The incidence of spinal cord injury (SCI) in the USA is estimated to be between 40 and 50 cases per million people per year (SCI Center 1998). Spinal cord injury results in long-term disability, often with profound effects on the quality of life of the affected individuals and their carers. In the USA, the lifetime medical costs resulting from spinal cord injury are estimated at nine billion dollars per year (Miller 1994). Existing data in developing countries are limited. A study from Beijing estimated the incidence of SCI at seven cases per million people per year (Wang 1990). Acute traumatic SCI occurs in about 3% of trauma admissions, and around half of these injuries involved the cervical spine (Burney 1993). In males under the age of 50, road traffic crashes are the most common cause of SCI (Burney 1993).
In response to the concern that an unstable spine will increase the frequency and severity of neurological injury, a number of approaches have been developed that aim to achieve spinal immobilisation. The two main methods are manual stabilisation and the use of orthotic devices such as backboards and splints, with a combination of adjuncts including cervical collars, sandbags and straps. Pre-hospital spinal immobilisation aims to stabilise the spine by restricting mobility, thus preventing secondary SCI during extrication, resuscitation, transport and evaluation of trauma patients with suspected spinal instability. It is estimated that 5% of trauma patients with cervical spinal injuries have missed or delayed diagnosis (Davis 1993), resulting in preventable mortality and morbidity. Occult cervical spine injuries may be more likely to be missed in obtunded patients with unstable spines, in whom it may be masked by the pain of multi-system injury and altered level of alertness. Spinal immobilisation is now routinely practised in the pre-hospital care of trauma patients and is widely recommended in a range of resuscitation guidelines (Advanced LS 1993, Advanced Paediatric Life Support, Pre-hospital Trauma Life Support, Advanced Life Support Group 1993, ACS 1997).
Despite the widespread use of spinal immobilisation, the clinical benefits of pre-hospital spinal immobilisation have been questioned. It has been argued that spinal cord damage is done at the time of impact and that subsequent movement is generally not sufficient to cause further damage (Hauswald 1998). Most trauma patients do not have spinal instability and, hence, will not benefit from spinal immobilisation. Nevertheless, largely in response to the fear of litigation, some five million patients in the US receive spinal immobilisation every year (Orledge 1998). However, there may be adverse effects. Observational studies have shown that rigid collars may cause airway difficulties, increased intracranial pressure (Davies 1996), increased risk of aspiration (Butman 1996), restricted respiration (Totten 1999), dysphagia (Houghton 1996) and skin ulceration (Hewitt 1994). Because any benefits of spinal immobilisation may be outweighed by the risks, the value of routine pre-hospital spinal immobilisation remains uncertain.
This systematic review aims to quantify the effect of different spinal immobilisation devices (including immobilisation versus no immobilisation) on their ability to immobilise the spine and on mortality, neurological injury, and adverse effects in trauma patients.
- To quantify the effect of spinal immobilisation versus no spinal immobilisation on mortality, neurological injury, spinal stability and adverse effects in trauma patients.
- To quantify the effect of different spinal immobilisation strategies on mortality, neurological injury, spinal stability and adverse effects in trauma patients.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials.
Types of participants
Trauma patients with suspected spinal cord injury.
Types of interventions
All strategies of spinal immobilisation including:
- backboards, mattress splints
- rigid and soft collars
- sandbags, straps or tapes
- collar and backboard combinations
- holding the head in the midline
- log rolling the patient.
Types of outcome measures
- Neurological injury.
- Degree of spinal stability.
- Adverse effects.
Search methods for identification of studies
We searched the following electronic databases;
- Cochrane Injuries Group's specialised register
- Cochrane Central Register of Controlled Trials (CENTRAL)
- National Research Register
These searches were last carried out in July 2007. The full search strategies are presented in the additional tables: Table 1 shows search strategies used previously in May 2003, Appendix 1 shows strategies used for the July 2007 update.
Searching other resources
Additionally all references in the background papers were checked and six authors contacted to identify potential published or unpublished data. Eight manufacturers of immobilisation devices were also contacted. There was no language restriction in any of the searches.
Data collection and analysis
Selection of studies
One author (IK) examined the electronic search results for reports of possibly relevant trials and these reports were then retrieved in full. One author (FB) examined 10% of the electronic search results to check for agreement on eligibility criteria. Two authors (FB, IK) applied the selection criteria independently to the trial reports, resolving disagreements by discussion with a third author (IR).
The following are the proposed methods which will be applicable if trials are found during subsequent updates of the review.
Data extraction and management
Two authors will independently extract data and information on the following:
- method of allocation concealment,
- number of randomised patients,
- type of participants,
- type of interventions,
- loss to follow-up,
- length of follow-up.
The authors will not be blind to the study authors or journal when doing this. Results will be compared and any differences resolved by discussion.
Where there is insufficient information in the published report, we will attempt to contact the trial authors for clarification.
Assessment of risk of bias in included studies
Since there is evidence that the quality of allocation concealment particularly affects the results of studies (Schulz 1995), two authors will score this quality on the scale used by Schulz as shown below, assigning C to poorest quality and A to best quality:
- A = trials deemed to have taken adequate measures to conceal allocation (that is, central randomisation; serially numbered, opaque, sealed envelopes; or other description that contained elements convincing of concealment)
- B = trials in which the authors either did not report an allocation concealment approach at all or reported an approach that did not fall into one of the other categories.
- C = trials in which concealment was inadequate (such as alternation or reference to case record numbers or to dates of birth).
If the method used to conceal allocation is not clearly reported, the trial author(s) will be contacted, if possible, for clarification.
Differences will be resolved through discussion.
We will assess the skewness of continuous data by checking the mean and standard deviation (if available). If the standard deviation is more than twice the mean for data with a finite end point, the data are likely to be skewed and it is inappropriate to apply parametric tests (Altman 1996). This is because the mean is unlikely to be a good measure of central tendency. If parametric tests cannot be applied, we will tabulate the data.
Assessment of heterogeneity
The groups of trials will be examined for statistical evidence of heterogeneity using a chi-squared test. If there is no obvious heterogeneity on visual inspection or statistical testing, pooled RR and 95% confidence intervals will be calculated using a fixed effects model.
The following comparisons are proposed;
- spinal versus no spinal immobilisation,
- different strategies of spinal immobilisation.
For dichotomous outcomes, such as death, the relative risk (RR) will be calculated with 95% confidence intervals, such that a RR of more than 1 indicates a higher risk of death in the first group named. The RR will be used as it is more readily applied to the clinical situation.
The effect of excluding trials judged to have inadequate (scoring C) allocation concealment will be examined in a sensitivity analysis.
Description of studies
No randomised controlled trials comparing the effect of spinal immobilisation strategies on trauma patients were found.
Risk of bias in included studies
Effects of interventions
Our search strategy identified 4453 potentially eligible reports. However, there were no trials meeting the inclusion criteria. A number of randomised controlled trials were identified comparing different spinal immobilisation strategies in healthy volunteers. The results of randomised controlled trials on healthy volunteers may provide some useful insights into their relative effectiveness in trauma patients. For this reason, although trials of healthy volunteers did not meet our inclusion criteria, we have summarised them in the additional tables ( Table 2) of the review.
We did not find any randomised controlled trials comparing different strategies of spinal immobilisation in trauma patients. The effect of spinal immobilisation on mortality, neurological injury, spinal stability and adverse effects in trauma patients therefore remains uncertain.
We screened 4453 potentially relevant papers, checked their reference lists and contacted experts in the field. We also contacted manufacturers of immobilisation devices for additional information. While it is possible that we might have missed a randomised controlled trial comparing spinal immobilisation techniques in trauma patients, we believe that, due to our thorough search strategy, this is unlikely.
The current protocol for pre-hospital spinal immobilisation has a strong historical rather than scientific precedent, based on the concern that a patient with an injured spine may deteriorate neurologically without immobilisation. The medical and legal concern of missing a cervical spinal injury has lent strong support for the conservative approach of liberal pre-hospital spinal immobilisation to almost all patients with trauma and possible neck injury, regardless of clinical complaint (Butman 1996). It is also suggested that iatrogenic cord damage could be reduced with better paramedic training and improved immobilisation procedures (Perry 1999). However, it has been argued that considerable force is required to fracture the spine at the initial impact, and that any subsequent movements of the spine are unlikely to cause further damage to the spinal cord (Hauswald 1998). It has also been suggested that pre-hospital spinal immobilisation has never been shown to affect outcome and that estimates in the literature regarding the incidence of neurological injury due to inadequate immobilisation may have been exaggerated (Hauswald 1998; Hauswald 2000). This calls into question the present routine use of pre-hospital spinal immobilisation.
For some patients, effective spinal immobilisation is prudent and can be vital to prevent the devastating effects of cord damage, yet for many the excessive use of this precaution may not be beneficial or necessary. It is estimated that over 50% of trauma patients with no complaint of neck or back pain were transported with full spinal immobilisation (McHugh 1998). Unwarranted spinal immobilisation can expose patients to the risks of iatrogenic pain, skin ulceration, aspiration and respiratory compromise, which in turn can lead to multiple radiographs, resulting in unnecessary radiation exposure, longer hospital stay and increased costs. The potential risks of aspiration and respiratory compromise are of concern because death from asphyxiation is one of the major causes of preventable death in trauma patients.
A set of highly sensitive clinical criteria has been developed and validated (Hoffman 2000) to identify trauma patients at low risk of spinal injury and rule out their need for radiography. These are trauma patients with absence of: neck pain or tenderness, altered level of consciousness, neurological deficit, evidence of intoxication and painful distracting injury. It has been suggested that a similar decision instrument could be developed for use in the pre-hospital setting, to establish the need to immobilise or not to immobilise (Domeier 1999). This is in addition to the criteria of mechanism of injury as the main determinant for out-of-hospital spinal immobilisation.
There are a lack of data from randomised controlled trials to support the practice of pre-hospital spinal immobilisation in trauma patients. While it may not be possible to conduct randomised controlled trials of spinal immobilisation versus no immobilisation in trauma patients, it may be feasible to consider such trials, comparing the different spinal immobilisation strategies, in outcomes of immobilisation efficacy, respiratory effects, tissue pressure and patient comfort in this target population. Results of randomised controlled trials on healthy volunteers may provide some useful insights into their relative effectiveness in trauma patients. For this reason although trials of healthy volunteers did not meet our inclusion criteria we have summarised them in the additional tables section of the review. For example in healthy volunteers, short-board technique was reported to be more efficient than collars alone in reducing spinal mobility (Cline 1985); vacuum mattress and padded backboards more comfortable than rigid backboards (Hamilton 1996; Hauswald 2000; Johnson 1996; Walton 1995). From these studies on healthy volunteers, it has been suggested that patients on whom spinal immobilisation has been used, and who are conscious, might reposition themselves to relieve the discomfort caused by ischaemia, which could theoretically worsen any existing spinal injuries. Patients who are unable to move or feel pain due to trauma are at risk of soft tissue injuries (Hauswald 2000).
Due to the absence of randomised controlled trials quantifying the effect of spinal immobilisation in trauma patients, and the possible adverse effects of its application, the value of routine pre-hospital spinal immobilisation remains uncertain.
Implications for practice
We found no randomised controlled trial which met our inclusion criteria in this review. The effect of pre-hospital spinal immobilisation on mortality, neurological injury, spinal stability and adverse effects in trauma patients therefore remains uncertain. Because airway obstruction is a major cause of preventable death in trauma patients, and spinal immobilisation (particularly of the cervical spine) can contribute to airway compromise, the possibility that immobilisation may increase mortality and morbidity cannot be excluded.
Implications for research
Large prospective studies are needed to validate the decision criteria for spinal immobilisation in trauma patients with high risk of spinal injury. In addition, randomised controlled trials to compare different immobilisation strategies on trauma patients need to be considered in order to establish an evidence base for the practice of pre-hospital spinal immobilisation.
We thank D Mohan, C Mock, R Norton and M Varghese of the WHO Pre-hospital Trauma Care Steering Committee for their comments and advice on the review.
We also thank R Wentz and K Blackhall for help with the searching, and Dr Jegede for help identifying useful background papers.
Finally, thanks to the authors of background papers and manufacturers for supplying additional information.
Data and analyses
This review has no analyses.
Appendix 1. Search strategy
July 2007 update search strategies
INJURIES SPEICALISED REGISTER
(spine or spinal) AND (immobile or immobilize or immobilization or stabili* or stable or brace or splint*)
MEDLINE 2007/June week 4
1.exp Spinal Injuries/
2.exp Spinal Cord Injuries/
3.((spine or spinal or cervix or cervical or lumbar or thora$) adj3 (injur$ or trauma$)).ab,ti.
7.exp Orthotic Devices/
8.(backboard$ or vacuum splint$ or neutral position or strapping or strapped or straps or spine board$ or tapes or taping or log roll$).ab,ti.
9.(headblock$ or sandbag$).ab,ti.
11.5 and 10
12.(randomised or randomized or randomly or random order or random sequence or random allocation or randomly allocated or at random or controlled clinical trial$).tw,hw.
14.12 or 13
15.exp models, animal/
17.exp Animal Experimentation/
18.exp Disease Models, Animal/
19.exp Animals, Laboratory/
22.20 not 21
23.14 not 22
24.11 and 23
EMBASE 2007/ week 27
1.exp Spinal Cord Injury/
2.exp Spine Injury/
3.((spine or spinal or cervix or cervical or lumbar or thora$ or neck) adj5 (injur$ or trauma$)).ab,ti.
8.(backboard$ or vacuum splint$ or neutral position or strapping or strapped or straps or spine board$ or tapes or taping).ab,ti.
9.(headblock$ or sandbag$ or orthosis or orthotic or brace$ or splint).ab,ti.
10.(immobili$ or mobility or stabili$ or collar$ or log roll$).ab,ti.
12.5 and 11
13.exp animal model/
16.exp Experimental Animal/
17.13 or 14 or 15 or 16
19.17 not 18
20.(randomised or randomized or randomly or random order or random sequence or random allocation or randomly allocated or at random or controlled clinical trial$).tw,hw.
21.exp clinical trial/
22.20 or 21
23.22 not 19
24.12 and 23
Central 2007, issue 2 and National Research Register 2007, issue 2
#1MeSH descriptor Spinal Injuries explode all trees #2MeSH descriptor Spinal Cord Injuries explode all trees
#3injur* and (spine or spinal or cervix or cervical or lumbar or thora* or neck)
#4trauma* and (spine or spinal or cervix or cervical or lumbar or thora* or neck)
#6(#1 OR #2 OR #3 OR #4 OR #5)
#7MeSH descriptor Immobilization explode all trees
#8MeSH descriptor Orthotic Devices explode all trees
#9immobili* or mobility or stabili* or collar* or orthotic or orthosis or brace* or splint*
#10backboard* or vacuum splint* or neutral position or strapping or strapped or straps or spine board* or tapes or taping or log roll*
#11headblock* or sandbag*
#12(#7 OR #8 OR #9 OR #10 OR #11)
#13(#6 AND #12)
#14(#13), from 2003 to 2007
www.clinicaltrials.gov and http://www.controlled-trials.com/mrct
(spine or spinal) AND ( immobile OR immobilize or immobilization or stabilize or stable or brace or splint ) [ALL-FIELDS]
spinal* immobil* trial*
spine* immobil* trial*
spinal immobil* random*
spine* immobil* random*
Last assessed as up-to-date: 30 June 2007.
Protocol first published: Issue 3, 2000
Review first published: Issue 2, 2001
Contributions of authors
IK helped to design the protocol, examined search results, applied inclusion criteria and wrote the review. FB examined search results, applied inclusion criteria, and helped to write the review. IR commented on the protocol and helped to write the review.
Declarations of interest
Sources of support
- Institute of Child Health, University of London, UK.
- Global Programme on Evidence for Health Policy (GPE), World Health Organisation, Switzerland.
Medical Subject Headings (MeSH)
MeSH check words