Benefits of Early Surgical Treatment for Patients with Multilevel Cervical Canal Stenosis of Acute Traumatic Central Cord Syndrome

Abstract

Introduction

Currently, there exists considerable debate surrounding the optimal treatment approaches for different subtypes of patients with spinal cord injury (SCI). The purpose of this study was to conduct a comparative analysis of the benefits associated with conservative treatment and treatments with different surgical periods for patients diagnosed with acute traumatic central cord syndrome (ATCCS) and multilevel cervical canal stenosis (CCS).

Methods

A retrospective cohort study was conducted, and 93 patients who met inclusion and exclusion criteria in our hospital between 2015 and 2020 were followed for a minimum duration of 2 years. Among them, 30 patients (Group A) received conservative treatment, 18 patients (Group B) received early surgery (≤7 days), and 45 patients (Group C) received late surgery (>7 days). The American Spinal Injury Association (ASIA) grade, Japanese Orthopedic Association (JOA) score, and recovery rate (RR) were evaluated. Multivariate linear regression was used to analyze prognostic determinants. Cost‐utility analysis was performed based on the EQ‐5D scale.

Results

The ASIA grade, JOA score, and RR of all three groups improved compared with the previous evaluation (P < 0.05). During follow‐up, the ASIA grade, JOA score, and RR of Group B were all better than for Group A and Group C (P < 0.05), while there was no significant difference between Group A and C (P > 0.05). The EQ‐5D scale in Group B was optimal at the last follow‐up. The incremental cost‐utility ratio (ICUR) of Group A was the lowest, while that of Group B compared to Group A was less than the threshold of patients’ willingness to pay. Age, initial ASIA grade, and treatment types significantly affected the outcomes.

Conclusions

Both conservative and surgical treatments yield good results. Compared with patients who received conservative treatment and late surgery, patients who received early surgery had better clinical function and living quality. Despite the higher cost, early surgery is cost‐effective when compared to conservative treatment. Younger age, initial better ASIA grade, and earlier surgery were associated with better prognosis.

This is a retrospective cohort study, the purpose of which is to compare the benefits associated with conservative treatment and treatments with different surgical periods for patients diagnosed with acute traumatic central cord syndrome (ATCCS) and multilevel cervical canal stenosis (CCS). From 2015 to 2020, 93 patients with ATCCS and CCS in two segments or above without cervical fracture or dislocation were analyzed. The American Spinal Cord Injury Association (ASIA) grade, the Japanese Orthopedic Association (JOA) score, and the JOA score recovery rate (RR) were evaluated. Multivariate linear regression was used to analyze prognostic determinants. Cost‐utility analysis was performed for each group based on the EQ‐5D scale, which was used to assess the living quality at the last follow‐up. After at least 2 years of follow‐up, we concluded that for patients with pre‐existing multilevel CCS without cervical fracture or dislocation of ATCCS, both conservative and surgical treatment yield good results. Compared with patients who received conservative treatment and late surgery, patients who received early surgery had better clinical function and living quality. Despite the higher cost, early surgery is cost‐effective when compared to conservative treatment. Younger age, initial better ASIA grade, and earlier surgery were associated with better prognosis.

Introduction

The global prevalence of spinal cord injury (SCI) ranges from 236 to 1009 per million, and the annual incidence is following a straight upward trend. ¹ Central cord syndrome is a manifestation of SCI, so cervical SCI may lead to different types of syndromes, one of which is acute traumatic central cord syndrome (ATCCS), which accounts for 70% of all incomplete cervical SCI. ² , ³ , ⁴ It is characterized by greater motor impairment of the upper extremities compared with that of the lower extremities, bladder dysfunction, and variable sensory loss below the level of injury, which was first described in 1954 by Schneider et al. ⁵ , ⁶ , ⁷ . Almost 50% of cases of ATCCS are due to hyperextension injuries. ⁸ , ⁹ In young patients, ATCCS is mostly caused by high‐energy events, whereas in those 45 years or older, it is more likely due to low‐energy events. ¹⁰ , ¹¹ According to previous research in patients with pre‐existing cervical canal stenosis (CCS), hyperextension injury exacerbates the loss of spinal volume and is more likely to lead to ATCCS. ¹² , ¹³ In contrast, for those with normal cervical canal diameter, whiplash injury is less likely to cause ATCCS. ¹⁴ It has been demonstrated that cervical degeneration worsens with age and the incidence of CCS increases. ¹⁵ , ¹⁶ Teresi et al. ¹⁷ reported that CCS occurs in 26% of asymptomatic older adults over the age of 64. With the rapid development of population aging, CCS is becoming more common in elderly people. As a result, more and more people are at risk of ATCCS due to CCS.

In recent years, the management of SCI has been controversial. The guidelines of the American Association of Neurological Surgeons (2002) stated that there is an insufficient level of evidence that early surgery results in better neurological outcomes for patients with acute SCI. ¹⁸ The updated guidelines (2013) removed the timing of surgery entirely because the evidence is so sparse that recommendations cannot be made with any degree of confidence without further study. ¹⁹ For ATCCS patients with cervical fracture or dislocation, early reports were relatively consistent, which advocated early surgical treatment to stabilize the cervical spine and relieve nerve compression. ²⁰ , ²¹ However, if the injury is not accompanied by cervical fracture or dislocation, treatment is controversial. Yamazaki et al. ¹³ evaluated prognostic factors for ATCCS in 47 patients due to CCS without cervical fracture. They recommend timely surgery, preferably within 2 weeks of injury, to achieve a better functional outcome. Early decompression surgery within 24 h after trauma was reported by La Rosa et al. ²² to be significantly more effective than late surgical treatment. Some research results were inconsistent with the above views. In 1998, Chen et al. ²³ published a retrospective study of 37 patients with ATCCS due to CCS without fractures and spinal cord compression. The improvement in functional recovery in the surgery group (81%) was more rapid and impressive than that in the control group (62%). However, functional recovery was almost identical in both groups after 2 years of follow‐up.

The controversy mentioned above in the treatment prompted us to conduct this retrospective comparative study to further compare the efficacy and cost utility of conservative, early surgical and late surgical treatments for patients with pre‐existing multilevel CCS without cervical fracture or dislocation of ATCCS. In addition, we analyze the factors that determine the prognosis of these patients.

Data and Methods

Selection Criteria

The inclusion criteria were as follows: (i) central cord syndrome; (ii) developmental cervical canal stenosis whose diameter of canal was less than 12mm in two segments or above; (iii) history of acute trauma; and (iv) patients who received methylprednisolone shock therapy within 8 h of injury.

The exclusion criteria were as follows: (i) cervical fracture and dislocation; (ii) American Spinal Injury Association (ASIA) Impairment Scale grade A or E; (iii) another type of serious injury occurring at the same time, such as brain injury, damage to major organs, or other spinal injuries; and (iv) patients who died or were unable to complete 24 months of follow‐up.

General Information

According to the inclusion and exclusion criteria, a total of 93 patients with multilevel CCS without cervical fracture or dislocation of ATCCS who were admitted to our hospital from January 2015 to December 2020 were enrolled (Fig. 1). Among them, 30 cases (Group A) received conservative treatment, 18 cases (Group B) received early surgical treatment (the time from injury to operation was ≤7 days), and 45 cases (Group C) received late surgical treatment (the time from injury to operation was >7 days). All patients received cervical spine X‐ray, CT, and MRI at admission. Patients in Groups B and C received cervical spine X‐ray and MRI after surgery. The basic data of patients in the three groups, including age, gender composition, causes of injury, the narrowest segment and degree of CCS, canal compression rate (CCR) (Fig. 2), the proportion of ossification of the posterior longitudinal ligament (OPLL) and myelomalacia, body mass index (BMI), comorbidities (including diabetes, hypertension, smoking, and alcoholism), length of stay (LOS), follow‐up period, and complications, are shown in Table 1.

Fig. 1.

The selection process of the included patients in this study.

Fig. 2.

T2‐weighted midsagittal magnetic resonance imaging of the cervical spine of a 54‐year‐old male patient. A is the maximum diameter of the cervical cord at the non‐compression level and B is the minimum diameter of the cervical spinal cord at the compression level. Canal compression rate (%) = (A − B) * 100%/B.

TABLE 1.

Demographics of three groups.

	Full sample (n = 93)	Group A (n = 30)	Group B (n = 18)	Group C (n = 45)	p‐value
Age (years)	59.95 ± 11.13	58.43 ± 12.25	59.78 ± 7.81	55.87 ± 11.52	0.127
Gender (Male/Female)	56/37	15/15	14/4	27/18	0.163
Causes of injury [cases (%)]					0.985
Traffic accident	37 (39.79%)	12 (40.00%)	7 (38.89%)	18 (40.00%)
Falling	31 (33.33%)	10 (33.33%)	6 (33.33%)	15 (33.33%)
Sports	12 (12.90%)	5 (16.67%)	2 (11.11%)	5 (11.11%)
Others	13 (13.98%)	3 (20.00%)	3 (16.67%)	7 (15.56%)
The narrowest segment [cases (%)]					0.985
C2–3	0 (0%)	0 (0%)	0 (0%)	0 (0%)
C3–4	16 (17.20%)	5 (16.67%)	4 (22.22%)	7 (15.56%)
C4–5	53 (56.99%)	16 (53.33%)	11 (61.11%)	26 (57.78%)
C5–6	20 (21.51%)	7 (23.33%)	3 (16.67%)	10 (22.22%)
C6–7	4 (4.30%)	2 (6.67%)	0 (0%)	2 (4.44%)
Canal diameter (mm)
Minimum	6.62 ± 11.07	6.87 ± 1.06	6.75 ± 1.14	6.53 ± 1.05	0.215
Maximum	10.45 ± 0.74	10.59 ± 0.87	10.51 ± 0.65	10.40 ± 0.73	0.404
CCR (%)	36.77 ± 8.38	35.20 ± 7.94	36.03 ± 8.49	37.32 ± 8.44	0.387
OPLL (n)	13 (13.98%)	3 (10.00%)	2 (11.11%)	8 (17.39%)	0.414
Myelomalacia [cases (%)]	27 (29.03%)	9 (30.00%)	5 (27.78%)	13 (28.89%)	0.986
BMI (kg/m2)	26.50 ± 3.93	26.65 ± 3.42	27.45 ± 4.03	26.02 ± 4.21	0.415
Comorbidities [cases (%)]
Diabetes	29 (31.18%)	7 (23.33%)	4 (22.22%)	18 (40.00%)	0.205
Hypertension	58 (62.37%)	16 (53.33%)	11 (61.11%)	31 (68.89%)	0.392
Smoking	45 (48.39%)	14 (46.67%)	9 (50.00%)	22 (48.89%)	0.971
Alcoholism	25 (26.88%)	6 (20.00%)	5 (27.78%)	14 (31.11%)	0.566
LOS (days)	16.20 ± 9.18	11.13 ± 10.11	12.00 ± 4.49	18.44 ± 9.10	0.001 a
Follow‐up (months)	42.18 ± 14.67	44.26 ± 14.81	40.89 ± 14.62	41.31 ± 14.78	0.641

Abbreviations: CCR, canal compression rate; LOS, length of stay; OPLL, ossification of the posterior longitudinal ligament.

^a Significant difference among the three groups (P _AB = 0.685, P _AC <0.001; P _BC <0.001).

The average age of the enrolled patients was 59.95 years. The most common cause of injury was traffic accidents, accounting for 39.79% of injuries. C4–C5 was the most vulnerable segment of CCS, and the mean CCR was 36.77%. Among these patients, the average LOS was 16.20 ± 9.18 days, and the average follow‐up time was 42.18 ± 14.67 months. There were no significant differences among the three groups in terms of age, gender, follow‐up period, causes of injury, maximal/minimal spinal canal diameter, the narrowest segment of the spinal canal, CCR, the proportion of OPLL and myelomalacia, BMI, and comorbidities (P > 0.05). Mean LOS was 18.44 ± 9.10 days in Group C, which was longer than 11.13 ± 10.11 days in Group A and 12.00 ± 4.49 days in Group B (P < 0.05).

Treatment Strategy

All patients were hospitalized in our hospital after trauma and received methylprednisolone shock therapy within 8 h of injury. The shock therapy was conducted as follows: 30 mg/kg methylprednisolone intravenously for the first 15 min, followed by 500 mL normal saline intravenously for 45 min, and 5.4 mg/kg/h methylprednisolone intravenously for the remaining 23 h (the concentration of methylprednisolone was 50 mg/mL i.v.). Subsequent symptomatic therapy and functional exercise were recommended for the three groups. After shock therapy, symptomatic therapy and functional exercise were recommended for the three groups depending on the progression of the patient's disease. Patients and their families made the final decision on whether or not to receive surgical treatment and when to perform surgery after the spinal surgeons made the appropriate recommendations based on the patient's different conditions.

Patients in Group A were instructed to limit cervical motion and stay in bed for at least 1 or 2 weeks. Patients in Group B received surgical treatment within 7 days after injury. According to the degree of injury and preoperative imaging assessment of CCS, anterior cervical corpectomy/discectomy and fusion (14 cases) or posterior cervical laminoplasty (four cases) was performed. Patients in Group C did not receive surgical treatment until more than 7 days after injury. According to the degree of injury and preoperative imaging assessment of CCS, anterior cervical corpectomy/discectomy and fusion (23 cases) or posterior cervical laminoplasty (22 cases) was performed. In Group B and Group C, we suggested two different surgical methods (anterior or posterior approaches), and there was no significant difference in the choice of surgical methods between the two groups (Supplementary material 1).

Evaluation Indicators

Neurological status was assessed using the ASIA grade and the Japanese Orthopedic Association (JOA) score, according to the International Standards for Neurological and Functional Classification of Spinal Cord Injury. In addition, the recovery rate (RR) was calculated based on the JOA score (Supplementary material 2).

As a dependent variable, the RR at the last follow‐up is regarded as an indicator to evaluate the prognosis for patients. The gender, age, spinal canal diameter (minimum), CCR, initial Asia grade, and types of treatment were added into the risk factor analysis as independent variables.

Cost‐utility analysis was performed using an EQ‐5D living quality scale‐based utility points system, and the total cost during hospitalization (C) was converted into 2019 US dollars. Patients were asked to fill out the EQ‐5D scale at the last follow‐up. The corresponding utility value (U) was obtained according to Chinese version of the EQ‐5D value set. ²⁴ U_A, U_B, and U_C represent the gained quality‐adjusted life years (QALYs) of patients in each group after treatment. Taking Groups A and B as an example, the incremental cost‐utility ratio (ICUR) of Group A is C_A/U_A, the ICUR of Group B is C_B/U_B, and the ICUR of B versus A is ICUR_B−A = (C_B − C_A)/(U_B − U_A) = △C_B−A/△U_B−A.

Statistical Methods

SPSS 26.0 statistical software (SPSS, Chicago, IL) was used for data analysis. The measurement data is expressed as mean ± standard deviation (⁻x ± s). One‐way ANOVA was used for comparison between groups. The paired sample t‐test was used for comparison in the same group. The X ²‐test was used for categorical variable data. Multivariate linear regression was used for correlation analysis. Dummy variable assignment was used if the independent variables were unordered multiple categorical variables. P < 0.05 indicated that the difference was statistically significant.

Results

Neurological Status Measured Using the American Spinal Injury Association Grade

The results of the ASIA grade are shown in Table 2. In all three groups, the ASIA grade at discharge, 6 months after discharge, and the last follow‐up showed significant improvement compared with that on admission (P < 0.05). At the time of admission and discharge, there was no significant difference in ASIA grade among the three groups (P > 0.05). At 6 months after discharge, the ASIA grade in Group B was significantly improved compared with Group A and Group C (P _AB <0.05, P _BC <0.05), while there was no significant difference between Group A and Group C (P _AC >0.05). Similarly, at the last follow‐up, the Asia grade of Group B was significantly improved compared with Group A and Group C (P _AB <0.05, P _BC <0.05), while there was no significant difference between Group A and Group C (P _AC= > 0.05).

TABLE 2.

Neurological status measured by the American Spinal Injury Association (ASIA) grade.

Group (n = 93)	Admission				Discharge				6 months				Final visit
	B	C	D	E	B	C	D	E	B	C	D	E	B	C	D	E
A (n = 30)	3	8	19	0	2	5	17	6 a	0	3	17	10 a , c	0	1	14	15 a , c
B (n = 18)	3	6	9	0	0	2	9	7 a	0	1	4	13 a , b	0	0	3	15 a , b
C (n = 45)	5	11	29	0	1	9	25	10 a	0	5	23	17 a , c	0	3	18	24 a , c

^a Significance compared with the previous time point, P < 0.05.

^b Significance compared with Group A, P < 0.05.

^c Significance compared with Group B, P < 0.05.

Clinical Function Measured by the Japanese Orthopedic Association Score

The JOA score results of patients are shown in Table 3. The JOA score at discharge, 6 months after discharge, and the last follow‐up of all three groups were significantly improved compared with that of the previous time point (P < 0.05). There was no significant difference in the JOA score among the three groups on admission and at discharge (P > 0.05). At 6 months after discharge, the JOA score of Group B (14.22 ± 1.83) was significantly higher compared with Group A (13.23 ± 2.18) and Group C (13.20 ± 1.97) (P _AB <0.05, P _BC <0.05), while there was no significant difference between Group A and Group C (P _AC >0.05). Similarly, at the last follow‐up, the JOA score of Group B (14.94 ± 1.63) was significantly higher than that of Group A (13.93 ± 2.07) and Group C (14.13 ± 2.04) (P _AB <0.05, P _BC <0.05), while there was no significant difference between Group A and Group C (P _AC >0.05). In addition, the RR of the JOA score of the three groups was significantly improved compared to the previous assessments at the time of discharge, 6 months after discharge, and the last follow‐up (P < 0.05). Group B (42.88% ± 10.00%) showed significantly better improvement at discharge compared with Group A (24.00% ± 16.45%) and Group C (23.67% ± 10.87%) (P _AB <0.05, P _BC <0.05), while there was no significant difference between Group A and Group C (P _AC >0.05). At 6 months after discharge, the RR of Group B (67.23 ± 12.82) was significantly higher compared with that of Group A (54.05 ± 14.23) and Group C (53.82 ± 12.39) (P _AB <0.05, P _BC <0.05), while there was no significant difference between Group A and Group C (P _AC >0.05). At the last follow‐up, the RR of Group B (76.09 ± 12.41) was still significantly higher than that of Group A (63.10 ± 15.54) and Group C (66.79 ± 16.72), while there was still no significant difference between Group A and Group C (P _AC >0.05).

TABLE 3.

Clinical function measured by the Japanese Orthopedic Association (JOA) score.

Group (n = 93)	Admission	Discharge		6 months		Final visit
	JOA score	JOA score	Recovery rate (%)	JOA score	Recovery rate (%)	JOA score	Recovery rate (%)
A (n = 30)	9.23 ± 2.61	10.80 ± 2.71 a	24.00 ± 16.45 a , c	13.23 ± 2.18 a , c	54.05 ± 14.23 a , c	13.93 ± 2.07 a , c	63.10 ± 15.54 a , c
B (n = 18)	9.00 ± 2.87	11.78 ± 2.40 a	42.88 ± 10.00 a , b	14.22 ± 1.83 a , b	67.23 ± 12.82 a , b	14.94 ± 1.63 a , b	76.09 ± 12.41 a , b
C (n = 45)	9.11 ± 2.54	10.89 ± 2.39 a	23.67 ± 10.87 a , c	13.20 ± 1,97 a , c	53.82 ± 12.39 a , c	14.13 ± 2.04 a , c	66.79 ± 16.72 a , c

^a Significance compared with the previous time point, P < 0.05.

^b Significance compared with Group A, P < 0.05.

^c Significance compared with Group B, P < 0.05.

Cost‐Utility Analysis

The average total hospitalization cost and ED‐5D utility value after treatment for the three groups are shown in Table 4. The cost for Group A was significantly lower than for Group B (P < 0.001) and Group C (P < 0.001), while there was no significant difference between Group B and C (P = 0.962). After treatment, patients of Group B gained 0.88 QALYs, significantly higher than Group A (0.71 QALYs, P = 0.002) and Group C (0.70 QALYs, P = 0.001), reflecting better living quality. There was no significant difference between Group A and Group C (P = 0.895). The ICUR_A was US$3588.71 per QALY, significantly lower than ICUR_B (US$15,236.14 per QALY, P < 0.001) and ICUR_C (US$20,899.46 per QALY, P < 0.001). The ICUR_B was significantly lower than the ICUR_C (P = 0.026). When the conservative treatment was considered as the baseline strategy, the ICUR_B−A = △C_B−A/△U_B−A = US$61,631.76 per QALY, ICUR_C−A = △C_C−A/△U_C−A = − US$1,044,585 per QALY. According to the recommendations from the World Health Organization, ICUR is generally considered to be worth the added cost when it is less than three times GDP per capita. ²⁵ Therefore in this study, thus, three times per capita GDP (27,214.09 × 3 = US$81,642.28) of the region (Suzhou) in 2019 was taken as the threshold of patients’ willingness to pay. Therefore, early surgical treatment can be considered cost‐effective compared to conservative treatment. In contrast, late surgical treatment was clearly not cost‐effective.

TABLE 4.

Cost‐utility analysis.

	Group A	Group B	Group C
Cost ($)	2562.99 ± 3521.49	13057.39 ± 3788.15 a	13008.84 ± 3720.28 a
Overall Utility	0.71 ± 0.20	0.88 ± 0.14 a	0.70 ± 0.19 b
ICUR ($ per QALY)	3588.71 ± 4499.41	15236.14 ± 5075.38 a	20899.46 ± 11929.75 a , b

^a Significance compared with Group A, P < 0.05.

^b Significance compared with Group B, P < 0.05.

Multivariate Linear Regression Analysis of Factors Associated with Japanese Orthopedic Association Score Recovery Rate

Factors including gender, age, spinal canal diameter (Minimum), CCR, initial ASIA grade, and three types of treatments were included in the multivariate linear regression analysis associated with JOA score RR at the last follow‐up. The results indicated that age, initial ASIA grade, and treatment options significantly affected the RR at the last follow‐up. As revealed in Table 5, the RR in Group B was 0.356 grades higher than that in Group A and 0.379 grades higher than that in Group C, which revealed that early surgical treatment played an important role in the recovery of neurological function. Thus, younger age, initial better ASIA grade, and earlier surgery were correlated with better prognosis.

TABLE 5.

Multivariate linear regression analysis of factors associated with JOA score recovery rate.

Parameters	Standardized coefficient	SE	p value
Gender	−0.123	0.011	0.053
Age	−0.206	0.001	0.001 a
Spinal canal diameter (Minimum)	−0.060	0.022	0.642
Canal compression rate	−0.042	0.281	0.749
Initial ASIA grade
(B: D)	−0.361	0.032	<0.001 a
(C: D)	−0.226	0.024	0.001 a
Group A: Group B	−0.356	0.035	<0.001 a
Group C: Group B	−0.379	0.027	<0.001 a

^a Statistically significant.

Discussion

In previous studies, surgeons had different opinions about whether to operate on patients with ATCCS. Until the mid‐20th century, surgical decompression of spinal cord injuries had been considered contraindicated because surgical contusion of the “fragile spinal” cord was suspected to cause further damage. ²⁶ , ²⁷ However, with the overall progress of medicine, many studies generally recommended surgery as soon as possible for ATCCS with clear evidence of spinal cord compression. ²⁸ , ²⁹ Furthermore, for those without cervical fracture or dislocation with multilevel CCS, the circumstances were more complicated and new controversy arose. This study retrospectively analyzed 93 ATCCS patients with multilevel CCS who received different treatments and were followed for at least 2 years and found that early surgery resulted in better clinical function and was more cost‐effective compared to conservative treatment and late surgery. Therefore, early surgery is recommended for ATCCS patients with multilevel CCS. In addition, the results of multivariate linear regression analysis showed that younger age, initial better ASIA grade, and earlier surgery were associated with better prognosis.

The Importance of Early Surgery for Long‐Term Prognosis of Acute Traumatic Central Cord Syndrome

Some studies suggested that surgical intervention should be performed early in ATCCS patients without cervical fracture or dislocation. ³⁰ , ³¹ , ³² Others have found the opposite: that there was no significant difference in long‐term neurofunctional recovery between surgical and conservative treatments. ²³ , ³³ Why are attitudes to surgery so different from study to study? Combined with the results of our study, we believe an important reason is that some studies did not distinguish between early and late surgery. Badhiwala et al. ³⁴ reached conclusions through the pooled analysis of individual patient data in 2020 that surgical decompression within 24 h of acute SCI is associated with improved sensorimotor recovery, and the first 24–36 h after injury appears to represent a crucial time window to achieve optimal neurological recovery with decompressive surgery following acute SCI. In our study, the efficacy of early and late surgery was quite different, which has also been demonstrated in the literature. ³⁵ , ³⁶ , ³⁷ Therefore, it is necessary to distinguish the time of surgical intervention when assessing the efficacy of surgery for ATCCS patients. At the same time, our study found no difference in the efficacy of late surgery compared with conservative treatment, which previous studies not directly compared. Therefore, our results further illustrated the importance of the timing of surgical intervention on the long‐term neurological function recovery of patients with pre‐existing multilevel CCS without cervical fracture or dislocation of ATCCS. In conclusion, we suggest that spinal surgeons should suggest early surgery when dealing with such patients. If conditions do not allow early surgery, our recommendation toward surgery is to be increasingly cautious with the extension of time because late surgery may not result in improvement in efficacy compared with conservative treatment. Why, then, did we choose to demark early and late surgery on day 7 in our study? Early, late, or delayed surgical intervention has been defined differently in the literature. ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ There is no consensus on the definition of early or late surgery because of the heterogeneity of patient characteristics and injury mechanisms, and it is difficult to define the precise timing of ATCCS surgical intervention. In this study, we excluded patients not only with cervical fracture and dislocation but also with ASIA grade A. Therefore, the condition of most patients on admission was relatively stable, which allowed patients and their families sufficient time to consider whether to undergo surgery. For elderly patients or those with other diseases, we first evaluate their physical condition and then consider treatment options. For these reasons, the time of surgical intervention for such patients in our hospital was usually no earlier than 3 days. Taken together, we determined that 7 days was an appropriate point for demarcation. In addition, there is no doubt that methylprednisolone plays an important role in the treatment of patients with SCI, and the use of methylprednisolone remains controversial. ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ According to the latest guideline of AOSpine, a 24‐h infusion of high‐dose methylprednisolone is suggested to be offered to adult patients within 8 h of acute SCI as a treatment option, but a 48‐h infusion of high‐dose methylprednisolone to adult patients with acute SCI is not suggested. ⁴⁶ Therefore, in this study, all patients included received methylprednisolone shock therapy within 8 h of injury, which made the comparison between the three groups more convincing.

Cost‐Utility of Early Surgery for Acute Traumatic Central Cord Syndrome Patients

In clinical practice, even though surgery is recommended by surgeons, the high cost of surgery has always led patients to hesitate when faced with a choice. Is the surgery cost‐effective? Not only do patients and their families ask, but clinicians often want to know how cost‐effective are different treatments. Through the analysis of cost utility, this study provides clinicians with a reference for families with economic problems in clinical practice. Conservative treatment resulted in a relatively good prognosis with the least cost, and from that standpoint alone is the most cost‐effective treatment. In analyzing the results of ICUR among different treatments, however, it is demonstrated that early surgery is the most cost‐effective approach under the expected payment threshold. Importantly, if the injury lasts longer than 7 days, we recommend cautious consideration of surgical treatment because of its poor cost‐utility. In terms of surgical timing, early surgery resulted in better cost‐utility. A cost‐utility analysis by Furlan and colleagues had similar results. ⁴⁷ When the delayed spinal decompression was considered as the baseline strategy, the early spinal decompression saved US$58,368,024.12 per QALY for patients with complete SCI and US$536,217.33 per QALY for patients with incomplete SCI.

Positive Factors Affecting the Long‐Term Prognosis of Neurological Function

Furthermore, the multivariate analysis revealed that factors significantly associated with JOA score RR at the last follow‐up were age, initial ASIA grade, and treatment types. Some authors have identified age‐related outcome differences. ⁴⁸ , ⁴⁹ Newey et al. ²⁷ reported that patients over 70 years of age showed poor neurological function recovery. Roth et al. ⁵⁰ also found that age was an adverse prognostic factor associated with functional outcomes. Although CCS is more likely to occur in the elderly, this does not seem to be sufficient to explain the results of this study, as we found that cervical spinal canal diameter is not a prognostic factor. Hence, we tend to attribute the result to the poor overall health and resilience of the elderly. In addition, it is not surprising that the ASIA grade of admission was shown to be a significant indicator. The initial neurological status has been reported in several studies to largely determine the neurological recovery of patients. ⁵⁰ , ⁵¹ Although different treatments had significant effects on the recovery of neurological function in these patients, early surgical treatment had the greatest impact, which confirmed the results of the previous intergroup comparison. The results of our study were generally consistent with those of Yamazaki et al. ¹³ and Chen et al. ⁵¹ on predictors of the prognosis of motor and function in patients with ATCCS.

Strengths and Limitations

This study conducted a study of a subtype of ATCCS patients. After following these patients for at least 2 years, it provides suggestions and a reference for the future treatment of ATCCS patients. This study confirmed that ATCCS patients with multilevel CCS who received early surgery not only had better neurological function, but in the long term, the treatment was also more cost‐effective than conservative treatment and late surgery. Additionally, this study determined that age and initial ASIA grade, as well as the time of surgery, are key factors affecting the long‐term prognosis of ATCCS patients with multilevel CCS. However, this study had several limitations. First, It was designed as a retrospective cohort study, and the sample size was relatively insufficient, especially in the early surgery group. Second, patients with late surgery may have more underlying diseases and poor recovery ability compared with other groups, which would make the neurological function results of these patients generally worse than those of the early surgery group. Thus, it is unconvincing to claim that the difference in efficacy is simply due to the different timing of surgery. Third, our study focused on the prevalence of diabetes and did not explore the management of diabetic patients or the presence or absence of peripheral neuropathy, which necessitates careful consideration and future exploration in similar research endeavors. Finally, we suggest that more caution is needed in interpreting the results of the cost‐utility analysis, not only because of the lack of sensitivity analysis and discounting but also because the cost during hospitalization was covered, while the actual cost includes the costs of rehabilitation and care after discharge and lost income.

Conclusions

For patients with pre‐existing multilevel CCS without cervical fracture or dislocation of ATCCS, both conservative treatment and surgical treatment can have a good recovery rate of over 60%. Compared with patients who received conservative and late surgical treatments, patients who received surgical treatment within 7 days had significantly better clinical function and living quality in the follow‐up period. No significant difference was found between conservative treatment and late surgical treatment. Younger age, initial better ASIA grade, and earlier surgery were associated with better prognosis. Despite the high cost, early surgical treatment is more cost‐effective when compared to conservative treatment.

Author Contributions

Conceptualization, J.Z. and T.L.; methodology, Q.Z.; software, W.H. and J.L.; validation, W.H., H.L., and H.Y.; formal analysis, Q.Z. and J.L.; investigation, J.Z.; resources, Q.Z.; data curation, W.H.; writing—original draft preparation, Q.Z.; writing—review and editing, J.Z. and T.L.; funding acquisition, T.L. All authors have read and agreed to the published version of the manuscript.

Conflict of Interest

The authors declare no conflict of interest.

Ethics Statement

Approval was obtained from the ethics committee of the First Affiliated Hospital of Soochow University (No. 2023294). The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Supporting information

Data S1. Supporting Information.