Literature review > Issue_2 > Review Sturm et al. 

Sturm et al. tested whether a diagnosis of bacterial vaginosis (BV) could be made from gram stains of the phosphate buffered saline fluid in which self-administered vaginal tampons were transported. The objective is clear and the study design is well planned in that they compared the diagnosis of BV by the tampon method (using a gram stain of the fluid and grading by the Nugent criteria) to the diagnosis of BV made from using the Amsel criteria as the reference method. Additionally they compared the tampon diagnosis to the other standard method for diagnosis of BV, namely of clinician collected vaginal gram stain using the Nugent scoring criteria. The study was performed in 84 women presenting to clinics for pregnancy related problems. Adequate sample size was studied and the use of 2 observers added to the power of the conclusions, as there was excellent agreement between the results of the Nugent scores from tampon fluid gram stain preparations and the gram stains of vaginal smears collected during the speculum examination for both of the two observers.

There were 21% of the 84 women who were diagnosed with BV using the Amsel criteria, with an overall sensitivity of 91.7% and specificity of 79.3% for the tampon collected fluid gram stain Nugent scores.

The authors also studied several preparations of tampon transport fluid for the tampon sample: unconcentrated, 25ul cytospin and 50ul cytospin, all of which showed excellent agreement compared to the speculum assisted collected vaginal gram stain, using a Nugent score or >6 as a diagnostic score of BV. No gain in diagnostic yield was obtained by the use of cytospin preparations, indicating that the gram stain could be made directly from the unconcentrated transport fluid.

There is much interest in the field of STD diagnostics today about the value of patient collected samples for the diagnosis of sexually transmitted infections [1]. Molecular testing has facilitated being able to use patient samples, which are self collected, such as urine and self administered vaginal swabs, as well as vulvar swabs and fluid from tampon samples. The authors have previously demonstrated the value of being able to use tampons for the diagnosis of chlamydia, gonorrhea, and trichomonas infections [2], and even on a smear made from a tampon [3]. Now they had added conclusive evidence to demonstrate that the smear for the diagnosis of BV can be made from the fluid, rather than the tampon, which preserves the integrity of the sample for amplified testing for other organisms, thus eliminating the possibility of contaminating the tampon sample.

It is significant that three sexually transmitted infections (STI) and also now BV can be made from a patient collected tampon sample placed into 10 ml of phosphate buffered saline. Use of such samples will facilitate large public health screening programs, where clinicians may not be available. Future studies comparing the diagnostic capabilities of such samples compared to self administered vaginal swabs need to be performed for the three STIs, as well as BV. Also needed are studies assessing the acceptability and feasibility of such tampon-collected vaginal samples from various population types of women.

References:

1. Gaydos, CA and Rompalo AM. The use of urine and self-obtained vaginal swabs for the diagnosis of sexually transmitted diseases. Current Infectious Disease Reports 4:148-157, 2002.

2. Tabrizi SN, Paterson BA, Fairly CK, et al. Comparison of tampon and urine as self-administered methods of specimen collection in the detection of Chalmydia trachomatis, Neisseria gonorrhoeae, and Trichomonas vaginalis in women. Int J STD AIDS 1998;9:347-9.

3. Wilkinson D, Ndovela N, Kharsany A, et al. Tampon sampling for the diagnosis of bacterial vaginosis: a potentially useful way to detect genital infections? J Clin Microbiol 1997;35:2408-9.

about SDI | newsletters | grants | publications | literature reviews

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