HIV-infected children in the Asia-Pacific region with baseline severe anemia: antiretroviral therapy and outcomes

The TREAT Asia Pediatric HIV Observational Database (TApHOD) is a longitudinal, multicenter, cohort study of HIV-infected children in Asia that has been described elsewhere [7]. For this analysis, we included children from 18 clinics in 6 countries: Thailand (n = 5), Malaysia (n = 4), Cambodia (n = 3), Vietnam (n = 3), Indonesia (n = 2), and India (n=1); these referral clinics routinely provide pediatric HIV clinical care and treatment. The database includes demographic characteristics, laboratory testing, and treatment information, which are collected from medical and clinical records and entered into computerized databases by trained staff. Data are transferred to the Kirby Institute twice yearly for cleaning and analysis.

For this report, we included children aged <18 years who initiated highly active ART (HAART; defined as a combination of ≥3 antiretrovirals) between

January 2003 and September 2013, and who had available baseline pre-ART hemoglobin (Hb) data. We defined severe anemia in accordance with the United States National Institutes of Health Division of AIDS (DAIDS) 2004 toxicity grading table: Hb <7.5 g/ dL for children ≥57 days [8]. Baseline values for Hb measurements were defined as the values closest to a window of 6 months before and 7 days after HAART initiation. Children were considered lost to follow-up if the time between their last visit and the date of the last data transfer was ≥12 months without documentation of transfer or death.

The baseline data included: age (<1, 1-5, 6-10, and ≥11 years), sex, height- and weight-for-age z-scores (within ± 3months of baseline), CD4 cell percentage (within ± 3 months of baseline), HIV viral load (within 3 months before baseline up to 2 weeks after baseline), WHO stage, initial ART regimen, year of ART initiation and country income status as defined by The World Bank [9].

Baseline categorical data are presented as frequencies (%) and continuous data as medians and interquartile ranges (IQR). Height-for-age z-scores for children <61 months and for children ≥61 months old were calculated using the WHO 2006 child growth standards and macros (ages 6 months to 5 years) [10] and the WHO 2007 child growth standards and macros (age 5-19 years) [11], respectively. Weight-for-age z-scores were calculated using the WHO child growth standards and macros for 1977 to allow for inclusion of all age groups [12].

Recovery from severe anemia, CD4%, AZT-based regimen modification (change to a regimen not containing AZT), death, and loss to follow-up were compared between those initiated on AZT- and non-AZT-containing regimens. All longitudinal analyses were performed on an intention-to-treat basis. Median CD4 percentage and the proportions “severe anemia-free” were evaluated in 6 ± 3-monthly intervals up to 3 years of follow-up (i.e., 3-9, 9-15, 15-21, 21-27, 27-33, and 33-39 months). If, for any given time interval, multiple values were recorded for a patient, the value closest to the 6-monthly time point was used. Kaplan–Meier method was used to evaluate the probability of AZT modification, death, and loss to follow-up over time. Time-to-event was left-censored. Right censoring occurred at the date of the last clinic visit. Stata version 12.1 was used for all statistical analyses.

This study was approved by the Institutional Review Boards at all of the data-contributing sites, the data management and analysis center (Kirby Institute, University of New South Wales, Sydney, Australia), and the coordinating center (TREAT Asia/amfAR, Bangkok, Thailand). The requirement for informed consent was waived because only anonymized patient data collected during routine patient care were used for analysis.

The median Hb 6 months after ART initiation was 10.7 g/dL (IQR 9.6–11.7, range 4.4–13.5) in the 81 of 130 children (62.3%) with available data; severe anemia persisted in 9.9%. At 36 months, 48 children had Hb tested and severe anemia persisted in 4.2% (median 11.9, IQR 10.7–12.6, range 6.6–14.0 g/dL). Of 130 children, 116 (89.2%) had a baseline CD4%, with a median of 5.9% (IQR1.0–13.9). At 6 months after ART initiation, the median CD4% (n = 91) was 16.0% (IQR 10.0–26.9), and at 36 months, median CD4% (n = 68) was 29.6% (IQR 25.9–36.0). By 36 months, growth improved with the weight-for-age z-score (n = 65) of < –2.5 decreased from 82.5% to 30.8% and height-for-age z-score (n = 65) of < –2.5 decreased from 71.0% to 36.9%.

By September 2013, 3 of the 18 children with severe anemia who started ART with an AZT-based regimen were switched off AZT at a rate of 6.3 (95% CI 2.0–19.6) switches per 100 patient-years. Two of these 3 children were switched at 1.6 months, and the other at 11.3 months (Table 2). By comparison, 60 of the 111 severely anemic children who started a d4T-based regimen were switched off d4T at a rate of 17.0 (95% CI 13.2-21.9) switches per 100 patient-years with a median time to d4T switch of 2.5 (IQR 1.5-4.2) years.

Twenty deaths occurred during the follow-up period at a rate of 3.5 (95% CI 2.3-5.5) per 100 patient-years. Those starting AZT-based first-line ART had a mortality rate of 10.8 (95% CI 4.8-23.9) deaths per 100 patient-years compared to 2.7 (95% CI 1.6-4.6) deaths per 100 patient-years among those who started d4T-based first-line ART. Loss to follow-up occurred in 15 children at a rate of 2.6 (95%CI 1.6-4.4) per 100 patient-years.