Diagnostic Accuracy of Magnetic Resonance Imaging in the Diagnosis of Placenta Accreta Spectrum: A Systematic Review and Meta-analysis

Abstract Objective To evaluate the diagnostic accuracy of magnetic resonance imaging (MRI) in diagnosing placenta accreta spectrum (PAS). Methods We conducted a comprehensive literature search from database inception to November 2023 using terms such as placenta creta, increta, percreta, PAS, MRI, and their respective Medical Subject Headings terms. All prospective and retrospective cohort, case-control, and cross-sectional studies involving prenatal magnetic resonance imaging diagnosis of PAS with subsequent pathological confirmation were included. Results A total of 40 studies encompassing 3664 women met the inclusion criteria, with 1894 cases confirmed pathologically as PAS. The overall sensitivity of MRI was 0.867 (95% confidence interval, 0.807–0.910), and the specificity was 0.860 (95% confidence interval, 0.799–0.905), with a correlation of 0.693 between sensitivity and specificity. The estimated odds ratio was 28.693 (95% confidence interval ( CI ): 14.463–56.924), the negative likelihood ratio was 0.178 (95% CI : 0.122–0.258), and the positive likelihood ratio was 4.316 (95% CI : 3.186–5.846). Analysis of individual MRI criteria revealed estimates of sensitivity, specificity, odds ratio, negative likelihood ratio, and positive likelihood ratio for abnormal placental bed vascularization as 0.500, 0.740, 2.788, 0.571, and 1.645 respectively; 0.384, 0.985, 6.270, 0.471, and 2.720 for bladder wall interruption; 0.766, 0.818, 13.638, 0.262, and 3.375 for the presence of dark intraplacental bands; 0.691, 0.913, 10.828, 0.352, and 3.361 for heterogeneous placenta; 0.688, 0.984, 34.886, 0.254, and 7.164 for indistinctive myometrium; 0.757, 0.864, 8.496, 0.362, and 2.778 for loss of retroplacental dark zone; 0.828, 0.593, 5.829, 0.329, and 1.766 for myometrial thinning; and 0.518, 0.916, 9.473, 0.411, and 3.526 for placental bulge, respectively. Conclusion MRI demonstrates significant utility in diagnosing PAS and its severity. It is recommended for use in all cases with inconclusive ultrasonographic findings. Registration Registration number CRD42021267501