Extent of Spinal Cord Decompression in Motor Complete (American Spinal Injury Association Impairment Scale Grades A and B) Traumatic Spinal Cord Injury Patients: Post-Operative Magnetic Resonance Imaging Analysis of Standard Operative Approaches

Abstract

Although decompressive surgery following traumatic spinal cord injury (TSCI) is recommended, adequate surgical decompression is rarely verified via imaging. We utilized magnetic resonance imaging (MRI) to analyze the rate of spinal cord decompression after surgery. Pre-operative (within 8 h of injury) and post-operative (within 48 h of injury) MRI images of 184 motor complete patients (American Spinal Injury Association Impairment Scale [AIS] grade A = 119, AIS grade B = 65) were reviewed to verify spinal cord decompression. Decompression was defined as the presence of a patent subarachnoid space around a swollen spinal cord. Of the 184 patients, 100 (54.3%) underwent anterior cervical discectomy and fusion (ACDF), and 53 of them also underwent laminectomy. Of the 184 patients, 55 (29.9%) underwent anterior cervical corpectomy and fusion (ACCF), with (26 patients) or without (29 patients) laminectomy. Twenty-nine patients (16%) underwent stand-alone laminectomy. Decompression was verified in 121 patients (66%). The rates of decompression in patients who underwent ACDF and ACCF without laminectomy were 46.8% and 58.6%, respectively. Among these patients, performing a laminectomy increased the rate of decompression (72% and 73.1% of patients, respectively). Twenty-five of 29 (86.2%) patients who underwent a stand-alone laminectomy were found to be successfully decompressed. The rates of decompression among patients who underwent laminectomy at one, two, three, four, or five levels were 58.3%, 68%, 78%, 80%, and 100%, respectively (p < 0.001). In multi-variate logistic regression analysis, only laminectomy was significantly associated with successful decompression (odds ratio 4.85; 95% confidence interval 2.2-10.6; p < 0.001). In motor complete TSCI patients, performing a laminectomy significantly increased the rate of successful spinal cord decompression, independent of whether anterior surgery was performed.

Keywords: ASIA Impairment Scale; MRI; decompression; spinal cord injury; trauma.

FIG. 1.

Flow diagram indicating the selection criteria for 184 American Spinal Injury Association Impairment Scale (AIS) grades A and B subaxial cervical spine traumatic spinal cord injury patients who had decompressive surgery following trauma.

FIG. 2.

Midsagittal plane of CT scan from subaxial cervical spine indicating methodology of measurements of sagittal diameter at three segmental levels and the subaxial cervical spine height from C2-T1.

FIG. 3.

Midsagittal subaxial CT and MRI cuts indicating morphology according to AOSpine Subaxial Cervical Spine Classification System. C, translation rotation and highly unstable injuries; A4, compression burst fractures; B2, flexion distraction without translation; B3, extension distraction without translation; A0, no imaging evidence of fracture dislocation.

FIG. 4.

Pre-operative (A) and post-operative (B) midsagittal magnetic resonance images indicating measured intramedullary lesion length a and b.

FIG. 5.

Midsagittal CT scan of cervical spine indicating (A) anterior cervical discectomy and fusion (ACDF); (B) anterior cervical corpectomy and fusion (ACCF); (C) Combined ACDF and three level laminectomy and posterior spinal fusion (PSF); (D) ACCF (two levels) and Laminectomy (two levels) and PSF; and (E) three levels of laminectomy and PSF.

FIG. 6.

Graph indicating progressively increased chances of success in decompression as the levels of laminectomy increase with surgical technique.

FIG. 7.

Midsagittal CT and MRI images of two subaxial cervical spine injuries without (upper rows) and with (lower rows) success in decompression. In the upper row a translation rotation injury (A and B, arrowhead and arrow, respectively) was managed with a single-level ACDF, while the lower row indicates another translation rotation and compression fracture (A and B) managed with a single-level of anterior cervical discectomy and fusion (C) and successful decompression (arrow).

FIG. 8.

Midsagittal CT and MRI of one level ACCF with and without success in decompression. (A-D) indicate a teardrop fracture (A and B arrows) indicating compressed spinal cord at C4 and beyond. (C) and (D) show one level corpectomy with no evidence of decompression (D). (E) and (F) indicate a C5 tear drop fracture managed with corpectomy and successful decompression of spinal cord (G and H arrows).

FIG. 9.

Midsagittal CT and MRI images belonging to two different patients managed by anterior cervical discectomy and fusion (ACDF) and laminectomy. Upper row indicates a C6/C7 translation rotation injury (A and B) with evidence of spinal cord compression. (C) and (D) show one level ACDF and two levels of laminectomy with inadequate decompression of the spinal cord at C4 and C5 and C7. (E-H) indicate a flexion compression injury and evidence of spinal cord compression at C5 and C6 segments. The patient had ACDF at C5/6 and C6/7 with three levels of laminectomy and posterior spinal fusion (G and H) with adequate decompression of the spinal cord (arrows).

FIG. 10.

Midsagittal CT and MRI images indicating wo vertical compression and teardrop fractures managed by corpectomy and laminectomy. Upper row plates (A-D) indicate C5 corpectomy and three-level laminectomies and posterior spinal fusion (PSF) with inadequate decompression at C3 and C7. Lower row (E-H) plates showing C5 corpectomy and three-level laminectomy and PSF with full decompression of the spinal cord.

FIG. 11.

Midsagittal views of two motor complete traumatic spinal cord injury patients with translation rotation (A-D) and no evidence of subaxial cervical spine fracture dislocations (E-H). Inadequate laminectomy and posterior spinal fusion (PSF) in the first patient did not decompress the cord adequately (at C7, plate D). Images in the lower row indicate that 4 level laminectomies and PSF completely decompressed the spinal cord.

FIG. 12.

Pre-clinical rodent model of experimental traumatic spinal cord injury indicating an ischemic necrotic spinal cord injury with rostral and caudal expansion over time. From Balentine, with permission.

FIG. 13.

Post-operative axial (A) and midsagittal (B) MRI images of a motor complete traumatic spinal cord injury indicating severe spinal cord swelling with no visualization of subarachnoid space in the region of laminectomy from C3-C6 (arrows).