Bacterial vaginosis Diagnosis, treatment and significance in gynecological practice

Abstract

Bacterial vaginosis (BV) is a syndrome which is characterized of a disagreeable vaginal discharge but not necessarily an increase in discharge and is present in nearly 20-30% of women attending gynaecological out-patient clinics. Some women have BV unknowingly but others suffer the disagreeable discharge as a lifelong nuisance and areconstantly visiting physicians without relief. This problem has even been regarded as an aesthetic problem and has received little attention, especially from gynaecologists. The diagnosis of BV is based on the fulfilment of three out of four clinical criteria as introduced by Amsel in 1983. Few studies have been carried out investigating whether BV is a risk factor for other gynaecological diseases such as bleeding disturbances or postoperative infections. The aim of this thesis was to investigate the role of BV in gyrtaecological diseases and to evaluate whether it is possible to make a reliable diagnosis of BV using saved smears, such as air-dried or Papanicolaou (PAP)-stained smears. This would make it possible to investigate a large material and enable retrospective investigations. Since vaginal leucocytosis has been regarded as a sign of genital infection, women with BY and leucocytosis have often been withdrawn from clinical studies. The presence of leucocytosis among women without BV was therefore also investigated to evaluate if this practice is justifiable. The diagnosis of BV using Amsel's criteria was compared with the detection of clue cells in air-dried vaginal smears or with the detection of clue cells in PAP-stained smears. Detection of clue cells in air-dried smears had a sensitivity of 96% and a specificity of 98% in diagnosing BY, and detection of clue cells in PAP-stained smears had a sensitivity of 90% and a specificity of 95%. Leucocytosis cannot be used to exelude BV as nearly 30 OJo of all women have leucocytosis regardless of whether BV is present or not and that healthy women can normally have transient leucocytosis during a menstrual cycle. The significance of BV in postoperative infections after gynaecological surgery and bleeding disorders was studied. BV was found to be a risk factor for pelvic inflammatory disease after first-trimester abortion, and postoperative infection following abdominal hysterectomy. A common complaint with a multifactorial etiology at a gynaecological out-patient department is bleeding disorder. Women with BV and with Mobiluncus were treated in a double-blind study with either metronidazole or with placebo. BV was successfully treated in 76% of the women. The bleeding disorder was regularized in all successfully treated women. Women with BV are at risk of developing a post-operative infection after first trimester abortion. One-hundred and seventy-four women with BV were treated in a doubleblind study with either metronidazole or placebo the week before the abortion. There were 12.2% post-operative infections in the placebo group compared to 3.6% in the group treated with metronidazole. The treatment thus reduced the incidence of infection by more than 3 times. Using simple clinical criteria it is possible to identify a risk group which if treated will reduce the incidence of postoperative infections after legal abortion. In Sweden with 36 000 abortions each year preoperative treatment of 10 000 women with BV will prevent 860 post-abortion PID's per year.