Shock/Sepsis/Trauma/Critical Care
Variability in CT imaging of blunt trauma among ED physicians, surgical residents, and trauma surgeons

An abstract of this work was presented as a poster at the 75th Annual Meeting of AAST and Clinical Congress of Acute Care Surgery held in Waikoloa, Hawaii.
https://doi.org/10.1016/j.jss.2017.02.015Get rights and content

Abstract

Background

Trauma triage decisions can be influenced by both knowledge and experience. Consequently, there may be substantial variability in computed tomography (CT) scans desired by emergency medicine physicians, surgical chief residents, and attending trauma surgeons. We quantified this difference and studied the effects of each group's decisions on missed injuries, cost, and radiation exposure.

Methods

All blunt trauma activations at an urban level 1 trauma center were studied over a 6-mo period. Three months into the study, a pan-scan protocol was introduced. Prior to CT imaging, providers separately completed a survey that asked which CT scans were desired for each patient. Based on the completed surveys, hypothetical missed injuries, radiation exposure, and cost were determined.

Results

The variability in the number of CT scans desired by each of the three providers and the resulting cost and radiation exposure were not statistically significant. Substantial variability was predominantly seen in the indications for the desired scans, with the difference between proportions ranging from 3.1%-68.7%. Agreement among the three providers was highest for head and c-spine scans (80%-100%) and lowest for maxillary face (57%-80%) and chest scans (52%-74%). Overall, the missed injury rate was similar for all the providers; chief residents missed significantly more major injuries than trauma attendings during the pan-scan period (P = 0.03).

Conclusions

Trauma training and level of training did not have a substantial effect on radiological decisions during the initial trauma assessment. This study sheds light on the growing uniformity among providers with regard to medical decision-making in the initial work-up of trauma.

Introduction

At designated trauma centers, trauma activations are managed by a trauma team.1, 2, 3, 4, 5 The trauma team leader is often a surgeon, but in some cases, the team is led by an emergency medicine physician (EMP).2, 6, 7, 8 After the initial assessment and resuscitation is completed and the patient is stabilized, diagnostic imaging is performed to identify injuries that are not evident clinically.

Computed tomography (CT) imaging is used routinely as a screening and diagnostic tool to identify injuries in trauma patients.9, 10, 11 However, there has been some debate as to whether or not the benefit outweighs the risk and higher cost. One of the major risks of CT imaging is the increase in radiation exposure to the patient, which has been shown to increase the long-term probability of developing cancer.12, 13, 14 Another threat is the increase in health care cost associated with a liberal approach to CT imaging.15 Although both of these factors may curtail the use of CT imaging on all trauma patients, they must be balanced against the risk of missed injuries.

There is no clear consensus as to when liberal CT imaging should be used for blunt trauma or which patients are the best candidates. Furthermore, there appears to be significant variability in the CT scans desired by EMPs, surgical chief residents (SCRs), and attending trauma surgeons (TSs) when trauma patients are seen in the trauma bay. Gupta et al. determined whether CT scans desired by surgeons or EMPs resulted in the diagnosis of clinically significant injuries.16 Ten percent of scans deemed unnecessary by EM physicians but necessary by TSs yielded abnormalities.16 EMPs were more conservative in their use of CT scans when compared to TSs.

Trauma triage decisions can have a major effect on patient outcomes. Both training and experience can influence decision-making during the initial trauma evaluation. The primary aim of this study is to quantify the differences among EMP, SCR, and TS on the utilization and performance of CT scans. A pan-scan (PS) protocol was introduced 3 mo into the study to determine if it would influence the providers' choices. Our secondary aim is to study the effects of each group's decisions on missed injuries, cost, and radiation exposure.

Section snippets

Methods

A 6-mo prospective study was performed on 426 blunt trauma patients that met activation criteria at an urban, academic level 1 trauma center. Throughout the study, the EMP, SCR, and TS were asked to complete a one page questionnaire as to which body regions required CT scanning based on their participation in the primary and secondary surveys of the patient. They were asked not to discuss their selections with one another. During the first 3 mo, trauma team leaders directed which scans were

Results

Four hundred and eighty-four blunt trauma activations occurred in the emergency department (ED) during the study period, and 426 patients were eligible for study inclusion. The survey was completed prospectively for 258 (61%) of these patients, 116 during the pre-PS period and 142 during the PS period. The mean age was 51 y (SD 21.7), and 169 (65.5%) were male. The majority (65.5%) of patients had an injury severity score within the range of 0-8, 22.9% were between 9 and 15, and 11.6% had an

Discussion

The near-universal, immediate availability of CT scanning continues to impact the initial management of blunt trauma patients. The science is still evolving as to whether or not a PS is indicated in the patient who is not obviously injured after a blunt trauma.17, 18, 19 Given the lack of consensus in the literature and the differences in training, it is reasonable to suppose that different types of physicians might make different decisions regarding the most appropriate CT scans for blunt

Acknowledgment

Authors' contributions: S.W.L., M.K.J., and S.D.S. contributed for study conception and design. M.K.J., S.W.L., T.R.K., R.J.R., and S.D.S. contributed for acquisition of data. M.K.J., S.W.L., and S.D.S. participated in analysis and interpretation of data. M.K.J., S.D.S., M.D.M., and J.A.M. drafted the manuscript. P.S.B., M.K.J., S.D.S., S.W.L., and R.J.R. performed critical revision of the article.

References (20)

There are more references available in the full text version of this article.
View full text