Retrospective Multivariate Analysis of Data from Children with Suspected Appendicitis: A New Tool for Diagnosis

Background. Decision-making for management may sometimes be difcult in acute appendicitis (AA). Various diagnostic scoring systems exist, but their sensitivity and specifcity rates are far from ideal. In this study, the determination of the predictors and the efect of radiological data and developing a new scoring system were aimed. Methods. Medical records of patients who were hospitalized for AA between February 2012 and October 2016 were retrospectively reviewed. All data were compared between patients with and without appendicitis. Te multivariate analysis was performed to defne signifcant variables and to examine the sensitivity and specifcity of each group of predictors including radiological data. A new scoring system (NSS) was formed and was compared with two existing scoring systems: pediatric appendicitis score (PAS) and Alvarado scoring system (ASS) by using reclassifcation method. Results. Negative appendectomy rate was 11.3%. Statistical analysis identifed 21 independently signifcant variables. Te heel drop test had the highest odds ratio. Sensitivity and specifcity rates of clinical predictors were 84.6% and 94.8%, respectively. Radiological predictors increased the sensitivity rate to 86.9%. Sensitivity and specifcity rates for PAS, ASS, and NSS were 86.8% and 83.9%, 84.7% and 81.6%, and 96.8% and 95.6%, respectively.Te “re-assessed negative appendectomy rate” was 6.2% and false positive results were remarkably more common in patients with duration of symptoms less than 24 hours. Conclusion. Radiological data improves the accuracy of diagnosis. Containing detailed clinical and radiological data, NSS performs superiorly to PAS and ASS, regarding sensitivity and specifcity without any age limitation.Te efciency of NSS may be enhanced by determining diferent predictors for diferent phases of the infammatory process.