Leptospirosis has emerged as an important urban health problem as slum settlements have expanded worldwide. Yet the dynamics of the environmentally transmitted Leptospira pathogen has not been well characterized in these settings. We used a stratified dense sampling scheme to study the dynamics of Leptospira abundance in surface waters from a Brazilian urban slum community. We collected surface water samples during the dry, intermediate and rainy seasons within a seven-month period and quantified pathogenic Leptospira by quantitative PCR (qPCR). We used logistic and linear mixed models to identify factors that explained variation for the presence and concentration of Leptospira DNA. Among 335 sewage and 250 standing water samples, Leptospira DNA were detected in 36% and 34%, respectively. Among the 236 samples with positive results geometric mean Leptospira concentrations were 152 GEq/mL. The probability of finding Leptospira DNA was higher in sewage samples collected during the rainy season when increased leptospirosis incidence occurred, than during the dry season (47.2% vs 12.5%, respectively, p = 0.0002). There was a marked spatial and temporal heterogeneity in Leptospira DNA distribution, for which type of water, elevation, and time of day that samples were collected, in addition to season, were significant predictors. Together, these findings indicate that Leptospira are ubiquitous in the slum environment and that the water-related risk to which inhabitants are exposed is low. Seasonal increases in Leptospira presence may explain the timing of leptospirosis outbreaks. Effective prevention will need to consider the spatial and temporal dynamics of pathogenic Leptospira in surface waters to reduce the burden of the disease.

Leptospirosis, a zoonosis caused by pathogenic Leptospira, primarily affects tropical, developing regions, especially communities without adequate sanitation. Outbreaks of leptospirosis have been linked with the presence of pathogenic Leptospira in water. In this study, we measured the physicochemical characteristics (temperature, pH, salinity, turbidity, electrical conductivity, and total dissolved solids (TDS)) of surface waters from an urban slum in Salvador, Brazil, and analyzed their associations with the presence and concentration of pathogenic Leptospira reported previously. We built logistic and linear regression models to determine the strength of association between physicochemical parameters and the presence and concentration of Leptospira. We found that salinity, TDS, pH, and type of water were strongly associated with the presence of Leptospira. In contrast, only pH was associated with the concentration of the pathogen in water. The study of physico-chemical markers can contribute to a better understanding of the occurrence of Leptospira in water and to the identification of sources of risk in urban slum environments.