Identifying reliable predictors of ovarian torsion in acute gynecological presentations: A retrospective case-control study

Objective: To identify clinical, laboratory, and imaging features that predict ovarian torsion in women undergoing laparoscopy for suspected adnexal torsion. Methods: This retrospective case-control study included 221 women aged 18 years and older who underwent laparoscopy for suspected ovarian torsion at Galilee Medical Center from 2011 to 2022. Demographic data, clinical presentation, imaging, laboratory findings, and intraoperative outcomes were extracted and compared between patients with confirmed torsion (n = 96) and those without (n = 125). Multivariable logistic regression was conducted to identify independent predictors. Results: Ovarian torsion was confirmed in 43.4 % of cases. Torsion was significantly associated with localized abdominal pain (82.3 % vs. 52.4 %, p < 0.001), vomiting (47.3 % vs. 20.0 %, p < 0.001), right-sided tenderness (51.0 % vs. 26.8 %, p = 0.002), and ovarian edema on ultrasound (18.0 % vs. 4.1 %, p = 0.001). Multivariable analysis identified four independent predictors: localized pain (OR 4.36, 95 % CI: 1.94–9.81, p < 0.001), vomiting (OR 2.38, 95 % CI: 1.16–4.89, p = 0.018), ovarian edema (OR 5.29, 95 % CI: 1.45–19.29, p = 0.012), and a non-significant trend toward higher White blood cells count (OR 1.09, 95 % CI: 0.99–1.21, p = 0.077). Conclusion: Localized pain, vomiting, and ultrasonographic ovarian edema independently predict ovarian torsion. Incorporating these markers into diagnostic pathways may improve early detection and guide surgical treatment. Prospective validation of predictive models is necessary.