Kidney function parameters in South African children living with HIV on antiretroviral therapy.

BackgroundWe evaluated longitudinal kidney function in children with perinatally acquired HIV (CWH) compared with HIV-negative controls.MethodsWe analyzed data from an observational study in Johannesburg, South Africa (2013-2018), including 220 CWH on non-tenofovir regimens and 220 controls. Baseline and follow-up visits with blood, urine, and dual-energy X-ray absorptiometry (DXA) were conducted a median of 3.7 years apart. Kidney function was assessed using serum creatinine-based (revised bedside Schwartz equation) and cystatin C-based (univariate CKiD cystatin C equation) estimated glomerular filtration rates (eGFR) and urinary microalbuminuria, proteinuria, and β2-microglobulin. We assessed between-group differences stratified by sex. Multivariable logistic regression was utilized to identify factors associated with microalbuminuria and proteinuria.ResultsAt baseline, mean age was 6.4 years among CWH and 7.0 years among controls. At baseline, the cohort included 220 CWH and 220 controls; 49.1% of CWH and 54.6% of controls were male. Among the 190 CWH and 111 controls who completed follow-up, 48.9% of CWH and 60.3% of controls were male. CWH had lower lean mass, and higher mean creatinine- and cystatin-C based eGFR than controls. Rate of change in eGFR did not differ between CWH and controls during follow-up. An eGFR <90 mL/min/1.73 m2 by cystatin C was observed in 13.3% male CWH vs 25.8% male controls (p=0.025) and 14.5% female CWH vs 25.9% female controls (p=0.049). At follow-up, proteinuria was detected in 20.7% of male CWH vs 7.5% male controls (p=0.030) and 34.4% female CWH vs 36.4% female controls (p=0.85). eGFR was >90 mL/min/1.73m2 in all children with proteinuria. Female sex was associated with higher odds of microalbuminuria (aOR 5.39, 95% CI 1.26-37.25) and proteinuria (aOR 2.59, 95% CI 1.44-4.76).ConclusionsAmong early treated children with perinatally acquired HIV on non-tenofovir regimens, eGFR was higher than in HIV negative controls, while annualized eGFR change and most urine biomarkers were similar by HIV status. However, a substantial proportion of both groups had eGFR below 90 mL/min/1.73 m2 and abnormal urinary markers, and male CWH had a higher prevalence of proteinuria than male controls; these findings suggest that a portion of the population have signs of abnormal kidney function that warrant further investigation and monitoring.