A Dual-Model Strategy for Pediatric Adnexal Torsion: Nomograms for Diagnosis and Necrosis Using Emerging...

To develop and validate preoperative nomograms for predicting pediatric adnexal torsion (AT) and post-torsion adnexal necrosis using clinical, laboratory, and ultrasonographic parameters.

Retrospective cohort study.

Single-center medical institution (Tianjin Children's Hospital).

A total of 186 girls (≤18 years) with suspected AT who underwent surgical exploration between March 2019 and January 2025.

Patients were randomly divided into training (n = 130) and test (n = 56) cohorts. Candidate variables including demographic characteristics, symptoms, laboratory indices, and ultrasound findings were screened using least absolute shrinkage and selection operator (LASSO) regression. Multivariable logistic regression models were constructed to predict AT and, among confirmed AT cases, adnexal necrosis. Model performance was evaluated by area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis.

Adnexal torsion was confirmed in 67 patients (36.0%), of whom 21 (31.3%) had pathological adnexal necrosis. For AT prediction, age, abdominal tenderness, systemic immune-inflammation index (SII), and the ultrasonographic whirlpool sign were identified as independent predictors and incorporated into a nomogram. This model demonstrated good discrimination (AUC 0.933 in the training cohort and 0.891 in the test cohort) with satisfactory calibration. Among patients with confirmed AT, pan-immune-inflammation value (PIV) and thrombin time (TT) independently predicted adnexal necrosis and were used to construct a separate necrosis nomogram, achieving an AUC of 0.811. Both models provided favorable net clinical benefit across clinically relevant threshold probabilities.

These nomograms provide a clinically accessible framework for preoperative risk stratification of pediatric AT and adnexal necrosis, leveraging routine clinical, laboratory, ultrasonographic, and emerging inflammatory indices.