Conclusion
this systematic review and meta‐analysis summarized the current outcome-predictive protocols for PUL. It suggests that M6 has considerable accuracy in diagnosis of EP in PUL. However, there are few studies of M6, and the negative impact of more complex prediction steps on clinical work is unknown. Using a single biomarker or a single type of biomarker, such as progesterone threshold method, has a higher average production utility, which indicates that it is more cost-effective.