Associação entre a prevalência da infecção e o risco para o adoecimento por Leishmania (Leishmania) infantum em áreas endêmicas distintas de Belo Horizonte, Minas Gerais, Brasil

Abstract

INTRODUCTION: Understanding the characteristics of infection and the importance of asymptomatic individuals in the epidemiology of visceral leishmaniasis in Brazil is essential for the identification and monitoring of transmission areas. This study evaluated the association between asymptomatic infection and the risk of illness in three endemic areas in Brazil. METHODS: An ecological study was conducted using a Bayesian approach to estimate the relative risk on the basis of the frequency of visceral leishmaniasis cases in the.Belo Horizonte city among 2008 and 2011. The results of this study allowed the identification of three coverage areas with low, medium, and high risk of illness, which were evaluated using a cross-sectional study to estimate the prevalence of asymptomatic infection due to L. infantum. The infection was identified using recombinant antigen K39 and soluble antigen enzyme-linked immunosorbent assays (ELISA-rK39 and ELISA-SLA). The prevalence rates of asymptomatic infection were estimated in different areas and were compared with the RR using the chi-square test (5%). A multivariate logistic regression and a multilevel logistic regression were used to calculate the association between asymptomatic infection and the variables analyzed. RESULTS: The map generated in the ecological study showed that the relative risk of illness due to visceral leishmaniasis was distributed heterogeneously in the municipality. From 147 coverage areas of Belo Horizonte, 76 (51.7%) coverage areas showed higher relative risk (1.01 to 3.03), 46 (31.3%) coverage areas had relative risk median (0.73 - 1.00), and 25 (17.0%) were considered lower risk (0,41- 0,72). A total of 935 children living in the three assessed areas participated in the cross-sectional study. The prevalence rates determined by area were 34.9% (95% confidence interval [CI], 29.740.4) in area 1, 29.3% (95% CI, 24.534.5) in area 2, and 33.6% (95% CI, 28.538.9) in area 3. However, no significant differences were observed in the prevalence rates of infection between these areas (p=0.29). After performing a multivariate logistic regression analysis with adjustment for age, the final model (p <0.05) for collective data the following variables remained: own car (Odds relative [OR] 0.69) and the presence of backyard in nearby residences (OR 1.18). The variables presence of yards in the neighborhood OR 1.14; 95% CI, 1.011.29; p=0.037) and younger age among children (OR 0.99; CI 95%, 0.980.99; p=0.008) remained in the final multi-level model. CONCLUSIONS: The results showed that the fact of having backyards in the neighborhood yard, increases by 14% (multilevel model) and 18% (multivariate logistic model) the chances of having an infection on that residence. The infection rates and the homogeneous distribution of asymptomatic individuals in the assessed areas indicate that infection due to L. infantum in areas with different degrees of risk to the clinical form of visceral leishmaniasis affects a significant percentage of children in the city of Belo Horizonte, regardless of the risk of illness.