Few interventions in trauma care are as reflexively applied - or as rarely questioned - as the cervical collar. From the moment that spinal injury enters the differential diagnosis, the collar goes on. It is placed by paramedics, maintained through the primary survey, kept in situ during imaging, and removed only once the spine has been formally cleared. For most trauma practitioners, this sequence is so deeply embedded in their clinical instinct that questioning it feels almost transgressive. The cervical collar is not merely a piece of equipment; it represents a philosophy of care - that in the face of uncertainty, immobilisation is safe, and movement is dangerous.
That philosophy is now under serious and sustained scrutiny. Over the past decade, a convergence of biomechanical research, clinical outcome data, and guideline revision has produced a body of evidence that challenges not only the effectiveness of hard cervical collars but their safety. The emerging picture is one in which routine collar application, far from being a neutral precautionary measure, carries its own measurable risks - and in certain patient groups, those risks are not trivial. The ATLS curriculum continues to teach cervical immobilisation as a cornerstone of trauma management, but the gap between what the manual recommends and what the evidence now supports is widening in ways that every candidate should understand.
What ATLS Teaches and Why
The ATLS rationale for cervical spine immobilisation rests on two premises: that a significant cervical spine injury may be present without neurological deficit in the acutely injured patient, and that uncontrolled movement of an unstable spine risks converting an incomplete neurological injury into a complete one. Both premises are biologically plausible. Cervical fractures without cord injury are well documented, and the mechanics of a displaced fracture segment impinging on the cord during patient movement are easy to conceptualise. The 11th edition of the ATLS Student Course Manual teaches manual inline stabilisation during airway management and collar application as part of the primary survey, and candidates are expected to know the clinical criteria that allow immobilisation to be removed - or withheld - in the alert, cooperative, neurologically intact patient.
The clinical decision rules that underpin selective immobilisation - principally the NEXUS criteria, validated by Hoffman and colleagues in the New England Journal of Medicine in 2000, and the Canadian C-Spine Rule, published by Stiell and colleagues also in the New England Journal of Medicine in 2003 - allow clinicians to identify low-risk patients in whom imaging and immobilisation can be safely avoided. Both rules have been prospectively validated in large cohorts and have high sensitivity for clinically significant cervical spine injury. Their existence already implies that routine immobilisation of all trauma patients is not the standard - yet in practice, particularly in the prehospital phase, the threshold for applying a collar remains low.
The Evidence Against Routine Collar Use
The first major challenge to routine cervical immobilisation came not from outcome data but from biomechanical research. The intuitive assumption that a rigid collar prevents cervical motion has proven to be substantially overstated. A series of cadaveric and volunteer studies, reviewed systematically by Kwan and colleagues in a Cochrane review examining pre-hospital spinal immobilisation, found that hard collars reduce - but do not eliminate - cervical motion in multiple planes. More importantly, in patients with grossly unstable fractures, collar application can paradoxically produce distraction forces that increase rather than reduce the displacement of fracture fragments. This is particularly relevant in injuries involving ligamentous disruption, where the collar's compressive and restrictive effects act on a spine that no longer has structural continuity.
The intracranial pressure question is more clinically urgent. Multiple studies have demonstrated that hard cervical collars, by compressing the external jugular veins and impeding venous drainage from the cranium, produce a measurable elevation in intracranial pressure. In healthy volunteers, the effect is modest and probably inconsequential. In patients with traumatic brain injury - precisely the population most likely to arrive immobilised after a high-energy mechanism - the consequences may be considerably more serious. Raised intracranial pressure in the context of head injury is an independent predictor of poor neurological outcome, and any intervention that compounds it, however modestly, warrants careful scrutiny. A frequently cited paper by Raphael and Chotai, published in Anaesthesia in 1994, quantified the ICP elevation associated with collar application and prompted a debate about routine use in head-injured patients that has only intensified as the neurointensive care literature has clarified the importance of venous outflow obstruction.
More recent work has sustained this concern. Hunt and colleagues, in research examining the haemodynamic effects of semi-rigid collars in patients with elevated ICP, found consistent elevations in intracranial pressure measurements following collar application, lending prospective clinical weight to what had previously been largely theoretical and experimental data.
Pressure Injuries, Agitation, and Practical Harms
Beyond the intracranial pressure problem, hard cervical collars generate a range of practical complications that are easy to dismiss in isolation but accumulate into a meaningful harm profile when considered together. Pressure injuries over the mandible, occiput, and clavicle develop with striking rapidity in patients who are immobilised for extended periods - a particular concern given the time pressures of major trauma centre throughput, where a patient may remain in a collar from point of injury through imaging, resuscitation, and operative stabilisation before any reassessment of the device occurs.
Collar-related agitation in confused or intoxicated patients presents a different but equally important problem. A patient who is combative and attempting to remove their collar is generating far greater cervical motion than one who is calm and unrestrained. Repeated studies examining the forces applied to the cervical spine during uncooperative patient management have found that immobilisation attempts in agitated patients can produce cervical loading that exceeds that of controlled movement in cooperative individuals. In this subgroup, the collar may be achieving the opposite of its stated purpose - restraining the practitioner's comfort rather than protecting the patient's spine.
The airway implications of collar application are also clinically significant in ways the ATLS curriculum acknowledges only partially. Mouth opening is restricted by a well-fitted hard collar, reducing the space available for laryngoscopy and increasing the difficulty of intubation. In a patient who requires rapid sequence induction, this restriction compounds the challenges already imposed by a potentially full stomach, haemodynamic instability, and a blood-filled airway. The recommendation that manual inline stabilisation be applied during intubation, with the front of the collar removed, represents a pragmatic solution - but it also concedes that the collar itself is an obstacle to one of the most time-critical interventions in trauma management.
How Guidelines Have Responded
The divergence between ATLS teaching and broader guideline evolution on this topic is instructive. The National Institute for Health and Care Excellence published guideline NG41 on spinal injury assessment in 2016 and updated it subsequently, reflecting a move away from universal prehospital immobilisation toward a more selective approach. NICE explicitly recognises that the risks of immobilisation - particularly in patients with airway compromise, agitation, or penetrating injury - may outweigh the potential benefits of spinal protection, and recommends that the decision to immobilise should be individualised rather than automatic.
This is a meaningful departure from the older reflexive approach. Penetrating trauma deserves specific mention here, because the evidence in this group is particularly strong. A retrospective analysis by Haut and colleagues, published in the Journal of Trauma in 2010 across a large US trauma database, found that spinal immobilisation in patients with penetrating trauma was associated with increased mortality compared with those not immobilised, even after controlling for injury severity. The proposed mechanism is that immobilisation delays definitive management, prolongs scene time, and diverts clinical attention without providing meaningful spinal protection in a patient group whose cervical spine injury rate is already low. Several major trauma systems have now adopted protocols that specifically exclude penetrating trauma from routine immobilisation pathways, a practice increasingly reflected in military pre-hospital care guidelines.
The Obtunded Patient: Where the Controversy Is Hardest
The most genuinely difficult territory in this debate is the patient who cannot be clinically assessed - the obtunded polytrauma patient, the severely intoxicated individual, or the patient with a significant distracting injury who cannot reliably report cervical pain. NEXUS and the Canadian C-Spine Rule both require a cooperative and alert patient for safe application. In the patient who does not meet these criteria, clinical clearance is not possible, and the question of whether to image and how to manage immobilisation until imaging is complete becomes considerably more complex.
CT cervical spine in this population has high sensitivity for bony injury, and the time from arrival to CT completion in most major trauma centres is now sufficiently short that prolonged pre-imaging immobilisation is rarely clinically necessary. The more contested issue is what to do when CT is negative but the patient remains obtunded and MRI - the investigation of choice for ligamentous injury - cannot be obtained acutely. Practice varies considerably between institutions, and the evidence base for prolonged immobilisation in CT-negative, clinically unassessable patients is weaker than most practitioners appreciate.
What the ATLS Candidate Must Know
For the examination, candidates should understand that ATLS 11th edition teaches cervical immobilisation as part of the primary survey approach to the major trauma patient, with manual inline stabilisation during airway manoeuvres and collar application as the default until clinical or radiological clearance can be achieved. The NEXUS criteria and an awareness of the Canadian C-Spine Rule represent the framework within which clinical clearance decisions are made.
What the thoughtful candidate should additionally understand is that this is one of the areas where ATLS teaching and broader clinical guideline evolution are travelling in somewhat different directions. The evidence that hard collars carry measurable risks - in terms of intracranial pressure, pressure injury, airway access, and patient agitation - is now sufficiently robust that its omission from examination teaching should not be taken to mean that it is clinically unimportant. In penetrating trauma in particular, the evidence argues against routine immobilisation with a clarity that has already influenced major prehospital guidelines internationally.
The cervical collar has saved lives. It has also been applied reflexively, universally, and without adequate consideration of the harms it generates in patients for whom the indication is uncertain. The trauma practitioner who understands both sides of that ledger - the protection it offers and the risks it confers - is equipped to make better decisions for individual patients than one who follows the protocol without interrogating its limits.