Prospective Pilot Derivation of a Decision Tool for Children at Low Risk for Testicular Torsion

Objective The purpose of this study was to derive a pilot clinical decision tool with 100% negative predictive value for testicular torsion based on prospectively collected data in children with acute scrotal pain. Methods This was a prospective cohort study of a convenience sample of newborn to 21‐year‐old males evaluated for acute (72 hours or less) scrotal pain at an urban children's hospital emergency department ( ED ). A pediatric emergency medicine fellow or attending physician documented history and examination findings on a standardized data collection form. The study investigators used ultrasound ( US ), operative reports, or clinical follow‐up to identify patients who had testicular torsion. Pearson's chi‐square test and odds ratios ( OR ) were used to identify factors associated with the diagnosis of testicular torsion. The authors also used a recursive partitioning model to create a low‐risk decision tool for testicular torsion. Results Of the 450 eligible patients, 228 (51%) were enrolled, with a mean (± SD ) age of 9.9 (±4.1) years, including 21 (9.2%, 95% confidence interval [ CI ] = 5.8% to 13.7%) with testicular torsion. The derived clinical decision tool consisted of three variables: horizontal or inguinal testicular lie ( OR  = 18.17, 95% CI  = 6.2 to 53.2), nausea or vomiting ( OR  = 5.63, 95% CI  = 2.08 to 15.22), and age 11 to 21 years ( OR  = 3.9, 95% CI  = 1.27 to 11.97). These variables had a sensitivity of 100% (95% CI  = 98% to 100%) and negative predictive value of 100% (95% CI  = 98% to 100%) for the diagnosis of testicular torsion. Conclusions Based on a decision tool derived with recursive partitioning, study patients with all of the following characteristics had no risk of testicular torsion: normal testicular lie, lack of nausea or vomiting, and age 0 to 10 years. Future research should focus on externally validating this tool to optimize emergent evaluation when testicular torsion is likely, while minimizing routine sonographic evaluation when patients are unlikely to have a serious condition requiring immediate management.