A Case-Control Study of Sonographic Maximum Ovarian Diameter as a Predictor of Ovarian Torsion in Emergency Department Females with Pelvic Pain

Author Department

Emergency Medicine

Document Type

Article, Peer-reviewed

Publication Date

7-2018

Abstract

BACKGROUND:

Color and power doppler ultrasound are commonly used in the evaluation of ovarian torsion but are unreliable. Because normal sized ovaries are unlikely to cause torsion, maximum ovarian diameter (MOD) could theoretically be used as a screening test in the ED. Identification of MOD values below which torsion is unlikely would be of benefit to providers interpreting radiology department or point-of care pelvic ultrasound.

OBJECTIVES:

To determine if sonographic maximum ovarian diameter (MOD) can be used as a screening tool to rule out torsion in selected patients.

METHODS:

Via a retrospective case-control study spanning a 14-year period, we examined the ultrasound characteristics of patients with torsion and age-matched controls, all presenting to the emergency department with lower abdominal pain and receiving a radiology department pelvic ultrasound for "rule-out torsion." Standardized data collection forms were utilized. Distributions of MOD were compared and sensitivity, specificity, and likelihood ratios were calculated for multiple cut-offs.

RESULTS:

We identified 92 cases of surgically confirmed ovarian torsion and selected 92 age-matched controls. In post-menarchal patients the sensitivity, specificity, +likelihood ratio, and -likelilhood ratio of a 3cm and 5cm MOD were 100% (96-100%), 30% (20-41%), 1.4 (1.3-1.7), 0, and 91% (83-97%), 92% (83-97%), 11.2 (5.5-22.9), .09 (.04-.19) respectively. The 5cm MOD, however, excluded an additional 52/84 (62%) of post-menarchal patients.

CONCLUSIONS:

A threshold MOD of 5cm on pelvic ultrasound may be useful to rule out ovarian torsion in post-menarchal females presenting with lower abdominal and pelvic pain. This article is protected by copyright. All rights reserved.

PMID

30044031

Share

COinS