BMC Gastroenterology (Jun 2023)

Development of a scoring model based on objective factors to predict gangrenous/perforated appendicitis

  • Toshimichi Kobayashi,
  • Eiji Hidaka,
  • Itsuki Koganezawa,
  • Masashi Nakagawa,
  • Kei Yokozuka,
  • Shigeto Ochiai,
  • Takahiro Gunji,
  • Toru Sano,
  • Koichi Tomita,
  • Satoshi Tabuchi,
  • Naokazu Chiba,
  • Shigeyuki Kawachi

DOI
https://doi.org/10.1186/s12876-023-02767-7
Journal volume & issue
Vol. 23, no. 1
pp. 1 – 7

Abstract

Read online

Abstract Background The mortality rate of gangrenous/perforated appendicitis is higher than that of uncomplicated appendicitis. However, non-operative management of such patients is ineffective. This necessitates their careful exam at presentation to identify gangrenous/perforated appendicitis and aid surgical decision-making. Therefore, this study aimed to develop a new scoring model based on objective findings to predict gangrenous/perforated appendicitis in adults. Methods We retrospectively analyzed 151 patients with acute appendicitis who underwent emergency surgery between January 2014 and June 2021. We performed univariate and multivariate analyses to identify independent objective predictors of gangrenous/perforated appendicitis, and a new scoring model was developed based on logistic regression coefficients for independent predictors. Receiver operating characteristic (ROC) curve analysis and the Hosmer–Lemeshow test were performed to assess the discrimination and calibration of the model. Finally, the scores were classified into three categories based on the probability of gangrenous/perforated appendicitis. Results Among the 151 patients, 85 and 66 patients were diagnosed with gangrenous/perforated appendicitis and uncomplicated appendicitis, respectively. Using the multivariate analysis, C-reactive protein level, maximal outer diameter of the appendix, and presence of appendiceal fecalith were identified as independent predictors for developing gangrenous/perforated appendicitis. Our novel scoring model was developed based on three independent predictors and ranged from 0 to 3. The area under the ROC curve was 0.792 (95% confidence interval, 0.721—0.863), and the Hosmer–Lemeshow test showed a good calibration of the novel scoring model (P = 0.716). Three risk categories were classified: low, moderate, and high risk with probabilities of 30.9%, 63.8%, and 94.4%, respectively. Conclusions Our scoring model can objectively and reproducibly identify gangrenous/perforated appendicitis with good diagnostic accuracy and help in determining the degree of urgency and in making decisions about appendicitis management.

Keywords