Twenty-one RCTs were included (n=1,466 women). Median sample size was 59 (range 15 to 180). Confidence intervals were generally wide, which reflected the small sample sizes.
Genital signs with disrupted epithelium (20 studies): A statistically significant increase in genital signs with disrupted epithelium was found for UC781 (RR 5.75, 95% CI 1.05 to 31.39; one study, three comparisons, n=36 treated and 12 controls) and Savvy (RR 3.67, 95% CI 1.02 to 13.14; one study, n=40). There was no significant difference between any other treatment and control.
Genital signs with intact epithelium (20 studies): A statistically significant increase in genital signs with intact epithelium was found for PRO2000 (RR 1.69, 95% CI 1.08 to 2.60; two studies, three comparisons, n=60 treated and 37 controls). There was no significant difference between any other treatment and control.
Urogenital symptoms (17 studies): A statistically significant reduction in urogenital symptoms was found for UC781 (RR 0.46, 95% CI 0.23 to 0.94; one study, three comparisons, n=36 treated and 12 controls). There was no significant difference between any other treatment and control.
Bacterial vaginosis (12 studies): A statistically significant reduction in bacterial vaginosis was found for dextrin sulphate compared to placebo or no gel (RR 0.61, 95% CI 0.42 to 0.88; two studies, five comparisons, n=36 treated and control group size ranged from 15 to 37). There was no significant difference between any between any other treatment and control.
No heterogeneity was found for many of the above analyses.
Laboratory parameters: Four studies reported no laboratory toxicities. Five studies reported no significant difference between treatment and control.
Other microbiology findings were reported.