IntroductionMaternal near miss (MNM) is a useful means to examine quality of obstetric care. Since the introduction of the WHO MNM criteria in 2011, it has been tested and validated, and is being used globally. We sought to systematically review all available studies using the WHO MNM criteria to develop global and regional estimates of MNM frequency and examine its application across settings.MethodsWe conducted a systematic review by implementing a comprehensive literature search from 2011 to 2018 in six databases with no language restrictions. The predefined data collection tool included sections on study characteristics, frequency of near-miss cases and study quality. Meta-analysis was performed by regional groupings. Reported adaptations, modifications and remarks about application were extracted.Results7292 articles were screened by title and abstract, and 264 articles were retrieved for full text review for the meta-analysis. An additional 230 articles were screened for experiences with application of the WHO MNM criteria. Sixty studies with near-miss data from 56 countries were included in the meta-analysis. The pooled global near-miss estimate was 1.4% (95% CI 0.4% to 2.5%) with regional variation in MNM frequency. Of the 20 studies that made adaptations to the criteria, 19 were from low-resource settings where lab-based criteria were adapted due to resource limitations.ConclusionsThe WHO MNM criteria have enabled the comparison of global and sub-national estimates of MNM frequency. There has been good uptake in low-resource countries but contextual adaptations are necessary.