Growth at 2 Years of Age in HIV-exposed Uninfected Children in the United States by Trimester of Maternal Antiretroviral Initiation

Abstract

Background: Abnormal childhood growth may affect future health. Maternal tenofovir (TFV) use was associated with lower body length and head circumference at 1 year of age in HIV-exposed uninfected (HEU) US children.

Methods: We studied 509 HEU children in the US-based Surveillance Monitoring of Antiretroviral Therapy Toxicities cohort whose HIV-infected mothers were not using antiretrovirals at the last menstrual period and began combination antiretroviral therapy (cART) in pregnancy (cART initiators). We examined adjusted associations between antiretrovirals and Centers for Disease Control 2000 growth Z scores at 2 years of age within trimester of cART initiation: weight (weight Z score), length (length Z score), weight-for-length [weight-for-length Z score (WFLZ)], triceps skinfold Z score (TSFZ) and head circumference (head circumference Z score).

Results: Mothers mean age was 28.6 years; 57% were black non-Hispanic and 19% delivered at <37 weeks gestation. At 2 years, mean weight Z score, length Z score, WFLZ and head circumference Z score were above average (P < 0.05), whereas TSFZ (P = 0.57) did not differ from average. WFLZ was >1.64 standard deviation (SD) (>95th percentile) in 13%. Among children of first-trimester cART initiators, TFV+emtricitabine-exposed children had slightly higher mean WFLZ (0.45 SD; 95% confidence interval: -0.10 to 1.00) and lower TSFZ (-0.55 SD; 95% confidence interval: -1.07 to -0.02) compared with zidovudine+lamivudine-exposed children. TSFZ was lower in those exposed to boosted protease inhibitors. In contrast, growth in children of second trimester cART initiators did not differ by antiretroviral exposures.

Conclusion: Growth was above average in HEU; 13% were obese. Maternal TFV use was not associated with lower length or head circumference at 2 years of age, as hypothesized, but may be related to greater weight among those exposed to cART early in pregnancy.

Figure 1

Flowchart for selection of HEU children with anthropometric measures at 2 years of age (+/−4 months) whose HIV-infected mothers initiated cART during this pregnancy as of April 1, 2013

Figure 2

This figure shows the adjusted differences in growth z-scores at 2 years of age (+/− 4 months) in HEU children for each ARV within each trimester of maternal cART initiation.^{1 1} ARV= antiretroviral type of regimen; cART = combination antiretroviral therapy; TFV = tenofovir; FTC = emtricitabine; ZDV = zidovudine; 3TC = lamivudine; bPI =Boosted protease inhibitor; Un-bPI = Unboosted PI; ATV = atazanavir; NFV = nelfinavir. WTZ =weight z; LNZ = length z; WFLZ = weight-for-length z; TSFZ = triceps skinfold z; HCZ = head circumference z. Adjusted models include the following covariates for each outcome by trimester of cART initiation: WTZ 1^st (site region, alcohol, tobacco, language at home); WTZ 2^nd (site region, language at home); LNZ 1^st (site region, alcohol, tobacco); LNZ 2^nd (site region, tobacco, language at home, living arrangement); WFLZ 1^st (site region, alcohol, language at home); WFLZ 2^nd (site region, tobacco, language at home); TSFZ 1^st (site region, alcohol, tobacco, birth year); TSFZ 2^nd (site region, living arrangement); HCZ 1^st (region, alcohol, tobacco); HCZ 2^nd (site region, tobacco, income, language at home). Marital status, race/ethnicity, birth outside the mainland US, and prior ARV use did not cause confounding and were not included.