Since the introduction of triple-drug combination antiretroviral therapy (cART), survival of HIV-infected children in resource-rich countries has substantially improved.1-4 The mortality of perinatally HIV-infected children in the United Kingdom and Ireland has declined from 9.3 deaths per 100 child-years at risk before 1997 to 2.0 deaths per 100 child-years in 2001-20021 and to 0.6 per 100 child-years during 2003-2006.2 An analysis based on data collected as part of the PACTG 219 and 219C cohort studies, which enrolled HIV-exposed and HIV-infected children in the United States, demonstrated a decline in mortality from 7.1 per 100 child-years in 1994 to 0.6 per 100 child-years in 2006.3 In many resource-limited countries in Asia, ART use for HIV-infected children began around 2002 but was not scaled up consistently across the region.5 Before that time, drug access was largely limited to monotherapy or dual-therapy with nucleoside reverse transcriptase inhibitors (NRTI). Although, responses to cART have been reported to be comparable with Western cohorts,6-13 the impact of frequent monotherapy or dual-therapy use in the past and survival outcomes of perinatally HIV-infected children across the region has not been reported.
In 2005, TREAT Asia's Pediatric Program was initiated by amfAR, The Foundation for AIDS Research, to provide the first platform from which pediatric HIV clinical providers and researchers in Asia could conduct regional-level observational research. In 2007, the group established the TREAT Asia Pediatric HIV Observational Database (TApHOD), a project designed to provide insights into improving treatment access and outcomes for HIV-positive children in Asia.14,15 In this analysis we describe survival and factors related with mortality in HIV-infected children managed in this regional network.
PATIENTS AND METHODS
TApHOD is a multicenter observational cohort study of infants and children living with HIV in Asia. TApHOD is a member cohort of the US National Institutes of Health's International Epidemiologic Databases to Evaluate AIDS. Data collection for TApHOD began in 2008, the methods of which have been described elsewhere.15 Participating sites are largely public referral hospitals based in urban settings. All eligible HIV-infected children receiving care at participating clinical sites are enrolled into the database. Inclusion criteria include (1) age at the first clinic visit of ≤18 years; and (2) confirmed diagnosis of HIV using age-appropriate tests; or (3) presumptive diagnosis according to World Health Organization criteria until confirmation.16
Institutional Review Board approval was obtained at participating sites and the data management and co-ordinating centers (University of New South Wales, Sydney; TREAT Asia, Bangkok). Because data are anonymously collected, informed consent was waived, unless locally required.
Data are extracted from the existing clinic databases or medical records, and put into a standardized Microsoft Access database or Excel file before being transferred electronically to the data management center for quality control and aggregation. Core variables include sociodemographic data, clinical history, dates and outcomes of clinic visits and hospitalizations, laboratory monitoring, ART history, and vital status. The cohort was prospectively constituted in 2008; retrospective clinical data were provided from the date of first entry into the clinic. Sites also provided data for those who died and were lost to follow-up (LTFU) during the retrospective period where available, thus allowing for complete site-level data ascertainment.
Data included in this analysis were transferred to TApHOD through March 2009, from children at 12 sites in 5 countries, including 5 sites in Thailand, 4 in Malaysia, and 1 each in India, Indonesia, and Cambodia. For the purpose of this analysis, children were excluded if they enrolled in the clinic during a time period for which complete treatment data could not be ascertained for those who died or were LTFU at their site.
Continuous variables were presented as means with standard deviations (SD) or medians with interquartile ranges (IQR), as appropriate. The incidence rates of mortality, the primary endpoint of interest, were calculated by dividing the number of deaths by the total number of person-years. For each member of the cohort, person-years at risk were measured from the start date of ART or the date of first clinic visit for children who were ART naive until the date of death or the date of the most recent clinic visit. LTFU was defined as having had no documented clinic visit for >12 months. We used an “intent-to-continue-treatment” approach and thus did not take into account changes to treatment, including treatment interruptions and termination of treatment after ART initiation. Kaplan-Meier methods were used to assess the cumulative probability of survival after the start of ART.
Univariate and multivariate Cox proportional hazard models were used to assess factors associated with mortality in children who received ART. Baseline covariates included in the univariate analysis were: ethnicity, weight-for-age and height-for-age Z scores, body mass index, WHO clinical staging (i.e., the most advanced stage documented prior to ART), hemoglobin level, and initial ART regimens. For height-for-age Z score, the WHO 2006/2007 Child Growth Standards17 were used. US CDC 2000 growth curves were used for weight-for-age Z scores, to allow for scoring children >10 years of age.18 CD4 cell percentage was treated as a time-dependent covariate (i.e., the individual values were continually updated throughout the follow-up period). Laboratory values from within 3 months before (all tests) and 1 month after ART initiation (except for CD4) were used as baseline values. ART was classified as monotherapy or dual-therapy, or cART. Due to concerns about the inconsistent HIV viral load data collection across the cohort, it was not included in the model. We considered demographic and clinical characteristics that met the criterion of p value <0.10 in univariate analysis to be potential risk factors in multivariate analysis. Multivariate models were then developed using a stepwise forward selection procedure. The significance of each variable was assessed with the likelihood ratio test. Statistical analyses were performed using Stata version 10 (StataCorp, College Station, TX).
Characteristics of the Study Population
Of the 2280 children included in this analysis, 1752 (77%) had been initiated on some level of ART, including cART in 1480 (84%) and mono- or dual-NRTI therapy in 272 (16%); 528 children were ART-naive (Table 1). Most children initiated on cART (97%) presented for care from 2002 onwards, whereas 70% of children initiated with monotherapy or dual-therapy presented for care before this date. The median age at the time of ART initiation was 6.5 years; 7.0 years for those initiated on cART and 2.1 years for those initiated on monotherapy or dual-therapy. In more than half of the children on cART and a fourth of those on monotherapy or dual-therapy, the baseline CD4 percentage was under 15%. The most commonly used first-line cART regimen was nonnucleoside reverse transcriptase (NNRTI)-based (93%), whereas 6% received protease inhibitor-based regimens. Among the NNRTI-based regimens, 67% were nevirapine based (NRTI backbone: 65% with stavudine, 35% with zidovudine), and 33% were efavirenz based (NRTI backbone: 56% with zidovudine, 42% with stavudine).
Duration of Follow-Up and Mortality Rate
By the end of March 2009, the median follow-up time for the 1752 children who had ever received ART was 3.1 (IQR: 1.5-4.8) years; 2.9 (IQR: 1.4-4.6) years for those who initiated cART; and 5.1 (IQR: 2.7-8.7) years for those who initiated monotherapy or dual-therapy. A total of 155 (8.8%) children were LTFU and 119 (6.8%) were transferred to other hospitals, giving a combined lost-to-program rate of 4.6 per 100 child-years [95% confidence interval (CI): 4.1 to 5.2]. Among these 1752 children, 115 (6.6%) died, giving a crude mortality rate of 1.9 per 100 child-years (95% CI: 1.6 to 2.4); 37% of deaths occurred during the first 3 months of ART. The median age at death for children who had ever received ART was 7.7 (IQR: 4.7-10.9) years; 7.8 (IQR: 4.8-10.9) years for those who initiated cART; and 5.8 (IQR: 2.7-7.9) years for those who initiated monotherapy or dual-therapy. The mortality rate decreased by half from 10.2 per 100 child-years (95% CI: 7.5 to 13.7) during the first 3 months of ART to 4.2 per 100 child-years (95% CI: 2.6 to 6.8) during the fourth to sixth months (Table 2). The rate fell further to 2.0 per 100 child-years (95% CI: 1.2 to 3.3) from 6 to 12 months and was 0.9 per 100 child-years (95% CI: 0.7 to 1.3) after 12 months. The Kaplan-Meier estimate of 5-year survival from the time of ART initiation was 91.7% (95% CI: 90.0% to 93.2%). Survival curves by baseline CD4 category are shown in Figure 1. Half of deaths in children with a baseline CD4 <5% occurred within 6 months after ART initiation and all of the reported deaths occurred within the first 2 years. The 2 most common causes of death reported were serious bacterial infections or pneumonia and opportunistic infections (Table 3). The baseline characteristics of children who died in each period are shown in Table 4. The median weight-for-age Z score in children who died during the first 3 months of ART was −5.1 (−6.9 to −3.4), which was much lower than those who died in other periods, but significant differences were not found between the 4 groups.
The 528 (23%) children who never received ART (Table 1) had a median follow-up time of 0.9 years (IQR: 0.1-2.5); 49 of them attended the clinic only once. Thirty-six children (6.8%) died after presenting to care, giving a crude mortality rate of 4.1 per 100 child-years (95% CI: 3.0 to 5.7). Their causes of death are described in Table 3. In the ART-naive group, 241 (46%) children were LTFU, and 34 (6%) were transferred to other clinical care programs, resulting in a combined lost-to-program rate of 31.5 per 100 child-years (95% CI: 28.0 to 35.5).
Predictors of Death After ART Initiation
The risk of death showed a significant decreasing trend as CD4 percentage increased (Table 5). The hazard ratio dropped from 33.85 (95% CI: 14.96 to 76.59) in children with a CD4 <5% to 4.02 (95% CI: 1.53 to 10.57) in children with a CD4 of 15%-19%, when compared with the reference category (CD4 ≥20%). The risk of death decreased by 10% for each unit increase in weight-for-age Z score (95% CI: 0.81 to 0.98). Mortality risk was significantly higher for those with WHO clinical stage 4 at baseline (HR: 4.78, 95% CI: 1.84 to 12.41).
Current ART Outcomes
By the end of March 2009, 1580 of 2280 children (69%) were still on active follow-up at the participating sites. A total of 1363 (86%) were on ART; of whom, 1041 (76%) were on NNRTI-based regimens and 7 (0.5%) were on mono- or dual-NRTI. The median age was 10.7 years (IQR: 7.4-13.5, range 8 months to 25.7 years). There were 217 (13%) children who had not yet started ART with a median age of 8.1 years (IQR: 5.7-10.7; range 1-21.7 years). The most recent CD4 was ≥15% in 85% of those on ART and 83% of ART-naive children and adolescents.
We found a mortality of 6.6% with 8.8% LTFU, giving an overall mortality rate of 1.9 per 100 child-years among Asian children who received any level of ART in this regional cohort. The 5-year survival rate from the time of ART initiation was 91.7% (95% CI: 90.0% to 93.2%). Mortality was highest in the first 3 months after ART initiation and fell with time on ART and was frequently caused by bacterial infections, Pneumocystis carinii pneumonia, and tuberculosis.
In comparison to our results, the KIDS-ART-LINC collaboration reported 6.9% mortality after ART initiation and 10.3% LTFU after 2 years in West Africa.19 Reduction in mortality with the availability of ART has been shown in several studies from developed countries1-4,20-22 and from resource-limited settings.19,23,24 Before ART became available to children in this region, survival of HIV-infected children with advanced stage disease was low. In a study of 353 symptomatic HIV-infected children cared for at Chiang Mai University Hospital, Thailand between 1989 and 1994, 42% of them died during their first hospital admission.25 In a study at a tertiary-care hospital in northeastern Thailand between 1989 and 1997, among 90 HIV-infected children with AIDS, 50% of them died with 30% LTFU. The 1-year survival rate from the time of first symptom was 75.3% (95% CI: 65.8% to 84.7%).26
In our study, the predictors of mortality in ART-experienced children included WHO clinical stage 4, low CD4% (as a time-dependent variable), and low weight-for-age Z score. Children who died in the first 3 months of ART had very low baseline weight-for-age Z score (ie: median −5.1), although not significantly lower than those who died during other periods. Malnutrition can play a critical role in HIV infection, as it impairs both humoral and cell-mediated immunity, although infection itself also contributes to malnutrition.27,28 The type of initial ART regimen (i.e., cART vs. mono- or dual-NRTI) was not associated with increased mortality. This may be related to subsequently switching from mono- or dual-NRTI regimens to cART during the follow-up period; as only 0.5% children who were previously initiated on mono- or dual-NRTIs remained on those regimens. Unlike studies from sub-Saharan Africa, we did not find baseline severe anemia as an independent risk factor for mortality.19
Excess early mortality in the first 3 months on ART has been reported both in children and adults.19,24,29,30 In a prospective cohort study of 135 HIV-infected Kenyan children initiating cART between August 2004 and December 2008 and followed for a median of 21 months, 20 (13%) children died; the overall mortality was 8.4 per 100 child-years.29 However, the mortality rate was 46 per 100 child-years in the first 4 months and dropped to 1.0 per 100 child-years from 4 to 24 months.29 A Medecins sans Frontières study including 3936 children <5 years of age in Africa and Asia reported 6.3% mortality at a median follow-up time of 10.5 months. The survival rate at 1 year was 93% (95% CI: 92% to 94%), and 55% of deaths and LTFU occurred during the first 3 months of ART.30
This study has several limitations. The data were both retrospectively and prospectively collected, therefore, specific causes of death and outcomes of children LTFU were not fully ascertainable. In addition, although there were at least 3 patients with suspected immune reconstitution syndrome, we were unable to assess the impact of immune reconstitution on mortality. The lack of data on adherence and viral load are other factors that limited our analysis. The children in our cohort were seen at regional pediatric referral centers, so results may not be generalizable to children receiving care in primary care centers or rural areas. Their acuity was high, as they frequently presented late to care and may have been transferred from centers less equipped to manage complicated opportunistic infections. However, some of these factors are challenging to control in the analysis of observational databases from a resource-limited setting. As the largest analysis of survival and mortality in HIV-infected children across Asia, these findings contribute important evidence on which to consider future management strategies.
Survival of HIV-infected children in Asia has improved substantially as ART has become more widely available in the region. The findings that advanced clinical and immunologic disease predicted mortality emphasize the urgent need for early identification of HIV-infected infants and children and access to cART before disease progression. The frequent association of low weight-for-age with poor outcomes in pediatric cohorts further highlights the importance of integrating nutritional interventions into ART programs to reduce early mortality.
The authors also wish to thank all the children and the staff in the participating centers who have given their time so generously during the course of this project.
1. Gibb DM, Duong T, Tookey PA, et al. Decline in mortality
, AIDS, and hospital admissions in perinatally HIV
-1 infected children in the United Kingdom and Ireland. BMJ
2. Judd A, Doerholt K, Tookey PA, et al. Morbidity, mortality
and response to treatment in children in the United Kingdom and Ireland with perinatally-acquired HIV
infection during 1996-2006: planning for teenage and adult care. Clin Infect Dis
3. 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
4. McConnell MS, Byers RH, Frederick T, et al. Trends in antiretroviral therapy
use and survival rates in HIV
-infected children and adolescents in the United States, 1989-2001. J Acquir Immune Defic Syndr
5. 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
6. Chearskul P, Chokephaibulkit K, Chearskul S, et al. Effect of antiretroviral therapy
in human immunodeficiency virus-infected children. J Med Assoc Thai
7. Puthanakit T, Oberdorfer A, Akarathum N, et al. Efficacy of highly active antiretroviral therapy
-infected children participating in Thailand's National Access to Antiretroviral Program. Clin Infect Dis
8. Puthanakit T, Aurpibul L, Oberdorfer P, et al. Hospitalization and mortality
-infected children after receiving highly active antiretroviral therapy
. Clin Infect Dis
9. Puthanakit T, Aurpibul L, Oberdorfer P, et al. Sustained immunologic and virologic efficacy after four years of highly active antiretroviral therapy
in human immunodeficiency virus infected children in Thailand. Pediatr Infect Dis J
10. Little KE, Bland RM, Newell ML. Vertically acquired paediatric HIV
infection: the challenges of providing comprehensive packages of care in resource-limited settings. Trop Med Int Health
11. Over M, Revenga A, Masaki E, et al. The economics of effective AIDS treatment in Thailand. AIDS
12. Kumarasamy N, Venkatesh KK, Devaleenol B, et al. Safety, tolerability and effectiveness of generic ART in HIV
-infected children in South India. J Trop Pediatr
13. Ciaranello AL, Chang Y, Margulis AV, et al. Effectiveness of pediatric antiretroviral therapy
in resource-limited settings: a systematic review and meta-analysis. Clin Infect Dis
14. Prasitsuebsai W, Bowen AC, Pang J, et al. TREAT Asia
Pediatric Program. Pediatric HIV
clinical care resources and management practices in Asia
: a regional survey of the TREAT Asia
pediatric network. AIDS Patients Care STDS
15. Kariminia A, Chokephaibulkit K, Pang J, et al. Cohort profile: The TREAT Asia
observational database. Int J Epidemiol
. 2010;1-10 doi:10.1093/ije/dyp358.
16. World Health Organization. WHO Case Definitions of HIV for Surveillance and Revised Clinical Staging and Immunological Classification of HIV-Related Disease in Adults and Children
. Geneva, Switzerland: World Health Organization; 2006.
17. WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards: Length/Height-for-Age, Weight-for-Age, Weight-for-Length, Weight-for-Height and Body Mass Index for-Age: Methods and Development
. Geneva, Switzerland: World Health Organization; 2006.
18. Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11
19. The KIDS-ART-LINC Collaboration. Low risk of death, but substantial program attrition, in pediatric HIV
treatment cohorts in Sub-Saharan Africa. J Acquir Immune Defic Syndr
20. de Martino M, Tovo PA, Balducci M, et al. Reduction in mortality
with availability of antiretroviral therapy
for children with perinatal HIV
-1 infection. Italian Register for HIV
Infection in Children and the Italian National AIDS Registry. JAMA
21. Sanchez JM, Ramos Amador JT, Fernandez de Miguel S, et al. Impact of highly active antiretroviral therapy
on the morbidity and mortality
in Spanish human immunodeficiency virus-infected children. Pediatr Infect Dis J
22. Viani RM, Araneta MR, Deville JG, et al. Decrease in hospitalization and mortality
rates among children with perinatally acquired HIV
type 1 infection receiving highly active antiretroviral therapy
. Clin Infect Dis
23. Fassinou P, Elenga N, Rouet F, et al. Highly active antiretroviral therapies among HIV
-1-infected children in Abidjan, Cote d'Ivoire. AIDS
24. George E, Noel F, Bois G, et al. Antiretroviral therapy
-1-infected children in Haiti. J Infect Dis
25. Sirisanthana V. Demographic and clinical characteristic of symptomatic vertical HIV
-infected children at Chiang Mai University Hospital. J Infect Dis Antimicrob Agents
26. Lumbiganon P, Kosalaraksa P, Loapaiboon M. Survival of children with AIDS: experience in a university hospital in northeast Thailand. J Med Assoc Thai
27. Katona P, Katona-Apte J. The interaction between nutrition and infection. Clin Infect Dis
28. Ivers LC, Cullen KA, Freedberg KA, et al. HIV
/AIDS, undernutrition, and food insecurity. Clin Infect Dis
29. Wamalwa DC, Obimbo EM, Farquhar C, et al. Predictors of mortality
-1 infected children on antiretroviral therapy
in Kenya: a prospective cohort. BMC Pediatr
. 2010;10:33. E-pub ahead of print.
30. Sauvageot D, Schaefer M, Olson D, et al. Antiretroviral therapy
outcomes in resource-limited settings for HIV
-infected children <5 years of age. Pediatrics
. 2010;125:e1039-e1047. E-pub.
APPENDIX I: THE MEMBERS OF THE TREAT ASIA PEDIATRIC HIV NETWORK
V. Saphonn* and S. Saramony, National Centre for HIV/AIDS Dermatology and STDs, Phnom Penh, Cambodia; U. Vibol* and S. Sophan, National Pediatric Hospital, Phnom Penh, Cambodia; J. Tucker, New Hope for Cambodian Children, Phnom Penh, Cambodia; F. J. Zhang and N Han, Beijing Ditan Hospital, Beijing, China; N. Kumarasamy* and S. Saghayam, Y. R. Gaitonde Centre for AIDS Research and Education, Chennai, India (supported by the Austrian AIDS Life Association); N. Kurniati* and D. Muktiarti, Cipto Mangunkusumo General Hospital, Jakarta, Indonesia; S. M. Fong* and M. Thien, Hospital Likas, Kota Kinabalu, Malaysia; N. K. Nik Yusoff* and L. C. Hai, Hospital Raja Perempuan Zainab II, Kelantan, Malaysia; K. Razali* and N. F. Abdul Rahman, Pediatric Institute, Hospital Kuala Lumpur, Kuala Lumpur, Malaysia; R. Nallusamy* and K. C. Chan, Penang Hospital, Penang, Malaysia; V. Sirisanthana* and L. Aurpibul, Chiang Mai University, Chiang Mai, Thailand; R. Hansudewechakul*† and A. Khongponoi, Chiangrai Prachanukroh Hospital, Chiang Rai, Thailand; P. Lumbiganon* and P. Kosalaraksa., Khon Kaen University, Khon Kaen, Thailand; G. Jourdain, Program for HIV Prevention and Treatment, Chiang Mai, Thailand; J. Ananworanich* ‡ and T. Puthanakit, T. Suwanlerk, The Netherlands, Australia, Thailand Research Collaboration (HIV-NAT), Bangkok, Thailand; K. Chokephaibulkit* and O. Wittawatmongkol, Siriraj Hospital, Mahidol University; H. K. Truong* and DAN Mai, Children's Hospital 1, Ho Chi Minh City, Vietnam; C. V. Do* and M.T. Ha, Children's Hospital 2, Ho Chi Minh City, Vietnam; V. H. Buy *and V. L. Nguyen, National Hospital of Pediatrics, Hanoi, Vietnam; N. O. Le and L. K. Do, Worldwide Orphans Foundation, Ho Chi Minh City, Vietnam; A. H. Sohn*, L. Messerschmidt, and J. Pang, TREAT Asia, amfAR-The Foundation for AIDS Research, Bangkok, Thailand; D. A. Cooper, M. G. Law*, and A. Kariminia, National Centre in HIV Epidemiology and Clinical Research, The University of New South Wales, Sydney, Australia.
*TApHOD Steering Committee member.
†Current Steering Committee Chair.