Complex US adnexal masses during pregnancy: Is pelvic MR imaging accurate for characterization?. Issue 93 (August 2017)

• Record Type: Journal Article
• Authors: Thomassin-Naggara, Isabelle; Fedida, Benjamin; Sadowski, Elisabeth; Chevrier, Marie-Claude; Chabbert-Buffet, Nathalie; Ballester, Marcos; Tavolaro, Sebastian; Darai, Emile
• Is Part Of: European journal of radiology. Issue 93(2017)
• Page Start: 200
• Page End: 208
• Publication Date: 2017-08
• Subjects: Pregnancy -- MRI -- Ovarian -- Complex; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2017.05.024
• Languages: English
• ISSNs: 0720-048X

Abstract

Highlights: MR imaging is highly accurate for the characterization of complex adnexal masses incidentally discovered during pregnancy. The sensitivities and specificities of prospective MRI were 66.7% (95% CI: 35.4%–87.9%) (6/9) and 96.3% (95% CI: 83.3%–99.4%). Utilizing MR imaging with the ADNEX MR-SCORE did not miss any cancers and showed an overall accuracy higher than 80%. Abstract: Objective: To retrospectively evaluate the accuracy of pelvic magnetic resonance (MR) imaging for the characterization of complex sonographic adnexal masses discovered in women during pregnancy. Study design: The study population comprised 31 pregnant women (median age: 32 years (range: 19–42); mean gestation age of 16 weeks) referred to our institution for MR imaging for characterization of an adnexal mass discovered incidentally during routine ultrasound (US) for other indications. The population comprised of 31 women, with 36 adnexal lesions (mean size: 103 mm [range: 20–290]), of which 27 were benign and 9 were malignant masses. Prospective US and MR reports were reviewed to determine the presence of a benign or malignant lesion. Two radiologists (R1 and R2) blinded to the final outcome, retrospectively evaluated the MR images using the criteria based on the ADNEX MR-SCORE and classified the lesion as benign or malignant. The reference standard was surgical pathology or at least a 1-year imaging follow-up. Results: Prospective US and MR imaging correctly identified the diagnosis in 27/36 (75%) (95% confidence interval (CI): 58.9%–86.2%) and in 32/36 (88.9%) (95% CI: 74.6%–95.6%) of lesions, respectively. MR imaging with ADNEX MR-SCORE allowed a correct diagnosis in 32/36 (88.9%) (95% CI: 74.6%–95.6%) of lesions for R1 and in 30/36 (83.3%) (95% CI: 68.1%–92.1%) of lesions for R2. The sensitivities and specificities of MR imaging using the MR ADNEX MR-SCORE were 100% (95% CI: 70.1%–1000%) for both readers and 85.1% (95% CI: 67.5%–94%) and 77.7% (95% CI: 59.2%–89.4%) for R1 and R2, respectively. No malignancy was classified as benign using MR criteria. The reproducibility between the two readers was almost perfect, with a kappa of 0.914. Conclusion: MR imaging is highly accurate for the characterization of complex adnexal masses incidentally discovered during pregnancy. Clinical relevance: MR imaging can accurately characterize adnexal masses in pregnancy and could be useful in opting for expectant management until delivery.