Article Text
PDF
Abstract
Therapeutic hypothermia is already widely acknowledged as an effective neuroprotective intervention, especially within the acute care setting in relation to conditions such as cardiac arrest and neonatal encephalopathy. Its multifactorial mechanisms of action, including lowering metabolic rate and reducing acute inflammatory cellular processes, ultimately provide protection for central nervous tissue from continuing injury following ischaemic or traumatic insult. Its clinical application within acute traumatic spinal cord injury would therefore seem very plausible, it having the potential to combat the pathophysiological secondary injury processes that can develop in the proceeding hours to days following the initial injury. As such it could offer invaluable assistance to lessen subsequent sensory, motor and autonomic dysfunction for an individual affected by this devastating condition. Yet research surrounding this intervention’s applicability in this field is somewhat lacking, the majority being experimental. Despite a recent resurgence of interest, which in turn has produced encouraging results, there is a real possibility that this potentially transformational intervention for treating traumatic spinal cord injury could remain an experimental therapy and never reach clinical implementation.
- clinical management
Statistics from Altmetric.com
Key messages
Acute traumatic spinal cord injury is a physically and psychologically devastating condition that predominately occurs in young adults following road traffic collisions, sports injuries and violence.
Therapeutic hypothermia has an established neuroprotective role within other acute care settings, such as cardiac arrest and neonatal encephalopathy, but not within spinal traumatic injury.
Although this intervention has the theoretical potential to limit pathophysiological secondary injury processes within this setting, the research is limited and inconclusive.
More research—especially large, multicentre, randomised controlled trials—is required before there is any likelihood of this experimental intervention reaching clinical implementation.
Introduction
Acute traumatic spinal cord injury is widely regarded as one of the most physically and psychologically devastating conditions for affected individuals.1 It predominately occurs in young adults, most notably men aged 20–44 years old,2 and typically as a result of road traffic collisions, sports injuries, falls and violence, with 51.6% of cases involving the cervical spine, 46.3% the thoracic spine and 2.1% the lumbosacral region.3 Whereas it has been acknowledged since the 19th century that this type of injury results in partial or complete disruption to neural pathways that transmit, modify and coordinate motor, sensory and autonomic control of organ systems,4 the actual mechanisms underlying neural tissue damage have only more recently started to be understood. Patient morbidity and mortality were originally viewed solely in relation to the severity of the initial injury, with surgical stabilisation of fractures being the only available intervention, and many early clinicians believing ‘higher-level’ cord injury to be untreatable.5 However, in the early 20th century theories, began to arrive that traumatic spinal cord damage resulted not just from the primary insult but also from a tissue-destructive secondary injury process in the ensuing hours and days, and approaches to managing this condition have since been revolutionised.1 5 As well as improved surgical techniques and the development of pharmacological interventions, other neuroprotective strategies have also started to emerge aimed at combating this secondary injury mechanism and thereby improving patients’ immediate and longer term prospects.2 6–9 It is one of these neuroprotective strategies that this report intends to focus on: that of therapeutic hypothermia, with the aim of establishing what it is and the rationale behind its use in this setting, whether this use is supported by research, and what the future holds.
An overview of therapeutic hypothermia
Therapeutic hypothermia, also known as ‘targeted temperature management’ (p1)10, is a neuroprotective intervention that has been used clinically since the middle of the 20th century, its first recorded use actually being for the slowing of metastatic breast cancer.11 Its application has subsequently become more focused on neural tissue preservation in the presence of cerebral ischaemia or trauma,12–14 and nowadays it is considered an essential strategy in the acute care setting in relation to conditions such as cardiac arrest and neonatal encephalopathy.15 Extensive, well-designed and sizeable randomised controlled trials, retrospective studies and meta-analyses have comprehensively validated therapeutic hypothermia’s application in these fields,16–23 and such is the belief in this intervention’s usefulness in preserving cerebral integrity following resuscitation from cardiac arrest that the European Resuscitation Council has even endorsed its use since 2010 within their Advanced Life Support guidelines.24 Whereas its early use involved lowering body temperatures to beneath 30°C using localised cooling, this was abandoned by the 1990s due to it inducing homeostatic protective mechanisms such as shivering that made temperature regulation difficult,25 26 and also due to it precipitating clinical complications. These included cardiac arrhythmias, hypotension, coagulopathies, electrolyte disorders and systemic infections, all of which were observed in well-conducted clinical and experimental trials.27–31 Nowadays, milder temperatures are used with more controlled systemic cooling methods, allowing for more precise maintenance of core temperatures at 32°C–34°C, and the subsequent avoidance of these previous issues.9 32 33
The mechanisms of action of therapeutic hypothermia are multifactorial and essentially relate to its ability to protect neural tissue in the central nervous system from continuing injury following ischaemic or traumatic insult and thereby minimise overall loss of neurological function.15 34 One of the most significant of these mechanisms is its capacity to decrease metabolic rate and so help conserve oxygen, glucose and overall energy levels for use by nervous tissue, small changes of which it is very sensitive to.10 This was demonstrated well by a comprehensive review of experimental studies, which revealed that hypothermia reduced metabolic rate in neural tissue in mammals by 2–4 times per 10°C reduction in temperature.35 Although criticism could be made as to how trustworthy and applicable these animal-based results are to humans, the number of well-designed and controlled experiments reviewed that produced concurrent results in mammalian subjects offers very persuasive evidence as to the effectiveness of this intervention here as well as its feasible relevance to humans.
In addition to reducing metabolic rate, therapeutic hypothermia has also been suggested to possess other neuroprotective qualities that can help to minimise damage to neural tissue following ischaemia or trauma. These comprise the reduction of acute swelling and inflammation, and the physically compressive effects that this can produce, as well as tissue damage from immunological reactions,9 10 and the inhibition of resultant pathophysiological cellular processes that ultimately lead to increased tissue disruption. These include free-radical generation and resultant oxidative stress, excitotoxicity and electrolyte imbalance, and apoptotic and necrotic cell death.26 32 34 Extensive experimental studies from the last 19 years have comprehensively evidenced the effectiveness of therapeutic hypothermia in controlling these damaging processes,36–49 and although once again liable to criticism due to being animal tissue-based, the use of mammals and good study designs tend to offset any major concerns.
Having looked at the neuroprotective mechanisms of action of therapeutic hypothermia, it is of no surprise that it is increasingly being considered for use in other clinical settings where cerebral nervous tissue is at increased risk of damage, such as traumatic head injury, stroke, cardiac surgery, and the surgical repair of cerebral and thoracoabdominal aortic aneurysms.9 50 51 In the same way, its use within acute traumatic spinal cord injury would also seem very plausible. However, to truly appreciate the benefits that this intervention might potentially offer, and hence any rationale for its use in the setting of acute spinal injury, a clearer understanding of the injury mechanism surrounding this devastating condition is first required.
Application of therapeutic hypothermia in acute spinal cord injury
Acute trauma to the spinal cord is incurred when the protective and stabilising bones and ligaments of the vertebral column fail through excessive movement, load or penetration, resulting in damage to the cord itself, ranging from a minor contusion through to a complete tear or transection.4 52 Depending on the vertebral level of the injury, its location within the spinal cord and hence which functionally specific areas of neuronal cell body-containing grey matter or axon-prolific white matter are involved, and the overall extent of the damage, the affected individual can experience anything from brief transient neurological deficit through to complete tetraplegia, loss of organ system control and ultimately death.53–55 Due to the complexity of neurological injury, the ‘ASIA Impairment Scale’, or AIS (see Table 1), was introduced by the American Spinal Injury Association (ASIA) in the 1990s. This scale simplistically grades the severity of spinal cord injury from A to E according to the extent of sensory and/or motor loss,56 with classification being advised at least 72 hours postinjury due to spinal shock often confusing the initial picture.4 54
The American Spinal Injury Association Impairment Scale: AIS A–E
As mentioned earlier, acute spinal cord injury is not believed to relate solely to the primary traumatic insult. During proceeding hours to days, a secondary injury process can develop that radiates out from the primary injury epicentre, affecting adjacent uninjured neural tissue, and ultimately increasing the overall lesion size and resultant neurological deficit.57 It is currently believed that this secondary injury process stems predominately from post-traumatic ischaemia caused by mechanical damage to microvasculature at the injury site, the previously mentioned cascade of pathophysiological cellular processes that result from neural tissue trauma and ischaemia, including free-radical-induced oxidative stress, excitotoxicity, and apoptosis and necrosis,52 58 ultimately exacerbating the injury to the spinal cord. This was first observed in postmortem studies conducted in Australia, Canada and America,59–62 and although these studies were small and few in number, their findings have since been corroborated by extensive, well-designed and more sizeable experimental studies and controlled trials.63–70
In addition to this, there is also a belief that acute inflammation also contributes to this secondary injury process. This normally protective mechanism is viewed as a ‘dual-edged sword’ (p33)71 in spinal cord injury, with the early inflammatory response causing increased tissue destruction through the activity of immunological agents such as neutrophils, macrophages and microglia, as well as damage due to physical compression. This theory, originally supported by limited experimental trials,72 73 has since become comprehensively endorsed by more recent and well-devised experimental and human studies74–79 establishing it alongside post-traumatic ischaemia as a further mechanism in this secondary injurious process.
As well as producing a clearer picture as to the existence and workings of this secondary injury process, these well-evidenced mechanisms also help to provide reasonable grounds for considering the use of therapeutic hypothermia and its previously mentioned neuroprotective properties in helping to manage this injury-worsening activity in acute spinal cord injury. Although unable to alter the initial damage caused, the ability to possibly attenuate any secondary injurious mechanisms, and thereby help preserve residual motor, sensory and autonomic neural pathways for affected individuals, is potentially invaluable, especially considering the poor prognosis that still currently surrounds this condition.2 5 80 However, as with any clinical intervention, clear evidence from well-designed studies and trials is needed before any consideration can be given to its implementation into clinical practice.81 82 As such, it is to this that this report must now turn so as to establish whether this intervention really does have a role to play in acute traumatic spinal cord injury.
Research surrounding the use of therapeutic hypothermia in acute spinal cord injury
Studies and trials surrounding the use of therapeutic hypothermia in traumatic spinal injury date from the 1960s, but it is only more recently that progress has started to be made on this front. Previous evidence predominately consists of animal-based experimental studies and limited human clinical trials, and although some of these investigations have produced interesting results, the majority of the research is far from reliable or consistent and so is of little value for establishing whether therapeutic hypothermia could be an effective intervention in this acute setting. For example, animal experiments from the 1960s and 1970s83–95 (see Table 2), although extensive and adequately controlled, and which produced reports of therapeutic hypothermia being beneficial in 12 of the 13 studies conducted,83–94 have been heavily criticised due to a lack of consistency in study methods and due to experimental designs having little relevance to human spinal cord injury or more modern clinical use of hypothermia.58 80 96 For instance, a wide variety of animals with differing spinal anatomy were used; the means of inducing spinal injury were varied and produced different levels of insult; the measurement of functional capacity in animals before and after experimentation was non-standardised; and the timings for cooling often differed. Not only this, but also localised cooling and temperatures beneath 12°C were used, which differs substantially from more modern clinical practice, and only thoracic spinal injuries were studied as opposed to cervical injuries as well, which decreases their relevance to human spinal cord trauma.
A summary of early experimental studies (1965–1978) into the effects of hypothermia on traumatically induced spinal cord injury
Early clinical trials were unfortunately of similar poor quality, the authors from each of the studies conducted from 1971 to 198497–103 openly admitting the limitations and statistical insignificance of their results due to small numbers involved, lack of controls and randomisation, concurrent use of surgical and pharmacological interventions, and large variation in patient ages, injury severity and duration of hypothermic intervention (see Table 3). Interestingly, all of the trials produced some degree of positive clinical outcome, with motor and sensory improvement being observed in cervical injuries as well as thoracic ones. However, due to there being no standardised neurological assessment tool available for use at baseline and at follow-up at this time, these results are impossible to truly evaluate, either individually or across the spectrum of studies. When further limitations are also considered, such as localised cooling being used with body temperatures beneath 7°C as opposed to more modern systemic cooling methods and milder temperatures of 32°C–34°C, then the substantially reduced usefulness of these studies can be more fully appreciated.
A summary of non-controlled early clinical trials (1971–1984) investigating the effects of hypothermia in spinal cord trauma
Despite this flawed earlier research, more recent animal-based experimental studies have proved to be of far greater use for considering the applicability of therapeutic hypothermia in acute traumatic spinal cord injury. As well as using same-species mammals with parallels to human spinal anatomy, systemic cooling to 32°C–34°C and good use of controls, studies have also adopted the use of a standardised locomotor-recovery measurement tool—the ‘open-field locomotor test’104—as well as detailed histological examination. As such, they have produced far more in-depth and reliable results, as well as much of it being very encouraging (see Table 4). For instance, studies measuring functional capacity in spinally injured rats unanimously found that cooled animals had improved locomotor ability scores over normothermic controls at follow-up,105–110 the difference being as great as 35%–40% in some studies.107 109 110 Interestingly, the results were statistically significant in all but one of the experiments, the authors of the stand-alone study attributing this outcome to their initial spinal injuries being excessively severe.106 Studies using histological examination have been equally as encouraging, with lesion progression in traumatically injured rat spinal cords being shown to be statistically significantly reduced in cooled rats versus controls in all conducted experiments,105 107–113 with differences in lesion size being as large as 16% in several studies.108–110 Furthermore, all of these studies also reported equal sparing of grey matter as well as white matter in cooled spinal cord samples, this helping to further endorse the neuroprotective efficacy of therapeutic hypothermia considering the normally heightened sensitivity of neuronal cell bodies in grey matter to trauma and ischaemia.52 54 57 However, criticism can still be levelled at these studies, such as the fact that rats were of significantly different weights in certain experiments,106 108 cervical as well as thoracic injuries were only sometimes employed,107 109 110 the mechanisms for inducing spinal injury were universally hugely variable, and the timings for cooling animals were only standardised in later studies.109 110 Clearly, these flaws in study design do attenuate these results to a degree, as does the simple fact that the studies relate to animals rather than directly to humans, and that they are controlled experiments with no relationship to the inherent variability of the clinical arena, but they nevertheless do still provide encouragement for continued research.
A summary of recent controlled experimental studies into the effects of hypothermia of traumatically induced spinal cord injury
More recent clinical trials, however, are still very much lacking and it is only in the last few years that attention surrounding this has been refocused. Many attribute this ‘pause’ in research to the sheer inadequacy of earlier clinical trials and subsequent development of other interventions for acute spinal cord injury, such as surgical decompression and stabilisation, and pharmacological therapies such as steroids.2 58 80 However, following a high-profile case in 2007, when an American football player was treated with therapeutic hypothermia following a ‘complete’ (AIS-A) cervical spinal cord injury, and who subsequently experienced rapid neurological improvement such that he regained sensory and partial motor function (AIS-D) within 1 year,114 a resurgence of interest has developed. Although this case has little stand-alone evidential value due to the outcome being confounded by concomitant use of decompressive surgery and steroid treatment,115 and due to it being a single, non-representative case involving a professional athlete, several clinical studies have since been instigated.116 117 These studies have, in turn, produced further encouraging results as to the efficacy of therapeutic hypothermia in acute spinal cord injury.
The first of these investigations, a controlled safety study that involved 28 age-matched and injury-matched patients with complete (AIS-A) cervical spinal cord injuries at baseline, 14 of whom received systemic cooling to 33°C for 48 hours, reported a statistically significant 42.8% conversion rate of at least one AIS grade in cooled patients versus a 33.7% conversion rate in the controls at follow-up.116 The study also importantly found that there was no difference in clinical complications between the two groups. These results were subsequently corroborated by a second identically designed study in 2013117. This study followed 70 patients and reported similar statistically significant results of a 43% improvement of at least one AIS grade in the 35 patients who underwent hypothermia versus a 35.5% improvement in the control group at follow-up, this study also finding no differences in clinical complications between groups. Due to both of these studies being non-randomised, relatively small, from the same institution, and ultimately the only studies of their kind in recent times, their usefulness is somewhat limited. However, the authors themselves declared that their purpose was principally to establish the safety and potential viability of an intervention such as this rather than anything else and thereby to hopefully stimulate further research in the format of sizeable, multicentre, randomised controlled trials.116 117
Having looked at the available evidence, it is clear that there is a need for further research, much of the available evidence being unsuitable for determining the true effectiveness of therapeutic hypothermia in acute spinal cord injury. Whereas more recent studies and trials have been encouraging, clearly demonstrating the potential for this intervention to affect this secondary injury process that exacerbates neural tissue damage and thereby overall neurological deficit following an initial insult to the spinal cord, their numbers and designs are still somewhat lacking. As such, with regard to future directions, this very much hangs in the balance. Through the carrying out of large, well-constructed randomised controlled trials, not only can therapeutic hypothermia’s effectiveness in acute spinal injury be more thoroughly investigated, but also the optimal design for its use can be established, including rates of cooling and rewarming, time windows for application and possible combinations with other already-established interventions such as pharmacological therapies.2 96 However, if future clinical trials are not initiated, then it is a real possibility that this neuroprotective treatment, which has already proven itself so beneficial within the settings of cardiac arrest and fetal encephalopathy, will remain an experimental therapy and never achieve clinical implementation.2 58 80
Conclusion
Having now considered therapeutic hypothermia in relation to acute traumatic spinal cord injury, it would seem, from a theoretical perspective, that this neuroprotective intervention has potential for application within this arena. This is principally due to its ability to seemingly attenuate the secondary injury process that plays such a significant role in exacerbating neural tissue damage in acute spinal cord injury, and thereby to lessen subsequent sensory, motor and autonomic dysfunction for an affected individual. Nevertheless, current evidence is desperately lacking, the conducting of large, multicentre, randomised controlled trials being ultimately needed in order for this intervention to move from being an experimental therapy to a clinical reality surrounding this devastating condition.
References
Footnotes
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.