Bacterial vaginosis (BV) is a polymicrobial infection caused by bacteria, including Gardnerella vaginalis, Mobiluncus species and anaerobic bacteria in the vaginal flora, instead of the dominant normal flora, mainly lactobacillus. Diagnosis has depended on the classical clinical presentation, which includes a homogenous vaginal discharge with a characteristic fishy odor, elevation of the vaginal pH and microscopic examination of clue cells, desquamated cells covered by the abnormal bacterial flora.1 Recently, Nugent et al's scoring system, which is based on an evaluation of the number of the bacteria with a certain morphology in the vaginal flora, was introduced for laboratory diagnosis.
2,3Risk factors for BV are still controversial. BV is considered a "sexually-associated" rather than "sexuallytransmitted" disease. In different studies, risk factors such as pregnancy, douching, intrauterine device usage (IUD) have been associated with BV.2,4 Although previously regarded as a harmless condition, recent studies have linked BV to several obstetrical and gynecological complications. 4,5 The relationship between contraceptive use and BV has been investigated in several different studies, which have identified IUD usage as a significant risk factor for BV. [5][6][7][8] In this cross-sectional study, we assessed the frequency of BV among patients attending a gynecology outpatient clinic for pelvic examination and investigated the association between current IUD usage and BV.
Materials and MethodsWe collected data from the medical charts and laboratory records for a convenience sample of non-pregnant women aged 15 to 45 years old that consecutively attended a gynecology outpatient clinic between January 1999 and December 2001. Women were included in the study regardless of whether they visited the clinic for routine gynecologic exam or were symptomatic (n=302). Only women who had menses at the time of pelvic examination were excluded from the study (n=21). Since 1995, this gynecology outpatient clinic has been collecting standard information about the genitourinary symptoms and current use of contraception methods of attending women using face-to-face medical interviews conducted prior to pelvic examination. Also, a physician blinded to the interview findings obtains vaginal specimens from the posterior vaginal fornix. Vaginal specimens were evaluated for BV using Nugent et al's scoring system on gram-stained smears. A 10-point scoring system was used in which a predominantly Lactobacillus flora (score 0-3) was called "normal', a mixed flora (score 4-6) was called "intermediate," and a score of 7-10 was considered BV 4,9,10 Data were analyzed using EpiInfo 5.0 statistical software and included frequency distributions, percentages, calculation of odds ratios (OR) and 95% confidence intervals (CI) for odd ratios.
ResultsOf the 281 patients, only 15 (5.3%) were asymptomatic and the rest reported at least one symptom (i.e., discharge, back pain, itching or burning) at the time of pelvic exam. Eighty-three women (29.5%) we...