Introduction: Most data on tuberculosis in human immunodeficiency virus (HIV)-infected children in Africa come from hospital-based and cross-sectional studies. Objectives: To estimate the incidence of tuberculosis in HIV-infected children participating in an observational cohort. Methods: HIV-infected children in Abidjan, Côte d'Ivoire, are followed in a prospective cohort. At enrollment, all children had a physical examination, CD4 lymphocyte counts, chest radiograph and a tuberculin test. Quarterly follow-up visits are organized. All patients with suspected tuberculosis undergo specific investigations including gastric aspiration and culture. All isolates are tested for susceptibility. Results: From October 2000 to December 2003, 129 girls and 153 boys were recruited. Of children without a current or previous diagnosis of tuberculosis, 6.5% (13 of 199) had a tuberculin test result of >5 mm, compared with 17.5% of children (10 of 57) with current or previous tuberculosis (P < 0.02). Forty-eight children (17%) had a history of treated tuberculosis, and 27 children were being treated for tuberculosis at enrollment or during the first month of follow-up. Eleven children were diagnosed with tuberculosis after the first month of follow-up, and the diagnosis of mycobacterial infection was confirmed in 7 cases. Of 5 tested isolates of Mycobacterium tuberculosis, 3 were resistant to at least 1 antitubercular drug. Cumulative incidence of tuberculosis was 2060/100,000 at 12 months, 3390/100,000 at 2 years and 5930/100,000 at 3 years. The 3-year risk was 12,400/100,000 in immunocompromised children (CD4 <15%) and 3300/100,000 in other children (P < 0.0001). Conclusion: The risk of tuberculosis among HIV-infected children in Côte d'Ivoire is strongly associated with the degree of immunodeficiency in HIV infection.
Objective: To estimate the frequency of human immunodeficiency virus type 1 (HIV-1) displaying genotypic drug resistance in highly active antiretroviral therapy (HAART)-treated children in Abidjan. Methods: Among the 269 HIV-1-infected children enrolled in the ANRS 1278 prospective observational cohort between October 2000 and September 2003, 115 [median age, 6.35 years (range, 1.2–15)] required treatment and received HAART for at least 6 months. Treatment consisted of 2 nucleoside analogue reverse transcriptase inhibitors associated with nelfinavir (70.5%) or efavirenz (29.5%). Plasma HIV-1 RNA and CD4+ T cell counts were determined at baseline and every 6 months thereafter. Genotypic resistance tests were performed in cases of virologic failure (viral load ≥3 log10 copies/mL) after at least 6 months of HAART. Results: After a median of 10.2 months of HAART, 66% (76 of 115) of children were in virologic success. Most of these children were infected with CRF02 strains. Twenty-seven viruses displayed resistance to at least 1 antiretroviral drug (27 of 38, 71%). Thirteen, 9 and 5 children had viruses with resistance to 1, 2 or 3 of the drugs included in their regimen, respectively. Resistance to lamivudine and/or to non-nucleoside analogue reverse transcriptase inhibitors was frequent among the 38 children in virologic failure. The 90M, 46L, 88S or 54V mutations were found in 11 (38%) of the 29 children taking nelfinavir. The overall frequency of viruses showing genotypic resistance to at least 1 antiretroviral drug was 23% (27 of 115) among the treated children. Conclusion: These results are similar to what is generally observed in industrialized countries. Despite these encouraging results, efforts are needed to maximize the long-term efficiency of treatment and to minimize the risk of emergence of drug resistance in treated children.
Most HIV vaccine trials in the world are conducted with clade B while most circulating viral strains in Africa are non-B subtypes. We determined whether CD8+ T cells from HIV-1 intersubtype CRF02_AG-infected Ivorian individuals were able to recognize clade B epitopes. CD8+ T cell responses of nine HIV-1 intersubtype CRF02_AG-infected Ivorian patients and nine HIV-1 subtype B-infected French patients were studied using pools of HIV-1 clade B peptides (110 well-defined HIV CD8+ T cell epitopes) in an ELISPOT IFN-γ assay. There was no difference in the number of recognized peptide pools between Ivorian and French cohorts (mean of four pools in both cases). Ivorian individuals had generated CD8+ T cell responses cross-reactive against HIV-1 subtype B and some individual peptides had been identified. Furthermore, sequence analysis of nef HIV genes of the Ivorian patients and nef cloning in two patients revealed very few variations between HIV- 1 intersubtype CRF02_AG and subtype B in nef immunodominant regions included in HIV clade B lipopeptide vaccines, currently tested in France.
The aim of this study performed in Abidjan, Côte d'Ivoire, was to describe the distribution of CD4+ T-cell lymphocytes (CD4) in HIV-1-infected (HIV+) pregnant women diagnosed during prenatal voluntary counseling and testing and to assess whether HIV-related immunodeficiency influenced the acceptance of an antiretroviral (ARV) package (zidovudine beginning at 36 weeks of amenorrhea plus intrapartum nevirapine) to prevent mother-to-child transmission. Between April and June 2002, a CD4 count was systematically performed in all HIV+ women (n=221) in 5 antenatal clinics carrying out voluntary counseling and testing. No difference in CD4 count was found in HIV+ women who did not return for their test result (n=50) and those who were informed of their positive serostatus (n=171) (median CD4 count: 389/mm3 vs. 420/mm3; P=0.19). We also found a lack of difference in CD4 count in those who accepted ARV (n=72) and those who did not but knew their HIV status (n=99) (median CD4 count: 405/mm3 vs. 425/mm3; P=0.47). The overall uptake of the intervention (31.9%) appeared to be independent of the maternal immune status.
