Efficacy and safety of emergency contraception interventions: a systematic review and Bayesian network meta-analysis

To assess the efficacy and safety of emergency contraception (EC) interventions.

Literature published until 28 December 2024 was retrieved from nine databases. Randomised controlled trials (RCTs) evaluating the efficacy of copper or levonorgestrel intrauterine devices (Cu-IUDs or LNG-IUDs), oral levonorgestrel (LNG), mifepristone (MFP), ulipristal acetate (UPA) and oral EC plus cyclooxygenase inhibitor (COX) were considered eligible. A Bayesian network meta-analysis was conducted to evaluate the relative risk (RR) with a 95% credible interval (CrI) for efficacy and adverse events. Evidence certainty was graded using the CINeMA GRADE application. Pairwise meta-analyses, robustness analyses and Bayesian meta-regressions were conducted to verify the reliability of the results.

Of 17 244 studies, 92 involving 37 260 women were included. With robustness verified, the Cu-IUD was the most effective EC treatment. No significant differences in EC efficacy were observed between LNG-IUD and all other treatments, but with low- or very-low-certainty evidence. With acceptable adverse events, co-treatment with LNG and COX (LNG-COX) was more efficacious than single-dose (RR=0.112, 95%CrI: 0.005, 0.686) or double-dose LNG (RR=0.095, 95%CrI: 0.004, 0.605) with high certainty. MFP (25 mg) and UPA had comparable EC efficacy (RR=0.782, 95%CrI: 0.425, 1.433), whereas MFP tended to induce advancement of the subsequent menstrual period.

Although the efficacy of LNG-COX was confirmed by high-certainty evidence, large-scale validation trials are required before it can be recommended clinically. Owing to within-study bias and imprecision, large RCTs are strongly recommended to assess the efficacy of ECs for interventions without direct comparisons.

CRD42024627671.