Episiotomy is performed during 30 to 35 percent of vaginal births, but its use has declined consistently in the past two decades. There has been concern that, compared with restrictive use, routine episiotomy is associated with a higher risk of anal sphincter and rectal injuries and that it precludes a woman from giving birth with an intact or minimally damaged perineum. There is no consensus about the use of episiotomy, illustrated by wide practice variation and a high rate of use in some settings. Hartmann and associates conducted a systematic review of studies of restrictive versus routine use of episiotomy.

Maternal outcomes included short-term complications such as degree of perineal injury and pain close to the time of birth, as well as longer-term complications such as urinary and fecal incontinence, pelvic floor defects, and sexual dysfunction.

In the strongest trial, which had a high gradient between restrictive and routine episiotomy use, the authors found that women in the restrictive-use group were more likely to have an intact peritoneum: 33.9 percent in this group had neither posterior perineal lacerations nor episiotomy, compared with 24.3 percent of patients in the routine-use group. Other studies showed similar results. Pain severity was similarly distributed between the routine and restrictive groups at 10 days postpartum. Another study, which used a visual analog scale, found that the routine-use group had more severe pain on the third postpartum day. Healing was the same between groups.

In terms of incision type, a single poor-quality study compared patients who had a midline incision with those who had a mediolateral incision. Midline incisions resulted in more extensions of the episiotomy into or through the sphincter; however, this group had less bruising of the perineum and resumed sexual intercourse earlier.

At three months, women who had episiotomy had weaker pelvic floor muscle strength compared with women who had spontaneous tears, but physical examination showed no difference in incontinence symptoms and degree of prolapse, making the significance of this finding unclear. Self-reports of urinary continence showed no difference between groups. Two studies showed an almost twofold increase in risk of incontinence of stool or flatus (relative risk: 1.91). Comparisons of sexual function showed that, in the short term, women in the restrictive-use group resumed sexual intercourse sooner than women in the routine-use group. Long-term outcomes of sexual function showed comparable results between groups.

The authors conclude that there is no maternal benefit from episiotomy, and in some cases postpartum injuries could have been averted had episiotomy not been routine. Limiting its use to fetal indications can reduce episiotomy rates to as little as 8 to 10 percent, but rates of less than 15 percent should be immediately realizable.