A 1-YEAR EVALUATION OF SYVA MICROTRAK CHLAMYDIA ENZYME-IMMUNOASSAY WITH SELECTIVE CONFIRMATION BY DIRECT FLUORESCENT-ANTIBODY ASSAY IN A HIGH-VOLUME LABORATORY

The Syva MicroTrak Chlamydia enzyme immunoassay (EIA; Syva Company, San Jose, Calif.) with cytospin and direct fluorescent-antibody assay (DFA) confirmation was evaluated on 43,630 urogenital specimens over a 1-year period in the Provincial Laboratory in Regina, Saskatchewan, Canada. This was a two-phase study intended to define a testing algorithm for Chlamydia trachomatis that would be both highly accurate and cost-effective in our high-volume (>3,000 tests per month) laboratory. The prevalence of C. trachomatis infection in our population is moderate (8 to 9%). In phase 1, we tested 6,022 male and female urogenital specimens by EIA. All specimens with optical densities above the cutoff value and those within 30% below the cutoff value were retested by DFA. This was 648 specimens (10.8% of the total). A total of 100% (211 of 211) of the specimens with optical densities equal to or greater than 1.00 absorbance unit (AU) above the cutoff value, 98.2% (175 of 178) of the specimens with optical densities of between 0.500 and 0.999 AU above the cutoff value, and 83% (167 of 201) of the specimens with optical densities within 0.499 AU above the cutoff value were confirmed to be positive. A total of 12% (7 of 58) of the specimens with optical densities within 30% below the cutoff value, were positive by DFA. In phase 2, we tested 37,608 specimens (32,495 from females; 5,113 from males) by EIA. Only those specimens with optical densities of between 0.499 AU above and 30% below the cutoff value required confirmation on the basis of data from phase 1 of the study. This was 4.5% of all specimens tested. This decrease in the proportion of specimens requiring confirmation provides a significant cost savings to the laboratory. The testing algorithm gives us a 1-day turnaround time to the final confirmed test results. The MicroTrak EIA performed very well in both phases of the study, with a sensitivity, specificity, positive predictive value, and negative predictive value of 96.1, 99.1, 90.3, and 99.7%, respectively, in phase 2. We suggest that for laboratories that use EIA for Chlamydia testing, a study such as this one will identify an appropriate optical density range for confirmatory testing for samples from that particular population.