Ultrasound and laboratory parameters in distinguishing complicated from uncomplicated appendicitis in children

Abstract

Introduction/Objective. To evaluate sonographic and laboratory findings as predictors of complicated and uncomplicated appendicitis in order to decide on further treatment options. Methods. This is a retrospective cohort study of 174 pediatric patients who had laboratory tests and ultrasound done before appendectomy in one year period. Results were compared with the intraoperative and histopathological findings of complicated (gangrenous or perforated) or uncomplicated (phlegmonous) appendicitis and assessed by binary logistic regression with backward elimination. The initial model included eight predictors. After backward elimination four remained: periappendiceal fluid, hyperechoic periappendiceal fat, white blood cell (WBC) count and C-reactive protein (CRP). The final model included the interaction between periappendiceal fluid and hyperechoic periappendiceal fat. Diagnostic performance of each parameter was presented with sensitivity and specificity. Results. Out of all patients, 86 had uncomplicated and 88 had complicated appendicitis (37 gangrenous and 51 perforated). In the final model three predictors were significantly associated with complicated appendicitis: interaction between periappendiceal fluid and hyperechoic periappendiceal fat, WBC count > 11 ? 109 / l, and CRP > 100 mg/l. Inclusion of interaction between periappendiceal fluid and hyperechoic periappendiceal fat excluded them as individual predictors. The maximum outside appendiceal diameter of more than 6mm had the highest sensitivity (93.2%), while wall thickness > 3 mm was the most specific (95.2%). Conclusion. Using periapendiceal fluid and hyperechoic periappendiceal fat as sonographic predictors and WBC and CRP as laboratory predictors can differentiate uncomplicated from complicated appendicitis in children and help a physician decide on antibiotic or surgical treatment.