A systematic review and meta-analysis of the factors associated with Leishmania infantum infection in dogs in Brazil

The risk factors associated with canine visceral leishmaniosis (CVL) in Brazil are unclear and controversial. The objectives of this systematic review and meta-analysis were to identify the best evidence available in this field and to determine the gaps in existing knowledge. Literature searches were carried out using four databases, the reference lists within articles, and references provided by experts in the field. Theoretical discussions or separate and independent meta-analyses of p-values or of effect sizes were used to pool information about each variable. Thirty-six articles were selected for detailed review, including 31 cross-sectional, two ecological and three cohort studies. The variables showing significant association with CVL were short hair, purebred, pen-domestic restricted (as compared with domestic-restricted dogs), and presence of green areas adjacent to home. The occurrence of CVL was also associated with the presence of domestic fowl in the home environment, with free dogs (as compared with restrained dogs), with male gender and with dogs >1 or 2 years of age, although these associations were not statistically significant. Due to the small number of publications, consistent results could not be obtained concerning the role of other factors. Most studies did not describe the criteria of eligibility and the process of selection of participants in sufficient detail and employed only one diagnostic test as proof of infection. Few studies controlled for confounding variables. No statistical evidence of publication bias was detected, but a great deal of information contained in the primary articles was lost because the results were not adequately described. The results of this review contribute to a better understanding of CVL and should assist in optimizing the development and implementation of control policies. Continuous actions, prioritizing dogs at higher risk and areas with higher abundance of green vegetation, together with policies to promote responsible dog ownership are mandatory. Problems concerning study design and data analysis described in the present study need to be taken into consideration in future studies. These must follow clear procedures to select participants and utilize standardized, valid and reliable diagnostic methods. The development of multivariate models and the use of the STROBE statement for description of the results should also be encouraged. Further research should investigate the patterns identified and prioritize CVL-related factors that have not been fully recognized or elucidated. Finally, ecological and cohort studies of CVL and investigations in other countries of Latin America are urgently required. (C) 2013 Elsevier B.V. All rights reserved.