a b s t r a c t

Background:

Pediatric cervical spine (CSI) and blunt cerebrovascular injuries (BCVI) are challenging to evaluate as they are rare but carry high morbidity and mortality. CT scans are the traditional imaging modality to evaluate for CSI/ BCVI, but involve radiation exposure.and potential future increased risk of malignancy. Therefore, we present re- sults from the implementation of a combined CSI/BCVI pediatric trauma clinical pathway to aid clinicians in their decision-making.

Methods:

We conducted a 2-year retrospective cohort study analyzing data pre and post implementation of the combined CSI/BCVI pathway. Data was obtained from a level 1 pediatric trauma center and included blunt trauma patients under the age of 14. We evaluated the use of cervical spine computed tomography (CT), CT angiography, and plain radiographs, as well as missed injuries and provider pathway adherence.

Results:

We included 358 patients: 209 pre-pathway and 149 post-pathway implementation. Patient mean age was 8.9 years and 61% were male (61% males). There were no significant differences in GCS, AIS, and ISS between pre and post pathway groups. Post pathway implementation saw reduced use of cervical spine CT, although this was not clinically significant (33% vs 31%, p = 0.74). However, cervical spine radiography use increased (9% vs 16%, p = 0.03), and there was also an increase in screening for BCVI injuries with higher use of CTA (5% vs 7%, p = 0.52). A total of 12 CSI and 3 BCVI were identified with no missed injuries. Provider adherence to the pathway was modest (54%).

Conclusion Implementation of a combined CSI/BCVI clinical pathway for Pediatric trauma patients increased screening radi- ography and did not miss any injuries. However, CT use did not significantly decrease and provider adherence was modest, supporting the need for further implementation analysis and larger studies to validate the pathway’s sensitivity and specificity for CSI/BCVI.

(C) 2021

1. Introduction

Unintentional injuries are the leading cause of death for children of all ages with over 140,000 Trauma victims under the age of 19 years in 2016 [1,2]. Although cervical spine injury (CSI) is rare in pediatric trauma patients (estimated incidence of 1-2%), it is nonetheless devas- tating with morbidity rates exceeding those of adults [3-5]. Therefore, a careful and evidence-based evaluation for CSI in pediatric trauma pa- tients is crucial as a missed injury could have significant sequelae.

* Corresponding author at: Department of Emergency Medicine, University of Florida College of Medicine - Jacksonville, 655 West 8^th St., Jacksonville, FL 32209, USA.

E-mail addresses: [email protected] (M. Schonenberg Llach), [email protected] (J.N. Fishe), [email protected] (B.K. Yorkgitis).

Computed tomography (CT) is the traditional imaging modality to de- tect CSI given its sensitivity to detect bony injuries, cost-effectiveness, wide availability, and rapid turnaround [3,6]. However CT, especially of the neck in younger pediatric patients, is associated with future potential malignancy stemming from radiation exposure [3,7,8]. Therefore, eval- uation for pediatric CSI must balance the risk of injury versus the risks of radiation exposure.

Like CSI, another rare but potentially devastating pediatric trau- matic injury is blunt cerebrovascular injury (BCVI). The estimated in- cidence of pediatric BCVI is 0.3%-0.9% [9]. However, the true incidence may be higher as BCVI symptoms may be confounded by other injuries and CT angiography (CTA), the gold standard for diag- nosis [10], is infrequently used in pediatric trauma patients [9,11]. As with CSI, undiagnosed and/or untreated BCVI can result in serious

https://doi.org/10.1016/j.ajem.2021.03.024

0735-6757/(C) 2021

complications, including stroke in 26-38% of cases [10-12]. Further complicating the diagnosis of BCVI is that its neurologic sequelae may be delayed by hours or even days [10-13].

Currently there are several validated clinical decision tools to screen adult trauma patients for CSI and help guide CT use [14-16]. However, those tools have produced mixed results when tested in a pediatric pop- ulation, likely owing to the lower incidence of PEdiatric CSI and ana- tomic differences that result in different injury patterns [5,17]. In the absence of a validated and widely promulgated pediatric CSI screening tool, individual institutions have developed their own clinical pathways to guide CSI imaging. Most publicly-available pathways are from high- resource, Tertiary care children’s hospitals (i.e., institutions with readily available magnetic resonance imaging (MRI) and/or pediatric surgical subspecialists such as neurosurgery and orthopedics) [18]. However, since the majority of children are seen in general emergency depart- ments [19], there may be significant barriers to widespread adoption of those pathways. Similarly, since CTA also exposes children to poten- tially harmful radiation, evaluation of BCVI in the pediatric trauma pa- tient is ideal for new clinical pathway development. While several pediatric BCVI Screening tools have been developed [20,21], to our knowledge there are no combined pediatric trauma CSI and BCVI pathways.

Therefore, this study’s objective was to investigate the effects of a new, combined CSI and BCVI pediatric trauma clinical pathway. The joint CSI/BCVI pathway was adopted at a general academic hospital (non-children’s hospital) that is a Level I adult and pediatric trauma center. This study presents patient outcomes and CT/CTA utilization be- fore and after the pathway’s implementation, as well as the pathway’s performance in identifying CSI/BCVI and provider adherence.

1. Methods
   1. Study design & setting

This was a 2-year retrospective cohort study analyzing data pre and post implementation of the combined CSI/BCVI clinical pathway. Uni- versity of Florida Health Jacksonville is the only combined Level I trauma center for adult and pediatric patients in Northeast Florida and South- east Georgia, caring for more than 250 pediatric trauma patients ages 0-14 years annually. The study institution’s pediatric trauma response involves co-management of patients by a trauma surgery attending physician who supervises surgery residents, and an emergency medi- cine or pediatric emergency medicine (PEM) attending who supervises emergency medicine residents and pediatric emergency medicine fel- lows. Pediatric subspecialty trained physicians (i.e., PEM attendings and fellows) are present every day from 3p to 12a, and also from Sunday to Thursday from 7a to 3p. The study was approved by the Institutional Review Board of the University of Florida. Data was primarily obtained from the institutional trauma registry, which is audited by registrars and reports to National Trauma Data Bank, supplemented with individ- ual chart reviews performed by the research team.

The joint pediatric trauma CSI/BCVI clinical pathway was developed after an extensive literature review of adult and pediatric literature on CSI/BCVI, as well as publicly available pathways and screening tools for the evaluation of CSI/BCVI. Prior to the implementation of the path- way, CSI and BCVI imaging was at the discretion of the treatment team (e.g., EM, PEM, and trauma surgery providers). Pathway development was a multidisciplinary effort between trauma surgery, pediatric emer- gency medicine, pediatric radiology, and neuroradiology. The final path- way utilized NEXUS, PECARN, and Denver screening criterion, and was designed to use resources at our own institution (a general academic hospital but not a dedicated children’s hospital). The pathway was pre- sented to staff during pediatric trauma conference, surgery and emer- gency medicine grand rounds, made available on the institutional intranet, and posted as large posters in the trauma bay and pediatric emergency room (Fig. 1). The pathway’s “go live” date was July 1, 2018.

• 1. Study population

This study included patients 14 years of age and younger who pre- sented to our trauma center or pediatric emergency department with complaints of blunt trauma during a 2-year period and who met criteria to be entered into the trauma registry (i.e., all who presented as a trauma alert to the trauma center or who were evaluated in the pediat- ric emergency department with traumatic injury and trauma was sub- sequently consulted). The “pre-pathway” group were seen between July 1st, 2017 to June 30th, 2018 and the “post-pathway” group were seen between July 1st, 2018 and June 30th, 2019. We excluded patients with penetrating trauma, those who were dead on arrival or who died at any time during their hospital admission. We also excluded those with high-risk congenital conditions at higher risk for Cervical spine injuries (i.e., Ehlers-Danlos syndrome, Down’s syndrome, etc.) and those who suffered from or were suspected to be non-accidental traumas. If it were unclear if non-accidental trauma was the cause for injury, the case was evaluated by two authors (B.K.Y. and J.N.F..) to determine if it would be included. If there was any disagreement, then a third author (M.S.L.) mediated the final decision.

• 1. Data analysis

The primary outcomes were CT cervical spine and CTA neck utiliza- tion (measured before and after pathway implementation). Secondary outcomes included screening cervical spine radiography utilization, in- cidence of CSI/BCVI, missed injuries, and provider adherence to the pathway in the post-implementation period. We performed descriptive statistics of patient and Encounter characteristics using frequencies and percentiles, means with standard deviations (SD), and medians with in- terquartile ranges (IQR), as appropriate based on normality determined by the Shapiro-Wilk test. For univariate comparisons between pre and post pathway patients, we used the chi square test (or Fisher exact- test for expected cell values less than 5) for categorical variables, the Student’s t-test for continuous variables with normal distributions, and the Wilcoxon rank sum test for continuous variables with non- normal distributions. All statistical analyses were performed using SAS v9.4 (Cary, NC).

1. Results

A total of 358 patients were included: 209 pre-pathway and 149 post-pathway implementation (Fig. 2). The mean age of patients was

8.1 years (SD 4.4), 61% were male, and most were African American (47%), followed by Caucasian (40%) and other races (12%). The majority were seen in the trauma center, except for 7% of patients who were ini- tially seen in the pediatric emergency department. Very few patients presented with Altered mental status (median Glasgow Coma Scale 14.6), and 63% patients did not complain of neck pain upon arrival. The Abbreviated Injury Scale and injury severity score (ISS) were comparable between the pre- and post-pathway groups (p = 0.49 vs p = 0.56, respectively). Disposition did not vary between pre- and post-pathway groups (admission rate 68% for pre-pathway vs 69% for post-pathway). Table 1 displays patient and encounter characteristics for overall, and for pre- and post-pathway groups.

Of note, many patients had at least one NEXUS (30%) or PECARN (75%) high risk criteria for CSI, but only 6% had any DENVER high risk criteria for BCVI (Table 1). Post-pathway, the use of cervical spine CT de- creased (33% vs 31%, p = 0.74) though the decrease was not statistically significant. However, cervical spine radiography use increased signifi- cantly (9% vs 16%, p = 0.03). Screening for BCVI increased with higher use of CTA cervical spine in the post pathway implementation group (5% vs 7%, p = 0.52), though that increase was not statistically significant.

Overall, there were a total of 12 CSI and 3 BCVI identified and there

were no Clinically significant injuries missed. In the pre-pathway, 3%

Fig. 1. Combined pathway for evaluation of pediatric cervical spine and blunt cerebrovascular injuries in blunt trauma patients.

of the total 209 patients were found to have CSI. Similarly, in the post-pathway group 3% of the total 149 patients were found to have CSI. BCVI was found in 1% of both pre- and post-pathway groups. Table 2 details the nature of patients’ CSI and BCVI inju- ries. Of the patients found to have a CSI, just over half presented with altered mental status (58%), and only 25% of patients complained of neck pain. However, CSI patients versus non-CSI patients had lower Glasgow coma scales (14 (IQR 8.5-14) vs 15 (IQR 15-15), p < 0.001) and significantly higher ISS (20.5 (IQR 14.5-27.5) vs 1 (IQR 1-5), p < 0.001).

When applying Nexus criteria, 83% of patients found to have a CSI screened positive for any one NEXUS criteria, compared to only 28% of patients with no injury (p < 0.001). The most common NEXUS criteria met for CSI patients was altered mental status (58% of all patients with CSI). Also of note, 100% of patients with CSI screened positive for any one PECARN criteria, with high-risk motor vehicle collision being the most frequent criteria met (75% of all patients with CSI). Of the 12 patients found to have a CSI, 58% had any one positive DENVER criteria. CTA was obtained in all 12 of those patients, 2 of whom were found to have BCVI.

Upon review, no return visits within 30 days of index hospital stay were noted for any CSI or BCVI that were not initially identified when our multi-hospital electronic medical record (EMR) system was que- ried. Provider adherence to the pathway (judged by at least one of the authors upon chart review) was modest (54% of all post-pathway pa- tients, N = 80). Of note, post-pathway patients whose provider followed the pathway had a 40% CT rate (32 patients out of 80), while the patients of providers who did not adhere to the pathway had a 20% CT rate (14 patients out of 69).

1. Discussion

The goal of this study was to evaluate the implementation of a novel, combined CSI and BCVI detection pathway in pediatric patients

evaluated for blunt traumatic injuries. We found the pathway was suc- cessful in detecting both CSI and BCVI in Pediatric blunt trauma patients and did not miss any clinically significant injuries. We also found a sig- nificant increase in the use of screening radiography of the cervical spine. However, provider adherence to the pathway was modest and while we did see a decrease in CT cervical spine rates, it was not signif- icant. However, future larger studies with greater patients and at multi- ple institutions may be more adequately powered to determine significant decreases in CT utilization in pediatric trauma patients. Re- gardless, any reduction in radiation of pediatric patients, with its poten- tial negative future consequences, could be considered clinically significant.

Our pathway to our knowledge is the first combined pediatric CSI and BCVI pediatric trauma screening clinical pathway. By using screen- ing tools (PECARN, NEXUS and Denver criteria) that are widely known to emergency and trauma providers and easy to apply, we were able to identify at-risk patients and evaluate them with resources common to most general community ED‘s (e.g. XR and CT). Importantly, all pa- tients who were found to have CSI had at least one positive PECARN criteria, which was consistent with the initial validation study of this screening tool [16]. Our observed increased sensitivity for detecting CSI may be due to combining PECARN’s criteria for high-risk injury mechanisms with NEXUS, in contrast to previous studies which applied NEXUS and failed to replicate it’s prediction power in adults with youn- ger Pediatric populations [7,16].

There has been recent renewed interest in the detection of pediatric BCVI, with newer pediatric-specific tools such as the UTAH and McGov- ern scores [20-22]. Although our clinical pathway utilized the Denver criteria for BCVI screening, newer data has shown that given the differ- ences in pediatric anatomy and injury patterns, the DENVER criteria might not be the most ideal for the pediatric population and might lead to overuse of CTA and unnecessary exposure to ionizing radiation [20,21]. However, some of those scores require more advanced imaging (such as MRA) for patients classified as low risk [22]. Our pathway was

Fig. 2. CONSORT diagram of Inclusion/exclusion criteria as applied to this study.

developed to be applicable to general settings such as free-standing community emergency rooms, using equipment that is readily available such as radiography and CT.

Despite successful identification of CSI and BCVI in our patients, we did not significantly reduce the use of cervical spine CT (33% to 31%). We did, however, nearly double the use of screening cervical radio- graphs (9% to 16%). CT has been the gold standard to identify CSI, its use is widely accepted, accessible, and has proven to be more sensitive in detecting CSI when compared to plain radiographs [23-25]. Addi- tionally, ED physicians, specifically those who are adult emergency medicine trained versus pediatric emergency medicine trained, are usually more familiar and more inclined to order a CT based on what screening tools such as NEXUS recommend, as opposed to using the two-step (XR and/or CT) approach our pathway proposes (and has been proven to be the preferred method in pediatric pa- tients) [6]. That could have contributed to the modest provider adher- ence to our protocol as our trauma center is staffed by a wide variety of clinicians, most trained primarily in adult emergency medicine [11,17]. Additionally, changing practice patterns can take years, and our post-implementation period was limited to 12 months of data col- lection. It is interesting that providers adherent to the pathway in the post-implementation period had higher rates of CT utilization than those who were non-adherent. However, more high-risk cases may have prompted providers to use the pathway, contributing to that dis- crepancy. Finally, because of its devastating consequences if missed and fear of potential litigation, clinicians may feel more comfortable ordering an initial CT rather than screening XR. Further studies are re- quired to standardize use of screening cervical spine XR versus imme- diate cervical spine CT and CTAs.

CSI is an infrequent but potentially devastating pediatric traumatic injury [3,5]. Additionally, CSI is a well-known risk factor for BCVI, specif- ically vertebral artery injury [26], and so developing a combined path- way to detect both CSI and BCVI helps ensure more complete evaluation of pediatric blunt trauma patients. As more screening for BCVI is done in trauma patients, the incidence of BCVI appears to be higher than originally reported [9,11,20,21]. Therefore, a combined pathway to detect both CSI and BCVI is of great utility for emergency and trauma clinicians caring for blunt trauma patients.

1. Limitations

This study has limitations that merit consideration. First, it is a retro- spective study at a single Level I pediatric center. However, this is the first combined CSI/BCVI pediatric trauma clinical pathway reported to our knowledge, which allows risk-stratified evaluation of both injuries simultaneously. We excluded patients who were dead on arrival or who died in our facility (N = 5) as the evaluation of moribund patients is very different than that of moderately injured/ill child, and the pathway’s main intent was to guide the more nuanced clinical scenar- ios. We also excluded patients with suspected or confirmed non- accidental trauma, as those cases have separate clinical guidelines with different injury patterns and injury etiologies. Patients with higher risk for CSI and BCVI, such as those with specific congenital conditions, were also excluded. In those cases, more aggressive evaluation might be needed due to their physical or developmental challenges and the higher risk for injury.

As with other pre/post pathway studies, changes in provider practice take time and we found modest adherence to our pathway, which

Table 1

Demographic and variable data between pre and post pathway implementation groups.

Variable	Total (358)	Pre (209)	Post (149)	p value & test
Age - mean (SD)	8.1 (4.4)	8.2 (4.6)	8.0 (4.3)	0.64 (t-test)
Male (%) Race White (%)	221 144	139 (66%) 89 (43%)	82 (55%) 55 (37%)	0.03 (CS) 0.15 (Fisher)
African-American (%)	170	92 (44%)	78 (52%)
Other (%)	44	28 (13%)	16 (11%)
Non-hispanic (%) Trauma level 1 (%)	323 82	184 (88%) 49 (23%)	139 (93%) 33 (22%)	0.10 (CS) 0.96 (CS)
2 (%)	167	98 (47%)	69 (46%)
3 (%)	83	50 (24%)	33 (22%)
Altered mental status (%)	23	12 (5%)	11 (7%)	0.50 (CS)
GCS - mean (SD)	14.6 (1.6)	14.6 (1.7)	14.6 (1.5)	0.97 (t-test)
NEXUS high-risk criteria Any (%)	109	64 (31%)	45 (30%)	0.93 (CS)
Midline (%)	30	15 (7%)	15 (10%)	0.33 (CS)
Altered (%)	27	15 (7%)	12 (8%)	0.76 (CS)
Intoxication (%)	1	1 (<1%)	0	0.58 (Fisher)
Focal neuro deficit (%)	3	1 (<1%)	2 (1%)	0.57 (Fisher)
distracting injury (%)	58	39 (19%)	19 (13%)	0.13 (CS)
PECARN high-risk criteria Any (%)	267	153 (73%)	114 (77%)	0.57 (CS)
Diving (%)	1	0	1 (<1%)	0.42 (Fisher)
Fall height (%)	8	7 (3%)	1 (<1%)	0.15 (Fisher)
Hanging (%)	1	1 (<1%)	0	0.58 (Fisher)
Hit by car (%)	45	29 (14%)	16 (11%)	0.38 (CS)
High-risk MVC (%)	213	116 (56%)	97 (65%)	0.07 (CS)
Denver criteria any high-risk (%)	23	12 (6%)	11 (7%)	0.53 (CS)
Cervical spine XR (%)	44	19 (9%)	25 (17%)	0.03(CS)
Cervical spine CT (%)	114	68 (33%)	46 (31%)	0.74 (CS)
Cervical spine CTA (%)	23	12 (6%)	11 (10%)	0.52 (CS)
Highest calculated AIS - median (IQR)	1 (1-4)	1 (1-4)	1 (1-4)	0.49 (Wilcoxon Rank Sum)
ISS - median (IQR) Disposition Admit (%)	1.5 (1-8) 247	2 (1-6) 143 (79%)	1 (1-9) 104 (70%)	0.56 (Wilcoxon Rank Sum) 0.71 (Fisher)
Discharge (%)	83	49 (23%)	34 (23%)
Transfer (%)	28	17 (8%)	12 (8%)
Provider adherence (%)	149		80 (54%)

GCS - Glasgow coma scale, XR - X-ray, CT - computed tomography, CTA - computed tomography angiography, AIS - Abbreviated injury scale, ISS - Injury severity score.

should temper interpretation of our results. However, with higher ad- herence we may have found a larger and more significant decrease in cervical spine CT utilization. Conversely, due to the nature of the pre/ post study design, there may have been trends or changes in provider

behavior related to our primary outcome present that were unrelated to our clinical pathway intervention. A large part of the study was extracting data from patient’s medical chart, and some charts contained missing data. To our knowledge we did not miss any clinically

Table 2

Description of diagnosed cervical spine and blunt cerebrovascular injuries in pediatric blunt trauma patients.

CSI BCVI Injury description

1. Yes No Fracture of C1 posterior arch and C7. Extensive edema of interspinous ligament. Superior translation of C6-7 facet joints, widening of C6-7 intervertebral disc space, and widening of C6-7 interspinous process distance.
2. Yes Yes Grade 1 right vertebral artery dissection. Disruption of anterior atlantooccipital membrane.
3. Yes No Acute avulsion fracture of the medial cortex of the left occipital condyle with atlantooccipital subluxation. Significant craniocervical ligamentous injury with disruption of the anterior longitudinal ligament, anterior atlantooccipital membrane, apical ligament, and avulsion of the tectorial membrane off the dorsal clivus. Probably torn alar ligaments. Posterior capsular and interspinous ligament injury at C1-2. Posterior ligamentous complex injury at C7-T1 with focal disruption of the ligamentum flavum, interspinous ligament, and supraspinous ligament. Additional interspinous and supraspinous ligamentous tear at T1-2 and probable multilevel partial tears of the interspinous ligament throughout the remaining cervical spine.
4. Yes No Disruption of the apical and right alar ligaments.
5. Yes No C7 spinal process fracture
6. Yes No Apical ligament disruption, injury to posterior interspinous ligament between C1 and C2.
7. Yes No Anterior-inferior teardrop fracture of the C2 vertebral body. Fracture involving bilateral pedicles with the mild displacement of the C2 vertebral ring. Fracture of left C2 foramen transversarium. Anterior subluxation of right C2 facet. Craniocervical junction ligamentous injury involving the anterior atlantooccipital membrane, alar ligament, and apical ligament. The anterior longitudinal ligament, posterior longitudinal ligament, posterior atlantoaxial ligament is disrupted at the C2 level.
8. Yes Yes Ligamentous injury at the craniocervical junction. Grade 2 vertebral artery dissection.
9. Yes No Ligamentous tears of apical ligament, alar ligaments and cruciate ligament.
10. Yes No C4 right anterior and superior endplate fracture.
11. Yes No Occipital bone fracture with edema of the spinal ligaments (Fluid is seen in the right atlanto-occipital joint.
12. No Yes Long segment dissection injury to the left intracranial ICA with severe near occlusive stenosis in the supraclinoid segment of the left Internal carotid artery.
13. Yes No Interspinous ligamental injury at levels C3-C6 CSI - cervical spine injury, BCVI - blunt cerebrovascular injury.

significant injuries with a chart review of any subsequent encounters within 30 days. However, it is unknown whether patients may have sought care at another health system or if there were clinically silent in- juries missed. This risk is mitigated as the EMR utilized for the study in- cludes the only pediatric neurosurgeons group providing care to the region including a free-standing pediatric hospital. Additionally, any symptomatic missed injuries would likely present within 30 days. How- ever, our follow up may have missed any asymptomatic injuries, but the clinical significance of those is unknown.

1. Conclusion

This study describes the successful implementation of a combined CSI and BCVI evaluation pathway in pediatric blunt trauma patients. Our pathway effectively detected those injuries and did decrease CT and increase screening XR use. This pathway is designed to be readily applicable to community and Tertiary care centers staffed by either adult or pediatric clinicians. More work is required to optimize the bal- ance of detecting high-risk injuries and while reducing unnecessary ion- izing radiation exposure in children.

Author contributions

M.S.L. contributed to the study concept and design, acquisition of the data, analysis and interpretation of the data, drafting of the manuscript and critical revision of the manuscript for important intellectual con- tent. J.N.F. contributed to the study design, analysis and interpretation of the data, statistical expertise, and critical revision of the manuscript for important intellectual content. B.K.Y. contributed to the study con- cept and design, interpretation of the data, and critical revision of the manuscript for important intellectual content.

Credit author statement

A novel combined CSI and BCVI pathway for pediatric trauma patients

• Maria Schonenberg Llach (MSL) contributed to the study concept and design, acquisition of the data, analysis and interpretation of the data, drafting of the manuscript and critical revision of the manuscript for important intellectual content.
• Jennifer Fishe (JNF) contributed to the study design, analysis and in- terpretation of the data, statistical expertise, and critical revision of the manuscript for important intellectual content.
• Brian Yorkgitis (BKY) contributed to the study concept and design, interpretation of the data, and critical revision of the manuscript for important intellectual content.

Declaration of competing interest

All authors (M.S.L., J.N.F., B.K.Y.) report no conflicts of interest to disclose.

Acknowledgements

The study investigators wish to acknowledge Yohan Diaz Zuniga from the University of Florida - Jacksonville Department of Surgery,

Division of Research for assistance with trauma registry data, and the Departments of Radiology, Surgery, and Emergency Medicine for their feedback during pathway development.

References

1. Heron M. National vital statistics reports volume 68, number 6, June 24, 2019, deaths: leading causes for 2017. Natl Vital Stat Rep. 2019;68 Page 17.
2. MC C, et al. National Trauma Data Bank: pediatric annual report 2016. Am Coll Surg. 2018.
3. Booth TN. Trauma: cervical spine. Pediatr Radiol. 2011;41(1):151.
4. Platzer P, Jaindl M, Thalhammer G, et al. Cervical spine injuries in pediatric patients. J

Trauma. 2007;62(2):389-96 [discussion 394-6].

1. Leonard JR, Jaffe DM, Kuppermann N, Olsen CS, Leonard JC. For the pediatric emer- gency care applied research network (PECARN) cervical spine study group. Cervical spine injury patterns in children [internet]. Pediatrics. 2014;133(5):e1179-88 Avail- able from https://doi.org/10.1542/peds.2013-3505.
2. Hannon M, Mannix R, Dorney K, Mooney D, Hennelly K. Pediatric cervical spine in- jury evaluation after blunt trauma: a clinical Decision analysis. Ann Emerg Med. 2015;65(3):239-47.
3. Schoneberg C, Schweiger B, Hussmann B, Kauther MD, Lendemans S, Waydhas C. Di- agnosis of cervical spine injuries in children: a systematic review. Eur J Trauma Emerg Surg. 2013;39(6):653-65.
4. Livingston MH, Igric A, Vogt K, Parry N, Merritt NH. Radiation from CT scans in pae- diatric trauma patients: indications, effective dose, and impact on surgical decisions. Injury. 2014;45(1):164-9.
5. Azarakhsh N, Grimes S, Notrica DM, et al. Blunt cerebrovascular injury in children: underreported or underrecognized? A multicenter ATOMAC study. J Trauma Acute Care Surg. 2013;75(6):1006-11 [discussion 1011-2].
6. Fenton SJ, Bollo RJ. Blunt cerebrovascular injury in children. Semin Pediatr Surg. 2017;26(1):2-7.
7. Cook MR, Witt CE, Bonow RH, et al. A cohort study of blunt cerebrovascular injury screening in children [internet]. J Trauma Acute Care Surg. 2018;84(1):50-7 Avail- able from: https://doi.org/10.1097/ta.0000000000001631.
8. Galardi MM, Strahle JM, Skidmore A, Kansagra AP, Guilliams KP. Cerebrovascular complications of pediatric blunt trauma. Pediatr Neurol. 2020;108:5-12.
9. Kopelman TR, Berardoni NE, O’Neill PJ, et al. Risk factors for blunt cerebrovascular in- jury in children: do they mimic those seen in adults? J Trauma. 2011;71(3):559-64 [discussion 564].
10. Hoffman JR, Mower WR, Wolfson AB, Todd KH, Zucker MI. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. N Engl J Med. 2000;343(2):94-9.
11. Stiell IG, Wells GA, Vandemheen KL, et al. The Canadian C-Spine Rule for radiography in alert and stable trauma patients. JAMA. 2001;286(15):1841-8.
12. Leonard JC, Kuppermann N, Olsen C, et al. Factors associated with cervical spine in- jury in children after blunt trauma. Ann Emerg Med. 2011;58(2):145-55.
13. Slaar A, Fockens MM, van Rijn RR, et al. Adherence to the guidelines of paediatric cervical spine clearance in a level I trauma centre: a single centre experience. Eur J Radiol. 2016;85(1):55-60.
14. Pannu GS, Shah MP, Herman MJ. Cervical spine clearance in pediatric trauma centers [internet]. J Pediatr Orthop. 2017;37(3):e145-9 Available from: https://doi. org/10.1097/bpo.0000000000000806.
15. Rui P, Kang K. National Hospital Ambulatory Medical Care Survey: 2017 Emergency Department Summary Tables. National Center for Health Statistics; 2014.
16. Ravindra VM, Bollo RJ, Sivakumar W, et al. Predicting blunt cerebrovascular injury in pediatric trauma: validation of the “Utah Score”. J Neurotrauma. 2017;34(2): 391-9.
17. Herbert JP, Venkataraman SS, Turkmani AH, et al. Pediatric blunt cerebrovascular in- jury: the McGovern screening score. J Neurosurg Pediatr. 2018;21(6):639-49.
18. Nagpal P, Policeni BA, Bathla G, Khandelwal A, Derdeyn C, Skeete D. Blunt cerebro- vascular injuries: advances in screening, imaging, and management trends. Am J Neuroradiol. 2017. https://doi.org/10.3174/ajnr.A5412 [Internet]. Available from:.
19. Holmes JF, Akkinepalli R. Computed tomography versus Plain radiography to screen for cervical spine injury: a meta-analysis. J Trauma. 2005;58(5):902-5.
20. Hale AT, Alvarado A, Bey AK, et al. X-ray vs. CT in identifying significant C-spine in- juries in the pediatric population. Childs Nerv Syst. 2017;33(11):1977-83.
21. Rana AR, Drongowski R, Breckner G, Ehrlich PF. Traumatic cervical spine injuries: characteristics of missed injuries. J Pediatr Surg. 2009;44(1):151-5 [discus- sion 155].
22. Berne JD, Cook A, Rowe SA, Norwood SH. A multivariate logistic regression analysis of risk factors for blunt cerebrovascular injury. J Vasc Surg. 2010;51(1):57-64.