METHODS: To better understand the burden of disease and associated risk factors, we evaluated 147 patients presenting with clinical leptospirosis to local hospitals in Sarawak, Malaysia for the presence of Leptospira and associated antibodies. Sera and urine specimens collected during the acute illness phase were assessed via a commercially available rapid diagnostic test (Leptorapide, Linnodee Ltd., Antrim, Northern Ireland), an ELISA IgM assay (Leptospira IgM ELISA, PanBio, Queensland, Australia) and a pan-Leptospira real-time PCR (qPCR) assay to estimate disease prevalence and diagnostic accuracy of each method. Microagglutination testing was performed on a subset of samples.

RESULTS: Overall, 45 out of 147 patients (30.6%) showed evidence of leptospires through qPCR in either one or both sera (20 patients) or urine (33 patients), and an additional ten (6.8%) were considered positive through serological testing, for an overall prevalence of 37.4% within the study population. However, each diagnostic method individually yielded disparate prevalence estimates: rapid test 42.2% for sera and 30.5% for urine, ELISA 15.0% for sera, qPCR 13.8% for sera and 23.4% for urine. Molecular characterization of a subset of positive samples by conventional PCR identified the bacterial species as Leptospira interrogans in 4 specimens. A multivariate risk factor analysis for the outcome of leptospirosis identified having completed primary school (OR = 2.5; 95 CI% 1.0-6.4) and weekly clothes-washing in local rivers (OR = 10.6; 95 CI% 1.4-214.8) with increased likelihood of leptospirosis when compared with those who had not.