A clinical prediction model for adnexal torsion in pediatric and adolescent population

Raanan Meyer*, Nir Meller, Aya Mohr-Sasson, Shlomo Toussia-Cohen, Daphna Amitai Komem, Roy Mashiach, Gabriel Levin

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Purpose: To describe the clinical characteristics of children and adolescents that underwent diagnostic laparoscopy for suspected adnexal torsion (AT), and to develop a prediction model for preoperative detection of AT among young women. Methods: A retrospective cohort study. We included all girls ≤18 years old with clinically suspected AT who underwent a diagnostic laparoscopy between 3/2011 and 6/2020. We compared patients with AT to those without AT and constructed a prediction model. Results: Overall, 120 children and adolescents with suspected AT were included in the study. Of those, AT was identified in 83 (69.2%). In a multivariate analysis, the following risk factors were independently associated with AT and included in the prediction model: absence of right lower quadrant tenderness upon examination [adjusted odds ratio (aOR) (95% Confidence interval (CI)) 3.23 (1.23–8.47), p = 0.017], platelets level >240 K [aOR (95% CI) 3.15 (1.19–8.36), p = 0.021], and neutrophils level >5.4 [aOR (95% CI) 2.71 (1.02–7.52), p = 0.046]. The rate of AT was 12.5% in cases without risk factors for AT, 56.7% with one, 68.8% with two, and 94.1% with three risk factors present, respectively. Conclusions: We have identified preoperative indicators independently associated with surgically confirmed AT in a large cohort of young women.

Original languageEnglish
Pages (from-to)497-501
Number of pages5
JournalJournal of Pediatric Surgery
Issue number3
StatePublished - Mar 2022


  • Abdominal pain
  • Adolescents
  • Children
  • Diagnostic laparoscopy
  • Prediction
  • Ultrasound


Dive into the research topics of 'A clinical prediction model for adnexal torsion in pediatric and adolescent population'. Together they form a unique fingerprint.

Cite this