New Clinical-Pathological Classification of Intraspinal Injury Following Traumatic Acute Complete Thoracic Spinal Cord Injury

AIS: American Spinal Injury Association Impairment Scale MRI: magnetic resonance imaging PLGA-PLL: Poly-lactic-co-glycolic acid-Poly-L-lysine SCI: spinal cord injury Traumatic spinal cord injury (SCI) can produce a sudden and devastating impact on quality of life due to severe motor, sensory, and autonomic dysfunction, including bowel, bladder, and sexual impairment. The adverse consequences of traumatic SCI result from a well-described pathological process involving primary and secondary mechanisms.1,2 The primary injury follows mechanical disruption of spinal cord tissue due to blunt or penetrating trauma. The initial local deformation and energy transfer results in cellular damage and death. A secondary phase of injury, involving inflammation, proapoptotic signaling, edema, reduced blood flow, and local tissue ischemia, results in further cellular necrosis, often culminating in the formation of large, fluid-filled cystic cavities.3-5 While a common pathological process underlies most traumatic SCIs, each individual SCI is unique. The anatomic location of the injury determines which segmental myotomes and sensory fields are impaired. In addition, nonclinical studies indicate that different spinal column injury patterns (ie, fracture dislocations vs burst fractures) result in different mechanical insults to the spinal cord, with corresponding differences in histopathology and behavioral outcome.6 Postmortem examination following autopsy and tissue fixation after various periods of survival with SCI (ranging from less than 24 h to many years) has also revealed heterogeneity in the pathology of human SCI;1,3,7,8 however, opportunities to observe spinal cord parenchyma following acute SCI, especially closed injuries, are rare. A more thorough understanding of the heterogeneity of pathology following human acute SCI may contribute to the development of future therapeutic interventions. The Neuro-Spinal Scaffold (InVivo Therapeutics, Cambridge, Massachusetts) is an investigational device that is intended to be implanted at the epicenter of an acute spinal cord contusion injury during open spine surgery.9 It is intended to act as a physical substrate for cell growth, appositional healing, and tissue remodeling, and to preserve the structural integrity of the cord. The Neuro-Spinal Scaffold is composed of the polymer Poly-lactic-co-glycolic acid-Poly-L-lysine (PLGA-PLL), which is a block copolymer of PLGA and PLL. PLGA is biodegradable and biocompatible. PLL contains lysine functional groups that promote cellular adhesion by creating a positively charged material substrate.10-12 The Neuro-Spinal Scaffold, formulated with the copolymer PLGA-PLL, is highly porous, biocompatible, biodegradable, and contains positively charged functional groups that promote cellular adhesion (Figure 1). In nonclinical models, the Neuro-Spinal Scaffold, surgically implanted at the epicenter of the wound after an acute spinal cord contusion injury, acts by appositional healing to spare spinal cord tissue, decrease post-traumatic cyst formation, decrease spinal cord tissue pressure, and facilitate neural regeneration.13 InVivo Therapeutics is currently enrolling patients with acute, complete thoracic SCI into the INSPIRE study (InVivo Study of Probable Benefit of the Neuro-Spinal Scaffold™ for Safety and Neurologic Recovery in Subjects with Complete Thoracic AIS A Spinal Cord Injury; ClinicalTrials.gov Identifier NCT02138110), designed to evaluate the safety and effectiveness of the Neuro-Spinal Scaffold for the treatment of complete T2-T12/L1 SCI. Implantation of the Neuro-Spinal Scaffold requires either durotomy and intra-arachnoidal dissection only, or in combination with myelotomy, depending on the presence or absence of cord laceration. The Neuro-Spinal Scaffold implantation procedure performed in the INSPIRE study provided a unique opportunity to directly observe pial and intramedullary spinal cord pathology in patients with acute traumatic thoracic American Spinal Injury Association Impairment Scale (AIS) grade A SCI within 96 h of injury, and to correlate the visible microsurgical cord pathology with neurological outcome following Scaffold implantation.FIGURE 1: The porous, cylindrical Neuro-Spinal Scaffold (InVivo Therapeutics) comes in 2 sizes, 2 mm diameter by 10 mm length and 3 mm diameter by 10 mm length. The Scaffold is designed for optimal fit in the intraspinal postirrigation cavity and can be trimmed if necessary to a specific length. The Neuro-Spinal Scaffold is expected to be resorbed from the site of implant within 4 to 8 wk. A, Macroscopic view of a representative 3 × 10 mm2 Scaffold. B, SEM cross-sectional image of a representative 2 × 10 mm2 Scaffold. Scale bar = 500 μm.METHODS The INSPIRE study is an open-label, nonrandomized, single-arm, multicenter Humanitarian Device Exemption Probable Benefit Study to evaluate the safety and probable benefit of the Neuro-Spinal Scaffold (InVivo Therapeutics) in up to 20 subjects. The INSPIRE study was approved by the US Food and Drug Administration (ClinicalTrials.gov Identifier NCT02138110) and by the Institutional Review Boards of the collaborating hospitals and medical centers. The primary objective was to evaluate whether the Neuro-Spinal Scaffold is safe and feasible for the treatment of complete SCI, as determined by no degradation in neurological level of injury, or sensory or motor function beyond that typically seen in patients with thoracic AIS A injuries. A secondary objective was to evaluate the clinical effectiveness of the Neuro-Spinal Scaffold, including improvement in AIS grade, sensory scores, motor scores, bladder and bowel function, Spinal Cord Independence Measure III, and Quality of Life Index (QLI-SCI III), as well as possibly decreased pain. Inclusion criteria for the study were neurological level of injury T2 to T12/L1, AIS A classification, 16 to 70 yr of age (inclusive), implantation of the Neuro-Spinal Scaffold occurring within 96 h from injury, nonpenetrating SCI that is no less than approximately 4 mm in diameter by magnetic resonance imaging (MRI), and informed consent obtained. Exclusion criteria included terminal illness, significant traumatic brain injury, coma, unreliable ISNCSCI, penetrating SCI, radiographic or visual evidence of parenchymal dissociation or anatomic transection where the injury completely disrupts a full cross-section of the spinal cord, requirement for long-term ongoing mechanical ventilation, clinically significant pre-existing neurological comorbidities, respiratory disease, or infection, documented immune deficiency disorders, recent significant substance abuse, or severe mental illness. At the time of this interim report (September, 2016), 10 patients with nonpenetrating complete SCI between T3-T11 had been enrolled. Following screening MRI and spine stabilization, durotomy and placement of dural tack-up sutures permitted inspection of the posterior medullary surface. In some cases, myelotomy was performed prior to Scaffold implantation. Images of the SCI epicenter were captured through a surgical microscope prior to implantation. RESULTS Neurological outcome was assessed in 8 patients (aged 18-55 inclusive, 6 males and 2 females) implanted with Scaffolds (InVivo Therapeutics) 9 to 83 h after injury (Table). Injuries were classified as “contusion-type” (5 patients; Figure 2) with an intact pial surface and peripheral white matter, or “compound-type” (3 patients; Figure 3) with elements of contusion, pial disruption, laceration, maceration, and damaged peripheral white matter but preserved cord continuity. Following dorsal myelotomy of contusion-type injuries, gentle irrigation of loose necrotic/hemorrhagic debris revealed an intramedullary cavity. Scaffolds implanted into the intramedullary cavity of a contusion-type injury were generally no longer visible due to overarching tissue but could be identified using ultrasound. Scaffolds implanted into compound-type injuries remained visible after implantation. Of patients classified as contusion-type injuries, 1 converted to AIS C by 1 mo, 1 converted to AIS B by 1 mo, and 1 converted to AIS B by 3 mo. Of the patients with compound-type injury, 1 converted to AIS B at 2 mo and 1 converted to AIS B at 6 mo (Table). Two additional patients died from causes deemed unrelated to Scaffold implantation.FIGURE 2: “Contusion-type” injuries (5 patients) displayed intact pial surface and peripheral white matter. Following dorsal myelotomy, gentle irrigation of loose necrotic debris revealed an intramedullary cavity.FIGURE 3: “Compound-type” injuries (3 patients) displayed elements of contusion, pial-disruption, laceration, maceration, and damaged peripheral white matter but preserved cord continuity.FIGURE 4: Following durotomy, 2 types of cord damage were observed. “Contusion-type” injuries (5 patients) displayed intact pial surface and peripheral white matter. “Compound-type” injuries (3 patients) displayed elements of contusion, pial disruption, laceration, maceration, and damaged peripheral white matter but preserved cord continuity.TABLE: The INSPIRE Study—Injury Type is not Predictive of AIS ConversionDISCUSSION Prior anatomic classifications of the pathology of human SCI have been performed following autopsy and tissue fixation.3,7 However, the microsurgical observations provided here are the first to attempt correlation of visual evidence of parenchymal injury with subsequent neurological outcome. We now report visual evidence of either “contusion injury” with central necrosis and peripheral tissue sparing, or “compound injury” with elements of contusion, focal laceration, and pial disruption, in patients within 96 h of severe (AIS A) traumatic acute thoracic SCI. Previously, Bunge et al3 classified human SCI pathology into 4 types. Contusion/cyst injury involves contusive injury leading to central hematomyelia that resolves over the course of weeks to a fluid-filled cyst surrounded by a surviving rim of peripheral white matter and pia. Cord maceration injury involves massive cord compression and distortion leading to loss of tissue topography. However, in some cases of maceration injury, fragments of spinal cord parenchyma maintained continuity across the damaged region. Cord laceration injury results from penetrating injuries such as gunshot wounds. Finally, in solid cord injury, the overall topology of the cord is retained, and the damage is largely limited to white matter tracts, with similarity to pathology observed following traumatic brain injury. While our observations largely align with those of Bunge et al,3 we did not observe 2 of the above injury types. The INSPIRE study includes only nonpenetrating SCIs, and excludes patients with complete parenchymal dissociation or anatomic transection. Accordingly, “cord laceration injuries” were not observed in the INSPIRE study. Similarly, Bunge et al3 observed “solid cord injuries” in patients with functionally incomplete injuries, which were excluded from the INSPIRE study. INSPIRE study patients with contusion-type injuries appear to correspond to contusion/cyst injury observed by Bunge et al.3 This type of injury is similar to the contusion animal model.14 INSPIRE study patients with compound-type injuries appear to correspond to the cord maceration injury type observed by Bunge et al.3 This type of injury has some similarity to the hemicordectomy animal model,15,16 but appears to combine elements of both contusion and maceration. For example, a compound-type injury may result from both contusion-induced grey matter injury, and focal maceration or internal (subdural) laceration of white matter from displaced bone. The term “compound” serves as an abbreviation for “compound contusion/maceration/laceration” type injury. An important distinction between contusion-type and compound-type injuries observed in the INSPIRE study is the status of the pia mater. In a contusion-type injury, the parenchyma is disrupted but the pia is intact. In contrast, in a compound-type injury, parenchymal disruption is accompanied by traumatically opening of the pia, similar to a compound bony fracture implies complication due to open skin. Spinal cords with contusion-type injuries required durotomy followed by myelotomy for implantation of the Neuro-Spinal Scaffold (InVivo Therapeutics). Myelotomy is a well-accepted procedure for the surgical resection of intramedullary spinal cord tumors.17 Moreover, beginning with Allen in 1914,18 numerous clincal cases of durotomies and myelotomies after acute SCI have been described in the US, Europe, Japan, and China.19-23 These authors typically report the evacuation of necrotic tissue, and hypothesize that the procedure is a potential means to reduce acute post-traumatic edema. Similarly, we observed that durotomy and dorsal myelotomy within 96 h in contusion-type injuries allowed gentle irrigation of loose necrotic debris to reveal an intramedullary cavity. The Neuro-Spinal Scaffold, once implanted into this cavity, was generally not visible due to encapsulation by surrounding white matter, but could be identified using ultrasound. In contrast, spinal cords presenting with compound-type injuries following durotomy did not require myelotomy for implantation of the Neuro-Spinal Scaffold. In these cases, the Scaffold was placed in apposition to the surviving fragments of the cord that crossed the damaged region, and continued to be visible after implantation. Both injuries were successfully implanted with the Scaffold, and AIS improvements were seen in both pathological types. Subsequently, 5 out of the 8 patients (63%) have converted from complete paraplegia (AIS A) to partial paralysis (AIS B or C) by 6 mo. Large, multinational, natural history databases generally indicate that only 15% to 16% of patients with complete (AIS A) thoracic injury will convert to an improved AIS grade within 6 mo after injury.24,25 However, neither injury type nor time to implant appears to be predictive of AIS conversion. CONCLUSION The surgical procedure for implantation of the Neuro-Spinal Scaffold (InVivo Therapeutics) provides a unique opportunity for direct, visual inspection of the acutely injured spinal cord revealing some attributes not clearly visible on initial MRI. Following durotomy, 2 forms of cord damage were observed. In contusion-type injuries, the cord surface is intact. In compound-type injuries, the pia is breached and there is visible but incomplete cord parenchymal separation (Figure 4). Both injuries were successfully implanted with the Neuro-Spinal Scaffold and AIS improvements were seen in both pathological types. Refinement of these observations will continue with future patient enrollment and could provide the basis of a novel clinical-pathological classification system. Disclosures This research was financially supported by InVivo Therapeutics. RT Layer, TR Ulich, KM Neff, and LK Masuoka are employees of InVivo Therapeutics.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.