Objective: To assess the predictive value of Doppler sonography in the diagnosis of ovarian torsion and to correlate Doppler results with surgical findings and various clinical characteristics.
Design: Retrospective study of discharged inpatients.
Setting: An academic community hospital.
Patient(s): Twenty-one patients with surgically confirmed ovarian torsion over an 8-year period.
Intervention(s): Data were collected on Doppler flow results, ultrasound and surgical findings, patient characteristics, and associated morbidity.
Main outcome measure(s): Accuracy of Doppler diagnosis as to presence of ovarian torsion.
Result(s): Twenty-one patients had surgically confirmed ovarian torsion. Doppler sonography was performed in 10 of the 21 patients. Doppler sonographic findings were normal in 60% (6 of 10), and abnormal (decreased or absent) in 40% suggestive of torsion. In cases involving ovulation induction, Doppler sonography findings were normal in 25% (1 of 4). Furthermore, the time to diagnosis of ovarian torsion (mean = 5.3 hours) and the time to hospital discharge (mean = 2 days) were both decreased when compared with instances when normal flow was detected by Doppler sonography (59 hours and 2.7 days, respectively).
Conclusion(s): Abnormal flow detected by Doppler sonography is highly predictive of adnexal torsion and is therefore useful in the diagnosis of ovarian torsion. However, when normal flow is detected by Doppler sonography, it does not necessarily exclude an ovarian torsion; in fact, torsion is missed in 60% of cases, and time to diagnosis in these cases is delayed. In cases of ovulation induction, sensitivity is increased to 75%.