The Cut-Off Values for Single Serum of Leptospirosis Detection
No Thumbnail Available
Date
2011-02-07
Journal Title
Journal ISSN
Volume Title
Publisher
Bulletin of the Department of Medical Sciences - วารสารกรมวิทยาศาสตร์การแพทย์
Abstract
Microscopic agglutination test (MAT) is the gold standard for serological diagnosis of leptospirosis. The criterion for definite diagnosis of current leptospiral infection is four fold rise in titer of paired sera or seroconversion. Making a diagnosis on results of a single MAT is difficult because of the uncertainty about the cut-off titer. The present study was conducted to determine the cut-off titer for a single MAT in areas of high and low endemicity for leptospirosis. A total of 1,447 serum samples were collected from 1,105 healthy individuals and 171 confirmed patients residing in areas of high ( northeast Thailand) and low endemicity (other regions). All the sera were tested by MAT. The optimal cut-off points for presumptive diagnosis of leptospirosis was selected by Receiver operating characteristics (ROC) analysis. In endemic area , the best cut-off being 1:100 where the sensitivity was (95% CI) 70.7 (64.3 -76.6) per cent and specificity was 95.0 (93.0 – 96.6) percent. LR+/ LR- was 45 .8 indicating reasonable separation between the positive and negative test results. In other regions, the best cut–off being 1:100 where the sensitivity was 68.3(52.2-76.5) percent, specificity was 96.5(94.5- 97.8) per cent , accuracy was 90.9(84.4- 93.0) percent and LR+/ LR- was 59.5. This indicates that the best cut-off titer to be used will be 1: 100 in both high and low endemic areas for presumptive diagnosis of leptospirosis. However, in low endemic areas, the suspected cases of leptospirosis with MAT titer 1:50 should be monitored for further investigation of the optimal cut-points. Key words: Leptospirosis, Microscopic agglutination test, cut-off titer,
Description
Keywords
Citation
Bulletin of the Department of Medical Sciences - วารสารกรมวิทยาศาสตร์การแพทย์; Vol.51 No.2 April - June 2009; 91-103