University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Survival of HIV-infected children
Survival of HIV-infected children: A cohort study from the Asia-Pacific region
Global Health Sciences Literature Digest
Published April 11, 2011
Journal Article

Lumbiganon P, Kariminia A, Aurpibul L, Hansudewechakul R, Puthanakit T, Kurniati N, Kumarasamy N, Chokephaibulkit K, Nik Yusoff NK, Vonthanak S, Moy FS, Mohd Razali KA, Nallusamy R, Sohn AH. Survival of HIV-infected children: A cohort study from the Asia-Pacific region. J Acquir Immune Defic Syndr. 2011 56:4,365-71.

In Context

The survival benefit associate with combination antiretroviral therapy (cART) among children in developed countries has been substantial and well described.(1,2) In resource-constrained countries in Asia, mono- and dual therapy were the only treatment options prior to 2002 at which time cART became available, although widespread scale up was inconsistent in the region.(3)


To describe the survival and factors associated with mortality among children in Asia


Public referral hospitals in urban areas in Thailand, Malaysia, India, Indonesia, and Cambodia

Study Design

Retrospective and prospective cohort


HIV-infected children aged 18 years and under who received care at participating clinical sites

Main Outcome Measures



Pediatric patients at participating sites were prospectively enrolled beginning in 2008 with retrospective collection of clinical data for patients who had been receiving care prior to 2008. Patients were excluded if they received care during periods when treatment data were not collected for patients who died or were lost to follow-up. Mortality rates were calculated by dividing the number of deaths by the total number of person-years of observation. Time at risk was measured starting from the date treatment was initiated or from at the first clinic visit. Patients were followed until death or the date of the last clinic visit. Patients who had not been seen at the clinic for more than 12 months were considered to be lost to follow-up.

An intent-to-treat approach was used in the analysis. Kaplan-Meier methods were used to estimate survival after initiation of antiretroviral therapy (ART). Predictors of mortality were measured using Cox proportional hazards models. Variables examined included ethnicity, weight-for-age and height-for-weight Z scores, body mass index, World Health Organization (WHO) clinical stage, hemoglobin level, initial ART regimen, CD4 percentages (treated as a time-dependent variable). The baseline laboratory values were those obtained three months before and one month after beginning ART. ART was classified as mono-therapy, dual therapy, or cART. Demographic and clinical variables with p values <0.10 in the univariate analysis were examined in the stepwise multivariate analysis.


There were 2,280 children included in the analysis of whom 1,752 (77%) had initiated ART; 1480 (84%) received cART and 272 (16%) had received either mono- or dual therapy. There were 528 ART-naïve patients. Ninety-seven percent of the patients who started cART presented for care in 2002 or later, while 70% of children who began treatment with mono or dual ART initiated treatment earlier. The median age at initiation of cART was 7.0 years and for initiation of mono- or dual therapy was 2.1 years. Mono- and dual therapy were started largely among children with baseline CD4 percentages less than 15%. Ninety-three percent of first-line cART was non-nucleoside reverse transcriptase inhibitor (NNRTI-based).

As of March 31, 2009 the median follow-up time for the children who had received ART was 3.1 years (interquartile range [IQR] 1.5-4.8); 2.9 years (IQR 1.4-4.6) for those who initiated cART, 5.1 years (IQR 2.7-8.7) for those who initiated mono-or dual therapy. There were 155 (8.8%) children who were lost to follow-up, and 119 (6.8%) transferred care for a total lost to the program rate of 4.6 per 100 person-years (py). Of the 1752 children at the start of observation, 115 (6.6%) died (mortality rate = 1.9 per 100 py). Thirty-seven percent of deaths occurred within the first three months of starting ART. The median age at death for children who had initiated cART was higher than the median age at death for those who began treatment with mono- or dual therapy; 7.8 (IQR: 4.8–10.9) and 5.8 (IQR: 2.7–7.9) respectively. The mortality rate declined from 10.2 per 100 person-years (95% CI 7.5 to 13.7) during the first 3 months of ART to 4.2 per 100 person-years (95% CI 2.6 to 6.8) during the fourth to sixth months, to 2.0 per 100 person-years (95% CI 1.2 to 3.3) from 6 to 12 months and to 0.9 per 100 person-years (95% CI 0.7 to 1.3) after 12 months. The probability of surviving five years after initiation of ART was 91.7% (95% CI 90.0, 93.2). Survival was worse for children with lower baseline CD4 percentages.

The median follow-up for the 528 patients who had never received ART was 0.9 years (IQR 0.1-2.50). Forty-nine only attended a clinic once and 36 (6.8%) died after receiving care. The mortality rate was 4.1 per 100 person-years (95% CI 3.0-5.7). Of the 528 ART-naïve patients, 241 (49%) were lost to follow-up and 34 (6%) transferred care which produced a total loss to the program of 31.5 per 100 person-years (95% CI 28.0-35.5).

The factors independently associated with death were CD4 percentages ([0-4%; HR 33.85, 95% CI 14.96-76.59]; [5-9%, HR 5.87, 95% CI 2.18-15.84]; [10-14%, HR 5.33, 95% CI 2.05, 13.84]; [15-19%, HR 4.02, 95% CI 1.53-10.57, >20, 1.0]), WHO clinical stage IV (HR 4.78, 95% CI 1.84-12.41), and weight-for-age Z score (HR 0.89, 95% CI 0.81-0.98).


The study findings support early diagnosis and initiation of cART in pediatric patients.

Quality Rating

Based on the Newcastle-Ottawa quality assessment scale, this was a good study. Information on the representativeness of the cohort was limited but given the diversity of countries from which patients are selected suggests that the population is fairly representative of children in HIV-care settings. Participants were drawn from same population. The ascertainment of clinical data appears to be inadequate given that information on adherence, viral loads, and immune reconstitution syndrome were not available.

The lack of association between use of cART and survival is unexpected. It is possible that this finding is an artifact from inadequate record keeping, biased follow-up, or other uncontrolled factors. Although the authors suggest that the use of mono- and dual- therapy contributed to this finding, it is in contrast to the numerous studies that demonstrate and independent effect of cART on survival. While it is evident that participants were alive at the start of the study, ascertainment of death was incomplete. Loss to follow-up was accounted for but may have introduced bias.

Programmatic Implications

The findings from this study support continued use of cART for children. The increased risk of death associated with initiating ART in the face of severe immune system compromise (i.e. higher risk of death was found with ART initiation at clinical stage IV and with lower baseline CD4 cell counts) supports efforts to increase early diagnosis and care. Findings from this study support those of other studies from resource constrained areas that found improved survival with cART and high mortality in the first 3 months of therapy.(4,5) The association between weight-for-age Z scores and mortality highlights the need to address nutrition as a key element in the care of HIV-infected children. Given that this study was unable to examine the possible effect of immune reconstitution syndrome added efforts to monitor its occurrence seems warranted.


  1. Judd A, Doerholt K, Tookey PA, et al. Morbidity, Mortality, and Response to Treatment by Children in the United Kingdom and Ireland with Perinatally Acquired HIV Infection during 1996—2006: Planning for Teenage and Adult Care. Clin Infect Dis. 2007;45:918–924
  2. Brady MT, Oleske JM, Williams PL, et al. Declines in mortality rates and changes in causes of death in HIV-1-infected children during the HAART era. J Acquir Immune Defic Syndr. 2010 Jan 1;53(1):86-94
  3. Chasombat S, Lertpiriyasuwat C, Thanprasertsuk S, et al. The national access to antiretroviral program for PHA (NAPHA) in Thailand. Southeast Asian J Trop Med Public Health. 2006;37:704–715
  4. KIDS-ART-LINC Collaboration. Low risk of death, but substantial program attrition, in pediatric HIV treatment cohorts in Sub-Saharan Africa 2008 Dec 15;49(5):523-31
  5. George E, Noel F, Bois G, et al. Antiretroviral therapy for HIV-1-infected children in Haiti. J Infect Dis. 2007;195:1411–1418