Prenatal identification of invasive placentation using magnetic resonance imaging: systematic review and meta-analysis.

OBJECTIVE To assess systematically the performance of prenatal magnetic resonance imaging (MRI) in diagnosing the presence, degree and topography of disorders of invasive placentation and to explore the role of the different MRI signs in predicting these disorders. The diagnostic accuracy of ultrasound and MRI in the detection of invasive placentation was also compared. METHODS MEDLINE, EMBASE, CINAHL and The Cochrane Library, including The Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects and The Cochrane Central Register of Controlled Trials, were searched electronically utilizing combinations of the relevant medical subject heading terms, keywords and word variants for 'invasive placentation' and 'magnetic resonance imaging'. Only prospective studies reporting a diagnosis of invasive placentation at the time of MRI and retrospective studies in which the radiologist was blinded to the final results were included in the analysis. The MRI signs explored were: uterine bulging, heterogeneous signal intensity, dark intraplacental bands on T2 weighted sequences, focal interruption of the myometrium and tenting of the bladder. Summary estimates of sensitivity, specificity, positive and negative likelihood ratios (LR+, LR-) and diagnostic odds ratio (DOR) were based, depending on the number of studies, upon DerSimonian-Laird random-effect or hierarchical summary receiver-operating characteristics models. RESULTS A total of 18 studies involving 1010 pregnancies at risk for invasive placentation were included. The overall diagnostic accuracy of MRI in detecting the presence of invasive placentation was: sensitivity, 94.4% (95% CI, 86.0-97.9%); specificity, 84.0% (95% CI, 76.0-89.8%); LR+, 5.91 (95% CI, 3.73-9.39); LR-, 0.07 (95% CI, 0.02-0.18); DOR, 89.0 (95% CI, 22.8-348.1). MRI had a high predictive accuracy in assessing both the depth and topography of placental invasion. All five MRI signs showed good predictive accuracy in the diagnosis of disorders of invasive placentation. There was no difference in either the sensitivity (P = 0.24) or the specificity (P = 0.91) between ultrasound and MRI for the detection of invasive placentation. CONCLUSIONS Prenatal MRI is highly accurate in diagnosing disorders of invasive placentation. Ultrasound and MRI have comparable predictive accuracy. Large population-based studies are needed in order to assess whether ultrasound can predict the depth and topography of placental invasion as reliably as can MRI.