Clinical judgment is superior to diagnostic tests in the management of pediatric small bowel injury

Traumatic solid organ injuries are easily recognizable on computed tomography (CT) scans and usually are treated nonoperatively. Small bowel injuries may be difficult to diagnose and require prompt operation. This study was done to assess the role of clinical examination versus diagnostic tests in evaluating these injuries. The medical records of all pediatric (< or = 18 years old) patients treated at a pediatric trauma center from 1984 to 1995 were reviewed. Statistical analysis was performed using SAS software, with P values of less than .05 considered significant. Small bowel injury occurred in 48 patients (21 blunt, 27 penetrating). Most blunt injuries were automobile-related (11 patients) or attributable to recreational activities (4) or bicycle accidents (2). Penetrating injuries were primarily caused by assaults with guns (21) or knives (4). All conscious patients with small bowel injury had abnormal physical examination findings at the time of presentation. Nineteen patients had generalized peritonitis, and 14 had localized abdominal tenderness. The serum amylase level was abnormal in 2 of 18 cases. Abdominal CT scans were obtained in six patients and showed the injury in only three. Peritoneal lavage (DPL), performed in 10 patients, led to the diagnosis in five. There was no significant difference in the complication rate (30%) between patients operated on immediately because of a diagnostic test result and those operated on later, after a period of clinical observation (P = 1.0, Fisher's Exact test). Associated injuries were common (60%) among both blunt and penetrating cases. In this nonoperative era of pediatric trauma care, small bowel injury is best diagnosed clinically. The physical examination is 100% sensitive in the conscious patient, and specificity is achieved by serial examination. Serum amylase, CT scan, and DPL are not reliable diagnostic tests to exclude these injuries. Patients can be observed until physical findings suggest bowel injury without increased morbidity. Associated injuries are common; thus, patients are best treated where multidisciplinary support is available.