a b s t r a c t

Background: Trauma patients sustaining Blunt injuries are exposed to multiple radiologic studies. Evidence indi- cates that the risk of cancer from exposure to ionizing radiation rises in direct proportion to the cumulative effec- tive dose (CED) received. The purpose of this study is to quantify the amount of ionizing radiation accumulated when arriving directly from point of injury to San Antonio Military Medical Center (SAMMC), a level I trauma center, compared with those transferred from other facilities.

Methods: A retrospective record review was conducted from 1st January 2010 through 31st December 2012. The SAMMC trauma registry, electronic medical records, and the digital radiology imaging system were searched for possible candidates. The medical records were then analyzed for sex, age, mechanism of injury, received directly from point of injury (direct group), transfer from another medical facility (transfer group), computed tomo- graphic scans received, dose-length product, CED of radiation, and injury severity score. A diagnostic imaging physicist then calculated the estimated CED each subject received based on the dose-length product of each com- puted tomographic scan.

Results: A total of 300 patients were analyzed, with 150 patients in the direct group and 150 patients in the trans- fer group. Both groups were similar in age and sex. Patients in the transfer group received a significantly greater CED of radiation compared with the direct group (mean, 37.6 mSv vs 28 mSv; P = .001). The radiation received in the direct group correlates with a lifetime attributable risk (LAR) of 1 in 357 compared with the transfer group with an increase in LAR to 1 in 266.

Conclusion: Patients transferred to our facility received a 34% increase in ionizing radiation compared with pa- tients brought directly from the injury scene. This increased dose of ionizing radiation contributes to the LAR of cancer and needs to be considered before repeating imaging studies.

Level of evidence: III.

Background

Trauma patients sustaining blunt injuries are exposed to multiple ra- diologic studies during their initial evaluation and throughout their

? Paper was presented to the Graduate Faculty of Baylor University in Partial Fulfillment of the Requirements for the Degree of Doctor of Science in Physician Assistant Studies in Emergency Medicine.

?? No conflicts of interest to declare. No funding received for this work.

* Corresponding author at: 10925 Bradbury Way, Peyton, CO 80831. Tel.:+1 915 319

3162 (cell), +1 719 559 3373 (home).

E-mail addresses: [email protected], [email protected] (K.A. Van Arnem), [email protected], [email protected] (D.P. Supinski), [email protected] (J.E. Tucker), [email protected] (S. Varney).

hospital stay. This radiation dose is cumulative and associated with an in- creased risk of cancer [1]. Not uncommonly, imaging is repeated in pa- tients who are transferred from outside facilities to the regional trauma center adding to the cumulative radiation exposure [2-4]. Studies report the cumulative effective dose (CED) of radiation in trauma patients as 6.8 to 22.7 mSv [5], but may be as high as 40.2 mSv in the first 24 hours [6].

The average effective dose of radiation from computed tomographic

(CT) scans ranges from 1 to 10 mSv, and it is estimated that we receive an average of 3 mSv of background radiation per year [7]. Table 1 shows typical radiation doses of common imaging studies as well as the equiv- alent annual background radiation [7]. To estimate the amount of radi- ation received from exposures, the National Research Council reviewed the current data on the health risks of low-level ionizing radiation and

http://dx.doi.org/10.1016/j.ajem.2016.09.018 0735-6757/

Table 1

Radiation dose comparison
Diagnostic procedure	Typical effective dose (mSv)a	No. of chest X-rays (PA film) for equivalent effective doseb	Period for equivalent effective dose from natural background radiationc
Chest x-ray (PA film)	0.02	1	2.4 d
Skull x-ray	0.1	5	12 d
lumbar spine	1.5	75	182 d
CT head	2	100	243 d
CT abdomen	8	400	2.7 y

Abbreviation: PA, posteroanterior.

^a Average effective dose in millisieverts (mSv) as compiled by Fred A. Mettler, Jr, et al., “Effective Doses in Radiology and Diagnostic nuclear medicine: A Catalog,” Radiology July 2008;248 (No. 1):254-263.

^b Based on the assumption of an average “effective dose” from chest x-ray (PA film) of 0.02 mSv.

^c Based on the assumption of an average “effective dose” from natural background radiation of 3 mSv per year in the United States.

published the Biological Effects of Ionizing Radiation VII (BEIR VII) Phase 2 report. The Council developed risk models and calculations, and iden- tified potential future research [1]. It concluded that a linear, dose- response relationship without a threshold exists between ionizing radi- ation and the development of solid organ and other cancers [1]. This risk of cancer increases with the amount of radiation exposure and is in- creased with a younger age at time of exposure [1].

Diagnostic imaging of blunt trauma patients, albeit critical to their timely evaluation, has increased dramatically due to convenience and speed of imaging modalities, particularly CT scanners. Computed tomo- graphic scans produce ionizing radiation, and the Cumulative dose in- creases one’s risk of developing cancer [1]. Clearly, when blunt trauma patients are deemed critically injured sufficiently to require transfer to a trauma center, some imaging studies may need to be repeated; how- ever, this may be better determined after examining the patient, reviewing the images and/or reports provided, and then considering the patient’s hemodynamic stability. The intent is to raise the decision to repeat radiographic imaging to a Conscious level instead of a reflex re- sponse when simply receiving a transferred patient. Our study objective was to quantify and compare the amount of radiation received by blunt trauma patients arriving directly from point of injury (direct group) to a level I trauma center compared with those transferred from other facil- ities (transfer group). We hypothesized that the transfer group would receive a 25% increased CED of radiation due to reimaging upon arrival to the regional trauma center. Our secondary objective was to estimate the patient’s increased lifetime attributable risk (LAR) of cancer due to the additional radiation received from one trauma incident.

Methods

Study design and setting

This study was a retrospective record review from 1st January 2010 through 31st December 2012 at a military level I trauma and referral center that sees approximately 75 000 military and civilian patients an- nually. It is the only level I trauma center in the Department of Defense located in the United States and receives 2300 to 2500 trauma patients annually, of which 40% are transferred from other facilities. The local in- stitutional review board approved the study.

Before any search for candidates or data were collected, a waiver for informed consent was granted to access patients’ health information.

Patient selection

The hospital trauma registry, electronic medical records, and Agfa IMPAX picture archiving and communications system (digital radiology imaging system) were searched for candidates satisfying inclusion and exclusion criteria. Included were all blunt trauma patients that present- ed as trauma patients, to include active duty military, reservists, Nation- al Guard, their dependents, and civilian emergencies. Any patient requiring immediate transfer to the operating room, any penetrating

trauma, or patients that died during their hospital stay were excluded from this study.

All patients meeting the inclusion and exclusion criteria were ob- tained from the Trauma Registry. Starting from 31st December 2012 and working back to 1st January 2010, we collected the first 150 pa- tients that were brought directly to our institution from point of injury. We then collected, in the same fashion, the first 150 patients with com- plete records that were transferred to our institution from other hospi- tals. The trauma records were paper charts filed in boxes by month and year of date of injury and stored in a secure room. Subjects brought di- rectly to San Antonio Military Medical Center (SAMMC) had all imaging studies performed at our facility and were located in the digital radiolo- gy system. The transfer records were reviewed and cross-referenced with the electronic medical record and IMPAX for additional data and CT scans. Each patient was assigned a unique number, and all identify- ing information was removed. This number was then added to the mas- ter data list to maintain the ability to cross-reference for additional information if necessary at a later time.

A single, trained abstractor reviewed the candidates’ records and en- tered the deidentified data on the prespecified case report form. Data points included sex, age, mechanism of injury (MOI), direct transfer from point of injury vs the name of the transferring medical facility, all ra- diologic imaging received, dose-length product (DLP) and/or CED from each CT scan, length of hospital stay , and injury severity score (ISS). As defined, DLP is the measurement generated from the CT scanner for each CT scan and gives an estimate of the total radiation dose.

Data analysis

The independent variable was mode of transfer (direct from point of injury to SAMMC vs those transferred from other facilities), and the de- pendent variable was the CED each subject received. We hypothesized that blunt trauma patients transferred from other hospitals would have a 25% increase in CED compared with those brought directly from the point of injury. We used descriptive statistics including means, proportions, and SDs. Categorical data were compared with the ?² test. Continuous data were compared using the Student t test for parametric data and both the Mann-Whitney rank sum test and Wilcoxon rank sum test for nonparametric data.

We used SPSS Sample Power, Version 2.0 (IBM Corporation, Armonk, NY) to estimate the sample size needed for a power of 80% with a level of confidence of 95%. As numerous studies report the CED of radiation in trauma patients to range from 6.8 to 22.7 mSv, we as- sumed this to be an average CED for our power analysis [5]. Based on a mean of 14.75 mSv and an SD of 3.975, an increase of 25% would pro- duce an effect size of 0.93 SD and would require a sample size of 20 sub- jects per group. We used 150 subjects per mode of transfer (300 total) to increase the power of our study.

A single, trained abstractor collected, reviewed, and entered the de- identified demographic and imaging data for each subject on a spread- sheet. A diagnostic imaging physicist performed the estimated dose calculations based on the dose report containing the DLP sent from the

CT system to IMPAX. When separately listed, the DLP for the CT scout image(s) for each study was eliminated, as some studies did not include the information and it contributes little to the total CED. At our institution, the CT scanner guidelines conform to the American College of Radiology and Joint Commission standards. It is not known, however, if all or some of the transferring facilities adhere to the same ACR and Joint Commission standards.

The DLP to effective dose calculation was made by multiplying the DLP by a standard conversion factor (mSv/mGy-cm) [8] for each particular area undergoing imaging (ie, head CT, chest CT, abdomen/pelvis CT, etc) to give an estimated dose in mSv for each individual CT scan. Some imag- ing studies that were performed at outside facilities were uploaded into our local IMPAX system, and those dosages were included. For those im- aging studies performed at outside facilities and not loaded into our IMPAX system, a standard dose estimate from the RADAR medical proce- dure dose calculator was used to estimate the dose received with each CT scan [9]. Tube current modulation (TCM) for dose reduction was used in approximately 84% of the studies performed on the direct group (528/ 628 studies); noting that TCM is used least for examinations of the head due to its relatively uniform thickness. Tube current modulation was used in at least 58% of the transfer group (531 studies with TCM and 387 without or unknown). Of the 81 CT examinations performed outside SAMMC and later entered into IMPAX, TCM was used in roughly half (38/ 81) of those examinations. We know that 43 examinations entered into IMPAX did not use TCM. As for the remaining 235 CT examinations from outside hospitals, we cannot determine whether TCM was used, but we assume that it was used in roughly half of the examinations.

On average, assuming a sex and age distribution similar to that of the entire US population, the BEIR VII lifetime risk model predicts that ap- proximately 1 person in 100 would be expected to develop cancer (solid cancer or leukemia) from a dose of 100 mSv above background. The LAR was calculated by multiplying the dose in mSv by 0.0001 mSv and taking the reciprocal of the result per the calculations proposed in the BEIR VII report [1].

Results

A total of 536 medical records were reviewed to obtain a total of 300 complete patient medical records. We reviewed 150 medical records for those patients brought directly from point of injury and 386 medical re- cords for patients transferred from an outside facility. Approximately 28% of the CT scans from the transferred group required the use of standard doses as those images and DLPs from other facilities were not available in our IMPAX system. These studies and images were not received upon pa- tient transfer to our institution but were annotated in their transfer paper- work as performed. In these cases, we used a standardized estimate to determine the amount of radiation received from studies before transfer.

Both groups were similar in age and sex (Table 2). Of note, the groups were statistically different, with the direct group having a longer LOS and lower ISS (Table 2). We further separated the ISS into 3 groups based on previous studies of low ISS (0-8), medium ISS (9-14), and high ISS (>=15). The most common mechanisms for blunt trauma overall were motor vehicle collision, fall b1 m, assaults, fall 1-6 m, and motorcycle crashes. The most common MOI by group (direct vs transfer) can be found in Table 3. The transfer group received a greater mean CED of radiation than did the direct group (37.6 vs 28 mSv, respectively; P = .001).

The results of CED by group are shown in Table 4.

A 2-factor analysis of variance was then performed on CED by transfer group and ISS group. The transfer group’s CED was greater than the direct group’s (P = .022). In addition, 2 or more of the ISS groups differed from each other (P = .007). Table 4 shows the CED by ISS category for each group (direct vs transfer). Figure further illustrates that for both the low and medium ISS categories, the transfer group had an increased CED, whereas in the high ISS category, the CED was greater for the direct group. The distributions of the CED in the direct and transfer groups as well as the ISS categories were skewed to the high side so the results of the

Table 2

Patient demographics and clinical characteristics by direct and transfer group

Group	Direct	Transfer	P
Sex, n (%)				.2a
Female	53.3 (35.3)	42 (28)
Male Age (y)	97 (64.7)	108 (72)		.3b
Mean (SD)	49.5 (23.5)	46.4 (22.8)
Median [IQR]	49.5 [27, 68]	39.5 [27,66]
LOS (d)				.04c
Mean (SD)	4.8 (6.2)	4.12 (8.9)
Median [IQR]	3 [1, 5]	2 [1, 4]
ISS			b.001c
Mean (SD)	8.7 (8)	11.8 (8.7)
Median [IQR]	5.5 [2, 13]	9 [5, 17]
a ?2 Test.

^b t Test.

^c Wilcoxon test.

post hoc tests were verified with the nonparametric Mann-Whitney rank sum test. This also showed a significant increase in CED of the transfer group compared with the direct group (P = .005), that the high ISS category CED was significantly greater than the low and medi- um ISS categories (P b .001), and that there was no significant difference between the CED of the low and medium ISS categories (P N .05). In Table 5, the LAR was calculated. The LAR was greater in the transfer group (1/266) compared with the direct group (1 in 357).

Discussion

We found that the transfer group had a 34% increase in CED of radi- ation compared with the direct group, with an absolute difference of 9.6 mSv. Although the mean ISS in both groups placed them in the medium ISS group, the transfer group had a higher mean CED in radiation (P =

.006). When comparing the groups based on ISS category, the trans- ferred subjects had an increased CED compared with the direct group in both the low ISS (0-8) and the medium ISS (9-14) groups. Interest- ingly, the direct group received an increase in CED when compared with the transfer group in the high ISS (>=15) category.

We would expect the CED to increase with increased ISS and in- creased hospital stay, which we found in the direct group. In the transfer group, the CED did increase from low ISS to medium ISS as well, but the high ISS category had a lower CED than the medium ISS. This, along with the increased CED in the direct group of the high ISS category compared with the transfer group, may be due to multiple factors. One factor may be that those patients brought directly to our institution were severely injured but stable enough to undergo CT imaging. Another reason may be that those patients in the high ISS category were transferred immedi- ately with no or only minimal imaging performed at the initial facility, or there may have been no need to obtain (additional) CT scans upon

Table 3

Mechanism of injury by transfer group

Assault	9 (6)	30 (20)	39 (13)
Bicycle	1 (0.7)	1 (0.7)	2 (0.7)
Fall 1-6 m	9 (6)	16 (10.7)	25 (8.3)
Fall b1 m	35 (23.3)	39 (26)	74 (24.7)
Fall N6 m	2 (1.3)	1 (0.7)	3 (1)
Fall-NFS	9 (6)	7 (4.7)	16 (5.3)
MVC	49 (32.7)	30 (20)	79 (26.3)
Motorcycle	13 (8.7)	9 (6)	22 (7.3)
Other blunt trauma	14 (9.3)	13 (8.7)	27 (9)
Pedestrian	9 (6)	4 (2.7)	13 (4.3)
Total	150	150	300

MOI Transfer group, n (%) Total, n (%) Direct Transfer

Abbreviations: MVC, motor vehicle collision; NFS, not further specified.

Table 4

CED by transfer and ISS

Direct vs ISS Transfer group Mean CED (mSv) SD n

Direct	Low ISS	23.697	14.6515	78
	Medium ISS	23.761	23.5634	41
	High ISS	44.465	29.7469	31
	Total	28.007	22.5652	150
Transfer	Low ISS	34.238	27.9694	52
	Medium ISS	42.089	37.6483	44
	High ISS	37.157	24.2601	54
	Total	37.592	29.9354	150

arrival to our facility. This could be remedied by selecting patients with similar ISS, reviewing only those radiologic studies that were performed in the first 24 hours after injury, or using patients with all available stud- ies in IMPAX to avoid further estimations.

The LAR of cancer also increases with an increased CED. With an in- crease in CED in the transfer group (37.6 mSv) compared with the direct group (28 mSv), we expect a subsequent increase in LAR. According to the BEIR VII report, there is a linear dose-response relationship between ionizing radiation and development of cancers. Griffey and Sodickson [10] estimated cumulative radiation-related cancer risks by converting each patient’s cumulative CT effective dose to estimated LAR using the standard- ized BEIR VII conversion of 0.0001/mSv (ie, 1 person in 10 000 would be ex- pected to develop cancer from a dose of 1 mSv above background). Although a person has a 42 in 100 chance of carcinogenesis from other sources [1], our study found an increase in LAR of cancer of 1 in 357 for the direct group vs the transferred group with an increase in LAR of cancer to 1 in 266. The increased risk may appear low but is still a source of in- creased risk that should be mitigated, especially in children given that the lifetime risk of cancer is inversely proportional to the patient’s age at the time of exposure. Children are more susceptible to ionizing radiation due to their body size and years of life remaining to develop cancer.

As improvements have been made to reduce the amount of radiation emitted from medical devices, we can take further steps to help reduce the amount of radiation patients receive, including eliminating unnec- essary imaging studies. Those patients brought directly to the level I trauma center received a mean of 28 mSv in radiation throughout their hospital stay and may be due, in part, to the eliminated need of re- peating CT scans. One way to accomplish this would be to institute a local-wide compatible software program to transmit images from one facility to another. Although the patient is en route to another facility, the receiving hospital providers can review images before the patient arrives. This in itself may create other problems to include 2 different ra- diology reports, the legality of radiologists “overreading” another radiologist’s read on the same CT scan and reimbursement for study in- terpretation, and some may request the study to be repeated altogether due to poor Image quality, insufficient view of a particular area, or change in patient’s clinical status.

90

Cumulative Effective Dose (mSv)

80

70

60

50 Direct

Table 5

Lifetime probability of risk of cancer from the distribution of CED in each group

Source Studies Dose (mSv) LAR

Mean Median Maximum Mean Median Maximum Direct 150 28.0 27.9 140.3 1 in 357 1 in 358 1 in 71

Transfer 150 37.6 30.9 203.4 1 in 266 1 in 324 1 in 49

Another reason for repeat imaging is a change in clinical condition. Many times there is no alternative to reimaging patients. Establishing guidelines for reimaging may hinder those medical providers from doing the best for their patients as each situation is individualized. Also, simply the act of transporting a patient from one facility to another can contribute to changes in clinical condition and subsequent need for re- peating CT scans. The need to transfer a patient to a higher level of care also implies that these patients are more critical and would potentially re- quire additional and repeat imaging. The patient’s status and clinician’s judgment dictate whether or not additional imaging is required, but we believe that alternatives to the current system and process need to be carefully considered to help reduce the patient’s cumulative radiation dose and their subsequent increased LAR of cancer.

As our study looked at only one small point in time, the LAR is cumu- lative throughout one’s lifetime. There are efforts currently to develop a personal medical imaging history card made available to coexist with each patient’s medical record that would act as a tally of total radiation received by medical imaging, as reported by the US Food and Drug Ad- ministration [7]. Increased awareness in the medical and radiology community continues to contribute to the modification and reduction of the radiation emitted from medical devices and the decrease in un- necessary imaging of patients.

Limitations

Our study had limitations. First, this was a retrospective study and may not have accounted for all of the imaging studies performed. Howev- er, radiologic studies are performed and tracked, so it is less likely that we missed many. There is an assumption that the data recorded were accu- rate and complete. Although some of the images from transfer facilities were transferred or uploaded into our system, some were not. Possible explanations include not sending a disk with the images, an incompatible disk or program, or a disk that was misplaced during patient transfer.

Another limitation was that an estimate was calculated for the CED of radiation each subject received. Although a better estimate was calcu- lated based on each individual study instead of using a standardized dose chart, a more effective estimate may be to use a dosimeter for each patient. We also used a standard dose for CT images that were not loaded into our IMPAX system for approximately 28% of the CT scans in the transfer group. It is possible that this underestimated the actual CED received, to include the mean CED in the high ISS category. A future recommendation would be to use subjects with all available CT scan information in the IMPAX system to ensure homogeneity throughout the CED calculations. An alternative collection method would be to use dosimeters. Nevertheless, because we used the same reference values for respective studies in all of these cases, this should minimize the amount of random error.

Next, the LAR calculations are based on estimates using the standard linear model in the BEIR VII report; however, it is the best estimate

available. These risk models are based on the Life Span Study cohort of

40

30

20

10

0

Low ISS Medium ISS High ISS

Transfer

Hiroshima and Nagasaki atomic bomb survivors [1].

Finally, we focused on CT scans and did not include other imaging studies (ie, plain films, fluoroscopic examinations, intravenous urograms, etc). Although these studies contribute to the CED of radia- tion, the amount of radiation is not as significant as that received from CT scans. As Kim et al [11] reported, CT scans constitute less than 10% of the total number of studies performed, but they accounted for two-

Figure. Cumulative effective dose compared with ISS group by direct vs transfer.

thirds of the radiation dose.

Conclusion

The CT scan is an invaluable tool for assessing blunt trauma patients. Not only did we find a 34% increase in CED of radiation of transferred pa- tients compared with the direct group, but we also found a subsequent increased LAR of cancer for the transfer group [12-17].

Author contribution

Kerri Van Arnem contributed in all aspects of the study including lit- erature search, study design, data collection, data analysis, data inter- pretation, writing, critical revision, and final approval for submission. David Supinski contributed to this study in literature search, study de- sign, critical revision, and final approval for submission. Jonathan Tucker contributed to this study in study design, data analysis, data interpreta- tion, critical revision, and final approval for submission.

Shawn Varney contributed to this study in study design, data analysis, Data interpretation, critical revision, and final approval for submission.

Acknowledgment

A special thank you is extended to John A. Nunez for his hard work and contribution of information to our study.

References

1. Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation, National Research Council (U.S.). Health risks from exposure to low levels of ionizing radiation: BEIR VII, phase 2. Washington, DC: National Academies Press; 2006 xvi.
2. Emick DM, Carey TS, Charles AG, et al. Repeat imaging in trauma transfers: a retro- spective analysis of computed tomography scans repeated upon arrival to a level I trauma center. J Trauma 2012;72(5):1255-62.
3. Ahmadinia K, Smucker B, Nash C, et al. Radiation exposure has increased in trauma patients over time. J Trauma 2012;72(2):410-5.
4. Kocher KE, Meurer WJ, Fazel R, et al. National Trends in use of computed tomogra- phy in the emergency department. Ann Emerg Med 2011;58(5):452-62.
5. Sharma OP, Oswanski MF, Sidhu R, et al. Radiation trends in trauma patients. Am Surg 2011;77:193-7.
6. Winslow JE, Hinshaw JW, Hughes MJ, et al. Quantitative assessment of diagnostic ra- diation doses in adult blunt trauma patients. Ann Emerg Med 2008;52(2):93-7.
7. United States Food and Drug Administration, Center for Devices and Radiological Health. Ra- diation emitting products and procedures. What are the radiation risks from CT? Available at www.fda.gov/radiation-emittingproducts/radiationemittingproductsandprocedures/ medicalimaging/medicalx-rays/ucm115329.htm. [Accessed August 9, 2013].
8. American Association of Physicists in Medicine (AAPM) report 96. The measure- ment, reporting, and management of radiation dose in CT. College Park, MD: AAPM; 2008.
9. RAdiation Dose Assessment Resource (RADAR). www.doseinfo-radar.com/ RADARHome.html. [Accessed March 2013].
10. Griffey RT, Sodickson A. Cumulative radiation exposure and cancer risk estimates in emergency department patients undergoing repeat or multiple CT. Am J Radiol 2009;192:887-92.
11. Kim PK, Gracias VH, Maidment ADA, et al. Cumulative radiation dose caused by ra- diologic studies in critically ill trauma patients. J Trauma 2004;57(3):510-4.
12. Sharma OP, Oswanski MF, Sidhu R, et al. Analysis of radiation exposure in trauma patients at a level I trauma center. J Emerg Med 2011;41:640-8.
13. Hui CM, MacGregor JH, Tien HC, et al. Radiation dose from initial trauma assessment and resuscitation: review of the literature. Can J Surg 2009;52(2):147-52.
14. Tien HC, Tremblay LN, Rizoli SB, et al. Radiation exposure from diagnostic imaging in severely injured trauma patients. J Trauma 2007;62(1):151-6.
15. Mueller DL, Hatab M, Al-Senan R, et al. Pediatric radiation exposure during the initial evaluation for blunt trauma. J Trauma 2011;70(3):724-31.
16. Kim SH, Jung SE, SH O, et al. Effects of a radiation dose reduction strategy for com- puted tomography in severely injured trauma patients in the emergency depart- ment: an observational study. Scand J Trauma Resusc Emerg Med 2011;19:67.
17. Salottolo K, Bar-Oh R, Fleishman M, et al. Current utilization and radiation dose from computed tomography in patients with trauma. Crit Care Med 2009; 37(4):1336-40.