Twenty RCTs met the inclusion criteria (n=2,851), 15 of which provided comparable pain score data to allow a meta-analysis. Of the 20 RCTs, 16 reported method of randomisation, 10 were double blind and all reported on withdrawals. Ten studies had a Jadad score of 3 or more and were considered high quality.
Use of local anaesthesia significantly reduced pain during outpatient hysteroscopy (SMD -0.54, 95% CI -0.86 to -0.23; 15 RCTs). There was substantial heterogeneity for this outcome; similar results and heterogeneity was observed in the subgroup of high quality trials.
When analysed separately, intracervical anaesthesia (SMD -0.36, 95% CI -0.61 to -0.10; three RCTs) and paracervical anaesthesia (SMD -1.28, 95% CI -2.22 to -0.35; five RCTs) significantly reduced pain, but transcervical anaesthesia (five RCTs) and topical anaesthesia (two RCTs) did not. The significant benefit of intracervical anaesthesia was lost when one study that reported data in a different manner was excluded. Heterogeneity was substantial for the analyses of paracervical anaesthesia. Meta-regression showed paracervical injection to be most effective. There was no significant difference between local anaesthesia and controls in terms of incidence of vasovagal events.
The funnel plot was reported as being asymmetrical. There was evidence that studies that showed no benefit may have been missed.