Changes in height, weight, head circumference, and MUAC z-scores with age in HIV-exposed infants.

Effects of tenofovir exposure on the different outcomes were similar adjusting for other factors at birth in Table 1; these cannot confound the relationship between growth and tenofovir exposure as there were no significant differences in these factors according to in utero tenofovir.aWould expect mean (standard deviation [SD]) of 0 (one) in a normal population: baseline is birth for weight and head circumference for age, week 12 for MUAC for age, and week 48 for height for age.bPredicted z-scores estimated from the fixed-effects population average fitted model at the ages above and back-transformed into sex-specific weights using the z-score formula.cMedian (IQR) per child with 0% or ≥90% in utero tenofovir exposure same as the whole group.dIdentical changepoints identified using profile log-likelihood as for the whole group.ep-Value from a 2 degree of freedom (df) Wald test (1 df for head circumference) comparing time trends before and after changepoint.fDriven by estimated small decline in height for age from 48–120 wk (and greater height for age at 48 wk) in those with ≥90% tenofovir in utero exposure (Figure 2): p = 0.38 comparing height-for-age trajectories after 120 wk.gp-Value from a 3 df Wald test (2 df for head circumference).