Résumé
Purpose: to assess the diagnostic accuracy of MRI in comparison with CT scan as a 2nd line examination of acute/subacute abdominopelvic pain in a prospective population of young women with a non-contributive ultrasound. The main objective was to demonstrate the non-inferiority of MRI diagnostic accuracy over CT in this context and to assess diagnostic performances of a conditional strategy using MRI rather than CT as a second line imaging procedure.Materials and Methods: this was a multicentric prospective non-inferiority comparative study between MRI and CT, conducted in French emergency departments, including non pregnant young women between 18 and 40 years who presented to the general emergency or emergency gynecologic department with non-traumatic abdominal pain lasting less than 5 days, with an inconclusive ultrasound, justifying prescription of additional CT scan. After written informed consent was obtained, CT and then, within 6 hours, abdomino-pelvic MR imaging were performed. Four trained radiologists specialized in emergency independently reviewed CT and MR images in a random order, blinded to the clinical outcome, using a standardized diagnoses list. A consensus review was done by a fifth trained radiologist using the same diagnosis list for CT and MRI in case of discrepancies.The gold standard or final diagnosis was done by a panel of experts after review of the patient's complete medical records and a telephone follow-up at 3 months. Statistical analysis was done to assess the non-inferiority regarding diagnostic accuracy of MRI compared to CT. The non-inferiority margin was set at 10%. Inter-observer agreement and diagnostic performance of a conditional imaging strategy (MRI then scanner when the MRI is normal) were estimated.Results: from June 2017 to July 2022, 149 patients were included and 133 patients were analyzed with a median age of 27 years (min-max: 18-40). The most common diagnoses were non-specific pain (30.1%), ovarian cyst rupture (12.8%), appendicitis (9.7%), other gynecologic causes (9.7%), and adnexal torsion (8.3%). 28 MRI was shown to have diagnostic accuracy of respectively, 84.2% (95% CI: 76.9-89.9) and 88% (95% CI: 81.2-93), when the right diagnosis was favored in MRI or in the consensual reading. In these two situations MRI was non inferior to CT. Results were similar in the per protocol and per protocol modified analysis. Inter-rater reliability value was substantial both for MRI 0.71 (95% CI: 0.63-0.79) and CT 0.78 (95% CI: 0.70-0.85). The conditional imaging strategy using MRI and CT when MRI is normal (non specific pain), had a performance accuracy of 91%.Conclusion: MRI, when it is carefully read, is not inferior to CT and might replace CT as a second line imaging modality in acute abdomen for young women if its accessibility is improved.