Nifedipine versus ritodrine for suppression of preterm labor; a meta-analysis

Background. Since large randomized clinical trials comparing the effectiven ess of nifedipine and ritodrine in the suppression of preterm labor are lac king, we performed a meta-analysis on the subject. Methods. We searched the databases Medline and EMBASE using the keywords 'n ifedipine', 'ritodrine' and 'randomized' or 'randomised'. The studies were scored for blinding, method of randomization and type of analysis ('intenti on-to-treat' versus 'par protocol'). Subsequently, two by two tables were c onstructed using 'delay of labor by 48 hours or more', 'delay of labor beyo nd 36 weeks gestation', perinatal mortality, respiratory distress syndrome and admission to a neonatal intensive care unit as end points. Homogeneity between the studies was tested with a Breslow-Day test. Pooled odds ratios were calculated in case homogeneity could not be rejected. Results. We could detect ten studies that were published between 1986 and 1 998, incorporating data of 681 patients. Nifedipine reduced the risk of del ivery within 48 hours compared to ritodrine, but this difference was not st atistically significant (odds ratio 0.85, 95% confidence interval 0.54 to 1 .1). Nifedipine also reduced the risk of delivery before 36 weeks compared to ritodrine, and this difference was statistically significant (odds ratio 0.59, 95% confidence interval 0.39 to 0.90). We are not aware of studies r eporting on long-term outcome. Conclusion. Since studies reporting on long-term outcome are lacking, the c hoice between nifedipine and ritodrine can only be based on obstetrical and short-term neonatal outcomes. From that perspective, nifedipine should be the drug of first choice for the suppression of preterm labor.