Long-term risk of mortality and loss to follow-up in children and adolescents on antiretroviral therapy in Asia.

Abstract

We described mortality and loss to follow-up (LTFU) in children and adolescents who were under care for more than 5 years following initiation of antiretroviral therapy (ART). Patients were followed from 5 years after ART until the earlier of their 25th birthday, last visit, death, or LTFU. We used Cox regression to assess predictors of mortality and competing risk regression to assess factors associated with LTFU. In total, 4488 children and adolescents initiating ART between 1997 and 2016 were included in the analysis, with a median follow-up time of 5.2 years. Of these, 107 (2.2%) died and 271 (6.0%) were LTFU. Mortality rate was 4.35 and LTFU rate 11.01 per 1000 person-years. Increased mortality was associated with AIDS diagnosis (adjusted hazard ratio [aHR] 1.71; 95% confidence interval [CI] 1.24-2.37), current CD4 count <350 cells/mm3 compared with ≥500 (highest aHR 13.85; 95% CI 6.91-27.76 for CD4 <200), viral load ≥10 000 copies/mL compared with <400 (aHR 3.28; 95% CI 1.90-5.63), and exposure to more than one ART regimen (aHR 1.51; 95% CI 1.14-2.00). Factors associated with LTFU were male sex (adjusted subdistribution hazard ratio [asHR] 1.29; 95% CI 1.04-1.59), current viral load >1000 copies/mL compared with <400 (highest asHR 2.36; 95% CI 1.19-4.70 for viral load 1000-9999), and ART start after year 2005 compared with ≤2005 (highest asHR 5.96; 95% CI 1.98-17.91 for 2010-2016). For children and adolescents surviving 5 years on ART, both current CD4 and viral load remained strong indicators that help to keep track of their treatment outcomes. More effort should be made to monitor patients who switch treatments.