Seven RCTs (n=446) were included.
There was no evidence of heterogeneity for any of the outcomes investigated.
Post-operative fever (4 studies): the pooled RR was 0.25 (95% confidence interval, CI: 0.14, 0.44).
Endometritis (4 studies): the pooled RR was 0.05 (95% CI: 0.01, 0.38).
Wound infection (2 studies presented data).
One study reported no wound infections in either group, while the second study reported a RR of 0.59 (95% CI: 0.24, 1.45) for wounds oozing or pus. Two other studies reported that there were no wound infections in the treatment group, but did not report the incidence of wound infections in the placebo groups.
Pooled (several types) infection (3 studies).
One study noted a non significant decrease in risk on morbidity that combined fever with 'other clinical signs of obvious infection'. The second study combined endometritis and wound infection, and also noted a non significant decrease. The third pooled all morbid events and noted a significant decrease (RR 0.28, 95% CI: 0.14, 0.54).