Risk Stratification of Adnexal Cysts and Cystic Masses: Clinical Performance of Society of Radiologists in Ultrasound Guidelines.

Abstract

Purpose To evaluate the performance of the 2010 Society of Radiologists in Ultrasound (SRU) consensus guidelines in the risk stratification of symptomatic and asymptomatic adnexal cysts. Materials and Methods An institutional review board-approved retrospective review was performed, including adnexal cysts detected with ultrasonography (US) with surgical diagnosis or at least 2 years of imaging or clinical follow-up, from January to June 2011. SRU management recommendations were scored as 0, no follow-up; 1, US follow-up; 2, magnetic resonance (MR) imaging follow-up; and 3, surgical evaluation. Distribution of outcomes (nonneoplastic cyst, benign neoplasm, malignant neoplasm) was compared in each rating group by using the Cochran-Armitage trend test. Where SRU guidelines allow more than one management option, they were classified as being interpreted in either a "surgically focused" environment, with limited MR imaging availability, or an "MR-capable" center, where MR imaging is selected whenever it is an option. Predictors of neoplasms and malignancy were evaluated by using multivariate logistic regression. Results A total of 570 cysts in 500 women aged 18-90 years (mean, 42 years) were included. There were 475 (83.3%) nonneoplastic cysts, 77 (13.5%) benign neoplasms, and 18 (3.2%) malignant neoplasms. Of the 500 women, 161 (32.2%) were asymptomatic. In the surgically focused interpretation of guidelines, proportions of any neoplasm and malignant neoplasm, respectively, were 1% and 0% in SRU 0, 17% and 1% in SRU 1, 48% and 0% in SRU 2, and 48% and 16% in SRU 3 (P < .0001 for both trends). In the interpretation of SRU guidelines with MR imaging when it was an option, proportions of any neoplasm and malignant neoplasm, respectively, were 1% and 0% in SRU 0, 17% and 1% in SRU 1, 38% and 5% in SRU 2, and 81% and 52% in SRU 3 (P < .0001, both trends) and 82 (89.1%) fewer benign cysts would have gone directly to surgical evaluation. In multivariate regression, SRU rating predicted both any neoplasm (odds ratio, 2.58; P < .0001) and malignant neoplasm (odds ratio, 4.94; P = .005). Conclusion SRU consensus guidelines effectively stratified the risk of neoplasia and malignancy. Selecting MR imaging when it is an option in the guidelines would have reduced the number of benign cysts sent for surgical evaluation. © RSNA, 2017.