The Effect of Primary Delivery of the Anterior Compared With the Posterior Shoulder on Perineal Trauma: a Randomized Controlled Trial

Background Approximately 85% of vaginal deliveries are accompanied by trauma to the genital tract, with a higher risk at the first compared to subsequent vaginal births [1,2]. Among primipara with vaginal deliveries, 86% sustain a vaginal or perineal tear, and 77% require suturing of a lesion [1,3].

Birth trauma is associated with both short- and long-term morbidity, including pain, discomfort, dyspareunia, and fecal incontinence, and perineal trauma may cause social problems and affect the psychological well-being of the mother [1,4]. The level of postpartum morbidity is related to the degree of trauma [5,6], and studies of preventive measures are therefore of interest.

Genital tract traumas are classified into subtypes according to the location and severity of the lesion. Most studies have evaluated the risk factors for 3rd- and 4th-degree perineal tears that include the anal sphincter complex, the so-called obstetric anal sphincter injuries (OASIS). Predisposing factors are increasing maternal age, heavier birthweight, longer duration of the second stage of labour, oxytocin augmentation, occiput posterior position, and instrumental delivery [7-9]. The risk increases with the number of different risk factors [9]. Protective factors are previous vaginal delivery, epidural analgesia, and multiparity [7-9]. Several perineal management techniques used during delivery have been studied, and a recent Cochrane review concluded that warm compresses and perineal massage seem to reduce the risk of OASIS [4]. The introduction of an interventional perineal protection programme also seems to reduce the incidence of OASIS [10-12].

Leading textbooks recommend primary delivery of the anterior shoulder by gentle traction if the shoulders are not delivered spontaneously [13,14]. However, if shoulder dystocia occurs, the recommended manoeuvres are primary delivery of the posterior arm or primary delivery of the posterior shoulder, with the woman positioned on her hands and knees (Gaskin's manoeuvre) [15,16]. A computer-simulated trial of the manoeuvres used during shoulder dystocia found that primary delivery of the posterior arm caused an 80% reduction in the delivery force and a 70% reduction in stretch to the brachial plexus [17]. Primary delivery of the posterior shoulder could therefore be of advantage during uncomplicated deliveries, but, to the knowledge of the authors, various methods of shoulder delivery have never been studied previously.

The objective of this trial is to evaluate the incidence and degree of perineal trauma after primary delivery of the anterior shoulder compared to primary delivery of the posterior shoulder during vaginal birth in primiparous women in a randomized controlled trial. The hypothesis is that primary delivery of the posterior shoulder reduces the rate and degree of perineal trauma.

Interventions The trial interventions take place during vaginal birth after the delivery of the head. The head is supported in the spontaneous rotation that occurs after its delivery, and the shoulders are delivered according to randomization. The intervention is either primary delivery of the anterior or the posterior shoulder as illustrated in the additional movie files (see additional files 1 and 2). It is performed by the midwife placing her hands around the head of the baby and applying gentle traction in the appropriate direction.

All midwives are trained in the two interventions by the primary investigator (HW) to secure uniform use of the techniques. Training sessions include an introductory video of the two methods and practical training on a birthing phantom (MODEL-med Sophie and her Mum Full Birth Obstetric Trainer, Carnegie, Australia). All midwives are also trained in evaluation and classification of perineal tears using an e-learning programme (GynZone ApS, Aarhus, Denmark). Primary training sessions (n = 14) took place from February to November 2013, and additional sessions are held every second month during the course of the trial to train new midwives and update previously trained midwives. Attendance at a training session accredits midwifes to deliver included patients.

The participants can deliver in the position they prefer, and if spontaneous delivery of the shoulders occurs, this is to be respected regardless of randomization. If during delivery the midwife judges that an alternative method of delivery is preferred with regard to the safety of the birthing mother, this overrules randomization. Episiotomy may be used in accordance to local guidelines as judged necessary by the midwife. In case of vacuum-assisted delivery, which according to guidelines is performed by physicians, the midwife delivers the shoulders. Any deviations from the protocol are registered on the clinical registration form.

Assessment After delivery of the placenta, a blinded midwife or doctor not otherwise involved in the delivery assesses the perineum and grades the perineal tears. Tears are sutured and officially classified and coded independently of the trial by the midwife responsible for the delivery or a doctor according to the hospital guidelines. Secondary outcomes are registered by the midwife responsible for the delivery.

Assessors of the primary outcome and the primary investigator are blinded to randomization.

Registered 3rd and 4th degree tears are validated through manual assessment of patient records. We assume that especially higher degree tears may be diagnosed after trial assessment during repair. In order not to underestimate the level of higher degree tears, data on all 3rd and 4th degree tears in our study population registered in the hospital register during the trial period will be retrieved after the end of the trial. These data will be validated against patient records and any additional confirmed cases of 3rd and 4th degree tears in our study population will be incorporated into the final dataset.

Discussion The purpose of this randomized controlled trial is to compare two methods of delivering the shoulders during vaginal delivery, with perineal trauma as the primary outcome.

The literature on delivery techniques is limited, and previous studies have primarily focused on their effect on OASIS. Most studies have been non-randomized, and different shoulder delivery techniques have not been evaluated previously. Thus, the strengths of this study are the randomized design, the intervention studied, i.e. the delivery of the shoulders at vaginal delivery, and the outcome of any perineal trauma.

The validity of this trial could be affected by the fact that several midwives perform the interventions. But numerous birth assistants are the reality at most centres, thereby increasing the external validity and generalizability of the trial. Additionally, it might be interpreted as a limitation that this is a single-centre trial, although it increases internal validity of the trial. tThe perineal tears are evaluated by several objective assessors (midwives or doctors), which possibly might affect the validity of the outcome assessment. We will try to overcome this issue by validating the registered higher degree tears (3rd and 4th degree) by examining the patient records. Additionally, we are planning to retrieve data on all 3rd and 4th degree tears in our study population during the study period from the hospital registers and validate this data against patient records. We assume that especially higher degree tears may be diagnosed after trial assessment. These tears will be registered centrally because registration commonly takes place after lesion repair. Thus, all higher degree tears included in our final analysis will be validated.

Most delivery assistance techniques used today are based on tradition and heritage and are not evidence based. This trial provides an example of how vaginal delivery techniques may be evaluated in a randomized controlled trial.

The results of this trial will clarify the role delivery of the shoulders has on perineal trauma and thereby add knowledge about recommended birthing techniques.

For futher details please see the published protocol:

Willer H, Aabakke AJ, Krebs L. The effect of primary delivery of the anterior compared with the posterior shoulder on perineal trauma: a study protocol for a randomized controlled trial. Trials 2014;15:291.