Neurological Recovery after Traumatic Cervical Spinal Cord Injury Is Superior if Surgical Decompression and Instrumented Fusion Are Performed within 8 Hours versus 8 to 24 Hours after Injury: A Single Center Experience

Abstract

A prospective study was performed to evaluate the impact of surgical decompression (SD) and instrumented fusion within 8 h versus 8–24 h after injury on neurological recovery after cervical traumatic spinal cord injury (tSCI) in patients operated on in the UMC Ljubljana, Slovenia. Only patients with the American Spinal Injury Association (ASIA) Impairment Scale (AIS) grades of A through C and with MRI-confirmed spinal cord compression were enrolled. The primary outcome was the change in AIS grade at the 6-month follow-up. Of the 48 enrolled patients, 22 patients who underwent surgery within 8 h (group 8 h) and 20 patients who underwent surgery between 8 and 24 h (Group 8–24 h) after injury concluded the study. At admission, there was no statistically significant difference in AIS grade between the study groups. At the 6-month follow-up, an improvement of at least two AIS grades was found in 45.5% of patients in group 8 h and in 10% of patients in group 8–24 h (p=0.017). The median improvement in the ASIA motor score was 38.5 (10.0–61.0) motor points in group 8 h and 15.0 (8.8–34.0) motor points in group 8–24 h (p=0.0468). In a multivariate analysis, adjusted for the preoperative AIS grade and the degree of spinal canal compromise, the odds of an at least two-grade AIS improvement were at least 106% higher for patients in group 8 h than for patients in group 8–24 h (odds ratio=11.08, p=0.004). No statistically significant difference was found in the rate of perioperative complications, pneumonia, and the number of ventilator-dependent days or the mortality between the groups. Our results suggest that the patients with tSCI who undergo SD within 8 h after injury have superior neurological outcomes than patients who undergo SD 8–24 h after injury, without any increase in the rate of adverse effects.

Introduction

Traumatic spinal cord injury (tSCI) is a catastrophic event with enormous personal, social, and economic impact.¹ Despite recent progress in understanding the pathophysiology of acute tSCI, treatment and protocols aimed at ameliorating the neurologic outcomes in patients remain inconclusive. Current concepts of the pathophysiology of acute tSCI indicate that both primary and secondary injuries are involved in neural tissue destruction.^2,3 The primary injury, usually caused by rapid spinal cord compression and contusion, also initiates a signaling cascade of downstream cellular mechanisms leading to a secondary injury. If the primary injury of neural tissue cannot be prevented, alleviating the secondary injury represents the only therapeutic target in patients with acute tSCI to date.^4,5 The evidence from animal studies suggests that the prolonged duration of spinal cord compression after tSCI exacerbates the secondary injury and is inversely related to neurological recovery.^6
–8 Despite the positive effects of acute spinal cord decompression on neurological recovery reported in these standardized preclinical studies, neurological benefits of early intervention remain elusive in the clinical setting.

According to a recent meta-analysis of clinical studies by van Middendorp and associates,⁹ the patients who underwent spinal surgery within 72 h after tSCI had significantly better neurological recovery outcomes and significantly shorter hospital admissions than the patients who underwent spinal surgery 1 week to 6–12 months after tSCI. Surprisingly, the criterion-based sensitivity analyses in that meta-analysis counterintuitively demonstrated greater neurological recovery in patients who underwent spinal surgery within 72 h after injury than in those who underwent spinal surgery within the first 24 h after injury.⁹ These results, however, are far from robust because these studies included patients with different severities of tSCI located at different levels of the spinal cord; the studies also demonstrated a high susceptibility to various sources of bias.⁹ In contrast, in two studies directly comparing neurological recovery between patients operated on within 24 h or between 24 and 72 h after tSCI, the neurological recovery was greater in the patients who underwent spinal surgery within 24 h, although the differences were not statistically significant.^10,11 In addition, in two recent studies comparing the effect of early surgical decompression within 24 h to later time frames, the decompression before 24 h after tSCI was associated with significantly improved neurological outcomes.^12,13

The effect of surgical decompression (SD) of the injured spinal cord on neurological recovery within different time windows during the first 24 h after injury remains poorly investigated. No statistically significant difference in neurological recovery was found between the patients with different tSCI levels who underwent SD within 8 h and those who underwent SD between 8 and 24 hours after tSCI.¹⁴ As suggested in a recent study, however, cervical facet dislocations should probably be reduced within 4 hours of injury to prevent permanent neurological damage.¹⁵ To the best of our knowledge, no study has yet compared the effects of SD of the spinal cord between different time windows within the first 24 h after injury on neurological recovery in patients with acute tSCI with cervical spine fractures or dislocations. Therefore, in this study, the hypothesis that neurological recovery 6 months after acute tSCI because of cervical spine fracture or dislocation is superior if SD and instrumented fusion are performed within the first 8 h than from 8 to 24 h after injury was tested. In addition, the effect of the severity of the tSCI (complete vs. incomplete) and the degree of spinal canal compromise on neurological outcome were examined.

Methods

A prospective cohort study of consecutive patients with acute cervical tSCI and fracture or dislocation of the subaxial cervical spine operated on within the first 24 h after the injury was performed from January 2007 to December 2012 at a single Level 1 Trauma Center (University Medical Centre Ljubljana, Ljubljana, Slovenia). Research ethics board approval was obtained before patient enrolment.

The neurological examination was performed according to the standards established by the American Spinal Injury Association (ASIA)¹⁶ first by a trained orthopedic trauma specialist who was blinded with respect to surgical treatment at the time of presentation at our institution or at the regional hospital and thereafter by the orthopedic spine surgeon who operated on the patient. Each patient underwent plain radiography, computed tomography (CT), and magnetic resonance imaging (MRI) of the cervical spine. The degree of spinal canal compromise was recorded according to the measurements described by Fehlings and colleagues,¹⁷ and re-examined after surgery with plain radiography or CT scan.

Before the enrolment, the set of inclusion and exclusion criteria was defined (Table 1). Only patients presenting with overall ASIA Impairment Scale (AIS) grades of A through C were enrolled to allow the comparison of a potential two AIS grade improvement in the cohort groups, as suggested by the Sygen trial.¹⁸ Only patients with spinal cord compression and spinal canal compromise of at least 25% on midsagital T2-weighted MRI¹⁷ were enrolled. All the enrolled patients were given methylprednisolone according to the recommendations of the third National Acute Spinal Cord Injury Study and were admitted to the same intensive care unit (ICU) for optimal postoperative care.^19,20

Table 1.

Inclusion and Exclusion Criteria

Inclusion criteria	Exclusion criteria
Age 16–85 years	Neurological deficits before injury
Initial AIS grade A–C	No evidence of fracture or dislocation
Fracture or dislocation C3–C7	Central cord syndrome
Neurological level of injury between C3 and C8	Life threatening injuries that prevent early decompression of the spinal cord
Spinal cord compression confirmed by MRI	Cognitive impairment preventing accurate neurologic assessment
Spinal canal compromise of at least 25%
Surgery within 24 h of tSCI

AIS, American Spinal Injury Association Impairment Scale; MRI, magnetic resonance imaging; tSCI, traumatic spinal cord injury.

According to the time elapsed from the injury to the SD, the cohort of consecutive patients was divided into two groups: the group in which SD was achieved within 8 h after injury (group 8 h) and the group in which SD was achieved from 8 to 24 h after injury (group 8–24 h). In spinal dislocations, the time of SD was defined as the time at which successful reduction resulted in spinal cord decompression. In patients with anterior spinal cord compression because of bony or disc material, the time of SD was defined as the time at which successful decompression through disc- or corpectomy resulted in spinal cord decompression. SD was followed by instrumented spinal fusion in all patients. Because all patients were operated on in the same institution, the same rationale of treatment regarding anterior, posterior, or combined surgical approaches was used. First, skeletal traction using a halo ring was applied in all patients in general anesthesia in the operating room. In spinal dislocations, closed reduction under fluoroscopic control was attempted. After fluoroscopic confirmation of a successful reduction, an anterior and/or posterior spinal fusion was performed, depending on the spinal injury pattern. In cases of unsuccessful closed reduction, the patients underwent open reduction through the posterior approach, followed by a posterior or a 360-degree fusion. The details of surgical procedures are presented in Table 2 and compared between the groups.

Table 2.

Surgical Procedure, Perioperative Data, and Length of Hospital Stay

Characteristic	Group 8 h	Group 8–24 h	p value
Surgical procedure
OR+ PF	1 (5)	0 (0)	0.196
OR+ PF+ ADF	3 (14)	0 (0)
CR+ PF	1 (5)	0 (0)
CR+ PF+ L	0 (0)	1 (5)
CR+ ADF+ PF	1 (5)	0 (0)
CR+ ACF	3 (14)	8 (40)
CR+ ACF+ PF	1 (5)	0 (0)
CR+ ACF+ PF+ L	0 (0)	1 (5)
CR+ ACF+ PPF	0 (0)	1 (5)
CR+ ADF	9 (41)	6 (30)
CR+ ADF+ PF	1 (5)	1 (5)
CR+ ADF+ PF+ L	0 (0)	1 (5)
CR+ ADF+ PPF	2 (9)	1 (5)
Laminectomy
Yes	0 (0)	3 (15)	0.099
Number of fused levels
1	10 (45)	12 (60)	0.673
2	11 (50)	7 (35)
3	1 (5)	1 (5)
Fusion rate
%	100	100	1
Surgical time
Min	161.8 (52.5)	169.0 (45.6)	0.4216
Blood loss
mL	347.7 (216.3)	387.5 (224.7)	0.4215
Perioperative complications
Surgical infection	0 (0)	1 (5)	0.476
CSF leak	2 (9)	0 (0)	0.489
Cardiovascular	0 (0)	2 (10)	0.221
Gastrointestinal	1 (5)	2 (10)	0.598
Length of HS
days	38.8 (24.0)	48.8 (40.3)	0.4271

OR, open reduction; PF, posterior fusion; ADF, anterior discectomy and fusion; CR, closed reduction; L, laminectomy; ACF, anterior corpectomy and fusion; PPF, postponed posterior fusion; CSF, cerebrospinal fluid; HS, hospital stay.

For each cohort, group medians (IQR) for numerical variables and counts (%) for categorical variables are presented.

All the patients started physical therapy during hospital stay. After cardiopulmonary stabilization, all patients were transferred to the same spinal cord rehabilitation clinic for a 6-month rehabilitation protocol. Six months after SCI, the neurological measurement was performed by a specialist of rehabilitation medicine in the rehabilitation clinic and an orthopedic spine surgeon specialist, both blinded with respect to surgical timing.

The primary outcome measure of interest was the ordinal change in the AIS grade at the 6-month follow-up. We also evaluated the change in the ASIA motor score (AMS)¹⁶ between admission and 6 months after tSCI. The 6-month period for follow-up was based on recommendations used in the National Spinal Cord Injury Study and Sygen trial.^18,19 In addition, the impact of the degree of spinal canal compromise and the severity of tSCI (complete vs. incomplete) on neurologic recovery were tested in a multivariate regression model, thus controlling for the change in time to surgery. The neurologic recovery was constructed as an at least a two-grade AIS improvement, the severity of tSCI as a complete tSCI group (AIS grade A) or incomplete tSCI group (AIS grades B and C), and the spinal canal compromise as increasing degrees of preoperative spinal canal compromise.

Additional clinical parameters collected at admission included patient age, sex, body mass index, smoking habits, presence of comorbidity (heart disease, high blood pressure, or diabetes), injury etiology, accompanying injuries, and transfer from other hospitals. The level and pattern of spinal injury were determined according to the classification of injuries of the cervical spine as proposed by Aebi and Nazarian and are presented in Table 3.²¹ Perioperative complications were identified and documented during the patients' hospital stay and on subsequent follow-ups and are presented in Table 2. The occurrence of pneumonia during the first month after injury and the number of ventilator-dependent days were also registered. All parameters were compared between the cohort groups.

Table 3.

Preoperative Demographic and Injury Characteristics of Patients

Characteristic	Group 8 h	Group 8–24 h	p value
Age
Years	44 (30.5–58.5)	52 (25.8–72.8)	0.715
Sex
Male	18 (82)	16 (80)	1
Body mass index	25.0 (2.6)	24.6 (2.9)	0.509
Smoking
Yes	6 (27)	7 (35)	0.7410
Comorbidity
Yes	3 (14)	7 (35)	0.741
Etiology
Assault	2 (9)	0 (0)	0.246
Diving	3 (14)	3 (15)
Fall	6 (27)	10 (50)
Motor vehicle accident	8 (36)	7 (35)
Sport	3 (14)	0 (0)
Transferred from other hospitals
Yes	0 (0)	12 (60)	<0.001
No	22 (100)	8 (40)
Pre-op AIS grade
AIS A	13 (59)	13 (65)	0.249
AIS B	5 (23)	1 (5)
AIS C	4 (18)	6 (30)
SCC
%	50 (30–50)	30 (30–50)	0.239
Level of spinal injury
C3–C4	1 (5)	3 (15)	0.308
C4–C5	4 (18)	2 (10)
C5-C6	5 (23)	7 (35)
C6-C7	10 (45)	7 (35)
C7-T1	2 (9)	1 (5)
Spinal injury pattern
A	4 (18)	4 (20)	0.308
B	5 (23)	1 (5)
C	13 (59)	15 (75)

AIS, American Spinal Injury Association Impairment Scale; SCC, spinal canal compromise.

For each cohort, group medians (IQR) for numerical variables and counts (%) for categorical variables are presented.

Statistical analysis

All analyses were performed using R environment.²² A statistical confidence alpha level of 0.05 was considered. The differences in patient demographics, injury characteristics, surgical procedures, and complications between the cohort groups were analyzed by using the Mann-Whitney U test for numerical variables and the Fisher exact test for categorical variables. The association between the improvement in AIS grade and the timing of surgery (group 8 h vs. group 8–24 h) was analyzed by using the Fisher exact test. Relative risks (RR) with confidence intervals were computed for evaluation of possible associations. The association between the change in AMS and the timing of surgery (group 8 h vs. group 8–24 h) was analyzed by the one-sided Mann-Whitney U test. The impact of the degree of tSCI and the degree of spinal canal compromise on neurologic recovery adjusted for the cohort groups was tested by using the Firth penalized logistic regression that allows for the inclusion of more variables in small sample settings.²³

Results

Study population

During the 6-year study period, a total of 66 consecutive patients with tSCI and fractures or dislocations of the subaxial cervical spine were operated on. In nine patients, surgery was not performed within 24 h after tSCI—in four cases of delayed transfer from other hospitals, in two cases of marked neurological recovery without spinal cord compression on MRI, in two cases of comorbidity and/or polytrauma, and in one case of missed SCI at admission in an obtunded patient. The remaining 57 patients were operated on within 24 h after tSCI, but 9 of them did not meet the study inclusion criteria. Of the remaining 48 patients, in 26 of them SD was achieved within 8 h after tSCI (group 8 h) and in 22 patients SD was achieved between 8 and 24 h after tSCI (group 8–24 h). In group 8 h, two patients who died from pulmonary failure, one patient in whom the decompression was inadequate, and one patient who was lost during follow-up were excluded from the study; 22 patients remained in this group. In group 8–24 h, one patient who died from pulmonary failure and one patient in whom decompression was inadequate, as determined by postoperative imaging, were excluded from the study, which left 20 patients in this group.

The differences in pre-operative AIS grade, degree of spinal canal compromise, spinal injury pattern, and level of spinal injury were not statistically significant between the groups (p>0.05). The differences in age, sex, body mass index, smoking habit, comorbidity, and injury etiology were not statistically significant between the groups (p>0.05), but there were statistically significantly more patients transferred from other hospitals in group 8–24h than in group 8 h (p<0.001). A list of demographic and injury characteristics of both group 8 h and group 8–24 h is provided in Table 3. Ten (45%) patients in group 8 h and 10 (50%) patients in group 8–24 h had pneumonia during the first month after injury (p>0.05). The median length of ventilator dependency was 6.5 (1–17) days in group 8 h and 5 (1.25–12) days in group 8–24 h (p>0.05). Two patients in group 8 h and one patient in group 8–24 h died because of cardiopulmonary failure before the conclusion of the study (p>0.05).

The median times to SD in group 8 h and group 8–24 h were 5 h (4–6) and 11 h (8.75–15), respectively. The differences in surgical procedures, number of laminectomies and fused levels, fusion rate, surgical time, blood loss, length of hospital stay, and complication rate were not statistically significant between the groups (p>0.05). A detailed list of surgical procedures, perioperative data, and the length of hospital stay are presented in Table 2.

Neurological outcome

The neurological improvement according to the change in AIS grade from the pre-operative assessment until the discharge from the rehabilitation unit 6 months after tSCI is represented in Tables 4 and 5. The distribution of AIS grades at admission was not statistically significantly different between group 8 h and group 8–24 h (p>0.05). Improvement of at least two AIS grades was found in 10 (45.5%) patients in group 8 h and in two (10%) patients in group 8–24 h (p=0.017) (RR 2.08; 95% confidence interval [CI]: 1.12, 3.87). An improvement of at least one AIS grade was found in 16 (72.7%) patients in group 8 h and in nine (45%) patients in group 8–24 h (p=0.115) (RR 1.81; 95% CI: 0.76, 4.30) (Fig. 1).

FIG. 1.

Recovery of American Spinal Injury Association Impairment Scale (AIS) grades at 6 months after acute traumatic spinal cord injury and surgical decompression within 8 h after injury (■ group 8 h) and from 8–24 h after injury (□ group 8–24 h). *Statistically significantly different between the groups (p=0.017) (relative risks 2.08; 95% confidence interval: 1.12, 3.87).

Table 4.

Changes in American Spinal Injury Association Impairment Scale Grade from Pre-Operative to 6 Months Follow-Up in Group 8 h

Preoperative AIS grade	A	B	C	D	E	Total
A	6	4	1	2		13
B				3	2	5
C				2	2	4

AIS, American Spinal Injury Association Impairment Scale.

Table 5.

Changes in American Spinal Injury Association Impairment Scale Grade from Pre-Operative to 6 Months Follow-Up in Group 8–24 h

Preoperative AIS grade	A	B	D	Total
A	11	1	1	13
B			1	1
C			6	6

AIS, American Spinal Injury Association Impairment Scale.

In the Firth penalized logistic regression model, the degree of tSCI (complete vs. incomplete) and the degree of spinal canal compromise were included in addition to the time elapsed between the injury and surgical intervention (Table 6). The odds of at least a two-grade AIS improvement were at least 106% (odds ratio [OR]=11.08, p=0.004) higher for those in group 8 h than for those in group 8–24 h after adjusting for pre-operative AIS grade and the degree of spinal canal compromise. The odds of at least a two-grade AIS improvement were not statistically significantly different between complete and incomplete patients with SCI (OR=0.26, p=0.087) and between those with different degrees of spinal canal compromise (OR=0.94, p=0.066), after the adjustment for the time interval to SD (group 8 h or Group 8–24 h) and the degree of spinal canal compromise or for the time to SD and pre-operative AIS grade, respectively.

Table 6.

Results of Penalized Logistic Regression Assessing the Effect of Surgical Decompression and Instrumented Fusion within 8 h Versus from 8 to 24 h after tSCI on Neurological Outcome, Adjusted for Preoperative AIS Grade and Spinal Canal Compromise

	Estimate	Lower 95%	Upper 95%	p value
Intercept	0.94	−1.74	3.92	0.495
Group 8 h	2.4	0.72	4.55	0.004
SCC	−0.06	−0.15	0	0.066
Complete tSCI	−1.36	−3.07	0.2	0.087

SCC, spinal canal compromise; tSCI, traumatic spinal cord injury.

In addition, we ran the analysis with exact numerical values of the time elapsed in hours from injury to SD (without categorizing the patients into group 8 h and group 8–24 h). The model fit and the results were very similar to the model described above with the only statistically significant variable being the hours to surgery. The odds of at least a two-grade AIS improvement were at least 2% (OR=0.83, p=0.029) lower for each additional hour from injury to surgery, after adjusting for the preoperative AIS grade and the degree of spinal canal compromise (Table 7).

Table 7.

Results of Penalized Logistic Regression Assessing the Effect of Surgical Decompression and Instrumented Fusion Performed At Different Time Intervals after Injury (Not Categorizing the Patients Into Group 8 h and Group 8–24 h), Adjusted for pre-operative American Spinal Injury Association Impairment Scale Grade and Spinal Canal Compromise

	Estimate	Lower 95%	Upper 95%	p value
Intercept	3.47	0.35	7.34	0.028
Hours to SD	−0.18	−0.42	−0.02	0.029
SCC	−0.05	−0.13	0.01	0.086
Complete tSCI	−1.28	−2.84	0.18	0.085

SD, surgical decompression; SCC, spinal canal compromise; tSCI, traumatic spinal cord injury.

The median value of AMS changed from admission to final follow-up at 6 months after tSCI from 17 to 51.5 points in group 8 h and from 10 to 22 points in group 8–24 h. The median improvement in AMS from the pre-operative assessment until 6 months after tSCI was 38.5 (10.0–61.0) motor points in group 8 h and 15.0 (8.8–34.0) motor points in group 8–24 h (p=0.0468).

Discussion

We performed a prospective cohort study at a single trauma center to evaluate the effect of the timing of SD within the span of 24 h after an acute tSCI and fracture or dislocation of the subaxial cervical spine on neurological recovery. In accordance with our hypothesis, we found a significantly higher proportion of patients with at least a two-grade AIS improvement at 6 months after tSCI in the group in which the patients underwent spinal surgery within 8 h after tSCI (45.5%) than in the group in which the patients underwent spinal surgery between 8 h and 24 h after tSCI (10%). Higher odds of at least a two-grade AIS improvement among the patients in group 8 h than in group 8–24 h were also found by using the penalized logistic regression model after the adjustment for the degree of pre-operative spinal canal compromise and severity of tSCI. Moreover, the adjusted analysis with the numerical values of the time elapsed from the injury to the surgery (without categorizing the patients into group 8 h and group 8–24 h) showed that the odds of at least a two-grade AIS improvement were significantly lower for each additional hour elapsing from injury to surgery. In addition, the median improvement in AMS from the preoperative assessment until 6 months after tSCI was statistically significantly higher in group 8 h than in group 8–24 h.

Studies focusing on the feasibility of SD within the first 24 h after tSCI showed that only between 24% and 51% of tSCI patients could undergo SD within the first 24 h and fewer than 10% of patients could receive surgical attention in the first 8 h after injury.^24
–26 Therefore, in the majority of recent clinical studies, the 24-h cutoff point was considered as a “practical” time window for early SD after tSCI.^12,13,27,28 Animal models, however, show a clear relationship between the neurologic recovery after tSCI and the timing of spinal cord decompression, with superior neurologic recovery after decompression within the first hours after injury, questioning the 24-h cutoff point as the “ideal” time for SD in humans.^6
–8 In the current study, the decision for the additional 8-h cutoff point was based on statistics of our center, showing that approximately one half of our patients with tSCI underwent surgery within 8 h after injury. Because the treatment approach at our institution is to operate on each patient with tSCI and spinal cord compression as soon as possible after tSCI, we believed that withholding early treatment from some patients for the purpose of randomization would be unethical. Therefore, the randomization of patients was determined primarily by the time elapsed from the injury to the patient's arrival at our institution. In this manner, we were able to perform a SD within 24 h after tSCI in 86.4% of all patients with fractures or dislocations of the subaxial cervical spine and acute tSCI. Of the 48 patients enrolled in the study, 54.2% had decompression within 8 h after tSCI.

We are not aware of any clinical study directly comparing the effect of SD on neurological recovery in cervical spine fractures or dislocations between different time frames within 24 h after tSCI. In the study examining the effect and safety of pharmacological therapy in acute tSCI, Pointillart and associates¹⁴ found no significant difference in the neurological recovery of the paraplegic and quadriplegic patients who underwent SD within 8 h and those who were either operated on between 8 and 24 h after tSCI or did not undergo surgery.¹⁴ As the primary interest of that study was to evaluate the effect of pharmacological therapy and not of the time elapsed from tSCI to the surgical intervention, it is difficult to compare its results with those of our study. In addition, because both quadriplegic and paraplegic patients were included in that study, the full potential of SD for neurological recovery in the quadriplegic patients might have been underestimated, because historical data show that they might achieve superior spontaneous recovery.²⁹ In contrast, the extent of the tSCI after the dislocation of the cervical spine sustained during rugby is reduced if spinal cord decompression by traction and closed reduction is achieved within 4 hours after injury, but not later.¹⁵ Moreover, in that study, the time to reduction was negatively correlated with the Frankel grade at discharge, with the cut-point most likely to predict complete paralysis on discharge after a delay of more than 7 hours.¹⁵ The results of that study are in line with our findings of a significantly greater rate of improvement for at least two AIS grades and a significantly higher improvement in AMS in patients operated on within 8 h after tSCI and significantly lower odds of comparable improvement for each additional hour to surgery. Both studies support the hypothesis that the decompression of the spinal cord within the first hours of tSCI results in superior neurological recovery than delayed decompression.

We found that 72.7% of all patients undergoing decompression within the first 8 h after tSCI improved by at least one AIS grade, which is in line with 67% of patients improving by one or more AIS grades with decompression by traction or surgery within 8 h of tSCI reported by McCarthy.³⁰ The improvement of at least two AIS grades in 23% of the complete tSCI patients in the group 8 h in our study is in accordance with 28% of the complete tSCI patients improving by at least two AIS grades in a recent study in which SD and cooling of the spinal cord was performed within 8 h after tSCI.³¹ In our entire cohort, 59.5% and 28.5% of all patients undergoing decompression within the first 24 h after tSCI improved by at least one or two AIS grades, respectively, which is in line with the recent report by Fehlings and colleagues,¹² who found an improvement of at least one AIS grade in 56.5% and an improvement of at least two AIS grades in 19.8% of the patients with cervical spine injury and tSCI operated within 24 h with similar inclusion and exclusion criteria. Therefore, we believe that the superior effect of SD on neurological recovery, if performed within 8 h after tSCI compared with SD performed between 8 to 24 h in our study, was rather because of the favorable effect of earlier surgery than because of an inferior quality in the management protocol of the group 8–24 h.

On the other hand, it would be possible that the superior neurological outcome in group 8 h vs. group 8–24 h was because of the higher rate of AIS grade B patients who may have higher chances of improvement for two AIS grades (i.e., from AIS B to AIS D) than AIS grade C patients (i.e., from AIS C to AIS E). The difference, however, in AIS grade distribution between groups at admission was not statistically significant. In addition, 40% of AIS grade B and 50% of AIS grade C patients in group 8 h recovered to AIS grade E, whereas no patient in group 8–24 h made such a recovery. Moreover, if all AIS grade B patients in group 8 h would have spontaneously recovered to AIS grade C with delayed treatment, still 32% of all patients in group 8 h would recover at least two AIS grades compared with 10% of all patients in group 8–24 h. Therefore, we believe that the superior neurological recovery in group 8 h was because of the shorter time to SD rather than the difference in the AIS grade distribution between groups at admission.

In our study, the severity of both tSCI and canal compromise seem to be negatively associated with neurological recovery for at least two AIS grades at 6 months after SD, although the association did not quite reach the level of statistical significance. The influence of these two variables, however, might become significant if the sample were larger—i.e., if the power of the statistical test were larger. In group 8 h, we found an improvement of at least two AIS grades in 23% of patients with complete and 77% of patients with incomplete tSCI. Accordingly, a more pronounced effect of decompression in incomplete than in complete tSCI patients was found by others.²⁷ These findings are in concordance with animal studies, showing that pressure and time work hand in hand in perpetuating tSCI.^6
–8,32 As suggested by Carlson et al. the beneficial effect of the decompression of the spinal cord within the first hours after tSCI on neurological recovery might be mediated by the restoration of spinal cord blood flow.^33,34

In our study, the delay of SD was particularly characteristic for the patients transferred from other hospitals. Therefore, none of them could undergo SD within 8 h after tSCI. Accordingly, the patients referred from other hospitals represented 60% of group 8–24 h, which showed a significantly worse neurological recovery rate. These findings support the recommendations in the Guidelines for the Management of Acute Cervical Spine and Spinal Cord Injuries, published in 2013 by the American Association of Neurological Surgeons/Congress of Neurological Surgeons Joint Guidelines Committee, stating that patients should be transported from the site of injury to the nearest medical facility capable of providing definitive care, where they have better chances for neurologic improvement and experience less complications.³⁵

In our study, no statistically significant differences in the rate of perioperative complications, pneumonia, number of ventilator-dependent days, and mortality was found between the two study groups. Similarly, no adverse effects of early surgery were found in other studies.^12,36
–38

Study limitations

We are aware of the fact that a study with 42 enrolled patients may lack statistical power. We believe, however, that the uniform treatment of all patients in a single spinal center with the same rationale of surgical treatment, intensive care treatment, and rehabilitation gives our study additional power because even minimal discrepancies in the treatment protocol, very probably arising if more patients from several additional centers were involved, may have unpredicted consequences. We also recognize that all patients transferred from other hospitals were allocated to group 8–24 h and none of them to group 8 h. We believe, however, that no other factors than the delay in the time to SD influenced outcome in these patients, because all patients were submitted to the same preoperative protocol regarding patient resuscitation, methylprednisolone administration, and spinal immobilization.

We recognize that the accuracy of the admission neurological examination may represent a problem, especially in patients in spinal shock. We tried to address this issue with a second neurological examination 72 h after surgery as suggested by Brown and coworkers,³⁹ but as many as 69% of patients in group 8 h and 70% of patients in group 8–24 h needed ventilatory support after the operation at 72 h after SCI. The neurological examination in intubated patients proved to be unreliable and was not incorporated in the study results. On the other hand, Burns and associates⁴⁰ found that a reliable initial examination (within 48 h of SCI) may be as reliable as a second, later examination (a week after SCI) to determine complete SCI, if the first examination is performed in a cooperative patient. Moreover, spinal shock is known to occur in virtually all patients with severe SCI, beginning within minutes after injury and continuing for up to 12 months.⁴¹ Therefore, given that in all patients in our cohort the first neurological examination was performed in the first hours after injury, we do not believe that an eventual spinal shock would have any significant influence on the difference in neurological recovery between the study groups.

Conclusion

Our results suggest that patients with tSCI and persistent spinal cord compression from fractures or dislocations of the lower cervical spine have a higher rate of an at least two-grade AIS improvement and a higher improvement in AMS at 6 months follow-up if SD of the spinal cord is performed within 8 h versus from 8 to 24 h after tSCI, without any increase in the rate of perioperative complications, pneumonia, and the number of ventilator-dependent days or the mortality. Moreover, each additional hour elapsing from injury to SD reduces the odds of at least a two-grade AIS improvement. In our cohort, a delay of SD was particularly characteristic for the patients transferred from other hospitals; therefore, a direct transfer of all tSCI patients from the site of injury to a specialized spinal center is strongly recommended. To define which additional factors affect neurological recovery after tSCI, further studies are necessary.

Footnotes

Acknowledgments

We are indebted to Professor Janez Sketelj for his critical review and constructive comments on the manuscript.

Author Disclosure Statement

No competing financial interests exist.

References

Ackery

, Tator

, and Krassioukov

(2004). A global perspective on spinal cord injury epidemiology. J. Neurotrauma, 21, 1355–1370.

Fehlings

M.G.

, and Sekhon

(2000). Cellular, ionic and biomolecular mechanisms of the injury process, in: Contemporary Management of Spinal Cord Injury: From Impact to Rehabilitation. Benzel

, and Tator

C.H.

(eds). American Association of Neurological Surgeons: Chicago, pps. 33–50.

Oyinbo

C.A.

(2011). Secondary injury mechanisms in traumatic spinal cord injury: a nugget of this multiply cascade. Acta Neurobiol. Exp., 71, 281–299.

Furlan

J.C.

, Noonan

, Cadotte

D.W.

, and Fehlings

M.G.

(2011). Timing of decompressive surgery of spinal cord after traumatic spinal cord injury: an evidence-based examination of pre-clinical and clinical studies. J. Neurotrauma, 28, 1371–1399.

, Walker

C.L.

, Zhang

Y.P.

, Shields

C.B.

, and Xu

X.M.

(2014). Surgical decompression in acute spinal cord injury: A review of clinical evidence, animal model studies, and potential future directions of investigation. Front. Biol., 9, 127–136.

Dimar

J.R.

2nd , Glassman

S.D.

, Raque

G.H.

, Zhang

Y.P.

, and Shields

C.B.

(1999). The influence of spinal canal narrowing and timing of decompression on neurologic recovery after spinal cord contusion in a rat model. Spine (Phila Pa 1976), 24, 1623–1633.

Delamarter

R.B.

, Sherman

, and Carr

J.B.

(1995). Pathophysiology of spinal cord injury. Recovery after immediate and delayed decompression. J. Bone Joint Surg. Am., 77, 1042–1049.

Carlson

G.D.

, Gorden

C.D.

, Oliff

H.S.

, Pillai

J.J.

, and LaManna

J.C.

(2003). Sustained spinal cord compression: part I.: time-dependent effect on long-term pathophysiology. J. Bone Joint Surg. Am., 85-A, 86–94.

van Middendorp

J.J.

, Hosman

A.J.

, and Doi

S.A.

(2013). The effects of the timing of spinal surgery after traumatic spinal cord injury: a systematic review and meta-analysis. J. Neurotrauma, 30, 1781–1794.

10.

McLain

R.F.

, and Benson

D.R.

(1999). Urgent surgical stabilization of spinal fractures in polytrauma patients. Spine (Phila Pa 1976), 24, 1646–1654.

11.

McKinley

, Meade

M.A.

, Kirshblum

, and Barnard

(2004). Outcomes of early surgical management versus late or no surgical intervention after acute spinal cord injury. Arch. Phys. Med. Rehabil., 85, 1818–1825.

12.

Fehlings

M.G.

, Vaccaro

, Wilson

J.R.

, Singh

A.,W.

, Cadotte

, Harrop

J.S.

, Aarabi

, Shaffrey

, Dvorak

, Fisher

, Arnold

, Massicotte

E.M.

, Lewis

, and Rampersaud

(2012). Early versus delayed decompression for traumatic cervical spinal cord injury: results of the Surgical Timing in Acute Spinal Cord Injury Study (STASCIS). PloS One, 7, e32037.

13.

Wilson

J.R.

, Singh

, Craven

, Verrier

M.C.

, Drew

, Ahn

, Ford

, and Fehlings

M.G.

(2012). Early versus late surgery for traumatic spinal cord injury: the results of a prospective Canadian cohort study. Spinal Cord, 50, 840–843.

14.

Pointillart

, Petitjean

M.E.

, Wiart

, Vital

J.M.

, Lassié

, Thicoipé

, and Dabadie

(2000). Pharmacological therapy of spinal cord injury during the acute phase. Spinal Cord, 38, 71–76.

15.

Newton

, England

, Doll

, and Gardner

B.P.

(2011). The case for early treatment of dislocations of the cervical spine with cord involvement sustained playing rugby. J. Bone Joint Surg. Br., 93, 1646–1652.

16.

American Spinal Injury Association (Revised 2000). G. American Spinal Injury Association: Chicago.

17.

Fehlings

, Rao

S.C.

, Tator

C.H.

, Skaf

, Arnold

, Benzel

, Dickman

, Cuddy

, Green

, Hitchon

, Northrup

, Sonntag

, Wagner

, and Wilberger

(1999). The optimal radiologic method for assessing spinal canal compromise and cord compression in patients with cervical spinal cord injury. Part 2: Results of a multicenter study. Spine (Phila Pa 1976), 24, 605–613.

18.

Geisler

F.H.

, Coleman

W.P.

, Grieco

, and Poonian

(2001). The Sygen multicenter acute spinal cord injury study. Spine (Phila Pa 1976), 26, Suppl 24, S87–S98.

19.

Bracken

M.B.

, Shepard

M.J.

, Holford

T.R.

, Leo-Summers

, Aldrich

E.F.

, Fazl

, Fehlings

, Herr

D.L.

, Hitchon

P.W.

, Marshall

L.F.

, Nockels

R.P.

, Pascale

, Perot

P.L.

Jr. , Piepmeier

, Sonntag

V.K.

, Wagner

, Wilberger

J.E.

, Winn

H.R.

, and Young

(1997). Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA, 277, 1597–1604.

20.

Bracken

M.B.

, Shepard

M.J.

, Holford

T.R.

, Leo-Summers

, Aldrich

E.F.

, Fazl

, Fehlings

M.G.

, Herr

D.L.

, Hitchon

P.W.

, Marshall

L.F.

, Nockels

R.P.

, Pascale

, Perot

P.L.

Jr. , Piepmeier

, Sonntag

V.K.

, Wagner

, Wilberger

J.E.

, Winn

H.R.

, Young

(1998). Methylprednisolone or tirilazad mesylate administration after acute spinal cord injury: 1-year follow up. Results of the third National Acute Spinal Cord Injury randomized controlled trial. J. Neurosurg., 89, 699–706.

21.

Aebi

, Nazarian

(1987). Klassification der Halswirbelsäulenverletzungen (Classification of injuries of the cervical spine). (Ger) Orthopade, 16, 27–36.

22.

R Development Core Team (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. ISBN 3-900051-07-0, URL Available at: http://www.R-project.org/. Accessed: April 6, 2015 .

23.

Firth

(1993). Bias reduction of maximum likelihood estimates. Biometrika, 80, 27–38.

24.

Botel

, Glaser

, and Niedeggen

(1997). The surgical treatment of acute spinal paralysed patients. Spinal Cord, 35, 420–428.

25.

Tator

C.H.

, Fehlings

M.G.

, Thorpe

, and Taylor

(1999). Current use and timing of spinal surgery for management of acute spinal surgery for management of acute spinal cord injury in North America: results of a retrospective multicenter study. J. Neurosurg., 91, Suppl 1, 12–18.

26.

W.P.

, Fehlings

M.G.

, Cuddy

, Dickman

, Fazl

, Green

, Hitchon

, Northrup

, Sonntag

, Wagner

, and Tator

C.H.

(1999). Surgical treatment for acute spinal cord injury study pilot study #2: evaluation of protocol for decompressive surgery within 8 hours of injury. Neurosurg. Focus, 6, e3.

27.

La Rosa

, Conti

, Cardali

, Cacciola

, and Tomasello

(2004). Does early decompression improve neurological outcome of spinal cord injured patients? Appraisal of the literature using a meta-analytical approach. Spinal Cord, 42, 503–512.

28.

Fehlings

M.G.

, Rabin

, Sears

, Cadotte

D.W.

, and Aarabi

(2010). Current practice in the timing of surgical intervention in spinal cord injury. Spine (Phila Pa 1976), 35, Suppl 21, S166–S173.

29.

Fawcett

J.W.

, Curt

, Steeves

J.D.

, Coleman

W.P.

, Tuszynski

M.H.

, Lammertse

, Bartlett

P.F.

, Blight

A.R.

, Dietz

, Ditunno

, Dobkin

B.H.

, Havton

L.A.

, Ellaway

P.H.

, Fehlings

M.G.

, Privat

, Grossman

, Guest

J.D.

, Kleitman

, Nakamura

, Gaviria

, and Short

(2007). Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials. Spinal Cord, 45, 190–205.

30.

McCarthy

M.J.

, Gatehouse

, Steel

, Goss

, and Williams

(2011). The influence of the energy of trauma, the timing of decompression, and the impact of grade of SCI on outcome. Evid. Based Spine Care J., 2, 11–17.

31.

Hansebout

R.R.

, and Hansebout

C.R.

(2014). Local cooling for traumatic spinal cord injury: outcomes in 20 patients and review of the literature. J. Neurosurg. Spine, 20, 550–561.

32.

Batchelor

P.E.

, Wills

T.E.

, Skeers

, Battistuzzo

C.R.

, Macleod

M.R.

, Howells

D.W.

, and Sena

E.S.

(2013). Meta-analysis of pre-clinical studies of early decompression in acute spinal cord injury: a battle of time and pressure. PloS One, 8, e72659.

33.

Carlson

G.D.

, Warden

K.E.

, Barbeau

J.M.

, Bahniuk

, Kutina-Nelson

K.L.

, Biro

C.L.

, Bohlman

H.H.

, and LaManna

J.C.

(1997). Viscoelastic relaxation and regional blood flow response to spinal cord compression and decompression. Spine (Phila Pa 1976), 22, 1285–1291.

34.

Carlson

G.D.

, Minato

, Okada

, Gorden

C.D.

, Warden

K.E.

, Barbeau

J.M.

, Biro

C.L.

, Bahnuik

, Bohlman

H.H.

, and Lamanna

J.C.

(1997). Early time-dependent decompression for spinal cord injury: vascular mechanisms of recovery. J. Neurotrauma, 14, 951–962.

35.

Walters

B.C.

, Hadley

M.N.

, Hurlbert

R.J.

, Aarabi

, Dhall

S.S.

, Gelb

D.E.

, Harrigan

M.R.

, Rozelle

C.J.

, Ryken

T.C.

, and Theodore

(2013). Guidelines for the management of acute cervical spine and spinal cord injuries: 2013 update. Neurosurgery, 60, 82–91.

36.

Campagnolo

D.I.

, Esquieres

R.E.

, and Kopacz

K.J.

(1997). Effect of timing of stabilization on length of stay and medical complications following spinal cord injury. J. Spinal Cord Med., 20, 331–334.

37.

Croce

M.A.

, Bee

T.K.

, Pritchard

, Miller

P.R.

, and Fabian

T.C.

(2001). Does optimal timing for spine fracture fixation exist?. Ann. Surg., 233, 851–858.

38.

Mirza

S.K.

, Krengel

W.F.

3rd , Chapman

J.R.

, Anderson

P.A.

, Bailey

J.C.

, Grady

M.S.

, and Yuan

H.A.

(1999). Early versus delayed surgery for acute cervical spinal cord injury. Clin. Orthop. Relat. Res., 359, 104–114.

39.

Brown

P.J.

, Marino

R.J.

, Herbison

G.J.

, and Ditunno

J.F.

, (1991). The 72-h examination as a predictor of recovery in motor complete quadriplegia. Arch. Phys. Med. Rehabil., 72, 546–548.

40.

Burns

A.S.

, Lee

B.S.

, Ditunno

J.F.

Jr. , and Tessler

(2003). Patient selection for clinical trials: the reliability of the early spinal cord injury examination. J. Neurotrauma, 20, 477–482.

41.

Ditunno

J.F.

, Little

J.W.

, Tessler

, and Burns

A.S.

(2004). Spinal shock revisited: a four-phase model. Spinal Cord, 42, 383–395.