Ultrasonographic features can predict outcome of conservative management of acute appendicitis in children.

Abstract

To determine whether specific ultrasonographic features can predict failure of conservative treatment of acute appendicitis. A 2-year retrospective study was conducted on children admitted with acute appendicitis. Those with uncomplicated appendicitis diagnosed solely by ultrasound, and treated conservatively, were followed 18-24m to assess treatment outcome. Management was considered successful if recurrent acute appendicitis was not observed during follow-up. Appendix diameter, wall thickness, presence of mucosal ulceration, hyperechogenic fat, free fluid, and lymph nodes were evaluated as potential discriminatory ultrasonographic predictors. T-tests, chi-square, sensitivity, specificity, and odds ratios were calculated. Out of 556 consecutive patients that were admitted with acute appendicitis, 180 (32%) managed conservatively. One hundred eleven (62%) imaged by US only. Ninety-two out of 111 (83%) were followed 18-24m to assess treatment outcome, and 19/111 (17%) were lost to follow-up. Conservative management was successful in 72/92 (78.2%), with treatment failure in 20/92 (21.8%) (5/92 (5.4%) with recurrent acute appendicitis and 15/92 (16.3%) underwent appendectomy). Of the ultrasonographic features studied, mucosal ulceration demonstrated statistically significant predictive value. Fifteen out of 20 (75%) treatment failures had mucosal ulceration, compared to 21/72 (29.2%) of the patients with successful treatment (p < 0.001). This yielded a positive odds ratio of 7.3 (2.3-22.6, 95% CI), 70.8% (58.9-80.9, 95% CI) specificity, and 75% (50.9-91.3, 95% CI) sensitivity. Positive predictive value was 41.6% (31.5-52.5, 95% CI) while intact mucosa had negative predictive value of 91% (82.4-95.6, 95% CI) for conservative management success. The presence or absence of appendiceal mucosal ulceration at ultrasound can predict conservative management outcome in the setting of acute appendicitis, potentially improving pediatric patient selection for conservative management.