Editorials

Midline versus mediolateral episiotomy

We still don't know which cut is better or how beneficial the procedure is

First described by a Scottish midwife in the 1740s, episiotomy was not used widely until the middle of the 20th century.¹ Prominent obstetricians in the United States argued that childbirth was a "decidedly pathological process" and that a small incision would speed labour, decrease trauma, and allow the perineum to be restored to nearly virginal condition after proper suturing. ^{2 3} This became standard practice in the United States and to a lesser degree in Europe throughout most of the century. The type of incision varied: in the United States, for example, midline episiotomy was preferred, in the United Kingdom the mediolateral procedure was standard. What is the evidence that routine episiotomy is beneficial or that one incision is better than another?

The first systematic review of this procedure was published in 1983.⁴ The evidence at that timethree studies with control groups and no randomised controlled trialsconcluded that "little research has been done to test the benefit of the procedure, and no published study could be considered adequate in its design and execution to determine whether hypothesized benefits do in fact result." The authors noted that the purported benefits of episiotomy, including prevention of third degree laceration, damage to the pelvic floor, and fetal injury (both mechanical and hypoxic), were plausible but unproved. However, they found that the risks of episiotomy, including the extension of the incision, unsatisfactory anatomical results, blood loss, pain, oedema, and infection, were serious.

A subsequent systematic review of the literature in 1995 found that episiotomies prevent anterior perineal lacerations (which result in minimal morbidity) but confer none of the other maternal or fetal benefits that are traditionally ascribed.⁵ The author argued that the incision substantially increased maternal blood loss, the average depth of posterior perineal injury, the risk of damage to the anal sphincter, the risk of improper healing of the perineal wound, and the amount of postpartum pain.

The Cochrane Collaboration's systematic review, last updated in May 1999, included six randomised controlled trials, all published since 1983.⁶ These trials compared the restricted use of episiotomies with routine use. Data from the six studies were combined: in the group routinely given episiotomies, 72.7% (1752/2409) of women in the routine use group had episiotomies while only 27.6% (673/2441) of women in the restricted group had episiotomies. Compared with routine use, the restricted use of episiotomy involved significantly less trauma to the posterior perineum, fewer sutures, and fewer complications of healing. The restricted use of episiotomy was associated with more trauma to the anterior perineum. There was no difference in the incidence of severe vaginal trauma, dyspareunia, urinary incontinence, or scores on measures of severe pain. The Cochrane reviewers concluded that restricted policies have some benefits when compared with routine episiotomy but called for further trials to address several unanswered questions, such as what the indications are for the restricted use of episiotomy in an assisted delivery, a preterm delivery, a breech delivery, and in predicted macrosomia and tears presumed to be imminent.

One of the greatest concerns is difficult to address in a randomised controlled trial: what is the relation, if any, between episiotomy and pelvic floor disorders later in life, especially urinary stress incontinence and relaxation of the pelvic floor? Although some obstetricians contend that episiotomy may help prevent these outcomes, there remains a need for epidemiological studies to examine this belief. ^{7 8}

The more pressing research need, however, is to evaluate which episiotomy technique (mediolateral or midline) provides the best outcome. There have been only two published trials that addressed this question, both of which were excluded from the Cochrane review because of poor methodological quality. ^{6 9 10} While it is not clear what the ideal rate of episiotomy might be for primiparous and nulliparous women, in Sweden 9% of primiparous women have episiotomies.¹¹

It is important to ascertain what the appropriate indications for episiotomy are and which is the best technique to use. The suggested advantages of the midline procedure include better sexual function in future and better healing, with improved appearance of the scar. On the other hand, the midline procedure may be associated with higher rates of extension and a coincident increase in perineal trauma. Although this question could be best addressed with randomised controlled trials, such trials rarely detect uncommon events. It is important, therefore, that cohort and case-control studies are designed to look at important but uncommon events such as severe perineal trauma and the development of rectovaginal fistulas.

As episiotomy has been more carefully studied, its routine use has been questioned and has declined in some settings.⁵ However, continued reassessment is needed because this promotes excellence in clinical practice and better outcomes for patients.¹² The relationship between a woman and her clinician should be built on trust, and the benefits and the risks of a procedure such as an episiotomy must be openly discussed to ensure truly informed consent.

The reexamination of the use of episiotomy that has occurred over the past 20 years underscores both the important role of systematic reviews in stimulating research and an often unappreciated issue in assessing procedures: what should be done with long standing procedures that have never been assessed using an evidence based approach. An important next step with episiotomy is to assess the relevant benefits of the midline versus the mediolateral technique. Randomised controlled trials should be conducted soon and their results disseminated broadly for the benefit of mothers and their children throughout the world.

Stephen B Thacker, director. 

Epidemiology Program Office, Centers for Disease Control and Prevention, Atlanta, GA 30333, USA (sbt1{at}cdc.gov)