Changes in Episiotomy Practice: Evidence-based Medicine in Action

Justin R Lappen; Dana R Gossett


Expert Rev of Obstet Gynecol. 2010;5(3):301-309. 

In This Article

Areas of Uncertainty

While a substantial body of evidence and the recommendation of various professional bodies support restricted rather than routine use of mediolateral episiotomy with spontaneous vaginal delivery, data remain inconclusive regarding the use of episiotomy in the setting of operative vaginal delivery. A recent national survey of obstetric providers in the UK and Ireland reported a preference for routine episiotomy for forceps delivery and restricted use for vacuum delivery.[49] The majority of providers (65.5%) perceived the relationship between operative vaginal delivery and anal sphincter tears as protective. However, various other studies have reported conflicting results.

A retrospective study of 323 operative vaginal deliveries by Robinson et al. in 1999 reported an increased risk of severe perineal laceration in vacuum deliveries with episiotomy compared with those without episiotomy with a relative risk of 3.7; however, there was no effect of episiotomy on perineal laceration in forceps deliveries.[50] A Scottish retrospective, population-based cohort study of over 2100 operative vaginal deliveries reported an increased risk of severe perineal laceration for operative vaginal deliveries with episiotomy relative to those without episiotomy (7.5 vs 2.5%; OR: 2.92).[51] A study published by Kudish and colleagues in 2006 noted a synergistic effect of operative vaginal delivery and midline episiotomy on the risk of anal sphincter laceration.[52] After controlling for age, ethnicity, birthweight and head circumference, the authors found that the joint use of episiotomy and forceps delivery was associated with a 20-fold increased risk of anal sphincter lacerations in nulliparous women and a 77-fold increased risk in multiparous women over spontaneous delivery without episiotomy. The authors concluded that this combination of operative procedures should be avoided whenever possible.

Recently, the British Journal of Obstetrics and Gynaecology published a series of articles with conflicting results in regard to the protective effect of mediolateral episiotomy performed during operative vaginal delivery. First, in 2007, de Leeuw et al. published a retrospective population-based study of over 28,000 operative vaginal deliveries from the Dutch National Obstetric Database.[53] In this study, the overall rate of episiotomy was 79% in vacuum extractions and 89% in forceps deliveries. Mediolateral episiotomy was highly protective for severe perineal laceration with both vacuum and forceps deliveries with ORs of 0.11 and 0.08, respectively. The number of episiotomies needed to prevent one anal sphincter injury in vacuum extractions was 12 and in forceps deliveries was five. In their conclusion, the authors advocate for liberal use of mediolateral episiotomy with operative vaginal delivery.

A year later, Murphy and colleagues published results of both a randomized control trial and a prospective cohort study of mediolateral episiotomy at the time of operative vaginal delivery.[54,55] A total of 200 nulliparous women were randomized to routine or restrictive episiotomy in the second stage of labor at the time when a provider determined that an operative vaginal delivery was required. Patients who did not desire randomization were recruited for the prospective cohort study (n = 1360 of which 1066 had episiotomy). In the randomized trial, routine episiotomy was not associated with a decreased rate of severe perineal laceration, primary postpartum hemorrhage or neonatal trauma; however, this pilot study was underpowered to demonstrate a difference in the primary outcome. The cohort study reported increased rates of postpartum hemorrhage, need for analgesia and perineal infection in the episiotomy group. Episiotomy did not impact the rate of severe obstetric laceration, shoulder dystocia or neonatal trauma. The authors concluded that episiotomy at the time of operative vaginal delivery may increase maternal risks without clear benefit and highlighted the need for a larger randomized clinical trial.

An additional complication in the study of mediolateral episiotomy, and a possible contributor to the mixed research data, is that not all 'mediolateral' episiotomies meet strict definitions of the procedure. In fact, one study demonstrated that only 46% of doctors and 33% of midwives were able to draw a pictogram of an 'ideal' episiotomy that started at the midline and had at least a 40° angle from the midline.[56] A subsequent clinical study examined actual measurements of episiotomies performed.[57] A total of 258 women agreed to participate, of whom 98 (41%) had mediolateral episiotomies. Of these, 58 were delivered by physicians and 40 by midwives. The investigators found that episiotomies cut by doctors were longer and deeper than those cut by midwives. Only 22% of episiotomies cut by doctors were truly mediolateral (between 40 and 60°), and none of the episiotomies cut by midwives met these criteria. The impact of operative technique may be substantial as an emerging body of literature provides evidence of a relationship between the angle of mediolateral episiotomy and the risk of anal sphincter damage.[58,59]

These differences in actual and intended incision angle may in part explain the different outcomes seen in different trials of mediolateral episiotomy. Therefore, an important question remains unanswered: would the routine use of a technically appropriate mediolateral episiotomy performed at the time of operative vaginal delivery reduce the incidence of severe obstetric laceration?

Other areas of uncertainty with regard to the use of episiotomy include its impact on long-term pelvic floor relaxation, pelvic organ prolapse, urinary incontinence and dyspareunia. Data outlined in this article suggest that episiotomy may increase the risk of short- and long-term dyspareunia and pelvic floor relaxation, but may have no effect on the risk for subsequent urinary incontinence or pelvic organ prolapse. However, most trials provide only short-term outcomes and do not report data beyond 1-year postpartum. Additional prospective studies are required to provide additional data on long-term outcomes.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.