O3-S6.06 Self-collection vs physician-collection for the detection of Chlamydia trachomatis , Neisseria gonorrhoeae , Trichomonas vaginalis and Mycoplasma genitalium

Abstract

Background: To determine the efficacy of self-collection for primary screening of an array of sexually transmitted infections. Methods: A cross-sectional baseline analysis was conducted on a cohort of 300 female sex workers (FSW) within an outpatient clinic from a suburban/slum area of Nairobi, from December 2009 to June 2010. APTIMA transcription-mediated amplification (TMA) assays (Gen-Probe Incorporated) were used for the detection of Chlamydia trachomatis (CT), Neisseria gonorrhoeae (GC), Trichomonas vaginalis (Trich) and Mycoplasma genitalium (MGen) infections and, were analysed in San Diego, California. The APTIMA COMBO 2 assay (AC2) was used for CT/GC, the APTIMA Trichomonas vaginalis assay for Trich, and an APTIMA Mycoplasma genitalium research assay for Mgen. FSW conducted a self-collected sample in privacy, based on standardised instructions, and then underwent a pelvic exam to obtain a physician-collected sample. Results: A total of 299 FSW (mean age of 30) participated, of which 15% were HIV-seropositive. MGen was the most common infection (12.7% in physician; 20.7% in self-collection), followed by Trich (7.7%; 9.7%), CT (4.3%; 5.4%), and GC (2.7%; 3.3%). For all STIs examined, the self-collected samples detected more positive cases than the physician-collected samples. No self-collected sample missed an infection that was picked by the physician sample. k for agreement between self-collected and physician-collected samples were high for CT: 0.89, GC: 0.89 and Trich: 0.87, and slightly lower for MGen: 0.73. Using physician sampling as the gold-standard, the sensitivity of self-collection was 100% for all STIs. The specificity for self-collection was high for CT, GC, and Trich (99.0%, 99.3% and 97.8% respectively) and slightly lower for MGen (91.6%). Conclusion A single self-collected sample appeared to perform with comparable sensitivity and specificity to that of physician-collected sampling for the detection of Chlamydia trachomatis, Neisseria gonorrhoeae, Trichomonas vaginalis and Mycoplasma genitalium.