Evidence on the effects of the second stage of labour per se on the pelvic floor prior to the birth of the fetus is limited. Studies to evaluate the impact of the second stage on the pelvic floor are challenging due to technical reasons and possible ethical concerns. With biomechanical modeling, changes on tissues of the pelvic floor have been studied to some extent. These are part of various changes to the pelvic floor during the different stages of labour. During the first stage there is increasing strength and frequency of uterine smooth muscle contractions, progressive dilatation of the cervix and descent of the fetal presenting part into the pelvis. The second stage of labour commences after the cervix is fully dilated and there is further descent of the fetal presenting part. This stage is complete when expulsion of the fetus from the introitus takes place with maternal voluntary efforts. There is progressive increase in the intrauterine pressures during uterine contractions and voluntary pushing efforts and an additional increase in intrauterine pressures has been estimated with ventouse traction of 113 N [1] and forceps traction of 200 N [2].

During the second stage of labour the descent of the fetal head causes progressive distension of the perineum and anal dilatation. The stage of fetal crowning occurs when the vulvovaginal opening is persistently dilated by the fetal head, the biparietal diameter has passed through the level of the ischial spines and there is no retraction of the fetal head between contractions. The perineum is thinned out and may undergo spontaneous tearing in primigravidae. This is less common in multiparous women. Episiotomy is usually performed at the stage of crowning to facilitate delivery with less risk of anterior perineal trauma.

The levator ani muscle complex consists of five parts and is responsible for closing the urogenital hiatus against the opening forces exerted by intraabdominal pressure. The muscle is distributed in a U-shape around the urethra, vagina and rectum, and the force caused during the muscle contraction compresses the rectum, vagina and urethra, from back to front. Using 3D/4D ultrasound imaging and dynamic MRI, detachment of the puborectalis muscle from its insertion to the inferior ramus of the pubic bone can be visualised and is noted in one third of all women having a normal vaginal birth [3]. However, 85–90 % of primiparous women can have a vaginal birth without disruption of the pubovisceral muscles, a result of compensatory stretching of the perineal body. This has been called the “fusible link hypothesis” [4]. Geometric models have suggested that muscle damage in second stage may be due to overstretching. The pubococcygeal/pubovisceral muscle, which is the most medial in the levator ani complex sustains the largest tissue strain with a stretch ratio of 3.26 [3]. MRI studies of simulated vaginal childbirth have also demonstrated the maximum stretch ratio to be 3.5 corroborating similar findings with the geometric models [4]. The maximum stretch ratio is usually achieved at the time of crowning of the fetal head, therefore factors which increase the stretch ratio such as instrumental deliveries and increased head circumference of the fetus or malposition would further increase the risk of levator ani trauma.

The pudendal nerve is also at risk of injury during vaginal delivery. It is located relatively superficially in the pelvis and is therefore prone to stretch damage during childbirth. Stretch injury of the pudendal nerve has been reported in 38–42 % of vaginal deliveries [5]. Although the nerve injury is usually reversible due to re-innervation, severe injuries or total transection may result in protracted restoration of nerve function as shown in simulated childbirth using rat models [6]. Pudendal nerve trauma during childbirth has been implicated in postpartum faecal and urinary incontinence and is usually attributed to increased stretch and pressure injury during the second stage of labour.

The American College of Obstetricians and Gynecologists defines prolonged second stage of labour as longer than 3 h in a nulliparous woman with regional anaesthesia and 2 h without regional anaesthesia [7]. Prolonged second stage of labour can be associated with malpositions, fetal macrosomia and relative cephalopelvic disproportion. A combination of raised intrauterine pressure in the second stage of labour due to intense contractions and maternal voluntary pushing efforts can lead to ischaemic injury of the pelvic nerves and muscles. This can lead to denervation injuries, which in some cases can be permanent, particularly if the active phase of the second stage of labour is prolonged. A neurophysiological study using concentric needle electromyography (EMG), pudendal nerve conduction tests and perineometer in 96 nulliparous women showed EMG evidence of re-innervation after vaginal delivery in 80 % of cases. Prolonged active second stage of labour (>1 h in primiparous women) and heavier babies showed the most significant EMG evidence of nerve damage, whereas passive second stage of labour did not increase the risk of denervation injury to the pelvic floor [8]. An MRI study in women 9–12 months after the first delivery showed that the use of forceps, anal sphincter rupture and episiotomy were associated with increased risk of levator defects. Women with levator injury were also found to have a 78 min longer second stage of labour in this study [9].

The relationship between epidural analgesia and perineal trauma during childbirth is not clear. It can be inferred that epidural analgesia may be associated with an increase in the rate of severe perineal trauma indirectly due to the increased likelihood of operative vaginal delivery and episiotomy associated with instrumental delivery. However, it can also be postulated that epidural analgesia in labour can exert a protective effect against severe perineal trauma by reducing the uncontrollable urge to push during delivery. To evaluate the independent risk of epidural analgesia in labour, logistic regression analysis was used in a study and epidural was not found to be an independent predictor of perineal injury [10]. A large population based study has shown that epidural analgesia in labour does not increase the risk of perineal trauma [11].

The effect of epidural analgesia on levator trauma is not clear. As epidural analgesia increases the duration of second stage with higher risk of instrumental delivery particularly forceps, it can potentially lead to levator ani injury and microtrauma. Prolonged pushing in second stage can lead to neuromuscular or vascular injury due to distension and stretching and longer second stage of labour has been associated with levator trauma [9]. However, intrapartum epidural may be beneficial by preventing premature pushing and may exert a protective effect on levator trauma by muscle relaxation of the pelvic floor with an effective epidural analgesia [12].

It is estimated that 85 % of women who deliver vaginally will suffer some degree of perineal trauma in the UK. The incidence of OASIS varies between 1.7 and 18 % [13–15]. Variations are related to the standards of reporting and training to recognise such tears and the methods of identifying OASIS. Various other factors, such as instrumental deliveries, use of episiotomy and parity, also affect the incidence of OASIS, hence the variation in the rates. Gurol-Urganci et al. reported that the rate of anal sphincter injuries tripled in England between 2000 and 2012 (1.8–5.9 %). Improved recognition of these tears, standardisation of the classification of perineal trauma and the decline in the use of routine episiotomy probably contributed to the increasing incidence of OASIS [16].

Episiotomy is one of the commonest procedures performed in labour. Albeit a fairly common practice, there is no robust scientific evidence in support of performing episiotomies and it seems that it has just crept into clinical practice. Episiotomy is usually performed to increase the vaginal orifice shortly prior to the delivery of the presenting part. It is also performed by obstetricians prior to instrumental delivery particularly forceps, breech vaginal deliveries and in deliveries where shoulder dystocia is anticipated such as fetal macrosomia. Episiotomy is also given for indications such as rigid and inflexible perineum particularly in primigravidae, to expedite vaginal delivery in cases of fetal distress during the second stage of labour, and to avoid multiple vaginal tears by performing a controlled surgical incision.

A Cochrane review has recommended the practice of restrictive episiotomy as it has the benefit of reducing severe perineal trauma and posterior perineal trauma although there was an increased risk of anterior perineal trauma [17]. This review included 8 studies with a total of 5541 women. In the routine episiotomy group, 75.15 % of women actually had episiotomies. In the group with restrictive episiotomy, 28.40 % had episiotomy. The restrictive episiotomy group had shown less risk of severe perineal trauma (RR 0.67, 95 % CI 0.49–0.91), less need for suturing (RR 0.71, 95 % CI 0.61–0.81) and lesser complications with wound healing (RR 0.69, 95 % CI 0.56–0.85). The use of restrictive episiotomy was associated with a higher incidence of anterior perineal trauma (RR 1.84, 95 % CI 1.61–2.10). This review did not recommend a specific type of episiotomy (midline versus mediolateral) and left the choice of type of episiotomy with the accoucheur due to lack of evidence.

Women delivering vaginally following intrauterine fetal demise appear to have a lower overall risk of perineal trauma compared to women with a live birth in a retrospective case matched study. This study included 323 women who delivered vaginally following intrauterine fetal death and was controlled for age, parity, gestational age and birth weight and excluded other significant factors, which contribute to perineal trauma such as instrumental delivery and episiotomy. The study concluded that women with intrauterine fetal death had a lower risk of perineal trauma (RR 0.16, 95 % CI 0.12–0.22) as well as lower risk of OASIS (RR 0.12, 95 % CI 0.03–0.50), which may be due to differences in biomechanics of childbirth in cases of intrauterine fetal demise [18].

Twin vaginal births generally tend to happen in earlier gestations than singleton vaginal births. It can be surmised that the lower birth weight and head circumference of twins may cause less perineal trauma than their singleton counterparts. Data looking at perineal trauma in twin pregnancies are limited. A recent retrospective cohort study [19] comparing twin vaginal deliveries (1538) and singleton vaginal deliveries (91,312) in a single tertiary unit identified nulliparity (twins adjusted OR 5.9, 95 % CI 1.7–20.9; singletons adjusted OR 3.9, 95 % CI 3.5–4.4), occipitoposterior position (twins adjusted OR 3.00, 95 % CI 1.1–8.0; singletons adjusted OR 1.6, 95 % CI 1.3–2.00), instrumental delivery (twins adjusted OR 4.3, 95 % CI 1.2–15.4, singletons adjusted OR 2.4, 95 % CI 2.2–2.6) and birth weight (twin adjusted OR 1.1, 95 % CI 1.0–1.2; singletons adjusted OR 1.07, 95 % CI 1.06–1.08) to be independent risk factors for OASIS both in vaginal twin and singleton deliveries. The authors also concluded that no single risk factor posed a higher risk in twins than in singleton pregnancy. The OASIS rate in twin vaginal deliveries (1.27 %) was approximately half than in singleton deliveries (2.55 %) in this study. This is not different to the risk factors which have already been identified to increase the risk of perineal trauma in singleton vaginal births however it is still clinically relevant in counseling women with twin pregnancies prior to making decisions about mode of delivery.