Head trauma: CT scan interpretation by radiology residents versus staff radiologists.

Abstract

To determine the rate and clinical outcome of discrepancies in interpretation by radiology residents and staff neuroradiologists of posttraumatic cranial computed tomographic (CT) scans. Prospective evaluation was performed for 419 consecutive emergency posttraumatic cranial CT studies that had been interpreted by radiology residents on call over a 16-month period. Discrepancies between the interpretations made by residents and those made by staff radiologists were divided into two groups: failure to recognize an abnormality (false-negative finding) and interpretation of normal as abnormal (false-positive finding). Discrepancies were considered major if they could affect patient care in the emergency setting and minor if they could not. Major and minor discrepancies were 1.7% and 2.6%, respectively, among interpretations made by residents and those by staff radiologists. Major discrepancies were four subdural hematomas, one pneumocephalus, one hemorrhagic contusion, and one subarachnoid hemorrhage. Minor discrepancies included six skull and five facial fractures. The discrepancy rate was statistically significantly higher (12.2%) when CT findings were abnormal than when they were normal (1.5%). No change in treatment was attributed to the delay in diagnosis. A low discrepancy rate was found between interpretations made by radiology residents and those made by staff neuroradiologists of posttraumatic cranial CT scans. There were no adverse clinical outcomes.