The antenatal HIV prevalence in South Africa was estimated at 29.5% in 2011 with 260,280 pregnant women requiring treatment for prevention of mother-to-child transmission (PMTCT) of HIV in 2010.1,2 The South African PMTCT guidelines were amended in 2010 aligning them with WHO recommendations,3,4 and the program is reported to have 98.8% coverage nationally for antenatal HIV testing, 78.3% CD4 cell count testing, 91.8% antiretroviral prophylaxis to mother and/or baby, and 89% in feeding advice.5,6 The high coverage and improved efficacy of the interventions has resulted in a substantial reduction in new pediatric HIV infections at 6 weeks of age from 5.8% in 2009 to 3.5% in 2010 and 2.7% in 20116,7; however, this still translates into thousands of infants infected annually.1,8 Understanding the shortcomings of PMTCT at the operational, community, and individual level will help to direct future policy decisions. By describing the timing of HIV diagnosis and profile of PMTCT uptake in women with newly diagnosed HIV-infected children younger than 2 years of age, we aimed to determine reasons for vertical HIV transmission and identify modifiable factors.
The Finding Infants with HIV Disease: Evaluation of Resistance (FInHDER) study was a prospective observational study conducted between January and December 2011 at 3 hospitals and 2 clinics in Johannesburg, South Africa. Newly diagnosed HIV-infected children (≤2 years) and their caregivers were recruited. Children were antiretroviral therapy (ART) naive or had received less than 2 weeks of ART.
The South African PMTCT guidelines were updated in 2010, and the study sample captured children born before and after this change.9,10 After April 2010 (Fig. 1), pregnant women were offered a single rapid HIV antibody assay at their first antenatal care (ANC) visit. If negative, a second test was offered after 32 weeks gestation. Positive results were confirmed (second rapid test) and CD4 cell count dictated further management. Women with CD4 ≤350 cells per cubic millimeter were initiated on lifelong ART. Women with CD4 >350 cells per cubic millimeter received twice-daily zidovudine (AZT) from 14 weeks gestation, with single-dose nevirapine (sd-NVP), 3-hourly AZT and single-dose emtricitabine/tenofovir during labor. HIV-exposed infants received daily-dose nevirapine (dd-NVP) for 6 weeks. Dd-NVP was continued throughout breastfeeding if mothers were not on ART. HIV-exposed infants had an HIV DNA polymerase chain reaction (PCR) test at their 6-week immunization visit. Before 2010, AZT was offered to women from 28 weeks gestation with sd-NVP during labor. Lifelong ART was limited to those with CD4 ≤200 cells per cubic millimeter. Infants received sd-NVP at birth and either 7 or 28 days of AZT depending on whether the mother had received prophylaxis for ≤4 weeks.9
Fieldworkers administered a structured questionnaire assessing obstetric history, ANC, timing of maternal diagnosis and access to PMTCT, and perceptions of the health care services. Economic factors included employment, residence, and amenities (electricity, water, and sanitation). Maternal psychological distress was measured according to the Kessler-10 system.11 Additional data were collected from medical records. A detailed review of each case identified psychosocial, health care system and biological factors that may have been implicated in transmission. The appropriateness of therapy was judged against the guidelines applicable during the pregnancy (2008 or 2010) using the recalled CD4 cell count from pregnancy or, if unavailable, the lowest CD4 cell count found on record in each case.
A sample size of 300 was calculated to address the expected prevalence of drug resistance (a separate aim of the study reported elsewhere).12 Data were collected on paper forms, entered into Microsoft Access, and analyzed using SAS (Version 9.3; SAS Institute Inc., Cary, NC). Categorical and continuous variables were compared using χ2 or Fisher exact tests, and t tests or Kruskal–Wallis methods, respectively. Breastfeeding duration was assessed by Kaplan–Meier analysis. The study was approved by the Human Research Ethics Committee of the University of the Witwatersrand (M101148) and the Columbia University Institutional Review Board. All caregivers provided written informed consent.
Between January and December 2011, 385 children were screened and 289 (75%) were enrolled (Fig. 2). Of those excluded, 15 refused consent, 37 had started ART >2 weeks prior, and 41 were discharged or died before enrollment. The 5 study sites contributed 1443 positive HIV PCR results within the district in 2011 (PCR reports, National Health Laboratory Service); the study sample consisted 18% of all positive PCR results in the public sector for the city of Johannesburg over the study period.
Timing of maternal HIV diagnosis, that is, (1) before, (2) during, or (3) after the index pregnancy, was determined in 276 mother–child pairs who were categorized accordingly. Thirty-three women (12%) were diagnosed with HIV before, 147 (53%) during, and 96 (35%) after the index pregnancy. Only 1 had been diagnosed with HIV during a previous pregnancy. One hundred eighty (65%) women were diagnosed with HIV before or during pregnancy and had the opportunity to receive PMTCT: 105 women (66%) and 139 (81%) infants received therapy as recommended by national guidelines. In 86 pairs (61%), both the mother and infant received appropriate treatment (P = 0.66 for diagnosis before or during pregnancy). Of the entire cohort, 81 (29%) mother–child pairs received no PMTCT intervention because of a maternal diagnosis only made postnatally.
The file review identified 256 (93%) subjects with at least 1 psychosocial, health care system or biological risk factor for vertical transmission (Table 3). In 162 (59%) cases, maternal and health care system factors coexisted. There were no significant differences based on timing of maternal diagnosis.
There were no differences in socioeconomic status or maternal stress levels between the 3 groups (Table 1). Access to health care was also similar. Most women were in a relationship with the child's father and had been for the preceding 2 years.
Women diagnosed before the index pregnancy were older (P < 0.0001), of higher parity (P = 0.013), and more likely to report the death of a previous child (P = 0.012) than women diagnosed during or after the index pregnancy. They were also more likely to report the HIV-related death of a partner or other relative (35% vs. 20% vs. 17%; P = 0.082). All had disclosed their HIV status to someone (partner or friends and family). In comparison, 17 (12%) women diagnosed during pregnancy and 17 (19%) diagnosed postpartum had not disclosed (P = 0.003).
ANC and PMTCT
Most women attended at least 1 antenatal visit (n = 259; 96%) at which 222 (92%) with unknown status had an HIV test (Table 2). The median gestation at first ANC visit was 24 weeks [interquartile range (IQR), 16–28], a median of 4 visits (IQR, 3–6) were attended, and 244 (96%) attended on ≥2 occasions. Birth outcomes (birthweight, premature birth) were not different between groups. Overall, 183 women (66%) attended ANC after the PMTCT guideline change in 2010.
Women diagnosed before the index pregnancy (n = 33) attended ANC before 20 weeks gestation more commonly (P = 0.028) and most received some form of maternal and/or infant prophylaxis (P < 0.0001 compared with women diagnosed after the index pregnancy). CD4 cell counts were assessed in 25 (89%) women at a median gestation of 21 weeks (IQR, 13–24). Twenty women (61%) qualified for lifelong ART although only 11 received it, and there was virological evidence of treatment failure in 10 of them. Of the 9 women who should have received ART, 3 reported not being offered ART, 5 reported they had not been emotionally ready to start, and 1 delivered at 36 weeks gestation before starting ART. Women in this group reported a marginally higher incidence of perinatal complications, for example, preterm or prolonged labor, premature rupture of membranes, umbilical cord prolapse, abnormal presentation, or bleeding (P = 0.056).
All 147 women diagnosed with HIV during the index pregnancy attended ANC, most (76%) after 20 weeks gestation. Ninety-one percent received some form of PMTCT (NVP, AZT, or ART); however, the applicable guidelines (2008 or 2010) were followed in only 86 women (66%). One hundred thirty-one (89%) had a CD4 cell count during pregnancy of whom 128 (98%) attended ANC at least twice (and should have received treatment according to the result). Twenty-four women (18%) initiated ART during pregnancy (median duration of treatment 11 weeks before delivery; IQR, 6–14). Twenty-one (31%) women who qualified for ART according to CD4 cell count criteria received only AZT. Forty-seven women with CD4 cell counts above the threshold received AZT (appropriate treatment); however, by the time of study entry, this had dropped below the threshold in 12 (26%). Nine women received only sd-NVP. A further 13 women took no PMTCT, the reasons stated were ART not dispensed (n = 6), limited understanding of PMTCT and nonadherence (n = 2), nonacceptance of HIV diagnosis (n = 2), delivered outside South Africa (n = 1), delayed diagnosis because of discrepant results (n = 1), and conflict with husband regarding medication (n = 1).
Women diagnosed with HIV after the index pregnancy (n = 96) were diagnosed at a median of 6 months postpartum (IQR, 3–15 months) and had attended at least 1 (n = 58; 60%), 2 (n = 50; 52%), or 3 (n = 44; 46%) immunization visits before infant HIV testing.
Twenty women diagnosed with HIV after delivery did not get tested antenatally. Nine did not attend ANC, mostly for reasons unknown, 2 reported being unaware of the pregnancy, 1 felt unprepared for motherhood, and 1 feared dismissal from work. Eleven women booked at ANC (median 24 weeks gestation; IQR, 20–28 weeks) but did not test: 5 were not offered an HIV test, 3 did not receive HIV test results, and 3 refused an HIV test.
Seventy-three women (87% of those attending ANC in this group) reported a negative antenatal HIV test. Apart from younger age, there were no other differences when compared with women who tested positive antenatally (Table 1). One woman insisted that she was only diagnosed postpartum, although antenatal maternal prophylaxis was documented as having been dispensed. Fifteen infants received NVP prophylaxis, 10 born to women who learned their status early postpartum. In 4 instances, infant NVP was documented as being issued, although their mothers denied receipt of the ANC HIV test result. One infant started NVP at 1 month of age after maternal diagnosis.
Feeding Choice and Infant Prophylaxis
Timing of maternal HIV diagnosis had a significant impact on feeding choice. Women diagnosed before the index pregnancy were less likely to breastfeed than women diagnosed during or after the index pregnancy (P < 0.0001) even if they had breastfed a previous child. Of those who initiated breastfeeding, 27% (women diagnosed before) continued until 6 months vs. 45% (women diagnosed during) vs. 56% (women diagnosed after pregnancy); P = 0.040. Approximately one third (n = 48, 36%) of women who breastfed reported mixed feeding, which did not differ between the groups.
There were 68 breastfed infants whose mothers were known to be HIV infected before or within 3 days of the infant's birth. Eleven (16%) received no prophylaxis postpartum (although 1 mother was on ART). Of the 57 (84%) infants who received any prophylaxis, 4 incorrectly received AZT and sd-NVP instead of dd-NVP after the 2010 change, 4 received only sd-NVP, and 44 received dd-NVP. Nine of these 44 babies (20%) received dd-NVP throughout breastfeeding, and in an additional 4 pairs, the mother was on ART. Thus, only 13 of 53 breastfed infants (25%) received optimal protection after the 2010 guideline change.
Maternal and Infant Outcome
Infants of mothers diagnosed before pregnancy were tested earlier compared with infants in the other 2 groups (P < 0.0001). More than two thirds of all children were unwell at diagnosis, the proportion being significantly higher (93%, P < 0.0001) where maternal diagnosis occurred postpartum. In 59% of those diagnosed after the index pregnancy, the child's clinical symptoms prompted the maternal HIV test. If postpartum maternal diagnosis is excluded, infants enrolled at the outpatient clinics were less likely to be ill at the time of diagnosis [2 (7%) vs. 185 (76%) at hospital; P < 0.0001] and tested earlier than infants enrolled in the hospital (6 vs. 14 weeks; P < 0.0001). There was no difference between infants of mothers diagnosed before vs. during pregnancy stratified by enrollment site.
Eight women (3%) died. The timing of HIV diagnosis could be established in 4 cases: 2 were diagnosed before pregnancy (CD4, 18 cells/mm3 in one and unknown in the other), 1 during pregnancy (CD4, 144 cells/mm3), and 1 after pregnancy (CD4, 186 cells/mm3). Two had started ART. The causes of death were unknown.
Although great improvements have been made in reducing mother-to-child HIV transmission in South Africa, the high HIV prevalence among pregnant women (29.5% in 2011)1 continues to result in perinatal infection. A combination of individual, biomedical, and health system factors contribute to PMTCT failures.
Reported ANC attendance was high, but most women presented for their first visit after 14–16 weeks gestation, resulting in over 25% not attending the 4 ANC visits recommended by the WHO.13 Because early gestation at first presentation to ANC is the strongest predictor of ART at delivery, and time on ART is inversely related to risk of vertical transmission,14–16 this results in suboptimal PMTCT. Depending on viral load, 12–16 weeks of maternal ART may be required to achieve viral suppression at delivery, longer than the median time on ART in our cohort.15,17,18 Late presentation for ANC in sub-Saharan Africa has been previously reported.14,19–21 Qualitative studies indicate that women (irrespective of HIV status) see little direct benefit of ANC unless they had experienced problems with the index or a previous pregnancy, often attending only to obtain the required documentation for delivery in a health care facility.19,20 Perceived poor quality of services, negative experiences with clinic staff, financial constraints, and distance from the clinic have been reported as contributing factors.19,21 Although not significant in our cohort (Table 1), these issues were reported by some individuals (Table 3) and further investigation is required.
Rates of antenatal HIV testing were high, consistent with a recent Medical Research Council PMTCT report that demonstrated an antenatal HIV testing rate of 98.8% nationally.6,22 According to the report, although 91.8% of HIV-infected mothers and their infants received PMTCT prophylaxis, only 52% received the regimens recommended by the national guidelines (compared with 61% in our study). The reasons for this are difficult to elucidate: individual maternal factors (denial, lack of “readiness”) and “stock-outs” of drugs (isolated individual reports) may account for a portion of these failures in our cohort. Starting prophylaxis late, not starting ART according to guidelines (ie, failure to initiate combination ART for women with low CD4 cell counts), and poor adherence to therapy were common factors. Simplifying the program and reducing the number of steps could reduce errors as well as ensure optimal PMTCT.23 South Africa updated the PMTCT guidelines again in March 2013 offering combination ART to all HIV-positive pregnant women irrespective of the CD4 cell count.24 Our data indicate that personal circumstance affected participation in PMTCT and adherence to therapy. Further study is required to address how and why personal contexts dictate health-seeking behavior and treatment adherence.
Rates of appropriate infant prophylaxis were high (81%), but only a quarter of breastfed infants received adequate protection for the duration of breastfeeding. Breastfeeding has been shown to confer a risk of approximately 1% per month in the absence of maternal ART or infant dd-NVP.25 This risk is higher among women with incident infections, a difficult group to reach being unaware of their HIV status. Mixed feeding is an additional risk for postnatal HIV transmission.26 In our study, there was no association between awareness of HIV status and exclusive feeding practices and mixed feeding rates were higher than those reported in a recently published survey (36% vs. 18%).27 Women diagnosed before pregnancy were less likely to breastfeed, and if they did, it was for a shorter duration. At the time, replacement feeding was supported by free infant formula, a service phased out in favor of exclusive breastfeeding with antiretroviral prophylaxis.28
Women aware of their HIV status before the index pregnancy were more likely to receive PMTCT and to test their infants early. This group was older, of higher parity, and had booked earlier at ANC thereby receiving treatment (AZT or ART) for longer. Positive associations between adherence to PMTCT prophylaxis and older maternal age have been previously reported.29,30 The death of a previous child may have influenced their health-seeking behavior, although it is unknown whether the deaths were HIV related. Infants of the mothers in this group tested earlier (at 7 weeks old, close to the recommended 6 weeks); yet, many (58%) were unwell at the time of diagnosis. This resonates with reports of high early mortality in perinatally infected infants (20% by 90 days).31 Despite early testing, morbidity in this group is high. In utero infection may be proportionately more common in the setting of good PMTCT coverage during late pregnancy and peripartum.32
Our data show that enrollment at the clinic rather than the hospital was associated with both earlier diagnosis and lower morbidity independent of the timing of maternal diagnosis, emphasizing the importance of early infant diagnosis at the clinics. It is unclear why infant diagnosis occurred later when women were diagnosed with HIV during pregnancy (median, 11 vs. 7 weeks if the maternal test was preconception). The reasons are probably multifactorial. One can speculate that women diagnosed before pregnancy were more willing to identify themselves as HIV infected at infant clinics. Perhaps there was more acceptance of the diagnosis, disclosure of HIV status to family and friends may have played a role. PMTCT information, including need for PCR testing, is documented in the patient-held Road-to-Health booklet, although it is often incomplete and anecdotal reports suggest that this is due to concern about inadvertent disclosure.
Most of the women diagnosed postpartum reported testing HIV negative during pregnancy. We relied on patient reporting of results (rapid tests are not recorded electronically), and feelings of guilt or denial surrounding HIV transmission may have influenced the responses. This could explain the discrepancies observed in women who reported testing HIV negative or not receiving an HIV result but for whom PMTCT-related medication is documented. Surveillance data in South Africa consistently demonstrates evidence of HIV exposure (as measured by the presence of serum antibodies to HIV) ranging from 4.1% to 6.9% in infants whose mothers reported a negative HIV test during pregnancy.6,33,34 Although blood-based rapid HIV tests have a high sensitivity, compliance with WHO standards and local quality assurance protocols cannot be guaranteed at each clinic and has resulted in field sensitivities as low as 87%–95%.6,35,36 Despite these uncertainties, it is likely that many of the reported negative results represent newly acquired infection during late pregnancy and breastfeeding. In Botswana, retesting negative women at delivery and at immunization visits identified an additional 1.3% and 2.4% of HIV-infected women, respectively.37,38 In Johannesburg, testing at delivery found the incidence of HIV infection after a reported negative result to be 4.2%.39 Maternal acquisition of HIV in late pregnancy confers a high risk of vertical transmission because of the high viral loads during primary infection,16,32,37,40–43 and postnatal acquisition of HIV was felt to account for up to 20% of breastfeeding-associated transmission in Zimbabwe.40 In the absence of a maternal diagnosis, infants cannot be identified as HIV exposed and will not be captured by the HIV PCR testing program. Repeat maternal testing is feasible, cost-effective, and acceptable to women both at the time of delivery and at immunization clinics.44,45 In South Africa, most women deliver in health care facilities and 95% attend the 6-week immunization visit.6,7,37–39,46,47 Repeat testing at these times has since been incorporated into updated PMTCT guidelines (April 2013).48 Postnatal maternal diagnosis had significant clinical implications and was often precipitated by the child's hospitalization. Delaying ART in HIV-infected infants results in significant morbidity and mortality.49–56
Limitations of our study include that not all infants screened were enrolled, some having already commenced ART or died. Our sample is not representative of every child diagnosed in the district during the study period. A control sample of HIV-exposed uninfected infants was not included for comparison. We relied on subject recall to establish the timing of maternal diagnosis, and refusal of antenatal HIV testing, which is associated with high HIV prevalence,57 may have been underreported. Lack of biological data meant that true incident infections after the first negative HIV test could not be confirmed. We did not have data on repeat antenatal HIV tests. Additionally, some subjects straddled the period of guideline change, which affected our study in 2 ways: the 2010 guidelines offered a more effective protocol and confusion during the changeover may have resulted in missed opportunities.
We highlight important gaps in the operational implementation of PMTCT guidelines. Early identification of HIV-infected pregnant women (preferably preconception) remains central as late maternal diagnosis denies PMTCT to a significant number of infants. Once identified, all components of the PMTCT cascade need to be implemented. An attempt should be made to sensitize staff to the interplay of multiple sociodemographic and biological factors surrounding HIV infection and pregnancy. Retesting HIV-negative women during late pregnancy, at delivery, and postnatally will help to identify infants at high risk. Mothers should be encouraged to attend early infant diagnosis at 6 weeks or at any time before that if their infant is unwell. Public education about of the benefits of obstetric care and early ANC is vital, whereas maternal feeding choices and exclusive feeding practices require support.
All participants are gratefully acknowledged. Many cases included stories of trauma yet also resilience. The study would not have been possible without the dedicated fieldworkers Phindile Moyo, Phumzile Nyamane, Jabulile Dlamini, Thuli Khumalo, Peace Khanyile, and support staff Mercy Mia. The National Health Laboratory Service provided useful regular updates on PCR testing rates. The expanding PMTCT services are a result of tireless work of the South African Department of Health and its partners toward improved outcomes.
1. National Department of Health. The National Antenatal Sentinel HIV and Syphilis Prevalence Survey. Pretoria, South Africa: National Department of Health; 2011.
2. UNAIDS. Global Report: UNAIDS Report on the Global AIDS Epidemic 2010. Geneva, Switzerland: UNAIDS; 2010.
3. World Health Organization. PMTCT Strategic Vision 2010-2015: Preventing Mother-to-Child Transmission of HIV to Reach the UNGASS and Millennium Development Goals: Moving Towards the Elimination of Paediatric HIV, December 2009. Geneva, Switzerland: World Health Organization; 2010.
4. World Health Organization. Rapid Advice: Use of Antiretroviral Drugs for Treating Pregnant Women and Preventing HIV Infection in Infants, Version 2. Revised 2010. ed. Geneva, Switzerland: World Health Organization; 2009.
5. Carlson C, Bannerman M, Barron P, et al.; for the Mid Year Review Team. National Strategic Plan 2007-2011, Mid Term Review 2010. Pretoria, South Africa: SANAC; 2009.
6. Goga AE, Dinh TH, Jackson DJ; for the SAPMTCTE study group. Evaluation of the Effectiveness of the National Prevention of Mother-to-Child Transmission (PMTCT) Programme Measured at Six Weeks Postpartum in South Africa, 2010: Medical Research Council; 2012.
7. Goga AE, Dinh TH, Jackson DJ; for the SAPMTCTE study group. Abstract WEPE173: Impact of the national prevention of mother-to-child transmission of HIV (PMTCT) program on perinatal mother-to-child transmission of HIV (MTCT) measured at six weeks postpartum, South Africa (SA). In: XIX International AIDS Conference. Washington, DC; 2012.
8. Statistics South Africa. Mid-Year Population Estimates 2011. Pretoria, South Africa: Statistics South Africa; 2011.
9. National Department of Health. Policy and Guidelines for the Implementation of the PMTCT Program. 2008 ed. Pretoria, South Africa: National Department of Health; 2008.
10. National Department of Health. Clinical Guidelines: PMTCT (Prevention of Mother-to-Child Transmission). 2010 ed. Pretoria, South Africa: National Department of Health; 2010.
11. Kessler RC, Andrews G, Colpe LJ, et al.. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32:959–976.
12. Kuhn L, Hunt G, Technau K, et al.. Abstract 951: Pre-treatment drug resistance mutations among HIV+ children <2 years of age who failed or missed PMTCT: Johannesburg, South Africa. In: Conference on Retroviruses and Opportunistic Infections. Atlanta, GA; 2013.
13. National Department of Health. Guidelines for Maternity Care in South Africa. Pretoria, South Africa: National Department of Health; 2007.
14. Stinson K, Boulle A, Coetzee D, et al.. Initiation of highly active antiretroviral therapy among pregnant women in Cape Town, South Africa. Trop Med Int Health. 2010;15:825–832.
15. Myer L. Initiating antiretroviral therapy in pregnancy: the importance of timing. J Acquir Immune Defic Syndr. 2011;58:125–126.
16. Johnson LF, Stinson K, Newell ML, et al.. The contribution of maternal HIV seroconversion during late pregnancy and breastfeeding to mother-to-child transmission of HIV. J Acquir Immune Defic Syndr. 2012;59:417–425.
17. Chibwesha CJ, Giganti MJ, Putta N, et al.. Optimal time on HAART for prevention of mother-to-child transmission of HIV. J Acquir Immune Defic Syndr. 2011;58:224–228.
18. Read PJ, Mandalia S, Khan P, et al.. When should HAART be initiated in pregnancy to achieve an undetectable HIV viral load by delivery? AIDS. 2012;26:1095–1103.
19. Myer L, Harrison A. Why do women seek antenatal care late? Perspectives from rural South Africa. J Midwifery Womens Health. 2003;48:268–272.
20. Simkhada B, Teijlingen ER, Porter M, et al.. Factors affecting the utilization of antenatal care in developing countries: systematic review of the literature. J Adv Nurs. 2008;61:244–260.
21. Mathole T, Lindmark G, Majoko F, et al.. A qualitative study of women's perspectives of antenatal care in a rural area of Zimbabwe. Midwifery. 2004;20:122–132.
22. Doherty T, Chopra M, Nsibande D, et al.. Improving the coverage of the PMTCT programme through a participatory quality improvement intervention in South Africa. BMC Public Health. 2009;9:406.
23. Kim MH, Ahmed S, Preidis GA, et al.. Low rates of mother-to-child HIV transmission in a routine programmatic setting in Lilongwe, Malawi. PloS One. 2013;8:e64979.
24. National Department of Health. The South African Antiretroviral Treatment Guidelines. 2013 ed. Pretoria, South Africa: National Department of Health, South Africa; 2013;12–13.
25. Bulterys M, Ellington S, Kourtis AP. HIV-1 and breastfeeding: biology of transmission and advances in prevention. Clin Perinatol. 2010;37:807–824. ix–x.
26. Iliff PJ, Piwoz EG, Tavengwa NV, et al.. Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. AIDS. 2005;19:699–708.
27. Goga AE, Doherty T, Jackson DJ, et al.. Infant feeding practices at routine PMTCT sites, South Africa: results of a prospective observational study amongst HIV exposed and unexposed infants—birth to 9 months. Int Breastfeed J. 2012;7:4.
28. The Tshwane declaration of support for breastfeeding in South Africa. S Afr J Clin Nutr. 2011;24:214.
29. Hassan AS, Sakwa EM, Nabwera HM, et al.. Dynamics and constraints of early infant diagnosis of HIV infection in Rural Kenya. AIDS Behav. 2012;16:5–12.
30. Peltzer K, Mlambo G. Factors determining HIV viral testing of infants in the context of mother-to-child transmission. Acta Paediatr. 2010;99:590–596.
31. Marston M, Becquet R, Zaba B, et al.. Net survival of perinatally and postnatally HIV-infected children: a pooled analysis of individual data from sub-Saharan Africa. Int J Epidemiol. 2011;40:385–396.
32. Lilian RR, Kalk E, Bhowan K, et al.. Early diagnosis of in utero and intrapartum HIV infection in infants prior to 6 weeks of age. J Clin Microbiol. 2012;50:2373–2377.
33. Rollins N, Little K, Mzolo S, et al.. Surveillance of mother-to-child transmission prevention programmes at immunization clinics: the case for universal screening. AIDS. 2007;21:1341–1347.
34. Moodley D, Esterhuizen TM, Pather T, et al.. High HIV incidence during pregnancy: compelling reason for repeat HIV testing. AIDS. 2009;23:1255–1259.
35. Black V, von Mollendorf CE, Moyes JA, et al.. Poor sensitivity of field rapid HIV testing: implications for mother-to-child transmission programme. BJOG. 2009;116:1805–1808.
36. World Health Organization. Overview of the Prequalification of diagnostics assessment process. Geneva, Switzerland: World Health Organization; 2011.
37. Lu L, Motswere-Chirwa C, Legwaila K, et al.. Abstract 15: HIV incidence in women during the first postpartum year: implications for PMTCT programs Francistown, Botswana, 2010. In: 3rd International Workshop on HIV Pediatrics. Rome, Italy: Reviews in Antiviral Therapy and Infectious Diseases 2011_8; 2011:17.
38. Lu L, Legwaila K, Motswere C, et al.. Abstract 91: HIV incidence in pregnancy and the first post-partum year and implications for PMTCT programs, Francistown, Botswana, 2008. In: 16th Conference on Retroviruses and Opportunistic Infections. Montreal, Canada; 2009.
39. Technau K. Can a Routine Peri-partum HIV Counselling and Testing Service for Women Improve Access to HIV Prevention, Early Testing and Treatment of Children? Johannesburg, South Africa: University of the Witwatersrand; 2009.
40. Humphrey JH, Marinda E, Mutasa K, et al.. Mother to child transmission of HIV among Zimbabwean women who seroconverted postnatally: prospective cohort study. BMJ. 2010;341:c6580.
41. Magder LS, Mofenson L, Paul ME, et al.. Risk factors for in utero and intrapartum transmission of HIV. J Acquir Immune Defic Syndr. 2005;38:87–95.
42. Liang K, Gui X, Zhang YZ, et al.. A case series of 104 women infected with HIV-1 via blood transfusion postnatally: high rate of HIV-1 transmission to infants through breast-feeding. J Infect Dis. 2009;200:682–686.
43. Lilian RR, Kalk E, Technau KG, et al.. Birth diagnosis of HIV infection on infants to reduce infant mortality and monitor for elimination of mother-to-child transmission. Pediatr Infect Dis J. 2013;32:1080–1085.
44. Kim LH, Cohan DL, Sparks TN, et al.. The cost-effectiveness of repeat HIV testing during pregnancy in a resource-limited setting. J Acquir Immune Defic Syndr. 2013;63:195–200.
45. Chi BH, Bolton-Moore C, Holmes CB. Prevention of mother-to-child HIV transmission within the continuum of maternal, newborn, and child health services. Curr Opin HIV AIDS. 2013;8:497–502.
46. Black S, Zulliger R, Marcus R, et al.. Abstract P38: Acceptability and challenges of rapid antiretroviral initiation during pregnancy in Cape Town, South Africa. In: 1st Southern African HIV Clinicians' Society Conference; 2012 25–28 November 2012; Cape Town, South Afrrica; 2012.
47. Day C, Barron P, Massyn N, et al.. District Health Barometer 2010/11. Durban, South Africa: Health Systems Trust; 2011.
48. National Department of Health. The South African Antiretroviral Treatment Guidelines 2013. 2013 ed. Pretoria, South Africa: National Department of Health; 2013.
49. Violari A, Cotton MF, Gibb DM, et al.. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med. 2008;359:2233–2244.
50. Bolton-Moore C, Mubiana-Mbewe M, Cantrell RA, et al.. Clinical outcomes and CD4 cell response in children receiving antiretroviral therapy at primary health care facilities in Zambia. JAMA. 2007;298:1888–1899.
51. Newell ML, Coovadia H, Cortina-Borja M, et al.. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet. 2004;364:1236–1243.
52. Bourne DE, Thompson M, Brody LL, et al.. Emergence of a peak in early infant mortality due to HIV/AIDS in South Africa. AIDS. 2009;23:101–106.
53. Dramowski A, Coovadia A, Meyers T, et al.. Identifying missed opportunities for early intervention among HIV-infected paediatric admissions at Chris Hani Baragwanath Hospital, Soweto, South Africa. S Afr J HIV Med. 2011;12:16.
54. Leyenaar JK, Novosad PM, Ferrer KT, et al.. Early clinical outcomes in children enrolled in human immunodeficiency virus infection care and treatment in lesotho. Pediatr Infect Dis J. 2010;29:340–345.
55. 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.
56. Wamalwa D, Benki-Nugent S, Langat A, et al.. Survival benefit of early infant antiretroviral therapy is compromised when diagnosis is delayed. Pediatr Infect Dis J. 2012;31:729–731.
57. Mseleku M, Smith TH, Guidozzi F. HIV seropositive in pregnant South African women who initially refuse routine antenatal HIV screening. BJOG. 2005;112:370–371.