Walenda C, Alain K, François R, Louise W, Marie-France A, Philippe M. Morbidity in HIV-1-infected children treated or not treated with highly active antiretroviral therapy (HAART), Abidjan, Côte d'Ivoire, 2000-04. J Trop Pediatr 2008 Dec 9. Epub.
Although no official data exist on the number of HIV-infected children with access to HAART in Côte d'Ivoire, UNAIDS estimates this figure to be around 1,800, or 4% of all HIV-infected children in the country. In Africa, data on morbidity in HIV-infected children over the age of 5 years are scarce and primarily comprised of hospital-based or cross-sectional studies describing symptoms and signs of pediatric infections. Data from developed countries show reduced morbidity among HIV-infected children after initiation of HAART and lower incidence of opportunistic and other infections than in the pre-HAART era.
To describe and compare morbidity in relation to progression of the disease in HIV-1-infected children in Côte d'Ivoire, according to symptoms and the presence or absence of HAART treatment
Outpatient day clinic for HIV-infected children in Abidjan, Côte d'Ivoire, set up for the purposes of the research project
Prospective cohort study
HIV-1 infected children between the ages of 18 months and 15 years old who were enrolled from October 2000 to December 2003 in the ANRS 1244/1278 observational cohort.
Occurrence of pathologic events, including ear, nose, and throat (ENT) diseases, chronic otitis, infectious skin diseases, prurigo, generalized pruritic dermatitis, diarrhea, isolated fever, documented malaria, bronchitis, pneumonia, tuberculosis, oral candidiasis and conjunctivitis; overall morbidity; HIV-related mortality; CD4 count; HIV-1 RNA viral load (VL); initiation and duration of HAART.
HIV-1 infected children were enrolled from October 2000 to December 2003 in the ANRS 1244/1278 cohort. Participants were monitored using standardized procedures in a dedicated outpatient clinic. All children underwent a clinical evaluation at baseline and then at quarterly visits throughout the period of the study. Children were also seen for any occurring diseases. A standardized clinical form was developed for registration of any diseases and health events during clinical examinations. Access to the outpatient clinic as well as related care and treatment was provided free of charge. Follow-up continued until September 2004.
CD4 T cell counts were performed at baseline and every 6 months thereafter. Percentages of CD4+ cells were measured by flow cytometry, and absolute CD4+ cell counts were determined using an automated blood cell counter. All children with a CD4 percentage below 25% were given daily cotrimoxazole prophylaxis. Plasma HIV-1 RNA assays were carried out at baseline and every 6 months thereafter. Beginning in April 2003, HIV-1 RNA VL measurements were determined using real-time RT-PCR. Eligibility criteria for initiation of HAART was either having CDC stage C infection or having a CD4 percentage below 15%.
The study population was divided into three groups. Group 1 included children who were symptomatic and were treated using HAART. This group was divided into two subgroups: before and after initiation of HAART. Group 2 included asymptomatic children who did not need HAART at any point during follow-up. Group 3 included symptomatic children who met the criteria to be treated at inclusion, but who were not treated, mainly because of rapid progression of the disease or difficulties and delays in the initiation of treatment. Duration of follow-up was calculated for each group by summing individual follow-up times.
A total of 273 children were included for analysis, including 148 boys (54%) and 125 girls (46%). Total follow-up time was 586 person-years and average follow-up was 2.1 years (range: 6 days to 4 years).
Group 1: There were 171 children included for analysis including 92 boys (54%) and 79 girls (46%). Mean age at inclusion was 6.25 years (range: 1.3 to 5.6 years, median 4.1 years). Duration of follow-up was 402 person-years: 74 person-years before treatment and 328 person-years under treatment. At the start of the study, mean absolute CD4 count was 504/mm3 and the mean percentage was 12.3%. Sixteen (9.3%) children from this group died during the study. A total of 2,483 pathologic events occurred during follow-up, corresponding to a global incidence of morbidity of 51 events per 100 person-months. When Group 1 was divided into two subgroups corresponding to before and after initiation of treatment, a statistically significant difference was observed in the occurrence of pathologic events: 732 pathologic events occurred prior to initiation of HAART (81.6 events per 100 person-months), and 1,751 events occurred after initiation of treatment (43.8 events per 100 person-months) (P<0.0001). Prior to initiation of HAART, the three most frequently occurring diseases were bronchitis (14.6 events per 100 person-months), diarrhea (10.2), and ENT diseases (10.0). Under HAART treatment, the three most frequently occurring diseases were bronchitis (9.9 events per 100 person-months), ENT diseases (6.4), and diarrhea (4.2). When the incidence of pneumonia before initiation of HAART was compared to that after initiation of HAART in this group, incidence was reduced by more than half (5.5 events per 100 person-months to 2.2 events per 100 person-months) in the treated group; this difference was statistically significant (P<0.0001).
Group 2: There were 62 children included for analysis, including 36 (58%) boys and 26 (42%) girls. Mean age at inclusion was 4.5 years (range: 1.6 to 10.3 years, median 3.8 years). Duration of follow-up was 143.3 person-years. At the start of the study, mean absolute CD4 count was 953/mm3 and the mean percentage was 23.8%. Seven (11.3%) children from this group died during the study. A total of 815 pathologic events occurred during follow-up, corresponding to an incidence of all-cause morbidity of 46.8 events per 100 person-months. The three most frequently occurring diseases were bronchitis (9.5 events per 100 person-months), ENT diseases (7.6), and skin infectious diseases (4.4).
Group 3: There were 40 children included for analysis, including 20 boys and 20 girls. Mean age at inclusion was 5.6 years (range: 1.3 to 15.4 years, median 4.1 years). Duration of follow-up was 40.3 person-years. At the start of the study, mean absolute CD4 count was 557/mm3 and the mean percentage was 10%. Twenty-three (57%) children from this group died during the study. A total of 334 pathologic events occurred during follow-up, corresponding to an incidence of all-cause morbidity of 69.3 events per 100 person-months. The three most frequently occurring diseases were bronchitis (10.6 events per 100 person-months), diarrhea (9.6), and ENT diseases (9.2).
The morbidity of HIV-1-infected children treated with HAART was similar to that of asymptomatic HIV-1-infected children. The incidence of morbid events, however, was significantly lower among asymptomatic children than among untreated, symptomatic children. Overall, morbidity and mortality were highest among untreated symptomatic children.
Fair. Given that this was an observational study and participants were not randomized into treatment arms, there was potential for confounding. Furthermore, children entered at various stages of HIV-1 infection, and complete medical histories were not available. Laboratory capacity to identify infectious pathogens was limited. Authors cite potential observation bias, as children receiving HAART in Group 1 were likely more closely followed than those in other groups. Additionally, the study did not actually show morbidity in untreated children since all children in the study received cotrimoxazole prophylaxis to prevent opportunistic infections. Morbidity in completely untreated children would likely be higher. Despite these limitations, the study was able to detect significant differences in morbidity between groups.
The authors conclude that results from this study provide evidence for the effectiveness of HAART in reducing mortality and morbidity in HIV-1 infected children in Africa. This adds to the body of evidence from both developed and developing countries on the effect of treatment on HIV-infected children, supporting the use of HAART. Given this combined evidence, HAART treatment and medical follow up for HIV-1 infected children in Africa should be continued and expanded to improve HIV-related morbidity and mortality in this population. The authors caution that, given the high rates of mortality in asymptomatic children observed in this study, further research should be conducted to review the criteria for initiation of HAART, with potential consideration towards starting HIV-infected children on HAART earlier to avert subsequent morbidity and mortality.