Mechanical methods
Membrane sweeping
Membrane sweeping is the most basic of mechanical methods of labour induction. It is performed by inserting a finger through the cervical os and sweeping it around between the chorion and the uterine wall to release prostaglandins.
Membrane sweeps, if performed more than once after 38 weeks' gestation, are shown to reduce the need for formal induction of labour by 40-50% and to increase the rate of spontaneous labour. The success of membrane sweeping is highest in multiparous women. In the United Kingdom, it is recommended that all women are offered regular membrane sweeps in the 2 weeks after their due date (8). If the cervix will not admit a finger it may not be possible to do a formal membrane sweep. In such cases, massaging around the cervix in the vaginal fornices may achieve a similar effect. A small amount of bleeding and cramping is common following the procedure.Amniotomy
Amniotomy, or artificial rupture of membranes (ARM), is rarely used alone for induction. If used in isolation, it is most effective in multiparous women with ripe cervixes (7). For these women, when used in combination with oxytocin prior to the start of spontaneous labour, ARM is as effective as vaginal dinoprostone (see later) in achieving vaginal delivery. Dinoprostone is generally given in preference as around half of those induced can progress to delivery without the need for an uncomfortable and intrusive intravenous infusion (8).
Amniotomy may be achieved using a variety of toothed instruments, but a plastic amniotomy hook is most commonly used to tear the membranes. If the fetal head is not engaged in the pelvis, a gradual release of liquor is required and this can be achieved by puncturing the membranes with a needle.
Amniotomy may be a useful adjunct in spontaneous labours to allow visualization of liquor and placement of a fetal scalp electrode in those at risk.
However, when performed early in labour or prior to labour, ARM can put a patient at risk of prolonged rupture of membranes and infection (7).Induction of labour when the head is high in the pelvis increases the risk of cord prolapse and malpresentation. Medical methods (e.g. dinoprostone) should therefore always be used initially. When ARM is necessary, this should be performed along with fundal pressure (a ‘controlled ARM'). With polyhydramnios, it is useful to keep the examining hand in the vagina until the liquor has drained and the head descended in case of cord prolapse. The woman should sit upright after the ARM to encourage fetal head descent. If the fetal head is very high, some practitioners prefer to conduct the ARM in theatre so that caesarean delivery can rapidly be carried out in the event of a cord prolapse.
Transcervical balloon catheters
Many versions of transcervical balloon catheters have been in use over the past few centuries. The most common version is a Foley catheter with a large reservoir tip which can hold up to 30-50 mL (typical Foley catheters hold approximately 10 mL). Transcervical catheters are inserted through the cervix manually or using an instrument such as ‘sponge holders’. When the position of the balloon is confirmed by inflation of the balloon and pulling back against the cervix, it is inflated to its desired capacity. The distal end of the catheter which hangs outside of the vagina is often taped to the inner thigh to apply tension against the cervix. However, this is probably not necessary as the method was highly effective in the PROBAAT (Dutch for ‘approved’) and INFORM randomized trials in which a single 30 mL balloon was used with no tension on the catheter (32, 33).
Some models of transcervical balloon have two separate balloons, one sits inside the internal cervical os and one in the vagina; the idea is to apply tension between the two balloons against the cervix. Studies comparing this balloon (commonly known by its trade name ‘Cook Balloon’ or the ‘Cervical Ripening Balloon’) to the traditional Foley catheter balloon in women with a previous caesarean section have shown the Foley carries a minimally shorter time to delivery but no difference in time from insertion to active labour, time from balloon expulsion to delivery, caesarean section, instrumental delivery, pain score, need for analgesia, or maternal satisfaction (34).
A more recent study, however, has found that that there may be a beneficial effect on both speed and vaginal birth rate, but only in nulliparous women (35). It must be noted that the double-balloon catheter is significantly more costly than the traditional Foley. Some studies have investigated an extra-amniotic saline infusion which runs while the transcervical balloon is in place. The typical rate is approximately 50 mL/hour. This has been shown to decrease the time from balloon insertion to expulsion as well as the time to delivery when used with both single- and double-balloon catheters (36).Transcervical catheters do not increase the risk of infection according to a meta-analysis (37). They are commonly maintained for up to 12 hours, but this time period was increased up to 4 days in the PROBAAT series of trials without any sign of excessive infective morbidity (32, 33). When the catheter falls out, ARM is performed and oxytocin commenced. Transcervical balloon catheters when compared head-to-head with vaginal misoprostol have longer induction to delivery times but lower rates of hyperstimulation and operative vaginal delivery with no difference in caesarean delivery rates (32).
When comparing mechanical methods to prostaglandins, a Cochrane review from 2012 found that mechanical methods resulted in a similar caesarean section rate, but with a lower risk of hyperstimulation than prostaglandins. No overall increased time to delivery was seen, but there were a larger proportion of multiparous women undelivered at 24 hours when compared to dinoprostone (38).
Laminaria
Laminaria tents are another form of mechanical cervical dilators. They are made from sterile seaweed or synthetic hydrophilic materials, and are introduced into the cervical canal. As these devices absorb water, they increase in diameter and so stretch the cervix. Laminaria appear to be as effective at labour induction as vaginal prostaglandins, but with a markedly reduced incidence of uterine hyperstimulation (38). This may make them safer for women who have had a previous caesarean section. They are commonly used in the outpatient setting for cervical ripening prior to dilation and evacuation of a miscarriage or pregnancy termination above 12 weeks’ gestation.