INTERVENTION: 100 000 IU vitamin D bimonthly, and 1000 mg calcium (2 chews) per day

CONTROL: double placebo

DURATION: bimonthly/ daily for 12 months.

Outcomes

PRIMARY OUTCOMES:

Serum 25 hydroxyvitamin D (25-OHD) concentrations

Serum and urine calcium

SECONDARY OUTCOMES:

HIV disease progression (CD4 count, viral load, ARV failure)

Adverse eventsNone
Notes

Participants were perinatally infected. HIV disease was classified by using Centers for Disease Control and Prevention criteria

Source of funding: National Institutes of Health

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised randomisation
Allocation concealment (selection bias)Low riskCentral allocation by study statistician
Blinding (performance bias and detection bias)
All outcomes
Low riskStudy personnel and participants were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskLow attrition (3 of 59 failed to complete the study)
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Austin 2006

Methods

Country: Canada

Setting: 22 outpatient clinics

Duration of recruitment: Aug. 1997 - May 1999

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

HIV positive, at risk for HIV disease progression based on ART status, CD4, and viral load; >18 years

EXCLUSION CRITERIA:

Continuing CD4 improvement on ART, severe pre-existing hepatic dysfunction, acute opportunistic infection, missed 2 earlier clinic visits without prior arrangement

Participants randomised:331

- 289 M and 42 F

- median age = 40 yrs (21-65) in treatment, 39 (22-63) in control

Participants analysed: 331

Loss to follow-up/ withdrawal/ death: 67

Discontinued intervention but remained in trial: 48

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Multivitamins (incl. vitamin A and trace elements) + natural mixed carotenoids (equivalent to 120000IU beta-carotene daily)

CONTROL: Multivitamins (incl. vitamin A and trace elements) without carotenoids

DURATION: four capsules daily for mean (s.d.)13 (6) months.

Outcomes

PRIMARY OUTCOMES: Mortality (time to death from AIDS-defining illness or any cause); time to new or recurrent AIDS-defining illness.

SECONDARY OUTCOMES: CD4 changes from baseline; viral load changes from baseline

Adverse eventsNone reported
Notes

Study closed early at mean 13 mo. follow-up

Number of patients on anti-retroviral therapy:
Supplement group: 137/ 165 (83%)
Control group: 148/ 166 (89%)

Source of funding: Canadian HIV Trials Network

Declared no conflicts of interest

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskCentralised block randomisation
Allocation concealment (selection bias)Low riskCentralised sequential randomisation
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskLosses to follow up explained and survival analysis used
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasHigh riskStudy closed prematurely due to interruption in supply of medication

Baeten 2002

Methods

Country: Kenya

Setting: hospital outpatient clinic

Duration of recruitment: Sep 1998 - Jun 2000

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

HIV-1 seropositive women attending Coast Provincial General Hospital outpatient clinics in Mombasa, Kenya

EXCLUSION CRITERIA:

age <18 or >45; pregnancy, or use of vitamin supplements or oral contraceptive pills

Participants randomised: 400

- 400 F

- median age = 28 yrs

Participants analysed: 354

Loss to follow-up/ withdrawal: 46

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Vitamin A (10 000 IU retinyl palmitate)

CONTROL: placebo

DURATION: daily for 6 weeks

Outcomes

PRIMARY OUTCOMES:

Vaginal HIV DNA and RNA

SECONDARY OUTCOMES:

Plasma viral load

CD4 and CD8 counts

Adverse eventsNone reported
Notes

Number of patients on anti-retroviral therapy:
Supplement group: 22/ 23 (85%)
Control group: 18/ 26 (78%)

Source of funding: Research grants from NIH, Univ of Washington, and Fogarty Int. Center; International AIDS Research and Training Program scholarships; Gen-Probe (reagents)

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated block randomisation
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskThose lost to follow up had more advanced HIV disease and vitamin A deficiency
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Bobat 2005

Methods

Country: South Africa

Setting: hospital outpatient clinic

Duration of recruitment: Mar - Dec 2003

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

Children aged 6-60 months with HIV-1 infection attending hospital clinic in Pietermaritzburg, South Africa

EXCLUSION CRITERIA: Children receiving ARVs

Participants randomised: 96

- 49 F and 46 M

- median age (zinc group): 40.1 months (27.4 to 48.4)

- median age (placebo group): 36.6 months (25 to 49.4)

Participants analysed: 85

Losses to follow-up/ withdrawal: 11

Exclusions post-randomisation: 0

Interventions

INTERVENTION:10 mg zinc sulphate

CONTROL: placebo

DURATION: daily for 6 months

Outcomes

PRIMARY OUTCOME:
Viral load

SECONDARY OUTCOMES:

% CD4 cells

Haemoglobin concentrations

Mortality

Morbidity (Watery diarrhoea; Pneumonia; URTI; Ear infection)

Adverse eventsNone
Notes

Concentrations of HIV-1 RNA in plasma were measured with a reverse transcriptase-PCR assay (COBAS AmpliPrep/Cobas
Amplicor HIV-1 Monitor version 1.5; Roche Molecular Systems, Branchburg, NJ, USA) at 1 month before randomisation, at the time of randomisation, and at 3, 6, and 9 months after the start of supplementation.

Source of funding: WHO; Johns Hopkins; Global Health Bureau; USAID; commercial (tablets)

Conflicts of interest: none

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised randomisation in blocks of size 8
Allocation concealment (selection bias)Unclear riskChildren allocated by investigator at hospital but concealment not explicitly stated
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskSmall loss to follow up and reasons given
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflict of interest

Burbano 2002

Methods

Country: USA

Setting: community-based clinic

Duration of recruitment: 1998 - 2000.

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

confirmed HIV, past or present use of illegal drugs, >= 18 years, adequate selenium status (> 85 microgram/l)

EXCLUSION CRITERIA: selenium deficient (< 85 microgram/l)

Participants randomised: 259

- 112 F

- median age = 40 yrs (range 24 to 54)

Participants analysed: 186

Loss to follow-up/ withdrawal: 73 at 12 months

Exclusions post-randomisation: 0

Interventions200 microgram selenium or placebo daily for 12 months.
Outcomes

PRIMARY OUTCOMES:

Number of hospital admissions
Type of hospital admissions
Risk of hospitalisation

SECONDARY OUTCOMES:

CD4 count

Hospitalisation cost
Plasma selenium

Adverse eventsNone reported
Notes

Number of patients on anti-retroviral therapy:
Selenium group: 64 (76%)
Control group: 60 (53%)

Number, type and duration of hospital admissions recorded 2 years prior and during study period. Medical records reviewed by team of physicians.

Source of funding: research grant and commercial (materials)

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskExclusions from the analysis (28%) not reported by intervention group
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Carcamo 2006

Methods

Country: Peru

Setting: Tertiary hospitals

Duration of recruitment: June 1998-Jan 2000

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

HIV-seropositive, persistent diarrhoea (>=7 days) without prior treatment

EXCLUSION CRITERIA: None stated

Participants randomised: 159

- 49 F and 110 M

- median age = 30 yrs (range 19-57) in Zinc group

- median age = 31 yrs (range 19-64) in placebo group

Participants analysed: 159

Loss to follow-up/ withdrawal: 51

Exclusions post-randomisation: 0

Interventions

INTERVENTION: zinc sulphate (100 mg)

CONTROL: placebo

DURATION: daily for 14 days

Outcomes

PRIMARY OUTCOMES:

Persistence of diarrhoea

Time until cessation of diarrhoea

SECONDARY OUTCOMES:

Plasma zinc and copper levels

Adverse eventsGastrointestinal symptoms (nausea, vomiting, abdominal pain, fever)
Notes

Sulfamethoxazole-trimethoprim prescribed for patients with enteric bacterial pathogens (23 in zinc group and 12 in placebo)

Source of funding:Fogarty IARTP grant; Univ. of Washington Center for AIDS Research; Centers for Disease Control and Prevention

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated blocked randomisation
Allocation concealment (selection bias)Low riskAssignment roll inaccessible to treatment allocators
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskHigh losses to follow up in both groups
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Coodley 1993

Methods

Country: USA

Setting: Hospital outpatient clinics

Duration of recruitment: not stated

Design: Randomised cross-over trial; no washout period

Participants

INCLUSION CRITERIA:

HIV-seropositive

EXCLUSION CRITERIA:

On other forms of vitamin A supplementation; significant hepatic or renal dysfunction; active opportunistic infection or fever

Participants randomised: 21

- 20 M and 1 F

- median age: not stated

Participants analysed: 17

Loss to follow-up/ withdrawal: 4

Exclusions post-randomisation: 0

Interventions

INTERVENTION: 60mg Beta-carotene

CONTROL: placebo

DURATION: three times daily for 4 weeks

Outcomes

PRIMARY OUTCOMES:

CD4 counts

SECONDARY OUTCOMES:

White blood cell count
Lymphocyte count
B-lymphocytes
Serum beta-carotene

Adverse eventsNo toxicity; skin discoloration in treatment group
Notes

CD4 count data reported as means and ranges**

16 patients received anti-retroviral therapy.

Source of funding: Hoffman La Roche Inc.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskLow attrition at 1 month; reasons given
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Coodley 1996

Methods

Country: USA

Setting: Hospital outpatient clinic and private practice

Duration of recruitment: not stated

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

HIV-seropositive; > 21 years

EXCLUSION CRITERIA:

Other forms of vitamin A supplementation 30 days prior to study; ART 60 days prior to study; significant hepatic or renal dysfunction; CD4 <50 or >600

Participants randomised: 72

- 63 M and 9 F

- median age: not stated

Participants analysed: 68 at 1 month; 50 at 3 months

Loss to follow-up/ withdrawal: 4 at 1 month; 22 at 3 months

Exclusions post-randomisation: 0

Interventions

INTERVENTION: 60mg Beta-carotene + multivitamins

CONTROL: placebo + multivitamins

DURATION: three times daily for 3 months

Outcomes

PRIMARY OUTCOMES:

CD4 counts

SECONDARY OUTCOMES:

T-cell counts

White blood cell counts

Natural killer cells

HIV p-24 antigen

Serum beta-carotene

Body weight

Karnofsky scores

Adverse eventsNone reported
Notes

Number of patients on anti-retroviral therapy:
Treatment group: 10 (28%)
Control group: 17 (47%)

Source of funding: research grant and commercial (materials)

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskHigh attrition at 3 months; reasons not given
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Coutsoudis 1995

Methods

Country: South Africa

Setting: Tertiary hospital study clinic

Duration of recruitment: April 1991-November 1993

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

Infants of HIV-infected women who had attended the antenatal clinic, delivered in hospital, and who lived within 10 miles of the hospital.

EXCLUSION CRITERIA:

Preterm infants

Participants randomised: 118

- 28 HIV-infected (13 in vitamin A group)

- 66 M and 52 F

- mean maternal age = 25 yrs (vitamin A) vs. 24.8 yrs (placebo)

Loss to follow-up:

17% (vitamin A) vs. 25% (placebo) at 6 months

36% (vitamin A) vs. 33% (placebo) at 12 months

58% (vitamin A) vs. 63% (placebo) at 18 months

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Vitamin A

CONTROL: placebo

DURATION: Repeat doses of 50 000 IU retinyl palmitate at 1 and 3 months and 100 000 IU at 6, 9,12 and 15 months

Outcomes

PRIMARY OUTCOMES:

Overall morbidity

SECONDARY OUTCOMES:

Acute diarrhoea
Persistent diarrhoea (>=7 days)
Hospitalised for diarrhoea
Thrush
Upper respiratory tract infection
Lower respiratory tract infection (LRTI)
Hospitalised for LRTI
Rash

Adverse eventsVomiting and bulging fontanelle
Notes

Monthly morbidity recall.

Multiple episodes of same condition in a single month counted as one episode (per 100 months).

Diagnosis of HIV infection in children based on positive HIV antibody test at 15 months (enzyme-linked immunosorbent assay
[ELISA], Abbott, N Chicago, Ill). Children who had lost maternal antibody by 15 months or sooner were diagnosed as uninfected.

Among the 11 deaths in children younger than 15 months, 9 were
diagnosed as HIV-infected on the basis of the following criteria: at least one HIV-related sign or symptom when last seen and death
from severe infection or persistent diarrhea beyond the first 4 weeks of life

Source of funding: Medical Research Council; University of Natal; Fogarty International Center; National Institute of Mental Health

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table
Allocation concealment (selection bias)Unclear riskThe capsules looked identical and were placed in number-coded envelopes from which they were removed
when appropriate.
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskSignificant losses to follow up insufficiently explained
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Fawzi 1998

Methods

Country: Tanzania

Setting: Hospital antenatal clinic

Duration of recruitment: Apr 1995 - Jul 1997

Design: Randomised trial with two-by-two factorial design

Participants

INCLUSION CRITERIA:

Pregnant women (12-27 weeks gestation) with confirmed HIV infection

EXCLUSION CRITERIA:

Non-resident in Dar es Salaam from recruitment until one year post-delivery

Participants randomised: 1085

- mean age = 24.7 yrs

Participants analysed: 1078

Loss to follow-up/ withdrawal: 54

Exclusions post-randomisation: 7 deaths

Interventions

INTERVENTION: Vitamin A alone (30 mg beta-carotene plus 5000 IU preformed vitamin A) OR vitamin A plus multivitamins (20mg vit B1, 20 mg vit B2, 25 mg vit B6, 100mg niacin, 50 microg vit B12, 500 mg vit C, 30 mg vit E and 0.8 mg folic acid) OR multivitamins without vitamin A

At delivery, women in both vitamin A groups were given an additional 200 000 IU vitamin A, while the other 2 groups received placebo.

CONTROL: placebo

DURATION: Daily for the duration of follow-up i.e. from enrolment until end of study (Aug. 2003).

Median follow-up was 71 months (IQR: 46 to 80) w.r.t. survival

Outcomes

PRIMARY OUTCOMES:

Maternal outcomes: Mortality and disease progression from AIDS-related causes; HIV-related complications; viral load; T-cell counts

Birth outcomes: Fetal death; low birth weight (< 2500g); preterm birth (< 37 weeks)

SECONDARY OUTCOMES:

Mortality among all live births and among HIV-infected infants Morbidity among infants (diarrhoea; respiratory tract infections)

Morbidity among mothers (hypertension; depression; social functioning)
Maternal weight gain during pregnancy
Postnatal child growth and development

CD4 counts of infants

Micronutrient status of infants

Haematologic status of children

Psychomotor development index
Mental development index

Adverse eventsNone reported
Notes

All women received 400mg ferrous sulphate and 5 mg folate daily, plus weekly doses of chloroquine antenatally

All infants received 100000 IU vitamin A at 6 months and 200000 IU every 6 months thereafter.

Vitamin A dropped from two of the regimens in Sept. 2000 and replaced with placebo due to safety concerns.

Source of funding: Fogarty International Center; National Institutes of Health; National Institute of Child Health and Human Development

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised factorial randomisation in blocks of 20
Allocation concealment (selection bias)Low riskSequentially numbered bottles coded identically
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskSurvival analysis used to impute missing outcome data
Selective reporting (reporting bias)Low riskRationale and design of study fully reported, and all outcomes were reported on
Other biasLow riskDeclared no conflict of interest

Fawzi 1999

Methods

Country: Tanzania

Setting: Hospital inpatient and outpatient follow-up

Duration of recruitment: April 1993 - March 1997.

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

Admitted to hospital for pneumonia; aged 6 - 60 months; no eye signs or symptoms of vitamin A deficiency

EXCLUSION CRITERIA:

treatment with vitamin A for 4 months prior to study entry; severe malnutrition; measles; pulmonary tuberculosis; diphtheria; whooping cough; xerophthalmia

Participants randomised:687

- 58 HIV-infected of 648 with known status (9%)

- 353 M and 295 F

- mean maternal age = 25.6 yrs (vitamin A group) vs. 26.2 (placebo)

Participants analysed: 648

Loss to follow-up/ withdrawal:76

Exclusions post-randomisation: 0

Interventions

INTERVENTION: vitamin A

CONTROL: placebo

DURATION: single dose on hospital admission, on day 2 and at 4 and 8 months after discharge (100,000 IU dose for infants; 200,000 IU for children)

Outcomes

PRIMARY OUTCOMES:

All-cause mortality

Cause-specific mortality (AIDS, diarrhoea, pneumonia, malaria, anaemia and other infections (measles, meningitis, dysentery, fever of unknown origin, malnutrition))

Diarrhoea
Acute respiratory infection

SECONDARY OUTCOMES:

Hospitalisation

Visits to health centre

Adverse eventsRespiratory infections; diarrhoea
Notes

Sera from children were tested for HIV antibodies by enzyme- linked immunosorbent assay and Western blot tests. For positive children <15 months of age, HIV infection was confirmed by amplified heat-denatured HIV-p24 antigen assays with confirmatory neutralization assays.

Cause of death ascertained by review of hospital records and home verbal autopsy questionnaire by two physicians. Discrepancies resolved by third physician.

Bi-weekly morbidity recall. No data provided on episodes of persistent diarrhoea or hospitalisation of HIV-infected children.

Source of funding: Thrasher Research Fund; International Development Research Center

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomisation in blocks of 20 insufficiently described
Allocation concealment (selection bias)Low riskVitamin A and placebo were dispensed out of a dropper from identical 25-ml opaque bottles that were labelled with one of four batch numbers. The batch number code was retained by the manufacturer until the end of the study
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskReasons for losses to follow up (n=76) not given
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Fawzi 2005

Methods

Country: Tanzania

Setting: Tertiary hospital antenatal clinic

Duration of recruitment: Sept 2000 - Oct 2002

Design: Randomised placebo-controlled trial

Participants

INCLUSION CRITERIA: Pregnant women (12-27 wk.), resident in Dar es Salaam for duration of study

EXCLUSION CRITERIA: None

Participants randomised: 400

- mean maternal age = 27 yrs

Participants analysed: 397

Loss to follow-up/ withdrawal: 3 deaths; 18 left study area after delivery

Exclusions post-randomisation: 3 left study area before delivery

Interventions

INTERVENTION: Zinc (25mg)

CONTROL: placebo

DURATION: daily until 6 weeks after delivery.

Outcomes

PRIMARY OUTCOMES:

Maternal haematological indicators at 6 weeks postpartum

Birth outcomes (duration, birthweight, birth length, head circumference, placental weight, preterm births, LBWs, SGAs, fetal loss, early child mortality

Maternal T-cell counts

Adverse eventsZinc had adverse effects on haemoglobin concentrations and other haematological indicators (RBC, PCV)
Notes

All women received weekly ferrous sulphate, folate, chloroquine phosphate, and multivitamin supplements (vitamins B, C, E).

Source of funding: National Institute of Child Health and Human Development

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskA randomisation list was prepared in blocks of 20
Allocation concealment (selection bias)Unclear riskAt enrolment, each eligible woman was assigned to the next numbered bottle of regimen.
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskLow loss to follow up, but CD4 data missing without explanation for a high proportion of both groups
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskStudy sponsors had no role in study design and reporting

Hanekom 2000

Methods

Country: USA

Setting: Hospital HIV clinic

Duration of recruitment: Not stated

Design: Randomised placebo-controlled trial

Participants

INCLUSION CRITERIA:

Clinic patients who were eligible for non-primary annual influenza vaccination

EXCLUSION CRITERIA:

Hypersensitivity to eggs, acute febrile illness, recent receipt of intravenous gammaglobulin or vaccination.

Participants randomised: 59 children

- M : F ratio = 0.6 (vitamin A) and 1.3 (placebo)

- median (range) age = 84 (31-209) months in vitamin A group; 77 months (25-142) in placebo group

Participants analysed: 59

Loss to follow-up/ withdrawal: 1

Exclusions post-randomisation: 0

Interventions

INTERVENTION: vitamin A (200 000 IU retinyl palmitate)

CONTROL: placebo

DURATION: daily for 2 days.

Outcomes

PRIMARY OUTCOMES:

Viral load changes after vaccination

Antibody levels (H1N1, H3N2) after vaccination

SECONDARY OUTCOMES:

T-cell counts

Vitamin A levels

Adverse eventsNone reported
Notes

HIV viral load (branch-chain DNA amplification, Chiton Corp, Emeryville, CA; lower limit of detection 500 copies/mL) measured on study days 0, 14, 28, and 42

All children received inactivated influenza vaccine on study day 14.

All were receiving anti-retroviral therapy. An unknown number who had ARV changes were excluded from viral load analysis.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated by pharmacy
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskOne exclusion due to incomplete follow up
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on funding sources or conflicts of interest

Humphrey 1999

Methods

Country: USA

Setting: HIV Clinic

Duration of recruitment: Jan - July 1996.

Design: Randomised placebo-controlled safety trial

Participants

INCLUSION CRITERIA: 18 to 45 years, CD4 > 200

EXCLUSION CRITERIA: pregnant or breastfeeding

Participants randomised: 40 women

- mean age (SD) in years = 36.2 (5.6) in vitamin A group and 33.2 (5.6) in placebo group

Participants analysed: 39

Loss to follow-up/ withdrawal: 1

Exclusions post-randomisation: 0

Interventions

INTERVENTION: 300 000 IU vitamin A.

CONTROL: placebo

DURATION: Single dose.

Outcomes

PRIMARY OUTCOMES:

Viral load

T-cell subsets (%CD4; %CD8 which are CD38+)

SECONDARY OUTCOMES:

Vitamin A status

Adverse eventsSigns or symptoms of toxicity (headache, nausea, vomiting, diarrhoea, fever)
Notes

Number of patients on anti-retroviral therapy:
vitamin A group: 12 (60%)
Control group: 7 (35%)

Source of funding: Paediatric AIDS Foundation grant

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskOne lost to follow up
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflict of interest

Hurwitz 2007

Methods

Country: USA

Setting: University clinic

Duration of recruitment: June 2001 - July 2005

Design: Placebo controlled trial

Participants

INCLUSION CRITERIA: aged 18-55 years; no history of major systemic disorders related to HIV; pre-menopausal and non - pregnant

EXCLUSION CRITERIA: on treatment for chronic conditions; selenium deficient

Participants randomised: 310

- mean age = 40.5 yrs

Participants analysed: 262

- 179 M and 86 F

Loss to follow-up/ withdrawal: 88

Exclusions post-randomisation: 48 pre-treatment

Interventions

INTERVENTION:Selenium (200 micrograms)

CONTROL: placebo

DURATION: daily for 9 months

Outcomes

PRIMARY OUTCOMES:

Viral load

CD4 count

Serum selenium

Adverse eventsNone
Notes

Patients on ARV: 105/141 (74%) in Se group; 87/121 (72%) in placebo group

Preliminary analysis at 9 months of an 18-month trial

Source of funding: National Institutes of Health

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised block randomisation
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskHigh unexplained losses to follow up, balanced between groups. Imputational analyses conducted.
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflict of interest

Hussey 1996

Methods

Country: South Africa

Setting: HIV clinic at a children's hospital

Duration of recruitment:1994-1995.

Design: Randomised placebo-controlled trial.

Lost to follow-up at 2 months:
Total sample: 1 (3%)

Intention-to-treat: not performed.

Participants

INCLUSION CRITERIA:

Child attendees at HIV clinic

EXCLUSION CRITERIA:

Acute infections, fever

Participants randomised: 75

- mean age = 17 mo.

Participants analysed: 75

Loss to follow-up/ withdrawal: 0

Exclusions post-randomisation: 0.

Interventions

INTERVENTION: 200 000 IU vitamin A

CONTROL: placebo

DURATION: daily for 2 days.

Outcomes

PRIMARY OUTCOMES:

T-Cell counts (absolute; CD4; CD56; CD29)

SECONDARY OUTCOMES:

Vitamin A levels

Adverse eventsNone reported
Notes

Conference abstract only

Children who received vitamin A were more immuno-suppressed

50% of children had vitamin A < 20 microgram/dl.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol and full report not available
Other biasUnclear riskDid not declare on funding sources or conflicts of interest

Jiamton 2003

Methods

Country: Thailand

Setting: Outpatient clinic

Duration of recruitment: Mar 2000 - Jan. 2001

Design: Placebo-controlled trial

Duration of recruitment: Mar 2000- Jan 2001.

Participants

INCLUSION CRITERIA: older than 18 years; 50<CD4<550;

EXCLUSION CRITERIA:
taking ARV or micronutrients for during month prior to enrollment

Participants randomised: 481

- 189 M and 292 F

- mean age = 32 yrs

Participants analysed: 481

Loss to follow-up/ withdrawal: 79 at 48 weeks

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Micronutrient supplement (3000 micrograms vitamin A, 6mg beta-carotene, 20 micrograms vitamin D, 80 mg vitamin E , 180 micrograms vitamin K, 400 mg vitamin C, 24mg vitamin B1, 15 mg vitamin B2, 40 mg vitamin B6, 30 microg vitamin B12, 0.1 mg folic acid, 40 mg pantothenic acid , 10 mg iron, 200 mg magnesium, 8 mg manganese, 30 mg zinc, 300 micrograms iodine, 3 mg copper, 400 micrograms selenium, 150 micrograms chromium, 60 mg cysteine)

CONTROL: placebo

DURATION: twice daily for 48 weeks.

Outcomes

PRIMARY OUTCOMES:

Mortality
Hospital admissions

SECONDARY OUTCOMES:

CD4 counts
Viral load

Adverse eventsA total of 137 minor adverse events reported (dizziness, drowsiness, nausea, rash, urine discolouration)
NotesSource of funding: Nestle Foundation; Vitabiotics
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskCentralised randomisation in blocks of 10
Allocation concealment (selection bias)Low riskInterventions packaged in identical coded bottles
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskThose lost to follow-up were sicker than those who remained in the trial (baseline median CD4 counts among those lost did not differ between groups. Survival analysis used to address missing outcome data.
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared independence of study investigators declared

Kaiser 2006

Methods

Country: USA

Setting: Four HIV study centres

Duration of recruitment: Jan. 2002- May 2003

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA:

On stable HAART regimen for >= 3 months; had developed symptoms of mitochondrial toxicity (distal symmetrical polyneuropathy (DSP)) from either stavudine and/or didanosine

EXCLUSION CRITERIA:

Pregnant; on treatment for active opportunistic infection or malignancy; vitamin B12 deficient; already taking more than one micronutrient pill per day

Participants randomised: 40

- 35 M and 5 F

- mean age = 46 yrs

Participants analysed: 40

Loss to follow-up/ withdrawal: 0

Exclusions post-randomisation: 0

Interventions

INTERVENTION:

Micronutrient supplement (33 ingredients)

CONTROL: placebo

DURATION: 2x daily for 12 weeks

Outcomes

PRIMARY OUTCOMES:

CD4 counts

Viral load

SECONDARY OUTCOMES:

Metabolic parameters

Neuropathy symptoms (DSP) score

General health status

Adverse eventsNone
NotesSource of funding:Bristol Myers-Squibb
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskMethod of block randomisation method not described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow up
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflict of interest

Kelly 1999

Methods

Country: Zambia

Setting: Home care service of tertiary hospital

Duration of recruitment: not stated

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA:

adults with persistent diarrhoea for more than 1month

EXCLUSION CRITERIA:

< 18 years, pregnancy, administration of antibiotics in the week prior to recruitment, Karnofsky scores > 80 or <50.

Participants randomised: 135

- 79 M and 56 F

- median age = 32.5 yrs (micronutrient); 34 (placebo)

Participants analysed: 106

Loss to follow-up/ withdrawal: 29

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Micronutrient supplement (10 500 IU vitamin A, 300mg vitamin C, 300mg vitamin E, 150 microg Selenium and 200mg Zinc sulphate)
Both treatment groups also received 5mg folic acid and 800mg albendazole twice daily.

CONTROL: placebo

DURATION: daily for 2 weeks

Outcomes

PRIMARY OUTCOMES:

Recovery from diarrhoea

- patient weeks with and without diarrhoea during 12 weeks follow up

- remission at 4 weeks

All cause mortality during first 4 weeks

Change in Body Mass Index and MUAC

Change in Karnofsky score

SECONDARY OUTCOMES:

Changes in CD4 and CD8 counts at 4 weeks
Changes in serum vitamin A and E after 4 weeks

Adverse eventsNone
NotesSource of funding: Smithkline Beecham
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskMethod not described
Allocation concealment (selection bias)Unclear riskNot described
Blinding (performance bias and detection bias)
All outcomes
High riskMicronutrient and placebo capsules were not identical; unclear whether providers and assessors were blinded
Incomplete outcome data (attrition bias)
All outcomes
High risk25% of patients were lost to follow-up due to death and the tradition of going back to the family home when terminally ill.
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflicts of interest.

Kupka 2008

Methods

Country: Tanzania

Setting: antenatal clinics

Duration of recruitment: Sept 2003 - July 2005

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

HIV-infected pregnant women (12-27 weeks gestation) seeking care

EXCLUSION CRITERIA:
Non-residents in Dar-es-Salaam or those not intending to stay until at least one year post-delivery

Participants randomised: 915

- mean age = 27.5 yrs

Participants analysed: 913

Loss to follow-up/ withdrawal: 0

Exclusions post-randomisation: 2

Interventions

INTERVENTION: Selenium (200 microgram selenomethionine)

CONTROL: placebo

DURATION: daily until 6 months post-delivery

Outcomes

PRIMARY OUTCOMES:

Viral load; CD4 counts; genital shedding of HIV-infected cells; risk of mastitis; birth weight; adverse pregnancy outcomes; maternal and infant mortality

SECONDARY OUTCOMES:

Haemoglobin concentrations; maternal morbidity

Adverse eventsNone
NotesSupported by the National Institute of Child Health and Human Development (NICHD R24 043555-05).
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised block randomisation
Allocation concealment (selection bias)Low riskStickers used to conceal numeric allocation codes
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators
Incomplete outcome data (attrition bias)
All outcomes
Low riskData on birth outcomes and birth weight missing for a low proportion of participants, and survival analysis used for mortality outcomes
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflict of interest

Luabeya 2007

Methods

Country: South Africa

Setting: Five rural primary care clinics

Duration of recruitment: June 2003 - Oct 2004

Design: Controlled trial

Participants

INCLUSION CRITERIA:

children 4-6 months old

EXCLUSION CRITERIA:

underweight for age; nutritional oedema; persistent diarrhoea; taking vitamin or micronutrient supplements in past month

Participants randomised: 373

- 32 HIV-infected; 154 born to HIV-infected mothers; 187 born to HIV-uninfected mothers

- 173 M and 162 F

- mean age = 5.5 months

Participants analysed: 335

Loss to follow-up/ withdrawal: 88

Exclusions post-randomisation: 0

Interventions

INTERVENTIONS:

Zinc (10mg) and multiple micronutrients (B vitamins; vitamins C, D, E, K; copper, iron, iodine) vs. zinc and vitamin A (1250 IU)

CONTROL:

Vitamin A

DURATION: daily for median duration of 14.9 months.

Outcomes

PRIMARY OUTCOMES:

Percentage of days of diarrhoea per child

SECONDARY OUTCOMES:

Severity of diarrhoea

Percentage of weeks with upper respiratory symptoms

Percentage of children who ever had pneumonia (maternal and field worker reports)

Adverse eventsVomiting reported post-supplementation in 9 children by fieldworkers
Notes

HIV testing of children was done between ages of 4 and 6 months using a quantitative HIV RNA assay (Nuclisens HIV-1 QT, Organon Teknika or Nuclisens EasyQ HIV-1, Biomerieux, Boxtel, The Netherlands)

Source of funding: National Institute of Health; Wellcome Trust

Declared no conflicts of interest

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised randomisation in blocks of 6
Allocation concealment (selection bias)Low riskPre-packed sequentially numbered study supplements
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing data addressed used appropriate statistical methods
Selective reporting (reporting bias)Low riskAll outcomes of interest were reported on
Other biasUnclear riskDelay in shipment of supplements prevented 243 children from receiving supplements for 11 weeks

Range 2006

Methods

Country: Tanzania

Setting: Five district health facilities

Duration of recruitment: August 2001 to July 2002

Design: Placebo-controlled 2x2 factorial trial

Participants

INCLUSION CRITERIA:

HIV-positive and HIV-negative persons aged >=15 years with sputum-positive pulmonary tuberculosis (new or relapsed cases)

EXCLUSION CRITERIA:

Patients who defaulted TB chemotherapy or those who remained smear-positive on chemotherapy (failure cases) and those with serious  tuberculosis or other disease unlikely to survive; pregnant and lactating women.

Participants randomised: 530

- 213 HIV-infected

- 325 M and 205 F

- mean age = 35.4 yrs

Participants analysed: 499

Loss to follow-up/ withdrawal: 77 within 244 days post-treatment

Exclusions post-randomisation: 31

Interventions

INTERVENTIONS:

Micronutrient supplement contained vitamin A (1.5mg), vitamin B1 (20 mg), vitamin B2 (20 mg), vitamin B6 (25 mg), vitamin B12 (50mg), folic acid (0·8 mg), niacin (40 mg), vitamin C (200 mg), vitamin E (60 mg), vitamin D3 (5mg), selenium (0·2 mg) and copper (5 mg), and Zn tablets contained 45 mg elementary Zn

CONTROL: placebo (2x2 factorial)

DURATION: daily for 8 months.

All patients received a standard 8 month TB chemotherapy regimen.

Outcomes

PRIMARY OUTCOMES:

All-cause mortality at 8 months

SECONDARY OUTCOMES:

Viral load

CD4 counts

Weight gain

Adverse eventsNone reported
NotesSource of funding: Danish International Development Assistance
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised randomisation
Allocation concealment (selection bias)Low riskSealed envelopes, codes unbroken until post-analysis
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskSurvival analysis was done but there were significant imbalances in losses between intervention groups
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflict of interest

Semba 1998

Methods

Country: USA

Setting: Community-based clinic

Duration of recruitment: Not stated

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA:

HIV-infected Intravenous Drug Users participating in ALIVE (AIDS Linked to Intravenous Experiences) Cohort (n=630); >= 18 years; not taking vitamin A supplements

EXCLUSION CRITERIA:

CD4 > 500 cells/mm3; pregnancy.

Participants randomised: 120

- 89 M and 31 F

- mean age = 38.2 yrs

- 50% treatment group vs. 43% placebo group on ART

Participants analysed: 120

Loss to follow-up/ withdrawal: 8.3% at 4 weeks

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Single dose of 200 000 IU vitamin A

CONTROL: placebo

Outcomes

PRIMARY OUTCOMES:

Viral load
CD4 count

SECONDARY OUTCOMES:
Serum vitamin A

Adverse eventsNone reported
NotesSource of funding: USAID
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table in blocks of 10
Allocation concealment (selection bias)Low riskSequentially numbered envelopes used to conceal allocation
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced across groups
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflicts of interest.

Semba 2005

Methods

Country: Uganda

Setting:Hospital clinic

Duration of recruitment:Jan. 1995 - June 1998

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA:

Children aged 6 months; resident near hospital for duration of trial

EXCLUSION CRITERIA:

Evidence of vitamin A deficiency

Participants randomised: 181 at age 15 months

- 90 M and 91 F

- mean age = 15 months

Participants analysed: 168

Loss to follow-up/ withdrawal: 0

Exclusions post-randomisation: 0

Interventions

INTERVENTION:

200 000 IU vitamin A

CONTROL: placebo

DURATION: every 3 months from 15 to 36 months.

Outcomes

PRIMARY OUTCOMES:

Mortality

SECONDARY OUTCOMES:
Morbidity (diarrhoea, cough, fever, ear discharge, hospitalisation)

Adverse eventsNone reported
Notes

Infants of HIV-positive women were tested for HIV-1 infection by using a p24 antigen assay (Coulter Diagnostics, Hialeah, FL, USA) until June 1996, after which time infants were tested for HIV-1 by using a qualitative assay for HIV-1 DNA polymerase
chain reaction (PCR; HIV-1 Amplicor, Roche Diagnostics, Indianapolis, IN, USA) or a quantitative assay for HIV-1 RNA PCR (Roche Amplicor Monitor, Roche Diagnostics, Branchburg, NJ, USA). All results were confirmed by serologic testing at ages 15 to 18 mo and by quantitative HIV-1 RNA PCR at age 15 mo (Roche Amplicor Monitor).

All children received daily prophylactic sulfamethoxazole- trimethoprim therapy

Source of funding: National Institutes of Health

Declared no conflicts of interest

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated, random codes
Allocation concealment (selection bias)Low riskSequentially numbered pill cards were used
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants, paediatrician and clinic staff were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskSurvival analysis used to account for differences in duration of follow up due to early termination of the trial
Selective reporting (reporting bias)Low riskStudy protocol not available but all stated outcomes of interest reported on
Other biasLow riskTrial stopped early due to change in national guideline on vitamin A supplementation

Semba 2007a

Methods

Country: USA

Setting: Study clinic

Duration of recruitment: Sept. 2002 - Aug. 2005

Design: Controlled trial

Participants

INCLUSION CRITERIA:

women >=18 years; history of injection drug use (IDU) within past 10 years; hepatitis C (HCV) antibody-positive; Karnofsky status >80%; serum ferritin <200 ng/ml

EXCLUSION CRITERIA:

Pregnant; history of liver failure, renal disease, interferon therapy for HCV; haemochromatosis; blood disorders

Participants randomised: 458

- mean age = 40 yrs

- 138 (30.1%) HIV-positive

Participants analysed: 115 at 12 months

Loss to follow-up/ withdrawal:151 (33%)

Exclusions post-randomisation: 0

Interventions

INTERVENTION:Micronutrients with iron (18 mg)

CONTROL: Micronutrients only

DURATION: daily for 12 months.

Outcomes

PRIMARY OUTCOMES:

Haemoglobin

Iron status

Plasma HCV

Viral load

Liver enzymes

Adverse eventsNot stated
Notes

On HAART: 27/69 (intervention) and 23/69 (control)

Trial stopped early due to slow recruitment.

Source of funding: National Institute on Drug Abuse; National Institute on Nursing Research

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised randomisation
Allocation concealment (selection bias)Low riskPre-packed sequentially numbered study supplements
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskHigh loss to follow up (27.7%) in both groups, and not reported by HIV status
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasUnclear riskDid not declare on conflicts of interest.

Semba 2007b

Methods

Country: Malawi

Setting: Eight community health centres

Duration of recruitment: July 1999 - October 2004

Design: Placebo-controlled

Participants

INCLUSION CRITERIA:

HIV-positive and HIV-negative adults with smear-positive pulmonary tuberculosis (new cases)

EXCLUSION CRITERIA:

Prior or current TB chemotherapy, prior vitamin supplements

Participants randomised: 1148

- 829 HIV-positive

- 336 M and 493 F

- mean age = 34 yrs

Participants analysed: 1148

Loss to follow-up: 103 in HIV-positive group (50 and 53 in micronutrient and placebo groups, respectively)

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Micronutrient supplement (vitamin A, C, D, E, B6, B12, Riboflavin, Thiamine, Niacin, folate, zinc, iodine, selenium)

CONTROL: placebo

DURATION: daily for 24 months..

All patients received a standard 8 month TB chemotherapy regimen.

Outcomes

PRIMARY OUTCOMES:

All-cause mortality

SECONDARY OUTCOMES:

Serum vitamin A, vitamin E and selenium

Adverse eventsNot reported
NotesSource of funding: National Institutes of Health and the Fogarty International Centre.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskBlocked randomisation
Allocation concealment (selection bias)Low riskpre-packed sequentially numbered study supplements
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskHigh unexplained loss to follow up but survival analysis was used
Selective reporting (reporting bias)Low riskStudy protocol available; all outcomes of interest were reported on
Other biasUnclear riskDid not declare on conflicts of interest.

Villamor 2008

Methods

Country: Tanzania

Setting: Five outpatient TB clinics

Duration of recruitment: April 2000 - April 2005

Design: Placebo-controlled

Participants

INCLUSION CRITERIA:

age 18-65 years; non-pregnant; no anti-TB treatment for > 4 weeks in previous year; Karnofsky score >=40%; plan to stay in Dar es Salaam for 2 years

EXCLUSION CRITERIA: none

Participants randomised: 887

- 471 HIV-positive

- 273 M and 198 F

- mean age = 34 yrs

Participants analysed: 1148

Loss to follow-up: 67 in HIV-positive group (33 and 34 in micronutrient and placebo groups, respectively)

Exclusions post-randomisation: 0

Interventions

INTERVENTION: Micronutrient supplement (retinol; vitamins B1, B2, B6, B12; niacin; vitamin C; vitamin E; folic acid; selenium)

CONTROL: placebo

DURATION: daily for 24 months..

All patients received DOTS anti-TB chemotherapy.

Outcomes

PRIMARY OUTCOMES:

Culture negativity at 1 month after initiation of treatment; mortality during at least 24 months of follow-up; TB recurrences.

SECONDARY OUTCOMES:

Changes from baseline in viral load, CD4 cell counts, and body weight.

Adverse eventsNone reported
NotesSource of funding: National Institute of Allergy and Infectious Diseases; US Department of Agriculture
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskcomputer-generated permuted blocks of 20, stratified by HIV status
Allocation concealment (selection bias)Unclear riskAll clinical and research staff were unaware of the subjects’ treatment assignment, but insufficient information provided
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskReasons not given for losses to follow-up, although appropriate statistical analyses were used
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflicts of interest

Wejse 2009

Methods

Country: Guinea-Bissau

Setting: TB clinics in urban disease surveillance site

Duration of recruitment: Nov 2003 - Dec 2005

Design: Placebo-controlled, parallel group

Participants

INCLUSION CRITERIA:

TB patients starting anti-TB treatment, >=15 years.

EXCLUSION CRITERIA:
None

Participants randomised: 367

- 222 M and 143 F

- mean age = 37.5 yrs

- 131 HIV-infected

Participants analysed: 365

Loss to follow-up/ withdrawal: 84

Exclusions post-randomisation: 2

Interventions

INTERVENTION: 100 000 IU cholecalciferol (vitamin D)

CONTROL: placebo

DURATION: at inclusion; 5 and 8 months after inclusion

Outcomes

PRIMARY OUTCOMES:
Reduction in a clinical severity score (TB score)

SECONDARY OUTCOMES:
12-month mortality

Adverse eventsMinor events reported; no difference between groups
NotesSource of funding: Aarhus University Hospital; Danish Research Council for Developmental Research
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated sequence
Allocation concealment (selection bias)Low riskIdentical, sequentially numbered containers were used.
Blinding (performance bias and detection bias)
All outcomes
Low riskPatients, staff and researchers
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskLoss to follow up unknown in HIV+ subgroup; survival analysis used
Selective reporting (reporting bias)Low riskProtocol available; all outcomes of interest reported on
Other biasLow riskHIV subgroup analyses not pre-specified but proportions were equally distributed; funder and provider had no role in study design

ZVITAMBO 2006

Methods

Country: Zimbabwe

Setting:Maternity clinics and hospitals

Duration of recruitment:Nov. 1997 - Jan. 2000

Design: Randomised trial with two-by-two factorial design

Participants

INCLUSION CRITERIA: Mother-infant pairs; HIV+ mothers; singleton infant with birthweight >=1500 g; resident in Harare after delivery

EXCLUSION CRITERIA: Mother or infant with acute life-threatening condition

Participants randomised: 4495 mothers

- mean age = 25.6 yrs

Participants analysed: 4495 infants of HIV+ mothers at baseline; 2582 infants assessable at 6 weeks for postnatal HIV infection

Loss to follow-up/ withdrawal: 139 infants excluded from 24-month mortality analysis due to loss to follow-up, death from unnatural causes, or missing dates

Exclusions post-randomisation: 0

Interventions

INTERVENTION:Single dose of maternal vitamin A (400 000 IU) and infant vitamin A (50 000 IU) OR maternal dose only OR infant dose only

CONTROL: placebo

DURATION: single dose <=96 hours after delivery

Outcomes

PRIMARY OUTCOMES:

Postnatal mother-to-child transmission (MTCT) of HIV

Child HIV-free survival

Infant mortality up to 24 months

Adverse eventsInfant mortality at 24 mo. approx. double in supplemented infants who were HIV negative at 6 weeks
Notes

Follow-up period curtailed due to economic conditions in Zimbabwe

Source of funding: Canadian International Development Agency, USAID, Gates Foundation; Rockefeller Foundation; BASF

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised factorial randomisation in blocks of 12
Allocation concealment (selection bias)Low riskSealed envelopes and encrypted computer files
Blinding (performance bias and detection bias)
All outcomes
Low riskParticipants and investigators were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskSurvival analysis used to include censored data
Selective reporting (reporting bias)Unclear riskInsufficient information; study protocol not available
Other biasLow riskDeclared no conflict of interest

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Austin 2003Trial terminated prematurely due to unstable intervention.
Constans 1996Pilot study; vitamin A group had more advanced disease at baseline
Kennedy 2000Included in Wiysonge 2005 review:
Vitamin A supplementation for reducing the risk of mother-to-child transmission of HIV infection
Kumwenda 2002Included in Wiysonge 2005 review:
Vitamin A supplementation for reducing the risk of mother-to-child transmission of HIV infection
Mehta 2009Observational analysis of perinatal outcomes and mortality vs. maternal vitamin D status in Tanzanian trial (Fawzi et al 1998)
Shor-Posner 2003No outcomes of relevance to this review

Characteristics of studies awaiting assessment [ordered by study ID]

Baum 2010

Methods

Country: USA

Setting:hospital outpatient clinic

Duration of recruitment:March 2002 - December 2005

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA: HIV-infected adults with low plasma zinc levels (<0.75 mg/L) and no history of endocrine or psychiatric disorders

EXCLUSION CRITERIA: premenopausal women who were pregnant or had an intention to become pregnant; plasma zinc levels <=0.35 mg/L at any time during the study

Participants randomised: 231

- 62 F and 169 M

- mean age = 42.7 yrs

Participants analysed:200

Loss to follow-up/ withdrawal:31

Exclusions post-randomisation: 0

Interventions

INTERVENTION:12 mg of elemental zinc for women; 15 mg for men

CONTROL: placebo

DURATION: 18 months

Outcomes

PRIMARY OUTCOMES:HIV disease progression, specifically immunological failure (CD4 < 200)

SECONDARY OUTCOMES:HIV viral load, morbidity, and mortality

Notes 

CIGNIS 2010

Methods

Country: Zambia

Setting:Public health sector clinic

Duration of recruitment:October 2005 to July 2009

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA: healthy 6 month old infants

EXCLUSION CRITERIA: no parental consent

Participants randomised:743

- mean age = 6 months

Participants analysed:576

Loss to follow-up/ withdrawal:167

Exclusions post-randomisation: 0

Interventions

INTERVENTION:micronutrient fortified porridge

CONTROL: conventionally fortified porridge

DURATION:12 months

Outcomes

PRIMARY OUTCOMES:stunting at age 18 months

SECONDARY OUTCOMES:hospital referral; death

Notes 

Mda 2010

Methods

Country: South Africa

Setting: academic hospital

Duration of recruitment: November 2005 and May 2007

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA: HIV-infected children aged between 4 mo and 2 y; admitted with diarrhoea or pneumonia to the paediatric wards of an academic hospital

EXCLUSION CRITERIA: diarrheal episode longer than 72 h on admission; pneumonia complicated by respiratory failure; children on ART or those who had received vitamin or micronutrient supplementation; children with chronic illness unrelated to HIV

Participants randomised:

-

Participants analysed:

Loss to follow-up/ withdrawal:

Exclusions post-randomisation: 0

Interventions 
Outcomes 
Notes 

Ndeezi 2010

Methods

Country: Uganda

Setting:Paediatric HIV clinics of the national referral hospital

Duration of recruitment:June 2005-June 2008

Design: Placebo-controlled trial

Participants

INCLUSION CRITERIA: children aged 1 - 5 years whose mothers had attended the clinic at least once, and who adhered to a regular study follow-up schedule for one year.

EXCLUSION CRITERIA: Children enrolled in other studies, those residing more than 15 kilometres from the clinic and those whose parents or caretakers were anticipating
moving from the study area

Participants randomised:847

- 56% were less than 36 months

Participants analysed:695

Loss to follow-up/ withdrawal:152

Exclusions post-randomisation: 0

Interventions

INTERVENTION: 2xRDA of 14 micronutrients

CONTROL: RDA of six multivitamins

DURATION:6 months

Outcomes

PRIMARY OUTCOMES:Mortality at 12 months

SECONDARY OUTCOMES:Growth (weight-for-height at 12 months); CD4 counts

Notes 

Ancillary