INTRODUCTION: The initial nonoperative management of perforated appendicitis fails in 15% to 25% of children. These children have complications and increased hospitalization. The purpose of this study was to identify predictors of failure. METHODS: Children with perforated appendicitis treated with antibiotics and intent for nonoperative management over a 4-year period were reviewed. Seventy-five children were identified and included in the study. Failure was defined as undergoing appendectomy before the initially planned interval. RESULTS: Nine (12%) of the patients required appendectomy sooner than initially planned. Age, presenting symptoms, physical examination findings, and white blood cell (WBC) count were similar in both success and failure groups. Absence of abscess and presence of appendicolith were both predictors of failure in a multivariate analysis, which included the presence of small bowel obstruction. The failed group had a longer median total length of stay (18 days [range, 4-67] vs 8 days [range, 4-31]; P = .002) and underwent 3 times as many computed tomography scans as successes (3 [range, 2-7] vs 1 [range, 0-5]; P < .001). CONCLUSION: Lack of abscess and presence of an appendicolith predict failure of nonoperative management of perforated appendicitis in children even when the effect of small bowel obstruction is accounted for. Children with these characteristics may benefit from alternative management strategies.
INTRODUCTION: The initial nonoperative management of perforated appendicitis fails in 15% to 25% of children. These children have complications and increased hospitalization. The purpose of this study was to identify predictors of failure. METHODS:Children with perforated appendicitis treated with antibiotics and intent for nonoperative management over a 4-year period were reviewed. Seventy-five children were identified and included in the study. Failure was defined as undergoing appendectomy before the initially planned interval. RESULTS: Nine (12%) of the patients required appendectomy sooner than initially planned. Age, presenting symptoms, physical examination findings, and white blood cell (WBC) count were similar in both success and failure groups. Absence of abscess and presence of appendicolith were both predictors of failure in a multivariate analysis, which included the presence of small bowel obstruction. The failed group had a longer median total length of stay (18 days [range, 4-67] vs 8 days [range, 4-31]; P = .002) and underwent 3 times as many computed tomography scans as successes (3 [range, 2-7] vs 1 [range, 0-5]; P < .001). CONCLUSION: Lack of abscess and presence of an appendicolith predict failure of nonoperative management of perforated appendicitis in children even when the effect of small bowel obstruction is accounted for. Children with these characteristics may benefit from alternative management strategies.