Can Chlamydial Cervicitis Influence Diagnosis of Bacterial Vaginosis?

Abstract

In the paper entitled “Evaluation of a Point-of-Care Test, BVBlue, and Clinical and Laboratory Criteria for Diagnosis of Bacterial Vaginosis,” by Bradshaw et al., published in the March 2005 issue of the Journal of Clinical Microbiology (2), the authors concluded that because a majority of women at greater risk of the sequelae of bacterial vaginosis (BV) are not in settings where conventional diagnostic methods are either practical or possible, they would greatly benefit from access to rapid and reliable point-of-care tests to improve the diagnosis and management of BV. This conclusion was based on an analysis of results of several clinical (Amsel [1]) and laboratory (Nugent [6]) criteria routinely recommended for diagnosis of BV with results of rapid tests. In our opinion, this conclusion does not apply to women with chlamydial cervicitis. The use of both Nugent and Amsel criteria simultaneously appears to be important for correct diagnosing of BV in this group of patients. To estimate the prevalence of BV by Amsel and Nugent criteria, we studied a group of 60 (mean age, 31.7 ± 6.82 years) nonpregnant women suspected for cervicitis. Patients with Neisseria gonorrhoeae, Trichomonas vaginalis, yeast infection, and human immunodeficiency virus infection were excluded from this study. The vaginal pH was measured using color strips. Three sterile cotton swabs (the first for Gram staining, the second for the KOH test, and the third for culturing of genital mycoplasmas) were used to obtain material from the posterior vaginal fornix, and one Dacron swab was used to obtain material from the endocervical canal for detection of Chlamydia trachomatis antigen by Chlamydia Direct IF (bioMerieux) (3). Evaluations of the vaginal Gram-stained smear and other criteria were performed by persons with more than 5 years of experience. C. trachomatis was confirmed in 31 (51.6%) cases, and genital mycoplasmas were cultured in 17 (28.3%) cases. BV was diagnosed in 14 (23.4%) of 60 examined women based on threeor four Amsel criteria (pH ≥4.5, thin milky homogenous discharge, presence of “clue cells,” and amine odor upon mixing vaginal fluid with 10% KOH) and was diagnosed in 6 of them (10%) according to Nugent scores (0 to 3, negative; 4 to 6, intermediate; 7 to 10, positive) (1, 6) (Table ​(Table1).1). In five out of the remaining eight women with Amsel-positive, Nugent-negative scores, coinfection with C. trachomatis was observed (Table ​(Table2).2). Interestingly, three or four Amsel criteria were positive among 57.1% of patients with intermediate Nugent scores and in 8.5% with Nugent negative scores. BV was diagnosed by Amsel criteria in four cases out of 47 women with normal flora (Nugent score, 0 to 3). In all four cases, coinfection with C. trachomatis (in one case, also with genital mycoplasmas) and absence of “clue cells” were detected, although another three Amsel criteria were positive (Table ​(Table2).2). Nugent-negative, Amsel-positive scores (based on positive Amsel criteria other than “clue cells”) very often resulted from contact bleeding and the presence of cervical mucopurulent contents, especially during chlamydial or mycoplasmal infection. That's why observation of “clue cells” in our opinion must be the main characteristic when Amsel criteria are used. This is confirmed by the next observation: among seven women with intermediate Nugent scores, in four cases BV was diagnosed by using Amsel criteria, in three cases the presence of “clue cells” was found, and in one case coinfection with C. trachomatis was found. Nugent intermediate scores are reported by many authors as abnormal vaginal flora and very often are accompanied by the presence of “clue cells” (2, 6). TABLE 1. Clinical and laboratory data of studied women TABLE 2. Characteristics of 14 women with BV, based on Amsel criteria It is a well-known fact that abnormal vaginal discharge is a symptom of many different pathological processes in women's genital tracts. Changing (increasing) vaginal pH may follow for many reasons, especially with chlamydial infection (4, 5, 7), the presence of cervical ectopia, contact bleeding, and others. Taking into account that chlamydial cervicitis can influence Amsel criteria, avoiding false-positive results and correct diag-nosing of BV using both Nugent and Amsel criteria (mainly “clue cells”) simultaneously is important for patients with cervicitis.