Comparison of the ligase chain reaction assay with an enzyme immunoassay to detect Chlamydia trachomatis in urine specimens from men

We evaluated the Abbott ligase chain reaction (LCR) assay in comparison with the routinely used method, Syva/Behring enzyme immunoassay (EIA), to detect Chlamydia trachomatis in urine specimens from men. A total of 397 consecutive urine specimens were tested from men who attended four urban sexual health clinics in Auckland, New Zealand, from March to April 1997. The urine specimens were tested by the LCR and the EIA methods, both utilising grey zones. The LCR grey-zone samples were confirmed by the direct fluorescent antibody (DFA) or a repeat LCR on a 1.3 dilution. All EIA results were confirmed by the DFA. The overall prevalence of chlamydia was 8.6 per cent (34/396), based on a positive LCR or confirmed positive EIA. The LCR detected 34 and EIA 22 of the total positives, giving sensitivities of 100 per cent (34/34) and 65 per cent (22/34) respectively. There were five LCR grey-zone specimens, three of which were confirmed positive, representing 9.7 per cent (3/31) more positive patients over and above those that were positive only above the cutoff. This study shows a high rate of chlamydia infection from urine samples in men, using LCR. The sensitivity of the assay can be further increased by incorporation of a grey-zone region. It is therefore recommended that highly sensitive DNA amplification tests such as LCR replace conventional methods for the diagnosis of chlamydia infections in men.