Background:
Optimal methods for diagnosis of rectal C. trachomatis infection are unclear.

Objective:
To define performance of culture and nucleic acid amplification tests (NAATS) for rectal chlamydial diagnosis.

Method:
Males and females >15 years old who acknowledged receptive anal sex (prior 2 months) and females presenting as a contact to gonorrhea, chlamydia or NGU or with untreated gonococcal or chlamydial infection were approached for enrollment. Four rectal swab specimens were collected for chlamydial culture, Gen-Probe TMA, Becton-Dickinson SDA and Roche PCR. Test performance was evaluated using a standard of any 2 positive of the three tests that excludes the evaluated test.

Result:
To date, 350 evaluable test sets have been collected (250 male and 90 female). Twenty-one (6.01%) specimens were culture positive. Preliminary blinded analysis using a gold standard of any 2 of 3 positive tests revealed the following sensitivities (95% CI): Test A 91.8% (80.4-97.7); Test B 80.0% (67.0-89.6); Test C 100% (92.1-100); Test D (culture) 33.9% (22.1-47.4). Specificity was calculated with results (95% CI) as follows: Test A 96.0% (93.1-97.9); Test B 98.3% (96.1-99.4); Test C 95.1% (92.0-97.2) and Test D (Culture) 99.7% (98.1-100).

Conclusion:
Currently available NAATs are substantially more sensitive than culture for the diagnosis of rectal chlamydial infection. The specificity of NAATs for diagnosis of rectal chlamydial infection, while less than that for culture, is acceptable.

Implications:
Culture-based diagnosis of rectal chlamydial infection results in a substantial number of missed infections. NAATS hold promise as a public health tool for the diagnosis and control of rectal chlamydial infection.