Ten RCTs (n=1,339)were included in the review.
Pre-term delivery (8 studies).
Progestational agents (17alpha-hydroxyprogesterone caproate and other forms of progesterone) were associated with a significantly reduced likelihood of pre-term birth compared with placebo (OR 0.45, 95% CI: 0.25, 0.80). The NNT to prevent one premature delivery was 10 (95% CI: 6, 24). The results were similar for studies of 17alpha-hydroxyprogesterone caproate only. There was evidence of statistical heterogeneity, hence a random-effects model was used. There was no evidence of publication bias and sensitivity analyses did not influence the results. The results of the meta-regression suggested that year, country, and type and dosage of progestational agents were not associated with the occurrence of pre-term birth.
Low birth weight (6 studies).
17alpha-hydroxyprogesterone caproate was associated with a significantly reduced likelihood of low birth weight compared with placebo (OR 0.50, 95% CI: 0.36, 0.71). The NNT to prevent one low birth weight was 12 (95% CI: 7, 43). No evidence of statistical heterogeneity was found, hence a fixed-effect model was used. There was no evidence of publication bias and sensitivity analyses did not influence the results.
Threatened premature labour (5 studies).
No statistically significant difference was found between progestational agents and placebo in the likelihood of hospital admission for threatened pre-term labour (OR 0.68, 95% CI: 0.35, 1.34). There was evidence of statistical heterogeneity, hence a random-effects model was used. There was no evidence of publication bias and sensitivity analyses did not influence the results.
Perinatal mortality (5 studies provided outcomes).
No statistically significant difference was found between progestational agents and placebo in the likelihood of perinatal mortality (OR 0.69, 95% CI: 0.38, 1.26). No evidence of statistical heterogeneity was found, hence a fixed-effect model was used. The results were similar for studies of 17alpha-hydroxyprogesterone caproate only. There was no evidence of publication bias and sensitivity analyses did not influence the results. The results of the meta-regression suggested that year, country, and type and dosage of progestational agents were not associated with the rate of perinatal mortality.
Respiratory distress syndrome (3 studies).
No statistically significant difference was found between progestational agents and placebo in the incidence of respiratory distress syndrome (OR 0.83, 95% CI: 0.25, 2.76). There was evidence of statistical heterogeneity, hence a random-effects model was used.