Background APTIMA Combo 2 (AC2) reliably detects Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) at extra-genital sites in men who have sex with men (MSM), but testing three separate samples (pharynx, urethra, rectum) is costly. Self-taken pharyngeal and rectal samples can be added to first-void urine (FVU) and tested as a single pooled sample (PS). We compared the sensitivity of PS with individual sample testing (SOC), and its acceptability to patients. In addition sample processing methods were compared (Method A: swabs and FVU into universal container to laboratory; Method B: APTIMA urine tube spiked with swabs to laboratory).

Methods MSM (symptomatic or CT/NG contacts) were recruited at two clinics. All were tested by PS and SOC; order of sampling was randomised. Demographics, sexual behaviour, symptoms, signs and acceptability were collected. Any positive AC2 test was considered a true positive. Each clinic used one processing method (not randomised).

Results 627/700 planned MSM recruited to date (43% HIV+); 65% symptomatic; 43% unprotected anal sex in the last month. 94 CT and 189 NG infections were detected (prevalence of CT and/or NG 46%; dual infection 6%). The sensitivity of PS and SOC to detect CT and NG was 92% and 96% respectively (p = 0.68). PS failed to detect 22 infections (CT = 6; 1 pharynx, 4 rectum, 1 urethra, and NG = 16; 11 pharynx, 4 rectum, 1 urethral).

Method A was used in 20/22 (91%) of missed cases; 21/22 (95%) were positive at a single site. SOC failed to detect 4 infections (CT = 3, NG = 1). 90% of MSM found self-taken sampling acceptable; 85% (n = 532) would be happy to take their own samples at home.

Conclusion PS compares well with SOC. It offers the potential for significant cost savings and easier home testing. Missed infections may be due to the method of sample processing or low organism numbers.