Spinal fractures are the third most common traumatic injury in older people, of which cervical spine injuries make up around 15%. They are predominantly seen in people living with frailty who fall from standing height. Spinal fractures in this patient group are associated with substantial morbidity and mortality (over 40% at 1 year). For many older people who survive, their injuries will be life changing. Practice between EDs varies significantly, with no universally accepted guidelines on either assessment, investigation or management specific to older people experiencing trauma. This expert practice review examines the current evidence and emergency management options in this patient group through clinical scenarios, with the aim of providing a more unified approach to management.
- spine and pelvis
- emergency department management
- accidental falls
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Trauma in older people is increasingly recognised as a significant challenge to healthcare systems in developed economies.1 In England, over 200 000 patients aged 65 years or older are admitted to hospital each year due to falls.2 In the UK, a fall from standing height is the most common mechanism of injury in major trauma.3 Spinal injuries represent the third most common injury sustained by this patient group (after head and thoracic trauma).3 Spinal injuries confer a 44% mortality rate at 1 year—eight times greater than in younger patients.4 The factors leading to the high incidence of these injuries in older people relate to issues more prevalent with ageing, including frailty, falls, polypharmacy and multimorbidity including osteoporosis.5
Cervical spine (c-spine) fractures are important to routinely assess for in all people presenting with falls. In the older population, 64% of these fractures occur in the upper c-spine, with odontoid peg fractures being the most common (up to 45%).6 Recognition of serious traumatic injury at the outset can be challenging, especially when older people are known to be less likely to receive expert clinical assessment with fewer older people receiving a trauma call compared with their younger counterparts.3 In addition, 20% of c-spine injuries are asymptomatic, making it necessary for clinicians to maintain a high index of suspicion.7
Clinical scenario one
An 85-year-old woman presents to the ED following a fall from standing while tending to her garden. She complains of exacerbation of her chronic neck pain but has no neurological symptoms. She has a history of osteoporosis, thoracic wedge fractures, osteoarthritis, hypertension and congestive cardiac failure. Her regular medications include co-codamol, furosemide, amlodipine, ramipril, bisoprolol, alendronic acid, Adcal D3 and aspirin. She is independent at home, but relies on her neighbour to perform the shopping, as she can only walk short distances with a stick.
On examination, she has a GCS of 15/15, no external head injury, a significant thoracic kyphosis and a loss of cervical lordosis. Midline c-spine tenderness is present with reduced neck movements. Neurological examination is normal. She has a laceration and haematoma to her right elbow. She is haemodynamically stable. You suspect a c-spine fracture.
How will you clear the c-spine? Which assessment tools will you use and are these validated in older people?
What immobilisation strategy will you use?
What imaging modality will you perform?
Cervical spine clearance tools
The Canadian C-spine (CCS) rule and the National Emergency X-Radiography Utilization Study (NEXUS) prediction rules (table 1) have been validated for screening c-spine injuries in alert and stable adults sustaining blunt neck trauma.8–10 These are frequently used in the ED and recommended by the National Institute for Health and Care Excellence (NICE) and ATLS.11 12 Both tools have good sensitivity, reducing the need for spinal immobilisation and imaging.13 14 In choosing which one to employ, the CCS rule has been found to be more sensitive (CCS: 100% vs NEXUS: 93.4%), while NEXUS is easier to use in clinical practice.10 15
Importantly, in older people the validity of these tools has still not been determined. The CCS rule identifies all patients over 65 years as high risk and recommends imaging for all.8 In order to reduce this imaging burden for older people, the NEXUS criteria studies have found good sensitivity in older patients.16 However, published case reports have described important c-spine fractures being missed.17 This has been borne out by a systematic review finding wide variation in the reported sensitivities of the NEXUS criteria in older people from 66% to 100%.18 While studies have suggested using modified criteria, these were too minor to change practice.19 20 Imaging the c-spine of every older person with a fall may seem excessive, and instead many clinicians opt to clear the spine clinically instead. However, clinical examination alone is unreliable with a sensitivity of 76.9% when performed by admitting clinicians.21 In view of the existing controversies in the literature, further research is required to validate and refine c-spine clearance tools in older people. We therefore advise liberal use of c-spine imaging in the interim.
Historically, there has been a tendency towards plain radiography in younger patients with single injury, and CT c-spine imaging in older populations. For decades, plain radiographs have been used for c-spine injury clearance. However, with studies demonstrating that 55.5% of clinically important fractures, including unstable injuries, are missed and a meta-analysis showing a 52% sensitivity, this approach in adults is no longer supported.22 23 Plain radiography in older patients is particularly difficult to interpret due to degenerative changes and fixed deformities, which alter the normal radiological landmarks used for c-spine fracture identification.24 However, we recognise imaging availability varies between departments and one study has advised either an accelerated plain radiography policy or the use of first-line CT in this patient group to reduce the time spent immobilised.25 While the risk of radiation is an important imaging consideration in younger patients with low-risk injuries, this is less of a concern in older people. C-spine CT should therefore be the first-line imaging choice in suspected c-spine fractures in older patients. If a c-spine fracture is found, further imaging should be considered: CT whole spine is the minimum recommended by NICE, as 45% of adults have a concomitant body injury and 10% have multilevel spinal injuries.11 26
The role of MRI in evaluating c-spine injuries in the older adults is currently unclear.22 One observational study found that CT imaging had a sensitivity of 98.5% and a specificity of 91% for clinically significant injuries, with false-negative cases displaying persistent neurological signs and symptoms confirmed on MRI.27 While MRI studies can identify c-spine injuries not visible on CT, these often include ligamentous or soft tissue injuries of undetermined clinical significance.28 In a meta-analysis, Malhotra et al found that 15% of negative CT scans had abnormalities on MRI; however, only 0.3% were unstable injuries.28 Several studies have demonstrated that in people with clinically important injuries that were not visible on CT imaging, abnormal neurology was always present.28 29 MRI should therefore be reserved for cases where there is persistent neurology which could be explained by a spinal cord injury (SCI), despite normal CT.
Immobilisation prior to imaging in suspected c-spine fractures
In adults with suspected c-spine fractures, traditional teaching suggests immobilising the c-spine with a hard collar and blocks on a long board to prevent neurological deterioration in unstable fractures.12
However, there is increasing debate on the efficacy of prehospital hard collars, with no direct evidence of prevention of spinal movement or reduced neurological injury.5 30 A systematic review by Oteir et al concluded that there was a lack of high level evidence on the effect of prehospital c-spine immobilisation on patient outcomes in adults.31
Pre-existing spinal deformities (in particular cervical lordosis and thoracic kyphosis) are common in older adults. Rigid collars can lead to inappropriate cervical hyperextension, with an increased risk of fracture displacement leading to worse neurological symptoms.5 Collars are often inappropriately fitted, with one study demonstrating that only 11% of emergency medicine practitioners were able to fit hard collars correctly.32 There is evidence that rigid cervical collars are associated with complications including pressure injuries, respiratory compromise and raised intracranial pressure.5 Furthermore, neck immobilisation in people suffering from long-term cognitive impairment or delirium can lead to significant distress and agitation.5
Recent guidelines have challenged historic dogma, with both NICE and London Major Trauma System 2021 guidelines advising a more pragmatic approach, recommending self-extrication, manual immobilisation or soft padding and tape in the most comfortable position for the patient.11 33 London Major Trauma System 2021 guidelines recommend that assessment, imaging and reporting of immobilised older patients should be completed within 2 hours of arrival in the ED. If immobilisation is required for longer, soft padded collars should be used.33
In view of the limited evidence to support hard collar immobilisation in suspected c-spine fractures, and the risks of active harm in older adults, we recommend that hard collars should not be used routinely prior to imaging in older adults in the ED. Immobilisation techniques should be patient-specific, with the aim of maintaining a comfortable position. Consideration should be given to delirium and long-term cognitive impairment, recognising where particular care is needed for those with neurological deficit.
Revisiting clinical scenario one
The presenting patient complains of pain on neck movement and a decision is made for immobilisation on a trolley using soft pads and tape in the most comfortable position possible. A CT is performed which shows a type I odontoid process fracture.
How would you manage this patient in the ED considering both fracture management and frailty risks?
Pain management is a key part of cervical fracture management, where poorly controlled pain can limit mobility and can lead to respiratory complications.5 However, pain is often under-reported by older adults, and underappreciated in this population by clinicians, commonly leading to undertreatment.34 Polypharmacy in older patients increases the risk of drug interactions and physiological changes of ageing influence the pharmacokinetics and pharmacodynamics of drug metabolism. Notably, hepatic and renal dysfunction are more prevalent in older age.35 These factors ultimately lead to a less predictable response to analgesia in older adults.
Although splinting and neck immobilisation does provide some non-pharmacological pain management, medication is usually required. Paracetamol is an effective and well-tolerated initial analgesic with minimal side effects; a dose reduction from 1 g to 500 mg four times a day is required for any adult <50 kg.35 Although a limited course of nonsteroidal anti-inflammatory drugs (NSAIDs) (with consideration of gastric protection) can be considered in specific circumstances, they are generally best avoided in older adults.35
Weak opioids are increasingly avoided in older adults. Tramadol has a highly deliriogenic side-effect profile and codeine non-responders number 25%–30% of the population, limiting its value. Low doses of strong opioids should be considered in severe acute pain due to cervical fracture. Regular low-dose immediate release morphine (eg, 2.5 mg oramorph four times a day+2 hourly PRN, or oxycodone equivalent in renal impairment) is effective and well tolerated. Prophylactic antiemetics should be co-prescribed and laxatives provided either on discharge for home use or on drug charts for admitted patients due to gastrointestinal side effects of opioids being ubiquitous. Special consideration should be made on the choice of drug administration route in immobilised patients who may struggle to swallow tablets safely. If there are no concerns of injuries below the neck, oral medication could be administered in a semi-Fowler’s position (upright sitting position at an angle between 15 and 45 degrees). In our experience, liquid preparations are often more easily tolerated. In cases where intravenous preparations are required in supine patients, low doses of intravenous morphine (1–2 mg, titrated to pain) or oxycodone in renal impairment can be considered.
Delirium and concordance
Delirium is a common complication of acute injury in the older population and is associated with significantly increased mortality and loss of independence. Agitation in hyperactive delirium can lead to poor tolerance of rigid collars, due to their restriction and discomfort.5 While delirium in the ED is often difficult to manage, initial management should include promotion of a calm, non-threatening environment, reorientation, appropriate hydration, 1:1 nursing and good analgesia to maximise comfort.36 Sedation in delirious older adults is associated with increased inpatient mortality and should therefore be reserved for situations in which the perceived risks to the patient (or others) outweigh the risk of sedation.5 If delirium is exacerbated by application of a rigid collar, it is important to reflect that the evidence base for rigid collar use is very limited. In these situations, removal of the collar may well represent the safest option, when the alternative is restraint by means of pharmacological sedation. If pharmacological sedation is deemed necessary to control agitation, the choice of agent should be based on patient comorbidities and investigations. Low-dose haloperidol (0.5 mg) can be used; however, in patients with prolonged QT, parkinsonism or with Lewy body dementia, low-dose lorazepam (0.5 mg) is an appropriate alternative.33 Further advice on the management of delirium in trauma patients can be found in the London Major Trauma System 2021 guidelines.33 Early discussion between the patient and carers with senior surgical colleagues, physicians, nurses and therapists is recommended to make best interest decisions.5 36
Immobilisation options following diagnosis of c-spine fractures
Non-surgical management aims to restrict spine movement and reduce the risk of neurological injury caused by displacement of bony fragments. Spinal immobilisation is conventionally achieved by utilisation of a rigid collar. Rigid collars have been shown to achieve similar outcomes to halo vest fixation, but with fewer complications, including respiratory distress, reduced mobility and dysphagia.24 However, the role of rigid collars is increasingly being challenged as few data exist to support their use, and an emerging evidence base describes their side-effect profile in older adults.
Complications of collars
Complications of collars in the older patient can be significant. Pressure injuries can arise as a direct consequence of the rigid collar in areas such as the chin, shoulders and back; alternatively, they may develop in conventional pressure areas such as the sacrum, elbows and heels, due to generalised immobility.5 The risk of pressure injury increases by 66% per day of cervical collar use and is associated with prolonged time to c-spine clearance.37 A review in 2019 found that dedicated c-spine collar protocols (including use of an air mattress, personalised collar choice, specific nursing care and a multidisciplinary team (MDT) approach) helped to reduce the risk of pressure injuries; this may be particularly pertinent when patients have prolonged ED stays.38 Other risks of rigid collars include respiratory compromise, swallowing impairment and increased risk of aspiration pneumonia.5 In addition, they can raise intracranial pressure through distortion of the jugular veins, which can worsen delirium, in particular in the presence of concomitant intracranial pathology.5 ED clinicians can minimise these risks by initiating early management discussions with surgical specialists and by ensuring admission to appropriately skilled wards.
While it is recognised that hard collars are likely to be more efficacious at restricting c-spine movement, soft foam collars may provide enough proprioceptive feedback to prevent clinically significant c-spine movement in older people.5 International opinion on the role for rigid collar immobilisation in the older adult is heterogeneous, and clinical trials are currently underway to inform evidence-based practice.39
Mobility and social considerations
C-spine management choices will have a significant impact on both mobility and independence, with many older people requiring supportive care immediately on discharge.5 Early physiotherapy and occupational therapy involvement is key, focusing on both returning to preinjury function and on future falls prevention. All patients must be assessed and receive a rehabilitation prescription within 24 hours of admission.40 Outcome measures are used to guide the development of problem lists, patient-centred goals and rehabilitation planning. There is currently no consensus on a core outcome set for this population, but a combination of global (eg, Functional Independence Measure), age-specific (eg, Elderly Mobility Scale), balance (eg, Berg Balance) and neck-specific measures (eg, Neck Disability Index, Neck Pain and Disability scale) may be used. Immobilisation may impact vision, balance and ability to self-care. Aids and adaptations are likely to be required. Collars will also require frequent changing, although older people may not be able to execute donning and doffing of rigid collars independently.41 Patient comorbidities, frailty and other injuries will have a significant impact on potential complications and recovery and must all be considered when having management discussions.
Considerations for onward referral for surgical fixation
The definitive management strategy of c-spine fractures in older people is significantly debated in the literature and varies from management with rigid collars or halo vests to surgical fixation. The ED clinician should be aware of factors which impact management decision-making when discussing these patients with surgical colleagues. Patient factors associated with worse outcomes following surgery include initial non-operative management, age ≥75 years, male gender and comorbidity (specifically including osteoporosis, dysphagia and cardiopulmonary disease).42 Fracture morphology also dictates management: surgical fixation in type II odontoid fractures can be considered, but only in patients able to tolerate surgery and at significant risk of instability (suggested by displacement or angulation and neurological deficit).43 Operative morbidity, such as a 50% loss of rotation when fusing posteriorly, may not be acceptable to patients. Fibrous union of the odontoid peg following non-surgical management may be adequate, especially in type I and III odontoid fractures.5 24 43
Lower c-spine fractures are less common in the older population and vary significantly in morphology and stability. There are a multitude of surgical management options which are dependent on fracture type and patient characteristics. Further discussion of these is beyond the scope of this article; however, they should be managed on a case-by-case basis with early surgical review.
In view of the complexity of the management choices, there should be early surgical referral (ideally within 4 hours of diagnosis).33 Review in the ED should include careful shared decision-making with the patient and carers.
Clinical scenario one summary
The presenting patient is immobilised with a Miami J collar and following early discussion with the surgical team is managed conservatively on the ward with input from the trauma geriatricians. She develops a pressure sore on her chin, but otherwise makes a good recovery and is discharged back to her home with a two times per day package of care.
Clinical scenario two
An 84-year-old woman presents to the ED by ambulance after having been found on the floor of her living room by her morning carer. She is more confused than normal and cannot recall the events leading to her fall. She complains of abdominal pain and feels weak and tired. She has recently been diagnosed with Alzheimer’s dementia and lives alone with four times a day care. Her medical history also includes osteoarthritis, spinal stenosis and bladder incontinence. Her regular medications are memantine, mirabegron and paracetamol. On examination, she is haemodynamically normal (heart rate 85 beats min-1, blood pressure 145/72 mm Hg, respiratory rate 18 min-1, without fever), GCS 14/15 and has a small laceration to her forehead with no other obvious injuries. She is weaker in her upper limbs bilaterally compared with lower limbs and displays reduced bicep reflexes bilaterally, with no altered sensation. Her abdomen is soft with suprapubic fullness and a bladder scan reveals she is in urinary retention. Rectal examination indicates slightly reduced anal tone, but intact sensation. There are no indications of an infection, with normal blood results, ECG and chest X-ray. A CT of head and c-spine is unremarkable.
In view of the history, what features are concerning and what further investigations are required?
How would you manage this patient in the ED?
Central cord syndrome
Central cord syndrome (CCS) is the most common incomplete SCI and is often secondary to c-spine hyperextension trauma, arising after falls in older patients with existing cervical stenosis. Only 10% are associated with c-spine fractures.44
First described by Schneider et al, the syndrome is characterised by motor weakness primarily in the upper limbs, with varied changes in limb sensation and bladder function.44 45 Cervical neck pain is not necessarily present nor required for diagnosis as the pathology is not directly related to bony injuries. Neurological injury occurs secondary to spinal cord pinching, as a result of ligamentum flavum or anterior osteophyte protrusion during cervical hyperextension.45 Sensory disturbance can present with bilateral loss of pain and heat sensation over the shoulders and upper limbs.45 Attempting to standardise diagnosis, van Middendorp et al surveyed specialist surgeons on diagnostic criteria for CCS; they reported that 75% considered motor impairments essential for diagnosis, while sensory disturbance and bladder dysfunction were less frequently considered diagnostic (39% and 24%, respectively).46 The authors concluded that a difference of 10 points in the American Spinal Injury Association score between upper and lower limbs was an essential diagnostic criterion.46 For emergency physicians, key to the diagnosis is having a high clinical index of suspicion, particularly in older people presenting with weakness of the upper limbs after falling.
Management of CCS in the ED
While early CT c-spine is important to identify bony injuries and is realistically achievable in the ED, MRI is the imaging modality required to determine the severity of cord injury.45
Non-surgical and initial management of CCS include immobilisation in a hard collar and careful haemodynamic control to ensure adequate perfusion of the spinal cord.45 47 Management should be patient-specific and will depend on the presence of concomitant spinal fractures, symptom severity and frailty.47 Surgical decompression can be considered in the context of persistent or progressive neurological signs and symptoms where there is evidence of persistent mechanical cord compression.47
Clinical scenario two summary
An MRI confirmed spinal cord compression related to a disc-osteophyte complex. Due to persistent compression and progressive neurological signs and symptoms, she was managed with spinal decompression surgery and had a slow recovery of neurological symptoms. She was discharged after 2 weeks to a rehabilitation unit.
C-spine injuries are common, particularly in older people, with a lack of well-evidenced recommendations on their emergency management. Falls from standing height are closely associated with advancing age and are already the most common cause of injuries meeting major trauma criteria. EDs must adapt to ensure the specific management needs of this patient group are met with pragmatic evidence-based practice.
While spinal clearance scoring systems have been validated in wider adult trauma, there is little evidence to support their use in the older patients. A liberal use of radiology is recommended to ensure occult injuries are not missed. Plain radiography is insufficiently sensitive to exclude fractures and CT imaging should be the first-line imaging modality. In patients presenting with abnormal neurology, MRI should be considered even where CT does not reveal injury. Hard collars in pre-diagnosis immobilisation are not recommended in older people, and if necessary, soft pads and tape should be used to maximise comfort. Ongoing management can be complex and should account for patient-specific factors, such as frailty, comorbidity, prognosis and quality of life. A multidisciplinary approach involving spinal surgeons, trauma geriatricians and dedicated therapists are needed to engage in careful shared decision-making with patients and their carers to best achieve optimal outcomes.
Patient consent for publication
Handling editor Mary Dawood
Twitter @BenchetritSandy, @julesblackham, @DrPhilipBraude, @ruth_halliday, @adampwilliams13, @EddCarlton
Contributors EC planned the manuscript following invitation from the editorial board and guarantees the overall content. SB, JB, PB, RH, DS, AW and EC contributed equally to the manuscript content and approved final submission.
Funding PB is directly funded by the Research Capability Funding North Bristol Trust Research and Innovation Department and EC by the NIHR Advanced Fellowship.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.