Abstracts

RETURN TO ABSTRACT LISTING


E1703. Evolving Utilization of Cervical Spine Imaging in the Emergency Department: A Novel Injury Severity Score Based Analysis
Authors
  1. Akram Khaja; Department of Radiology and Imaging Sciences, Emory University School of Medicine
  2. Michal Horný; Department of Radiology and Imaging Sciences, Emory University School of Medicine
  3. Patricia Balthazar; Department of Radiology and Imaging Sciences, Emory University School of Medicine
  4. Tarek Hanna; Department of Radiology and Imaging Sciences, Emory University School of Medicine
  5. Paul Harkey; Department of Radiology and Imaging Sciences, Emory University School of Medicine
  6. Alexander Villalobos; Department of Radiology and Imaging Sciences, Emory University School of Medicine
  7. Richard Duszak; Department of Radiology and Imaging Sciences, Emory University School of Medicine
Objective:
In recent decades, the overall use of cervical spine imaging in the emergency department (ED) setting has increased, with a disproportionate increase in computed tomography (CT) over radiography. Much of this is likely in the setting of trauma. Adherence to validated clinical decision rules intended to ameliorate inappropriate imaging has been variable, with speculation of disproportionate inappropriate imaging use in the setting of minor trauma. We therefore studied evolving patterns of cervical spine imaging in ED patients specifically presenting with trauma focusing on modality, geography, and severity of injury.

Materials and Methods:
Using 2009 to 2015 IBM® MarketScan® Commercial Database, we identified all ED trauma encounters, associated cervical spine imaging services, and all related diagnosis codes. Using an International Classification of Disease (ICD), 9th revision, to Injury Severity Score (ISS) conversion tool validated in the surgical literature, we calculated an ISS for each ED encounter up through the end of ICD-9 in the third quarter of 2015 and studied imaging utilization trends and correlations with injury severity.

Results:
Between 2009 Q1 and 2015 Q3, a total of 38,668,483 ED visits were identified in our dataset, with 9,183,237 (23.7%) involving trauma. Of all trauma encounters, 6,313,013 (68.7%) had an ISS < 3 (“minor trauma”) and the remaining 2,870,224 had an ISS > 3 (“non-minor trauma”). Cervical spine imaging use per 100 trauma encounters increased 51.3% (from 2.9% to 4.4%). Radiography use increased 3.8% (from 1.5% to 1.5%) and CT use increased 100.5% (from 1.4% to 2.8%). In patients with minor injuries, overall cervical spine imaging use increased 39.7% (from 1.3% to 1.8%), radiography use increased 17.0% (from 0.9% to 1.1%), and CT use increased 92.6% (from 0.4% to 0.8%). In patients with non-minor injuries, overall cervical spine imaging use increased 61.3% (from 1.5% to 2.5%), radiography use decreased 18.7% (from 0.5% to 0.4%), and CT use increased 103.7% (from 1.0% to 2.0%). Despite the overall national increase in cervical spine imaging over time, utilization declined in 16 states. The greatest increase was observed in Alabama (+328% overall; +268% in minor and +387.6% in non-minor trauma). The greatest decline was observed in West Virginia (-38.7% overall; -43.4% in minor and -35.2% in non-minor trauma).

Conclusion:
In the setting of trauma, ED cervical spine imaging use varies markedly at the state level. Substantial increases in use have been driven largely by increased CT usage. Contrary to speculation about disproportionate inappropriate imaging use in the setting of minor trauma, increases were greatest in the setting of non-minor trauma. Our novel use of an ICD-derived ISS imaging utilization stratification tool demonstrates its feasibility for use in this and related future trauma imaging utilization research.