Early infant breastfeeding practices, and predictors of breastfeeding cessation, in HIV-uninfected and HIV-infected mothers on antiretroviral treatment: a prospective cohort study

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Background: Optimal breastfeeding promotes child health and survival globally. Pro-breastfeeding HIV infant feeding policy changes in 2010, and the roll-out of universal maternal anti-retroviral therapy (ART) since 2013, have created an opportunity for re-establishment of breastfeeding among HIV exposed infants in resource-limited settings. Yet data are limited on breastfeeding practices under these policies. This study aimed to evaluate breastfeeding practices and predictors of breastfeeding cessation among women living with HIV (WLHIV) receiving universal ART, and a comparator group of HIV-negative (HIV-) women in Cape Town, South Africa. Methodology: This secondary data analysis used deidentified data from two parallel, prospective cohort studies conducted at the Gugulethu Midwife and Obstetric Unit between 2013 and 2017; the unit is certified as Baby-Friendly. Utilizing the same research staff, measurement tools and procedures, the studies followed pregnant women (HIV-; and WLHIV initiating universal ART) through delivery. HIVexposed infants received nevirapine and co-trimoxazole prophylaxis. At the early neonatal visit (± 7 days), breastfeeding mother-infant pairs were eligible to continue follow-up (visits at 6 weeks; 3, 6, 9, 12 months). At each study visit, trained field workers administered 24-hour recall questionnaires to assess infant feeding and asked about changes in feeding practices since the preceding visit. Exclusive breastfeeding (EBF) was defined as feeding infants no other food or drink other than breastmilk, except for medicines, vitamins and minerals. Breast feeding (BF) was censored at last visit with reported BF. We used Cox proportional hazards regression to assess factors associated with cessation of exclusive and all breastfeeding among HIV-negative children of both groups of women (HIV-exposed uninfected, HEU and HIV-unexposed, HU children), expressed as crude (HR) or adjusted hazard ratios (aHR). Results: Overall, 872 breastfeeding mother-infant pairs (HEU, n=461; HU, n=411) contributed 925 person-years of follow-up. All WLHIV initiated ART in pregnancy (at ART initiation, median HIV viral load 3.97 log10 copies/mL; median CD4 cell count, 354 cells/uL); 352/461 (76%) of WLHIV had viral load < 1 hour; overall, 788/867, 91%) was more common among HU than HEU infants (95% vs 87%, p<0.0001). Among infants who ever received EBF (754/872, 86%), median EBF duration was 1.4 (0.2-3.1) months. HEU vs HU infants were more likely to ever EBF (91% vs 81%, p <0.0001), with longer durations of EBF (median 1.5 vs 1.4 months, p=0.01; HR for EBF cessation, 0.78 [95% CI 0.67-0.9]). The overall median duration of any breastfeeding was 6.0 (IQR 1.5-12.0) months, with earlier cessation among HEU (median 3.9 months) than HU infants (median 9.0 months); HR for BF cessation 1.87 (95% CI 1.56-2.24). Lactation issues (for example cracked nipples or engorgement) were common in the first 6 weeks postpartum (reported by 143 of 872, 16%), with greater risk among WLHIV than HIV- women (20% vs 12%, p=0.003). Lactation problems (vs none reported) were associated with increased relative hazard for EBF cessation among both HEU (HR 1.48, 95% CI 1.13- 1.96) and HU infants (HR 1.63, 95% CI 1.17-2.26); and for BF cessation among HEU infants (HR 2.45, 95% CI 1.85-3.24). Study limitations include reliance on maternal recall, and inability to assess exact date of breastfeeding cessation. Conclusion: Breastfeeding practices remain poor among both WLHIV and HIV- women, despite probreastfeeding HIV policy changes. An alarmingly high proportion of WLHIV reported lactation problems including cracked nipples. These data highlight an urgent need for lactation support among peri-urban women in our setting, especially those living with HIV, both to improve overall breastfeeding practices to maximize child health and to prevent breastmilk-associated transmission of HIV.