Introduction of a pan-scan protocol for blunt trauma activations: what are the consequences?

    Introduction of a Pan-scan protocol for blunt trauma activations: what are the consequences? Melissa K. James PhD, Sebastian D. Schubl MD, Michael P. Francois BS, Geoffrey K. Doughlin MD, FACS, Shi-Wen Lee DO PII: DOI: Reference:

S0735-6757(16)30605-2 doi: 10.1016/j.ajem.2016.09.027 YAJEM 56130

To appear in:

American Journal of Emergency Medicine

Received date: Revised date: Accepted date:

4 August 2016 13 September 2016 14 September 2016

Please cite this article as: James Melissa K., Schubl Sebastian D., Francois Michael P., Doughlin Geoffrey K., Lee Shi-Wen, Introduction of a Pan-scan protocol for blunt trauma activations: what are the consequences?, American Journal of Emergency Medicine (2016), doi: 10.1016/j.ajem.2016.09.027

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Introduction of a Pan-scan protocol for blunt trauma activations: what are the consequences? Melissa K. James, PhD, 1,3Sebastian D. Schubl, MD, 4Michael P. Francois, BS; 1,2Geoffrey K. Doughlin, MD, FACS, 2Shi-Wen Lee, DO

T

1

Departments of Surgery and 2Emergency Medicine, Jamaica Hospital Medical Center, Jamaica, NY, USA; 3Department of Surgery, University of California at Irvine Medical Center, Orange, CA, USA; 4Ross University, School of Medicine, Dominica, W.I.

SC

RI P

1

Contact information: Melissa K. James: [email protected]

MA

NU

Running title: Pan-scan vs. selective CT scanning in blunt trauma

ED

Sebastian D. Schubl: [email protected]

Michael P. Francois: [email protected]

PT

Geoffrey K. Doughlin: [email protected]

CE

Shi-Wen Lee: [email protected]

AC

Disclosure: All authors declare no conflicts of interest and have no financial ties to disclose.

Corresponding Author: Shi-Wen Lee, DO Department of Emergency Medicine Jamaica Hospital Medical Center 8900 Van Wyck Expressway, Jamaica, NY 11418 [email protected] Phone: 718-206-6026 Fax: 718-206-6085

ACCEPTED MANUSCRIPT Abstract Study Objective: The aim of this study is to determine if the introduction of a pan-scan protocol

findings, treatment times, radiation exposure, and cost.

RI P

T

during the initial assessment for blunt trauma activations would affect missed injuries, incidental

SC

Methods: A six month prospective study was performed on blunt trauma patients at a Level 1

NU

trauma center. During the last three months of the study a pan-scan protocol was introduced to the trauma assessment. Categorical data were analyzed by Fisher’s exact test and continuous

MA

data were analyzed by Mann-Whitney non-parametric test. Results: There were a total of 220 patients in the pre-panscan period and 206 patients during the

ED

pan-scan period. There was no significant difference in injury severity or mortality between the

PT

groups. Introduction of the pan-scan protocol substantially reduced the incidence of missed injuries from 3.2% to 0.5%, the length of stay in the ED by 68.2 minutes (95% CI: -134.4 to -

CE

2.1), and the mean time to the first operating room visit by 1465 minutes (95% CI: -2519 to -

AC

411). In contrast, fixed CT scan cost increased by $48.1 (95% CI: 32 to 64.1) per patient, however, total radiology cost per patient decreased by $50 (95% CI: -271.1 to 171.4). Additionally, the rate of incidental findings increased by 14.4% and the average radiation exposure per patient was 8.2 mSv (95% CI: 5.0 to 11.3) greater during the pan-scan period. Conclusion: Although there are advantages to whole body computed tomography, elucidation of the appropriate blunt trauma patient population is warranted when implementing a pan-scan protocol. Keywords: selective CT, computed tomography, pan-scan, whole body computed tomography, blunt trauma

ACCEPTED MANUSCRIPT Introduction The use of pan-computed tomography (also known as a pan-scan, PS) as standard of care for

RI P

T

trauma activations is becoming more prevalent. However, many trauma centers still use selective computed tomography to assess trauma patients. Selective computed tomography is generally

SC

directed by the trauma team leader and is based on the mechanism of injury, the physical exam findings, and the results of trauma bay imaging including plain film and a focused assessment

NU

with sonography for trauma (FAST). With the lower utility of CT scanning in penetrating trauma

MA

patients, the debate over the use of PS is isolated to victims of blunt trauma. There are often no obvious signs of injury in patients with blunt trauma though the mechanism may suggest high

ED

risk. Pan-computed tomography may be more appropriate in assessing blunt trauma injuries.[1] At some trauma centers, a PS is routinely used as a diagnostic tool to identify traumatic injuries.

PT

Several European registry-based studies have identified and confirmed a decrease in mortality at

CE

centers where a PS is utilized. However, there are several concerns with the designs of these studies.[2, 3] Additionally, there has been increasing concern on whether or not the benefits

AC

outweigh the risks.[4, 5] Patients subjected to PSs are exposed to a higher dose of radiation which can pose long-term cancer risk.[6-8] The increased use of PSs has also been critiqued due to its cost; however there is insufficient evidence to conclude that utilizing a PS as the standard of care for trauma patients increases total cost of care.[9, 10] On the other hand, PSs can be beneficial to trauma patients as it is considered more sensitive and accurate in detecting multiple injuries.[11, 12] This may be an important factor in decreasing the mortality rate in severely injured trauma patients as it allows early identification of injuries.[13-17] Some studies even demonstrate that a PS decreases the time it takes to identify injuries allowing for more timely treatment, and shorter stay in the Emergency Department (ED) and hospital.[2, 11, 12, 18] Additionally, some argue that

ACCEPTED MANUSCRIPT contrary to the belief that PSs increase exposure to radiation, if a PS is performed early in the evaluation process it may decrease the overall number of scans the patient receives and hence

RI P

T

decrease the overall level of radiation exposure.[2] Moreover, a PS may identify unanticipated injuries, which may have been missed if only

SC

selective scanning was chosen. Asha et al. examined missed injuries before and after the introduction of a PS protocol.[7] There was no significant benefit to utilizing a PS to detect

NU

missed injuries however; the overall incidence of missed injuries was minimal.[7] Another group

MA

attempted to determine if PSs were overused for blunt trauma patients via surveys of emergency medicine physicians and trauma surgeons.[19, 20] Prior to CT imaging both groups were asked to

ED

predict the site of injury and choose between whole body CT and selective CT. Whole body scanning detected missed injuries, 0.3% of which needed immediate intervention, suggesting that

PT

pan-computed tomography should be used for blunt trauma patients.[20]

CE

At our institution we utilize selective scanning as our standard of care for trauma activation

AC

patients. The aim of this study is to determine if implementing pan-computed tomography as the standard of care for trauma patients during the initial assessment would affect the number of missed injuries, radiation exposure, and hospital cost. Secondarily, we hypothesized that if PSs are utilized as part of the initial evaluation of all blunt trauma patients, ED disposition, treatment and/or discharge may be faster, thereby reducing cost of care. Methods Study design This prospective study took place over a six month period in the Emergency Department (ED) at an urban, academic Level 1 trauma center. Blunt trauma patients who met criteria as trauma

ACCEPTED MANUSCRIPT activations were included in the study. During the last three months of the study, a pan-scan (PS) protocol was introduced during the initial trauma assessment as the standard of care for all blunt

T

trauma activations. This study was approved by the institution’s review board, patient consent

RI P

was not required.

SC

Standard evaluation of trauma patients was performed, which included history, physical examination, anteroposterior chest and pelvis radiographs, and FAST. The PS protocol was

NU

developed with the assistance of the radiology department and introduced three months into the

MA

study. A PS included imaging of the head, cervical-spine, chest, and abdomen/pelvis. Imaging of the maxillary face was optional. Each CT scan was a separate protocol and a set of images

ED

generated as a no single run PS protocol was designed. Head, maxillary face and cervical-spine imaging was done without intravenous contrast, while chest and abdomen/pelvis imaging was

PT

performed with contrast. CT scans were read by the radiology attending and by an off-site

CE

radiology service during off hours. All scans read off-site were confirmed the following day by the radiology attending. In all cases the chief trauma resident and trauma attending reviewed the

Setting

AC

CT imaging and reports.

Jamaica Hospital has three tiers of trauma activation. Trauma activations are managed by a team consisting of an emergency physician, an attending trauma surgeon, a resident trauma team, an anesthesia attending, a trauma program manager or nursing supervisor, a respiratory therapist, a radiology technician, social services, two ED registered nurses, and security. The trauma surgeon and the anesthesia attending are only required to be present at tier 1 activations. For tier 2 activations, the chief trauma resident is required to be present while the presence of the trauma

ACCEPTED MANUSCRIPT surgeon and attending anesthesiologist is optional unless deemed necessary. Tier 3 trauma activations are consults evaluated by the trauma surgery resident and were not included in the

RI P

T

study. Selection of Participants

SC

All blunt trauma patients over the age of 18 years who were treated as trauma activations during

NU

the study period were included. Patients were excluded if they were under 18 years of age, expired in the trauma bay prior to CT scanning, pregnant, transferred directly to the operating

MA

room without CT scanning, or were downgraded from trauma activation status before CT scanning.

ED

Identification of “Missed Injuries”

PT

The final decision to image the patient was made by the trauma team leader, meaning the trauma

CE

surgeon or chief resident. Any injuries not detected in the initial scans ordered during trauma assessment but later identified were considered missed injuries. Extremity injuries were excluded

AC

except for proximal humerus and proximal femur injuries which were considered missed if no chest or pelvis CT were ordered respectively. Injuries that were clinically obvious or noted on plain film were not included. A trauma surgeon and ED attending independently reviewed electronic medical records to identify missed injuries. Missed injuries were categorized into minor and major. Minor was defined as injuries that would be generally manageable on an outpatient basis if they were found in isolation. While major was defined as an injury that in and of itself would require admission or operation.

ACCEPTED MANUSCRIPT Incidental Findings Incidental findings were defined as any unexpected finding not related to trauma. Exclusions

RI P

T

included age appropriate changes, mild degenerative joint disease, signs of previous surgery or trauma, and findings already known from previous imaging. Minor incidental findings were

SC

defined as findings that were clinically insignificant or findings not requiring intervention or

intervention or follow-up within 30 days.

MA

Other Variables

NU

follow-up within 30 days. Major incidental findings were defined as those requiring immediate

Patient electronic medical records were reviewed for demographics, CT findings, and patient

ED

disposition. Age, gender, weight, height, Injury Severity Score (ISS) (based on AIS 2005),

PT

Glasgow Coma Scale (GCS), mortality, initial pulse rate, initial respiratory rate, initial systolic BP, and alcohol and drug use were obtained from the electronic medical records. Length of stay

CE

in the hospital, surgical intensive care unit (SICU), and ED were also recorded. Fixed costs were

AC

broadly defined as those that are unaffected by the patient volume while variable costs change with volume, meaning they would descend to zero with no patients. Cost data was obtained from our institution’s cost center database. Radiation Radiation dose for each patient was obtained from the dose report for each CT scan obtained by the patient. Radiation doses were obtained for scans ordered during the trauma assessment and for scans ordered after the assessment including after hospital admission. The total dose length product (DLP, mGy/cm) was used to calculate the effective dose (E, mSv) using the formula E=DLP x k where k is the tissue weighting factor based on the body region scanned. Based on

ACCEPTED MANUSCRIPT the 2007 International Commission on Radiation Protection recommendations, the following k values were used: head - 0.002, neck - 0.0059, chest - 0.014, abdomen - 0.015. [21] The CT

RI P

T

scanner (GE Optima CT660) was unchanged during the study period. Data analysis

SC

To analyze data GraphPad Prism 6 (GraphPad Software Inc., La Jolla, CA) was used, p <0.05

NU

was considered significant. Categorical variables are expressed as count (n) and percent (%) and continuous variables are expressed as mean and standard deviation (SD) or median and

MA

interquartile range (IQR) where relevant. Significance testing was done using Fisher’s exact test on categorical variables. For continuous variables, the sample was first assessed for normality

ED

using the Shapiro-Wilk normality test followed by the Mann-Whitney non-parametric test. The Welch’s t test was used to calculate a 95% confidence interval (CI) for the difference between

CE

Results

PT

the means.

AC

Characteristics of study subjects There were a total of 608 trauma activations during the six month study period. Penetrating trauma accounted for 121 of the activations, blunt trauma accounted for 484 and burns accounted for three trauma activations. Of the 484 blunt trauma activations, 426 were eligible for study inclusion. A total of 220 (51.6%) patients were treated during the pre-PS period and 206 (48.4%) patients during the PS period. The mean age of the 426 eligible patients was 48.8 years (SD: 21.5) and 285 (66.9%) were male. Most patients (88.3%) had a GCS in the range of 12-15. A total of 50 (11.7%) patients required intubation in the trauma bay. The average length of stay, for

ACCEPTED MANUSCRIPT the 270 (63.4%) admitted patients, was 4.5 days (SD: 8.3). The median injury severity score (ISS) was 4 (IQR: 1, 9) and the mortality rate was 2.8%. See Table 1 for additional details.

RI P

T

Main Results

A total of seven (3.2%) patients during the pre-PS period and one (0.5%) patient during the PS

SC

period had missed injuries. Only one injury was missed per patient. There was a 2.7% decrease

NU

in the number of missed injuries during the PS period vs. the pre-PS period. Patients during the pre-PS period were 6.7 times (95% CI: 0.82 to 55.3) more likely to have a missed injury when

MA

compared to the patients in the PS period. During the pre-PS period, three (42.9%) of the missed injuries were considered major. Major injuries decreased to 0% during the PS period. The single

ED

missed injury during the PS period was considered minor. There was a 14.4 % increase in the number of patients with incidental findings and the average number of incidental findings

PT

increased by 1.3 (95% CI: 0.8 to 1.8) during the PS period. Additional information is presented

CE

in Tables 2 and 3.

AC

To determine if introduction of a PS protocol changed trauma management/outcome various time intervals were analyzed (Table 4). Introduction of a PS protocol increased the length of stay in the hospital by 1.3 days (95% CI: -0.33 to 2.88) and the SICU by 2.3 days (95% CI: -2.62 to 7.23). However, this increase was not significant. When patients were stratified based on an ISS > 8, there was still no significant difference in the length of stay. The average time spent in the ED was significantly reduced by 68.2 minutes (95% CI: -134.4 to -2.1) during the PS period when compared to the pre-PS period. Moreover, patients with an ISS > 8 spent substantially less time in the ED (-182 minutes, 95% CI: -326.4 to -37.63) during the PS period. The introduction

ACCEPTED MANUSCRIPT of the PS protocol also significantly reduced the time to the operating room by 1,465 minutes (95% CI: -2,519 to -411). Detailed information is presented in Table 4.

RI P

T

The cost of implementing a PS protocol was also evaluated (Table 5). Introduction of the PS protocol decreased the average radiology cost per patient by $50 (95% CI: -271.1 to 171.4).

SC

However, the average cost of patient care increased by $4,971 (95% CI: -1,100 to 11,042) during the PS period. When radiology cost was separated into fixed vs. variable and CT vs. x-ray, the

NU

introduction of the PS protocol increased the total fixed CT cost per patient by $48.1 (95% CI:

MA

32 to 64.1) (Table 5). Radiation exposure was also analyzed (Table 6). The radiation dose for each CT scan performed on a patient during and after the trauma assessment was obtained. As

ED

the number of CT scans increased during the PS period, the average radiation dose per patient

PT

increased by 8.2 mSv (95% CI: 5.0 to 11.3). Limitations

CE

The main limitation of the study is that it was done at a single center; therefore the results cannot

AC

be generalized. This was not a randomized study, which would have been a better approach to this study. In addition, the PS protocol was not a single pass scan, each scan was done separately. Using a single run protocol might have reduced the radiation exposure. Another limitation was the cost data, the radiology cost was not separated into ED vs. non-ED radiology cost. However, the majority of scans were ordered in the ED prior to hospital admission. Lastly, in the context of trauma, radiologists may have excluded some minor incidental findings while focusing on traumatic injuries.

ACCEPTED MANUSCRIPT Discussion There is debate over whether or not the benefits of pan-computed tomography outweigh the

RI P

T

risks. Generally, PSs are reported to decrease mortality, missed injuries, and treatment times at the expense of increased radiation, incidental findings, and hospital costs. No single study has

SC

attempted to address all these factors in a prospective analysis to determine what the effects would be on each factor with the introduction of a PS paradigm in a previously selective scan

NU

environment. A recent randomized study, focused on severely injured patients, addressed some

MA

of these factors but did not address time to the operating room, radiology cost, missed injuries, and incidental findings.[17] In this study, we determined that a PS protocol during the initial

ED

trauma assessment minimized missed injuries, and decreased various time intervals while increasing incidental findings, radiation exposure, and scanning cost. Mortality was not affected.

PT

Injuries were considered missed if they were not identified during the initial trauma assessment

CE

but were later found by either x-ray or CT. Prior to the introduction of the PS protocol, there

AC

were seven patients with missed injuries. After the PS was introduced, only one patient had a missed injury. Of the eight patients with missed injuries, two (25%) had a PS and six (75%) were selectively imaged or not imaged at all. It is important to note that the one patient with a missed injury, during the PS period, had a PS. However, an orbital roof fracture was only detected on a later CT image of the maxillary face. If a PS was initially done during trauma triage many of the missed injuries would have been detected earlier. This suggests that a PS may decrease the number of missed injuries in blunt trauma patients. The impact of pan-CT on incidental findings has not been widely reported. As expected, the increase in scans during the PS period increased the number of incidental findings. However,

ACCEPTED MANUSCRIPT implementation of a pan-scan protocol did not significantly increase the number of clinically relevant incidental findings. During the pan-scan period, five incidental findings required follow-

T

up within 30 days. In our population, there was a high rate of incidental findings, which poses a

RI P

concern. We predominantly serve uninsured and under-served communities that have limited access to primary preventative care, which may account for the high percentage of patients with

SC

incidental findings. The discovery of incidental findings in trauma patients poses a burden on the

NU

health care system and increases cost due to the additional evaluation required to prevent adverse outcomes. However, the major concern is the unnecessary expenditure for inconsequential

MA

incidental findings identified on whole body CT.

ED

Several studies have demonstrated that pan-computed tomography decreased the length of stay in the hospital and the length of time to diagnosis and treatment. The introduction of the PS

PT

protocol did reduce the time spent in the ED and the time to the first operating room visit. These

CE

two time intervals are likely indicators of the time to diagnosis and treatment and their reduction during the PS period in the face of no other changes strongly implies a faster progress of patients

time to care.

AC

through the hospital system. Our study confirms that pan-computed tomography accelerates the

Contrary to other studies, the mortality rate did not significantly decline after the introduction of the PS protocol. In fact, during the PS period the mortality rate doubled, but this increase was not significant. This is likely due to the relatively small number of patients that are included in this study; however larger randomized studies had similar results.[17] Of note, injury severity was similar before and after the introduction of the PS protocol. There was a 6.4% increase in patients with an initial GCS of 3 during the PS period, which may account for the twofold increase in mortality. In this study, there was no apparent correlation between CT scans and

ACCEPTED MANUSCRIPT mortality. Previous studies that found this association are registry based and had two confounding factors in their design. First, the most critically ill patients, who have the highest

T

mortality, cannot tolerate any CT scans and are therefore relegated to the non-PS group. Second,

RI P

newer centers that deliver state of the art care are more likely to have a CT scanner integrated into their emergency department. In theory, patients at these centers will have a better survival

SC

rate because a disproportionately larger percentage of patients will receive a PS compared to

NU

those at older less advanced hospitals, by this means skewing the populations.

MA

A PS is considered a useful tool for patients who have a diminished level of consciousness and who are severely injured or are suspected to have multiple injuries. In our population, 11.7% of

ED

the patients had a GCS < 12. However, when patients were sub-grouped based on GCS < 12 there was no change in time to care or number of missed injuries after the introduction of the PS

PT

protocol. Injury severity was similar between the two groups. As expected, patients with an

CE

injury severity score greater than eight had a shorter length of stay in the ED. After the introduction of the PS protocol, the average length of stay in the ED further decreased by three

AC

hours for patients with the most severe injuries. Therefore, injury severity but not GCS may be a better indication for a PS. Presently, injury severity is calculated retrospectively, therefore new prospective ways of determining injury severity in the trauma bay are needed. It is important to note that 44.1% of the patients during the pre-PS period had a PS and 8.25% of the patients during the PS period did not have a PS. With the introduction of the PS protocol, there was a shift from 55.9% to 8.25% of blunt trauma activations receiving selective CT scanning and from 44.1% to 91.75% receiving a PS. This ~48% increase was sufficient to decrease missed injuries by 3.1%, length of stay in the ED by 68.2 minutes, and time to the first

ACCEPTED MANUSCRIPT operating room visit by 24.4 hours. This suggests that there is a significant benefit to utilizing pan-computed tomography in blunt trauma patients.

RI P

T

Since the BMI of the groups pre- and post- PS introduction was similar, the specific radiation dose for each patient was determined. In general, our population was overweight therefore

SC

radiation exposure was high. However, the average radiation exposure increased by 8.2 mSv during the PS period. This was due to the increase in the number of scans ordered during the

NU

trauma assessment. The total number of scans ordered after the trauma assessment did not

MA

change after the introduction the PS protocol. Many of the scans ordered later in the hospital stay were repeat head CT scans for known traumatic brain injuries and CT scans of the chest and

ED

abdomen/pelvis. In our population, the average effective radiation dose for a PS was 23.8 mSv which is similar to reported averages of 24 mSv.[7, 22] According to Verdun et al., an effective

PT

dose of 10-100 mSv is considered low risk (10-3) for cancer, while >100 mSv is considered

CE

moderate risk (>10-2). [23] In our study, 99.5% of patients were low risk for cancer and 0.5% had a moderate risk of cancer, an insubstantial number when extrapolated to the entire population of

AC

blunt trauma patients nation-wide. However, these results do not take into account age, gender, and natural risk which all contribute to the risk of death from cancer. Additionally, the overall radiation exposure could have been significantly reduced by the utilization of a defined single run PS protocol as opposed to simply performing each individual component of the PS separately and combining them. Eliminating the overlap between scans and the reduced radiation needed for such a CT can dramatically reduce exposure. Not much has been done to determine if selective scanning is more cost effective than whole body CT. In this study, both total cost and radiology cost were analyzed. The average cost to treat a blunt trauma activation patient increased by $4,971 during the PS period. However, many

ACCEPTED MANUSCRIPT factors, such as co-morbidities, may contribute to this increase. The demographics and medical status of patients in both the pre-PS and PS periods were similar; however the small size of the

T

groups indicates that a few outliers could have greatly affected the data. In the PS group there

RI P

was an increase in the number of patients with a GCS of 3, the number of patients requiring intubation, the number of patients who went directly to the OR from the ED, and the length of

SC

stay in the hospital. Whether these factors facilitated an increase in the cost is unknown, but each

NU

of these factors can negatively impact cost. On the other hand, the introduction of the PS protocol did not have a significant effect on the total radiology cost, which is where an effect of

MA

PS should most readily be evident. As expected, the fixed CT cost increased during the PS period. When radiology cost was separated into CT and diagnostic x-ray cost, the cost of

ED

diagnostic x-rays decreased during the PS period. This suggests that there may have been a

PT

decrease in the number of x-rays performed during the PS period, which cancelled out the

CE

increase in the fixed CT cost.

In conclusion, despite the outcomes of the study it is difficult to justify the use of a PS on every

AC

blunt trauma patient. For intoxicated and intubated patients a PS may be necessary since performing a physical exam may be challenging. In our population, ~33% of the patients were positive for alcohol use; ~ 9% were positive for illegal drug use, and ~12% were intubated in the trauma bay. In trauma care, over-triage and under-triage of patients can be a problem. The introduction of a PS protocol raised the possibility of over-triage in some patients. In our study, 67.4% of patients had an ISS < 8, 65 % of which had a PS. Not to mention, only 30% of the patients who had a PS had an injury in two or more body regions and only 2.8% had an injury in all the four components of a PS. Additionally, there were only five major incidental findings identified during the PS period, with none requiring immediate intervention. Moreover, ~ 80%

ACCEPTED MANUSCRIPT of the scans done during the PS period were negative, which makes it difficult to justify the necessity of a PS for all blunt trauma activations. This is especially true when cost and the

RI P

T

potential long-term risk of radiation exposure are taken into consideration.

Study concept and design – MKJ, SDS, SWL, GKD

NU

Data collection – MKJ, SWL, SDS, MPF

SC

Author Contribution:

MA

Data analysis & interpretation – MKJ, SDS, SWL

ED

Manuscript preparation & critical revision – MKJ, SDS, SWL Acknowledgements

PT

The authors thank Gideon Yoeli, MD from the Department of Radiology at JHMC for his advice

AC

CE

on radiation dose and incidental findings.

ACCEPTED MANUSCRIPT References [1.]

Salim A, Sangthong B, Martin M, Brown C, Plurad D, Demetriades D: Whole body

RI P

T

imaging in blunt multisystem trauma patients without obvious signs of injury: results of a

[2.]

SC

prospective study. Arch Surg 2006, 141(5):468-473; discussion 473-465.

Surendran A, Mori A, Varma DK, Gruen RL: Systematic review of the benefits and

NU

harms of whole-body computed tomography in the early management of multitrauma

MA

patients: are we getting the whole picture? J Trauma Acute Care Surg 2014, 76(4):1122-

[3.]

ED

1130.

Gunn ML, Kool DR, Lehnert BE: Improving Outcomes in the Patient with Polytrauma: A

McCollough CH, Guimaraes L, Fletcher JG: In defense of body CT. AJR Am J

AC

[4.]

CE

53(4):639-656, vii.

PT

Review of the Role of Whole-Body Computed Tomography. Radiol Clin North Am 2015,

Roentgenol 2009, 193(1):28-39.

[5.]

Marin D, Nelson RC, Rubin GD, Schindera ST: Body CT: technical advances for improving safety. AJR Am J Roentgenol 2011, 197(1):33-41.

[6.]

Hui CM, MacGregor JH, Tien HC, Kortbeek JB: Radiation dose from initial trauma assessment and resuscitation: review of the literature. Can J Surg 2009, 52(2):147-152.

ACCEPTED MANUSCRIPT [7.]

Asha S, Curtis KA, Grant N, Taylor C, Lo S, Smart R, et al.: Comparison of radiation exposure of trauma patients from diagnostic radiology procedures before and after the

[8.]

RI P

T

introduction of a panscan protocol. Emerg Med Australas : EMA 2012, 24(1):43-51.

Beatty L, Furey E, Daniels C, Berman A, Tallon JM: Radiation Exposure From CT

Beinfeld MT, Wittenberg E, Gazelle GS: Cost-effectiveness of whole-body CT

MA

[9.]

NU

Experience. CJEM 2015, 17(6):617-623.

SC

Scanning in the Resuscitative Phase of Trauma Care: A Level One Trauma Centre

Lee WS, Parks NA, Garcia A, Palmer BJ, Liu TH, Victorino GP: Pan computed

PT

[10.]

ED

screening. Radiology 2005, 234(2):415-422.

tomography versus selective computed tomography in stable, young adults after blunt

CE

trauma with moderate mechanism: a cost-utility analysis. J Trauma Acute Care Surg

[11.]

AC

2014, 77(4):527-533; discussion 533.

Weninger P, Mauritz W, Fridrich P, Spitaler R, Figl M, Kern B, et al.: Emergency room management of patients with blunt major trauma: evaluation of the multislice computed tomography protocol exemplified by an urban trauma center. J Trauma 2007, 62(3):584591.

ACCEPTED MANUSCRIPT [12.]

Wurmb TE, Fruhwald P, Hopfner W, Keil T, Kredel M, Brederlau J, et al.: Whole-body multislice computed tomography as the first line diagnostic tool in patients with multiple

[13.]

RI P

T

injuries: the focus on time. J Trauma 2009, 66(3):658-665.

Hutter M, Woltmann A, Hierholzer C, Gartner C, Buhren V, Stengel D: Association

SC

between a single-pass whole-body computed tomography policy and survival after blunt

NU

major trauma: a retrospective cohort study. Scand J Trauma, Resusc Emerg Med 2011,

[14.]

MA

19:73.

Kimura A, Tanaka N: Whole-body computed tomography is associated with decreased

ED

mortality in blunt trauma patients with moderate-to-severe consciousness disturbance: a

Huber-Wagner S, Biberthaler P, Haberle S, Wierer M, Dobritz M, Rummeny E, et al.:

CE

[15.]

PT

multicenter, retrospective study. J Trauma Acute Care Surg 2013, 75(2):202-206.

AC

Whole-body CT in haemodynamically unstable severely injured patients-a retrospective, multicentre study. PloS One 2013, 8(7):e68880.

[16.]

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: a systematic review and meta-analysis. J Trauma Acute Care Surg 2014, 77(4):534-539.

[17.]

Sierink JC, Treskes K, Edwards MJ, Beuker BJ, den Hartog D, Hohmann J, et al.: Immediate total-body CT scanning versus conventional imaging and selective CT

ACCEPTED MANUSCRIPT scanning in patients with severe trauma (REACT-2): a randomised controlled trial.

van Vugt R, Kool DR, Deunk J, Edwards MJ: Effects on mortality, treatment, and time

RI P

[18.]

T

Lancet 2016.

management as a result of routine use of total body computed tomography in blunt high-

[19.]

NU

SC

energy trauma patients. J Trauma Acute Care Surg 2012, 72(3):553-559.

Tillou A, Gupta M, Baraff LJ, Schriger DL, Hoffman JR, Hiatt JR, et al.: Is the use of

Gupta M, Schriger DL, Hiatt JR, Cryer HG, Tillou A, Hoffman JR, et al.: Selective use of

PT

[20.]

ED

Trauma 2009, 67(4):779-787.

MA

pan-computed tomography for blunt trauma justified? A prospective evaluation. J

computed tomography compared with routine whole body imaging in patients with blunt

[21.]

AC

CE

trauma. Ann Emerg Med 2011, 58(5):407-416 e415.

The 2007 Recommendations of the International Commission on Radiological Protection. ICRP publication 103. Ann ICRP 2007, 37(2-4):1-332.

[22.]

Sierink JC, Saltzherr TP, Wirtz MR, Streekstra GJ, Beenen LF, Goslings JC: Radiation exposure before and after the introductionof a dedicated total-body CT protocol in multitrauma patients. Emerg Radiol 2013, 20(6):507-512.

ACCEPTED MANUSCRIPT Verdun FR, Bochud F, Gundinchet F, Aroua A, Schnyder P, Meuli R: Quality initiatives* radiation risk: what you should know to tell your patient. Radiographics 2008,

CE

PT

ED

MA

NU

SC

RI P

T

28(7):1807-1816.

AC

[23.]

ACCEPTED MANUSCRIPT Table 1. Characteristics of the eligible patients.

PRE-PAN SCAN

PAN SCAN PERIOD

PERIOD

(N= 206)

T

ALL PATIENTS

N

RI P

VARIABLE

SC

(N = 220)

426

48.8 (21.5)

48.4 (21.2)

49.1 (21.8)

MALE

426

285 (66.9)

135 (61.4)

150 (72.8)

FEMALE

426

141 (33.1)

85 (38.6)

56 (27.2)

WEIGHT (LBS)

418

170.7 (38.6)

168 (36)

173.6 (41.1)

HEGHT (INCHES)

408

66.9 (5.5)

66.2 (3.7)

67.7 (6.9)

406

26.9 (5.3)

27.0 (5.0)

26.9 (5.6)

426

18 (4)

19 (5)

18 (3)

426

89 (18)

90 (18)

88 (18)

426

144 (27)

144 (26)

143 (29)

426

-

-

-

3

19 (4.5)

3 (1.4)

16 (7.8)

4-7

7 (1.6)

6 (2.7)

1 (0.5)

8 - 11

24 (5.6)

12 (5.4)

12 (5.8)

12 - 15

376 (88.3)

199 (90.4)

177 (85.9)

101 (23.7)

42 (19.1)

59 (28.6)

INITIAL GCS

MA

ED

AC

INITIAL SYSTOLIC BP

CE

INITIAL RESPIRATORY RATE

TIER 1 ACTIVATION

PT

BMI (KG/M2 )

INITIAL PULSE

NU

AGE (YEARS)

426

ACCEPTED MANUSCRIPT 426

325 (76.3)

178 (80.9)

147 (71.3)

ALCOHOL USE

401

133 (33.2)

74 (33.6)

59 (28.6)

DRUG USE

275

26 (9.4)

4 (2.7)

22 (10.7)

INTUBATED IN TRAUMA BAY

426

50 (11.7)

ED DISPOSITION

426

-

RI P

T

TIER 2 ACTIVATION

28 (13.6)

-

-

131 (59.5)

139 (67.5)

90 (68.7)

89 (64)

75 (27.8)

36 (27.5)

39 (28)

16 (6.0)

5 (3.8)

11 (7.9)

156 (36.6)

89 (40.4)

67 (32.5)

426

92 (21.6)

46 (20.9)

46 (22.3)

426

-

-

-

MINOR (0 - 8)

287 (67.4)

142 (64.5)

145 (70.4)

MODERATE (9-15)

89 (20.9)

52 (23.6)

37 (18)

SEVERE (16-24)

40 (9.4)

20 (9.1)

20 (9.7)

CRITICAL (≥ 25)

10 (2.3)

6 (2.7)

4 (1.9)

270 (63.4)

NU

ADMITTED

SC

22 (10)

179 (66.3)

SICU

ED

DIRECT TO OPERATING ROOM

MA

FLOOR

CE

# OF PATIENTS REQUIRING SURGERY

PT

DISCHARGED

AC

INJURY SEVERITY SCORE

LENGTH OF STAY (DAYS)

426

4.5 (8.3)

3.9 (6.8)

5.2 (9.6)

MORTALITY

426

12 (2.8)

4 (1.8)

8 (3.9)

BMI, body mass index; BP, blood pressure; GCS, Glasgow coma scale; ED, emergency department; SICU, surgical intensive care unit; N, number in population. Categorical variables are expressed as count (%) and continuous variables are expressed as mean (SD).

ACCEPTED MANUSCRIPT Table 2. Analysis of missed and incidental injuries pre- and post-PS introduction. PAN SCAN PERIOD (N=206)

(n, %)

(n, %)

NUMBER OF PATIENTS WITH MISSED INJURIES

7 (3.2)

1 (0.5)

NUMBER OF MISSED INJURIES

7 (1.25)

1 (0.2)

OR (95% CI)

P VALUE

RI P

T

PRE-PAN SCAN PERIOD (N=220)

INJURIES

0.069

6.2 (0.76 to 50.51)

0.074

1 (100)

0.4 (0.01 to 14.1)

1.000

0 (0)

2.3 (0.07 to 76.7)

1.000

137 (62.3)

158 (76.7)

0.5 (0.33 to 0.76)

0.002*

TOTAL: 349

TOTAL: 590

MEAN: 1.6 (SD: 1.9)

MEAN: 2.9 (SD: 3.1)

1.3 (0.8 to 1.8)

INCIDENTAL, MINOR

348 (99.7)

585 (99.15)

2.9 (0.3 to 25.6)

0.421

INCIDENTAL, MAJOR

1 (0.3)

5 (0.85)

0.3 (0.04 to 2.9)

0.421

3 (42.9)

PT

CE

NUMBER OF PATIENTS WITH INCIDENTAL FINDINGS

AC

NUMBER OF INCIDENTAL FINDINGS

NU

MISSED, MAJOR

MA

4 (57.1)

ED

MISSED, MINOR

SC

6.7 (0.82 to 55.3)

DIFFERENCE: <0.0001*

OR, odds ratio; CI, confidence interval; N, number of patients in population; n, count; SD, standard deviation; * denotes p<0.05

ACCEPTED MANUSCRIPT Table 3. Missed Injuries. TYPE OF MISSED INJURY

IMAGING ON ARRIVAL

OTHER INJURIES

Motorcyclist vs. motor vehicle

CT Head, Maxillary Face

Subdural hematoma with midline shift of 1cm, vault of skull fracture, tibia fracture, clavicle fracture, facial laceration

Fall

CT Head

(4 , bilateral 5 )

Driver in motor vehicle accident

CT Head, Cspine, Abdomen/Pelvis

Intertrochanteric femur fracture

Pedestrian vs. motor vehicle

TREATMENT OF MISSED INJURY

DISPOSITION

SC

Rib fracture (1)

RI P

T

MECHANISM OF INJURY

Expired

Concussion

None

Transferred

Concussion, contusion of knee, rib th fractures (7-10 )

None

Home

None

None

ORIF Hip

Home

Fall

None

None

ORIF Hip/Femur

Rehabilitation

Nasal bone fracture

Fall

CT Head, Cspine

Open wound of nose

Closed reduction of fracture

Home

Orbital roof fracture

Pedestrian vs. motor vehicle

CT Head, Cspine, Chest, Abdomen/pelvis

Facial laceration

None

Home

th

Subtrochanteric femur fracture

MA

ED

th

PT

Rib fractures (3)

CE

Subcapital femur fracture

AC

(12 )

NU

None

th

None

Home

RI P

CT Head, Cspine, Chest, Abdomen/pelvis

Fall

SC

Temporal bone fracture

Subarachnoid hemorrhage, dislocation of finger PIP joint, rib fractures, lung contusion/laceration, pneumothoraces, transverse processes vertebra fracture, shattered spleen and left kidney

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

ORIF, open reduction and internal fixation; CT, computed tomography. All “missed” injuries were later identified by either x-ray or CT, after evaluation by the trauma team.

ACCEPTED MANUSCRIPT Table 4. Analysis of various time intervals pre- and post-PS introduction.

P VALUE

(95% CI)

(MEAN, SD)

SC

(MEAN, SD)

DIFFERENCE

T

TIME VARIABLES

PAN SCAN PERIOD (N=206)

RI P

PRE-PAN SCAN PERIOD (N=220)

1.3 (-0.33 to 2.88)

0.239

5.2 (0.99 to 9.34)

0.060

9.7 (12.7)

2.3 (-2.62 to 7.23)

0.855

11.8 (13.8)

3.6 (-2.32 to 9.46)

0.411

459.1 (422.8)

390.9 (256.2)

-68.2 (-134.4 to 2.1)

0.026*

411.0 (614.5)

229.0 (164.2)

-182.0 (-326.4 to -37.63)

0.0008*

TIME TO 1 OPERATING ROOM VISIT (MINS)

2946 (2953)

1481 (2047)

-1465 (-2519 to 411)

0.002*

- DIRECT TO OPERATING ROOM (MINS)

100.2 (43.8)

136.5 (68.2)

36.3 (-25.65 to 98.16)

0.304

2946 (3112)

1621 (2353)

-1325 (-2721 to 70.74)

0.012*

LENGTH OF STAY (DAYS)

5.2 (9.6)

NU

3.9 (6.8)

LENGTH OF STAY (DAYS) 7.7 (8.9)

12.9 (14.4)

LENGTH OF STAY IN SICU (DAYS)

AC

LENGTH OF STAY IN ED (MINS)

CE

LENGTH OF STAY IN ED (MINS)

(ISS >8)

8.2 (11.6)

PT

(ISS>8)

7.4 (10.9)

ED

LENGTH OF STAY IN SICU (DAYS)

MA

(ISS >8)

ST

st

TIME TO 1 OPERATING ROOM VISIT (MINS) (ISS >8)

ISS, injury severity score; SICU, surgical intensive care unit; ED, emergency department, MINS, minutes; CI, confidence interval; SD, standard deviation; N, number of patients; * denotes p<0.05

ACCEPTED MANUSCRIPT Table 5. Cost analysis pre- and post-PS introduction.

COST VARIABLES

PAN SCAN PERIOD (N=206)

DIFFERENCE P VALUE (95% CI)

T

PRE-PAN SCAN PERIOD (N=220)

(MEAN, SD)

HOSPITAL STAY COST ($)

13,573 (25,507)

18,544 (36,800)

RADIOLOGY COST ($)

1,362 (1,297)

1,312 (1,017)

FIXED RADIOLOGY COST ($)

618 (565)

VARIABLE RADIOLOGY COST ($)

744 (735)

CT SCAN COST ($)

361 (205)

RI P

(MEAN, SD)

0.010*

-50 (-271.1 to 171.4)

0.629

47 (-55.4 to 148.5)

0.042*

647 (511)

-96 (-216.3 to 23.53)

0.371

415 (132)

54 (21.3 to 86.5)

<0.0001*

174 (97)

222 (70)

48.1 (32 to 64.1)

<0.0001*

VARIABLE CT SCAN COST ($)

187 (109)

194 (63)

6.8 (-10 to 23.7)

0.079

DIAGNOSTIC X-RAY COST ($)

1,001 (1,195)

897 (957)

-104 (-309.3 to 101.9)

0.213

FIXED DIAGNOSTIC XRAY COST ($)

445 (517)

443 (474)

-1.5 (-95.9 to 92.9)

0.581

VARIABLE DIAGNOSTIC X-RAY COST ($)

557 (679)

453 (483)

-103.2 (-214.9 to 8.5)

0.066

NU MA

665 (506)

ED

PT

CE

AC

FIXED CT SCAN COST ($)

SC

4971 (-1100 to 11042)

ACCEPTED MANUSCRIPT CI, confidence interval; SD, standard deviation; CT, computed tomography; N, number in population; * denotes p<0.05

PAN SCAN PERIOD

RI P

PRE-PAN SCAN PERIOD (N=220)

# OF PATIENTS RECEIVING SELECTIVE SCANS

123 (55.9%)

# OF PATIENTS RECEIVING PAN SCAN

97 (44.1%)

CT SCANS ORDERED DURING

Total: 737

Total: 883

TRAUMA ASSESSMENT

ED

SC

(N=206)

Mean: 3.3

Mean: 4.3

AC

CT SCANS ORDERED AFTER TRAUMA ASSESSMENT

RADIATION DOSE (mSv)

TOTAL SCANS

RADIATION DOSE (mSv)

NU

17 (8.25%)

MA

PT

CE

RADIATION DOSE (mSv)

T

Table 6. Analysis of radiation exposure pre- and post-PS introduction.

189 (91.75%)

DIFFERENCE P VALUE

(95% CI)

47.6% (38.7 to 56.6)

< 0.0001*

47.6% (38.7 to 56.6)

< 0.0001*

0.94 (0.7 to 1.1)

<0.0001*

8.4 (5.8 to 10.9)

<0.0001*

0.581

18.4 (SD:15.1)

26.8 (SD:12.0)

Total: 47

Total: 46

Mean: 0.2

Mean: 0.2

0.01 (-0.1 to 0.12)

1.5 (SD:7.3)

1.3 (SD:6.9)

-0.2 (-1.6 to 1.1)

0.566

Total: 784

Total: 929 0.9 (0.7 to 1.2)

<0.0001*

Mean: 3.6

Mean: 4.5

19.9 (SD:18.9)

28.1 (SD:14.3)

8.2 (5.0 to 11.3)

<0.0001*

mSv, millisieverts; CI, confidence interval; SD, standard deviation; N, number of patients in population; #, number; *denotes p<0.05