A validated predictive model for adnexal torsion pre-operative diagnosis

Area of Science:

• Gynecology
• Surgical Oncology
• Emergency Medicine

Background:

• Adnexal torsion (AT) is a gynecologic emergency requiring prompt diagnosis.
• Accurate pre-operative diagnosis of AT can be challenging, potentially leading to delays in treatment.
• Developing simple, effective predictive tools is crucial for improving patient outcomes.

Purpose of the Study:

• To develop and validate a straightforward predictive model for the pre-operative diagnosis of adnexal torsion (AT).
• To identify key clinical and sonographic parameters associated with AT.
• To enhance the accuracy of pre-operative AT diagnosis in women undergoing laparoscopy.

Main Methods:

• Retrospective cohort study including 669 women undergoing laparoscopy for suspected AT (January 2011-June 2020).
• Comparison of pre-operative characteristics between women with and without surgically confirmed AT.
• Multivariate regression analysis to identify independent predictors of AT.

Main Results:

• The study identified vomiting, a neutrophil-to-lymphocyte ratio > 3.5, and sonographic finding of an enlarged ovary as independent predictors of AT.
• Positive predictive values (PPV) for AT diagnosis increased with the number of risk factors present, reaching 93.1% in the derivation cohort.
• The validated model demonstrated good predictive performance, with PPVs ranging from 66.6% to 82.6% based on the number of risk factors.

Conclusions:

• A simple predictive model incorporating vomiting, neutrophil-to-lymphocyte ratio, and ovarian size has been developed and validated for pre-operative AT diagnosis.
• This model can assist clinicians in evaluating patients with suspected AT, potentially improving diagnostic accuracy.
• Implementation of this model may lead to more timely interventions and better patient outcomes for adnexal torsion.