Paediatric HIV infection is a global public health crisis. An estimated 530 000 new infections occurred in children in 2006, more than 90% in sub-Saharan Africa . The natural history of paediatric HIV infection in infants has been well documented in North American and European cohorts, showing a characteristic bimodal pattern of disease progression [2,3]. Prior to the era of antiretroviral therapy (ART), 20–30% of children developed AIDS or died before 2 years of age; the remainder progressed more slowly to AIDS at a mean of 5–6 years [2,3]. Studies from Kenya, Malawi, Rwanda, Côte d'Ivoire and South Africa show much faster disease progression, with 26–45% mortality by 1 year of age, 35–59% by 2 years of age and up to 89% by 3 years of age [4–8].
Programmes being implemented in resource-limited countries to reduce mother-to-child transmission (MTCT) include those based on the HIV Network for Prevention Trials (HIVNET) 012 protocol. Single-dose nevirapine (sd-NVP) administered to mothers at onset of labour and to infants within 72 h of birth reduces MTCT by approximately 50% in a breastfeeding population . Although new World Health Organization (WHO) guidelines for prevention of MTCT  advocate use of a three-part regimen of antenatal zidovudine (from 28 weeks of gestation), intrapartum zidovudine, lamivudine and single-dose NVP, and postnatal ART to mother (zidovudine and lamivudine for 7 days) and child (sd-NVP, plus zidovudine for 7 days), it is recognized that sd-NVP alone will continue to be used as a minimum in settings where capacity is limited. The natural history of paediatric HIV may be altered by this intervention, since sd-NVP only reduces intrapartum (IP) MTCT, not affecting intrauterine (IU) MTCT . IU-infected infants themselves tend to progress more rapidly to disease [11–13]. Additionally, reduction of MTCT by ART may restrict transmission so that it only occurs from mothers with particularly high viral loads, with correspondingly rapid disease progression in their infants [14–16]. Where combination therapy is used, it may be anticipated that, while MTCT will be reduced further, the transmissions that do occur may increasingly be in the setting of very advanced maternal disease.
The changing natural history of paediatric HIV infection in sub-Saharan Africa is of relevance in defining the optimal parameters for the use of ART, which remain unclear, especially in infants. No studies to date have described the virological and immunological progression of infants infected despite sd-NVP, or addressed whether ART is effective in this setting. With new WHO guidelines recommending treatment thresholds for ART initiation in resource-limited settings , it is important to assess whether such an approach is likely to be successful.
This study describes the natural history in the first year of life of a cohort of HIV-infected infants in KwaZulu-Natal Province, South Africa. At the end of 2005, there were an estimated 5.5 million people living with HIV in South Africa . HIV prevalence in antenatal attendees in KwaZulu-Natal, the worst affected province, increased from < 1% in 1990 to 39.1% in 2005 . In 2001, the HIVNET 012 regimen was implemented to reduce MTCT.
The infants in this report form part of a study designed to assess the feasibility of ART delivery in early infancy and to compare immediate versus delayed ART in perinatally infected infants. Infected infants were randomized blindly at enrolment 2:1 to immediate versus delayed ART, and the results of the study will be reported upon completion. This current report focuses first on MTCT in the whole study cohort and, second, on the natural history of disease progression in the 20 infants randomized to delayed treatment. This delayed treatment group provided an opportunity to characterize in detail the immunological decline of infants infected following perinatal PMTCT interventions in a resource-limited setting.
The study was approved by the Biomedical Research Ethics Committee of the University of KwaZulu-Natal, Durban and the Institutional Review Board of the Massachusetts General Hospital, Boston, Massachusetts, USA.
Study subjects and diagnosis of infant HIV infection
Antenatal mothers were recruited from October 2002, and paediatric study subjects were enrolled between July 2003 and September 2005, at St Mary's Hospital, Mariannhill, Durban, and Prince Mshiyeni Hospital, Umlazi, Durban. Women attending antenatal clinics were provided with voluntary counselling and testing for HIV-1 infection after 28 weeks of gestation. HIV-1-infected mothers giving written consent to participate provided blood samples for viral load and CD4 cell count measurement at median 35 weeks of gestation [interquartile range (IQR), 32–36]. All other aspects of perinatal care were undertaken according to local guidelines. At this time, sd-NVP was the only intervention available for prevention of MTCT. Mothers were given sd-NVP (200 mg) to self-administer at the onset of labour and infants were given sd-NVP (2 mg/kg) by a nurse within 72 h of birth. After counselling, mothers were encouraged to make their own choice about method of feeding. HIV infection was not an indication for Caesarean section, but rates were high, consistent with previous reports from Durban . Peripheral venous blood was obtained from infants on days 1 and 28 of life to identify IU or IP infection, respectively , and repeated as soon as feasible (3–8 days later) to confirm the diagnosis. Among mothers who chose to breastfeed, infants identified as IP infected may, in some cases, have been infected early postpartum through breast milk transmission. Virological and immunological data on all mothers recruited to the study are presented in this report.
Infants were eligible if they were born at one of the two study hospitals, had confirmed IU/IP HIV infection and care-givers agreed to administer ART. Infants were excluded if they were born at < 37 weeks of gestation, were < 2 kg birth weight or had evidence of congenital abnormalities. Forty-three infants were randomized at diagnosis to immediate ART and 20 infants to deferred ART; this was done by drawing an envelope provided by the trial coordinators in Boston, USA. The 20 infants randomized to delayed ART form the basis of this report. An additional 12 infants were identified as infected with HIV but were ineligible because of prematurity (eight), intrauterine growth restriction (one), hydrocephalus (one), failure to return for infant HIV test result (one) and relocation to a different area (one). A cohort of HIV-uninfected infants (negative by polymerase chain reaction at 1 year) born to HIV-infected mothers at Cato Manor Clinic, Durban were followed up at 6 weeks and at 3, 6, 9 and 12 months of age as part of a separate study. Although no clinical data were available from these infants, they provided comparative CD4 cell count data for HIV-uninfected African infants.
Infant follow-up and treatment
Infants were followed up in a dedicated outpatient clinic and had viral load and CD4 cell count measurement at each visit. Outpatient visits were monthly for the first year, and every 2 months thereafter. All infants received co-trimoxazole prophylaxis from 6 weeks to 1 year of age and had free outpatient and inpatient treatment of illness. Criteria for commencing ART in infants randomized to deferred treatment were a confirmed CD4 cell% ≤ 20% unexplained by a treatable or self-limiting intercurrent illness, or progression to paediatric stage III disease or advanced paediatric stage II disease,as defined by the WHO . If the CD4 cell% fell to ≤ 20%, the result was confirmed on a second sample taken 4 weeks later, before starting therapy. The ART regimen used was initially zidovudine, lamivudine, nelfinavir and nevirapine. Nevirapine was discontinued once the viral load became undetectable (< 50 copies/ml). A four drug, three class combination was chosen because of the uncertain implications of sd-NVP-exposure peripartum. Nelfinavir was chosen at the time the protocol was written as the protease inhibitor of choice because of availability, safety and proven efficacy in infected neonates .
HIV-1 RNA and CD4 cell% determination
Diagnosis of HIV-1 infection utilized reverse-transcriptase polymerase chain reaction on 500 μl plasma. Viral load measurement was undertaken using the Roche Amplicor assay version 1.5 (Roche Molecular Systems, Branchburg, New Jersey, USA). CD4 cell% was determined by flow cytometry.
Statistical tests used are indicated in the text.
Mother-to-child transmission rates
All 740 babies born to 719 HIV-infected mothers were included in the MTCT analysis. There were 75 transmissions, which gave an overall MTCT rate of 10.3% (69% IU, 31% IP; Fig. 1). The calculated IU MTCT rate was 7.1% and the IP MTCT rate 3.2%.
There was no significant difference in maternal age, intake of sd-NVP, gestational age or birth weight in the uninfected, IU and IP groups (Table 1). Vaginal delivery occurred in 71% of women in the IP group, 58% in the IU group and 59% in the uninfected group (differences not significant), consistent with vaginal delivery being a risk factor for IP transmission . There was no significant difference in mode of feeding between IU-infected, IP-infected and uninfected infants in the first month of life (Table 1).
Effect of maternal viral load and CD4 cell count on transmission
The median viral load in mothers of HIV-infected infants was higher than in mothers of HIV-uninfected infants (99 650 versus 26 750 copies/ml; P < 0.001, Mann–Whitney). Median viral load was higher in mothers of IP-infected than IU-infected infants (279 000 versus 86 600 copies/ml; P = 0.039; Fig. 2a). However, viral loads in mothers of both groups of infected infants were significantly higher than in mothers of uninfected infants (IU versus no transmission, P = 0.002; IP versus no transmission, P < 0.001; Fig. 2a). Median CD4 cell counts in mothers of IP-infected infants were significantly lower than in mothers of uninfected infants (200 versus 394 cells/μl; P < 0.001, Mann–Whitney; Fig. 2b), and those in mothers of IU-infected infants were lower than those in mothers of uninfected infants (327 versus 394 cells/μl; P = 0.05, Mann–Whitney; Fig. 2b).
Changes of viraemia in the first year of life
Monthly viral load data were analysed from the 20 infants randomized to deferred ART and from the 43 infants randomized to immediate ART before they started therapy. The median viral load of IU-infected infants on day 0–1 was 155 000 copies/ml and this fell to 6510 copies/ml on confirmatory HIV testing at day 5 (IQR, 3–6; Fig. 3a). In all cases where the timing of infant NVP administration was recorded (59/61 IU-infected infants), this was a median 5 days (IQR, 3–7) prior to the confirmatory blood test and was, therefore, likely to have caused the fall in viral load in the first week of life. In 13/34 (38%) infants, this day 5 viral load was beneath the limit of detection (3.6 log10 or 4000 copies/ml) that would have been available using analysis of dried blood spots (50 μl whole blood)  and in one infant was undetectable (< 50 copies/ml).
The initial viral load in IP-infected infants (day 28) was higher than in IU-infected infants (median 585 000 versus 155 000 copies/ml). Peak viraemia, defined as highest viral load in the first 6 months of life, was significantly higher in IP-infected infants than in IU-infected infants (5 160 000 versus 984 000 copies/ml; P < 0.001, Mann–Whitney; Fig. 3b). The median viral load of untreated infants declined marginally from 1 724 000 copies/ml between 1 and 6 months of age to 1 320 000 copies/ml by 6–12 months of age (Fig. 3c). However, artefact contributes to this apparent decline since infants with high viral loads are lost as they start ART because of immunological failure.
Changes in CD4 cell count in the first year of life
Monthly CD4 cell data from 20 HIV-infected infants randomized to deferred treatment were analysed to characterize immunological decline in this cohort. HIV-infected infants showed rapid progression to immunosuppression compared with uninfected infants (Fig. 4a,b). Median CD4 cell% at birth was 47% and seven infants (35%) had progressed to a CD4% of ≤ 20% by 3 months; this increased to 14 (70%) by 6 months and 16 (80%) by 1 year (Fig. 4a,c). Time to CD4% ≤ 20% was directly related to maternal CD4 cell count (Spearman r = 0.51; P = 0.02; Fig. 4d). There was no correlation between viral load at any time in infancy and immunological decline, although the number of infants in the deferred treatment group was small.
Morbidity and mortality in the first year of life
Infants whose CD4 cell% fell to ≤ 20% were started on ART, and thus the natural history of clinical disease progression has been modified in this cohort. However, in the first 6 months, before ART was started, five infants (25%) required hospital admission (four for treatment of infections and one for severe thrombocytopenia) and all 20 children required outpatient treatment of infections on a median 3.5 occasions (range, 1–10; Fig. 5a). Only one child had weight persistently below the 3rd centile (Fig. 5b). One child (5%) died of acute gastroenteritis at 6 months of age. Although treatment criteria had been reached, ART had not yet been started at the time of death.
This study describes rapid disease progression in a cohort of infants infected with HIV following perinatal PMTCT interventions and has implications for ART roll-out programmes in resource-limited settings, given the large number of infants who will reach criteria for treatment.
The overall MTCT of 10.3% in this cohort is a reduction of approximately 50% from pre-HIVNET 012 transmission rates of 19% in non-breastfed Durban infants . In this group, 31% of infants were infected IP and 69% IU. This reversed IU/IP transmission ratio occurs because NVP acts only to reduce IP transmission . In this cohort, sd-NVP reduced the viral load by a median of 1.4 log copies/ml between day 1 and day 5. Women whose infants were infected IP despite sd-NVP tended to have a particularly high viral load, which correlates strongly with disease progression in infected infants [14–16]. Furthermore, the majority of infants are now infected IU, and several studies have described more rapid disease progression in this group [11–13].
The NVP-induced 25-fold reduction of viral load in IU-infected infants in the first days of life is of significance where filter paper methods of HIV diagnosis using dried blood spots are employed. Up to 38% of IU infections would be undetected by dried blood spots testing at day 5 in this cohort . Collection of dried blood spots should, therefore, either be undertaken prior to infant NVP administration or delayed until the potential effects of NVP have passed in order to reduce underdiagnosing of IU-infected infants. Use of dried blood spots testing to diagnose IP infection at day 28 does not suffer from the same limitation.
Viral loads followed the kinetics of previously-described infant cohorts [16,27–30]. The peak viraemia is strikingly high (median 5 160 000 copies/ml IP; 984 000 copies/ml IU), with median viral load maintained > 1 000 000 copies/ml throughout the first 6 months (median 1 724 000 copies/ml beyond the first month of life). Overall, viral loads are higher than those reported in other African infants [16,27] and are similar to those seen in the subset of Western infants defined as rapidly progressing [11,28]. This is likely to reflect the high viral loads of mothers of infected infants, since sd-NVP has reduced the proportion of women transmitting with low viral loads.
A strength of this study, despite the small number of subjects, is the detailed characterization of immunological decline in this setting. Infants showed rapid progression to immunosuppression: in 35% of infants the CD4 cell% declined to ≤ 20% within 3 months of birth, and in 70% by 6 months. New guidelines from South Africa  and the WHO  recommend treatment for infants at CD4 cell thresholds of < 30–35%, and < 25%, respectively. If a 25% threshold was used in this cohort, 55% of infants would require treatment by 3 months of age, and 85% by 6 months.
These data present a dilemma for healthcare providers and policy makers in resource-limited settings, since the vast majority HIV-infected infants would qualify for ART within 6 months of life using the new WHO guidelines. Starting ART immediately upon diagnosis may be preferable to reduce costs of monitoring, enable treatment to start before immunological decline has occurred and minimize loss to follow-up. The challenge then becomes how to limit treatment failure in infants on ART in the absence of access to repeated CD4 cell count and viral load monitoring.
This study demonstrates that use of a perinatal regimen that effectively blocks IP transmission shifts the timing of infection in those infants who fail prophylaxis to the IU-related transmission time period, and that IP transmission occurs increasingly in the setting of very advanced maternal disease (high viral load and low CD4 cell count). It is critical, therefore, that women have access to CD4 cell count testing in pregnancy and that ART is made available for women with advanced disease, both to improve their own health and to prevent transmission more effectively than current perinatal regimens. Because those infants who fail PMTCT strategies are likely to progress rapidly, early testing must be offered to HIV-exposed infants in resource-limited settings, and early ART be made available. Additionally, alternative strategies to lifelong ART need to be developed to address the burgeoning paediatric HIV epidemic.
Wendy Mphatswe, Natasha Blanckenberg and Gareth Tudor-Williams contributed equally to this study.
The HPP study group: Ayanda Cengimbo, Prakash Jeena, Nonhlanhla Nene, Danni Ramduth, Nigel Rollins, Hoosen Coovadia, Krista Dong; contributions from Kesia Mgwenya, Thandi Cele, Thandi Sikhakhane, Maud Mbambo, Thandekile Phahla, Deli Sindane, Thobekile Sibaya, Nicky Linda, Pretty Siphengane at the clinics; and Katherine Luzuriaga for helpful discussions of the manuscript are all gratefully acknowledged.
Sponsorship: This study was supported by a grant from the Secure the Future Initiative, Bristol Myers Squibb; Wellcome Trust; Doris Duke Charitable Foundation; and the Mark and Lisa Schwartz Foundation. The funders did not participate in the design and conduct of the study; collection, management, analysis and interpretation of the data; or preparation, review, or approval of the manuscript.
2. Barnhart HX, Caldwell MB, Thomas P, Mascola L, Ortiz I, Hsu HW, et al
. Natural history of human immunodeficiency virus disease in perinatally infected children: an analysis from the Pediatric Spectrum of Disease Project. Pediatrics 1996; 97:710–716.
3. Blanche S, Tardieu M, Duliege A, Rouzioux C, Le Deist F, Fukunaga K, et al
. Longitudinal study of 94 symptomatic infants with perinatally acquired human immunodeficiency virus infection. Evidence for a bimodal expression of clinical and biological symptoms. Am J Dis Child 1990; 144:1210–1215.
4. Dabis F, Elenga N, Meda N, Leroy V, Viho I, Manigart O, et al
. 18-Month mortality and perinatal exposure to zidovudine in West Africa. AIDS 2001; 15:771–779.
5. Mbori-Ngacha D, Nduati R, John G, Reilly M, Richardson B, Mwatha A, et al
. Morbidity and mortality in breastfed and formula-fed infants of HIV-1-infected women: A randomized clinical trial. JAMA 2001; 286:2413–2420.
6. Newell ML, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, Dabis F. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 2004; 364:1236–1243.
7. Spira R, Lepage P, Msellati P, van de Perre P, Leroy V, Simonon A, et al
. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda. Mother-to-Child HIV-1 Transmission Study Group. Pediatrics 1999; 104:e56.
8. Taha TE, Kumwenda NI, Broadhead RL, Hoover DR, Graham SM, van der Hoven L, et al
. Mortality after the first year of life among human immunodeficiency virus type 1-infected and uninfected children. Pediatr Infect Dis J 1999; 18:689–694.
9. Guay LA, Musoke P, Fleming T, Bagenda D, Allen M, Nakabiito C, et al
. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999; 354:795–802.
10. World Health Organization. Antiretroviral Drugs for Treating Pregnant Women and Preventing HIV Infection in Infants in Resource-limited Settings: Towards Universal Access. Recommendations for a Public Health Approach. Geneva: World Health Organization; 2006. Accessed 19 October, 2006 at http://www.who.int/hiv/pub/guidelines/WHOPMTCT.pdf
11. Dickover RE, Dillon M, Leung KM, Krogstad P, Plaeger S, Kwok S, et al
. Early prognostic indicators in primary perinatal human immunodeficiency virus type 1 infection: importance of viral RNA and the timing of transmission on long-term outcome. J Infect Dis 1998; 178:375–387.
12. Kuhn L, Steketee RW, Weedon J, Abrams EJ, Lambert G, Bamji M, et al
. Distinct risk factors for intrauterine and intrapartum human immunodeficiency virus transmission and consequences for disease progression in infected children. Perinatal AIDS Collaborative Transmission Study. J Infect Dis 1999; 179:52–58.
13. Mayaux MJ, Burgard M, Teglas JP, Cottalorda J, Krivine A, Simon F, et al
. Neonatal characteristics in rapidly progressive perinatally acquired HIV-1 disease. The French Pediatric HIV Infection Study Group. JAMA 1996; 275:606–610.
14. Abrams EJ, Wiener J, Carter R, Kuhn L, Palumbo P, Nesheim S, et al
. Maternal health factors and early pediatric antiretroviral therapy influence the rate of perinatal HIV-1 disease progression in children. AIDS 2003; 17:867–877.
15. Ioannidis JP, Tatsioni A, Abrams EJ, Bulterys M, Coombs RW, Goedert JJ, et al
. Maternal viral load and rate of disease progression among vertically HIV-1-infected children: an international meta-analysis. AIDS 2004; 18:99–108.
16. Rouet F, Sakarovitch C, Msellati P, Elenga N, Montcho C, Viho I, et al
. Pediatric viral human immunodeficiency virus type 1 RNA levels, timing of infection, and disease progression in African HIV-1-infected children. Pediatrics 2003; 112:e289.
17. World Health Organization. Antiretroviral Therapy of HIV Infection in Infants and Children in Resource-limited Settings: Towards Universal Access. Recommendations for a Public Health Approach. Geneva: World Health Organization; 2006. Accessed on 19th October, 2006 at http://www.who.int/hiv/pub/guidelines/WHOpaediatric.pdf
20. Bobat R, Coovadia H, Coutsoudis A, Moodley D. Determinants of mother-to-child transmission of human immunodeficiency virus type 1 infection in a cohort from Durban, South Africa. Pediatr Infect Dis J 1996; 15:604–610.
21. Bryson YJ, Luzuriaga K, Sullivan JL, Wara DW. Proposed definitions for in utero
versus intrapartum transmission of HIV-1. N Engl J Med 1992; 327:1246–1247.
22. World Health Organization. Scaling up Antiretroviral Therapy in Resource-limited Settings, 2003 Revision. Geneva: World Health Organization; 2003.
23. Luzuriaga K, McManus M, Catalina M, Mayack S, Sharkey M, Stevenson M, Sullivan JL. Early therapy of vertical human immunodeficiency virus type 1 (HIV-1) infection: control of viral replication and absence of persistent HIV-1-specific immune responses. J Virol 2000; 74:6984–6991.
24. The mode of delivery and the risk of vertical transmission of human immunodeficiency virus type 1–a meta-analysis of 15 prospective cohort studies. International Perinatal HIV Group
. N Engl J Med
25. Cassol S, Gill MJ, Pilon R, Cormier M, Voigt RF, Willoughby B, et al
. Quantification of human immunodeficiency virus type 1 RNA from dried plasma spots collected on filter paper. J Clin Microbiol 1997; 35:2795–2801.
26. Coutsoudis A, Pillay K, Kuhn L, Spooner E, Tsai WY, Coovadia HM. Method of feeding and transmission of HIV-1 from mothers to children by 15 months of age: prospective cohort study from Durban, South Africa. AIDS 2001; 15:379–387.
27. Biggar RJ, Janes M, Pilon R, Miotti P, Taha TE, Broadhead R, et al
. Virus levels in untreated African infants infected with human immunodeficiency virus type 1. J Infect Dis 1999; 180:1838–1843.
28. De Rossi A, Masiero S, Giaquinto C, Ruga E, Comar M, Giacca M, et al
. Dynamics of viral replication in infants with vertically acquired human immunodeficiency virus type 1 infection. J Clin Invest 1996; 97:323–330.
29. Palumbo PE, Raskino C, Fiscus S, Pahwa S, Fowler MG, Spector SA, et al
. Predictive value of quantitative plasma HIV RNA and CD4+ lymphocyte count in HIV-infected infants and children. JAMA 1998; 279:756–761.
30. Shearer WT, Quinn TC, LaRussa P, Lew JF, Mofenson L, Almy S, et al
. Viral load and disease progression in infants infected with human immunodeficiency virus type 1. Women and Infants Transmission Study Group. N Engl J Med 1997; 336:1337–1342.
31. Southern African HIV Clinicians Society. Antiretroviral Treatment in Children, Draft 2005. Pretoria: Southern African HIV Clinicians Society; 2005.