• 
• 

Three Part Question

In [blunt adult trauma patients] does [spinal immobilisation versus no immobilisation] affect the risk of [secondary neurological deterioration]

Clinical Scenario

You are asked to assess a 32-year-old who has waited two hours after being fully immobilised following a RTC. He has midline cervical tenderness and undergoes CT as X-rays are inadequate. You wonder whether immobilisation is doing more harm than good and if it actually prevents secondary neurological deterioration (SND).

Search Strategy

MEDLINE 1946-Oct 2017 including MEDLINE in process and other non-indexed citations, and Embase using the Ovid interface on ATHENS were searched. The Cochrane Library, Scopus and Pubmed were also searched.
[cervical vertebrae/in OR ((cervical or c) adj2 spin*).ti,ab OR exp “spinal cord injuries’ OR exp “spinal injuries” OR exp “spinal fractures” OR (spinal adj2 injur*).ti,ab] AND [exp “immobilization” OR (neck adj2 collar*).ti,ab OR (collar or brace or immobili*).ti,ab]

Search Outcome

A total of 3391 articles were identified of which 3250 were excluded. 141 full text articles were assessed for eligibility, of which 11 were relevant and included in the quantitative synthesis.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Hauswald et al. 1998 Malaysia/USA	Blunt trauma patients: 334 immobilized in USA and 120 not immobilized in Malaysia	Retrospective chart review (1988-1993) 2-	Secondary neurological deterioration following immobilisation versus non immobilisation	21% of immobilised patients suffered neurological disability compared to 11% of non-immobilised patients	Retrospective. Different populations compared. Failure to adjust for injury severity scores. Not fully blinded. Patients excluded if died at scene or during transport. No power calculation.
				Immobilisation associated with higher risk of neurological disability than non-immobilisation in association with all spinal injury (OR 2.03; 95% CI 1.03 – 3.99; p =0.04) and when limited to cervical spine injury (OR 1.52; 95% CI 0.64 – 3.62).
Lin et al. 2001 Taiwan	8633 patients involved in low speed motorcycle trauma 50.7% immobilised 49.3% not immobilised	Retrospective review of trauma registry (Jan 1st 2008 – Dec 31st 2009) 2-	Epidemiology, and management of cervical spine injuries	63 had CSI	Retrospective. Single centre. Broad exclusion criteria. Data related to SND not explicitly published. Non-immobilised group small.
			Efficiency and results of immobilisation	80.9% were immobilised (12 were not)
				74.6% had neurological deficit
				No significant correlation of CSI between immobilised and non immobilised patients (p = 0.896)
Sussman BJ 1977 USA	12 cases brought forward for litigation over 5-year period	Retrospective Case Series 3	Adequacy of care	Care inappropriate in all 12 cases – multiple reasons	Retrospective. Single centre. Broad exclusion criteria. Data related to SND not explicitly published. Non-immobilised group small. Retrospective. Highly selective patient group. Old study – practices different from current.
			Causes of deterioration	67% deteriorated as a result of inappropriate care
				9 patients inadequately immobilised – 4 probably deteriorated as a result
Toscano 1988 Australia	123 patients admitted to spinal injuries unit (assessed within 7 days of admission)	Retrospective case series 22-month period (1983-1984) 3	Presence and degree of neurological deterioration between time of injury and time of admission to spinal unit (Frankel Classification)	32 (26%) sustained major neurological deterioration	Retrospective. Researcher not blinded. Potential recall bias as conducted via interview. Confounders not adjusted for. Inadequate description of study sample
			Site of deterioration	53.1% deteriorated at the local hospital, 28.1% during transfer from scene to local hospital, and 6.3% during transfer from local hospital to Spinal Unit
			Cause of deterioration	59% of deteriorations attributed to no/inadequate immobilisation
				Risk of secondary neurological deterioration increases with the length of time it takes clinicians to suspect SCI.
Harrop et al. 2001 USA	182 patients with ASIA A complete cervical SCI presenting to regional SCI centre (1993-1999)	Retrospective case series 3	Identify patients at risk of secondary neurological deterioration after complete cervical SCI	12 (6%) deteriorated within 30 days	Retrospective. Small sample size. Single specialist centre limits generalizability.
			Identify causes of secondary deterioration in early, delayed and late groups.	4 of 5 deteriorating in first 24 hours attributed to effects of immobilisation/ non-immobilisation (2 with ankylosing spondylitis after immobilising, 2 after failure to immobilise due to agitation/obesity) 5th patient deteriorated due to vertebral artery injury
Thumbikat et al 2007 USA	18 patients with Ankylosing.Spondylitis. with major neurological deficits at admission to Spinal Injuries Centre. (1996-2005)	Retrospective case series 3	Identify cause of injuries	12 of 15 trauma patients deteriorated neurologically	Retrospective. Lacks sufficient detail. Specialist patient group.
			Identify effectiveness of management and outcome	At least 3 deteriorated due to overextension of spine associated with immobilisation.
Todd et al. 2015 UK	59 cases brought forward for litigation (2001-2011)	Retrospective case series 3	Cause of neurological deterioration	27 (46%) patients deteriorated neurologically	Retrospective. Highly selected patient group, not representative of SCI patients. Selection Bias as patients involved in litigation likely to have poor outcomes.
			Whether neurological deterioration was preventable	23 probably deteriorated due to failure to immobilise
				Neurological deterioration probably avoidable in 25 patients
Oto et al 2015	41 cases from 12 English Language papers	Case review 3	Review published reports of secondary neurological deterioration after blunt trauma	Published cases are rare and poorly documented.	Subject to publication bias. Significant missing data and rarity of cases meant firm conclusions couldn’t be made.
			Describe its nature, context and risk factors	Deterioration occurred in 26 cases in the ED and 13 pre-hospitally.
				Identifiable cause of deterioration in 12 cases, further 3 deteriorated after imaging
				7 cases of deterioration within the ED probably movement-provoked, none of the 13 who deteriorated pre-hospitally were thought to be movement provoked.
				May be some association with iatrogenic manipulation in patients with altered conscious level or ankylosing spondylitis, but it remains unclear whether SND is linked to patient movement or if this is preventable by immobilisation.
Stroh and Braude. 2001 USA	504 patients with CSI transported to 5 different hospitals by EMS. (July 1990 – June 1996)	Retrospective chart review 2-	Sensitivity of selective immobilisation protocol	99% sensitive	Retrospective. Fail to publish data on outcomes in immobilised group. Missing data in non-immobilised group.
			Safety of this protocol	1.8% (9 patients) not immobilised (2 refused, 2 unable to immobilise due to kyphosis/combative, 5 missed)
				1 adverse outcome (76yr old with neurology following chiropractor manipulation. Had no change in neurology between EMS and ED assessments)
				No neurological deterioration was associated with failure to immobilise patients with CSI
Domeier et al. 2005 USA	13357 trauma patients presenting to 5 hospitals (415 with Spinal Injury)	Prospective 2-	Whether a selective immobilisation protocol can be instituted without causing harm	3% had spinal injuries	Fail to publish data on outcomes in immobilized group.
				8% of spinal injuries were not immobilised
				No non-immobilised patients with spinal injury sustained cord injury, therefore protocol use advocated
Meyers et al. 2009 USA	942 patients with trauma transferred to a level 1 trauma centre. 43 with acute spinal fracture (1year study period)	Retrospective cohort 2-	Pre-hospital assessment and immobilisation	7 fractures missed by protocol and not immobilised	Retrospective. Lack of blinding. Only patients transferred to trauma centre included. No data on patients not undergoing radiological study.
			Neurological status at discharge of patients not immobilised	No non-immobilised patients with spinal fracture had spinal cord injury or persistent neurology

Comment(s)

Spinal cord injury and resultant neurological deficit is a severe and life-threatening complication of trauma associated with significant litigation. Cervical spine injury occurs in 2-4% of trauma patients, of which around 20% have SCI. Spinal immobilisation was adopted in the 1970s as it was believed to facilitate extrication, minimise spinal movement, and prevent SND. This was based on case reports from the 1960s. However, growing evidence suggests cervical immobilisation is not a benign intervention and it is no longer recommended in penetrating trauma. As a result the dogma of spinal immobilisation in blunt trauma has also been questioned. Only level 3 and 2- evidence was identified that addressed the question. This comes from three types of study: direct comparison, case series and comparison of Emergency Medical Services (EMS) protocols. Whilst case series have linked SND to excessive spinal movement and failure to immobilise, studies comparing EMS protocols found that non-immobilised patients with spinal fractures did not deteriorate. Similarly, direct comparison studies conclude that the risk of neurological deterioration is the same or lower when patients are not immobilised. Since substantial force is required to injure the spine, small movements occurring subsequently are unlikely to have significant adverse effects and muscle contractions should be sufficient to maintain stability in awake patients. The evidence suggests that the number needed to treat with immobilisation to prevent SND is high. Although the number needed to harm is unknown, immobilisation is associated with significant adverse effects. These include pain, raised intracranial pressure, tissue ischaemia, restriction of breathing, and increased radiological investigation. Hauswald et al also attributed higher risk of SND in their immobilised group to delays in resuscitation. Collars are frequently poorly fitted which may exaggerate vertebral mal-alignment and even well fitted collars may increase distraction between vertebrae resulting in secondary injury. Certainly, immobilisation appears to increase the risk of SND in patients with Ankylosing Spondylitis. With a paucity of evidence, it is difficult to completely discount a practice that is almost universally adopted. However, with the clear risks associated with immobilisation, continuing this standard of care may well be detrimental to trauma patients and further high quality research in the area is needed. (Definitions/Abbreviations: Spinal Injury (SI)– spinal fracture with/without cord damage; Spinal cord injury (SCI) – injury to cord; Secondary neurological deterioration (SND) – neurological deterioration occurring after initial injury; Emergency Medical Services (EMS))

Clinical Bottom Line

There is insufficient evidence to reach firm conclusions as to whether immobilisation reduces the risk of SND compared to non-immobilisation in adult blunt trauma patients, and the risks and benefits of immobilisation should be balanced on a case-by-case basis ensuring appropriate priority is given to resuscitation.

References

1. Hauswald M, Ong G, Tandberg D, Omar Z Out-of-hospital spinal immobilization: its effect on neurologic injury. Acad. Emerg. Med. 1998; 5: 214–9.
2. Lin HL, Lee WC, Chen CW, Lin TY, Cheng YC, Yeh YS, et al Neck collar used in treatment of victims of urban motorcycle accidents: over- or underprotection? Am. J. Emerg. Med. 2011; 29: 1028–1033
3. Sussman BJ Proceedings of the annual scientific meeting of the international medical society of paraplegia heald at stoke Mandeville from 28-30 July 1977 (Part II): Fracture dislocation of the cervical spine: a Paraplegia 1977; 16 (1978-79): 15-38
4. Toscano J. Prevention of neurological deterioration before admission to a spinal cord injury unit. Paraplegia. 1988; 26: 143–150.
5. Harrop JS, Sharan AD, Vaccaro AR, Przybylski GJ. The cause of neurologic deterioration after acute cervical spinal cord injury. Spine. 2001; 26: 340–6.
6. Thumbikat P, Hariharan RP, Ravichandran G, McClelland MR, Mathew KM. Spinal cord injury in patients with ankylosing spondylitis: a 10-year review. Spine. 2007; 32: 2989–2995.
7. Todd NV, Skinner D, Wilson-MacDonald J. Secondary neurological deterioration in traumatic spinal injury. Bone Joint J. 2015; 97-B: 527-531
8. Oto B, Corey DJ, Oswald J, Sifford D, Walsh B. Early Secondary Neurologic Deterioration After Blunt Spinal Trauma: A Review of the Literature. Acad. Emerg. Med. 2015; 22: 1200-1212
9. Stroh G, Braude D. Can an out-of-hospital cervical spine clearance protocol identify all patients with injuries? An argument for selective immobilization. 
 Ann. Emerg. Med. 2001; 37: 609–15.
10. Domeier RM, Frederiksen SM, Welch K. Prospective performance assessment of an out-of-hospital protocol for selective spine immobilization using clinical spine clearance criteria. Ann. Emerg. Med. 2005; 46: 123–31.
11. Myers LA, Russi CS, Hankins DG, Berns KS, Zietlow SP. Efficacy and compliance of a prehospital spinal immobilization guideline. Int. J. Emerg. Med. 2009; 2: 13–17.