Indications and Implications of Episiotomy

Many factors colluded to make this the most common operation in obstetrics.¹ When childbirth occurred at home with a lay accoucheur, variable lighting, no standardized suture material, and generally small infants, episiotomy was rare. In a giant cultural shift early in the 20th century, childbirth became a medical procedure, largely taking place in hospitals with professional attendants. Furthermore, improved understanding of nutrition and better pregnancy dating capabilities have led to a small but significant increase in birthweight.

The primary impetus for episiotomy was the widespread use of forceps for assisting delivery. Maternal soft tissues rarely withstood implementation of forceps without laceration, and the greatest proponents of forceps use became equally fervent about the value of a clean, straight incision in terms of safety and ease of repair.

Coupled with hospitalization for childbirth (and by some accounts the primary driving force behind that shift) was the increasing use of anesthesia/analgesia. Although demonstrating marvelous benefits, adequate pain relief seems to carry the inexorable burden of lessening expulsive efforts. Thus, delivery was prolonged at the very time in history when those supervising its progress were people of professional stature and having multiple professional obligations. It must have been difficult indeed for a man of Dr. Pomeroy's (1867–1925) reputation to sit at a patient's perineum for 1 or 2 hours, held captive by maternal expulsive efforts and the caprice of nature. The temptation to bring control into the process would surely have been irresistible.

Add to these factors the availability of suture material, strong and ready lighting, burgeoning knowledge about asepsis and surgical technique, and the natural compassion for a suffering patient, and the phenomenal rise in popularity of episiotomy can readily be understood.

Like much of modern obstetrics, this practice was based on recommendations of experts rather than on principles of scientific investigation. It gained the stature of tradition within 30 years, and even as late as 1989, Williams Obstetrics made only passing mention of opposition to its routine use.² Despite a steady decline in forceps use, and anesthetic techniques that in theory should allow for an optimally controlled spontaneous delivery, the belief has been deeply ingrained in the professional ethos that episiotomy provides benefits not otherwise achievable for mother and infant.

It took the consumerist movement of the 1970s to shake this conviction. The desire to control the birthing process had now possessed patients as well as obstetricians, and increasing conversational freedom allowed the complications of medicalizing childbirth to gain widespread media attention. In this second “cultural revolution,” women emphatically declared their need for a delivery that is not only safe but also personal and comforting. Among many other elements of “traditional” medical care that came into question was the routine use of episiotomy.

In addition to the consumerism movement, the scientific community had also begun to hold itself to a higher standard of accountability than mere conformity to consensus or expert opinion. The move toward evidence-based medicine demanded that any intervention be proven to hold greater merit than risk, and the practice of episiotomy came under scrutiny in the mid-1980s. Many trials suggested less benefit and more harm than had previously been recognized, and the medical community began a continuing shift toward reserving episiotomy for particular indications. Unfortunately, most of the studies on the subject suffer from serious design flaws that prohibit a full understanding of the circumstances under which an episiotomy might, indeed, be of benefit.  Nonetheless, there has been a clear shift in practice away from the routine use of episiotomy by more recently trained obstetricians.³

Several indications have been used as empiric reasons for performance of an episiotomy.^{4, 5} One advantage is reduction of trauma to the fetal head, particularly in vulnerable premature infants. Another proposed advantage is shortening the second stage of labor, thereby providing respite for mother and baby from the exhaustive work of delivery. It is presumed that a shorter second stage will result in less infant hypoxia, less sepsis, and less maternal infection as well as the de facto benefit of “getting it over with.”

Another argument in favor of episiotomy is concern over integrity of the pelvic floor. Prolonged labors and large infants are known to be risk factors for subsequent disorders of pelvic floor anatomy and function. By providing greater outlet dispensability without stretching, it is felt that innervation and anatomic relationships might be better preserved.

Episiotomy is often recommended in the event of fetal distress and shoulder dystocia to deliver the infant more rapidly. The term “soft-tissue dystocia” was coined to encompass the notion that the perineal body may impede labor progress to a measurable and on occasion detrimental degree. Relief of this dystocia by episiotomy allows for prompt delivery of the infant.

Lastly, episiotomy is considered to be indicated if a significant spontaneous laceration appears otherwise unavoidable, which includes most cases in which forceps are used. Some include use of a vacuum extractor as carrying higher potential for laceration, and would consider an episiotomy to be of benefit. One of the common exhortations of residents in the mid-1980s was “a cut is faster to repair than a tear!”

Each of these indications has some indirect evidence in support of its value. In the studies cited in the next section, each has been considered as an “indicated” use of episiotomy, in contradistinction to the procedure's “routine” use. It bears comment, however, that no single indication has had the support of a prospective, randomized controlled trial with regard to measurable change in outcome based on providing or withholding the intervention.

The increased scrutiny regarding use of episiotomy has failed to confirm its purported advantages, and indeed has pointed to diametrically opposed outcomes to those presumed. The most obvious instance of this is the claim of protection against unplanned perineal trauma. Woolley believes this question to have the most extensive research base of any part of the episiotomy debate.⁴ Although this protection has been claimed for episiotomy since De Lee's time,⁶ there is ample literature to support the assertion that episiotomy increases propensity for third- and fourth-degree extensions and other lacerations. The Cochrane Library has reviewed the prospective randomized trials on restrictive versus routine use of episiotomy. In the six studies that met their inclusion criteria, the pooled results showed that selective use resulted in less posterior perineal trauma, less suturing, and fewer healing complications. There was, however, more anterior trauma, and no difference in severe trauma, dyspareunia, urinary incontinence, or pain.⁷ A more recent systematic review of the literature confirms lack of benefit from liberal use of episiotomy with regard to perineal laceration severity.⁸ These studies suffer from lacking a “no episiotomy” group, thereby allowing the confounding issue of whether the very things considered “indications” for episiotomy may account for the trauma rather than the episiotomy itself.

In a large retrospective review, episiotomy was identified as a risk factor for severe perineal trauma independent of birth weight and operative intervention.⁹ Thorp and co-workers restricted indications for episiotomy to fetal distress and planned operative delivery, and found a significant decline in major perineal trauma compared to more liberal use.¹⁰ This was a prospective but non-randomized trial, and compares the experience of a single operator with other residents in his program. The confounders are obvious in that study. A historic review of anal sphincter lacerations in one large delivery unit before and after implementation of a restrictive policy toward episiotomy performance had similar findings, with reduction in sphincter damage of approximatly 50%.¹¹  This study used entirely historical controls.  In a large database review, mediolateral episiotomy was found to protect the perineum from severe laceration, whereas midline episiotomy increased trauma substantially.¹² A large review of operative vaginal deliveries also found midline episiotomy to have a higher and mediolateral a lower relative risk of severe trauma than no episiotomy at all.¹³

In addition to causing more perineal trauma initially, it appears that episiotomy wound healing may be somehow different from that of spontaneous lacerations. Women in several studies reported less pain and faster return of function with spontaneous tears, though longer follow-up times appeared to negate these differences.⁴

The issue of long-term pelvic floor protection has been highly controversial, and is the most difficult arena of study in this field. In a retrospective cohort study comparing episiotomy versus spontaneous laceration, the rates of fecal incontinence at 3 and 6 months were significantly higher in the episiotomy patients.¹⁴ This study suffers from all the flaws inherent in a retrospective design. With a relatively short follow-up of 9–12 months postpartum, DeLancey's group demonstrated a significant correlation of anal sphincter disruption with subsequent anal incontinence,¹⁵ which indirectly implicates episiotomy in light of the above studies. In contrast, a gastroenterology study found the odds ratio of a sphincter defect to be 16 with a perineal tear, and only 6.6 with an episiotomy.¹⁶  In one short-term follow-up study patients randomized to restrictive or more liberal use of episiotomy were followed at a mean of 7 months with urodynamics and anal manometry.  No benefits were conferred by liberal use of episiotomy in this study.¹⁷

With regard to urinary incontinence, the picture is even more confusing. Danish investigators studying risk factors for “lower urinary tract symptoms” identified both lesion of sphincter ani and episiotomy to have minor association.¹⁸ Definitions and methodology of this study make generalization highly problematic. A Swedish questionnaire study sought to identify determinants of stress incontinence and concluded that episiotomy was not correlated.¹⁹ This investigation suffers from imprecise terminology and recall bias, and finds several counterintuitive results (no correlation of incontinence with birth weights of children, increased incontinence in an estrogen replacement subset) that urge caution in interpretation of its findings.

In what is probably the largest retrospective evaluation in the literature, Swedish researchers linked three national registries to identify patients receiving surgery for urinary incontinence and then evaluated retrievable variables believed to affect this condition. In contrast to the above studies, they found a negative association between episiotomy and subsequent incontinence surgery, and no association at all with large perineal tear.²⁰

Short-term studies are hardly sufficient to demonstrate improvement or detriment in long-term outcome measures such as pelvic floor relaxation and development of anal and/or stress urinary incontinence. There is also no realistic way to control for subtle details of episiotomy repair technique. Lastly, confounding variables are only now beginning to be understood, making any study results inherently suspect. This may explain the contradictory findings of the studies on the subject.⁴

Literature examining use of labor epidurals has called into question the notion that shortening the second stage is of any tangible benefit in an otherwise uncomplicated labor.^{21, 22} There has been no difference in Apgar scores or cord pH values of infants whose mothers' labors were allowed to progress beyond the traditional limits versus those delivered by strict active management.⁴ Similarly, protection of the fetal head appears to have little to do with widening the outlet. High inverse correlation with gestational age and the occasional occurrence of intracranial hemorrhage even after cesarean section strongly argues for the primary problem being fetal rather than maternal.²³ In Woolley's extensive review of the subject, four retrospective studies were cited that failed to show an advantage to episiotomy with respect to reduction in incidence of fetal intracranial hemorrhage.⁴

There is little question that on occasion use of an episiotomy hastens delivery. Whether this is of clinical value during an occurrence of “fetal distress” or whether the same results could be achieved through other means are still unanswered questions. Woolley's review, in fact, suggests no difference in 5-minute Apgar scores or in the occurrence of shoulder dystocia based on performance of episiotomy.⁴ With respect to shoulder dystocia particularly, a robust retrospective review showed no outcome difference based on episiotomy performance²⁴ and a small randomized study comparing fetal manipulation maneuvers alone wtih episiotomy alone with both maneuvers and episiotomy found that adding the episiotomy conferred little if any benefit regarding brachial plexus damage, while anal sphincter tears were significantly higher in the episiotomy groups.²⁵ 

The current state of knowledge regarding the effects of episiotomy allows for very little dogma and raises many answerable questions. The overwhelming preponderance of recent literature argues against “routine” or “prophylactic” use of episiotomy. Is there a role for this procedure at all?

Other than the long-term issues related to pelvic floor integrity and function, each of the touted indications for episiotomy is relatively well defined and measurable. What is lacking is a randomized, controlled, prospective trial in which one arm receives episiotomy for particular defined indications, and the other receives no episiotomy at all under any circumstances. There is sufficient literature support at this time to allow this design consideration as an ethical and valuable contribution to the body of knowledge. Work by Klein,²⁶ Robinson,²⁷ Low ²⁸, and Howden³ shows that performance of episiotomy has more to do with accoucheur than any other variable. This suggests that “indications” are in the eye of the beholder. Perhaps it is time to move beyond the question “What are the appropriate indications for episiotomy?” to the more fundamental question “Is there an appropriate indication for episiotomy?” Until this question is answered, prudent practice demands conservative use of the procedure with clear documentation of reasons for its use and repair technique.