Mortality in the year following antiretroviral therapy initiation in HIV-infected adults and children in Uganda and Zimbabwe

Abstract

Background: Adult mortality in the first 3 months on antiretroviral therapy (ART) is higher in low-income than in high-income countries, with more similar mortality after 6 months. However, the specific patterns of changing risk and causes of death have rarely been investigated in adults, nor compared with children in low-income countries.

Methods: We used flexible parametric hazard models to investigate how mortality risks varied over the first year on ART in human immunodeficiency virus-infected adults (aged 18-73 years) and children (aged 4 months to 15 years) in 2 trials in Zimbabwe and Uganda.

Results: One hundred seventy-nine of 3316 (5.4%) adults and 39 of 1199 (3.3%) children died; half of adult/pediatric deaths occurred in the first 3 months. Mortality variation over year 1 was similar; at all CD4 counts/CD4%, mortality risk was greatest between days 30 and 50, declined rapidly to day 180, then declined more slowly. One-year mortality after initiating ART with 0-49, 50-99 or ≥ 100 CD4 cells/μL was 9.4%, 4.5%, and 2.9%, respectively, in adults, and 10.1%, 4.4%, and 1.3%, respectively, in children aged 4-15 years. Mortality in children aged 4 months to 3 years initiating ART in equivalent CD4% strata was also similar (0%-4%: 9.1%; 5%-9%: 4.5%; ≥ 10%: 2.8%). Only 10 of 179 (6%) adult deaths and 1 of 39 (3%) child deaths were probably medication-related. The most common cause of death was septicemia/meningitis in adults (20%, median 76 days) and children (36%, median 79 days); pneumonia also commonly caused child deaths (28%, median 41 days).

Conclusions: Children ≥ 4 years and adults with low CD4 values have remarkably similar, and high, mortality risks in the first 3 months after ART initiation in low-income countries, similar to cohorts of untreated individuals. Bacterial infections are a major cause of death in both adults and children; targeted interventions could have important benefits.

Figure 1.

Kaplan-Meier mortality 1 year after antiretroviral therapy (ART) initiation according to age and pre-ART CD4 count. Abbreviations: ART, antiretroviral therapy; ARROW, Antiretroviral Research for Watoto; DART, Development of Antiretroviral Therapy in Africa.

Figure 2.

Daily risk of death and survival through 1 year on antiretroviral therapy (ART) according to age and pre-ART CD4 count. Flexible parametric model [20, 21] on log-normal scale with 1 interior knot. Points show times when deaths occurred. Abbreviations: ART, antiretroviral therapy; ARROW, Antiretroviral Research for Watoto; DART, Development of Antiretroviral Therapy in Africa; PY, person-years.

Figure 3.

Daily risk of death and survival through 1 year before and on antiretroviral therapy (ART). Flexible parametric model [20, 21] on log-normal scale with 1 interior knot. Points show times when deaths occurred. Fewer than 40 children aged 1–3 years with 0%–4% pre-ART CD4 count were enrolled in the 3Cs4kids study (data not shown). Abbreviations: ART, antiretroviral therapy; PY, person-years.