Mechanical dilatation of the cervix at non-labour caesarean section for reducing postoperative morbidity

During non-labour caesarean, some obstetricians routinely dilate the cervix from above--using finger, sponge forceps or other instruments because they believe that the cervix of women without labour pain is undilated and may cause obstruction of blood or lochia drainage. However, mechanical cervical dilatation using sponge forceps or a finger during caesarean section may result in contamination by vaginal micro-organisms during dilatation, and increase the risk of infection or cervical trauma.

To determine the effects of mechanical dilatation of the cervix during elective/non-labour caesarean section on postoperative morbidity.

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 August 2011).

All randomised or quasi-randomised controlled trials comparing intraoperative cervical dilatation using a finger, sponge forceps or other instrument during non-labour caesarean section versus no mechanical dilatation.

Two review authors independently assessed studies for inclusion, assessed the risk of bias of each included study and extracted data.

We included three trials with a total of 735 women undergoing elective caesarean section. Of these women, 338 underwent intraoperative cervical dilatation with a double-gloved index digit inserted into the cervical canal to dilate, and 397 did not undergo intraoperative cervical dilatation. Three included trials had moderate-to-high risk of bias. None of the three trials reported our primary outcome of postpartum haemorrhage. In one study of 400 women, blood loss was significantly lower in the cervical dilatation group compared with the no dilatation group (mean difference (MD) -48.49 ml, 95% confidence interval (CI) -88.75 to -8.23). The incidence of febrile morbidity and haemoglobin concentrations in the postoperative period in women undergoing intraoperative cervical dilatation was not significantly different from those who did not receive cervical dilatation (risk ratio (RR) 1.07, 95% CI 0.52 to 2.21 (three trials, 735 women) and MD -0.05 g/dl, 95% CI -0.17 to 0.06 (two trials, 552 women), respectively. There were no significant differences in wound infection, change of haemoglobin level, hematocrit level at postoperative period, endometritis, infectious morbidity, or urinary tract infection. There was a significant difference in operative time, which was reduced in cervical dilatation group (MD -1.84 mins, 95% CI -2.21 to -1.47 (one study, 400 women), but this is of doubtful clinical significance.

There was insufficient evidence of mechanical dilatation of the cervix at non-labour caesarean section for reducing postoperative morbidity. Further randomised controlled trials with adequate methodological quality comparing intraoperative cervical dilatation using a finger, sponge forceps or other instrument during non-labour caesarean section versus no mechanical dilatation for reducing postoperative morbidity are needed.