Monitoring Spinal Cord Hemodynamics with Near-Infrared Spectroscopy After Acute Spinal Cord Injury

The spinal cord injury (SCI) community is justifiably excited about the promising neuro-protective and neuroregenerative treatments emerging from research laboratories around the world. In the efforts to improve outcome after SCI, there is a clear need to establish novel treatment approaches that can tackle the many complex biological problems that hinder recovery after this devastating injury. However, there is also a compelling (albeit arguably less 'novel and innovative') need to optimize the management approaches to SCI that are currently implemented in clinical practice, as these clinical practices can impact the neurologic outcome of individuals who suffer an SCI today. Our initiative is focused on such a clinical practice: the hemodynamic management of acute SCI.

Because trauma to the spinal cord disrupts its blood supply and leaves it vulnerable to further ischemic damage, current practice guidelines recommend that clinicians try to improve blood delivery to the cord by augmenting the mean arterial blood pressure (MAP) to 85-90 mmHg for the first 7 days after injury. Aside from urgent surgical decompression, this hemodynamic management approach is one of the only things that clinicians can do to potentially improve outcome after acute SCI. While most institutions (including ours) adhere to these guidelines, it is also apparent from clinical studies that the neurologic outcome of many acute SCI patients is not actually improved by such MAP augmentation. If the spinal cord blood supply is disrupted after injury and augmenting the MAP can improve this, why is it difficult to demonstrate a neurologic benefit with this approach?

One of the fundamental difficulties with this current hemodynamic management approach is that when clinicians augment the MAP, they have no way of knowing whether this is actually improving the oxygenation and blood supply to the injured cord. All they can measure is the general MAP and the arterial oxygen saturation of the individual and hope that by adjusting these to a certain level, the condition of the injured spinal cord will be improved. This inability to monitor what is happening in the spinal cord is a fundamental limitation to our current approach to hemodynamic management in acute SCI, and it leaves many basic questions unanswered. For a specific cord-injured patient, is a MAP of 85 mmHg actually improving the blood supply and oxygenation in the spinal cord? Should the MAP actually be higher? Or alternatively could it be lower? Is it worth administering higher and higher doses of vasopressor drugs to drive the MAP up? Or is there a threshold at which further vasopressors might drive the MAP up but have no appreciable physiologic benefit to the spinal cord (or be even potentially damaging by causing worse hemorrhage and edema)? These are all questions that cannot be addressed without a way of actually measuring what is occurring within the injured spinal cord.

Therefore, we have worked to develop a sensor to measure oxygenation and blood flow of the injured spinal cord. This sensor shines near-infrared light into the spinal cord, and based on the absorption of this light within the cord, it can provide a measure of blood and oxygen delivery to the tissue. This concept of using 'near-infrared spectroscopy' or 'NIRS' to non-invasively monitor blood and oxygen levels within the underlying tissue is already in widespread clinical use for hemodynamic monitoring of the brain; however, it has not been developed for the injured spinal cord. The NIRS sensor that we have developed is meant to be laid upon the membrane (called the 'dura') that encases the spinal cord and surrounding cerebrospinal fluid (CSF), obviating the need to insert a probe through the dura that might risk injuring the cord and causing a leakage of CSF. The NIRS sensor then emits near-infrared light through the injured spinal cord beneath it to measure oxygen and blood supply. We have evaluated the NIRS sensor in a large animal (pig) model of SCI, where we took advantage of the larger, more human-like size of the pig spinal cord to evaluate how well the sensor measured oxygen and blood supply of the injured spinal cord. By comparing the non-invasive NIRS measurements of oxygen and blood supply against similar measurements derived from invasive probes inserted directly into the spinal cord, we found that the NIRS sensor can accurately monitor these physiologic parameters within the cord.

With this promising data and our refined NIRS system, we now seek to evaluate this NIRS sensor in a small safety and feasibility study of acute SCI patients. Our plan is to recruit individuals with cervical or thoracic SCI who are to undergo posterior decompressive surgery, and then apply the NIRS sensor to the exposed spinal cord intra-operatively. We would evaluate these individuals for up to 7 days post-injury and assess the safety and feasibility of conducting such monitoring with the NIRS sensor. We will also assess how effective the NIRS sensor is at measuring changes in oxygen and blood flow by comparing these measures with continuous measurements of systemic MAP and arterial oxygen saturation over this 7-day period. With this, we will be able to determine if the NIRS sensor can provide meaningful physiologic information about the injured spinal cord that can then be utilized by a clinician to guide the hemodynamic management approach for that given patient. We believe such a tool is imperative for clinicians to have if they are to fully take advantage of the therapeutic opportunity that exists in the hemodynamic management of acute SCI.