Self-collected Vaginal Swabs for the Detection of Multiple Sexually Transmitted Infections in Adolescent Girls

Abstract

Objective: To evaluate the use of self-collected vaginal swabs to test for Neisseria gonorrhoeae, Chlamydia trachomatis, and Trichomonas vaginalis, and to describe the acceptability of this technique to adolescents.

Design: Comparison of a new protocol for sexually transmitted infection (STI) testing with the current standard of care, using the same subjects. Survey of attitudes regarding the self-collection technique.

Setting: A juvenile correctional facility in western Pennsylvania.

Participants: Convenience sample of 133 detainees, 12–17 years of age.

Intervention: Girls were invited to undergo STI screening using a self-collected vaginal swab. Polymerase chain reaction was used to test this specimen for each of the above three infections. Pelvic exams and endocervical testing were performed at the discretion of the physician performing the intake physical, independent of the study.

Main Outcome Measures: The number of infections diagnosed using the self-testing protocol, compared to the number diagnosed using standard practices; acceptability of the self-collection technique.

Results: Twenty-four percent of sexually active subjects had one or more infections diagnosed by self-testing: 11.3% had C. trachomatis, 8.5% had N. gonorrhoeae, and 11.7% had T. vaginalis. Only 30% of subjects with infections had pelvic exams while detained; therefore 70% of girls with infections would have been missed in the absence of the self-testing option. The self-collection technique was acceptable to 95% of subjects.

Conclusions: STI testing using self-collected vaginal specimens is highly acceptable to adolescent girls, and can dramatically increase the detection rate for these three treatable infections when pelvic exams are not performed.

Introduction

Sexually transmitted infections (STIs) are common in adolescents, with an estimated 25% of sexually experienced teens developing at least one STI prior to graduating from high school.¹ Neisseria gonorrhoeae and Chlamydia trachomatis can cause infections that may progress to pelvic inflammatory disease (PID) and its sequelae, including chronic pelvic pain, ectopic pregnancy, and infertility. Screening for C. trachomatis in at-risk populations has been shown to reduce the incidence of PID.² Trichomonas vaginalis, a sexually transmitted parasite, is known to cause vaginitis and an increased risk of premature labor in pregnant women; it may also increase transmission of other sexually transmitted pathogens including human immunodeficiency virus (HIV).3, 4, 5 All of these infections are asymptomatic in a significant proportion of infected women. However, even asymptomatic infections can be associated with considerable morbidity.

Studies of teens in juvenile detention centers demonstrate that this is a particularly high-risk population, with STI prevalence ranging from 9% to 48%.6, 7, 8 Many of these recent studies used nucleic acid amplification (NAA) tests such as polymerase chain reaction (PCR) or ligase chain reaction (LCR) on urine samples to screen for N. gonorrhoeae and C. trachomatis. The sensitivity of LCR for urine specimens compared to culture or NAA of endocervical specimens is 50%–95% for gonococcal infections and 79%–96% for chlamydial infections.9, 10, 11, 12, 13, 14, 15 Though methodologies among these studies vary, most demonstrate sensitivities in the 85%–95% range. In the detention setting, time and personnel constraints may limit the ability to perform speculum exams for endocervical screening of all female detainees. The prospect of performing noninvasive tests for these two pathogens is therefore attractive, as it may substantially increase the yield of infections identified and treated.

Several studies have shown that vaginal introital specimens obtained by patients are useful in the diagnosis of N. gonorrhoeae and C. trachomatis when analyzed using PCR or LCR.16, 17, 18, 19, 20, 21 Since both of these organisms infect the columnar epithelial cells of the endocervix and cause cellular damage, it is not surprising that their nucleic acids can subsequently be detected in vaginal fluid. The reported sensitivities and specificities of this method, when compared to endocervical specimens, are 81%–100% and 93%–99%, respectively. All of the above studies included adolescent as well as adult women. The vaginal introitus has also been shown to be an appropriate testing site for T. vaginalis using PCR or culture.22, 23, 24 PCR testing for N. gonorrhoeae, C. trachomatis, and T. vaginalis can be performed using a single swab specimen. The present study describes the use of this technique to screen female adolescent detainees for all three organisms in a setting where universal screening with speculum exams is not feasible. We hypothesized that more infections would be diagnosed and treated when this screening was available than would be identified by standard care practices, and that self-testing for STIs would be highly acceptable and preferable in this population.