Twenty-four studies reporting on 23 cohorts (four RCTs, and nine observational and 10 retrospective cohorts) were included in the review (958 patients, range 20 to 122). No evidence of bias was found for most studies; some did not report adverse effects or had inadequate description of participant characteristics. Follow-up ranged from one month to over five years.
The improvement rate was 75% (95% CI 69 to 81; 12 studies) for the short term, 73% (95% CI 62 to 83; 10 studies) for the mid term, and 64% (95% CI 57 to 71; 10 studies) for the long term. The cure rate was 43% (95% CI 33 to 54; 13 studies) for the short term, 37% (95% CI 28 to 46; 11 studies) for the mid term, and 36% (95% CI 27 to 46; 11 studies) for the long term. Significant heterogeneity was found for all the analyses. Publication bias was observed for the long term, but not for the short and mid term.
Seventeen studies reported reinjection rates for patients who were not cured after the first injection. These found that higher reinjection rates were associated with better improvement rates for the long term, but not the short or mid term.
The reported adverse effects were temporary urinary retention, urge incontinence, urinary tract infections, temporary dysuria, and transient haematuria. No serious adverse events were reported.