Side-effects of oral misoprostol in the third stage of labour--a randomised placebo-controlled trial

Misoprostol, an inexpensive, stable, orally active prostaglandin analogue, has been suggested for use in the prevention of postpartum haemorrhage. Potential side-effects, however, need to be quantified.

To compare the rate of postpartum shivering and pyrexia following oral misoprostol 600 micrograms and placebo.

A double-blind placebo-controlled trial. Women in labour were randomly allocated to receive either misoprostol 600 micrograms orally or placebo after delivery. Conventional oxytocics were given immediately if blood loss was thought to be more than usual. Side-effects were recorded. Postpartum blood loss in the first hour was measured by collection in a special flat plastic bedpan.

The labour ward of an academic hospital in Johannesburg, with 7,000 deliveries per annum.

Shivering and pyrexia.

The groups were well matched. Misoprostol use was associated with more shivering (44% versus 11%, relative risk (RR) 4.03, 95% confidence interval (CI) 2.85-5.70), pyrexia > or = 37.8 degrees C (38% v. 6%, RR 6.23, CI 3.89-9.97), 1-hour systolic blood pressure > or = 140 mmHg (33% v. 25%, RR 1.32, CI 1.03-1.70), and diastolic blood pressure > or = 90 mmHg (10.5% v. 3.0%, RR 3.44, CI 1.67-7.11). There were no other significant differences. The study was not designed to be large enough to assess a difference in blood loss > or = 1,000 ml (9% v. 9.7%, RR 0.93, CI 0.56-1.53). Possible effects on blood loss may have been obscured by the lesser use of additional oxytocics in the misoprostol group (14% v. 18%, RR 0.78, CI 0.54-1.13).

This study has shown the association of postpartum oral misoprostol 600 micrograms with shivering, pyrexia and hypertension. The increased blood pressure, as for the trend towards increased abdominal pain, may be secondary to the uterotonic effect of misoprostol. Large randomised trials are needed to assess the effectiveness of misoprostol in the prevention of postpartum haemorrhage, against which the disadvantages demonstrated here can be weighed.