Abstract Body

Background: Due to limited access to dual x-ray absorptiometry (DXA), few studies have described bone quality among HIV+ children in resource-limited settings (RLS), where the majority of HIV-infected children reside. Quantitative ultrasonography (QUS) is portable, relatively inexpensive, quick, simple to use and does not involve radiation exposure, making it well-suited for monitoring and evaluating bone acquisition in RLS. Although data are limited, QUS appears to have good agreement with DXA. We compare bone quality by QUS of South African HIV+ children receiving antiretroviral therapy (ART) to healthy HIV- children recruited as a control group.

Methods: Data were obtained from the bone sub-study (79% of target sample accrued) of CHANGES (Childhood HAART Alterations in Normal Growth, Genes, and aGing Evaluation Study), a longitudinal study of perinatally HIV+ children and HIV- controls in Johannesburg, South Africa. Weight- (WAZ), height- (HAZ), and BMI-for-age (BAZ) z-scores were calculated using WHO standards; CD4 and HIV-1 RNA levels were measured. Speed of sound (SOS) and broadband ultrasound attenuation (BUA) at the heel/calcaneus were measured by QUS (Lunar Achilles Insight). Calcaneus stiffness index (SI) was calculated as per manufacturer: SI = (0.67 x BUA + 0.28 x SOS) – 420.

Results: 206 HIV+ children (49.5% male) and 140 controls (55% male) were included in this analysis. HIV+ children were younger than controls (mean age 6.4±1.3 vs 7.2±1.5 years, p<0.01). All HIV+ children were on ART (52% PI-based, 48% NNRTI-based regimens; none receiving tenofovir; mean duration on ART 5.7±1.1 years), 72% had undetectable HIV-1 RNA levels, and median CD4% was 37.7. HIV+ children had lower mean WAZ (-0.8 vs -0.4, p<0.01) and HAZ (-1.4 vs -0.9, p<0.01) than controls. Median SOS was similar between HIV+ and HIV- children (1580 vs 1573, p=0.15). Unadjusted mean BUA was lower in HIV+ than HIV- children (71.5 vs 84.3, p<0.01) and remained lower after adjustment for age, sex, weight, and height (p<0.01). Unadjusted mean SI was also lower in HIV+ than HIV- children (78.5 vs 82.0, p=0.03), remaining lower after adjustment for age, sex, weight, and height (78.6 vs 81.9, p=0.048).

Conclusions: In this South African sample of school aged children, lower indices of bone quality by QUS are detectable among HIV+ children receiving ART compared to HIV- controls. QUS may prove to be a valuable method to assess bone quality and acquisition in HIV-infected children in RLS.