Secondary Logo

Journal Logo

Gastroenteritis and Early Weaning

Breastfeeding, Mother-to-Child HIV Transmission, and Mortality Among Infants Born to HIV-Infected Women on Highly Active Antiretroviral Therapy in Rural Uganda

Homsy, Jaco MD, MPH*; Moore, David MD, MPH; Barasa, Alex MBChB; Were, Willi MBChB; Likicho, Celina RN, RM; Waiswa, Bernard MBChB, MPH; Downing, Robert; Malamba, Samuel PhD*; Tappero, Jordan MD, MPH; Mermin, Jonathan MD, MPH§

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: January 2010 - Volume 53 - Issue 1 - p 28-35
doi: 10.1097/QAI.0b013e3181bdf65a
  • Free
  • Press Release



Highly active antiretroviral therapy (HAART) and formula feeding have practically eliminated mother-to-child transmission (MTCT) of HIV in the United States and Europe.1 However, HIV remains a major cause of pediatric morbidity and mortality in sub-Saharan Africa. UNAIDS estimated that nearly half a million children under 15 acquired HIV and 330,000 died of AIDS in this region in 2007, representing 90% of all new pediatric infections and deaths worldwide.2 A similar picture prevails in Uganda, where an estimated 140,000 children were living with HIV in 2008, 95% of whom were infected through MTCT, and a third of these likely acquired the infection through breastfeeding.3

Current efforts to expand access to HAART in sub-Saharan Africa are accompanied by the expectation that HAART will drastically reduce MTCT of HIV as has occurred in Europe and the United States. However, in Uganda as in the vast majority of sub-Saharan Africa, pregnant women continue to deliver without knowledge of their HIV serostatus and the vast majority breastfeed their babies.2,3 Results of observational studies investigating the impact of maternal HAART on late postnatal MTCT of HIV suggest that maternal HAART can reduce but does not eliminate MTCT-with postnatal transmission rates between 4-6 weeks and 6 months ranging from 0.9% to 2.3%4-8-and may improve the HIV-free survival of infants born to mothers taking HAART for their own health.8

The safety and efficacy of using HAART for pre- and postnatal transmission prophylaxis have recently been investigated in 3 randomized controlled trials.9-11 An important question addressed by these trials is how effective HAART can be when used prophylactically and what impact it can have on the HIV-free survival of breastfeeding infants. The Breastfeeding, Antiretroviral and Nutrition (BAN) study in Malawi compared 6-month postnatal maternal HAART prophylaxis with daily infant nevirapine (NVP) for 6 months and a control arm receiving no antiretroviral (ARV) prophylaxis, all with or without nutritional supplementation. Initial results showed a 4.9% in utero transmission at 1 week and an additional 3% postnatal transmission at 6 months in the maternal HAART arm, with a similar result for the infant NVP arm.9 In the Mma Bana trial in Botswana, HIV-infected pregnant women with CD4 ≥200 cells/mm3 were randomized to receive either triple nucleoside reverse transcriptase inhibitor (NRTI)- or protease inhibitor-containing HAART and were compared with a control group of women with CD4 <200 cells/mm3 receiving the standard NRTI/nonnucleoside reverse transcriptase inhibitor regimen. Results showed an overall MTCT rate of 1%, a 0.9% rate of breastfeeding transmission in the triple NRTI arm and no breastfeeding transmission in the 2 other arms.10 Lastly, the World Health Organization (WHO) Kesho-Bora trial which compared triple ARV prophylaxis from 28-36th gestational week until a maximum of 6 months in pregnant women with 200-500 CD4 cells/mm3 postpartum to antenatal AZT from 28-36th gestational week plus single-dose nevirapine intrapartum and a dual tail therapy (AZT + 3TC) from onset of labor to 1 week postpartum in breastfeeding women reported a 42% reduction in infant HIV infections and a 36% increase in infant HIV-free survival at 12 months in the triple ARV prophylaxis group.11

The objective of this study was to assess the rate of MTCT of HIV and mortality in relation to the type and duration of breastfeeding among infants born to a cohort of HIV-infected women taking HAART for their own health in rural Uganda.



Participants in this study were part of a randomized clinical trial (RCT) initiated in March 2003 evaluating clinical and laboratory-based strategies for monitoring antiretroviral therapy among 1100 HIV-infected 18-49 years old adults receiving HAART including 733 women in Tororo and Busia Districts in rural Uganda. Index clients were recruited from the Tororo branch of The AIDS Support Organization, a national nongovernmental organization providing medical and psychosocial care and support for people living with HIV/AIDS.

RCT Study Enrolment and Follow-Up Procedures

Clients were determined to be HAART eligible if they had CD4 cell counts ≤250 cells/μL or WHO stage III or IV disease. Lamivudine, stavudine, and nevirapine were used as the primary HAART regimen in parallel with daily cotrimoxazole prophylaxis. All index clients were visited at home on a weekly basis by trained lay workers. Every week, these home visitors checked on the participants' clinical status, delivered antiretroviral drugs and cotrimoxazole, and checked on drug adherence. Every month, they monitored participants' weight and women's pregnancy status. And every 3 months, they bled patients for CD4 counts and viral loads.

All RCT index clients, their children and HIV-infected household members were offered free comprehensive HIV care, treatment, and prevention services by the study, including supportive counseling, prevention of mother-to-child transmission (PMTCT) prophylaxis, HAART, and clinic referrals as required. All HIV-infected household members and HIV-exposed children of index clients were provided daily cotrimoxazole as of 6 weeks of age, screened for HAART eligibility, and treated if eligible for HAART. All clients were given insecticide-treated bed nets for malaria prevention, and a safe drinking water vessel with a supply of chlorine to reduce the risk of diarrheal disease.

MTCT Study Enrolment and Follow-Up Procedures

All HIV-infected female index clients who delivered one or more live infants between March 1, 2003 and January 1, 2007 were enrolled in this study. The HIV status of these women's had been established long before their pregnancy. Pregnancies were identified by asking all HIV-infected women index clients about their last normal menstrual period at monthly home visits. Women whose last normal menstrual period occurred ≥30 days before the visit were offered a urine pregnancy test and those with a positive result were provided with PMTCT counseling, were offered ARV prophylaxis at home and were referred to the antenatal care clinic of Tororo District Hospital that was adjacent to the RCT study clinic. The hospital midwives further counseled and educated women about breastfeeding, safe weaning, and replacement feeding and as of 2005, invited all newly delivered HIV-infected pregnant women to partake in a nutrition education class where safe replacement food preparation was demonstrated and discussed. Mothers received additional nutritional and food preparation advice, counseling, and referral from field officers during regular home visits. Neither the hospital nor the study provided any food supplementation to breastfeeding or nonbreastfeeding mothers.

All HIV-infected pregnant women were provided with 2 mg/kg of NVP syrup with instructions to administer it as a single dose to their infant within 72 hours of birth. After September 2005, the NVP syrup dose was supplemented with AZT syrup 4 mg/kg b.i.d. to be administered to the infants for 7 days postpartum (or 28 days if the mother had started HAART <1 month before delivery) as per the revised Uganda MOH PMTCT guidelines.12 All mothers were counseled to exclusively breastfeed for 3-6 months, followed by rapid weaning and safe replacement feeding as per the MOH guidelines at the time. Infants were passively followed-up until to July 31, 2007.


All participants provided written consent to participate in the RCT study in 1 of 6 local languages. The RCT study and the MTCT substudy were approved by the Science and Ethics Committee of the Uganda Virus Research Institute, the Uganda National Council of Science and Technology, and the institutional review boards of the Centers for Disease Control and Prevention (CDC), Atlanta, GA, and of the University of California, San Francisco, CA, USA.

Data Collection

Baseline clinical and demographic data of women participants were collected through standardized questionnaires at the time of enrolment in the RCT study. Infants' dates of birth were determined through home visits by field officers and double checked in hospital records when available. Blood samples for CD4+ T lymphocyte cell counts and HIV polymerase chain reaction (PCR) testing were collected every 3 months and transported to CDC-Uganda laboratories at the Uganda Virus Research Institute in Entebbe. Infant HIV DNA PCR testing was offered to all mothers as their children reached 6 weeks of age. Testing of breastfed infants was scheduled quarterly thereafter until ≥6 weeks after complete cessation of breastfeeding. Children under 18 months of age were tested by both antibody testing and HIV DNA or RNA PCR as part of a separate study.13 Children 18 months old and older were tested by antibody testing only. All infant testing required the mother to bring their infant to the RCT study clinic for bleeding. Babies who did not present for testing were not traced as they were not RCT index cases.

Breastfeeding status, type, and date of breastfeeding cessation were recorded at the time of regular home visits. Midwives-counselors explained to mothers that exclusive breastfeeding meant that nothing but breast milk was given to the baby and that mixed feeding started the moment the baby started receiving any amount of additional liquid or solid food, except for medicines. Mothers of deceased children were asked about the feeding status of their child at the time of death and their weaning (defined as complete breastfeeding cessation) date before that time, if applicable. When a mother had difficulty remembering dates, adult members of her household were asked to help with their recollection. Infant and maternal deaths were recorded by field officers through weekly home visits. Clinical conditions present at the time of infant deaths were abstracted from the RCT clinical records and/or by reports obtained retrospectively from mothers.

Laboratory Testing

CD4+ counts were measured using standard procedures (TRUCount on FACSCalibur, Becton-Dickinson, La Jolla, CA). Qualitative HIV PCR testing (Amplicor HIV-1 DNA PCR Test, version 1.5; Roche, Branchburg) was performed on dried blood spots collected on specimen filter paper. Viral load measurements using quantitative RNA PCR (Cobas Amplicor HIV-1 RNA Monitor Test, version 1.5; Roche) were done on EDTA blood samples. All positive PCR results were confirmed by repeat bleeding and testing. HIV enzyme-linked immunosorbent assay (ELISA) antibody testing was performed in parallel on 2 plasma specimen (Vironostika HIV-1 Plus O Microelisa System, Biomérieux, Durham, NC; Murex rapid microfiltration HIV-1 ELISA assay, Murex Biotech Ltd, Dratford, UK).


We examined the effects of maternal HAART and exclusive and total breastfeeding durations on HIV MTCT and infant survival between the time mothers started HAART and July 31, 2007 including all live births up to January 1, 2007. Infants were considered to be born to a mother receiving HAART if the mother had commenced HAART before the baby's birth. Duration of total breastfeeding referred to the combined duration of exclusive breastfeeding and mixed feeding for all infants, and was equivalent to the age at which infants were completely weaned. Deceased children who were unable to breastfeed because of their sickness were not treated as “weaned” unless their mother stated she had taken clear action to wean them before, and independent of, their illness.

Infant survival in relation to breastfeeding status was analyzed using Kaplan-Meier methods. Child-time at risk was defined by whether a child was still breastfeeding, had ceased breastfeeding, or had never been breastfed at the time of death. Two-sided P values were computed using a log-rank test of the probability of survival during the time at risk (log-rank test of equality of survivor functions).

In bivariate analysis, we assessed the role of potential predictors of infant mortality. The Kruskal-Wallis test was used for analysis of continuous variables and medians, and the chi-square, or Fisher Exact Test, for analysis of categorical variables. Cox proportional hazards modeling was used to identify factors associated with survival. The final model which best fit the data were chosen using the Akaike Information Criterion14 as a measure of the goodness of fit of the model. All analyses were done using SAS version 9.1 (SAS Institute, Cary, NC).


A total of 118 infants were born to 102 women on HAART during the study period, including 4 pairs of twins and 12 siblings (12 mothers had a second pregnancy, 2 of whom produced twins). The median follow-up of infants was 18.1 months [interquartile range (IQR) 9.3-26.0]; 84% (80/95) of the children who were alive by the end of follow-up were followed for 12 months or more (Table 1).

Descriptive Characteristics of Children and Their Mothers by Child Survival Status

Mothers' median time on HAART preceding delivery was 20.3 months (IQR 12.1-28.2); 19 of them started HAART during pregnancy and had been on treatment for a median time of 5.2 months (IQR 2.9-6.3) before delivery. Nine mothers had detectable viral loads ranging from 5150 to 402,000 cps/mL within a median time of 25 days (IQR 9-43) around the time of delivery (Table 1).

Ninety-two percent (109/118) of women breastfed exclusively their infants for a median duration of 4.0 months (IQR 3.0-6.0) and stopped breastfeeding at a median age of 5.0 months (IQR 3.0-7.0). At 6 months of age, 48% (57/118) of babies had been weaned while 25% (29/118) were still being exclusively breastfed and 20% (23/118) were receiving mixed feeding (breast milk and additional foods or liquids) (Table 1).

A total of 114 (97%) infants were tested by HIV DNA or RNA PCR, none of whom were found to be positive by the end of follow-up or death (Tables 1 and 2). Four infants died before ever being tested by HIV PCR at ages ranging between 12 and 64 days; one of them had had a negative HIV ELISA test result 4 days after weaning at the age of 19 days, whereas the other 3 had a positive ELISA test result (Table 2).

Individual Characteristics of Deceased Children (n = 23)

Eighty-six percent (98/114) of the PCR test results were final as they were obtained from infants who either were never breastfed or had ceased breastfeeding for 6 weeks or more (Table 1). Four of the 16 nonfinal PCR results were from live infants who were still breastfeeding at the time of their last PCR testing and 12 were from deceased infants, 9 of whom were still breastfeeding at the time of testing, whereas 3 had been weaned less than 6 weeks before testing (Table 2). Eleven of these 16 babies (69%) with nonfinal HIV PCR results died with severe diarrhea and/or vomiting (Table 2).

Twenty-three (19%) of the 118 infants died during follow-up. Three of the deaths occurred among twins and 4 of the dead children were the offspring of a second pregnancy that their mother experienced while on HAART. Median age at death was 3.7 months (IQR 1.9-8.3; Table 1). No HIV infection was identified among the 19 infants tested by PCR before their death, however, 12 of these results and the ELISA antibody negative result could not be considered final for the reasons mentioned above. The median durations of exclusive and total breastfeeding among the dead infants were both 3.0 months with IQRs of 1.5-3.0 and 1.5-4.4, respectively (Table 1). Three of the 23 deceased babies were never breastfed, 13 babies died after a median time of 6.7 weeks (IQR 3-14.5) after being weaned, whereas 5 died while being exclusively breastfed and 2 died while being mixed fed (Table 2). In the week preceding their death, 15 (65%) of these 23 infants had symptoms of gastroenteritis (diarrhea and/or vomiting), 9 had signs of respiratory illness (dyspnea, pneumonia, cough), and 2 experienced fever and/or convulsions unaccompanied by any of the previous symptoms (Table 2).

When infant mortality was analyzed in relation to breastfeeding status, there was a significantly lower probability of survival among infants who had been weaned or never breastfed (P = 0.0018, log-rank test of survivor function) as compared with those who were still breastfeeding by the end of follow-up (Fig. 1). Using Cox proportional hazards model to analyze factors associated with mortality in 102 first born babies (Table 3), bivariate analysis showed that infant mortality was inversely associated with the duration of either exclusive breastfeeding (EBF) [hazard ratio (HR) = 0.69 per month of EBF, 95% confidence interval (CI) 0.54-0.86, P = 0.002] or total breastfeeding (TBF) (HR = 0.70 per month of TBF, 95% CI, 0.57-0.86, P < 0.001), whereas a borderline association existed with maternal death (HR = 6.19, 95% CI, 0.81-47.2, P = 0.08). There was no significant association between infant mortality and maternal viral loads, CD4 counts closest to delivery or maternal marital status.

Cox Proportional Hazard Analysis of Factors Associated With Mortality in First Born Babies of Women on HAART (n = 102)
Kaplan-Meier estimates of infant survival in relation to breastfeeding status at the time of death.

As the Akaike criterion showed that the model which best fit the data combined the duration of total breastfeeding and maternal death, we factored 2 versions of the final multivariate model: one with continuous and another with discrete durations of total breastfeeding ranging from 4 to 7 months (Table 3). In both versions, a strong and significant inverse association was detected between the duration of total breastfeeding and infant mortality, which was maintained after adjusting for maternal death. In the model using the duration of total breastfeeding as a continuous variable, each additional month of breastfeeding reduced the infant mortality risk by 24% (adjusted HR = 0.71; 95% CI, 0.57-0.87). Using discrete durations of total breastfeeding showed that breastfeeding for less than 6 months accounted for a 6-fold higher mortality risk (adjusted HR = 6.19; 95% CI, 1.41-27.0, P = 0.015) as compared with breastfeeding for more than 6 months. Maternal death was also associated with infant mortality after controlling for the duration of breastfeeding in either version of the model but this association did not reach statistical significance (highest adjusted HR = 3.80; 95% CI, 0.50-29.1, P = 0.20).


In this cohort of 118 HIV-exposed infants born to mothers on HAART in rural Africa, no mother-to-child HIV transmission event was observed among 114 infants who were tested by PCR at least once, 93% of whom were breastfed. However, infant mortality was high (19%), was strongly associated with a shorter duration of total breastfeeding, and was closely preceded by severe diarrhea and vomiting in 65% of the cases. These findings support the possibility that the majority of these deaths may have been caused by severe gastroenteritis following early weaning. This observation is corroborated by the associations between breastfeeding and survival among young children in Africa and HIV-exposed children in particular,15-19 and data from other perinatal cohorts in Kenya, Malawi, Uganda,20-23 and an infant diarrhea outbreak in Botswana.24

Our finding that any breastfeeding shorter than 6 months was associated with a 6-fold increased risk of mortality mirrors the level of protection of breastfeeding against diarrhea reported by a large WHO study on the role of breastfeeding on the prevention of infant mortality.25 This result suggests that 6 months of breastfeeding may represent a critical point before which weaning posed a significant risk to the survival of these infants, thus supporting current WHO infant feeding recommendations for PMTCT.

The finding that no MTCT event occurred in this study is striking, but limited by the fact that the HIV status of 20 children was not final. Thus it is not possible to definitely rule out that these children may have been infected and died of HIV-associated illnesses. However, 9 of the 16 deceased babies with a missing or a nonfinal PCR test results were weaned at a very young age (median 2.0 months, IQR 1.3-4.0), 2 were receiving mixed feeding at the time of death, one was never breastfed, and 69% of them (11/16) died with severe diarrhea and/or vomiting. Therefore, whether or not these babies were infected with HIV, severe gastroenteritis following early weaning remains a plausible explanation for their premature death.

The number of deaths among infants born to mothers on HAART was unexpectedly high, whereas the number of deaths among adults in the RCT cohort was low.26 By comparison, Uganda infant mortality rate was 76 per 1000 live births during the study period.27 In the survival analysis, the risk of death was significantly higher among infants who were weaned early or had never been breastfed. In multivariate analysis, no factor other than the shorter duration of breastfeeding (ie, neither the mothers' CD4 counts closest to delivery nor their marital status or survival) explained this outcome. The effect of maternal morbidity and mortality on their child's survival is well documented28-30 and is reflected in the nearly 4-fold hazard ratio associated with maternal death in the multivariate analysis. The lack of statistical significance of this result is likely due to the small number of maternal deaths in this cohort.

The durations of exclusive (EBF) and total breastfeeding (which includes mixed feeding) were not found to have a different effect on mortality. The most likely explanations for this finding are the short breastfeeding duration and the small sample size of the study, which may have limited the power to detect an effect of EBF versus mixed feeding. Moreover, most women followed the feeding recommendations in force at the time and weaned their child rapidly thereby reducing the duration of mixed feeding, as shown by the small difference between the median durations of exclusive and total breastfeeding. In other larger studies looking at longer breastfeeding durations in women not on HAART, EBF during the first 6 months of life has been shown to be associated with a reduced risk of postnatal HIV transmission as compared with mixed feeding, but the role of EBF in increasing HIV-free survival is less clear,16,30,32-35 probably because the mode of infant feeding is one of several factors that influence the HIV-free survival of infants born to HIV-infected women. Other factors include the types of maternal and infant postnatal MTCT prophylaxis, treatment adherence, and maternal morbidity and mortality.9-11,28-30,35-39

This study was limited by its observational nature, the lack of adjustment for the mothers' socioeconomic status (SES) and the infants' birth weight in the multivariate analysis, and a possible social desirability bias from mothers from whom breastfeeding data were collected retrospectively. Practically all women participants lived in similar low SES conditions, with no running water, using pit latrines and with no or very low income, however poor hygiene or nutrition in the deceased babies' households may have increased the risk of severe gastroenteritis. Low birth weight may also have predicted mortality in this cohort but birth weight data were not accessible as women lived as far as 100 km away from Tororo hospital and delivered in a number of different health centers or dispensaries, or at home. Lastly, social desirability may have skewed the women's recall on duration of exclusive breastfeeding and thus underestimated the duration of mixed feeding.

Our data point to the extreme vulnerability of young infants weaned early in settings where breastfeeding is the only safe and affordable way of providing adequate infant nutrition. The high rates of gastroenteritis and mortality observed occurred despite all families receiving weekly home visits by trained workers and/or counselors as well as a care package including daily cotrimoxazole prophylaxis, and a safe water vessel with chlorine. Given the very rudimentary conditions most of the study participants lived in, it is likely that the rate of adverse outcomes might have been even higher without these preventive measures. For this reason, following the findings of this study, the cohort counselors were sensitized to the high morbidity and mortality risk faced by babies weaned prematurely and were urged to emphasize the new WHO HIV and Infant Feeding recommendations31 when counseling pregnant women or new mothers.

In conclusion, in this Ugandan cohort, maternal HAART provided very high protection against ante- and postnatal HIV transmission but not against infant mortality, and early weaning was the single factor associated with this latter finding. These results emphasize the importance of identifying and treating as promptly as possible HIV-infected pregnant women as well as women of reproductive age eligible for HAART. Furthermore, they suggest that in settings where breastfeeding is the safest and most feasible infant feeding option, the best strategy to prevent MTCT and increase HIV-free survival of exposed infants is to actively and unequivocally promote exclusive breastfeeding for at least 6 months for all HIV-exposed infants. Otherwise, even the most effective PMTCT prophylaxis may fail to prevent or reduce the risk of gastroenteritis and death of HIV-free infants due to premature weaning and/or unsafe replacement feeding.40 Recent data have highlighted the feasibility and potential benefits of prolonged exclusive breastfeeding for HIV-exposed children.15,16,34

Larger studies on infant mortality in relation to mothers' treatment or prophylaxis regimen and breastfeeding duration are warranted to further define the precise breastfeeding duration that maximizes HIV-free survival of infants in resource-limited settings.


We thank Tororo District Hospital, The AIDS Support Organization, and CDC Tororo staff for providing care and support to mothers and infants of the Home-Based AIDS Care project. We are indebted to Dr. Beryl West from CDC-Uganda for helping with laboratory logistics, and to Juliet Namugga and Harriet Turyahebwa from CDC Informatics Unit for helping with data entry and cleaning. Last, we are grateful to Drs. Nathan Schaffer, Tracy Creek, Mary Glenn Fowler, and to Rachel King for their useful comments on the manuscript.


1. European Collaborative Study. Mother-to-child transmission of HIV infection in the era of highly active antiretroviral therapy. Clin Infect Dis. 2005;40:458-465.
2. UNAIDS. 2008 Report on the Global AIDS Epidemic. WHO Library Cataloguing-in-Publication Data: UNAIDS/08.25E/JC1510E. WHO, Geneva, Switzerland.
3. Ministry of Health Uganda and ORC Macro. Uganda HIV/AIDS Sero-Behavioral Survey 2004-2005. Kampala and Calverton: Uganda Ministry of Health and ORC Macro; March 2006.
4. Thomas T, Masaba R, Ndivo R, et al. Prevention of Mother-to-Child Transmission of HIV-1 among breastfeeding mothers using HAART: The Kisumu Breastfeeding Study, Kisumu, Kenya, 2003-2007. Presented at: 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2007; Boston, MA. Session 14, Abstract 45aLB.
5. Kilewo C, Karlsson K, Ngarina M, et al. Prevention of mother to child transmission of HIV-1 through breastfeeding by treating mothers prophylactically with triple antiretroviral therapy in Dar es Salaam, Tanzania-the MITRA PLUS study. Presented at: 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention, Sydney, Australia, July 22-25, 2007. Abstract TuAX101.
6. Towne-Gold B, Ekouevi DK, Viho I, et al. Antiretroviral treatment and prevention of peripartum and postnatal HIV transmission in West Africa: evaluation of a two-tiered approach. PLoS Med. 2007;4:1362-1373.
7. Ekouevi D, Coffie P, Benaud B, et al. Antiretroviral therapy in pregnant women with advanced HIV disease and pregnancy outcomes in Abidjan, Cote d'Ivoire. AIDS. 2008;22:1815-1820.
8. Taha T, Kumwenda J, Kafulafula G, et al. Maternal highly active antiretroviral treatment (HAART): does it improve child survival? Abstract MOPEB083 presented at: 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention; July 19-22, 2009; Cape Town, South Africa.
9. Chasela C, Hudgens M, Jamieson D, et al. Both maternal HAART and daily infant nevirapine (NVP) are effective in reducing HIV-1 transmission during breastfeeding in a randomized trial in Malawi: 28 week results of the Breastfeeding, Antiretroviral and Nutrition (BAN) Study. Abstract WELBC103 presented at: 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention; July 19-22, 2009; Cape Town, South Africa.
10. Shapiro R, Hughes M, Ogwu A, et al. A randomized trial comparing highly active antiretroviral therapy regimens for virologic efficacy and the prevention of mother-to-child HIV transmission among breastfeeding women in Botswana (The Mma Bana Study). Abstract WELBB101 presented at: 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention; July 19-22, 2009; Cape Town, South Africa.
11. Kesho Bora Study Group. Triple-antiretroviral (ARV) prophylaxis during pregnancy and breastfeeding compared to short-ARV prophylaxis to prevent mother-to-child transmission of HIV-1: the Kesho Bora randomized controlled clinical trial in five sites in Burkina Faso, Kenya and South Africa. Abstract LBPEC01 presented at: 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention; July 19-22, 2009; Cape Town, South Africa.
12. Ministry of Health Uganda, AIDS Control Programme. National Guidelines for the Prevention of Mother to Child HIV Transmission. Revised ed. August 2006.
13. Menzies N, Homsy J, Chang Pitter JY, et al. The cost-effectiveness of routine rapid HIV antibody testing before DNA-PCR testing for early infant HIV diagnosis in resource-limited settings. Pediatr Infect Dis J. 2009;28:819-825.
14. Akaike H. A new look at the statistical model identification. IEEE Trans Automatic Control. 1974;19:716-723.
15. Rollins NC, Becquet R, Bland RM, et al. Infant feeding, HIV transmission and mortality at 18 months: the need for appropriate choices by mothers and prioritization within programmes. AIDS. 2008;22:2349-2357.
16. Kuhn L, Aldrovandi GM, Sinkala M, et al. Zambia Exclusive Breastfeeding Study. Effects of early, abrupt weaning on HIV-free survival of children in Zambia. N Engl J Med. 2008;359:130-141.
17. Thior I, Lockman S, Smeaton LM, et al. Mashi Study Team. Breastfeeding plus infant zidovudine prophylaxis for 6 months vs formula feeding plus infant zidovudine for 1 month to reduce mother-to-child HIV transmission in Botswana: a randomized trial: the Mashi Study. JAMA. 2006;296:794-805.
18. Obimbo EM, Mbori-Ngacha DA, Ochieng JO, et al. Predictors of early mortality in a cohort of human immunodeficiency virus type 1-infected African children. Pediatr Infect Dis J. 2004;23:536-543.
19. Kagaayi J, Gray RH, Brahmbhatt H, et al. Survival of infants born to HIV-positive mothers, by feeding modality, in Rakai, Uganda. PLoS One. 09 Dec 2008, doi:10.1371/journal.pone.0003877.
20. Kourtis A, Fitzgerald D, Hyde L, et al., and the BAN Study Team. Diarrhea in uninfected infants of HIV-infected mothers who stop breastfeeding at 6 months: The BAN Study Experience. Poster 772 presented at: 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, CA.
21. Thomas T, Masaba R, van Eijk A, et al. Rates of diarrhea associated with early weaning among infants in Kisumu, Kenya. Poster 774 presented at: 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, CA.
22. Kafulafula G, Hoover DR, Taha TE, et al. Frequency of Gastroenteritis and Gastroenteritis-Associated Mortality with Early Weaning in HIV-uninfected Children born to HIV-Infected Women in Malawi. J Acquir Immune Defic Syndr. 2010;53:6-13.
23. Onyango-Makumbi C, Bagenda D, Mwatha A, et al. Early Weaning of HIV-Exposed Uninfected Infants and Risk of Serious Gastroenteritis: Findings from Two Perinatal HIV Prevention Trials in Kampala, Uganda. J Acquir Immune Defic Syndr. 2010;53:20-27.
24. Creek T, Kim A, Lu L, et al. Hospitalization and Mortality Among Primarily Non-Breastfed Children During a Large Outbreak of Diarrhea and Malnutrition in Botswana, 2006. J Acquir Immune Defic Syndr. 2010;53:14-19.
25. WHO Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality. Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries: a pooled analysis. Lancet. 2000;355:451-455.
26. Mermin J, Were W, Ekwaru JP, et al. Mortality in HIV-infected Ugandan adults receiving antiretroviral treatment and survival of their HIV-uninfected children: a prospective cohort study. Lancet. 2008;371:752-759.
27. Uganda Bureau of Statistics, Ministry of Health and Macro International Inc., USA Uganda Demographic and Health Survey 2006. Uganda Ministry of Health, August 2007, Kampala, Uganda.
28. Brahmbhatt H, Kigozi G, Wabwire-Mangen F, et al. Mortality in HIV-infected and uninfected children of HIV-infected and uninfected mothers in rural Uganda. J Acquir Immune Defic Syndr. 2006;41:504-508.
29. Zaba B, Whitworth J, Marston M, et al. HIV and mortality of mothers and children: evidence from cohort studies in Uganda, Tanzania, and Malawi. Epidemiology. 2005;16:275-280.
30. Kuhn L, Kasonde P, Sinkala M, et al. Does severity of HIV disease in HIV-infected mothers affect mortality and morbidity among their uninfected infants? Clin Infect Dis. 2005;41:1654-1661.
31. WHO, UNICEF, UNAIDS & UNFPA. &UNFPA. HIV and Infant Feeding: New evidence and programmatic experience. Report of a Technical Consultation held on behalf of the Inter-agency Task Team (IATT) on Prevention of HIV Infections in Pregnant Women, Mothers and their Infants HIV and Infant Feeding, Geneva, Switzerland, 25-27 October 2006. WHO Press 2007, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland.
32. Nduati R, Richardson BA, John G, et al. Effect of breastfeeding on mortality among HIV-1 infected women: a randomized trial. Lancet. 2001;357:1651-1655.
33. Iliff PJ, Piwoz EG, Tavengwa NV, et al. ZVITAMBO study group. Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. AIDS. 2005:19: 699-708.
34. Coovadia HM, Rollins NC, Bland RM, et al. Mother-to-child transmission of HIV-1 infection during exclusive breastfeeding in the first 6 months of life: an intervention cohort study. Lancet. 2007;369:1107-1116.
35. Horvath T, Madi BC, Iuppa IM, et al. Interventions for preventing late postnatal mother-to-child transmission of HIV. Cochrane Database Syst Rev. 2009:CD006734.
36. Kumwenda NI, Hoover DR, Mofenson LM, et al. Extended antiretroviral prophylaxis to reduce breast-milk HIV-1 transmission. N Engl J Med. 2008;359:119-129.
37. Bedri A, Gudetta B, Isehak A, et al. Six Week Extended-Dose Nevirapine (SWEN) Study Team. Extended-dose nevirapine to 6 weeks of age for infants to prevent HIV transmission via breastfeeding in Ethiopia, India, and Uganda: an analysis of three randomised controlled trials. Lancet. 2008;372:300-313.
38. Taha TE, Kumwenda NI, Hoover DR, et al. The impact of breastfeeding on the health of HIV-positive mothers and their children in sub-Saharan Africa. Bull World Health Organ. 2006;84:546-554.
39. Becquet R, Bequet L, Ekouevi DK, et al. ANRS 1201/1202 Ditrame Plus Study Group. Two-year morbidity-mortality and alternatives to prolonged breast-feeding among children born to HIV-infected mothers in Côte d'Ivoire. PLoS Med. 2007;4:e17.
40. Mofenson L. Antiretroviral prophylaxis to reduce breast milk transmission of HIV type 1: new data but still questions. JAIDS. 2008;48:237-240.

HIV; highly active antiretroviral therapy; breastfeeding; mother-to-child transmission; infant mortality; Uganda; Africa

© 2010 Lippincott Williams & Wilkins, Inc.