Objective: To prospectively validate the Pediatric Appendicitis Score (PAS), developed on a cohort of children with abdominal pain suggestive of appendicitis, in unselected children with abdominal pain who present to the emergency department.
Study design: Over a 19-month period, we prospectively recruited children 1 to 17 years old who came to our tertiary pediatric emergency department, with a chief complaint of abdominal pain of duration less than 7 days. PAS components included fever >38 degrees C, anorexia, nausea/vomiting, cough/percussion/hopping tenderness (2 points), right-lower-quadrant tenderness (2 points), migration of pain, leukocytosis >10 000 cells/mm(3), and polymorphonuclear neutrophilia > 7500 cells/mm(3). A follow-up call was made to verify final outcome. Sensitivity, specificity, and the receiver operating characteristic curve of the PAS with respect to diagnosis of appendicitis were calculated.
Results: We collected data on 849 children. 123 (14.5%) had pathologic study-proven appendicitis. Mean (median, range) score for children with appendicitis and without appendicitis was 7.0 (7, 2-10) and 1.9 (1, 0-9), respectively. If a cutoff PAS of <or=2 was used to discharge patients without further investigation, only 3 (2.4%) with appendicitis would be sent home. If a PAS of >or=7 was used to take children to the operating room without further investigation, only 29 (4%) would not have appendicitis. For the PAS the area under the receiver operator curve was 0.95.
Conclusions: The PAS is useful, because a value <or=2 (found in 73% of children without appendicitis) has high validity for ruling out appendicitis, and a score >or=7 (found in 61% of children with appendicitis) has a high validity for predicting the presence of appendicitis. Children with PAS of 3 to 6 (37% with appendicitis and 23% without appendicitis in this study) should undergo further investigation such as observation, ultrasonography, or computed tomography.