Development and validation of a real-time PCR for detection of pathogenic leptospira species in clinical materials.

Abstract

Available serological diagnostics do not allow the confirmation of clinically suspected leptospirosis at the early acute phase of illness. Several conventional and real-time PCRs for the early diagnosis of leptospirosis have been described but these have been incompletely evaluated. We developed a SYBR Green-based real-time PCR targeting secY and validated it according to international guidelines. To determine the analytical specificity, DNA from 56 Leptospira strains belonging to pathogenic, non-pathogenic and intermediate Leptospira spp. as well as 46 other micro-organisms was included in this study. All the pathogenic Leptospira gave a positive reaction. We found no cross-reaction with saprophytic Leptospira and other micro-organisms, implying a high analytical specificity. The analytical sensitivity of the PCR was one copy per reaction from cultured homologous strain M 20 and 1.2 and 1.5 copy for heterologous strains 1342 K and Sarmin, respectively. In spiked serum & blood and kidney tissue the sensitivity was 10 and 20 copies for M 20, 15 and 30 copies for 1342 K and 30 and 50 copies for Sarmin. To determine the diagnostic sensitivity (DSe) and specificity (DSp), clinical blood samples from 26 laboratory-confirmed and 107 negative patients suspected of leptospirosis were enrolled as a prospective consecutive cohort. Based on culture as the gold standard, we found a DSe and DSp of 100% and 93%, respectively. All eight PCR positive samples that had a negative culture seroconverted later on, implying a higher actual DSp. When using culture and serology as the gold standard, the DSe was lower (89%) while the DSp was higher (100%). DSe was 100% in samples collected within the first--for treatment important--4 days after onset of the illness. Reproducibility and repeatability of the assay, determined by blind testing kidney samples from 20 confirmed positive and 20 negative rodents both appeared 100%. In conclusion we have described for the first time the development of a robust SYBR Green real-time PCR for the detection of pathogenic Leptospira combined with a detailed assessment of its clinical accuracy, thus providing a method for the early diagnosis of leptospirosis with a well-defined satisfactory performance.