Share this article on:

Breastfeeding, HIV status and weights in South African children: a comparison of HIV-exposed and unexposed children

Patel, Devena,b; Bland, Rutha,c; Coovadia, Hoosend; Rollins, Nigele; Coutsoudis, Annad; Newell, Marie-Louisea,b

doi: 10.1097/QAD.0b013e3283345f91
Epidemiology and social

Objectives: To examine growth of children by maternal and infant HIV status allowing for infant feeding mode.

Design: Women enrolled into a nonrandomized intervention cohort.

Methods: Children of HIV-infected and uninfected women weighed and assessed for HIV status, monthly: from birth to 9 months; quarterly: 10–24 months. Daily infant feeding practices recorded at weekly intervals. Weight-for-age z-scores of children born to HIV-infected mothers compared with the reference population of children of HIV-uninfected mothers. Changes in z-scores over age were examined by HIV infection status and infant feeding practice using linear mixed effects models.

Results: The 1261 children of HIV-infected mothers grew as well as the reference group of 1061 children of HIV-uninfected mothers, irrespective of feeding mode. z-scores for HIV-infected children were consistently lower than those of HIV-exposed but uninfected children: a difference of 420 g for male children and 405 g for female children at 52 weeks of age. Breastfed HIV-infected infants had consistently higher z-scores for weight, especially during first 6 weeks (difference of 130 g for male children; 110 g for female children). In an adjusted regression analysis, maternal mid-upper arm circumference, CD4 cell count, infant birth weight and HIV status had the biggest impact on infant growth (z-score coefficient: 0.38 for mid-upper arm circumference ≥28.35 vs. <25.7 cm; P < 0.001; −0.32 for CD4 cell count <200 vs. ≥500; P = 0.001; −2.01 for birth weight <2500 vs. ≥2500 g; P < 0.001; −0.20 for infected vs. uninfected children; P < 0.001).

Conclusion: Optimal early feeding practices ameliorate the effect of being born to an HIV-infected mother and strengthen the recommendation of exclusive breastfeeding for HIV-infected women in terms of long-term child health.

aAfrica Centre for Health and Population Studies, University of KwaZulu-Natal, Mtubatuba, KwaZulu-Natal, South Africa

bCentre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, University College London, London, UK

cDivision of Developmental Medicine, University of Glasgow Medical Faculty, Glasgow, UK

dCentre for HIV/AIDS Networking

eDepartment of Paediatrics and Child Health, University of KwaZulu-Natal, Durban, KwaZulu-Natal, South Africa.

Received 25 August, 2009

Revised 18 October, 2009

Accepted 20 October, 2009

Correspondence to Ruth Bland, Africa Centre for Health and Population Studies, University of KwaZulu-Natal, PO Box 198, Mtubatuba, KwaZulu-Natal 3935, South Africa. Tel: +27 35 550 7500; fax: +27 35 550 7565; e-mail:

Back to Top | Article Outline


Poor growth has been reported in HIV-infected children and is associated with decreased survival [1–6]. In untreated children, HIV impact on growth is considerable, but HAART may ameliorate this effect [5]. Furthermore, compared with infants of HIV-uninfected mothers, all infants of HIV-infected mothers have significantly lower mean birth weights, especially when maternal HIV is advanced [2,7–12].

Far less is known about the effect of maternal HIV infection on subsequent growth of HIV-exposed, but uninfected, children, particularly from Africa. European cohorts have described normal growth in HIV-exposed children compared with children born to HIV-uninfected mothers [1,5] as has a very small study from Rwanda in the late 1980s [13]. However, small studies from Durban, South Africa [3] and Zambia [14] have suggested lower weights in HIV-uninfected infants born to HIV-infected mothers compared with population standards or infants of HIV-uninfected mothers. Further, results from another recently published trial from Zambia show marked declines in weight-for-age z-scores in HIV-uninfected children of HIV-infected mothers between 4.5 and 15 months, though the decline was less marked in infants who were breastfed [15]. In large African cohorts, growth has not been compared with that of children born to HIV-uninfected mothers from the same population, nor have growth patterns over age been based on appropriate population-based reference standards [16,17].

Infant feeding also significantly affects growth. The new WHO growth standards, launched in 2006 and designed to show how children should grow with optimal nutrition, acknowledge early differences in growth patterns between breastfed and nonbreastfed infants, with the mean weight of breastfed infants being above the median of the American standards (with most children formula fed), during the first half of infancy [18]. Early life growth patterns of uninfected children of HIV-infected mothers, by feeding patterns over age, have not been described in detail [14].

Recommendations on infant feeding for HIV-infected mothers have largely been based on risks of HIV transmission and early morbidity [19]. Information on infant growth associated with early feeding practices is important for policy makers and provides further information to balance the risks and benefits associated with breastfeeding – that is, acquisition of HIV through breast milk vs. the morbidity associated with avoidance or early cessation of breastfeeding.

Extensive breastfeeding support was provided to women in the Vertical Transmission study (VTS) [20,21], which thus provides a unique opportunity to explore the effect of exclusive breastfeeding (EBF) in the first 6 months of life and the impact of maternal HIV infection, on growth, in terms of weight, in the first 24 months of life. Weight is examined as it is a useful medium and long-term indicator of outcome; feasible to measure in resource-limited settings; and interpretable in global terms. We assessed factors affecting weights of children of HIV-infected mothers with respect to standards from a directly comparable reference group of children born to HIV-uninfected mothers from the same area.

Back to Top | Article Outline

Population and methods

HIV-infected and HIV-uninfected pregnant women were enrolled into the VTS, a nonrandomized intervention cohort study, from seven rural, one semi-urban and one urban primary healthcare clinics in KwaZulu-Natal (2001–2004). Women were counseled on infant feeding choices antenatally and supported postnatally by lay counselors during home visits [20]. An independent group of field workers conducted home visits weekly from birth to 9 months and documented all feeds and morbidity episodes for each day of the preceding week. HIV prevalence rate among pregnant women in the study area was approximately 40% [22]. High rates of EBF were achieved, with a median duration of approximately 175 days, irrespective of maternal HIV status [20].

After delivery, mothers and infants attended study clinics, monthly from 6 weeks to 9 months and then quarterly to 2 years. At each visit, the child was assessed by a study nurse who recorded a medical history and performed a physical examination. In addition, HIV status was determined by quantitative HIV RNA assay. For this analysis, infants were categorized as uninfected, infected perinatally (positive PCR at 4–8 weeks after delivery) or infected postnatally (positive PCR after 8 weeks of age) [21,23,24]. Gestational age was based on both the date of the last menstrual period and palpation by the midwife. Prematurity and severe prematurity were defined as deliveries 37 or less and 34 or less weeks gestation, respectively.

Anthropometric measurements were carried out at each visit by trained study clinic assistants, who had completed secondary school and received standardized anthropometry training. Strict adherence to the techniques taught was emphasized during regular in-service training and quality assurance checks were conducted by the clinic manager throughout the study period.

All clinics used the same measuring equipment: a scale with precision within 10 g for newborns; and for older children, portable electronic weighing scales with taring capability and a precision to 100 g (UNICEF Electronic Scale 890), as recommended by WHO [25]. The clinic manager checked calibration using standard metal weights at approximately weekly intervals.

Other than at birth, when infants were placed in a weighing tray, children were held by their mothers for weighing. First, a mother's weight was recorded and then the scale tared and the baby given to her to hold. Each child had two weights recorded; the mean value was later calculated during analysis. If the two recorded weights had a difference greater than 100 g, a third measurement was taken and the two weights within 100 g were recorded [25]. Weights were measured at scheduled clinic visits at birth, 6, 10, 14, 18, 22 and 26 weeks and then 7, 8, 9, 12, 15, 18, 21 and 24 months. Children's lengths and head circumferences were also measured at scheduled visits, but this analysis examines weights only, for the reasons given above.

Anthropometric measurements were also taken on mothers. The mid-upper arm circumference (MUAC), taken at 6 weeks postdelivery, using a 5-mm blank lead-in millimeter graded tape measure, has been used in these analyses [25].

At the time of the study, HAART was not available for treatment to delay HIV disease progression for either women or children, but single-dose nevirapine was provided to all HIV-infected women and their infants to reduce peripartum mother-to-child transmission [26]. The Biomedical Research Ethics Committee, University of KwaZulu-Natal, approved the study; all participating women gave written informed consent.

Back to Top | Article Outline

Data management

Data, recorded on individual study forms, were scanned using optical imaging recognition software (Teleform, San Diego, California, USA), thus minimizing data entry errors.

To be eligible for this analysis, maternal HIV status had to be known and the child had to have a gestational age at birth between 18 and 45 weeks and at least one growth measurement after birth. Gestational age was assessed by the study nurse at the antenatal booking visit using a combination of date of last menstrual period and abdominal palpation; ultrasounds were not routinely conducted. This resulted in a small proportion of liveborn children with extreme gestational ages at delivery who were excluded from analysis.

Only growth measurements on HIV-uninfected children born to HIV-uninfected mothers contributed data to the reference population. Children weighing less than 2500 g at birth were excluded if they were born preterm, but included if born at term. To avoid the influence of measurement outliers in the construction of the weight-for-age standards, seven (<0.1%) observations falling outside ±4 SDs of the sample median were also excluded.

Back to Top | Article Outline

Statistical methods

Weight-for-age z-scores

z-scores for weight were calculated to examine levels and patterns of growth among children born to HIV-infected mothers. z-scores, indicating how many SDs an observation is above or below the mean, were constructed to allow comparison between children born to infected mothers and the reference population (HIV-uninfected children born to uninfected mothers). Weight-for-age standards for male and female children were constructed separately by fitting the Box-Cox normal distribution (LMS method) using the GAMLSS (Generalized Additive Models for Location, Scale and Shape) method [27]. The resulting growth standards for weight-for-age were then used to obtain z-scores for each weight measurement for children born to HIV-infected women by age and sex.

Back to Top | Article Outline

Examining deviations in growth from the reference population

After adjusting for infant's birth weight and maternal MUAC measured at 6 weeks postpartum, z-scores over age were examined by HIV-infection status, infant feeding practice and maternal CD4 cell count using linear mixed effects (LME) models. A mixed effects model assumes that each parameter is the sum of both a fixed and random component, wherein the fixed components are the same for all individuals in the population and the random components differ between individuals [28]. The use of a LME model, therefore, accounts for individual variability through the random effects, but allows for a population average to be estimated via the fixed effects. The model selection was based on the Akaike Information Criterion (AIC) [29]. The following variables were considered in the LME regression model and retained if inclusion resulted in an improved AIC: mother's MUAC, as a proxy for maternal nutrition status, maternal CD4 cell count closest to delivery, type of water source and latrine facility and rural/urban residence. The child's sex was not significant and its inclusion did not result in an improvement in fit of the model.

Calculations were done in R version 2.6.0 [30] using the function GAMLSS and LME.

Back to Top | Article Outline


A total of 1498 children were born to HIV-infected mothers. Thirty-six pairs of twins and 43 children with reported extreme gestational ages were excluded (<18 weeks or >45 weeks), leaving 1383 children of HIV-infected mothers. Of these, 1261 with at least one weight measurement after birth are included.

A total of 1432 children were born to HIV-uninfected mothers, 1250 had information on the child's HIV status and 1231 were HIV-negative (19 were HIV-infected, born to mothers with primary HIV infection). Twenty-one pairs of twins and 61 children with extreme gestational ages were excluded (<18 weeks or >45 weeks). A further 48 (4.1%) preterm children with a birth weight less than 2500 g and 19 (1.8%) preterm children with missing birth weights were also excluded leaving 1061 children for analysis.

Compared with HIV-uninfected women, HIV-infected women were older, more likely to deliver premature and low birth weight infants and to have fewer clinic visits (Table 1). Deaths among HIV-infected women and their infants were more likely than those in HIV-uninfected women. Sociodemographic characteristics of HIV-infected and HIV-uninfected women were similar.

Table 1

Table 1

Back to Top | Article Outline

Growth of children born to HIV-infected women

Figure 1 shows the weight z-scores for infected and uninfected children of HIV-infected mothers. z-scores for HIV-infected children are below the reference standard; at 6 weeks, the difference was 240 g for male children and 200 g for female children and at 52 weeks, 420 g for male children and 405 g for female children.

Fig. 1

Fig. 1

A total of 910 (75%) children, with 9411 measurements, born to HIV-infected mothers were included in a multivariable model; 300 (25%), with 2154 measurements, were excluded because of missing data on birth weight, maternal CD4 cell count and MUAC. There were no significant differences between those excluded from this analysis and those included with respect to the mean z-score at birth (mean −0.22 vs. −0.24; P = 0.77) or gestational age at delivery (36.9 vs. 37.7 weeks; P = 0.44), though those excluded had significantly fewer study visits [median 2 (interquartile range; IQR 1–9) vs. 11 (IQR 7–15); P < 0.0001].

Of the variables considered for inclusion in the model, a linear term in age, birth weight, maternal MUAC, maternal CD4 cell count closest to delivery, infant feeding type and infant HIV infection status were required in the final model, with the presence of an interaction effect between age and birth weight also improving the model fit significantly. The adjusted coefficients of change in z-score for weight are presented in Table 2. Again, the reference group is composed of HIV-uninfected children of HIV-uninfected mothers. HIV-exposed children, whose mothers were at a less advanced stage of disease, were more consistently of a more appropriate weight for their age than those whose mothers had lower CD4 cell counts (Table 2). Children born to mothers with a MUAC at least 28.35 cm were estimated to have a weight z-score 0.38 higher than that of a child born to a mother with a MUAC less than 25.7 cm, which at birth equates to being on the 58th weight percentile, as opposed to the 42nd. Overall, children given breast milk had an average z-score 0.048 higher compared with children receiving no breast milk.

Table 2

Table 2

To explore further the differences between HIV-uninfected and HIV-infected children, Fig. 2a shows the unadjusted z-scores for weight for HIV-uninfected children of HIV-infected mothers to oscillate around zero and this indicates that these children's growth is close to that of the reference population, irrespective of feeding mode. On the other hand, in HIV-infected infants, breastfeeding was associated with consistently higher z-scores than other feeding modes, more pronounced in the first 4 months, and particularly in the first 6 weeks, where the difference was 130 g for male children and 110 g for female children (Fig. 2b). This difference was more marked for infants infected perinatally; breastfed infants demonstrated substantially better growth in the first 6 weeks compared with those receiving no breast milk (data not shown).

Fig. 2

Fig. 2

Sociodemographic characteristics were examined to allow for differences in environmental and economic conditions (e.g. water or fuel), which may impact on complementary feeding preparation and subsequent child growth; these were not statistically different between HIV-infected and HIV-uninfected mothers (see Table 1). Most women had access to piped water, but few had piped water in their own homestead. Figure 3 shows the unadjusted z-scores of HIV-exposed children by water supply; the most marked differences were between 1 and 2 years of age when children with piped water in their homes had higher z-scores compared with those with other water supplies, irrespective of feeding modality.

Fig. 3

Fig. 3

Back to Top | Article Outline

Growth of children born to HIV-infected and HIV-uninfected women

Between 3 and 12 months of age, weights of children born to infected mothers differed significantly from those born to uninfected mothers (data not shown). On average, the weight of a child of an HIV-infected mother was 0.19 lower z-score than that of an HIV-uninfected child of an HIV-uninfected mother; this equates to the 42nd weight percentile based on reference curves created from the weight data on HIV-uninfected children born to uninfected women.

Back to Top | Article Outline


Growth of HIV-negative children of HIV-infected and uninfected mothers

This large African cohort of children, with repeated growth measurements over 2 years and rigorous collection of infant feeding data, allows detailed examination of weight gain in children of HIV-infected women compared with the reference population of infants born to HIV-uninfected women from the same setting. These novel analyses allow for maternal circumstances and environmental factors. The objective of this paper was not to describe the growth of children in rural KwaZulu-Natal and to compare this to international standards or the growth of children elsewhere, but to evaluate whether maternal HIV infection status and the mode of feeding made a difference to the longitudinal weights of children. For these reasons, we chose to develop our own reference standard. An alternative approach would have been to calculate weight-for-age z-scores using the new WHO growth standards [18], a methodology recently used in a cohort of HIV-exposed children from Zambia [15].

One of the key findings of this study is that the growth of HIV-exposed but HIV-uninfected children was as good as that of the reference population. This is a significant and new [3,14] result for two reasons: first, up to 40% of infants are born to HIV-infected mothers in southern Africa, of whom the majority will be exposed but uninfected. The growth and development of these children is important, but often overlooked. Second, previous studies have shown that uninfected-exposed children may be at greater risk of morbidity and mortality compared with infants of HIV-uninfected mothers, particularly in poor, rural settings like ours [31]. Undernutrition is a significant predictor of child mortality in developing countries [32–35], in areas of low [36] and high HIV prevalence [37,38] and is associated with poor early development [39]. We postulate that EBF for 6 months may compensate for vulnerabilities experienced by HIV-exposed-uninfected children [37,40], which may include maternal morbidity, decreased caring ability for infants and increased exposure to opportunistic infections [41,42].

Back to Top | Article Outline

Growth and infant feeding

Most infants in our study received breast milk [20]. The few HIV-infected women who opted for formula feed were carefully counseled and had better socioeconomic circumstances than those choosing for breastfeed [43], so were more likely to provide replacement feeds in an hygienic way. EBF for 6 months is critical for child survival [44–48]; international guidelines now recommend 6 months of EBF for HIV-exposed infants, unless they can be provided with replacement feeds in an ‘acceptable, feasible, affordable, sustainable and safe’ manner [19]. We suggest that the optimal feeding practices of our women, together with support and other appropriate health interventions, resulted in the good growth of the HIV-exposed-uninfected children, true for infants who were breastfed or formula-fed. Overall, children given breast milk had an average weight 0.048 higher z-score compared with children receiving no breast milk, after allowing for birth weight, maternal MUAC, HIV infection status and maternal CD4 cell count. (300 children were excluded from the multivariate model because of missing data, as detailed in the Results section; as there were no significant differences with respect to birth weight or gestational age, it is unlikely that the exclusion of these children will have substantially affected the overall results). These findings further strengthen the recommendation of EBF for HIV-infected women [19] – not only is transmission lower than in mothers who mixed breastfed [21,49], and mortality from infectious diseases lower (in some settings) than in those infants who were formula-fed [45,50], but their long-term health in terms of growth is improved.

Back to Top | Article Outline

Growth of children and maternal health and water supply

The birth weight of children born to HIV-infected mothers was lower than that of infants born to HIV-uninfected mothers, in line with previous studies [2,7,12]. Children of larger women, measured by MUAC, were consistently heavier for age from birth than children of smaller women; and infants born to HIV-infected mothers with advanced disease (CD4 cell count<200 cells/μl) were always lighter, in accord with at least one previous study [51]. To improve survival and longer term health of children, HIV-infected pregnant women with low CD4 cell counts need to be identified and fast-tracked for HAART to improve their virological, immunological and nutritional status prior to delivery [40].

There was no evidence to suggest that a potentially hazardous water supply would be associated with poor growth in the first 6 months, most likely because most women EBF. Once complementary feeds were introduced at 6 months of age, children in homes with piped water had the best z-scores for weight. Piped water not only provides a hygienic water supply with less risk of diarrhea for the child, but it is also a proxy for other socioeconomic circumstances, and those with piped water were probably more able to afford better quality complementary foods. However, this was only a trend that was unadjusted and based on small numbers, and further studies are required to substantiate this finding.

Back to Top | Article Outline

Growth of HIV-infected children

Not surprisingly, HIV-infected children were consistently lighter from birth compared with uninfected-exposed children, irrespective of feeding practice. Although the birth weights of infected children were lower than those of exposed-uninfected children, their weights continued to diverge from the early weeks to around 6–9 months of age, when the differences plateau. Data from Europe show continued divergence of weights of HIV-infected children until they commence on HAART when their growth improves [52–54]. During the study period, HAART was not available for HIV-infected infants in this area, resulting in early mortality of approximately half the infected children. HAART is now available in South Africa; the major challenge is identifying infected infants as early as possible and initiating treatment.

Back to Top | Article Outline


These encouraging results show that with good counseling for appropriate feeding choices and support for optimal feeding practices, HIV-uninfected children of HIV-infected women grew as well as those of uninfected women. These findings further strengthen the recommendations of EBF for HIV-infected women in resource-poor settings, for long-term child health.

Back to Top | Article Outline


We are grateful to all women and children enrolled in the study; the field, clinic, laboratory and data management staff at the Africa Center, including the managers (Thembi Blose, Zanele Fakude, Cookie Govender, Nqobile Mkhwanazi, Londiwe Mthethwa, Samukelisiwe Mtshali and Ntombizodumo Mkwanazi); Colin Newell who assisted with data management; Dave Perlman and Kobus Herbst for their assistance in database design; the Community Liaison Office at the Africa Center and the Community Advisory Board for their guidance and feedback throughout the study.

D.P. contributed to the analysis of data and drafted manuscript. R.B. contributed to the design and implementation of study, analysis of data and drafted manuscript. H.C. and A.C. contributed to the design of study and reviewed manuscript. N.R. contributed to the design and implementation of study and reviewed manuscript. M.-L.N. contributed to the design of study, analysis of data and drafted manuscript.

Wellcome Trust (Grant #063009/Z/00/2) and Africa Center (Grant #050534) supported this study.

Back to Top | Article Outline


1. Agostoni C, Zuccotti GV, Gianni ML, D'Auria E, Giovannini M, Riva E. Body mass index development during the first 6 months of life in infants born to human immunodeficiency virus-seropositive mothers. Acta Paediatr 1998; 87:378–380.
2. Arpadi SM. Growth failure in children with HIV infection. J Acquir Immune Defic Syndr 2000; 25(Suppl 1):S37–S42.
3. Bobat R, Coovadia H, Moodley D, Coutsoudis A, Gouws E. Growth in early childhood in a cohort of children born to HIV-1-infected women from Durban, South Africa. Ann Trop Paediatr 2001; 21:203–210.
4. Moye J Jr, Rich KC, Kalish LA, Sheon AR, Diaz C, Cooper ER, et al. Natural history of somatic growth in infants born to women infected by human immunodeficiency virus. Women and Infants Transmission Study Group. J Pediatr 1996; 128:58–69.
5. Newell ML, Borja MC, Peckham C. Height, weight, and growth in children born to mothers with HIV-1 infection in Europe. Pediatrics 2003; 111:e52–e60.
6. Saavedra JM, Henderson RA, Perman JA, Hutton N, Livingston RA, Yolken RH. Longitudinal assessment of growth in children born to mothers with human immunodeficiency virus infection. Arch Pediatr Adolesc Med 1995; 149:497–502.
7. Braddick MR, Kreiss JK, Embree JB, Datta P, Ndinya-Achola JO, Pamba H, et al. Impact of maternal HIV infection on obstetrical and early neonatal outcome. AIDS 1990; 4:1001–1005.
8. Bulterys M, Chao A, Munyemana S, Kurawige JB, Nawrocki P, Habimana P, et al. Maternal human immunodeficiency virus 1 infection and intrauterine growth: a prospective cohort study in Butare, Rwanda. Pediatr Infect Dis J 1994; 13:94–100.
9. Rollins NC, Coovadia HM, Bland RM, Coutsoudis A, Bennish ML, Patel D, Newell ML. Pregnancy outcomes in HIV-infected and uninfected women in rural and urban South Africa. J Acquir Immune Defic Syndr 2007; 44:321–328.
10. Taha TE, Dallabetta GA, Canner JK, Chiphangwi JD, Liomba G, Hoover DR, Miotti PG. The effect of human immunodeficiency virus infection on birthweight, and infant and child mortality in urban Malawi. Int J Epidemiol 1995; 24:1022–1029.
11. Thorne C, Patel D, Newell ML. Increased risk of adverse pregnancy outcomes in HIV-infected women treated with highly active antiretroviral therapy in Europe. AIDS 2004; 18:2337–2339.
12. Wei R, Msamanga GI, Spiegelman D, Hertzmark E, Baylin A, Manji K, Fawzi WW. Association between low birth weight and infant mortality in children born to human immunodeficiency virus 1-infected mothers in Tanzania. Pediatr Infect Dis J 2004; 23:530–535.
13. Lepage P, Msellati P, Hitimana DG, Bazubagira A, Van Goethem C, Simonon A, et al. Growth of human immunodeficiency type 1-infected and uninfected children: a prospective cohort study in Kigali, Rwanda, 1988 to 1993. Pediatr Infect Dis J 1996; 15:479–485.
14. Makasa M, Kasonka L, Chisenga M, Sinkala M, Chintu C, Tomkins A, Filteau S. Early growth of infants of HIV-infected and uninfected Zambian women. Trop Med Int Health 2007; 12:594–602.
15. Arpadi S, Fawzy A, Aldrovandi GM, Kankasa C, Sinkala M, Mwiya M, et al. Growth faltering due to breastfeeding cessation in uninfected children born to HIV-infected mothers in Zambia. Am J Clin Nutr 2009; 90:344–353.
16. Cole TJ. The LMS method for constructing normalized growth standards. Eur J Clin Nutr 1990; 44:45–60.
17. Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 1992; 11:1305–1319.
18. de Onis M, Garza C, Onyango AW, Borghi E. Comparison of the WHO child growth standards and the CDC 2000 growth charts. J Nutr 2007; 137:144–148.
19. WHO. WHO HIV and Infant Feeding Technical Consultation held on behalf of the inter-agency task team on prevention of HIV infections in pregnant women, mothers and their infants. Geneva, October 2006.
20. Bland RM, Little KE, Coovadia HM, Coutsoudis A, Rollins NC, Newell ML. Intervention to promote exclusive breast-feeding for the first 6 months of life in a high HIV prevalence area. AIDS 2008; 22:883–891.
21. Coovadia HM, Rollins NC, Bland RM, Little K, Coutsoudis A, Bennish ML, Newell ML. 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.
22. Rice BD, Batzing-Feigenbaum J, Hosegood V, Tanser F, Hill C, Barnighausen T, et al. Population and antenatal-based HIV prevalence estimates in a high contracepting female population in rural South Africa. BMC Public Health 2007; 7:160.
23. Alioum A, Cortina-Borja M, Dabis F, Dequae-Merchadou L, Haverkamp G, Hughes J, et al. Estimating the efficacy of interventions to prevent mother-to-child transmission of human immunodeficiency virus in breastfeeding populations: comparing statistical methods. Am J Epidemiol 2003; 158:596–605.
24. Alioum A, Dabis F, Dequae-Merchadou L, Haverkamp G, Hudgens M, Hughes J, et al. Estimating the efficacy of interventions to prevent mother-to-child transmission of HIV in breast-feeding populations: development of a consensus methodology. Stat Med 2001; 20:3539–3556.
25. de Onis M, Onyango AW, Van den Broeck J, Chumlea WC, Martorell R. Measurement and standardization protocols for anthropometry used in the construction of a new international growth reference. Food Nutr Bull 2004; 25:S27–S36.
26. KwaZulu Natal Provincial Department of Health. Protocol for the phased implementation of a comprehensive package of care for the prevention of mother to child transmission of HIV in KwaZulu Natal. 4th ed. Pietermaritzburg, South Africa; 2003.
27. Rigby RA, Stasinopoulos DM. Generalised Additive Models For Location, Scale And Shape (with discussion). Appl Statist 2005; 54:507–554.
28. Pinheiro JC, Bates DM. Mixed effects models in S and S-Plus. Springer Publishing; 2002.
29. Akaike H. A new look at the statistical model identification. IEEE Trans Automatic Contr 1974; 19:716–723.
30. R Development Core Team (2007). R: a language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. ISBN 3-900051-07-0, URL
31. Marinda E, Humphrey JH, Iliff PJ, Mutasa K, Nathoo KJ, Piwoz EG, et al. Child mortality according to maternal and infant HIV status in Zimbabwe. Pediatr Infect Dis J 2007; 26:519–526.
32. Black RE, Morris SS, Bryce J. Where and why are 10 million children dying every year? Lancet 2003; 361:2226–2234.
33. Caulfield LE, de Onis M, Blossner M, Black RE. Undernutrition as an underlying cause of child deaths associated with diarrhea, pneumonia, malaria, and measles. Am J Clin Nutr 2004; 80:193–198.
34. Kaler SG. Diseases of poverty with high mortality in infants and children: malaria, measles, lower respiratory infections, and diarrheal illnesses. Ann N Y Acad Sci 2008; 1136:28–31.
35. Rice AL, Sacco L, Hyder A, Black RE. Malnutrition as an underlying cause of childhood deaths associated with infectious diseases in developing countries. Bull World Health Organ 2000; 78:1207–1221.
36. Yoon PW, Black RE, Moulton LH, Becker S. The effect of malnutrition on the risk of diarrheal and respiratory mortality in children <2 y of age in Cebu, Philippines. Am J Clin Nutr 1997; 65:1070–1077.
37. Chatterjee A, Bosch RJ, Hunter DJ, Fataki MR, Msamanga GI, Fawzi WW. Maternal disease stage and child undernutrition in relation to mortality among children born to HIV-infected women in Tanzania. J Acquir Immune Defic Syndr 2007; 46:599–606.
38. Fawzi WW, Herrera MG, Spiegelman DL, el Amin A, Nestel P, Mohamed KA. A prospective study of malnutrition in relation to child mortality in the Sudan. Am J Clin Nutr 1997; 65:1062–1069.
39. Grantham-McGregor S, Cheung YB, Cueto S, Glewwe P, Richter L, Strupp B. Developmental potential in the first 5 years for children in developing countries. Lancet 2007; 369:60–70.
40. Kuhn L, Thea DM, Aldrovandi GM. Bystander effects: children who escape infection but not harm. J Acquir Immune Defic Syndr 2007; 46:517–518.
41. McNally LM, Jeena PM, Gajee K, Thula SA, Sturm AW, Cassol S, et al. Effect of age, polymicrobial disease, and maternal HIV status on treatment response and cause of severe pneumonia in South African children: a prospective descriptive study. Lancet 2007; 369:1440–1451.
42. Thea DM, St Louis ME, Atido U, Kanjinga K, Kembo B, Matondo M, et al. A prospective study of diarrhea and HIV-1 infection among 429 Zairian infants. N Engl J Med 1993; 329:1696–1702.
43. Bland RM, Rollins NC, Coovadia HM, Coutsoudis A, Newell ML. Infant feeding counselling for HIV-infected and uninfected women: appropriateness of choice and practice. Bull World Health Organ 2007; 85:289–296.
44. Creek T, Arvelo W, Kim A, Lu L, Bowen A, Finkbeiner T, et al. Role of infant feeding and HIV in a severe outbreak of diarrhoea and malnutrition among young children – Botswana 2006 [abstract 770. 2007]. CROI: 14th Conference on Retroviruses and Opportunistic Infections.
45. Kuhn L, Aldrovandi GM, Sinkala M, Kankasa C, Semrau K, Mwiya M, et al. Effects of early, abrupt weaning on HIV-free survival of children in Zambia. N Engl J Med 2008; 359:130–141.
46. WHO. Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries: a pooled analysis. WHO Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality. Lancet 2000; 355:451–455.
47. WHO. Nutrient adequacy of exclusive breastfeeding for the term infant during the first six months of life. Geneva: World Health Organization; 2002.
48. WHO. Global strategy for infant and young child feeding. Geneva: World Health Organization; 2003.
49. Iliff PJ, Piwoz EG, Tavengwa NV, Zunguza CD, Marinda ET, Nathoo KJ, et al. Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. AIDS 2005; 19:699–708.
50. Thior I, Lockman S, Smeaton LM, Shapiro RL, Wester C, Heymann SJ, et al. 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.
51. Kuhn L, Kasonde P, Sinkala M, Kankasa C, Semrau K, Scott N, 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.
52. Weight, height and human immunodeficiency virus infection in young children of infected mothers. The European Collaborative Study. Pediatr Infect Dis J 1995; 14:685–690.
53. Nadal D, Steiner F, Cheseaux JJ, Rudin C. Ritonavir promotes increased growth in HIV-infected children. Paediatric AIDS Group of Switzerland. AIDS 1998; 12:2356–2357.
54. Steiner F, Kind C, Aebi C, Wyler-Lazarevitch CA, Cheseaux JJ, Rudin C, et al. Growth in human immunodeficiency virus type 1-infected children treated with protease inhibitors. Eur J Pediatr 2001; 160:611–616.

exclusive breastfeeding; growth; HIV; infant feeding; LMS method; rural Africa; weight

© 2010 Lippincott Williams & Wilkins, Inc.