Western pediatric cervical spine study: an observational prospective Western Pediatric Surgery Research Consortium and Western Trauma Association multicenter study protocol

Abstract

Background

In pediatric patients presenting after blunt trauma, the incidence of clinically significant cervical spine injury (CSI) is approximately 1%. Although multidetector CT (MDCT) is widely accepted for cervical spine clearance in adults, its utility in children remains uncertain. The purpose of this study is to evaluate the sensitivity of MDCT in identifying clinically significant CSI in pediatric patients, stratified by age groups: adolescents (12 years to 17 years) and children (0 years to 11 years).

Methods

We designed a national, multicenter, prospective observational study to assess the diagnostic performance of MDCT in detecting clinically significant CSI in pediatric trauma patients. Data are being collected from participating centers on imaging findings, clinical outcomes, and the presence of confirmed CSIs after blunt trauma. The analysis plan includes stratification by age group and calculation of sensitivity, specificity, and predictive values of MDCT for CSI.

Results

This abstract outlines the background, methodology, and analytical framework of the study. Data collection and analysis are ongoing. This study represents the largest known cohort of pediatric patients undergoing cervical spine imaging after blunt trauma and is the first prospective study of its kind focused exclusively on this population.

Conclusions

This study will provide critical data to inform guidelines for pediatric cervical spine clearance after blunt trauma. By evaluating the sensitivity of MDCT in detecting clinically significant CSI, the findings aim to support evidence-based imaging strategies and improve the safety and efficiency of pediatric trauma care.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• The incidence of clinically significant cervical spine injury (CSI) after blunt trauma in children is low, around 1%. Although multidetector CT (MDCT) is widely used and validated in adults for cervical spine clearance, there is no consensus on its reliability or appropriateness in the pediatric population. The lack of high-quality prospective data has left a gap in evidence-based guidelines for pediatric imaging.

WHAT THIS STUDY ADDS

• This study is the first large-scale, prospective, multicenter evaluation of MDCT performance for cervical spine clearance in children. It will provide robust, age-stratified data on the sensitivity of MDCT for detecting clinically significant CSI in both children and adolescents.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

• Results from this study may guide future imaging protocols, reduce unnecessary radiation exposure in children, and help develop national guidelines for pediatric cervical spine clearance. It may also inform future research priorities and standardize care practices across trauma centers.

Introduction

Background and rationale

In pediatric patients presenting after blunt trauma, the incidence of clinically significant cervical spine injury (CSI), defined as an injury requiring operation or immobilization with a halo, is approximately 1%. The National Emergency X-Radiography Utilization Study (NEXUS) identified a 0.98% incidence rate of CSI,¹ and studies using the National Trauma Databank (NTDB) report similar rates ranging from 1.3% to 1.6%.^{2 3} These results are similar to adult trauma patients, with a reported incidence of clinically significant CSI of 2% in blunt trauma.⁴ In 2015, the Eastern Association for the Surgery of Trauma published an evidence-based guideline recommending clearance of the adult cervical spine (CS) in obtunded blunt trauma patients without lateralizing signs on neurologic examination after negative multidetector CT (MDCT) without the addition of MRI.⁵ This recommendation was based on a review of 12 studies including a total of 1017 obtunded adult patients without a reliable physical examination and no apparent lateralizing signs on neurologic examination who had normal MDCTs. Among these patients, 9% had stable ligamentous CSIs detected on MRI that were not identified on MDCT; however, there were no unstable CSIs not seen on MDCT.⁵ These guidelines were transformative and changed practice in many adult trauma centers. In 2016, the Western Trauma Association published a prospective study evaluating CS clearance in 10 765 adult blunt trauma patients. They found the sensitivity of MDCT was 98.5% in identifying clinically significant CSI, as defined as a CSI requiring surgical fixation or halo placement.⁶ All the missed injuries occurred in elderly patients with degenerative disc disease and/or symptomatic central cord syndrome—two disease states not typically seen in children.

The prevailing concern among providers caring for pediatric blunt trauma patients is that MDCT alone is inadequate for screening and ruling out pediatric CSI due to the ligamentous laxity and bone immaturity seen in children.² Expert consensus from the Pediatric Cervical Spine Clearance Working Group recommends clearance of the pediatric CS and removal of the cervical collar (c-collar) in patients with normal radiographs or MDCT in combination with a normal physical examination (pain-free neck motion in a patient that is neurologically intact) or a normal MRI.⁷ This protocol has been adopted and implemented by many centers located in the USA who care for pediatric trauma patients.

Missing a clinically significant CSI may lead to neurologic compromise; however, most children with blunt trauma are not found to have CSIs. Prolonged CS immobilization with hard c-collars can lead to iatrogenic complications such as c-collar associated pressure ulceration, patient and family anxiety, discomfort, nursing challenges, difficulty with airway and respiratory management, and increased intracranial pressure.^{8 9} A safe, reliable, and fast means of evaluating for CSI is therefore needed. There are multiple imaging modalities currently used for CSI screening and clearance, and each has its own risk-to-benefit profile. CS X-rays can be obtained quickly with minimal radiation exposure, but this practice has been largely abandoned in the adult trauma evaluation.¹⁰ MDCT provides a relatively fast and reliable method of screening for CSI, but there is higher radiation exposure which may increase the lifetime risk of malignancy.¹¹ Additionally, advancements in MDCT image quality with 64-slice helical scans with sagittal and coronal reconstructions have led to a far superior image quality compared with single-slice CT.¹² Although MRI is radiation-free, a number of concerns question its practicality of use, including (a) availability, (b) need for sedation or anesthesia, (c) dependent on patients status/stability, (d) length of performance of the study, and/or (e) increased cost. In addition, MRI has been reported to have a high false positive rate in diagnosing CSI, which may result in unnecessary CS immobilization.¹³ Limited sequence MRI is an emerging imaging technique that may be more sensitive than MDCT for detecting CSI and being faster than conventional MRI and mitigating the need for anesthesia. The calculus for balancing the risks of radiation exposure, timeliness, and accessibility of imaging against the risk of a missed CSI differs between children and adults.

Russell et al recently conducted a large single-center retrospective study evaluating the sensitivity of MDCT for detecting clinically significant CSIs in pediatric blunt trauma patients.⁸ The study included approximately 4500 pediatric patients and found that MDCT was 100% sensitive for diagnosing CSIs that required intervention, including operation or fixation with halo placement. Additionally, in another retrospective study, Derderian et al found MDCT has a 100% sensitivity in detecting unstable (radiologic findings of disruption of two or more contiguous spinal columns) and clinically significant (requiring operation or halo) CSIs.¹⁴ Qualls et al evaluated 63 children admitted to the intensive care unit with both an MRI and MDCT and found no missed unstable injuries on the MRI.^{14 15} Although these data are promising, these studies lack generalizability due to their retrospective nature, single pediatric institutional design, and inadequate power. Moreover, several studies suggest MDCT alone is inadequate to screen for CSI in pediatric trauma patients.^{16 17} More research is needed to create an evidence-based protocol for CS clearance in children and adolescents.

In 2022, the Western Pediatric Surgery Research Consortium (WPSRC) and began collaborating on the ‘Western Pediatric Cervical Spine Study’, a prospective observational study evaluating the accuracy of CS imaging modalities in blunt pediatric trauma. Here, we report the study aims and hypotheses, study design, and analytical plan.

Aims and hypotheses

The primary aim of the Western Pediatric Cervical Spine Study is to determine the sensitivity of MDCT in identifying clinically significant CSI in children. The primary outcome measure is clinically significant CSI, which is defined as a CSI requiring an operation or immobilization with halo placement. We hypothesize that the MDCT is highly sensitive in identifying clinically significant CSI in the adolescent patient population. The secondary aims are to: (1) determine the sensitivity of MDCT in detecting clinically significant CSI in children (0 years to 11 years old); (2) determine the specificity, positive predictive value (PPV), and negative predictive value (NPV) of MDCT in identifying clinically significant CSI in adolescents (12 years to 17 years old) and children (0 years to 11 years old); (3) determine the sensitivity, specificity, PPV, and NPV of CS X-ray and MRI in detecting clinically significant CSI in children (0 years to 11 years of age) and adolescents (12 years to 17 years old); (4) assess the 90-day CS clearance practices of patients who are discharged with CS precautions and CS immobilization with a hard c-collar who did not meet the strict definition of clinically significant CSI; (6) determine the contemporary pattern of pediatric CSIs and develop a consensus of classifying CS ligamentous injuries as stable versus unstable injuries based on radiographic findings, and (7) assess the incidence of and risk factors for CS hard collar-associated pressure ulceration.

Methods and analysis

Study design

This is a multicenter, prospective observational study that will enroll adolescents (12 years to 17 years old) and children (0 years to 11 years old) undergoing CS imaging (X-ray, MDCT, MRI) as part of their initial trauma evaluation for a blunt mechanism of injury. The study began patient enrollment on August 1, 2022, and we estimate completing patient enrollment on August 1, 2026.

The study design and rationale were developed in collaboration with the WPSRC in September 2021 and presented to the WTA Multicenter Trials Committee at the 51st annual meeting in February 2022, in Big Sky, MT. Participating trauma centers were primarily recruited through national society meetings, including the WTA, the American College of Surgeons Clinical Congress, the Pediatric Trauma Society, the Western Pediatric Trauma Conference, and the American Association for the Surgery of Trauma, with the goal of enlisting 60 to 80 trauma centers for patient enrollment. Study site recruitment was closed on January 1, 2024. Each study site is obtaining or has already completed internal review board (IRB) approval for the study locally. We intend to use the Standards for Reporting of Diagnostic Accuracy Studies reporting guidelines for manuscript preparation.

Setting

This study is being conducted in the USA and includes patients receiving treatment at a variety of trauma centers, including adult trauma centers, mixed adult and pediatric trauma centers, and free-standing pediatric centers. Given that many injured children are evaluated at centers that do not specialize in the care of children, we are including both adult and pediatric level I, II, and III trauma centers to make the findings generalizable. Information regarding trauma verification level, number of yearly trauma activations and pediatric trauma activations, utilization of board-certified pediatric radiologists, and CS clearance protocols will be collected from each participating trauma center.

Currently, 73 trauma centers have been enlisted and are in various stages of study onboarding, with 51 sites actively enrolling patients. The geographic distribution of participating trauma centers as of May 2024 is represented in figure 1.

Figure 1. Location and type of the 73 participating trauma centers.

Participants

Study participants are identified by daily screening of the trauma registry at participating trauma centers. Patients meeting the following criteria are included in the study:

1. < 18 years old.

2. Blunt force trauma mechanism of injury.

3. Received any CS imaging (XR, MDCT, and/or MRI) within 24 hours of arrival at the trauma center for evaluation of CSI.

4. Received a skeletal survey, including a CS X-ray, as part of a suspected non-accidental-trauma evaluation.

Exclusion criteria for the study include patients presenting with penetrating trauma to the head or neck or poor-quality imaging that is not repeated. For MDCT, a poor-quality image is defined as <64-slice, imaging cuts >3 mm, no reformats, and/or significant motion artifact. In addition, patients with a documented history of congenital spine anomalies, spinal operation, trisomy 21, osteogenesis imperfecta, or other skeletal abnormalities will be excluded. Participants will not be excluded if they are interfacility transfers unless the imaging meets the above exclusion criteria. Many pediatric trauma patients are first examined and triaged at an external facility without pediatric trauma resources prior to evaluation and definitive care at a trauma center with pediatric resources.¹⁸,²¹ A waiver of consent for study enrollment was granted by the IRB at participating study sites.

Data sources, variables, and outcomes

Patients are identified prospectively at each site and data is collected by existing records. The patient’s treatment team (trauma surgery, neurosurgery, orthopedics, radiologist) is contacted if data points are not readily available or if there is conflicting data in the electronic medical record (EMR). Demographic data and injury characteristics are collected, including age, sex, ethnicity, race, mechanism of injury, and associated injuries. Information regarding the patient’s initial trauma evaluation, including intubation prior to arrival, initial Glasgow Coma Scale (GCS) score, systolic blood pressure, heart rate, and neurologic examination is recorded. All CS imaging (X-ray, MDCT, MRI, flexion-extension films) within the first 24 hours of arrival at the trauma center is reviewed. Any abnormality identified on imaging is documented.

The primary outcome of this study is clinically significant CSI. Characteristics of all CSI and their management are reviewed and recorded; injury characteristics include the level of CS fracture, presence of a ligamentous injury, and the occurrence of atlanto-occipital dislocation. For this study, we have not defined which ligamentous injuries are classified as ‘stable’ versus ‘unstable’ CSI, but rather defer the classification to each trauma center’s impression. Definitive management of CSI will be abstracted from the medical chart, including surgical fixation, halo placement, use of a cervical thoracic orthosis and/or c-collar. The EMR is reviewed at 3 months after hospital discharge for each patient discharged in a hard collar who did not undergo surgical fixation or halo placement to assess for additional interventions, length of c-collar use, CS clearance practices, and incidence of associated complications such as pressure injury. Figure 2 contains all data points that are collected.

Figure 2. Data collection form. GCS, Glasgow Coma Scale; ICU, intensive care unit.

Data collection and management

Deidentified data are collected at each participating trauma center and managed using Research Electronic Data Capture (REDCap) electronic data capture tools hosted at the University of Utah, located in Salt Lake City, UT.^{22 23} REDCap is a secure, web-based software platform designed to support data capture for research studies, providing (1) an intuitive interface for validated data capture, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless data downloads to common statistical packages, and (4) procedures for data integration and interoperability with external sources.

For data quality and integrity, the data are regularly examined by each participating trauma center’s study coordinators. In addition, 10% of all study records from each site will be chosen randomly and assigned for an audit of data entry accuracy after 6 months of participation or 100 patients of data entry.

Sample size

It is imperative that the imaging modality used to screen for clinically significant CSI have a very high sensitivity (99%), as missing an injury could result in permanent paralysis or death. Given the difference in the pediatric CS maturity by patient age, we intended to determine the sensitivity of the MDCT scan in two distinct patient populations: adolescents (12 years to 17 years old) and children (<12 years old). Because this is a non-consenting study and all the data for the primary endpoint will be collected prior to patient discharge, we do not anticipate a significant drop-out rate and therefore, the sample size reflects the actual enrollment size. To detect a sensitivity of 99%, with a precise 95% CI of 95% to 100%, a total of 11 300 patients are needed for the adolescent-aged cohort. The sample size was calculated using the 2-tailed exact Clopper-Pearson interval method, a common method for calculating binomial CIs.²⁴,²⁶ Given the rarity of CSI and decreased use of MDCT in children due to radiation, we are willing to accept a less precise 95% CI, as we anticipate fewer children will undergo MDCT evaluation for CS clearance, to ensure study feasibility. Using the same methodology, we determined a sample size of 4400 children is needed to detect CSI on MDCT with a sensitivity of 95% with a less precise 95% CI of 90% to 100%. We plan to keep the study open to all patients until we enroll 11 300 adolescents and 4400 children.

Analysis plan

Patient demographics, mechanisms of injury, and injury characteristics will be characterized with descriptive statistics. We plan to report means and SD or medians with IQRs based on the normality of the distribution of each variable. The analysis of clinically significant CSI detected by MDCT imaging was based on a previous study conducted by Inaba et al among adult blunt trauma patients.⁶

We plan to assess the sensitivity of the MDCT scan in detecting clinically significant CSI in each age cohort. True positives are defined as a clinically significant CSI that was identified on MDCT as either a significant injury, an abnormal finding, or an equivocal finding prompting further imaging evaluation. True negatives are defined as a patient not having a clinically significant CSI, and we will assess if the true negative was detected correctly if the MDCT did not register findings in any of the above categories. If the MDCT registers a positive finding when a patient did not have a clinically significant CSI, this will correspond to a false positive. If the MDCT does not register a positive finding, when a patient did have a clinically significant CSI, this will correspond to a false negative. We will include exact Clopper-Pearson 95% CIs with each estimate. Test characteristics (sensitivity, specificity, PPV, and NPV) of X-ray and MRI in detecting clinically significant CSI will be assessed in both age cohorts. True positives and negatives will be defined like the definitions used for the MDCT.

Univariate logistic regressions will be conducted among the subset of subjects with positive CSIs, and separately among the subset of subjects with negative CSIs to examine the impact of patient and trauma center characteristics on the sensitivity and specificity, respectively. In particular, we are interested in the impact of trauma center designation (pediatric vs adult), trauma center level, radiology training (board-certified pediatric vs adult radiologist), patient age, and admission GCS. Factors that are found to have a significance of p value <0.10 will be included in multivariable logistic regression models for sensitivity and specificity.

The disposition of patients who did not receive CS clearance prior to discharge will be assessed 90 days after discharge. The mechanism of CS clearance will be summarized. In addition, univariable and multivariable analysis will be conducted comparing the demographic and injury characteristics of patients who did not receive CS clearance prior to discharge to those who had CS clearance during hospitalization. The incidence of pressure ulcerations associated with hard collar application will be determined. Risk factors in the development of pressure ulceration will be analyzed using a univariable and multivariable analysis. Multivariable models will be constructed in each case by including risk factors with p value<0.10 in the univariable analysis.

Discussion

Current pediatric guidelines recommend obtaining an MRI for CS clearance in the obtunded patient (GCS<9) in the setting of a normal MDCT.⁷ We are conducting the first prospective multicenter study to establish the utility of X-ray and MDCT in screening for clinically significant CSI in pediatric blunt trauma. This study will determine if there is a need for additional CS imaging in pediatric patients who have a normal MDCT. We will provide a detailed description of patient characteristics, injury characteristics, CSI screening modality, CS clearance practices, and their relationship to CSI. CS X-ray and MDCT imaging discrepancies with MRI will also be discussed.

This study is powered to assess the test characteristics of MDCT in two patient populations: adolescents (12 years to 17 years old) and children (<12 years old). Anatomic differences exist between the pediatric and adult CS, including incomplete ossification, a unique vertebral configuration, and ligamentous laxity.²⁰ These differences in pediatric and adult CS configuration are thought to be prominent until approximately 9 years of age.^{2 27 28} Additionally, the pattern of CSI has been reported to be different in children (<10 years of age) and adolescents (10 years to 17 years), with children having a greater percentage of CSI occurring between cervical vertebrae C to C4 than adolescents. Although the definition of an adolescent is variable, we chose the cut-off of 12 by expert consensus. The results of this study will be used to either reinforce or change guidelines on the clearance of the CS in pediatric patients. Furthermore, the findings of this study may suggest creating separate CS guidelines for children and adolescents.

The proposed study design has multiple strengths. First, the study includes trauma centers with different designations and includes adult and pediatric designated trauma centers. As previously reported, 75% to 85% of all pediatric trauma patients receive care at non-pediatric hospitals.¹⁸²⁹,³¹ We included adult and pediatric-designated trauma centers to ensure the generalizability of the study. In addition, by including trauma centers with varying designations, we can analyze the impact that care has on detecting clinically significant CSIs. Second, the prospective observational nature of the study allows for increased accuracy and completion of data points to be abstracted. In instances of missing data in the EMR, members of the research team at each participating trauma center may obtain this information by contacting the study subject’s clinical team. Finally, the study will follow subjects discharged in a hard c-collar for 90 days to understand their disposition and CS clearance method. 

To conclude, this article presents the rationale and design of the Western Pediatric Cervical Spine Study: An Observational Prospective WPSRC and WTA multi-institutional study in evaluating the utility of X-ray and MDCT in the screening of CSI in pediatric blunt trauma patients. After completion of the study, we will provide guidance in the clearance of the pediatric CS.

Dissemination

The study results will be analyzed and prepared for publication in a peer-reviewed international journal and presented at national meetings. The results will be presented at the WTA annual meeting.

Data availability statement

Data sharing not applicable as no datasets were generated and/or analyzed for this study.