E5015. Diagnostic Accuracy of the Follicular Ring Sign in the Sonographic Diagnosis of Ovarian Torsion
Authors
Rabab Al Mansoori;
McMaster University
Jonathan Bellini;
McMaster University
Ankush Jajodia;
McMaster University
Stefanie Lee;
McMaster University
Objective:
This study aims to assess the diagnostic accuracy of the sonographic follicular ring sign (FRS) among adult female emergency room patients receiving pelvic ultrasound for suspected ovarian torsion.
Materials and Methods:
A retrospective study was conducted of all adult female patients presenting with pelvic pain to emergency room departments at all Hamilton Health Sciences sites who received pelvic ultrasound for suspicion of ovarian torsion over a 5-year period from 2014–2019. Exclusion criteria included patients who were not biologically female at birth and patients who had bilateral oophorectomy. Clinical, surgical, and pathologic reports for a minimum of 1 year following presentation were retrieved from the EMR and reviewed to determine the clinical outcome as the reference standard. Deidentified ultrasound images of all patients who went to surgery, as well as a control subset of negatives, were independently reviewed by two readers (blinded to clinical outcome) for the presence or absence of the FRS. Discrepancies were resolved by an experienced third reader.
Results:
There were 1566 patients who met the inclusion and exclusion criteria and were reviewed for clinical outcomes. Of the 44 patients who underwent surgery, 27 cases were positive for ovarian torsion, and 17 cases were negative. Alternate diagnoses included hemorrhagic cysts, tubo-ovarian abscesses, or endometriosis. The sensitivity of FRS among surgically confirmed cases was 48.2% (95% CI: 28.7 to 68.1%), with a positive predictive value of 6.7%, due to the low prevalence. There were 128 patients who did not go on to surgery reviewed as a control, and all were negative for FRS. The combined specificity was 98.6% (95% CI: 95.1–99.8%), with an NPV of 99%. The sonographic FRS has a high specificity and moderate accuracy (63.6%) in predicting ovarian torsion in adult patients.
Conclusion:
The FRS is a relatively discrete sonographic finding that may improve accuracy in diagnosis of ovarian torsion in adult emergency patients, in combination with other sonographic features.
