Hypothermia and Human Spinal Cord Injury: Updated Position Statement and Evidence Based Recommendations from the AANS/CNS Joint Section on Disorders of the Spine Peripheral Nerves
John E. O'Toole, Marjorie C. Wang, and Michael G. Kaiser
Grade I - There is insufficient evidence to recommend for or against the practice of either local or systemic therapeutic hypothermia as a treatment for acute spinal cord injury.
Grade C - There is level IV evidence based on one retrospective comparative cohort study and one prospective cohort study to suggest that systemic modest hypothermia might be applied safely to this population.
Both local and systemic hypothermia have been of interest for decades as potential therapies for acute spinal cord injury (SCI). In 2007 the American Association of Neurological Surgeons (AANS)/Congress of Neurological Surgeons (CNS) Joint Section on Disorders of the Spine and Peripheral Nerves and Joint Section on Trauma released a position statement and evidence-based review on hypothermia after SCI. In that review, Resnick et al found a lack of evidence to either support or refute the use of local or systemic hypothermia for acute SCI in humans. The reviewers advocated for controlled clinical trials investigating the safety and efficacy of this intervention prior to its adoption in clinical practice. In an effort to keep the position statement current, an ad hoc committee was formed to generate an updated evidence-based recommendation founded upon a review of the literature from the intervening time period since the 2007 statement.
A computerized search of the National Library of Medicine database was performed using PubMed with the search terms "hypothermia AND spinal cord injury". The search was limited to the years 2005 to present since the prior review covered 1965-2005. One hundred and thirty-one references were obtained. The titles and abstracts of these references were then reviewed, and all publications not pertaining to the clinical use of hypothermia after acute SCI in humans were eliminated including laboratory, preclinical, and vascular surgical reports. Only papers published in English were included. Case series and case reports as well as systematic reviews/meta-analyses were included, but general review papers were excluded. The bibliographies of selected papers were also reviewed for additional references. This yielded four publications of relevance.
The four new publications included one case report, one retrospective feasibility case series, one retrospective comparative case series and one pooled retrospective and prospective case series. The details and critique of the evidence can be found in the attached evidentiary table (Table 3). Briefly, the case report from Cappuccino et al described the treatment of a professional football player who sustained a blunt cervical SCI (ASIA A) during play that was treated with systemic hypothermia one day after undergoing anterior-posterior decompression and fusion for C3-4 dislocation. He eventually recovered to ASIA D by four months postoperatively, and the authors felt the degree of recovery was more than would be expected in the absence of hypothermia. Unfortunately, this single case example (level IV evidence) provides inadequate evidence to judge the safety or efficacy of hypothermia in this clinical situation. The remaining three studies were all published from the same institution and all included the same retrospectively reviewed cohort of 14 patients with complete (ASIA A) acute cervical SCI treated with operative decompression and stabilization followed by 48 hours of modest (32-34ÂºC) systemic hypothermia via an intravascular cooling catheter. The first report from Levi et al in 2009 was a technical feasibility and early safety study that provides level IV evidence that the authors' method of hypothermia was reproducible and that systemic hypothermia can be used safely in acute SCI patients. The second report from Levi et al in 2010 examined this same cohort of patients but compared them to a similar group of SCI patients who did not undergo systemic hypothermia in an attempt to establish baseline safety for this intervention. The authors found no statistically significant difference in complications between the groups except for an increased incidence of pleural effusions and anemia in the hypothermia group. The authors concluded that systemic hypothermia for acute SCI is safe and that phase 2 and 3 trials are feasible. This study suffers from limitations, outlined in the evidentiary table that downgraded its level of evidence to IV. It therefore provides low-level evidence that hypothermia may be applied safely to acute spinal cord injury patients. The final report from this group, Dididze et al in 2013 , presented a pooled analysis of the previously reported retrospective cohort of 14 patients with an additional prospectively treated cohort of 21 patients all undergoing systemic hypothermia in which they investigated clinical outcomes and complications. Comparison of pre- and post-treatment ASIA scores at 12 months revealed that 43% of patients improved at least 1 ASIA grade at follow-up (35% when excluding 4 patients that spontaneously improved in first 24 hours). Most common complications were pulmonary, as seen previously. Overall, 14% had venous thromboembolic events (VTE) (24% in prospective group, none in the smaller retrospective cohort). The authors conclude that systemic endovascular hypothermia for cervical acute SCI is safe and results in higher rates of neurological improvement than seen in previously reported population studies on SCI. As with the prior publications, the absence of a true control group precludes the formulation of definitive inferences on the actual safety or efficacy of systemic hypothermia for acute cervical SCI. This study provides low-level (level IV) evidence for the safety of modest systemic hypothermia in this patient population. Conclusions: Scientific studies have documented a potential benefit of systemic hypothermia in animal models of acute spinal cord injury; however there remains a paucity of clinical evidence to recommend for or against the practice of either local or systemic hypothermia for acute SCI in humans. The level IV evidence suggesting the safety of modest systemic hypothermia is promising, but controlled, comparative clinical studies investigating safety and efficacy must be performed prior to the introduction of hypothermia in the routine clinical care of patients with acute SCI.
Directions for Future Research
Further research is essential to determine if the preclinical promise of systemic hypothermia for acute spinal cord injury can be realized in humans. If prospective randomized controlled trials prove too challenging to conduct in this patient population, prospective comparative cohort studies (ideally at multiple centers) should be conducted to define the effectiveness and safety of this intervention.
Download pdf including evidentiary summary