Comparison of three commercially available amplification assays, AMP CT, LCx, and COBAS AMPLICOR, for detection of Chlamydia trachomatis in first-void urine.

We compared the Gen-Probe transcription-mediated amplification assay (AMP CT), the Abbott LCx assay, and the Roche COBAS AMPLICOR assay for the detection of Chlamydia trachomatis in a mixed population in urine samples. First-void urine, urethral specimens, and cervical specimens in females were obtained from 1,000 patients (544 males and 456 females) visiting the outpatient sexually transmitted disease clinic of our hospital. The prevalence of C. trachomatis infection was 7.7% as determined by tissue culture of urethral and cervical specimens. The sensitivities of LCx, COBAS AMPLICOR, and AMP CT compared to cell culture were 79, 86, and 78%, respectively. Sensitivity and specificity were recalculated by using a new "gold standard", i.e., a sample was considered to be true positive if two or more techniques yielded positive results. Specimens positive only by cell culture or positive in only one commercial amplification technique were retested by a previously described in-house PCR. After discordance analysis the sensitivities of LCx, COBAS AMPLICOR, and AMP CT were 84, 93, and 85%, respectively. Specificity exceeded 99% for all three assays. With each method the sensitivity was lower for urine samples from females compared to urine samples from males. By application of this new gold standard, existing differences between methods are highlighted; future evaluations of new techniques should be validated against two or more amplification assays.