SGEM#493: You Can’t Hold Me Down with Spinal Motion Restrictions

Date: November 8, 2025 Reference: Millin M, et al., Prehospital Trauma Compendium: Prehospital Management of Spinal Cord Injuries – A NAEMSP Comprehensive Review and Analysis of the Literature, Prehospital Emergency Care, Aug 2025. Guest Skeptic: Clay Odell, BSN, NRP, RN, is a Paramedic Firefighter with Newport (NH) Fire-EMS. He is a past Chief of the NH Bureau of Emergency Medical Services. Previous positions he held are Trauma System Coordinator for the State of NH, Executive Director of Upper Valley Ambulance in Fairlee VT, and a flight crew member at Dartmouth Hitchcock Advanced Response Team. Clay has been a paramedic since 1985 and has been a registered nurse since 1997. Case: Your EMS unit responds to a 911 call for a hunting accident. You arrive to find the patient sitting on the tailgate of his truck. He tells you he fell out of his deer stand, approximately 20 feet, landing on his head. He walked out of the woods about a mile after the fall. His chief complaint is head and neck pain. He has a Glasgow Coma Scale (GCS) score of 15, a hematoma and laceration above his left eye, and he is quite tender over the cervical spine region. You observe your colleagues trying to apply a rigid cervical collar without moving the patient’s neck. It goes about as well as usual, meaning the patient's head is moved a bit. Maybe more than a bit. He then decides he hates the collar and rips it off. He adamantly refuses all attempts to apply a soft collar or improvised towel roll. On arrival at the trauma center, you give a handoff report, and the team leader demands to know why the patient is not in a collar. The patient overhears this and rather profanely informs everyone, "you ain't putting no *F-ing* collar on me". Background: Prehospital care for suspected SCI has two competing imperatives, limiting secondary cord injury while avoiding iatrogenic harm. Historically, EMS prioritized rigid immobilization (long backboard + rigid cervical collar) based on the fear that post-injury movement could precipitate delayed neurologic deterioration. In the last two decades, emergency care has shifted toward selective spinal motion restriction (SMR) and earlier collar removal when appropriate. This reflects a better understanding of risk, test performance, and harms from prolonged immobilization. In the ED, validated decision tools (NEXUS and Canadian C‑spine Rule) help identify very‑low-risk patients who do not need imaging; when imaging is needed, modern multidetector CT outperforms plain radiography for clinically significant cervical spine injury (CSI). A large Western Trauma Association cohort (10,276 patients) found CT sensitivity of 98.5% with a negative predictive value of ~100% for clinically significant injuries. The misses were rare and occurred in patients with focal neurologic deficits, who then warranted an MRI [1]. Prehospital protocols increasingly emphasize minimizing time on a backboard and avoiding prolonged collar use because of pressure injury and other morbidities. A systematic review by the East Association for the Surgery of Trauma (EAST) supports collar removal after a negative high-quality CT in an obtunded adult. They highlight the downstream harm from extended immobilization without added benefit [2]. Pediatrics and geriatrics remain special populations. NEXUS shows high sensitivity in children, though confidence intervals are wider in the very young.  Clinicians should have a low threshold for imaging in older adults, who are vulnerable to serious cervical spine injuries from low-energy mechanisms [3]. Clinical Question: In trauma patients with potential SCI, what is the evidence that post‑injury movement causes delayed neurologic deterioration, and what are the benefits and harms of prehospital spinal immobilization and SMR? Reference: Millin M, et al., Prehospital Trauma Compendium: Prehospital Management of Spinal Cord Injuries – A NAEMSP Comprehensive Review and Analysis of the Literature, Prehospital Emergency Care, Aug 2025. Population: Human (and some animals) with potential spinal cord injuries that addressed one of four predefined questions(pathophysiology of delayed neurologic injury, harms of immobilization, effectiveness of immobilization/SMR at limiting movement, and the relationship to delayed neurological injury). Exclusions: Manikin/simulation‑only studies, abstracts without full publication, editorials, other reviews/meta-analyses, guidelines, book chapters, and non-English full manuscripts. Intervention: Prehospital spinal immobilization (long backboard + cervical collar) and SMR (cervical collar ± vacuum splint/ambulance cot) Comparison: No immobilization vs immobilization or alternative immobilization strategies. Outcome: Primary Outcome: Effectiveness of immobilization/SMR at limiting movement and relationship to delayed neurological injury. Secondary Outcomes: Harms of immobilization/SMR (airway/respiratory compromise, raised ICP, pressure ulcers, delays to care, increased imaging, pain), and associations between hypoperfusion/hypoxia and worse neurologic outcomes. (See Tables 2-4 summarized in the paper.) Type of Study: Comprehensive review with no meta-analysis Authors’ Conclusions: There are no data in the published literature to support spinal immobilization and spinal motion restriction as standard of care. Efforts aimed to reduce the use of cervical collar should be considered, and the use of backboards and full body vacuum splint should be limited to the point in time of active patient extrication.  Quality Checklist for Therapeutic Systematic* Reviews: (Yes/No/Unsure) * This is the list we use for SR, while the study in question was called a “Comprehensive Review.” The main question being addressed should be clearly stated. Yes The search for studies was detailed and exhaustive. Yes Were the criteria used to select articles for inclusion appropriate? Yes Were the included studies sufficiently valid for the type of question asked? Unsure Were the results similar from study to study? No Were there any financial conflicts of Interest? No Results: The review synthesized 115 manuscripts spanning prehospital and hospital settings, adult and pediatric populations, and diverse geographies and time periods. The authors did not pool a single set of patient demographics due to the heterogeneity of designs and questions. Key Results: No definitive evidence that routine immobilization/SMR prevents delayed neurologic injury, while harms are common. Primary Outcome: Limiting movement (biomechanical surrogates) results are mixed. No randomized trials were found showing that collars/boards prevent secondary spinal cord injury. In the best-known observational comparison (immobilized vs. not), immobilization was associated with more neurologic disability. A before-and-after system study (moving from routine backboards to SMR) found no increase in delayed deficits. Secondary Outcomes: Harms were common with negative airway/respiratory effects, increased ICP,skin breakdown, delays in care, increased imaging, and pain. Dr. Millin is an emergency and EMS physician, retired from full-time faculty at Johns Hopkins University in Baltimore, and now runs a non-profit wilderness EMS agency in Maryland. He is also the lead author of this comprehensive review, and we asked him five nerdy questions.  Heterogeneity: The review necessarily combined diverse designs (case series, retrospective cohorts), different exposures (movement vs perfusion/hypoxia), and varied outcomes (motion metrics vs patient-important neurologic outcomes). Such heterogeneity limits inferences, and the authors appropriately did not attempt to meta-analyze the data. Risk‑of‑Bias: While selection and abstraction processes are detailed, the report does not describe a formal risk‑of‑bias (RoB) tool. This is a common expectation in systematic reviews. They could have used the ROBINS‑I, which helps assess the risk of bias in the results of non-randomized studies that compare the potential benefit and harm of two or more interventions [4]. Without a structured RoB assessment, confidence in causal interpretations is limited. Screening: The title/abstract was screened by a single reviewer. This may increase the chance of missed eligible studies. Best practice is for two independent reviewers to screen titles and abstracts and to review full texts independently, resolving discrepancies through discussion or a third reviewer. This process improves the validity, transparency, and reproducibility of systematic reviews [5,6,7]. Language & Publication Bias: The authors limited their search to English-language full texts. This can introduce language bias. In addition, narrative syntheses with small or observational studies are also vulnerable to publication bias, with negative studies less likely to be published. Surrogates Outcomes. Several included papers focus on movement reduction as a surrogate for patient-important outcomes (neurologic deficits). Surrogate-oriented outcomes (SOO) do not reliably translate into patient-oriented outcomes (POOs). Comment on Authors’ Conclusion Compared to SGEM Conclusion:  I, Clay Odell, believe the results, and I think this is likely as good evidence as we’re ever going to get regarding the clinical utility (or not) of cervical collars. This may be sufficient evidence for some EMS medical directors to reduce the emphasis on c-collars, but there will remain skeptics. I wonder if the NAEMSP will adopt these conclusions into their official position statement. We agree that routine, prolonged immobilization is not evidence-based and carries harm, and that minimizing backboard time is good practice. We would frame the conclusion more cautiously and prioritize oxygenation/perfusion, use manual stabilization/SMR selectively,