There were 5 RCTs (n=4,650) of episiotomy included; 1 RCT (n=333) of spontaneous delivery versus forceps ; 7 RCTs (n=2,582) of forceps versus vacuum extraction; 3 RCTs (n=1,547) of perineal massage ; 9 RCTs of birthing position, 7 with supporting furniture (n=2,958) and 2 without (n not reported); 1 RCT (n not reported) of whirlpool baths; 1 RCT (n=5,471) of head flexion; and 2 RCTs (n=382) of pushing instructions.
Episiotomy (5 RCTs, n=4,650).
The weighted risk difference in perineal traumas that required suturing was -0.23 (95% CI: -0.35, -0.11) in favour of avoiding episiotomy, compared with liberal episiotomy (n=4,631). This is equivalent to an absolute decrease of 23% in the risk of sutured perineal trauma. Avoiding routine episiotomy in 4.4 women would prevent one case of perineal trauma that required suturing. The risk difference was similar when the only trial of median episiotomy was excluded. There was significant heterogeneity between the trials that was not eliminated by excluding any one trial. When including only the 3 trials with a high quality score, the risk difference was reduced to -0.14 (95% CI: -0.27, +0.01), but there was still significant heterogeneity. No difference was shown in anal sphincter trauma (n=4,650).
Operative vaginal delivery (8 RCTs).
In one RCT (n=333) of spontaneous delivery versus forceps, the risk difference for perineal trauma was -0.11 (95% CI: -0.18, -0.04) in favour of spontaneous delivery. This equates to one in nine women who give birth without forceps avoiding a third degree tear.
Vacuum extraction caused less anal sphincter trauma than forceps delivery (risk difference -0.06, 95% CI: -0.10, -0.02), based on 7 RCTs (n=2,582). One anal sphincter tear would be avoided for every 18 women whose delivery was assisted by vacuum rather than forceps. Heterogeneity between the trials was not significant.
Perineal massage (3 RCTs, n=1,547).
In nulliparous women, perineal massage in the weeks before giving birth protected against trauma to the perineum (risk difference -0.08, 95% CI: -0.12, -0.04). One case of perineal trauma that required suturing would be avoided for every 13 women who did prenatal perineal massage. Heterogeneity between the trials was not significant.
No RCTs of easing back the perineum in the second stage of labour were found.
Positioning for birthing (9 RCTs).
No statistically-significant difference was found in perineal trauma when comparing the supported upright position versus recumbent position of the mother during the second stage of labour (risk difference 0.02, 95% CI: -0.05, +0.09), based on 7 RCTs (n=2,958). Heterogeneity between the trials was significant. Analysis of only the 3 good-quality trials (n=1,825) eliminated the heterogeneity without changing the findings (risk difference -0.03, 95% CI: -0.07, +0.01). The effect of birthing position on perineal trauma could not be assessed adequately from the 2 trials that compared the two positions (recumbent and upright) without the use of supporting furniture.
Whirlpool baths (1 RCT, n not reported).
The RCT of using a jacuzzi during labour (but not for birth) found fewer operative deliveries and less perineal trauma in the jacuzzi group (no further details were reported).
Flexion of the head (1 RCT, n=5,471).
A large RCT of infant head flexion and perineal support, as applied by a midwife, versus no such support, found no statistically-significant difference in sutured perineal trauma (risk difference 0.01, 95% CI: -0.02, +0.04).
Minimising pushing (2 RCTs, n=382).
The 2 RCTs that compared instructed with spontaneous pushing found no difference in perineal trauma (the data were not reported).