BACKGROUND:
Decompression of the injured spinal cord confers neuroprotection. Compared with timing of surgery, verification of surgical decompression is understudied.
OBJECTIVE:
To compare the judgment of cervical spinal cord decompression using real-time intraoperative ultrasound (IOUS) following laminectomy with postoperative MRI and CT myelography.
METHODS:
Fifty-one patients were retrospectively reviewed. Completeness of decompression was evaluated by real-time IOUS and compared with postoperative MRI (47 cases) and CT myelography (4 cases).
RESULTS:
Five cases (9.8%) underwent additional laminectomy after initial IOUS evaluation to yield a final judgment of adequate decompression using IOUS in all 51 cases (100%). Postoperative MRI/CT myelography showed adequate decompression in 43 cases (84.31%). Six cases had insufficient bony decompression, of which 3 (50%) had cerebrospinal fluid effacement at >1 level. Two cases had severe circumferential intradural swelling despite adequate bony decompression. Between groups with and without adequate decompression on postoperative MRI/CT myelography, there were significant differences for American Spinal Injury Association motor score, American Spinal Injury Association Impairment Scale grade, AO Spine injury morphology, and intramedullary lesion length (IMLL). Multivariate analysis using stepwise variable selection and logistic regression showed that preoperative IMLL was the most significant predictor of inadequate decompression on postoperative imaging (P = .024).
CONCLUSION:
Patients with severe clinical injury and large IMLL were more likely to have inadequate decompression on postoperative MRI/CT myelography. IOUS can serve as a supplement to postoperative MRI/CT myelography for the assessment of spinal cord decompression. However, further investigation, additional surgeon experience, and anticipation of prolonged swelling after surgery are required.
KEY WORDS: Intraoperative ultrasound, MRI, Spinal cord decompression, Spinal cord injury
ABBREVIATIONS:
- ACDF
anterior cervical discectomy and fusion
- AMS
ASIA motor score
- ASIA
American Spinal Injury Association
- CST
closed skeletal traction
- DISH
diffuse idiopathic skeletal hyperostosis
- IMLL
intramedullary lesion length
- IOUS
intraoperative ultrasound
- LL
lower limit
- MVC
motor vehicle collision
- OPLL
ossified posterior longitudinal ligament
- UL
upper limit.
Spinal cord injury is one of the most formidable challenges of modern medicine. Although modern instrumentation techniques can provide long-term restitution and stability to the injured spinal column, no pharmaceuticals or cell-based therapies have been shown to definitively improve long-term outcomes.1 It remains to be seen whether promising efforts to regenerate neural tissue can enhance recovery.2-6 Therefore, spinal cord decompression remains the only validated means of neuroprotection at this time.
The transfer of kinetic injury to the spinal cord precipitates immediate injury to the parenchymal epicenter with loss of vascular and axonal integrity.7 This is followed by a cascade of secondary mechanisms.8-16 Persistent compression causes additional cytotoxic and vasogenic edema in a positive feedback loop.17,18 Spinal cord swelling extends at rates of 918 μm/hour in motor complete injuries and 21 μm/hour in American Spinal Injury Association (ASIA) Impairment Scale (AIS) grades C and D.19 Preclinical studies have shown that surgical decompression can mitigate secondary injury through restoration of blood flow and evoked potentials.20,21
Although much attention has been dedicated to timing of surgery,22 the technique and verification of surgical decompression itself are understudied. It is our impression that, for many years, a standard if unarticulated definition of decompression was the re-establishment of anatomic alignment with internal fixation. This may date to Crutchfield,23 who in 1954 wrote: “The simplest, safest, and most effective way to decompress and protect the spinal cord is to reestablish the normal alignment of the spine.” However, a recent study showed that only 121 of 184 motor complete patients (66%) were adequately decompressed on postoperative MRI, with laminectomy being the only significant predictor of adequate decompression.24
MRI provides excellent spatial resolution and, as such, has been used in the judgment of spinal cord decompression.24-26 Although MRI provides superior resolution and assessment, compared with real-time intraoperative ultrasound (IOUS) it has several drawbacks. MRI is time-consuming and expensive, lacks real-time evaluation, and is unavailable in many centers throughout the world. Computed tomography (CT) myelography provides excellent delineation of neural elements but is also time-consuming and carries risk of infection.
IOUS has been used in spinal cord injury since the early 1980s.27 Early IOUS pioneers28-40 described the appearance of the spinal cord and canal, identified cord motion, and reported its utility in traumatic and other spinal disorders. Mirvis and Geisler41 reported IOUS findings in 29 patients with cervical spinal cord injury. They graded lesions by echogenicity on a scale from 0 to 4 and reported a correlation with ASIA motor score. However, no study has compared adequacy of spinal cord decompression as judged by real-time IOUS with postoperative MRI and CT myelography. We hypothesized that assessment of spinal cord decompression by IOUS is comparable with these modalities.
METHODS
Design
A retrospective analysis of prospectively collected data was performed. This study was Institutional Review Board–approved with consent exemption.
Primary Objective
The primary objective of the study was to compare the judgment of cervical spinal cord decompression using real-time intraoperative ultrasound with postoperative MRI and CT myelography.
Inclusion/Exclusion
Two patients were excluded because of cerebrospinal fluid leak precluding IOUS visualization and postoperative hematoma.
Cohort
Fifty-one consecutive patients were included. Each patient was admitted to a Level I trauma center with cervical spinal cord injury between June 2021 and March 2022.
Admission Scenario
Patients underwent scene evaluation, resuscitation, and transfer on a rigid backboard with neck immobilized. Primary and secondary surveys were completed by an admitting trauma surgery team. CT was acquired, and neurosurgery was consulted. Using the International Standards for Neurological Classification of Spinal Cord Injury, a complete neurological assessment including ASIA motor score (AMS) and AIS grade was documented by either a senior neurosurgery resident or a neurotrauma nurse practitioner. Injury morphology was characterized using AO Spine subaxial cervical spine injury classification.42
MRI and CT Myelography
Acquisition of postoperative MRI/CT myelography is routinely performed at our institution. In this study, 47 patients underwent preoperative MRI, 3 underwent preoperative CT myelography, and 1 had neither MRI nor CT myelography before surgery. CT myelography was performed because of implanted devices. Four patients presented in a delayed fashion (ie, the time from injury to preoperative MRI/CT myelography is greater than 110 hours). The average and median time from injury to preoperative imaging of the remaining 47 patients was 15.55 and 10.48 hours, respectively. Preoperative and postoperative intramedullary lesion lengths (IMLLs) were determined by calculating mean measurements on T2-weighted sequences of 3 independent reviewers. The presence or absence of ossified posterior longitudinal ligament, diffuse idiopathic skeletal hyperostosis, fracture remote from the epicenter, and preinjury cervical fusion were recorded.
Closed Skeletal Traction
Attempted reduction with closed skeletal traction (CST) was performed before surgery in 6 patients. Gardner-Wells tongs, incremental weight, and Stryker frame (5 patients) or Jackson Table (Mizuho OSI) (1 patient) were used with C-arm fluoroscopy. CST was successful in 2 cases of unilateral locked facet. CST was unsuccessful in 2 cases of unilateral locked facet and one case of bilateral locked facets (reduction was completed by an open posterior approach). In one teardrop fracture, CST improved alignment before corpectomy.
Preoperative Planning and Surgical Technique
Based on preoperative imaging and clinical examination, a surgical plan was made by the neurosurgical team (comprising one of 7 attending neurosurgeons, 1 of 4 chief residents, and one of 4 senior residents). Each patient underwent laminectomy, including 35 with and 16 without posterior fusion. In general, one level of laminectomy was proposed for each 15-mm segment of T2 signal change. Eleven patients also underwent anterior surgery, including 8 anterior cervical diskectomy and fusion and 3 corpectomy. Anterior cervical diskectomy and fusion was performed after laminectomy in 4 patients. Laminectomy for each level was performed from lateral mass to lateral mass; the width of every laminectomy was measured at the midpoint of each skeletal segment, and mean values were calculated.
Judgement of Decompression by IOUS
After completing the planned number of laminectomy levels, the surgical field was filled with saline and the spinal cord and surrounding structures were visualized using IOUS. Each case used the Aloka Prosound Alpha 7 Ultrasound System (Hitachi Aloka Medical, Ltd) (cumulative range 4.40-13.33 MHz). Real-time images in both long and short axes were generated along the entire length of the laminectomy bed. Depth and gain were adjusted as needed by an assistant. Images were periodically frozen or printed for closer review. Decompression was judged as adequate when there was visualization of subarachnoid space both dorsal and ventral to the spinal cord along the entire laminectomy bed.43 If the sagittal profile of the cervical spine was felt to prevent dorsal migration of the cord at certain segments, dorsal subarachnoid space alone was considered sufficient. If IOUS demonstrated inadequate decompression after completing the preoperative laminectomy plan, further laminectomy was performed until the surgical team was satisfied with additional IOUS evaluation. No case left the operating room without the impression of adequate decompression using IOUS. The surgical team's impression was recorded before postoperative MRI or CT myelography.
Judgment of Decompression on Postoperative MRI and CT Myelography
We used postoperative MRI and CT myelography as the arbiter of decompression because they provide direct visualization of the spinal cord in relation to surrounding structures. We did not use a criterion of decompression based on the surgical technique of bony decompression alone. For postoperative MRI (47 patients) and CT myelography (4 patients), reviewers were instructed to judge decompression using both axial and sagittal T2 sequences on MRI or axial and sagittal CT myelography images as adequate when there was visualization of subarachnoid space around the entire ventral surface of the cord, around the entire dorsal surface of the cord, or both along the entire length of the laminectomy bed containing the injured spinal cord segment. All studies were independently judged by 5 reviewers, including 3 attending neurosurgeons, 1 chief resident, and 1 attending trauma radiologist for a total of 255 judgments. Consensus of adequate/inadequate decompression was established by at least 4/5 (80%) rater agreement. Cases with 3/5 (60%) agreement underwent adjudication including re-review of imaging and discussion to establish a final consensus.
Statistical Analysis
For judgment of decompression on postoperative MRI/CT myelography, percent agreement and Fleiss44' kappa were calculated. In the comparison of 2 groups, (1) patients with adequate decompression on both IOUS and postoperative MRI/CT myelography and (2) patients with adequate decompression on IOUS but inadequate decompression on postoperative MRI/CT myelography, univariate analysis was performed using the 2-tailed t-test for continuous variables and Fisher's exact test for categorical variables. For multivariate analysis, stepwise variable selection based on AIC with subsequent logistic regression (WALD test) was performed. A professional statistician performed basic and advanced statistical analysis using R version 4.1.0 (General Public License). Two other authors performed basic statistical analysis using Stata/SD 15.1 (StataCorp LLC).
RESULTS
IOUS demonstrated inadequate decompression after the preoperatively planned number of laminectomy levels in 5 cases (9.80%), prompting an additional half level of laminectomy in each, for a final judgement of adequate decompression in all 51 cases (100%).
Determination of Adequate Decompression Using Postoperative MRI and CT Myelography
For determination of decompression on postoperative MRI (47 cases) and CT myelography (4 cases), 40/51 cases had 5/5 (100.00%) agreement among the 5 reviewers, 7 cases had 4/5 (80.00%) agreement, and 4 cases had 3/5 (60.00%) agreement. The latter 4 cases were adjudicated. The percent agreement was 89.80%. Fleiss' Kappa was 0.61, consistent with good or substantial agreement. Four cases with postoperative CT myelography were judged to have adequate decompression. Final consensus was 43/51 (84.31%) with adequate decompression and 8/51 (15.69%) with inadequate decompression. A case with adequate decompression on both IOUS and postoperative MRI is shown in Figure 1.
FIGURE 1.
A 64-year-old male presented after a motor vehicle collision with AO Spine morphology type C (ASIA/AIS 28B). A, Preoperative MRI before closed skeletal traction, posterior fixation, and multilevel laminectomy. B1 and B2, Long-axis IOUS images at rostral and caudal ends of the laminectomy bed, respectively (inverted). B3 and B4, Short-axis IOUS images showing ventral and dorsal subarachnoid space within the middle of the laminectomy bed. C, Postoperative MRI demonstrating adequate decompression of the spinal cord. AIS, ASIA Impairment Scale; ASIA, American Spinal Injury Association; IOUS, intraoperative ultrasound; MVC, motor vehicle collision.
Demographic, Clinical, and Radiographic Variables
Mean demographic, clinical, and radiographic variables are presented in Table 1. Univariate analysis between groups with and without adequate decompression on postoperative MRI and CT myelography is also shown in Table 1. There were several statistically significant differences. The group with adequate decompression on postoperative imaging had higher admission AMS and AIS grade, lower percentage of AO Spine injury morphology type C, and smaller preoperative and postoperative IMLL.
TABLE 1.
Statistical Analysis of Demographics, Clinical Presentation, and Radiographic Parameters
| Decompressed on IOUS and postoperative MRI/CT myelography | Decompressed on IOUS but not postoperative MRI/CT myelography | Total | P value | |
|---|---|---|---|---|
| Age (mean/SD) | 59.44 (14.26) | 55.86 (21.58) | 58.88 (15.42) | .55 |
| Sex, No. (%) | ||||
| Male | 34 (79.07) | 6 (75.00) | 40 (78.43) | |
| Female | 9 (20.93) | 2 (25.00) | 11 (21.57) | 1.00 |
| Mechanism, No. (%) | ||||
| Fall, ground level | 20 (46.51) | 3 (37.50) | 23 (45.10) | |
| Fall, from height | 12 (27.91) | 2 (25.00) | 14 (27.45) | |
| MVC | 9 (20.93) | 1 (12.50) | 10 (19.61) | |
| Sport | 2 (4.65) | 2 (25.00) | 4 (7.84) | .29 |
| Admission AMS (mean/SD) | 50.47 (35.60) | 17.13 (22.50) | 45.24 (35.85) | .0142 |
| Admission AIS grade, No. (%) | ||||
| A | 7 (16.28) | 5 (62.50) | 12 (23.53) | |
| B | 5 (11.63) | 1 (12.50) | 6 (11.76) | |
| C | 9 (20.93) | 1 (12.50) | 10 (19.61) | |
| D | 22 (51.16) | 1 (12.50) | 23 (45.10) | .027 |
| Admission AIS grade, No. (%) | ||||
| A and B (motor complete) | 12 (27.91) | 6 (75.00) | 18 (35.29) | .017 |
| C and D (motor incomplete) | 31 (72.09) | 2 (25.00) | 33 (64.71) | |
| AO morphology, No. (%) | ||||
| A0 | 25 (58.14) | 3 (37.50) | 28 (54.90) | |
| A4 | 1 (2.33) | 0 (0.00) | 1 (1.96) | |
| B2 | 1 (2.33) | 0 (0.00) | 1 (1.96) | |
| B3 | 13 (30.23) | 2 (25.00) | 15 (29.41) | |
| C | 3 (6.97) | 3 (37.50) | 6 (11.77) | .20 |
| AO Spine injury morphology, No. (%) | ||||
| C | 3 (6.98) | 3 (37.50) | 6 (11.76) | .042 |
| All others (A0, A4, B2, and B3) | 40 (93.02) | 5 (62.50) | 45 (88.24) | |
| AO Spine injury modifier, No. (%) | ||||
| None | 32 (74.41) | 5 (62.50) | 37 (72.55) | |
| M2 | 3 (6.98) | 0 (0.00) | 3 (5.88) | |
| M3 | 5 (11.63) | 1 (12.50) | 6 (11.77) | |
| F4 | 3 (6.98) | 2 (25.00) | 5 (9.80) | .32 |
| IMLL (preoperative) (mean/SD) | 32.01 (19.75) | 56.00 (23.43) | 36.27 (22.19) | .0043 |
| IMLL (postoperative) (mean/SD) | 39.92 (29.65) | 71.93 (28.05) | 45.37 (31.53) | .0074 |
| OPLL, No. (%) | ||||
| Yes | 8 (18.60) | 0 (0.00) | 8 (15.69) | |
| No | 35 (81.40) | 8 (100.00) | 43 (84.31) | .33 |
| DISH, No. (%) | ||||
| Yes | 9 (20.93) | 2 (25.00) | 11 (21.57) | |
| No | 34 (79.07) | 6 (75.00) | 40 (78.43) | 1.00 |
| Remote fracture (from the epicenter), No. (%) | ||||
| Yes | 2 (4.65) | 1 (12.50) | 3 (5.88) | |
| No | 41 (95.35) | 7 (87.50) | 48 (94.12) | .41 |
| Preinjury cervical fusion, No. (%) | ||||
| Yes | 3 (6.98) | 0 (0.00) | 3 (5.88) | |
| No | 40 (93.02) | 8 (100.00) | 48 (94.12) | 1.00 |
AIS, ASIA Impairment Scale; AMS, ASIA motor score; CT, computed tomography; DISH, diffuse idiopathic skeletal hyperostosis; IMLL, intramedullary lesion length; IOUS, intraoperative ultrasound; MVC, motor vehicle collision; OPLL, ossified posterior longitudinal ligament.
Continuous variables were evaluated using the t-test (2-tailed), and categorical variables using Fisher's exact test.
Bold indicates statistical significance.
Variables of the Surgical Technique
Mean values for variables of the surgical technique are presented in Table 2. The mean and median number of laminectomy levels for all patients were 3.74 (SD 1.11) and 3.5, respectively. IOUS led to an additional half level of laminectomy in 5 cases (9.80%), of which 2 (40%) were not adequately decompressed on postoperative MRI. There were no statistically significant differences between groups for the mean number of laminectomy levels, laminectomy width, posterior fusion, rostral-most and caudal-most level of laminectomy, and anterior fusion before or after laminectomy.
TABLE 2.
Statistical Analysis of Variables of the Surgical Technique
| Decompressed on IOUS and postoperative MRI/CT myelography | Decompressed on IOUS but not postoperative MRI/CT myelography | Total | P value | |
|---|---|---|---|---|
| Levels of laminectomy (mean/SD) | 3.83 (1.07) | 3.28 (1.28) | 3.74 (1.11) | .21 |
| Average laminectomy width (all levels) (mm) (mean/SD) | 17.12 (1.68) | 17.26 (2.89) | 17.14 (1.88) | .85 |
| Minimum laminectomy width (mm) (mean/SD) | 15.05 (2.73) | 15.99 (4.69) | 15.20 (3.07) | .43 |
| Maximum laminectomy width (mm) (mean/SD) | 18.47 (2.03) | 18.20 (3.08) | 18.43 (2.19) | .74 |
| Rostral-most laminectomy width (mm) (mean/SD) | 16.97 (1.89) | 15.86 (3.99) | 16.80 (2.32) | .22 |
| Caudal-most laminectomy width (mm) (mean/SD) | 15.94 (2.07) | 16.58 (1.95) | 16.04 (2.05) | .43 |
| Posterior fusion with laminectomy, No. (%) | ||||
| Yes | 27 (62.79) | 7 (87.50) | 34 (66.67) | |
| No | 16 (37.21) | 1 (12.50) | 17 (33.33) | .17 |
| IOUS findings led to further laminectomy at the time of surgery, No. (%) | ||||
| Yes | 3 (6.98) | 2 (25.00) | 5 (9.80) | |
| No | 40 (93.02) | 6 (75.00) | 46 (90.20) | .17 |
| Rostral-most level of laminectomy (total and partial) | ||||
| C1 | 1 | 0 | 1 | |
| C2 | 5 | 2 | 7 | |
| C3 | 21 | 3 | 24 | |
| C4 | 9 | 2 | 11 | |
| C5 | 6 | 1 | 7 | |
| C6 | 1 | 0 | 1 | .87 |
| Rostral-most laminectomy was partial or total, No. (%) | ||||
| Partial | 6 (13.95) | 2 (25.00) | 8 (15.69) | |
| Total | 37 (86.05) | 6 (75.00) | 43 (84.31) | .56 |
| Caudal-most level of laminectomy (total and partial) | ||||
| C4 | 1 | 1 | 2 | |
| C5 | 6 | 2 | 8 | |
| C6 | 11 | 2 | 13 | |
| C7 | 20 | 3 | 23 | |
| T1 | 5 | 0 | 5 | .50 |
| Caudal-most laminectomy was partial or total, No. (%) | ||||
| Partial | 19 (44.19) | 3 (37.50) | 22 (43.14) | |
| Total | 24 (55.81) | 5 (62.50) | 29 (56.86) | 1.00 |
| Anterior fusion, No. (%) | ||||
| Yes | 7 (16.28) | 3 (37.50) | 10 (19.61) | |
| No | 36 (83.72) | 5 (62.50) | 41 (80.39) | .18 |
| Anterior fusion, No. (%) | ||||
| ACDF | 5 (11.63) | 2 (25.00) | 7 (13.73) | |
| Corpectomy | 2 (4.65) | 1 (12.5) | 3 (5.88) | |
| None | 36 (83.72) | 5 (62.50) | 41 (80.39) | .2 |
| Anterior fusion was after laminectomy, No. (%) | ||||
| Yes | 4 (9.30) | 0 (0.00) | 4 (7.84) | |
| No | 39 (90.70) | 8 (100.00) | 47 (92.16) | 1.00 |
ACDF, anterior cervical diskectomy and fusion; CT, computed tomography; IOUS, intraoperative ultrasound.
Continuous variables were evaluated using the t-test (2-tailed), and categorical variables were evaluated using Fisher's exact test.
Variables of Timing
There were no statistically significant differences between groups for variables of timing including time from injury to preoperative MRI/CT myelography, time from injury to surgery, and time between preoperative and postoperative MRI/CT myelography (Table 3).
TABLE 3.
Statistical Analysis of Timing of Injury, Imaging, and Surgery
| Decompressed on IOUS and postoperative MRI/CT myelography | Decompressed on IOUS but not postoperative MRI/CT myelography | Total | P value | |
|---|---|---|---|---|
| Time from injury to preoperative MRI/CT myelography, h, No. (%) | ||||
| Less than 8 | 16 (38.10) | 4 (50.00) | 20 (40.00) | |
| >8-24 | 16 (38.10) | 3 (37.50) | 19 (38.00) | |
| >24 | 10 (23.80) | 1 (12.50) | 11 (22.00) | .89 |
| Time from injury to surgery, h, No. (%) | ||||
| Less than 16 | 4 (9.30) | 2 (25.00) | 6 (11.76 | |
| >16-32 | 15 (34.89) | 4 (50.00) | 19 (37.26) | |
| >32 | 24 (55.81) | 2 (25.00) | 26 (50.98) | .17 |
| Time between preoperative and postoperative MRI/CT myelography (mean hours /SD) | 91.88 (224.72) | 109.31 (224.85) | 94.67 (222.53) | .84 |
| Time from surgery to postoperative MRI/CT myelography (mean hours /SD) | 27.34 (28.41) | 18.13 (8.50) | 25.90 (26.45) | .37 |
CT, computed tomography; IOUS, intraoperative ultrasound.
Continuous variables were evaluated using the student's t-test and categorical variables were evaluated with fisher's exact test.
Multivariate Analysis and Prediction Modeling
AMS, AIS grade, AO morphology, and IMLL were selected for multivariate analysis. For the primary outcome (adequate decompression on postoperative MRI/CT myelography), stepwise variable selection based on Akaike Information Criterion with subsequent logistic regression (Wald method) showed that preoperative IMLL was the most significant predictor of inadequate decompression (Table 4; P = .024, OR = 0.957). A graphical representation of this model shown in Figure 2 demonstrates that with increasing preoperative IMLL, the likelihood of adequate decompression on postoperative imaging decreases.
TABLE 4.
Stepwise Multivariate Analysis for AO Spine Injury Morphology (Type C vs Non–Type C) and Preoperative IMLL
| Coefficient | SD | Z-value | P value | Odds ratio | LL, 95% CI | UL, 95% CI | |
|---|---|---|---|---|---|---|---|
| Intercept | 3.746 | 1.017 | 3.498 | .000 | 42.368 | 5.192 | 345.763 |
| AO Spine morphology type C | −1.591 | 1.008 | −1.577 | .115 | 0.204 | 0.028 | 1.471 |
| Preoperative IMLL | −0.043 | 0.019 | −2.259 | .024 | 0.957 | 0.922 | 0.994 |
IMLL, intramedullary lesion length; LL, lower limit; UL, upper limit.
Bold indicates statistical significance.
FIGURE 2.
Graphical representation of the stepwise multivariate model for AO Spine subaxial injury morphology (type C vs non–type C) and preoperative IMLL demonstrates decreasing likelihood of adequate decompression on postoperative MRI/CT myelography despite adequate decompression on intraoperative ultrasound. The shaded areas represent confidence regions. CT, computed tomography; IMLL, intramedullary lesion length.
Reasons for Inadequate Decompression on Postoperative Imaging
Reasons for inadequate decompression on postoperative MRI in 8 cases are presented in Table 5. Six cases had inadequate bony decompression, of which 3 (50%) had persistent compression at >1 level, with C2-3 being the most common (4 cases). Two cases had adequate bony decompression but severe circumferential swelling with complete intrathecal displacement of subarachnoid space near the injury epicenter. Cases with insufficient laminectomy and severe circumferential swelling are shown in Figures 3 and 4, respectively.
TABLE 5.
Reasons for Lack of Decompression on Postoperative MRI
| Judgment of decompression on IOUS but not postoperative MRI | N = 8 | Comment |
|---|---|---|
| Insufficient laminectomy (inadequate bony decompression) | N = 6 | Level of insufficient laminectomy was • C2-3 in 4 cases, • C3-4 in 3 cases, • C4-5 in 1 case, • C7-T1 in 1 case. |
| Spinal cord edema (adequate bony decompression) | N = 2 | Judged to have adequate decompression on IOUS without the need for expansion duraplasty at the time of surgery. |
IOUS, intraoperative ultrasound.
FIGURE 3.
A 44-year-old male presenting after MVC with AO Spine morphology type B3 (ASIA/AIS 12A). A, Preoperative MRI. B1-B5, intraoperative ultrasound images demonstrating adequate decompression after anterior fixation and multilevel laminectomy. C, Postoperative MRI showing inadequate bony decompression at 2 sites (red arrows). AIS, ASIA Impairment Scale; ASIA, American Spinal Injury Association; MVC, motor vehicle collision.
FIGURE 4.
A 44-year-old male presenting after shallow dive with AO Spine injury morphology type C (teardrop fracture with retrolisthesis) (ASIA/AIS 8A). A, Preoperative MRI was performed before closed skeletal traction, anterior corpectomy, posterior fixation, and multilevel laminectomy. B1-B3, Short-axis IOUS images near the rostral, midpoint, and caudal aspects of the laminectomy bed, respectively. B4, Long-axis IOUS image within the laminectomy bed. C1 and C2, Postoperative MRI demonstrating inadequate decompression because of severe circumferential spinal cord swelling. The absence of both thecal sac deformation from external pressure and complete lack of CSF space near the injury epicenter is shown in the axial MR image (red arrow). AIS, ASIA Impairment Scale; ASIA, American Spinal Injury Association; CSF, cerebrospinal fluid; IOUS, intraoperative ultrasound.
DISCUSSION
Patients with severe clinical injuries and longer IMLL were significantly more likely to have inadequate decompression on postoperative imaging despite surgeon impression of adequate decompression using IOUS. Although the real-time capability of IOUS lends itself to ameliorating an inadequate preoperative plan, 2/5 (40%) cases that underwent additional laminectomy remained inadequately decompressed on postoperative MRI.
There are several factors that may influence these findings. First, patients with severe spinal cord injuries may continue to swell after surgery in the sagittal and axial dimensions at a much higher rate than those with less severe injuries. Real-time IOUS may therefore accurately visualize de facto decompression at the time of surgery, but over time, continued swelling in severe cases may result in delayed recompression at the rostral/caudal extent of the laminectomy bed or within the thecal sac. Second, patient positioning may play a role in the discrepancy. The cord could change position when transitioning from the prone position of surgery to the supine position of imaging. With this in mind, we advocate that IOUS should be checked after final rod tightening in cases requiring posterior fixation because a change in sagittal alignment could alter the status of the spinal cord. Finally, the morphology of the posterior elements of the cervical spine may play a role although none was shown to have a significant relationship in this series. This is suggested by inadequate bony decompression at C2-3 in 4 cases, perhaps because of the C2 bifid; however, there was inadequate bony decompression at several other levels.
Preoperative planning based on admission MRI, injury morphology, and neurological examination remains indispensable. With the currently available technology in most centers, IOUS cannot overcome poor preoperative planning because of its inability to visualize stenosis beyond an intact laminar segment. Preoperative IMLL was the most significant predictor of inadequate decompression on postoperative imaging. Despite visualization of adequate decompression with IOUS, additional swelling should be anticipated in the most severe cases. Two cases with severe circumferential swelling would have benefited from expansion duraplasty but were judged to not require it at the time of surgery. A randomized trial of standard treatment vs standard treatment plus duraplasty is currently ongoing (DISCUS, Duroplasty for Injured Cervical Spinal Cord with Uncontrolled Swelling); however, here, IOUS does not play a role in patient selection. Because duraplasty carries risks including cerebrospinal fluid leak, wound breakdown, and infection, future studies of IOUS could be performed to assess its ability to distinguish patients who would benefit from duraplasty from those in whom it is not required.
Establishing real-time IOUS as an acceptable alternative to postoperative MRI and CT myelography would provide several advantages, including decreased cost in finance and time and accessibility throughout the world. This study represents a preliminary step in understanding the utility and limitations of IOUS in the judgment of cervical spinal cord injury through a comparison with postoperative MRI and CT myelography. With knowledge of anticipated swelling, further surgeon experience, and improving technology, we believe that IOUS can become an established supplement to postoperative MRI for the judgment of spinal cord decompression, but this will require careful attention to the pitfalls that our study has demonstrated. Further research may reveal additional shortcomings not demonstrated in this study.
Future Directions
A randomized study comparing radiographic and clinical outcomes with and without IOUS is anticipated. The parameters of IOUS that predict need for expansion duraplasty require further investigation. Using contrast enhancement and Doppler technology, a series of questions arise regarding the relationships between spinal cord perfusion/blood flow, in vivo monitoring (ie, perfusion pressure), lesion morphology and expansion rate, adequacy of decompression, and clinical outcome.
Limitations of This Study
This is a retrospective, single-center investigation without randomization of IOUS. Statistical analysis is based on a relatively small sample and might have insufficient power to detect variables with small effects. Although the study demographics are largely consistent with current trends, there were more AIS grades C/D than A/B. There is currently an insufficient long-term clinical follow-up of at least 6 months for all study patients, and therefore, the primary focus was on verification of surgical decompression rather than clinical benefit. Judgment of decompression with intraoperative IOUS was dependent on individual surgical teams, any permutation of which could have provided differing intraoperative judgments. Postoperative imaging was acquired at nonstandardized time points and might have preceded or followed time of peak swelling.
CONCLUSION
Real-time IOUS provided good but not excellent (84.31%) judgment of cervical spinal cord decompression compared with postoperative MRI/CT myelography. There were significant relationships between markers of injury severity and likelihood of inadequate decompression on postoperative imaging. IOUS is a promising modality that can supplement postoperative MRI and CT myelography, but several factors including the influence of injury severity must be considered. Further research is required to establish the role of IOUS within the greater paradigm of cervical spinal cord injury management.
Footnotes
This research was presented in abstract form at the 90th Annual Meeting of the American Association of Neurological Surgeons, April 29 to May 2, 2022, in Philadelphia, PA and was selected to receive the Charles Tator Spinal Cord Injury Resident Research Award.
Contributor Information
Jesse A. Stokum, Email: jstokum@som.umaryland.edu.
Abdul-Kareem Ahmed, Email: AKAhmed@som.umaryland.edu.
Chixiang Chen, Email: chixiang.chen@som.umaryland.edu.
Aaron Wessell, Email: apwessell@gmail.com.
Gregory Cannarsa, Email: gregory.cannarsa@swedish.org.
Nicholas Caffes, Email: ncaffes@som.umaryland.edu.
Jeffrey Oliver, Email: joliver@som.umaryland.edu.
Joshua Olexa, Email: JOlexa@som.umaryland.edu.
Phelan Shea, Email: PShea@som.umaryland.edu.
Mohamed Labib, Email: mlabib@som.umaryland.edu.
Graeme Woodworth, Email: gwoodworth@som.umaryland.edu.
Alexander Ksendzovsky, Email: aksendzovsky@som.umaryland.edu.
Uttam Bodanapally, Email: ubodanapally@umm.edu.
Kenneth Crandall, Email: kcrandall@som.umaryland.edu.
Charles Sansur, Email: csansur@som.umaryland.edu.
Gary Schwartzbauer, Email: gschwartzbauer@som.umaryland.edu.
Bizhan Aarabi, Email: baarabi@som.umaryland.edu.
Funding
This study did not receive any funding or financial support.
Disclosures
The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES
- 1.Zipser CM, Cragg JJ, Guest JD, et al. Cell-based and stem-cell-based treatments for spinal cord injury: evidence from clinical trials. Lancet Neurol. 2022;21(7):659-670. [DOI] [PubMed] [Google Scholar]
- 2.Sköld MK, Fehlings MG. Editorial: karolinska institutet 200-year anniversary symposium on injuries to the spinal cord and peripheral nervous system-an update on recent advances in regenerative neuroscience. Front Neurol. 2017;8:510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ahuja CS, Nori S, Tetreault L, et al. Traumatic spinal cord injury-repair and regeneration. Neurosurgery. 2017;80(3S):S9-S22. [DOI] [PubMed] [Google Scholar]
- 4.Baptiste DC, Tighe A, Fehlings MG. Spinal cord injury and neural repair: focus on neuroregenerative approaches for spinal cord injury. Expert Opin Investig Drugs. 2009;18(5):663-673. [DOI] [PubMed] [Google Scholar]
- 5.Baptiste DC, Fehlings MG. Pharmacological approaches to repair the injured spinal cord. J Neurotrauma. 2006;23(3-4):318-334. [DOI] [PubMed] [Google Scholar]
- 6.Bastidas J, Athauda G, De La Cruz G, et al. Human Schwann cells exhibit long-term cell survival, are not tumorigenic and promote repair when transplanted into the contused spinal cord. Glia. 2017;65(8):1278-1301. [DOI] [PubMed] [Google Scholar]
- 7.Muradov JM, Ewan EE, Hagg T. Dorsal column sensory axons degenerate due to impaired microvascular perfusion after spinal cord injury in rats. Exp Neurol. 2013;249:59-73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Carlson SL, Parrish ME, Springer JE, Doty K, Dossett L. Acute inflammatory response in spinal cord following impact injury. Exp Neurol. 1998;151(1):77-88. [DOI] [PubMed] [Google Scholar]
- 9.Simard JM, Tsymbalyuk O, Keledjian K, Ivanov A, Ivanova S, Gerzanich V. Comparative effects of glibenclamide and riluzole in a rat model of severe cervical spinal cord injury. Exp Neurol. 2012;233(1):566-574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Simard JM, Popovich PG, Tsymbalyuk O, Gerzanich V. Spinal cord injury with unilateral versus bilateral primary hemorrhage--effects of glibenclamide. Exp Neurol. 2012;233(2):829-835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Simard J, Woo S, Schwartzbauer G, Gerzanich B. Sulfonylurea receptor 1 in central nervous system injury: a focused review. J Cereb Blood Flow Metab. 2012;32(9):1699-1717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Woo SK, Kwon MS, Ivanov A, Gerzanich V, Simard JM. The sulfonylurea receptor 1 (Sur1)-transient receptor potential melastatin 4 (Trpm4) channel. J Biol Chem. 2013;288(5):3655-3667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Luo J, Shi R. Diffusive oxidative stress following acute spinal cord injury in Guinea pigs and its inhibition by polyethylene glycol. Neurosci Lett. 2004;359(3):167-170. [DOI] [PubMed] [Google Scholar]
- 14.Abraham J, Balasubramanian AS, Theodore DR, Nagarajan S, Apte CA, Chandi S. Spinal cord edema, 5-hydroxytryptamine, lipid peroxidation, and lysosomal enzyme release after acute contusion and compression injury in primates. Cent Nerv Syst Trauma. 1985;2(1):45-60. [DOI] [PubMed] [Google Scholar]
- 15.Tator CH, Deecke L. Studies of the treatment and pathophysiology of acute spinal cord injury in primates. Paraplegia. 1973;10(4):344-345. [DOI] [PubMed] [Google Scholar]
- 16.Choo AM, Liu J, Dvorak M, Tetzlaff W, Oxland TR. Secondary pathology following contusion, dislocation, and distraction spinal cord injuries. Exp Neurol. 2008;212(2):490-506. [DOI] [PubMed] [Google Scholar]
- 17.Saadoun S, Papadopoulos MC. Targeted perfusion therapy in spinal cord trauma. Neurotherapeutics. 2020;17(2):511-521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Varsos GV, Werndle MC, Czosnyka ZH, et al. Intraspinal pressure and spinal cord perfusion pressure after spinal cord injury: an observational study. J Neurosurg Spine. 2015;23(6):763-771. [DOI] [PubMed] [Google Scholar]
- 19.Le E, Aarabi B, Hersh DS, et al. Predictors of intramedullary lesion expansion rate on MR images of patients with subaxial spinal cord injury. J Neurosurg Spine. 2015;22(6):611-621. [DOI] [PubMed] [Google Scholar]
- 20.Aki T, Toya S. Experimental study on changes of the spinal-evoked potential and circulatory dynamics following spinal cord compression and decompression. Spine (Phila Pa 1976). 1984;9(8):800-809. [DOI] [PubMed] [Google Scholar]
- 21.Vlajić I. Microangiographic observations of morphological vessel changes after experimental spinal cord trauma. Adv Neurol. 1978;20:451-460. [PubMed] [Google Scholar]
- 22.Badhiwala JH, Wilson JR, Witiw CD, et al. The influence of timing of surgical decompression for acute spinal cord injury: a pooled analysis of individual patient data. Lancet Neurol. 2021;20(2):117-126. [DOI] [PubMed] [Google Scholar]
- 23.Crutchfield WG. Skeletal traction in treatment of injuries to the cervical spine. J Am Med Assoc. 1954;155(1):29-32. [DOI] [PubMed] [Google Scholar]
- 24.Aarabi B, Olexa J, Chryssikos T, et al. 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. J Neurotrauma. 2019;36(6):862-876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Aarabi B, Akhtar-Danesh N, Chryssikos T, et al. Efficacy of ultra-early (<12 h), early (12-24 h), and late (>24-138.5 h) surgery with magnetic resonance imaging-confirmed decompression in American spinal injury association impairment scale grades A, B, and C cervical spinal cord injury. J Neurotrauma. 2020;37(3):448-457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Aarabi B, Akhtar-Danesh N, Simard JM, et al. Efficacy of early (≤24 hours), late (25-72 hours), and delayed (>72 hours) surgery with magnetic resonance imaging-confirmed decompression in American spinal injury association impairment scale grades C and D acute traumatic central cord syndrome caused by spinal stenosis. J Neurotrauma. 2021;38(15):2073-2083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ganau M, Syrmos N, Martin AR, Jiang F, Fehlings MG. Intraoperative ultrasound in spine surgery: history, current applications, future developments. Quant Imaging Med Surg. 2018;8(3):261-267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Dohrmann GJ, Rubin JM. Intraoperative ultrasound imaging of the spinal cord: syringomyelia, cysts, and tumors—a preliminary report. Surg Neurol. 1982;18(6):395-399. [DOI] [PubMed] [Google Scholar]
- 29.Rubin JM, Dohrmann GJ. Work in progress: intraoperative ultrasonography of the spine. Radiology. 1983;146(1):173-175. [DOI] [PubMed] [Google Scholar]
- 30.Rubin JM, Dohrmann GJ. The spine and spinal cord during neurosurgical operations: real-time ultrasonography. Radiology. 1985;155(1):197-200. [DOI] [PubMed] [Google Scholar]
- 31.St Amour TE, Rubin JM, Dohrmann GJ. The central canal of the spinal cord: ultrasonic identification. Radiology. 1984;152(3):767-769. [DOI] [PubMed] [Google Scholar]
- 32.Jokich PM, Rubin JM, Dohrmann GJ. Intraoperative ultrasonic evaluation of spinal cord motion. J Neurosurg. 1984;60(4):707-711. [DOI] [PubMed] [Google Scholar]
- 33.Rubin JM, Dohrmann GJ. Intraoperative spinal ultrasonography. J Belge Radiol. 1986;69(1):9-27. [PubMed] [Google Scholar]
- 34.Dohrmann GJ, Rubin JM. Cervical spondylosis and syringomyelia: suboptimal results, incomplete treatment, and the role of intraoperative ultrasound. Clin Neurosurg. 1988;34:378-388. [PubMed] [Google Scholar]
- 35.Dohrmann GJ, Rubin JM. Neurotraumatology: ultrasonic evaluation of the central nervous system. Neurol Res. 1997;19(3):317-322. [DOI] [PubMed] [Google Scholar]
- 36.Dohrmann GJ, Rubin JM. History of intraoperative ultrasound in neurosurgery. Neurosurg Clin N Am. 2001;12(1):155-166, ix. [PubMed] [Google Scholar]
- 37.Montalvo BM, Quencer RM. Intraoperative sonography in spinal surgery: current state of the art. Neuroradiology. 1986;28(5-6):551-590. [DOI] [PubMed] [Google Scholar]
- 38.Quencer RM, Montalvo BM, Eismont FJ, Green BA. Intraoperative spinal sonography in thoracic and lumbar fractures: evaluation of Harrington rod instrumentation. AJR Am J Roentgenol. 1985;145(2):343-349. [DOI] [PubMed] [Google Scholar]
- 39.Montalvo BM, Quencer RM, Green BA, Eismont FJ, Brown MJ, Brost P. Intraoperative sonography in spinal trauma. Radiology. 1984;153(1):125-134. [DOI] [PubMed] [Google Scholar]
- 40.Montalvo BM, Quencer RM, Brown MD, et al. Lumbar disk herniation and canal stenosis: value of intraoperative sonography in diagnosis and surgical management. AJNR Am J Neuroradiol. 1990;11(1):31-40. [PMC free article] [PubMed] [Google Scholar]
- 41.Mirvis SE, Geisler FH. Intraoperative sonography of cervical spinal cord injury: results in 30 patients. AJR Am J Roentgenol. 1990;155(3):603-609. [DOI] [PubMed] [Google Scholar]
- 42.Vaccaro AR, Koerner JD, Radcliff KE, et al. AOSpine subaxial cervical spine injury classification system. Eur Spine J. 2016;25(7):2173-2184. [DOI] [PubMed] [Google Scholar]
- 43.Tat J, Tat J, Yoon S, Yee AJM, Larouche J. Intraoperative ultrasound in spine decompression surgery: a systematic review. Spine (Phila Pa 1976). 2021;47(2):E73-E85. [DOI] [PubMed] [Google Scholar]
- 44.Fleiss J. Measuring nominal scale agreement among many raters. Psychol Bull. 1971;76(5):378-382. [Google Scholar]