Acceptability and efficacy of vaginal self-sampling for genital infection and bacterial vaginosis: A large, cross-sectional, non-inferiority trial

Abstract

ObjectiveScreening for genital infection (GI), bacterial vaginosis (BV), sexually transmitted infection (STI) and asymptomatic carriage of group B streptococcus (GBS) in pregnant women is a common reason for medical appointments. Objectives were first to determine the non-inferiority of vaginal self-sampling compared with vaginal/cervical classical sampling to screen for GIs, bacterial vaginosis (BV), STIs, and GBS asymptomatic carriage in pregnant women; second to determine the feasibility of vaginal self-sampling.MethodsVaginal self-sampling (VSS) and vaginal/cervical classical sampling (VCS) of 1027 women were collected by health care professionals and simultaneously carried out on each patient. Bacterial infection, yeast infection, Chlamydia trachomatis, Neisseria gonorrhoea, Mycoplasma genitalium, Trichomonas vaginalis and Herpes simplex virus types were systematically screened in both paired VSS and VCS samples.ResultsStatistical tests supported the non-inferiority of VSS compared with VCS. Agreements between VCS and VSS remained high regardless of the type of studied infection. VSS had successful diagnostic performances, especially for Predictive negative value (PNV) (over 90%) for all studied infections. Most participants (84%) recommended the use of VSS.ConclusionsThis study remains the most exhaustive in screening for GI, BV, STI agents and asymptomatic GBS carriage. Given its efficacy and acceptability, VSS seems to be a viable alternative to classic physician sampling among women in the general population. This study provides evidence that vaginal self-sampling can be used as a universal specimen for detection of lower genital tract infections in women.Study Identification numberID-RCB 2014-A01250-4