Literature review > Issue_6 > Review on Shafer et al. 

The paper by Shafer et al. addresses a very important and common issue in screening for Chlamydia trachomatis and Neisseria gonorrhoeae in asymptomatic women; which sample provides the best sensitivity and specificity for detection of infection and yet maintains acceptability to the person. The population they chose was non-health care seeking young women entering the military. As part of their routine physical they obtained a pelvic examination, but the study design was to compare endocervical specimens obtained by the physician to noninvasive specimens including first-void urine samples and self-collected vaginal swabs. The concept was to see which of these three specimens would be most reliable and acceptable to the women in detecting C. trachomatis and N. gonorrhoeae. In addition, they sought to determine the prevalence of Trichomonas vaginalis by culture. This issue of noninvasive screening for both men and women for sexually transmitted infections has been addressed by previous publications but only a few actually did a careful comparison of the three different collection methods in an asymptomatic population of women for all three pathogens.

The participants in the study, which was a cross-sectional design, included 1841 sexually active women from all 50 states, Guam, Puerto Rico, and the U.S. Virgin Islands. The median age was 18 years, 90% were less than 21 years old, and 92% were unmarried. 43% were of racial/ethnic minority and 44% reported more than four lifetime partners. All of the specimens were processed and tested for C. trachomatis and N. gonorrhoeae using the ligase chain reaction (LCR) (LCx, Abbott Laboratories). Unfortunately, this assay is no longer available and hence the generalizability of this study for this particular assay is no longer applicable. LCx was removed from the market by the FDA due to inherent problems in the manufacturing process. Nevertheless, other nucleic acid amplified assays are still available on the market that could be used in this type of population and consequently the results may be applicable to other nucleic acid amplified tests.

Another interesting aspect of this particular study was that discordant analysis of discrepant results between the three different collection methods was not performed. Rather, the investigators stated that performance of the sample was determined using any positive test by any collection method as the true positive since the ultimate goal was to find out which method was most sensitive and specific using one particular assay, rather than comparing different diagnostic assays.

Overall, the prevalence in this population for C. trachomatis and N. gonorrhoeae was 11.6% and 2.4%, respectively. No one single specimen type could detect all of the positive samples. For C. trachomatis the self-collected vaginal swab detected 81% of all positive specimens compared to 72% and 65% for urine and endocervical swabs, respectively. For N. gonorrhoeae the self-collected vaginal swab detected 72% of all positive samples compared to 40% and 24% for endocervical swab and urine collection, respectively. In summary, for both C. trachomatis and N. gonorrhoeae, self-collected vaginal swabs were by far more sensitive in detecting these two pathogens than either endocervical or urine alone. By combining urine specimen with the self-collected vaginal swab, one could boost the detection of C. trachomatis to 94% whereas by combining the self-administered swab with the endocervical swab could boost the detection rate of N. gonorrhoeae to 93%. Thus, in an asymptomatic population the self-administered vaginal swab is highly acceptable to women and is sensitive at least using LCR. If it is combined with urine collection, then one gains additional sensitivity using two noninvasive screening methods.

With the limitation that the investigators used LCR, which is no longer available, the results of this study do provide important public health information for screening asymptomatic women at risk for sexually transmitted infections. It is likely with the high sensitivity of other nucleic acid amplified tests, that these findings could be applied to routine screening of asymptomatic women in different non-clinic venues as well as in traditional clinic settings. Transcription-mediated amplification (Gen-Probe) and strand displacement amplification (SDA) (Becton Dickinson) and polymerase chain reaction (PCR) (Amplicor, Roche Diagnostics) are three additional nucleic acid amplified tests which are still on the market and which could be utilized in different settings. Each has its own advantages and limitations and similar studies, particularly with the TMA Gen-Probe assay, underscore the high sensitivity of the assay in using noninvasive self-administered vaginal swabs and urine for chlamydia and gonorrhea. Additional studies continue with the other additional diagnostic tests. Studies such as these once again underscore the importance of routine screening of both asymptomatic and symptomatic women with new sensitive molecular diagnostic assays.

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001