Identification of blunt abdominal injuries in children.

BACKGROUND The use of computed tomography (CT) to screen for injuries in pediatric blunt abdominal trauma (BAT) is increasing, concurrent with increasing concern over long-term risk of radiation-associated malignancies. We proposed to determine features that could be identified in the early assessment of these patients, which can predict the likelihood of clinically important intra-abdominal injuries warranting imaging by CT. We further queried if these were discrepant from factors associated with the decision to obtain an abdominal CT. METHODS Data of patients admitted with BAT to one of two Level I pediatric trauma centers were reviewed retrospectively. Clinical, laboratory, radiographic, and epidemiologic data were collected. Logistic regression was used to determine associations between pre-CT findings and ultimate diagnoses of "notable" or "clinically important" intra-abdominal injuries. Similar analyses were performed to determine which findings were associated with actually receiving an abdominal CT scan. RESULTS Of 571 patients, 37% had a notable intra-abdominal injury and 18% a clinically important intra-abdominal injury. After adjusting for all covariates, hematuria (gross or microscopic), elevated serum alanine aminotransferase, and documentation of clinically concerning abdominal findings upon examination remained significant predictors (odds ratio (OR), 3.5; 95% confidence interval [CI], 1.8-6.8; OR, 10.9; 95% CI, 2.5-47, respectively) of a clinically important injury. Undergoing a CT head and the presence of hematuria were significantly associated with obtaining a CT of the abdomen (OR, 3.4; 95% CI, 1.5-7.7; OR, 2.9; 95% CI, 1.1-7.3, respectively), while concerning abdominal findings and decreased Glasgow Coma Scale (GCS) score were not. CONCLUSION Clinical variables may be used to predict intra-abdominal injuries after pediatric BAT that may warrant imaging with CT scanning. Combined with findings from similar studies, it may be possible to derive and validate a decision-making rule both sensitive and specific in predicting the need for abdominal CT scanning in these patients. LEVEL OF EVIDENCE Prognostic study, level III.