Age,Not MRI T2 Signal Intensity,Predicts Neurological Recovery After Surgery for Cervical Spinal Cord Injury Without Radiographic Evidence of Trauma

Abstract

Study Design

Retrospective Cohort Study.

Objectives

To evaluate the prognostic value of preoperative increased signal intensity (ISI) grade on T2-weighted magnetic resonance imaging (MRI) and to identify risk factors associated with poor neurological recovery after surgery in adult cervical spinal cord injury without radiographic evidence of trauma (SCIWORET).

Methods

A total of 128 consecutive SCIWORET patients who underwent surgical treatment between January 2016 and June 2023 were retrospectively analyzed. Patients were classified into 3 groups (Grade 0, 1, and 2) according to ISI grade on preoperative MRI. Neurological function was assessed using the Japanese Orthopaedic Association (JOA) score and recovery rate (RR). Multivariate logistic regression analysis identified predictors of poor RR (< 50%), and receiver operating characteristic (ROC) analysis determined the optimal age cutoff for prognosis.

Results

Signal intensity alterations were observed in 111 patients (86.7%). Higher ISI grades correlated with lower preoperative JOA scores (r = −0.303, P < 0.001) but not with postoperative RR (r = −0.067, P = 0.450). Multivariate analysis identified age as the only independent predictor of poor RR (OR = 1.10, 95% CI: 1.04-1.16, P < 0.001). ROC curve analysis yielded an optimal age cutoff of 58.5 years.

Conclusions

Preoperative ISI grade reflected the severity of spinal cord injury but did not predict postoperative neurological recovery. Older age (> 58.5 years) independently predicted poorer outcomes, highlighting the importance of integrating patient age into prognostic counseling and individualized surgical decision-making in SCIWORET.

Keywords

increased signal intensity spinal cord injury without radiographic evidence of trauma neurological recovery

Introduction

Cervical spinal cord injury without radiographic evidence of trauma (SCIWORET) is a distinct clinical entity characterized by neurological deficits occurring in the absence of spine fracture or dislocation on conventional radiological images, including X-rays and computed tomography (CT) scans.^1,2 Initially described in the pediatric population,³ SCIWORET has been increasingly recognized in adults, particularly those with preexisting cervical spinal stenosis or ossification of the posterior longitudinal ligament (OPLL), typically following minor trauma.^4-8 For patients presenting with severe symptoms or poor response to conservative treatment, surgical decompression remains the primary intervention to improve neurological function.⁹

Diagnosis of SCIWORET relies primarily on clinical manifestations and neurological examination, supported by radiographic evaluation. Owing to its superior soft-tissue contrast and lack of radiation exposure, magnetic resonance imaging (MRI) has become an indispensable tool for evaluating the extent of spinal cord injury.¹⁰ MRI can detect intramedullary abnormalities such as increased signal intensity (ISI) on T2-weighted images (T2WI), which may reflect edema, ischemia, or gliosis.^11,12 Although MRI T2 signal change is a well-recognized prognostic biomarker in degenerative cervical myelopathy,^13,14 its predictive value in adult SCIWORET remains uncertain and inconsistently reported.^7,15-17 These inconsistencies may stem from heterogeneity in study populations, variability in MRI timing, and the lack of a standardized grading system.

Yukawa et al¹⁸ proposed a T2-weighted MRI classification system that stratified ISI into 3 grades (Grade 0, 1, and 2). Despite its widespread adoption, its prognostic utility in adult SCIWORET has not been robustly validated in large cohorts.¹⁶ Therefore, this study aimed to systematically evaluate the association between preoperative ISI grade and postoperative neurological outcomes, and to identify risk factors for poor neurological recovery. We hypothesized that higher ISI grades would be associated with worse neurological recovery after surgical decompression.

Methods

Study Design

This retrospective cohort study included consecutive patients who underwent surgical treatment for cervical SCIWORET at our institution between January 2016 and June 2023. The diagnosis was established based on clinical symptoms, signs, and radiographic evaluations, including X-ray, CT, and MRI. The study protocol was approved by the institutional review board and conducted in accordance with the principles of the Declaration of Helsinki. Informed consent was waived due to the retrospective design of the study.

Patient Inclusion and Exclusion Criteria

Inclusion criteria were as follows: (1) age > 18 years; (2) a documented history of trauma; (3) neurological deficit without evidence of spinal fracture or dislocation on radiographic imaging; (4) absence of cervical-related symptoms or signs before trauma; and (5) a minimum follow-up of 2 years with complete medical records. Exclusion criteria were: (1) radiological evidence of trauma (fracture, dislocation, or subluxation); (2) concurrent major traumatic injuries outside the cervical spine; (3) history of cervical myelopathy before trauma; (4) spinal deformity, infection, tumor, or previous spinal surgery; (5) incomplete medical records or death before the final follow-up. Patients were stratified into 3 groups according to preoperative T2-weighted MRI findings: Grade 0 (no ISI change), Grade 1 (mild or obscure ISI), and Grade 2 (intense or bright ISI) (Figure 1).

Figure 1.

ISI grade on sagittal T2-weighted MRI. Representative sagittal T2-weighted MRI images showing different ISI grades: (A) Grade 0, none; (B) Grade 1, light or obscure; and (C) Grade 2, intense or bright. ISI: increased signal intensity; MRI: magnetic resonance imaging

Data Collection and Outcome Evaluation

Baseline demographic and clinical data were collected, including age, body mass index (BMI), comorbidities (hypertension and diabetes mellitus), presence of ossification of the posterior longitudinal ligament (OPLL), ISI grade, and symptom duration. Surgical details, including surgical approach (anterior or posterior), operative time, and estimated blood loss, were also recorded. Anterior approaches included anterior cervical discectomy and fusion (ACDF) and anterior cervical corpectomy and fusion (ACCF), whereas posterior approaches included laminoplasty and laminectomy. The surgical approach was selected based on imaging findings, lesion extent, site of compression, and severity of neurological impairment. Additionally, the interval between trauma and preoperative MRI acquisition was recorded for all patients. The median interval was 1.9 months (interquartile range, 0.8-5.0 months), with a range from 8 days to 28 months.

Neurological function was assessed using the Japanese Orthopaedic Association (JOA) scoring system for cervical myelopathy. The recovery rate (RR) was calculated according to Hirabayashi et al’s¹⁹ formula: RR = (postoperative JOA score − preoperative JOA score)/(17 − preoperative JOA score) × 100%. A poor outcome was defined as RR < 50%. The minimal clinically important difference (MCID) for RR was set at 52.8%, as described by Kato et al.²⁰

Statistical Analysis

Two independent observers evaluated preoperative ISI grade, discrepancies were resolved through consensus. Interrater reliability was assessed using the intraclass correlation coefficient (ICC), with ICC > 0.75 indicating excellent agreement.²¹ Statistical analyses were performed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism version 9.0 (GraphPad Software, San Diego, CA, USA). Continuous variables were expressed as mean ± standard deviation (SD), and categorical variables as counts (percentages). Differences among the 3 ISI grade groups were analyzed using one-way analysis of variance (ANOVA) or the Kruskal-Wallis H test, depending on data distribution. Bonferroni correction was applied for multiple comparisons, adjusting the significance threshold to P < 0.017 (0.05/3).²² Categorical variables were compared using the Chi-square test. Spearman’s rank correlation coefficient assessed correlations. Variables with P < 0.1 in univariate analysis or clinical relevance were entered into multivariate logistic regression to identify predictors of poor RR. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive value of age, and the Youden index was used to determine the optimal cutoff. A two-tailed P < 0.05 was considered statistically significant.

Results

Demographic Characteristics

A total of 161 consecutive patients with SCIWORET underwent cervical spine surgery performed by our team between January 2016 and June 2023. Of these, 128 patients were included in the final analysis. The demographic and clinical characteristics of the cohort were summarized in Table 1. The mean age was 54.5 ± 11.2 years (range, 18-80 years), with 97 males (75.8%) and 31 females (24.2%). The mean BMI was 25.1 ± 3.5 kg/m². Hypertension and diabetes mellitus were observed in 27.3% and 20.3% of patients, respectively. OPLL was present in 60 patients (46.9%). Preoperative MRI revealed ISI changes in 111 patients (86.7%), including 63 (49.2%) with Grade 1, and 48 (37.5%) with Grade 2 ISI. Interobserver reliability for ISI grade was excellent (ICC = 0.974, 95% CI: 0.963-0.982, P < 0.001). The mean follow-up duration was 69.6 ± 27.2 months. The Mean preoperative and final follow-up JOA scores were 11.4 ± 2.8 and 14.6 ± 2.2, respectively. Yielding an average improvement of 3.2 ± 1.9 points and a mean RR of 57.8% ± 25.1%. 44 patients (34.4%) underwent anterior approaches (ACDF or ACCF), and 84 patients (65.6%) underwent posterior approaches (laminoplasty or laminectomy). At final follow-up, 97 patients (75.8%) achieved good neurological recovery, and 78 patients (60.9%) reached the MCID threshold for RR.

Table 1.

Demographic Characteristics of the Cohort

Variables	Values (N = 128)
Age (years)	54.5 ± 11.2
Sex
Male	97 (75.8%)
Female	31 (24.2%)
BMI (kg/m²)	25.1 ± 3.5
Hypertension	35 (27.3%)
Diabetes mellitus	26 (20.3%)
OPLL	60 (46.9%)
ISI grade
Grade 0	17 (13.3%)
Grade 1	63 (49.2%)
Grade 2	48 (37.5%)
Follow-up duration (months)	69.6 ± 27.2
Surgical approach
Anterior (ACDF or ACCF)	44 (34.4%)
Posterior (laminoplasty or laminectomy)	84 (65.6%)
Preoperative JOA score	11.4 ± 2.8
Final follow-up JOA score	14.6 ± 2.2
JOA score improvement	3.2 ± 1.9
RR (%)	57.8 ± 25.1
Poor RR	31 (24.2%)
Achieve MCID for RR	78 (60.9%)

BMI: body mass index; OPLL: ossification of the posterior longitudinal ligament; ISI: increased signal intensity; ACDF: anterior cervical discectomy and fusion; ACCF: anterior cervical corpectomy and fusion; JOA: Japanese Orthopaedic Association; RR: recovery rate; MCID: minimal clinically important difference.

Comparisons Among Different ISI Grades

As shown in Table 2, significant differences were observed among ISI groups in symptom duration, surgical approach, and JOA score. Symptom duration > 1 month was more frequent in the Grade 1 (55.6%) and Grade 2 (60.4%) groups than in the Grade 0 (23.5%) group. Posterior approaches predominated in the Grade 1 (69.8%) and Grade 2 (70.8%) groups, whereas anterior approaches were more common in the Grade 0 (64.7%) group. Preoperative JOA score was significantly higher in the Grade 0 group compared with the Grade 2 group (12.8 ± 1.9 vs 10.6 ± 3.0, P < 0.001). Similarly, the final JOA score differed significantly between the Grade 0 and Grade 2 groups (15.5 ± 1.1 vs 14.0 ± 3.0, P = 0.008). However, JOA improvement and RR did not differ significantly among the 3 groups (both P > 0.05). Across all groups, JOA score improved significantly after surgery (Figure 2). The proportion of patients achieving MCID for RR did not differ among groups (70.6%, 58.7%, and 60.4%; P = 0.670).

Table 2.

Comparisons of Demographics and Neurological Outcomes Among Different ISI Grades

	Grade 0 (N = 17)	Grade 1 (N = 63)	Grade 2 (N = 48)	P
Age (years)	50.7 ± 9.6	54.8 ± 12.1	55.4 ± 10.2	0.316
Sex				0.987
Male	13 (76.5%)	48 (76.2%)	36 (75.0%)
Female	4 (23.5%)	15 (23.8%)	12 (25.0%)
BMI (kg/m²)	25.2 ± 3.1	24.5 ± 3.6	25.9 ± 3.4	0.104
Hypertension	2 (11.8%)	17 (27.0%)	16 (33.3%)	0.229
Diabetes mellitus	5 (29.4%)	14 (22.2%)	7 (14.6%)	0.371
OPLL	9 (52.9%)	25 (39.7%)	26 (54.2%)	0.275
Surgical approach				0.018^*
Anterior (ACDF or ACCF)	11 (64.7%)	19 (30.2%)	14 (29.2%)
Posterior (laminoplasty or laminectomy)	6 (35.3%)	44 (69.8%)	34 (70.8%)
Symptom duration > 1 month	4 (23.5%)	35 (55.6%)	29 (60.4%)	0.028^*
Operative time (min)	100.8 ± 53.1	102.1 ± 32.3	90.1 ± 27.2	0.169
Estimated blood loss (mL)	128.2 ± 124.3	163.9 ± 150.0	141.6 ± 125.7	0.550
Follow-up duration (months)	82.6 ± 24.4	65.4 ± 26.7	70.7 ± 27.8	0.054
Length of hospital stay (days)	4.9 ± 1.9	5.3 ± 2.9	5.2 ± 2.4	0.991
Preoperative JOA score	12.8 ± 1.9	11.6 ± 2.6	10.6 ± 3.0	0.002^**
Follow-up JOA score	15.5 ± 1.1	14.8 ± 1.5	14.0 ± 3.0	0.035^*
JOA score improvement	2.7 ± 1.5	3.2 ± 2.1	3.4 ± 1.9	0.459
RR (%)	63.7 ± 25.4	57.4 ± 26.6	56.3 ± 23.1	0.673
Achieve MCID for RR	12 (70.6%)	37 (58.7%)	29 (60.4%)	0.670

ISI: increased signal intensity; BMI: body mass index; OPLL: ossification of the posterior longitudinal ligament; JOA: Japanese Orthopaedic Association; RR: recovery rate; MCID: minimal clinically important difference. ^* represents P < 0.05, ^** represents P < 0.01, ^*** represents P < 0.001.

Figure 2.

Changes in JOA scores between preoperative and final follow-up across different ISI grades. Comparison of preoperative and final follow-up JOA scores in patients stratified by ISI grade. JOA: Japanese Orthopaedic Association; ISI: increased signal intensity

Correlation Analysis

As summarized in Table 3. ISI grade was negatively correlated with preoperative JOA score (r = −0.303, P < 0.001) and final follow-up JOA score (r = −0.212, P = 0.016), although the correlation coefficients were low. No significant correlations were found between ISI grade and either JOA score improvement (r = 0.107, P = 0.229) or RR (r = −0.067, P = 0.450).

Table 3.

Correlation Analysis Between Neurological Outcomes and ISI Grade

VS. ISI grade	r	P
Preoperative JOA score	−0.303	<0.001^***
Final follow-up JOA score	−0.212	0.016^*
JOA score improvement	0.107	0.229
RR	−0.067	0.450

ISI: increased signal intensity; JOA: Japanese Orthopedic Association; RR: recovery rate. ^* represents P < 0.05, ^** represents P < 0.01, ^*** represents P < 0.001.

Predictors of Poor Recovery Rate

Univariate and multivariate logistic regression analyses identified predictors of poor RR (Table 4). In univariate analysis, age (P < 0.001) and surgical approach (P = 0.007) were associated with poor RR. After multivariate adjustment, only age (OR = 1.10, 95% CI: 1.04-1.16, P < 0.001) remained an independent predictor. ROC curve analysis demonstrated that age had good discriminatory power for poor RR (area under the curve, AUC = 0.755, 95% CI: 0.659-0.851, P < 0.001). The optimal cutoff determined by the Youden index was 58.5 years yielding a sensitivity of 67.7% and specificity of 73.2% (Figure 3).

Table 4.

Univariate and Multivariate Logistic Regression Analyses for Poor RR

	Univariate analysis		Multivariate analysis
	OR (95% CI)	P	OR (95% CI)	P
Age (years)	1.11 (1.05-1.17)	<0.001^***	1.10 (1.04-1.16)	<0.001^***
Sex
Male	Reference
Female	1.39 (0.56-3.46)	0.474
BMI (kg/m²)	0.94 (0.84-1.06)	0.348	0.89 (0.77-1.04)	0.135
Hypertension	0.90 (0.36-2.26)	0.825
Diabetes mellitus	1.93 (0.76-4.91)	0.170	1.62 (0.53-4.98)	0.398
OPLL	1.28 (0.57-2.89)	0.544	0.91 (0.34-2.42)	0.853
ISI grade
Grade 0	Reference
Grade 1	3.00 (0.62-14.47)	0.171	1.98 (0.35-11.32)	0.441
Grade 2	2.23 (0.44-11.29)	0.332	1.96 (0.32-12.05)	0.470
Preoperative JOA score	1.00 (0.86-1.06)	0.993	1.08 (0.90-1.29)	0.398
Surgical approach
Anterior	Reference
Posterior	7.92 (1.77-35.39)	0.007^**	1.44 (0.47-4.41)	0.527

RR: recovery rate; OR: odds ratio; CI: confidence interval; BMI: body mass index; OPLL: ossification of the posterior longitudinal ligament; ISI: increased signal intensity; JOA: Japanese Orthopaedic Association. ^* represents P < 0.05, ^** represents P < 0.01, ^*** represents P < 0.001.

Figure 3.

ROC curve for age predicting poor recovery rate. ROC curve showing the predictive ability of age for poor recovery rate. ROC: receiver operating characteristic; AUC: area under the curve

Subgroup Analysis by Age and Surgical Approach

Subgroup analysis based on the optimal age cutoff (Table 5) revealed that younger patients achieved significantly higher final JOA scores (15.0 ± 2.0 vs 14.0 ± 2.4, P = 0.003), greater JOA score improvement (3.6 ± 1.9 vs 2.6 ± 1.9, P = 0.004), and higher RR (65.2 ± 20.6 vs 46.0 ± 26.5, P < 0.001) compared with older patients. Younger patients also had a higher rate of achieving MCID (67.9% vs 47.8%, P = 0.026). Comparative analysis by surgical approach (Table 6) showed that anterior approaches were more frequently performed in younger patients and were associated with shorter operative time, less estimated blood loss, and a lower incidence of poor RR (13.6% vs 29.8%, P = 0.043). Although RR was higher in the anterior group, the difference was not statistically significant (P = 0.214).

Table 5.

Subgroup Analysis According to the Optimal Age Cutoff for Neurological Outcomes

	Younger age (<58.5 years) (N = 81)	Older age (>58.5 years) (N = 47)	P
Age (years)	48.2 ± 8.7	65.4 ± 5.0	<0.001^***
Preoperative JOA score	11.4 ± 2.7	11.4 ± 2.9	0.920
Final follow-up JOA score	15.0 ± 2.0	14.0 ± 2.4	0.003^**
JOA score improvement	3.6 ± 1.9	2.6 ± 1.9	0.004^**
RR (%)	65.2 ± 20.6	46.0 ± 26.5	<0.001^***
Achieve MCID for RR	55 (67.9%)	22 (47.8%)	0.026^*
Poor RR	10 (12.4%)	21 (44.7%)	<0.001^***

OPLL: ossification of the posterior longitudinal ligament; ISI: increased signal intensity; JOA: Japanese Orthopaedic Association; RR: recovery rate; MCID: minimal clinically important difference. ^* represents P < 0.05, ^** represents P < 0.01, ^*** represents P < 0.001.

Table 6.

Comparative Analysis Between Anterior and Posterior Approaches

	Anterior approaches	Posterior approaches	P
Age (years)	49.4 ± 10.9	57.1 ± 10.4	<0.001^***
Hypertension	8 (18.2%)	27 (32.1%)	0.092
Diabetes mellitus	7 (15.9%)	19 (22.6%)	0.370
OPLL	15 (34.1%)	45 (53.6%)	0.036^*
ISI grade			0.018^*
Grade 0	11 (25.0%)	6 (7.1%)
Grade 1	19 (43.2%)	44 (52.4%)
Grade 2	14 (31.8%)	34 (40.5%)
Preoperative JOA score	12.1 ± 2.4	11.0 ± 2.9	0.011^*
Final follow-up JOA score	15.3 ± 1.1	14.2 ± 2.5	0.010^*
JOA score improvement	3.2 ± 2.0	3.2 ± 1.9	0.599
RR (%)	62.8 ± 22.7	55.2 ± 25.9	0.214
Poor RR	6 (13.6%)	25 (29.8%)	0.043^*
Operative time (min)	82.3 ± 40.4	105.3 ± 27.6	<0.001^***
Estimated blood loss (mL)	71.5 ± 106.9	192.3 ± 134.2	<0.001^***

OPLL: ossification of the posterior longitudinal ligament; ISI: increased signal intensity; JOA: Japanese Orthopaedic Association; RR: recovery rate. ^* represents P < 0.05, ^** represents P < 0.01, ^*** represents P < 0.001.

MRI Timing and Subgroup Analysis

Most patients (89.1%) underwent preoperative MRI more than 2 weeks after trauma, whereas only 14 patients (10.9%) received MRI during 8 to 14 days after trauma. A subgroup comparison between early (≤ 2 weeks) and delayed (> 2 weeks) MRI demonstrated no statistically significant differences in age, sex, BMI, preoperative JOA score, ISI grade, operation time, estimated blood loss, and neurological outcomes (P > 0.05) (Table 7).

Table 7.

Comparative Analysis Between Early and Delayed MRI

	Early MRI (N = 14)	Delayed MRI (N = 114)	P
Age (years)	52.0 ± 6.4	55.0 ± 11.4	0.138
Sex			0.556
Male	12 (85.7%)	85 (74.6%)
Female	2 (14.3%)	29 (25.4%)
BMI (kg/m²)	24.2 ± 3.2	25.2 ± 3.5	0.310
ISI grade			0.719
Grade 0	1 (7.1%)	16 (14.1%)
Grade 1	8 (57.1%)	55 (48.3%)
Grade 2	5 (35.7%)	43 (37.7%)
MRI timing (months)	0.3 ± 0.0	4.5 ± 5.9	<0.001^***
Operative time (min)	110.9 ± 50.3	95.8 ± 31.6	0.544
Estimated blood loss (mL)	231.4 ± 236.2	140.9 ± 118.3	0.364
Preoperative JOA score	10.2 ± 4.1	11.5 ± 2.5	0.299
Final follow-up JOA score	13.8 ± 3.8	14.7 ± 1.9	0.694
JOA score improvement	3.6 ± 2.1	3.2 ± 1.9	0.444
RR (%)	56.0 ± 25.7	58.0 ± 25.1	0.954
Poor RR	2 (14.3%)	29 (25.4%)	0.556
Achieve MCID for RR	8 (57.1%)	69 (60.5%)	0.807

MRI: magnetic resonance imaging; BMI: body mass index; ISI: increased signal intensity; JOA: Japanese Orthopaedic Association; RR: recovery rate; MCID: minimal clinically important difference. ^* represents P < 0.05, ^** represents P < 0.01, ^*** represents P < 0.001.

Discussion

MRI has become indispensable in evaluating SCIWORET, owing to its superior soft-tissue contrast and ability to delineate intramedullary lesions.¹¹ T2-weighted sagittal and axial images are particularly valuable for assessing the extent of spinal cord injury.²³ However, the relationship between ISI grade on MRI and postoperative neurological outcomes in SCIWORET remains incompletely understood.^15,24 In this retrospective cohort of 128 surgical SCIWORET patients, preoperative MRI ISI grade correlated with baseline neurological severity but failed to predict postoperative neurological recovery, whereas age emerged as the only independent prognostic factor.

Signal intensity alterations were observed in 111 patients (86.7%) in our study. Higher ISI grades were associated with lower preoperative JOA scores, reflecting more severe initial spinal cord injury. This supported previous histopathologic and imaging evidence suggesting that T2 hyperintensity represented edema, ischemia, gliosis, or demyelination.²⁵ Our findings were consistent with Machino et al,¹⁵ who reported that ISI grade correlated with baseline neurological status. However, ISI grade did not influence postoperative neurological outcome in our cohort. In our cohort, the Grade 0 group demonstrated better final follow-up JOA score, whereas the Grade 1 and Grade 2 groups showed comparable outcomes. RR was comparable among the 3 groups. Prior studies reported divergent results, with some suggesting that higher ISI grade predicted worse outcomes,^15,16,26 whereas others found no correlation.²⁷ These discrepancies may be explained by the timing of MRI acquisition. Early MRI (especially within 3 days after trauma) may underestimate the extent of spinal cord injury and has been shown to correlate poorly with subsequent neurological outcomes.^27,28 Instead, delayed MRI tends to capture more stable and mature pathological changes.⁷ Moreover, Ouchida et al⁷ demonstrated that many Grade 1 lesions evolved into Grade 2 on delayed MRI obtained after 2 weeks, which was consistent with the broader traumatic spinal cord injury, where MRI T2 hyperintensity showed variable prognostic value depending on timing and injury phase.^23,29 The MRI scans in our cohort were performed significantly later than those in many previous studies which primarily focused on the hyper-acute phase (within 1 week).^15,16,30 In our cohort, only 14 patients (10.9%) underwent MRI within 2 weeks after trauma (with the earliest at 8 days), while the vast majority (89.1%) were scanned beyond 2 weeks after trauma. Furthermore, our subgroup analysis revealed no significant differences between early and delayed MRI groups, indicating that the signal alterations we observed represented more stabilized pathological changes. It should be noted that ISI on T2WI reflects a broad spectrum of compressive pathologies with variable recuperative potential. It is a nonspecific manifestation that may indicate either reversible changes such as edema and ischemia or irreversible damage including gliosis and myelomalacia. When the signal intensity represents minor pathological changes that resolve after decompression, it is unlikely to serve as an independent prognostic factor.³¹

Age demonstrated a strong and independent association with recovery. Patients older than 58.5 years exhibited significantly lower RR and were less likely to reach MCID thresholds than younger individuals. These findings align with previous reports suggesting that aging adversely affects spinal cord recovery. Qi et al³² revealed that age negatively influenced neurological prognosis in SCIWORET. Kwon et al³⁰ similarly identified older age as a negative predictor of neurological recovery in SCIWORET patients with OPLL. Age-related reductions in vascular perfusion, neural plasticity, and regenerative capacity may impair recovery following surgical decompression.^33,34 Moreover, elderly patients often have multilevel spondylosis, OPLL, or diffuse idiopathic skeletal hyperostosis,² which narrow the spinal canal and exacerbate spinal cord compression even under minor trauma.^35,36

In our cohort, although anterior procedures appeared to yield better neurological recovery in the unadjusted analysis, this observation must be interpreted cautiously. Posterior decompression was performed predominantly in older patients with multilevel stenosis or OPLL, whereas anterior approaches were more common in younger individuals with focal disc compression. Age and underlying cervical pathology are well-recognized determinants of neurological recovery independent of surgical technique.^37,38 After adjusting for these covariates in the multivariate analysis, surgical approach was no longer an independent predictor of outcome. Therefore, the apparent superiority of anterior surgery in unadjusted comparisons likely reflected differences in baseline patient characteristics rather than a causal effect of the surgical approach itself. Patient-specific factors, particularly age, may outweigh surgical technique in determining recovery potential in SCIWORET.

Several limitations should be acknowledged. First, its retrospective design may introduce selection bias, particularly regarding surgical approach. Second, our qualitative ISI grade criteria may lack the sensitivity of quantitative signal intensity measurements. Third, the single-center design may limit the generalizability. Finally, we did not account for potential confounding by rehabilitation protocols, which could influence long-term outcomes. Future studies employing quantitative imaging metrics and multicenter prospective designs are warranted to validate these findings.

Despite these limitations, our study provided valuable insight into the prognostic role of MRI findings and patient factors in SCIWORET. ISI grade reflected baseline spinal cord impairment and should not be used alone as a determinant of surgical decision-making. In contrast, age should be considered during perioperative counseling, expectation setting, and surgical decision-making.

Conclusions

Our study demonstrated that although MRI findings reflected the severity of spinal cord injury in patients with SCIWORET, preoperative ISI grade did not predict postoperative neurological recovery. Instead, age was identified as the most important independent risk factor for poor neurological outcomes following surgical decompression. Patients older than 58.5 years exhibited significantly lower recovery rates and were less likely to achieve clinically meaningful improvement. These findings could assist clinicians in providing more accurate prognostic counseling and in tailoring surgical strategies to individual patient characteristics.

Footnotes

ORCID iDs

Juncai Lei

Panpan Hu

Xiaoguang Liu

Ethical Considerations

This study was approved by the Ethics Committee of Peking University Third Hospital (No. S20250729) and conducted in accordance with the principles of the Declaration of Helsinki.

Consent to Participate

Informed consent was waived due to the retrospective nature of the study.

Consent for Publication

All identifiable individual data presented in this article were obtained with informed consent from the participants.

Author Contributions

Juncai Lei, Panpan Hu, and Yan Li contributed to the study design, data analysis, and manuscript drafting. Xiaoguang Liu and Feng Wei conceived the study, supervised data analysis, and revised the manuscript. Hua Zhou and Zhongjun Liu contributed to study design and statistical analysis. Fengliang Wu, Ben Wang, Lei Dang, Yanchao Tang, Liang Jiang, and Xiao Liu assisted in data collection and analysis. All authors read and approved the final manuscript. Juncai Lei, Panpan Hu, and Yan Li contributed equally and share first authorship. Xiaoguang Liu and Feng Wei shared corresponding authorship.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Natural Science Foundation of China (No. 82372451) and Beijing Natural Science Foundation (No. 4252025).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated and analyzed during the current study are available from Peking University Third Hospital. However, restrictions apply to the availability of these data, which were used under institutional license and are not publicly accessible.*

References

Wyndaele

. SCIWORA. Spinal Cord. 2016;54(10):755. doi:10.1038/sc.2016.141

Nakamura

Teraguchi

Nakata

, et al. Associated factors of cervical spinal cord injury without radiographic evidence of trauma: a retrospective study. Cureus. 2025;17(3):e80006. doi:10.7759/cureus.80006

Pang

Wilberger

. Spinal cord injury without radiographic abnormalities in children. J Neurosurg. 1982;57(1):114-129. doi:10.3171/jns.1982.57.1.0114

Yang

Wang

Zhang

Liu

Cao

. Spinal cord injury without radiographic abnormality (SCIWORA) in Tianjin, China: a single-center report of 101 cases. Eur J Med Res. 2025;30(1):361. doi:10.1186/s40001-025-02627-y

Cao

Tang

Jia

Chen

. Risk factors for poor prognosis of spinal cord injury without radiographic abnormality associated with cervical ossification of the posterior longitudinal ligament. BioMed Res Int. 2022;2022:1572341. doi:10.1155/2022/1572341

Kimura

Shiraishi

Sawamura

Sugawara

Inoue

Takeshita

. Ossification of the anterior longitudinal ligament affects the severity and distribution of neurological deficits following spinal cord injury without radiological abnormality. J Orthop Sci. 2024;29(3):718-725. doi:10.1016/j.jos.2023.03.006

Ouchida

Yukawa

Ito

, et al. Delayed magnetic resonance imaging in patients with cervical spinal cord injury without radiographic abnormality. Spine. 2016;41(16):E981-E986. doi:10.1097/BRS.0000000000001505

Koyanagi

Iwasaki

Hida

Akino

Imamura

Abe

. Acute cervical cord injury without fracture or dislocation of the spinal column. J Neurosurg. 2000;93(1 Suppl):15-20. doi:10.3171/spi.2000.93.1.0015

Fehlings

Tetreault

Hachem

, et al. An update of a clinical practice guideline for the management of patients with acute spinal cord injury: recommendations on the role and timing of decompressive surgery. Glob Spine J. 2024;14(3_suppl):174S-186S. doi:10.1177/21925682231181883

10.

Ohshio

Hatayama

Kaneda

Takahara

Nagashima

. Correlation between histopathologic features and magnetic resonance images of spinal cord lesions. Spine. 1993;18(9):1140-1149. doi:10.1097/00007632-199307000-00005

11.

Guan

Zheng

, et al. A critical appraisal of clinical practice guidelines for diagnostic imaging in the spinal cord injury. Spine J. 2023;23(8):1189-1198. doi:10.1016/j.spinee.2023.03.003

12.

Wei

Cao

Tian

Yuan

. Clinical and radiologic results of anterior cervical discectomy and fusion for cervical spondylotic myelopathy in elderly patients with T2-weighted increased signal intensity. World Neurosurg. 2018;112:e520-e526. doi:10.1016/j.wneu.2018.01.071

13.

Zhang

Shen

Wang

Ding

. Magnetic resonance T2 image signal intensity ratio and clinical manifestation predict prognosis after surgical intervention for cervical spondylotic myelopathy. Spine. 2010;35(10):E396-399. doi:10.1097/BRS.0b013e3181c6dbc4

14.

Tetreault

Kopjar

Vaccaro

, et al. A clinical prediction model to determine outcomes in patients with cervical spondylotic myelopathy undergoing surgical treatment: data from the prospective, multi-center AOSpine North America study. J Bone Joint Surg Am. 2013;95(18):1659-1666. doi:10.2106/JBJS.L.01323

15.

Machino

Ando

Kobayashi

, et al. MR T2 image classification in adult patients of cervical spinal cord injury without radiographic abnormality: a predictor of surgical outcome. Clin Neurol Neurosurg. 2019;177:1-5. doi:10.1016/j.clineuro.2018.12.010

16.

Machino

Yukawa

Ito

, et al.

Can magnetic resonance imaging reflect the prognosis in patients of cervical spinal cord injury without radiographic abnormality?

Spine. 2011;36(24):E1568-1572. doi:10.1097/BRS.0b013e31821273c0

17.

Segi

Nakashima

Ito

, et al. Early versus delayed surgery for elderly traumatic cervical spinal injury: a nationwide multicenter study in Japan. Glob Spine J. 2024;16:21925682241227430. doi:10.1177/21925682241227430

18.

Yukawa

Kato

Yoshihara

Yanase

Ito

. MR T2 image classification in cervical compression myelopathy: predictor of surgical outcomes. Spine. 2007;32(15):1675-1678. doi:10.1097/BRS.0b013e318074d62e

19.

Hirabayashi

Miyakawa

Satomi

Maruyama

Wakano

. Operative results and postoperative progression of ossification among patients with ossification of cervical posterior longitudinal ligament. Spine. 1981;6(4):354-364. doi:10.1097/00007632-198107000-00005

20.

Kato

Oshima

Matsubayashi

Taniguchi

Tanaka

Takeshita

. Minimum clinically important difference and patient acceptable symptom state of Japanese orthopaedic association score in degenerative cervical myelopathy patients. Spine. 2019;44(10):691-697. doi:10.1097/BRS.0000000000002928

21.

Zhou

Lin

Zhang

, et al. Correlation and reliability of cervical sagittal alignment parameters between plain radiographs and multipositional MRI images. Spinal Cord. 2023;61(5):307-312. doi:10.1038/s41393-023-00895-1

22.

Perneger

. What’s wrong with Bonferroni adjustments. BMJ. 1998;316(7139):1236-1238. doi:10.1136/bmj.316.7139.1236

23.

Freund

Seif

Weiskopf

, et al. MRI in traumatic spinal cord injury: from clinical assessment to neuroimaging biomarkers. Lancet Neurol. 2019;18(12):1123-1135. doi:10.1016/S1474-4422(19)30138-3

24.

Matsushita

Maeda

Mori

, et al.

Can the acute magnetic resonance imaging features reflect neurologic prognosis in patients with cervical spinal cord injury?

Spine J. 2017;17(9):1319-1324. doi:10.1016/j.spinee.2017.05.009

25.

Zhong

Wang

Huang

Luo

. Risk factors for rapid progressive neurological deterioration in patients with cervical spondylotic myelopathy. J Orthop Surg Res. 2021;16(1):75. doi:10.1186/s13018-021-02227-6

26.

Martinez-Perez

Munarriz

Paredes

Cotrina

Lagares

. Cervical spinal cord injury without computed tomography evidence of trauma in adults: magnetic resonance imaging prognostic factors. World Neurosurg. 2017;99:192-199. doi:10.1016/j.wneu.2016.12.005

27.

Liu

Zhao

Wang

. Early MRI finding in adult spinal cord injury without radiologic abnormalities does not correlate with the neurological outcome: a retrospective study. Spinal Cord. 2015;53(10):750-753. doi:10.1038/sc.2015.45

28.

Boese

Nerlich

Klein

Wirries

Ruchholtz

Lechler

. Early magnetic resonance imaging in spinal cord injury without radiological abnormality in adults: a retrospective study. J Trauma Acute Care Surg. 2013;74(3):845-848. doi:10.1097/TA.0b013e31828272e9

29.

Miyanji

Furlan

Aarabi

Arnold

Fehlings

. Acute cervical traumatic spinal cord injury: MR imaging findings correlated with neurologic outcome--prospective study with 100 consecutive patients. Radiology. 2007;243(3):820-827. doi:10.1148/radiol.2433060583

30.

Kwon

Shin

Lee

Cho

. Prognostic factors for surgical outcome in spinal cord injury associated with ossification of the posterior longitudinal ligament (OPLL). J Orthop Surg Res. 2015;10:94. doi:10.1186/s13018-015-0235-3

31.

Tetreault

Dettori

Wilson

, et al. Systematic review of magnetic resonance imaging characteristics that affect treatment decision making and predict clinical outcome in patients with cervical spondylotic myelopathy. Spine. 2013;38(22 Suppl 1):S89-110. doi:10.1097/BRS.0b013e3182a7eae0

32.

Cao

Xia

, et al. Does cervical curvature affect neurological outcome after incomplete spinal cord injury without radiographic abnormality (SCIWORA): 1-year follow-up. J Orthop Surg Res. 2022;17(1):361. doi:10.1186/s13018-022-03254-7

33.

Nouri

Tetreault

Singh

Karadimas

Fehlings

. Degenerative cervical myelopathy: epidemiology, genetics, and pathogenesis. Spine. 2015;40(12):E675-693. doi:10.1097/BRS.0000000000000913

34.

Rozzelle

Aarabi

Dhall

, et al. Spinal cord injury without radiographic abnormality (SCIWORA). Neurosurgery. 2013;72(Suppl 2):227-233. doi:10.1227/NEU.0b013e3182770ebc

35.

Morishita

Naito

Hymanson

Miyazaki

Wang

. The relationship between the cervical spinal canal diameter and the pathological changes in the cervical spine. Eur Spine J. 2009;18(6):877-883. doi:10.1007/s00586-009-0968-y

36.

Fehlings

Tetreault

Riew

, et al. A clinical practice guideline for the management of patients with degenerative cervical myelopathy: recommendations for patients with mild, moderate, and severe disease and nonmyelopathic patients with evidence of cord compression. Glob Spine J. 2017;7(3 Suppl):70S-83S. doi:10.1177/2192568217701914

37.

Chen

Liu

Dong

, et al. A comparison of the anterior approach and the posterior approach in treating multilevel cervical myelopathy: a meta-analysis. Clin Spine Surg. 2017;30(2):65-76. doi:10.1097/BSD.0000000000000398

38.

Nouri

Martin

Nater

, et al. Influence of magnetic resonance imaging features on surgical decision-making in degenerative cervical myelopathy: results from a global survey of AOSpine international members. World Neurosurg. 2017;105:864-874. doi:10.1016/j.wneu.2017.06.025