RESUMO
Different algorithms have been proposed to increase the diagnostic capacity of syphilis. We analyzed three common algorithms for detecting suspected syphilis cases in low prevelance Turkish population. The study included a total of 340 clinical serum samples from adults throughout Turkey, who had syphilis as a clinical preliminary diagnosis and were positive on at least one of the following tests: Rapid Plasma Reagin (RPR), Treponema pallidum Haemagglutination test (TPHA) and FTA-abs Ig. In adittion to percent agreement, kappa coefficients were calculated to compare the conformity between the three algorithms. Both the reverse and the ECDC algorithms had higher diagnostic efficacy than the conventional algorithm. The sensitivity/specificity/ accuracy of conventional, reverse and ECDC algorithms were 51.3%/86.1%/55%; 80.9%/86.1%/81.4% and 80.9%/100%/82.9% respectively. The interrater reliability was moderate for conventional-reverse algorithm (73.53%; к=0.484; 95%CI=0.41-0.56; p=0.001) and conventional-ECDC algorithm (72.06%; к=0.454; 95% CI= 0.37-0.54; p=0.001), and near perfect for reverse-ECDC algorithm (98.53%; к=0.963; 95% CI=0.93-0.99; p=0.0001). Our data support the use of ECDC algorithm in serological diagnosis of syphilis. It may increase the diagnostic capacity if treponemal tests are used for screening, and then positive results are confirmed with a different and second treponemal test.