Sanjay Dhall, MD

Acute Traumatic Spinal Cord Injury and Antiquated Viewpoints: An Honest Assessment of an Organized Neurosurgery Position Statement

Neurological recovery following acute spinal cord injury (SCI) is often viewed as limited, and long-term neurological sequelae of SCI represent a leading cause of chronic disability worldwide. In addition to the catastrophic disability that often results, healthcare for patients with chronic SCI has been reported to cost as much as $1.6 billion just in the United States. As such, much effort has been focused on improving existing interventions such as medical and surgical management, as well as identifying new therapeutics such as devices or drugs to treat SCI. Currently, there are no FDA-approved treatments for SCI, with surgical decompression and medical management representing the mainstay of acute SCI care.

In recent years, a number of studies have attempted to study the relationship between the timing of surgical decompression and neurologic recovery after SCI. While some studies in the late 1990s defined “early surgery” as within 72 hours of injury, most studies of this millennium have used 24 hours as the definition. In the past few years, a handful of studies have attempted to study “ultra-early” surgery for SCI occurring within 12 hours of injury.

Recently, AO Spine published a clinical practice guideline on a variety of topics, including the timing of surgery for acute SCI, recommending early surgery (within 24 hours of injury) as an option.

In response, the American Association of Neurological Surgeons (AANS), Congress of Neurological Surgeons (CNS), the AANS/CNS Joint Section on Disorders of the Spine and Peripheral Nerves (DSPN), and the AANS/CNS Joint Section on Neurotrauma and Critical Care published a position statement. In this position statement, these leaders of organized neurosurgery concluded that the “ability to draw conclusions regarding the timing of surgical intervention that will have a meaningful impact on the patient’s neurological outcome” is limited by deficiencies in the current evidence. While the above was not an inaccurate statement, organized neurosurgery took it a step further and declared that “there is no prospective class I evidence that supports a recommendation for the timing of surgical intervention”.

It is certainly true that the current body of literature on surgical timing in SCI covers a heterogeneous patient population, uses variable outcome measures, and perhaps even inconsistent treatment protocols, as suggested by our national neurosurgery leadership. But do these “deficiencies” really limit our ability to draw any conclusions about the timing of surgery and neurologic recovery as they suggest?

The Published Data:

The STASCIS trial was a prospective multicenter observational study that compared neurologic outcomes, complications, and mortality between early (<24 hours) and late (>24 hours) surgery for acute SCI. The authors reported that patients in the early arm were 2.8 times more likely to improve by 2 or more AIS grades than the late group. While a substantial and impactful study, this trial did not collect or report within the 24-hour group.

Specifically, the question of whether surgery within less than 24 hours such as 8, 12, or 24 hours would have an impact on neurologic recovery remained unanswered. In 2015, Jug and colleagues retrospectively compared neurologic recovery after SCI in surgery within 8 hours versus within 24 hours. They concluded that the odds for at least a 2 AIS grade improvement were 106% higher in the 8-hour group as compared to the 24-hour group. While compelling, this study had a significant limitation in that the entire 24-hour group was comprised of patients transferred in from outside institutions where the initial hemodynamic management was unclear and may have contributed to these results. Similarly, Grassner et al. studied neurologic outcomes at one year comparing surgery within 8 hours versus 24 hours for acute SCI. They also concluded that the early group (8 hours) was better in multiple ways, not only in neurologic recovery but also in terms of independence as assessed by the spinal cord independence measure (SCIM).

In 2018, Burke et al. published a retrospective comparison of surgery performed ultra-early (<12 hours) as compared to early (12-24 hours) and late (>24 hours) for acute SCI. Unlike previous studies, all of these patients were treated in the same institution using the same protocols. The three groups were not significantly different in terms of age, injury severity score (ISS), and admission ASIA impairment scale (AIS). They reported that the ultra-early group experienced significantly better neurologic improvement in terms of AIS grade (1.3 versus 0.5 in the early group). Similarly, they demonstrated that 88.9% of the AIS A patients in the ultra-early group converted to a better grade as compared to 30.8% of the early/late group AIS A patients.

Subsequently, Aarabi et al. published a study in 2020 with conflicting results on ultra-early surgery for SCI. While they concluded that ultra-early surgery was not associated with improved neurologic recovery, there were significant confounders to be considered. The distribution of motor incomplete (AIS C) patients was, by the authors’ statistics, significantly different between the groups. Specifically, only 5 of the 32 ultra-early patients (15.6%) were motor incomplete as compared to 17 of the 40 later patients (42.5%). It is well established in the SCI literature that motor complete SCI is associated with far lower rates of neurologic recovery than incomplete.

An Inconvenient Truth?

The data above is far from ideal, as is nearly all of the spine and neurosurgery literature. However, the preponderance of evidence suggests that SCI patients who are surgically decompressed earlier coincidentally have a greater degree of neurologic recovery. Undoubtedly, this represents a major paradigm shift in the care of SCI.

Whilst neurosurgeons are accustomed to treating high intracranial pressure secondary to intracranial hematomas as a surgical emergency, this has not been the historical precedent in the treatment of acute traumatic spinal cord injury. These surgical procedures are frequently longer and more complex than trauma craniotomies and require more support and infrastructure that require investment from healthcare systems. This paradigm shift toward urgent or perhaps even emergent surgery for SCI will undoubtedly result in more pressure towards us to act quicker. However, this inconvenience is superseded by the clinical benefit to our patients, which is clear even with the current body of literature. Larger trials will undoubtedly be done, but in the interim, if there is even a suggestion of quicker action saving neurological function, we should change our current practice.

Final Thoughts:

The position statement from our national neurosurgical leaders pointed to a lack of level 1 evidence to support early surgery for SCI. Level 1 evidence is defined typically as at least one, if not more, prospective randomized trials. Nearly every single procedure offered to neurosurgical patients is not supported by Level 1 evidence. This application of this highest standard would, if applied universally, suggest that we as neurosurgeons stop the most basic and common neurosurgical operations due to lack of Level 1 evidence.

As the largest state neurosurgical organization, we should encourage our national parent organizations, CNS, AANS, and the DSPN, to amend this erroneous and anachronistic position statement to better reflect what current evidence suggests.