a b s t r a c t

Objective: It is unclear if additional Computerized tomography imaging is warranted after injuries are identified on CT in blunt trauma patients. The objective of this study was to determine the incidence and significance of injuries identified on secondary CT imaging after identification of injuries on initial CTs in blunt trauma patients.

Methods: This was a retrospective cohort study at an academic Level 1 trauma center with a two-tiered Trauma system. Inclusion criteria: age >= 18, level 2 Trauma activation, injury identified on initial CT, and secondary CTs ordered. Secondary injuries were categorized as resulting in: no changes, minor changes, or major Changes in management.

Results: 537 patients underwent 1179 initial CT scans which identified 744 injuries. There were 1094 secondary CTs which identified 143 additional injuries in 94 (18%) patients. 9 (1.7%) patients had at least one major management change and 64 (12%) had at least one minor management change. Rib fracture(s) was the most common injury on secondary scans [45/143 (32%)]. The major management changes were: Tube thoracostomy for pneumothorax (4 patients), blood transfusion for hemoperitoneum (1 patient), surgery for acetabular fracture (1 patient), thoracolumbar brace for spine fracture (2 patients) and angiography for splenic injury (1 patient).

Conclusion: While a significant proportion of patients (18%) had injuries on secondary CT, only 1.7% of patients had a resultant major management change. Future research is warranted to determine the need for additional CT imaging after an initial selective imaging strategy in blunt trauma patients.

(C) 2019

Introduction

Computed tomography has become an essential diagnostic tool for the evaluation of blunt trauma patients in the Emergency Department (ED) [1]. The availability and accuracy of this imaging modality allows for rapid identification of life-threatening injuries, making it an optimal tool for the evaluation of trauma patients and contributing to a dramatic increase in CT utilization in the ED [1,2]. Increased imaging, however, leads to increased costs, resource utilization, and ED length of stay [3,4]. In addition, increased ionizing radiation exposure to patients undergoing CT imaging [5] may contribute to higher rates of fatal and non-fatal cancers [6]. Finally, studies have reported rates of unnecessary or negative CT imagining studies as high as 42% [7], with rates of

? Abstract presented at the Society of Academic Emergency Medicine Annual Meeting May 16, 2018 in Indianapolis, IN, United States of America.

* Corresponding author at: Cooper University Hospital, One Cooper Plaza, K152, Camden, NJ 08103, United States of America.

E-mail address: [email protected] (R. Byrne).

negative CTs for patients undergoing imaging based solely on mechanism of injury as high as 80-96% [8]. Identifying means to safely reduce unnecessary CT imaging in trauma patients evaluated in the ED is therefore of public health importance.

At our institution, patients who do not meet criteria for level one trauma activation undergo selective CT imaging based on history, physical exam, and mechanism of injury. If an injury is identified, patients frequently receive additional CT imaging of other body areas prior to patient disposition, often completing the “whole body CT” (WBCT). These additional CT images are obtained to rule out potential concomitant occult injuries missed by selective imaging. Obtaining secondary CT imaging after identifying an injury is not unique to our institution and also frequently occurs upon transfer to a trauma center [9,10]. It is currently unclear if secondary CT imaging to identify occult injuries is warranted.

Two recent retrospective studies found a low incidence of injuries identified on secondary CT scans obtained after identifying an injury on selective CT imaging [9,10]; however these results are limited secondary to small sample size (total n = 216), and one study only

https://doi.org/10.1016/j.ajem.2019.158432

0735-6757/(C) 2019

included patients receiving initial CT imaging of the head and cervical spine. Neither study included a detailed analysis of the significance of secondary imaging regarding patient-oriented changes in clinical management.

The objectives of this study were (1) to determine the management changes resulting from injuries discovered on secondary imaging (primary outcome) and (2) to determine the incidence of occult injuries identified on secondary CT imaging among a low risk trauma cohort who had any injury identified on initial selective CT imaging (secondary outcome).

Methods

Study design and setting

This study is reported in accordance with the Strengthening of Observational Studies in Epidemiology (STROBE) Statement [11]. Our institutional review board approved this study with waiver of informed consent. This was a retrospective cohort study of trauma patients evaluated in the ED of a single urban, academic, level 1 trauma center. Our institution has an annual ED census of approximately 80,000 and employs a two-tier trauma activation system according to an institutional protocol (Supplemental Table 1). Patients not meeting level 1 activation criteria are evaluated by an Emergency Medicine physician (EP), with or without a resident physician (level 2 activation). Patients evaluated by residents are, by departmental protocol, presented to the attending physician before any CT orders are placed. The EP attending makes all initial imaging and Treatment decisions and may also make the decision to obtain additional (secondary) CTs based on results of initial imaging or changes in patient clinical status. If an injury warranting admission is identified, the trauma team is consulted and further decisions regarding secondary imaging are at the discretion of the attending trauma surgeon. Patients with injuries not requiring admission may also undergo secondary imaging deemed necessary by the EP.

Subject selection

We queried our electronic medical record EPIC(TM) (EPIC Systems Corporation, WI) to identify patients evaluated at our facility between January 2013 and December 2015 that met the following inclusion criteria; 1) age >= 18; 2) evaluated by an EP as a level 2 trauma activation;

3) underwent initial selective CT imaging of at least one core body part (head, cervical spine, chest, abdomen/pelvis) for the evaluation of trauma; 4) had at least one traumatic injury identified on initial CT imaging; and 5) had secondary CT imaging of at least one core body part ordered at least 15 min after but b12 h after initial scan orders. Similar time frame methodology has previously been used to define secondary CT imaging [10]. We excluded patients who were initially evaluated by a trauma surgeon, patients receiving an initial WBCT, patients with no injury identified on initial CT imaging, patients whose secondary CT imaging only included extremities or CT angiography, patients transferred from outside facilities, and patients whose only secondary CT imaging was a repeat of the same body area. Injuries identified on an adjacent body area were classified as occult injuries; for example, a CT of the cervical spine identifying a pneumothorax was classified as occult if a CT of the chest was not also ordered during initial scans.

Data collection

Two investigators reviewed the electronic medical record for each subject and abstracted the data. Both abstractors had previous experience using EPIC and underwent a formal training session, including performing joint data extraction on a set of practice medical records to ensure uniform handling of data. A standardized data

extraction form and predefined definition of variables were used for all data collection. The abstractors held periodic meetings to review coding rules and to monitor performance [12]. We calculated inter- observer agreement using the kappa statistic between the two abstractors based on a 10% sample of cases selected at random.

We recorded demographics, medical comorbidities, initial vital signs, mechanism of injury, and injury severity score (ISS). We captured all initial and secondary CT imaging performed as well as all initial plain film X-rays and recorded the time in minutes from the initial CT orders to the secondary CT orders. We recorded the number and types of injuries identified on both primary and secondary CT imaging. We entered all data into a Research Electronic Data Capture (REDCap, Vanderbilt University, TN) database [13] and exported into Stata/SE

14.1 for Mac, StataCorp LP (College Station, TX, USA) for analysis.

Outcome measures

The primary outcome was the incidence and types of management changes based on results of secondary scans, and the secondary outcome was the incidence of injuries identified on secondary scans. Management change was categorized as major, minor, or no change based on a priori criteria (Supplemental Table 2) and assigned by two investigators: an EP and a trauma surgeon. A third abstractor (a second trauma surgeon) was available to adjudicate discrepancies in categorization of management changes.

Data analysis

We began the analysis with descriptive statistics. We displayed categorical data as counts and proportions. We described continuous data as mean values and standard deviation (SD) or median values and interquartile range (IQR), based on distribution of data. Injuries identified on initial and secondary CT imaging were tabulated according to type of CT imaging (head, cervical spine, chest, abdomen/pelvis combination, pelvis alone, thoracic spine, lumbar spine).

We calculated the proportion [with 95% confidence intervals (CI)] of patients with: an injury identified on secondary CT, patients who had a resultant minor management change, and patients who had a resultant major management change.

Results

During the 36-month study period, a total of 627 charts were screened for inclusion and 537 patients were included in the final cohort (Fig. 1). Table 1 displays patient characteristics for the entire cohort. The most common mechanism of injury was ground level fall [189 (35%)]. Only 49 (9%) patients were on chronic anticoagulation and 102 (19%) chronic Antiplatelet therapy. Nine of these patients were on both anticoagulation and antiplatelet therapy. Warfarin was the most common anticoagulant [35/49 (71%)] and aspirin was the most common antiplatelet agent [90/102 (88%)]. On presentation to the ED, 88 (16%) patients had tachycardia (heart rate >= 100 beats/min) and only two (0.4%) patients were hypotensive (systolic blood pressure b 90). The median (IQR) ISS for the entire cohort was 9 (5-13). inter-observer agreement among data abstractors was excellent for all variables tested (? N 0.90), specifically for injury identified on secondary CT imaging (? = 1) and classification of management change (? = 1).

There were 1179 initial CT images obtained, which identified 744 injuries in the 537 patients (Table 2). The majority of patients had an initial CT of the head and cervical spine [448 (83%)]. The most common injury identified on initial CT imaging was intracranial extra-axial hemorrhage [208 (39%)]. The median (IQR) time from ordering initial CT imaging to the orders for secondary CT imaging was 103 (64-159) min.

There were 1094 secondary CTs performed which identified 143 injuries in 94 [18% (95% CI 14%-21%)] patients. 448 patients (83%) received secondary CTs which completed the WBCT. Table 3 displays injuries identified on secondary CT imaging among the entire cohort. The majority of patients had a secondary CT of the chest, abdomen, and pelvis [405 (75%)]. The most common injury identified on secondary CT imaging was rib fracture(s) [48 (34%)]. Of the 425 patients with a secondary CT of the chest, 154 (36%) had a plain film x-ray of the chest ordered as part of the initial workup. Of the 48 patients with rib fracture(s) identified on secondary CT imaging, 28 (58%) patients had an initial plain film x-ray of the chest. Of the 11 patients with a pneumothorax identified on secondary CT of the chest, 8 (73%) had an initial plain film x-ray of the chest. None of the seven patients who had a pelvic fracture identified on secondary CT imaging of the pelvis had an initial plain film of the pelvis ordered.

Epic EMR Query 637 patients
		Exclusions: 28 <18 years old 37 no injury on initial CT 2 evaluated primarily by trauma surgery 17 extremity injury only on initial CT 20 WBCT done initially 19 no CT of core body part on secondary CT 4 secondary CT repeat of injured body area
*537 patients analyzed

Of 537 patients, 64 [12% (95% CI 9-15%)] had at least one management change based on the results of secondary CT imaging; 9 [1.7% (95% CI 0.8-3%)] had major changes and 55 [10% (95% CI 8-13%)] had minor changes. The most common management change was follow-up imaging of an injury [51/64 (80%)], of which the majority were for patients with rib fractures [31/51 (61%)]. The full list of management changes is listed in Table 4. There were 10 major management changes in 9 patients: thoracostomy tube in four patients, surgery for acetabular fracture in one patient, admission to the hospital for one patient, thoracolumbar bracing for unstable spine fractures in 2 patients, and transfusion of one unit PRBC plus angiography for splenic

Fig. 1. Flowchart of study patients listing reasons for exclusion from analysis. *Several patients met multiple exclusion criteria.

Table 2

Injuries identified on initial Computerized tomography imaging.

Injury All subjects

n = 537

n, (% of subjects with injury)

Table 1

Baseline characteristics for all subjects.

Variable All subjects

n = 537

Head CT (448 ordered)

Extra-axial hemorrhage	208 (39)
Intra-axial hemorrhage	76 (14)
Skull fracture	54 (10)
Other	25 (5)

Age [years, median (IQR)] Female [n (%)]	60 (41-78) 236 (44)	Cervical spine CT (399 ordered) cervical spine fracture	91 (17)
Pre-existing comorbidities [n (%)]		Rib fracture	15 (3)
Diabetes	91 (17)	Pneumothorax	10 (2)
Known coronary artery disease	76 (14)	pulmonary contusion	2 (0.4)
Hypertension	243 (45)	Other	24 (4)
Malignancy	22 (4)	Chest CT (59 ordered)
Renal insufficiency	26 (5)	Pneumothorax	8 (1)
Pulmonary disease	68 (13)	Pulmonary contusion	2 (0.4)
Cerebral vascular disease	33 (6)	Clavicle fracture	3 (0.6)
Congestive heart failure	26 (5)	Rib fracture	31 (6)
Dementia	31 (6)	Thoracic spine fracture	8 (1)
Charlson comorbidity score [median (IQR)]	0 (0-1)	Lumbar spine fracture	9 (2)
Anticoagulation [n (%)]	49 (9)	Other	12 (2)
Antiplatelets [n (%)] Mechanism of injury [n (%)]	102 (19)	Abdominal/pelvis CT (55 abdominal/55 pelvis ordered)
Ground level fall	189 (35)	liver injury	3 (0.6)
Fall from height	99 (18)	Splenic injury	6 (1)
Fall, unspecified	34 (6)	Pelvic fracture	12 (2)
MVC	69 (13)	Free fluid	1 (0.2)
MCC	22 (4)	Pneumothorax	1 (0.2)
Bicycle crash	15 (3)	Pulmonary contusion	1 (0.2)
Assault, blunt	82 (15)	Rib fracture	4 (0.7)
Assault, penetrating	1 (0.2)	Other	13 (2)
Diving	1 (0.2)	Facial bone CT (131 ordered)
Unknown	24 (4)	Facial bone fracture	95 (18)
Presenting vitals [median (IQR)]		Retrobulbar hematoma	5 (1)
Heart rate Respiratory rate Systolic blood pressure	82 (72-95) 18 (18-20) 142 (127-161)	Other Thoracic spine CT (12 ordered) Thoracic spine fracture	3 (0.6) 6 (1)
Diastolic blood pressure Saturation of arterial oxygen Injury severity score [median (IQR)]	82 (73-92) 98 (96-99) 9 (5-13)	Rib fracture Lumbar spine CT (20 ordered) Lumbar spine fracture	1 (0.2) 14 (3)
MVC = motor vehicle crash; MCC = motorcycle crash; IQR = interquartile range; SD = Other 1 (0.2)

standard deviation. CT = computerized tomography.

Table 3

Injuries identified on secondary computerized tomography (CT) imaging.

Injury All subjects

n = 537

n, (% of subjects with injury)

Head CT (77 ordered)

Extra-axial hemorrhage 1 (0.2)

Intra-axial hemorrhage 0

Skull fracture 0

Other 0

Cervical spine CT (121 ordered)

Cervical spine fracture 1 (0.2)

Other 2 (0.4)

Chest CT (425 ordered)

Pneumothorax	9 (2)
Pulmonary contusion	4 (1)
Clavicle fracture	11 (2)
Rib fracture	45 (8)
Thoracic spine fracture	9 (2)
Lumbar spine fracture	6 (1)
Aortic injury	1 (0.2)
Other	15 (3)
Abdominal/pelvis CT
(447 abdominal/443 pelvis ordered)
Liver injury	2 (0.4)
Splenic injury	3 (0.6)
Renal injury	1 (0.2)
Pelvic fracture	7 (1)
Free fluid	1 (0.2)
Pneumothorax	2 (0.4)
Pulmonary Contusion	1 (0.2)
Rib fracture	3 (0.6)
Other	16 (3)
Facial bone CT (8 ordered)
Facial bone fracture	1 (0.2)
Thoracic spine CT (11 ordered) Thoracic spine fracture	1 (0.2)
Lumbar spine CT (5 ordered)
Lumbar spine fracture	1 (0.2)

Table 4

Management changes based on secondary computerized tomography (CT) imaging.

Management change Frequency^a (% total patients)

Major n = 9 patients

Thoracostomy tube	4 (0.7)
surgical fixation for acetabular fracture	1 (0.2)
Blood product transfusion	1 (0.2)
Admission to hospital	1 (0.2)
TLSO brace for unstable fracture	2 (0.4)
Angiography: embolization of splenic injury Minor n = 64 patients	1 (0.2)
Follow-up imaging of injury	51 (10)
TLSO brace for comfort	3 (1)
Sling for scapula or clavicle fracture	10 (2)
Upgrade to trauma ICU	2 (0.4)

^a Frequency count does not equal number of patients as several patients had multiple management changes.

injury in one patient. Mechanism of injury, results of initial and secondary CT imaging, and management changes for patients with major management changes based on secondary scans are described in Table 5.

Discussion

In this retrospective cohort study, we sought to determine the incidence of occult injuries identified on secondary CT imaging among a low risk (level 2 trauma activation) population who had injuries identified by an initial selective CT imaging strategy, as well as the management implications resulting from those injuries. This is the largest study to date to analyze injuries identified on secondary CT imaging as well as the subsequent changes to patient management.

In our study, 1094 secondary CT scans identified 143 additional injuries in a substantial proportion (18%) of patients. Of note, 53/143 (37.5%) of the additional injuries identified on secondary CT imaging had plain films ordered of the injured body area, making these unlikely to be truly occult injuries (i.e. clinically unsuspected based on history and physical exam) injuries. Our classification of these injuries as occult, therefore, represents the most conservative approach to the interpretation of the data, assuming that most or all of the injuries could have been missed on plain film imaging. Moreover, most additional injuries led to minor or no changes in patient management. Rib fracture(s) alone accounted for 48 (34%) of all injuries on secondary CT scans, with the management change in nearly all these cases being a simple repeat plain film of the chest typically performed on the following hospital day.

Of the 9 patients with a major management change based on secondary CT, only 1 had a potentially life-threatening injury requiring intervention (embolization for splenic injury). Three of the 4 patients requiring tube thoracostomy had a chest x-ray ordered as part of their initial management, making it unlikely that these were truly clinically unsuspected. All 3 pneumothoraxes were detected on final radiology interpretations of the plain films, and it is unlikely that a clinically significant pneumothorax would have been missed by the EP. None of the patients had a Tension pneumothorax. The patient with an acetabular fracture had a plain film of the affected hip performed initially, and even if missed would have ultimately been detected after the patient attempted to ambulate. One patient received one unit of Packed red blood cells for unclear reasons but was never hypotensive and never proceeded to surgical intervention. Two patients were placed in thoracolumbosacral orthotic devices for spinal fractures. These were adjudicated to represent a major Change in management for these patients as their fractures were both 2-column injuries and potentially unstable. Neither patient, however, had a spinal cord injury or went on to receive surgical fixation. Finally, the patient requiring angiography for splenic injury after a fall down stairs was initially intoxicated and difficult to evaluate clinically. After achieving sobriety, a documented repeat exam noted complaints of abdominal pain and tenderness on exam, and the patient was noted to be tachycardic. A reasonable, routine re-examination of any similar patient would have prompted further imaging and at no point was this patient hemodynamically unstable.

One approach to low risk blunt trauma patients is to initially obtain CTs of select body areas based on clinical exam. The limitations of physical examination, however, raises concern for missed clinically significant occult injuries when performing selective CT imaging [14,15]. In addition, numerous studies examining the utility of the whole body CT (WBCT) have reported increased identification of injuries resulting in changes in patient management [16,8,17,18], while several meta-analyses have found an association with decreased mortality in blunt trauma patients who receive immediate WBCT versus selective CT [19,20,21,22]. It is important to note that the studies demonstrating decreased mortality with WBCT typically included a more severely injured cohort of patients than in our study, with median ISS in the 24-28 range compared to our median ISS of 9. An ISS of 9 is better reflective of the injury severity typically encountered by most EPs in most emergency departments, limiting the applicability of WBCT studies focusing on severely injured patients seen in Level 1 trauma centers. Conversely, the lone randomized controlled trial directly comparing selective CT with WBCT failed to find a Mortality benefit [23]. This ongoing controversy leads many EPs to perform selective CT imaging based on clinical examination, especially among lower risk patients typically encountered in non-Level 1 trauma centers. Other studies examining the utility of secondary CTs have found similar results. In the study by Heller MT et al., patients transferred to a trauma center undergoing secondary CTs had a 7% incidence of minor injuries, and no Major injuries [9]. This study was limited by the small sample of patients (101), and a reliance on record review to

Table 5

Mechanism, identified injuries, and management changes among subjects who had a major management change.

Age/sex/mechanism	Initial injury	Secondary scan: Injuries	Major management change resulting from secondary CT	Initial plain films ordered of injured body part
93 yo male unknown	C-spine CT:	Chest CT: Rib fracture with ptx	Tube thoracostomy	cxr
70 yo male fall from	Hydropneumothorax Cervical spine CT: Bilateral	Chest CT: Rib fx, ptx	Tube thoracostomy	None
standing 68 yo male fall from	apical ptx Head CT: Orbital fx	Abdomen/pelvis CT: Multiple pelvic fx	ORIF left	L hip
height 78 yo male fall from	Cervical Spine CT: PTX	Chest CT: Rib fx, ptx, clavicle fx	acetabulum Tube thoracostomy	cxr

standing

51 yo female fall from standing

Head CT: Multiple facial fx Abdomen/pelvis CT: rib fx, grade III splenic laceration, hemoperitoneum. Angiography

intervention PRBC transfusion

None

49 yo fall from height

CT cspine - Fx of L posterior

first rib

CT a/p: splenic laceration Admission to

hospital

None

78 yo male fall from standing

CT head - SDH CT a/p: Fx of T8/T9 vertebral bodies TLSO brace for 2 column injury

None

83 yo female fall down steps

CT head - SDH CT chest, a/p - multiple rib fx, clavicle fx, L2 vertebral compression fx, L1 inferior endplate fx

TLSO brace for 2 column injury

Shoulder

56 yo male bicycle accident

C-spine CT: Rib fx, bilateral tiny apical ptx

Chest/abdomen/pelvis CT: Grade 4 right hepatic laceration, Pulmonary lacerations, pulmonary contusions, tiny bilateral ptx, rib fx

Tube thoracostomy cxr

c-spine: cervical spine, CT: computed tomography, cxr: chest X-ray, L: left, SDH: subdural hematoma, ptx: pneumothorax, fx: fracture, ORIF: open reduction internal fixation, TLSO: thoracic lumbar sacral orthosis, PRBC: packed red blood cells.

determine which secondary CTs were not felt to be clinically indicated by the trauma team. Further, it is unclear how many patients received secondary CTs prior to transfer, meaning most injuries may have already been identified by the transferring physician and excluded from analysis.

In another study by Kelleher et al., 115 patients underwent secondary CT of the chest and abdomen/pelvis after identification of an injury on head or c-spine CT [10]. None of the secondary scans identified any additional injuries. Similar to the Heller study, many of the patients were transferred from another facility, and patients who received secondary imaging of the torso at the transferring institution were excluded from analysis. Further, this was not a severely injured cohort, as indicated by the predominant mechanism of injury (80.9% ground level fall) and mean ISS of 9.3. These factors may help explain the 0% incidence of injuries on secondary CTs in this study compared to our data.

A recent prospective study attempted to determine rates of missed injuries on potentially unnecessary CT imaging in blunt trauma patients by prospectively surveying emergency physicians and trauma surgeons regarding the perceived necessity of ordered CT scans [24]. The authors found few clinically important injuries on imaging deemed not necessary by the emergency physicians, however the premise of this study is limited by the design: physicians only gave an opinion on whether certain CTs were required, with the full knowledge that the scans would be performed regardless. This may have led to a false confidence in labeling CTs unnecessary, as the actual consequences of potential missed injuries did not play a role in their decision making. Our study, although retrospective, is likely closer to real-world practice as EPs made definitive decisions on initial selective imaging based on history and physical examination.

The only randomized trial to date examining outcomes in patients undergoing WBCT vs selective CT did not show any difference in mortality between groups [23], demonstrating that a selective imaging approach is reasonable in blunt trauma patients. Numerous studies have identified clinical decision tools to help physicians determine when it is appropriate to forgo CT imaging, such as the Canadian C- spine rule [25], the Canadian CT Head rule [26], and the NEXUS Chest CT instrument [27]. Other preliminary studies have begun to explore criteria for determining which patients can safely undergo selective CT imaging as opposed to WBCT [7].

This study has several important limitations. First, this was a single center, retrospective study. Second, we selected patients for analysis who had at least one injury identified on initial selective CT imaging and subsequently had a secondary CT. We did not we include patients who had an initial injury on initial selective CT imaging and did not have secondary imaging and thus some occult injuries may have been missed. Similarly, we did not assess for patients who may have had injuries missed on initial CT scans and were discharged but returned to the ED and subsequently had injuries identified on repeat imaging. Third, patients with secondary scans ordered at least 15 min but b12 h after initial scans were included in analysis. The 15-minute window was chosen to ensure capture of subjects that had additional imaging ordered very quickly after initial imaging orders based on physician or radiology technician identification of injuries during primary scans. This short window may have resulted in the inclusion of some patients whose secondary CT orders were placed before patient had their first studies performed (and were therefore part of the initial CT orders). Our median time to placement of secondary CT orders, however, was 103 (64-159) min, making it unlikely that this contributed significantly to our results. Fourth, the reasons why secondary CT scans were ordered on this patient cohort are unclear. Possible contributing factors may have included: EM physician suspicion for occult injuries based on the results of initial CTs, consultant or trauma attending request, or a change in patient status increasing suspicion for injuries. Change in a patient’s clinical status would certainly warrant additional CT scans, and if these patients had been excluded from analysis, the rate of truly occult (clinically unsuspected) injuries on secondary CTs would have been even lower. Finally, changes in management were determined in a retrospective fashion which may have been influenced by abstractor bias. Management changes, however, were strictly defined a priori by the EP and trauma surgeon investigator and inter-observer agreement for management change was perfect.

Conclusion

Our study provides evidence that while the overall incidence of traumatic injuries identified on secondary scans is significant (18%), the proportion of patients in this low risk cohort requiring a major change to their clinical management is very small (1.7%). Further work is warranted to better define criteria for determining the need

for secondary CT imaging in patients who have an injury identified by an initial selective CT imaging strategy.

Grant support

None.

Author contributions

RB, JPH, MK, AP, and BWR conceived and designed the study. MK performed all EMR inquiries and established the patient database. BWR created and maintained the REDCAP database. RB and AP collected and recorded all study data. BWR performed all statistical analysis. RB drafted the manuscript, and all authors contributed substantially to its revision. RB takes responsibility for the paper as a whole.

Declaration of competing interest

None.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2019.158432.

References

1. Broder J, Warshauer DM. Increasing utilization of computed tomography in the adult emergency department, 2000-2005. Emerg Radiol. 2006;13(1):25-30. https://doi. org/10.1007/s10140-006-0493-9.
2. Kocher KE, Meurer WJ, Fazel R, Scott PA, Krumholz HM, Nallamothu BK. National trends in use of computed tomography in the emergency department. Ann Emerg Med. 2011;58(5):452-462.e3. https://doi.org/10.1016/j.annemergmed.2011.05.020.
3. Lehtimaki T, Juvonen P, Valtonen H, Miettinen P, Paajanen H, Vanninen R. Impact of routine contrast-enhanced CT on costs and use of Hospital resources in patients with acute abdomen. Results of a randomised clinical trial. Eur Radiol. 2013;23(9): 2538-45. https://doi.org/10.1007/s00330-013-2848-4.
4. Smits M, Dippel DWJ, Nederkoorn PJ, et al. Minor head injury: CT-based strategies for management-a cost-effectiveness analysis. Radiology. 2010;254(2):532-40. https://doi.org/10.1148/radiol.2541081672.
5. Asha S, Curtis KA, Grant N, et al. Comparison of radiation exposure of trauma patients from diagnostic radiology procedures before and after the introduction of a panscan protocol. Emerg Med Australas. 2012;24(1):43-51. https://doi.org/10. 1111/j.1742-6723.2011.01504.x.
6. Brenner DJ, Elliston CD. Estimated radiation risks potentially associated with full- body CT screening. Radiology. 2004;232(3):735-8. https://doi.org/10.1148/radiol. 2323031095.
7. Davies RM, Scrimshire AB, Sweetman L, Anderton MJ, Holt EM. A decision tool for whole-body CT in major trauma that safely reduces unnecessary scanning and associated radiation risks: an initial exploratory analysis. Injury. 2016;47(1):43-9. https://doi.org/10.1016/j.injury.2015.08.036.
8. Salim A. Whole body imaging in blunt multisystem trauma patients without obvious signs of injury. Arch Surg. 2006;141(5):468. https://doi.org/10.1001/archsurg.141.5. 468.
9. Heller MT, Kanal E, Almusa O, et al. Utility of additional CT examinations driven by completion of a standard trauma imaging protocol in patients transferred for minor trauma. Emerg Radiol. 2014;21(4):341-7. https://doi.org/10.1007/s10140- 014-1200-x.
10. Kelleher MS, Gao G, Rolen MF, Bokhari SAJ. Completion CT of chest, abdomen, and pelvis after acute head and cervical spine trauma: incidence of acute traumatic findings in the setting of low-velocity trauma. Radiology. 2016;279(2):395-9. https://doi.org/10.1148/radiol.2015151509.
11. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12): 1495-9. https://doi.org/10.1016/j.ijsu.2014.07.013.
12. Gilbert EH, Lowenstein SR, Koziol-McLain J, Barta DC, Steiner J. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med. 1996; 27(3):305-8. https://doi.org/10.1016/S0196-0644(96)70264-0.
13. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)-a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42 (2):377-81. https://doi.org/10.1016/j.jbi.2008.08.010.
14. Schurink GWH, Bode PJ, van Luijt PA, van Vugt AB. The value of physical examination in the diagnosis of patients with blunt abdominal trauma: a retrospective study. Injury. 1997;28(4):261-5. https://doi.org/10.1016/S0020-1383(97)00007-7.
15. Neeki MM, Hendy D, Dong F, et al. Correlating abdominal pain and intra-Abdominal injury in patients with blunt abdominal trauma. Trauma Surg Acute Care Open. 2017;2(1). https://doi.org/10.1136/tsaco-2017-000109.
16. Deunk J, Brink M, Dekker HM, et al. Routine versus selective computed tomography of the abdomen, pelvis, and lumbar spine in blunt trauma: a prospective evaluation. J Trauma Inj Infect Crit Care. 2009;66(4):1108-17. https://doi.org/10.1097/TA. 0b013e31817e55c3.
17. Moussavi N, Davoodabadi AH, Atoof F, Razi SE, Behnampour M, Talari HR. Routine chest computed tomography and hospital admission time and complications in patients with blunt trauma. Arch Trauma Res. 2015;4(2). https://doi.org/10.5812/ atr.25299v2.
18. Self ML, Blake A-M, Whitley M, Nadalo L, Dunn E. The benefit of routine thoracic, abdominal, and pelvic computed tomography to evaluate trauma patients with Closed head injuries. Am J Surg. 2003;186(6):609-14. https://doi.org/10.1016/j. amjsurg.2003.08.003.
19. Caputo ND, Stahmer C, Lim G, Shah K. Whole-body computed tomographic scanning leads to better survival as opposed to selective scanning in trauma patients. J Trauma Acute Care Surg. 2014;77(4):534-9. https://doi.org/10.1097/TA. 0000000000000414.
20. Jiang L, Ma Y, Jiang S, et al. Comparison of whole-body computed tomography vs selective radiological imaging on outcomes in major trauma patients: a meta- analysis. Scand J Trauma Resusc Emerg Med. 2014;22(1):54. https://doi.org/10. 1186/s13049-014-0054-2.
21. Chidambaram S, Goh EL, Khan MA. A meta-analysis of the efficacy of whole-body Computed tomography imaging in the management of trauma and injury. Injury. 2017;48(8):1784-93. https://doi.org/10.1016/j.injury.2017.06.003.
22. Healy DA, Hegarty A, Feeley I, Clarke-Moloney M, GRACE PA, Walsh SR. Systematic review and meta-analysis of routine total body CT compared with selective CT in trauma patients. Emerg Med J. 2014;31(2):101-8. https://doi.org/10.1136/ emermed-2012-201892.
23. Sierink JC, Treskes K, Edwards MJR, et al. Immediate total-body CT scanning versus conventional imaging and selective CT scanning in patients with severe trauma (REACT-2): a randomised controlled trial. Lancet. 2016;388(10045):673-83. https://doi.org/10.1016/S0140-6736(16)30932-1.
24. Gupta M, Schriger DL, Hiatt JR, et al. selective use of computed tomography compared with routine whole body imaging in patients with blunt trauma. Ann Emerg Med. 2011;58(5):407-416.e15. https://doi.org/10.1016/j.annemergmed. 2011.06.003.
25. Stiell IG. The Canadian C-Spine Rule for radiography in alert and stable trauma patients. JAMA. 2001;286(15):1841. https://doi.org/10.1001/jama.286.15.1841.
26. Stiell IG, Wells GA, Vandemheen K, et al. The canadian CT head rule for patients with minor head injury. Lancet. 2001;357(9266):1391-6. https://doi.org/10.1016/S0140- 6736(00)04561-X.
27. Robert M, Langdorf MI, Nishijima D, et al. Derivation and validation of two Decision instruments for selective chest CT in blunt trauma : a Multicenter prospective observational study ( NEXUS Chest CT). PLoS Med. 2015;12(10):1-17. https://doi. org/10.7272/q6w66hpn.