Hospitalization in South African Adolescents With Perinatally Acquired HIV on Antiretroviral Therapy

Abstract

Background: Little is known about hospitalization in African adolescents with perinatally acquired HIV (PHIV+ adolescents). We described the incidence and causes of hospitalization in participants enrolled in the Cape Town Adolescent Antiretroviral Cohort in South Africa.

Methods: Data collected from July 2013 to October 2018 from PHIV+ and HIV- adolescents were analyzed. Participants were assessed every 6 months and data on intercurrent hospitalization were abstracted. Causes of hospitalizations were classified according to ICD-10 codes. Descriptive statistics, time-to-event analysis and Poisson regression were used to describe causes and incidence and to determine incidence rate ratios for factors associated with hospitalization.

Results: Five hundred fifteen PHIV+ and 109 HIV- participants had a median follow-up of 4.1 years [interquartile range (IQR): 3.7-4.6]. At enrollment HIV+ participants had a median duration of ART of 7.6 years (IQR: 4.6-9.2), median CD4 count of 713 cells/mm (IQR: 561.0-957.5) and 387 (75%) had a viral load <50 copies/mL. There were 149 hospital admissions over 64 months. Crude incidence rates for hospitalization were 6.6 [95% confidence interval (CI): 5.7-7.8] and 2.2 (95% CI: 1.2-4.3) per 100-person-years (P = <0.01) in HIV + and HIV-, respectively. Ninety of 149 (60%) admissions in HIV+ participants were classified as non-infectious, 36/149 (24%) were infectious and 23/149 (16%) were "other HIV-related" or "unknown." Older age (15-19 years) and maintaining a CD4 >500 cells/cm were associated with decreased risk of hospitalization: adjusted incidence rate ratios of 0.61 (CI: 0.44-0.86, P = <0.01) and 0.68 (CI: 0.49-0.94, P = 0.02), respectively.

Conclusions: PHIV+ adolescents had a high incidence of hospitalization despite ART. Strategies addressing infectious and non-infectious morbidity must be strengthened.