急性虫垂炎患者（40歳以上）における偶発的な虫垂腫瘍の予測のための新しいスコア：多施設共同後ろ向きコホート研究

This multicenter retrospective cohort study aimed to develop a novel five-parameter scoring model to predict incidental appendiceal neoplasia in patients aged ≥40 years diagnosed with acute appendicitis. Previous literature has reported a significant increase in the risk of neoplasia, particularly after age 40, and this cut-off value was used as the basis for our study design.

A multicenter retrospective cohort analysis was conducted across six tertiary hospitals between January 2019 and December 2024. Adult patients aged ≥40 years who underwent appendectomy with preoperative contrast-enhanced computed tomography (CT) and had available laboratory data were included. Predictive variables were identified using multivariate logistic regression. The scoring system incorporated age, female sex, appendix diameter, absence of CT wall enhancement, and low neutrophil count. Diagnostic performance was assessed via receiver operating characteristic (ROC) analysis.

Of 2,143 patients analyzed, 122 (5.7%) had incidental neoplasia. The scoring system yielded an area under the ROC curve of 0.641 (95% confidence interval: 0.594-0.690). Using a cutoff score ≥3, sensitivity was 56.6%, specificity 65.4%, positive predictive value 9%, and negative predictive value 96.1%. Patients with a maximum score of 5 had a 33.3% incidence of neoplasia.

This five-parameter scoring system demonstrates a high negative predictive value and may help identify low-risk patients in whom standard appendectomy can be safely performed. This model, which allows low-risk patients to be safely excluded thanks to its high negative predictive value, provides an innovative contribution to clinical decision-support systems.