The utilization of computed tomography in the emergency department and appropriateness as determined by the american college of radiology appropriateness criteria: 2000 versus 2003

Abstract

Study objectives: We compare the change in use of computed tomography (CT) in the emergency department (ED) overall and by CT scan type for 2000 versus 2003. CT scan use was assessed for appropriateness using the American College of Radiology (ACR) Appropriateness Criteria. Methods: We performed a retrospective study of all CT scans ordered on ED for the first quarters of 2000 and 2003 at an urban Level I trauma center with 54,000 emergency visits per year. Information was obtained from the hospital information services database. A retrospective assessment of appropriateness of all CT scans ordered in the ED for a corresponding week from each period (March 20 to 26, 2000, and March 23 to 29, 2003) was performed using the ACR Appropriateness Criteria. Results: The overall ED CT scan use rates for 2000 and 2003 were 10 CT scans per 100 patient visits and 19 per 100, respectively. The 2000 versus 2003 rates for discharged were 6 per 100 and 12 per 100 and for admitted were 19 per 100 and 37 per 100. CT scan use by CT type increased from 2000 to 2003 as follows: head CT 45%, abdomen 114%, chest 331%, cervical spine 1077%, and other 67%. Head and abdominal CT scans accounted for the majority of CT scans performed in 2000 (89%) and 2003 (77%). The appropriateness determination using the ACR Appropriateness Criteria demonstrated the following distribution: 61%, 14%, 14%, and 11%. The CT scans covered by the ACR Criteria (Appropriate and categories) for 2000 versus 2003 revealed no significant difference in scan appropriateness ( P =.981). The specific ACR guidelines used included Neurology 42% (109/262), Urology 16% (41/262), Cardiovascular 8% (22/262), and Gastrointestinal 9% (24/262). Cardiovascular and Gastrointestinal guideline use demonstrated a significant change from 2000 to 2003 (2/22 versus 20/22, P =.004; and 1/24 versus 23/24, P =.0005, respectively). Neurology and Urology guideline use and and categories demonstrated no significant change in use. Of the Inappropriate CT scans in 2000 all were head CTs (12/12), 50% (6/12) were performed for possible transient ischemic attack (ACR level=4), and 33% (4/12) were for new-onset vertigo in elder patients (ACR level=4). Those considered Inappropriate in 2003 were primarily head CTs (88% [21/24]), with 58% (14/24) performed for possible transient ischemic attack, 8% (2/24) for new-onset vertigo in elder patients, and 1 for classic appendicitis (ACR level=4). Most Not Covered in 2000 and 2003 were with acute change in mental status (75.9% [22/29]). Conclusion: CT scan use at our institution nearly doubled between 2000 and 2003. Although all CT scan types increased in use, the greatest increases were seen in cervical spine and chest. Head and abdominal CT scans constituted the majority (81%) of scans. The majority of CT scans performed were considered Appropriate or fell outside the ACR criteria as a part of either an established Trauma Protocol or Not Covered by the criteria. Fourteen percent of scans were Inappropriate by ACR criteria. Over 50% of these were performed for possible transient ischemic attack (ACR level=4) and illustrate a disparity between our practice and the ACR guidelines, with the current standard of care in emergency medicine recommending the performance of an emergency CT in these cases. Cardiovascular and Gastrointestinal ACR guideline use increased significantly.