Development and Internal Validation of a New Predictive Score for Complicated Acute Appendicitis in a Tunisian Population

Abstract

Background: Distinguishing uncomplicated acute appendicitis (UAA) from complicated acute appendicitis (CAA) remains a major diagnostic challenge even after performing cross-sectional imaging with direct therapeutic implications. This study aimed to develop and validate a predictive score for CAA in a Tunisian clinical setting.
Methods: A retrospective study was conducted on 326 patients who underwent surgery for acute appendicitis. Clinical, biological, and radiological data collected at admission were analyzed using univariate and multivariate logistic regression to identify independent predictive factors for CAA. A predictive score was then constructed and its diagnostic performance assessed using the ROC curve.
Results: The prevalence of CAA was 43.6% (N = 142). Six independent predictive factors were identified: age ≥ 40 years (AOR = 3.1; p = 0.002), fever ≥ 38.5°C (AOR = 2.3; p = 0.02), leukocyte count ≥ 15,000/mm³ (AOR = 3.2; p = 0.002), CRP ≥ 100 mg/L (AOR = 7.4; p < 0.001), free fluid in the right iliac fossa or pelvis (AOR = 2.1; p = 0.039), and a collection in the right iliac fossa on imaging (AOR = 7.5; p < 0.001). The score ranged from 0 to 20, with a median of 4 (IQR 2–12). At the optimal cut-off of ≥ 6, it yielded a sensitivity of 80%, specificity of 90%, positive and negative predictive values of 85% each, and an overall accuracy of 84.9% (AUROC = 0.90).
Conclusion: This new predictive score, incorporating six clinical, biological, and radiological parameters available at admission, provides a ....(abstract truncated at 250 words).