RESUMEN
BACKGROUND: Up to 15% of deaths of people living with HIV is attributable to meningeal cryptococcosis, with nearly 75% occuring in sub-Saharan Africa. Although rare in children, it is a major cause of morbidity and mortality in people living with HIV. A strong association between cryptococcal antigenemia and the development of meningeal cryptococcosis has been shown in adults. Thus, in 2018, the World Health Organization published an updated version of its guidelines for the diagnosis, prevention and management of cryptococcal infection in adults, adolescents and the HIV-infected child. GOAL: To determine the prevalence of cryptococcal antigenemia and to identify its determinants in children infected with HIV. METHODS: An analytical cross-sectional study was carried out at the approved treatment center of Laquintinie hospital in Douala over a period of 4 months. Children were recruited consecutively after informed parental consent. Cryptococcal antigenemia and CD4 assay were performed using a Cryptops® immunochromatographic rapid diagnostic test and flow cytometry, respectively. The data collected included the socio-demographic, clinical and paraclinical variables of the children, as well as their antecedents. Data analysis was performed using Epiinfo software version 3.1 and SPSS 21.0. The significance threshold was set at 5%. RESULTS: A total of 147 children were enrolled. The mean age was 9.8 ± 4.09 years. The majority were on antiretroviral therapy (142, 96.60%). Only 13 (8.80%) were in severe immunosuppression. No child showed signs of meningeal cryptococcosis. The prevalence of cryptococcal antigenemia was 6.12%. Severe immunosuppression [OR: 10.03 (1.52-65.91), p = 0.016] and contact with pigeons [OR: 9.76 (1.14-83.65), p = 0.037] were independent factors significantly associated with the carriage of the cryptococcal antigen. CONCLUSION: We recommend screening for cryptococcal antigenemia and routine treatment with fluconazole of all HIV positive children with cryptococcal antigen whether symptomatic or not.