Objective: The objective of our study was to review the efficacy and safety of atosiban and nifedipine using both direct and indirect evidence.
Data sources: We performed electronic (e.g. MEDLINE, EMBASE, CENTRAL) and hand searches (last search August 2011) to identify randomized controlled trials (RCTs) comparing atosiban with nifedipine directly, or either drug with betamimetics, in order to allow for indirect analyses.
Methods of study selection: A review was conducted of published, unpublished, and ongoing RCTs comparing atosiban with nifedipine directly, or either drug with betamimetics, in pregnant women under threat of preterm labor.
Tabulation, integration, and results: Data were extracted for an intention-to-treat analysis, and meta-analysis was performed using a random-effects model. Data from direct and indirect evidence were combined using inverse variance. Four RCTs were identified that compared atosiban with nifedipine, with another 31 RCTs comparing either drug with betamimetics. Data for the majority of outcomes were unavailable. When evidence was available, the meta-analysis showed that there were no clear differences in the significant effects exerted by the two drugs. All safety outcomes were in favor of atosiban: there were lower incidences of adverse drug reactions, flushing, gastrointestinal tract upset, hypotension, palpitation, and tachycardia in women prescribed atosiban, with the exception of nausea, which was more frequent in such women. On combining direct and indirect evidence, we also identified a higher probability of treatment failure within 7 days of initiation of treatment (birth or alternative tocolytic therapy) and higher incidence of headache in women receiving nifedipine.
Conclusion: Both drugs seem to have a similar positive effect on prolongation of pregnancy; however, the safety profile of atosiban is considerably better.