Background

Breast stimulation has been suggested as an effective means of inducing labour. It is both an inexpensive and non‐medical intervention which allows women greater control over the induction process. It is not clear how breast stimulation increases uterine contractions, however it has been shown to be effective for contraction stress tests and for the augmentation of labour.

This is one of a series of reviews of methods of cervical ripening and labour induction using a standardised methodology.

Objectives

To determine the effectiveness of breast stimulation for third trimester cervical ripening or induction of labour in comparison with placebo/no intervention or other methods of induction of labour.

Search methods

The Cochrane Pregnancy and Childbirth Group trials register, the Cochrane Controlled Trials Register and bibliographies of relevant papers. Last searched: April 2001.

Selection criteria

(1) randomised controlled trials comparing breast stimulation used for third trimester cervical ripening or labour induction with no intervention or other methods listed above it on a predefined list of labour induction methods; (2) random allocation to the treatment or the control group; (3) adequate allocation concealment; (4) violations of allocated management not sufficient to materially affect conclusions; (5) clinically meaning outcome measures reported; (6) data available for analysis according to the random allocation; (7) missing data insufficient to materially affect the conclusions.

Data collection and analysis

A generic strategy has been developed to deal with the large volume and complexity of trial data relating to labour induction. This involved a two‐stage method of data extraction. The initial data extraction was done centrally.

Main results

Six randomised controlled trials were included in the analysis involving 719 women.

When trials comparing breast stimulation with no intervention were analysed there was a significant reduction in the number of women not in labour at 72 hours (62.7% versus 93.6%, relative risk (RR) 0.67, 95% confidence interval (CI) 0.60 ‐ 0.74). However, this result did not remain significant in women with an unfavourable cervix. There were no instances of uterine hyperstimulation. A reduction in the rate of postpartum haemorrhage was reported (0.7% versus 6%, RR 0.16, 95% CI 0.03 ‐ 0.87). No significant difference was detected in the caesarean section rate (9% versus 10%, relative risk RR 0.90, 95% CI 0.38 ‐ 2.12) or rates of meconium staining. Three perinatal deaths were reported (1.8% versus 0%, RR 8.17, 95% CI 0.45 ‐ 147.77). This finding should be interpreted with caution. The three deaths occurred in the breast stimulation arm of a three‐armed trial (breast stimulation versus no intervention versus oxytocin) of questionable quality. The study was very small (n = 57), involved only high risk women, and the method of randomisation was not reported.

When comparing breast stimulation with oxytocin alone the analysis found no difference in caesarean section rates (28% versus 47%, RR 0.60, 95% CI 0.31 ‐ 1.18). No difference was detected in the number of women not in labour after 72 hours (58.8% versus 25%, RR 2.35, 95% CI 1.00 ‐ 5.54). There was no significant difference in the rates of meconium staining. There were four perinatal deaths. Three in the breast stimulation arm, and one in the oxytocin arm of the three‐armed trial noted above (17.6% versus 5%, RR 3.53, 95% CI 0.40‐30.88).

Authors' conclusions

Breast stimulation cannot be fully evaluated on efficacy and safety grounds from the included studies. Whilst it would appear beneficial in terms of a reduction in the number of women not in labour after 72 hours, and a reduction in postpartum haemorrhage, until safety issues have been fully evaluated it should not be considered for use in a high risk population.