LAUSR

OBJECTIVE To define the prevalence of early cardiac dysfunction in children and young adults with perinatally acquired HIV and predictors of cardiac function. DESIGN Cross-sectional design. METHODS Early cardiac dysfunction was defined as left ventricular (LV) global longitudinal strain z-score less than -2 or myocardial performance index at least 0.5 with normal LV ejection fraction. Regression models were fitted to assess the relationship between measures of cardiac function and HIV RNA levels, clinical variables, and markers of inflammation. RESULTS Six-hundred and forty-three individuals (mean age 14.1 ± 5.2 years) were enrolled. The average time on combination antiretroviral treatment was 6.8 ± 3.6 years. Nearly 28% of individuals met criteria for early cardiac dysfunction. Individuals with early cardiac dysfunction were older (15.3 vs. 13.5 years, P < 0.001), had more frequently detectable HIV RNA (52.5 vs. 41.7%, P = 0.018), were more likely exposed to azidothymidine or zidoudine (55.6 vs. 41.2%, P = 0.002), and had higher median level of plasma IL-6 concentrations (1.00 vs. 0.88 pg/ml, P = 0.011). Multivariable models show LV ejection fraction negatively associated with HIV RNA levels [β -0.18; 95% confidence interval (CI) -0.33, -0.03] and azidothymidine or zidoudine exposure (β -1.75; 95% CI -2.62, -0.88) and positively associated with proportion of life on combination antiretroviral treatment (β 2.65; 95% CI 0.90, 4.41). Higher myocardial performance index was positively associated with serum inflammation marker (IL-6 β 0.01; 95% CI 0.0001, 0.001). Left ventricular global longitudinal strain was not significantly associated with clinical and laboratory variables of interest. CONCLUSION Over one-quarter of children and young adults living with HIV demonstrated evidence of cardiac dysfunction, which may be associated with increasing levels of systemic inflammation.