Sage Journals: Discover world-class research

Abstract

Study Design

A prospective cohort study.

Objectives

To report a new index, the realigned K-line, for predicting surgical outcomes after laminoplasty in patients with degenerative cervical myelopathy (DCM).

Methods

One hundred twenty-eight patients with DCM undergoing laminoplasty were enrolled from January 2018 to April 2021 in our department. A realigned K-line was defined as the line connecting the midpoints of the spinal cord between C2 and C7 on realigned T1-weighted magnetic resonance imaging. The minimum interval between the anterior compression factors of the spinal cord and the realigned K-line (INT_rea), and the modified K-line (INT_mod) were measured. A logistic regression analysis was performed to identify factors associated with unsatisfactory surgical outcomes. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) was applied to evaluate the reliability of the multivariate logistic regression model.

Results

Univariate analysis showed that the score for the bladder function section of the Japanese Orthopedic Association Cervical Myelopathy Evaluation Questionnaire, numeric rating scale scores for arm pain, and INT_rea might be related to the Japanese Orthopaedic Association (JOA) recovery rate (RR) not achieving the minimal clinically important difference (MCID) (P < .05). Only INT_rea (odds ratio = .744, P < .05) was an independent preoperative factor related to the JOA RR not achieving the MCID (area under the curve, .743). A cutoff of 5.0 mm for INT_rea had an accuracy of 71.9% and specificity of 80.3% for predicting the JOA RR not achieving the MCID.

Conclusions

INT_rea is an independent preoperative risk factor related to the JOA RR not achieving the MCID in patients with DCM. A cutoff point of 5.0 mm is most appropriate for alerting spine surgeons to a high likelihood of the JOA RR not achieving the MCID.

Keywords

degenerative cervical myelopathy realigned k-line modified k-line laminoplasty anterior compression of the spinal cord postoperative clinical outcomes

Introduction

Laminoplasty is an effective and relatively safe non-fusion operation for the treatment of degenerative cervical myelopathy (DCM), and its long-term neurological recovery is promising.^1-7 Laminoplasty mainly drifts the spinal cord backward through a “bow-string effect” based on cervical lordosis, thereby achieving indirect decompression of the compressed spinal cord.^8,9

Several preoperative factors have been reported to compromise the efficacy of this indirect decompression.^10-13 Preoperative kyphosis and significant anterior compression of the spinal cord, such as cervical disc herniation and ossification of the posterior longitudinal ligament (OPLL), can lead to residual anterior compression of the spinal cord (ACS) after laminoplasty. These studies independently evaluated the 2 aforementioned factors but did not consider their combined effect on postoperative clinical outcomes. Fujiyoshi et al.¹⁴ first analyzed these 2 factors on a neutral lateral radiograph and combined them into 1 parameter, which they named the K-line, that they used to select the surgical approach in patients with OPLL. However, anterior compression factors of the spinal cord, such as intervertebral disc herniation and spur formation, are unclear on plain radiographs.

Magnetic resonance imaging (MRI) is optimal because it has better visibility than plain radiography and can achieve accurate cervical sagittal parameters. Taniyanna et al.¹⁵ reported a modified K-line connecting the midpoint of the spinal cord between C2 and C7 on preoperative T1-weighted MRI (T1MRI). Their results showed that the minimum interval (INT_min) between the modified K-line and anterior compression factors is an important prognostic indicator of postoperative ACS. However, the value of INT_min is easily affected by the curvature of the cervical spine; the study did not include the effects of gravity and load-bearing on cervical curvature, and the Cobb angle on supine MRI scans may be lower than that on plain radiographs.^15,16 Moreover, the supine posture with a coil on the neck during MRI may lead to a different curvature of the cervical spine. The advent of upright MRI has made it possible to address these problems. However, upright MRI is a new technique that has not yet become routine compared with conventional MRI, and few imaging centers in the world perform upright MRI. In brief, MRI is still a practical and convenient tool for K-line measurement and may be used for postoperative prediction, but further study is needed for better accuracy. Using plain radiographs as references may be a solution; however, to our best knowledge, there is no reference to sagittal parameters of the cervical spine on plain radiographs in reports of the modified K-line.

Previous studies have reported that Surgimap allows surgeons to simulate sagittal correction and graphically present the expected sagittal alignment.^17,18 The present study aimed to report a new index, the realigned K-line, for predicting the surgical outcomes after laminoplasty in patients with DCM and to validate the hypothesis that cervical curvature will affect the measurement of INT.

Methods

Ethics Statements

This study was implemented and reported in line with the Standard Protocol Items: Recommendations for Interventional Trials 2013 Statement.¹⁹ It was also approved by our institutional review board (number: GDREC2017293H). Written informed consent was obtained from all patients before collecting their clinical and radiological information.

Study Design and Population

Patients who underwent laminoplasty under the diagnosis of multilevel DCM and who had complete imaging data were consecutively enrolled in this prospective cohort study from January 2018 to April 2021. The exclusion criteria were as follows: (1) patients with a history of other neurological diseases that may affect their neurological function, such as stroke, Parkinson's disease, or poliomyelitis, and (2) previous cervical surgery. All patients were followed for >1 year. Preoperative MRI and X-ray data, clinical outcome measurements (the Japanese Orthopaedic Association [JOA] score, Japanese Orthopaedic Association Cervical Myelopathy Evaluation Questionnaire [JOACMEQ] score, and numeric rating scale (NRS) scores of neck and arm pain) were collected.

Operative Technique

All surgeries were performed by a single senior surgeon. We conducted laminoplasty as described in previous studies.^20-22 The patients were operated on in a prone neutral position, and then the C3 to C7 laminae were exposed by dissection of both paracervical muscles through a posterior midline incision. A high-speed air-burr drill was used to create grooves at the junction of the lamina and facet joints, and the ventral cortex of each lamina was cut. Instead of being cut, the ventral cortex of the other lamina was grooved as a hinge, and the cut lamina of the opposite lamina was raised and secured in an expanded position with titanium miniplates and screws. The laminoplasty procedure was plated alone, and no bone graft or fusion equipment was placed on the hinge or open side.

Definition of the Realigned K-line

To draw the realigned K-line, we first realigned the cervical alignment on MRI, with sagittal alignment on an X-ray as a reference. A previous study showed that the midpoints of the spinal cord at C2 and C7 are fixed before and after laminoplasty.²³ Among the intervals between the modified K-line and anterior compression factors of the spinal cord at each segment in the magnetic resonance (MR) image, only the lowest interval (INT_min) was significantly correlated with neurological improvement after laminoplasty.²³ Based on this theory, we selected 3 points on the mid-sagittal MR image as the anchor points for realignment: A. the midpoint of the spinal canal at C2, B. the midpoint of the spinal canal at C7, and C. the posterior edge of the intervertebral space at the most severely compressed segment. Then, we drew 2 lines between AC and BC, and the intersection of these 2 lines created an angle (∠ACB), which correlated with the curvature of the cervical spine. In the same way, we drew the aforementioned 2 lines and the angle (∠A'C'B') on the X-ray of the same patient. We used the simulated osteotomy function of Surgimap software¹⁸ (New York, NY, USA) to process the MR image into a realigned MRI scan so that the included angle between the realigned MRI (∠ACB) was equal to that angle on the X-ray (∠A'C'B'). Therefore, the relative positional relationship within the 3 anchor points was similar between the realigned MRI scan and preoperative X-ray. Finally, we drew a K-line on the realigned MRI scan as previously described and called it the “realigned K-line.” Based on the realigned K-line, we defined INT_rea as the minimum interval between the realigned K-line and anterior compression factors on the realigned MRI scan (Figure 1(A)-(C)).

Figure 1.

Schematic drawings of the “realigned K-line.” A: The most severely compressed segment is C5/6 on MRI. The 3 anchor points are as follows: A) the midpoint of the spinal canal at C2, B) the midpoint of the spinal canal at C7, and C) the posterior edge of the intervertebral space at the most severely compressed segment. The intersection of line AC and line BC forms an angle (∠ACB). The modified K-line is defined and the INT_mod is measured. B: In the same way, draw the 2 aforementioned lines and the angle (∠A'C'B') on the X-ray of the same patient. C: A simulated osteotomy is performed on the T1MRI using Surgimap software (we refer to the MRI scan after a simulated osteotomy as the “realigned MRI scan”) so that the ∠ACB on the realigned MRI scan is equal to the ∠A'C'B' on the X-ray. The realigned K-line is defined and the INT_rea is measured. The C2-7 lordosis angle (CL, the cobb angle between the inferior endplate of C2 to the inferior endplate of C7) was measured on MRI, radiography, and realigned MRI, respectively. MRI, magnetic resonance imaging; T1MRI, T1-weighted magnetic resonance imaging; INT_rea, the minimum interval between the realigned K-line and anterior compression factor; INT_mod, the minimum interval between the anterior compression factors of the spinal cord and the modified K-line.

Measurements

Radiological Measurements

We created a realigned MRI scan with all preoperative MRI scans, and INT_rea and INT_mod were measured. The cervical lordotic angle (C2–C7 Cobb angle) was measured using radiography, MRI, and realigned MRI to confirm the efficacy of this method. Moreover, INT_rea was measured separately on 50 randomly selected realigned MRI scans by 2 independent spine surgeons to evaluate inter-observer reliability, and then 1 of the observers analyzed the remaining images. The same observer measured INT_rea twice on 50 randomly selected realigned MRI images, at 2-week intervals, to evaluate intra-observer reliability.

There are 2 ways to measure the INT_rea. The first method is to realign at the most severely compressed segment on the mid-sagittal MRI, copy the relative position relationship between the K-line and the posterior edge of intervertebral space at the most severely compressed segment from radiography to MRI, and then measure INT_rea. The second method is to realign the cervical curvature at each segment on the mid-sagittal MRI to restore the true cervical curvature, and then measure INT_rea. The first method is simple and can obtain the relative position relationship between the K-line and the posterior edge of the intervertebral space at the most severely compressed segment, making the measurement of INT_rea more accurate than the second method. The second method can better restore cervical curvature, but it is more complex than the first method. We randomly selected 50 patients to compare the 2 methods. (Supplementary File 1, Fig 1).

Clinical Measurements

The JOA score, JOACMEQ score, and NRS scores of neck and arm pain were used to evaluate the neurological outcomes before and after laminoplasty. The JOA recovery rate (RR) was evaluated using the Hirabayashi method as follows: RR = [postoperative JOA score – preoperative JOA score]/[17–preoperative JOA score] × 100%.²⁴ The JOACMEQ defines effective treatment as when (1) the post-treatment score increases by ≥ 20 points compared to the pretreatment score, or (2) the pretreatment score is < 90 points and the post-treatment score is ≥ 90 points. Patients with pre-and post-treatment scores ≥90 were excluded from further analysis.²⁵ The efficacy rate was calculated as follows: [(number of patients judged “effective”)/[(total number of patients in the group) − (number of patients whose pre- and post-treatment scores are both ≥90)].

The minimal clinically important difference (MCID) was defined as the smallest change that can be recognized as “clinically meaningful” by a patient, to evaluate the success of interventions. A previous study demonstrated that the minimum clinically important recovery rate was 52.8%.²⁶ In this study, we used 52.8% as a reference criterion to measure whether the changes in the JOA scores were significant enough to make a clinical difference. The patients were classified into 2 groups: the JOA RR achieved the MCID (minimal change group) and JOA RR did not achieve the MCID (no change group). We also evaluated the therapeutic effect based on whether the changes in JOACMEQ scores met the criteria for effective treatment.

Statistical Analysis

Numerical variables are described as mean ± standard deviation (SD), and the differences were analyzed using a Student's t-test. Categorical variables are shown as frequencies and percentages, and differences were compared using the Pearson’s chi-squared test or Fisher’s exact test. Multivariate logistic regression analysis was performed to identify potential independent preoperative factors correlated with the JOA RR not achieving the MCID. Preoperative factors with P-values < .1 in univariate analyses were entered into the multivariate analysis. The intraclass correlation coefficient (ICC) was calculated to evaluate the inter-observer and intra-observer reliabilities of INT_rea. A receiver operating characteristic (ROC) curve was used to measure the reliability of the multivariate logistic regression model. The area under the ROC curve (AUC) was used to test the overall ability of the multivariate logistic regression model to predict the JOA RR not achieving the MCID.

The minimum sample size for this study was 70, according to Flahault et al. (with a sensitivity and specificity of 90% and consideration that the lower 95% confidence limit should not be < .75).²⁷ Owing to the small amount of missing data, no imputation was performed.

All statistical analyses were performed using SPSS statistical software, version 26 (IBM Corp, Armonk, NY, USA). Statistical significance was defined as a P-value < .05.

Results

Characteristics of the study Participants

A total of 143 patients who met the inclusion criteria during the study period were enrolled. During the follow-up period, 3 patients died and 2 had a stroke after surgery. Another 10 patients were lost to follow-up. Thus, 15 patients were excluded, leaving 128 patients for the primary analysis (Figure 2). Table 1 shows the patients’ baseline demographic and clinical characteristics.

Figure 2.

Flow-chart of patient inclusion.

Table 1.

Baseline demographic and clinical characteristics of the 128 patients.

	Total (N = 128)	Minimal change group (N = 66) JOA RR ≥ MCID		P value
Age (Y)	64.3 ± 11.0	62.7 ± 12.2	66.0 ± 9.4	.09
Gender	85M/43F	46M/20F	39M/23F	.42
BMI	23.9 ± 3.4	23.8 ± 3.7	24.1 ± 3.1	.61
Duration of symptom (M)	24.7 ± 28.9	23.0 ± 28.1	26.5 ± 29.8	.49
Smoking history				.55
Yes	53(41.4%)	29(43.9%)	24(38.7%)
No	75(58.6%)	37(56.1%)	38(61.3%)
Comorbidities (%)	63(49.2%)	30 (45.5%)	33(53.2%)	.38
Hypertension	56(43.8%)	27(40.9%)	29(46.8%)
Diabetes	19(14.8%)	7(10.6%)	12(19.4%)
Coronary heart disease	10(7.8%)	2(3%)	8(12.9%)
C2-7 Cobb angle
X-ray	16.7 ± 11.4	19.1 ± 11.5	14.2 ± 10.8	.013
MRI	9.1 ± 12.0	8.7 ± 12.2	9.6 ± 12.0	<.001∗
Realigned MRI	13.7 ± 9.9	14.9 ± 10.3	12.5 ± 9.5	.11#
INT_mod	4.1 ± 2.7	4.2 ± 2.5	3.9 ± 2.9	.47
INT_rea	5.9 ± 3.0	6.9 ± 2.7	4.7 ± 2.9	<.001
Pre-op JOA	11.7 ± 2.1	11.7 ±2.0	11.7 ± 2.2	.977
Pre-op NRS for neck	2.9 ± 3.1	2.7 ± 3.0	3.1 ± 3.1	.462
Pre-op NRS for arm	2.1 ± 2.9	1.5 ± 2.5	2.7 ± 3.1	.020
Average JOA recovery rate (%)	59.8%	79.5%	41.8%	<.001
Effective rate for the JOACMEQ (%)
Cervical spine	63.0% (29/46)	65.4% (17/26)	60.0% (12/20)	.71
Upper extremity	67.0% (63/94)	85.4% (41/48)	47.8% (22/46)	<.001
Lower extremity	58.1% (54/93)	79.1% (34/43)	40.0% (20/50)	<.001
Bladder	73.5% (36/49)	93.3% (14/15)	64.7% (22/34)	.08
QOL	30.7% (39/127)	41.5% (27/65)	19.4% (12/62)	.007

	Pre-op	Post-op	P value	Pre-op	Post-op	P value	Pre-op	Post-op	P value
NRS for neck	2.9 ± 3.1	0.8 ± 1.4	<.001	2.7 ± 3.0	0.8 ± 1.4	<.001	3.1 ± 3.1	0.7 ± 1.5	<.001
NRS for arm	2.1 ± 2.9	0.5 ± 1.5	<.001	1.5 ± 2.5	0.1 ± 0.5	<.001	2.7 ± 3.1	0.8 ± 2.0	<.001
JOA	11.7 ± 2.1	14.9 ± 1.5	<.001	11.7 ± 2.0	15.8 ± 0.9	<.001	11.7 ± 2.2	13.9 ± 1.4	<.001

^∗X-ray compared to MRI in all patients; #X-ray compared to realigned MRI in all patients. MCID, minimal clinically important difference; BMI, body mass index; INT_rea, minimum interval between the anterior compression factors of the spinal cord and realigned K-line on realigned magnetic resonance imaging; INT_mod, minimum interval between the anterior compression factors of the spinal cord and modified K-line on magnetic resonance imaging; JOA, Japanese Orthopaedic Association; RR, recovery rate; NRS, numerical rating scale; JOACMEQ, Japanese Orthopaedic Association Cervical Myelopathy Evaluation Questionnaire; QOL, quality of life; MRI, magnetic resonance imaging; Pre-op, preoperatively; Post-op, post-operatively; MCID, minimal clinically important difference.

The study population of 128 patients included 85 men (66.4%), with a mean (±SD) age of 64.32 (±11.0) years. Sixty-six patients (51.6%) who achieved the MCID were classified into the minimal change group and 62 patients (48.4%) who did not achieve the MCID were classified into the no change group. None of the patients had worse neurological outcomes postoperatively than preoperatively. The C2–C7 Cobb angles were found to be nearly equivalent between the realigned MRI and X-ray images and were strongly correlated (r = .953, P < .0001).

Reliability of the Realigned K-line

The ICC was calculated to evaluate the inter-observer and intra-observer reliabilities of the INT_rea measurement. According to the ICC, the inter-observer and intra-observer reliability values for the INT_rea measurements were .904 and .985, respectively (Table 2), indicating acceptable reliability of the INT_rea measurements.

Table 2.

Inter- and intra-observer reliabilities for the INT_rea measurements.

	INT_rea measures
Inter-observer reliability (ICC),N = 50INT_rea	.904
Intra-observer reliability (ICC),N = 50INT_rea	.985

ICC, intraclass correlation coefficient; INT_rea, the minimum interval between the realigned K-line and anterior compression factor.

Additionally, we found no significant difference in the INT_rea values and C2-7 Cobb angles obtained by realigning at the most severely compressed segment and realigning the cervical curvature at each segment. However, it is simpler to realign at the most severely compressed segment than at each segment, and a relative position relationship between the K-line and the posterior edge of the intervertebral space at the most severely compressed segment can be obtained. Therefore, our method of realigning at the most severely compressed segment is simple and feasible (Supplementary File 2, Table 1).

Relationship between preoperative factors and the JOA RR not achieving the MCID

Regarding prognostic indicators of clinical outcomes after laminoplasty, our results revealed that INT_rea was significantly correlated with the JOA RR. Univariate analyses showed that there were significant differences between the 2 groups regarding the C2-7 Cobb angle on radiography, score of the bladder function section of the JOACMEQ, preoperative NRS score for arm pain, and INT_rea (all, P < .05).

Five preoperative factors with P-values < .10 in univariate analyses (including age, C2-7 Cobb angle on X-ray, score of the bladder function section of the JOACMEQ, preoperative NRS score for arm pain, and INT_rea) were selected for the multivariate logistic regression model. The results showed that only INT_rea was an independent preoperative factor related to the JOA RR not reaching the MCID (Table 3). The AUC of the multivariate logistic regression model was .746 (Fig. 3).

Table 3.

Results of logistic regression analysis of the preoperative risk factors for the JOA RR not reaching the MCID.

Variables	OR	95% CI	P value
Age	1.025	.989–1.064	.180
C2-7 Cobb angle on X-ray	1.006	.957–1.058	.807
Bladder function in the JOACMEQ scores	.973	.940–1.066	.973
Preoperative NRS for the arm	1.135	.984–1.310	.082
INT_rea	.732	.596–.899	.003

OR, odds ratio; CI, confidence interval; INT_rea, the minimum interval between the anterior compression factors of the spinal cord and realigned K-line on realigned magnetic resonance imaging; NRS, numerical rating scale; JOACMEQ, Japanese Orthopaedic Association Cervical Myelopathy Evaluation Questionnaire; JOA RR, Japanese Orthopaedic Association recovery rate; MCID, minimal clinically important difference.

Figure 3.

ROC curve of the multivariate logistic regression model. The AUC of .745 indicates a fair ability to discriminate the JOA RR not reaching the MCID. ROC, receiver operating characteristic; AUC, area under the receiver operating characteristic curve; JOA, Japanese Orthopaedic Association; RR, recovery rate; MCID, minimal clinically important difference.

Moreover, the ROC curve analysis for INT_rea showed an AUC of .718 (95% confidence interval [CI] = .628-.807, P < .001; Supplementary File 1, Figure 2). The cut-off value of INT_rea for the JOA RR not achieving the MCID was 5.0 mm, with a sensitivity of 62.9% (95% CI = 49.7%–74.6%), specificity of 80.3% (95% CI = 68.3%–88.7%), positive predictive value of 75.0% (95% CI = 60.8%–85.5%), and negative predictive value of 69.7% (95% CI = 58.0%–79.5%).

Regarding the postoperative clinical outcomes, we found that 66 patients (51.6%) achieved the MCID. Seventy-six (59.5%) patients had an INT_rea value ≥5.0 mm. Compared with patients with an INT_rea value >5.0 mm, patients with an INT_rea <5.0 mm had a higher risk of the JOA RR not achieving the MCID after laminoplasty (OR: 6.913, 95% CI = 3.119-15.323, P < .001; Supplementary File 2, Table 2).

We also investigated the predictive performance of INT_rea for effective treatment after surgery based on whether the changes in the JOACMEQ scores met the criteria for effective treatment. ROC curve analysis of INT_rea showed that INT_rea could predict ineffective treatment of upper extremity function, with an AUC of .748 (95% CI = .659-.836, P < .001; Supplementary File 1, Figure 3).

Discussion

To our knowledge, this is the first study to realign the cervical curvature on MRI by simulating osteotomy, measuring the sagittal parameters, and discussing the relationship between these parameters and the surgical outcomes after laminoplasty. The major result of the present study was finding that INT_rea may be an independent preoperative factor related to the JOA RR not reaching the MCID. In addition, we confirmed the hypothesis that the measurement of INT on mid-sagittal MRI requires the addition of true curvature information of the cervical spine.

The postoperative prognosis of patients treated with laminoplasty for DCM can be affected by various clinical factors.^28-32 To better predict the clinical efficacy of laminoplasty, Fujiyoshi et al.¹⁴ developed the K-line on lateral neutral radiographs. However, anterior compression factors of the spinal cord such as intervertebral disc herniation and spur formation, are poorly visible on plain radiographs.²³ In contrast, MRI has better visibility than plain radiographs and can accurately measure cervical sagittal parameters. Taniyanna et al.¹⁵ developed a modified K-line on T1MRI and demonstrated that an INT_min value of 4.0 mm can predict residual anterior compression of the spinal cord after laminoplasty with 80% sensitivity and 80.6% specificity.

Although MRI offers better soft tissue imaging than plain radiography, it is rarely able to maintain the cervical spine in a neutral position during MRI, and the cervical curvature can easily change when patients move from an upright position to a supine position. Moreover, the supine posture with a coil on the neck during MR may lead to a different curvature of the cervical spine. An abnormal cervical curvature will not only compromise the effect of indirect decompression but also affect the minimum interval value and thus the accuracy of its prediction of postoperative clinical efficacy. Taniyanna et al.¹⁵ also found an effect of the aforementioned factors on the cervical curvature. They believe that there was a significant correlation between the C2–C7 Cobb angle on MRI and that on plain radiography; therefore, their findings were not affected. However, a recent study reported that despite the significant correlation between the C2–C7 Cobb angle on supine MRI and that on plain radiography, the mean measurement of the C2–C7 Cobb angle was still statistically different between MRI and plain radiography.³³ Several studies have reported that supine MRI is not a substitute for plain radiography for assessing cervical alignment.^16,34,35 Our study also revealed that the mean C2–C7 Cobb angle was different between radiography and MRI. The advent of upright MRI has made it possible to address these problems. However, upright MRI is a new technique that has not yet become routine compared with conventional MRI, and few imaging centers in the world perform upright MRI. In short, MRI is a practical and convenient tool for K-line measurement and may be used for postoperative prediction, but further study is still needed for better accuracy. To consider the influence of gravity and load bearing on cervical curvature while eliminating the effect of neck coil on cervical curvature, we realigned the cervical curvature on MRI by simulating osteotomy with the X-ray as a reference, so that the cervical curvature on the realigned MRI scan was similar to the cervical curvature on the radiograph. The results revealed that the C2–C7 Cobb angle was nearly equivalent between the realigned MRI scan and X-ray.

Next, we developed a realigned K-line on the realigned MRI and measured the minimum interval between the realigned K-line and anterior compression factors (INT_rea). We then analyzed the relationship between preoperative factors and postoperative clinical outcomes. Based on the univariate analysis, we found that patients whose JOA RR did not reach the MCID tended to have a smaller preoperative C2-7 angle, smaller INT_rea, worse bladder function according to the JOACMEQ score, and higher NRS score for arm pain. This result was similar to that of existing studies, where the minimum interval was significantly associated with the JOA RR.^15,23,36 To investigate the independent preoperative factors for predicting the JOA RR not reaching the MCID, we conducted a multivariate logistic regression analysis, which revealed that only INT_rea was a significant factor. To test the reliability of the multivariate logistic regression model, we used the ROC curve and its AUC, which could assess the ability of this model to predict the JOA RR not reaching the MCID. Our model showed an AUC of .746, indicating fair discrimination ability. Previous studies have shown that a prediction model would be considered reasonable with a value of >.7.³⁷ We analyzed the relationship between the preoperative INT_rea value and JOA RR to assess what values of INT_rea are best to prevent the JOA RR from not reaching the MCID. Based on the receiver operating characteristic curve, a cutoff value of 5.0 mm was found to predict the JOA RR not achieving the MCID. The result reveals that an INT_rea value <5.0 mm may be a risk factor for the JOA RR not achieving the MCID. Compared with patients with an INT_rea value >5.0 mm, patients with an INT_rea value <5.0 mm had a higher risk of the JOA RR not achieving the MCID. Moreover, our results revealed that INT_rea can predict ineffective treatment of upper extremity function based on the JOACMEQ score.

This study found that INT_rea was an independent preoperative factor related to the JOA RR not reaching the MCID, but INT_mod had no significant correlation with the postoperative JOA RR. However, Taniyanna et al.²³ believed that INT_min was closely related to the JOA RR after surgery. We believe that the reason for this difference is that we restored the true curvature of the cervical spine on MRI by referring to the X-ray before measuring INT_rea. The intra- and inter-observer reliabilities of the INT_rea measurement on realigned MRI were substantial, supporting that the INT_rea measurement is a reliable and reproducible means of communication between different observers. Therefore, we suggest that clinicians restore the true cervical curvature on MRI before measuring the sagittal parameters used to predict clinical outcomes after laminoplasty. Realigning the cervical curvature by simulating osteotomy is a simple and easy method.

Our study has several limitations. Firstly, this was a single-center prospective study with a small number of patients. Therefore, a multicenter prospective study with a larger sample size is needed to further clarify the validity and practicability of INT_rea for predicting the clinical outcomes of patients with DCM after laminoplasty. Second, owing to the different imaging principles of MRI and radiography, the magnification of the images is also different. The lines from the midpoint of C2 to the posterior edge of the intervertebral space at the most severely compressed segment and from the midpoint of C7 to the posterior edge of the intervertebral space at the most severely compressed segment could not be made identical on realigned MRI and radiography. However, because a significant correlation between the 2 lines on realigned MRI and radiography was observed in this series, the findings may still be considered significant (Supplementary File 3). Third, this study focused on the predictive performance of INT_rea for clinical outcomes after laminoplasty. Furthermore, the most severely compressed segment on MRI does not always correspond to that responsible for cervical myelopathy. Therefore, we did not assess the predictive performance of INT_rea for inadequate decompression of the segment responsible for cervical myelopathy. We will conduct further research on this aspect.

Conclusions

This study’s results suggest that INT_rea may be an independent preoperative risk factor related to the JOA RR not achieving the MCID in patients with DCM. A more satisfactory clinical effect will be obtained after laminoplasty in DCM patients with an INT_rea value >5.0 mm. On this basis, the realigned K-line may inform clinical decision-making and guide individualized treatments for DCM.

Supplemental Material

Supplemental Material - Accuracy of Realigned K-line for Predicting Surgical Outcomes After Laminoplasty in Patients With Degenerative Cervical Myelopathy

Supplemental Material for Accuracy of Realigned K-line for Predicting Surgical Outcomes After Laminoplasty in Patients With Degenerative Cervical Myelopathy by Qifei Duan, Guoyan Liang, Chong Chen, Yongyu Ye, Xiaoqing Zheng, Changxiang Liang and Yunbing Chang in Global Spine Journal

Supplemental Material

Supplemental Material - Accuracy of Realigned K-line for Predicting Surgical Outcomes After Laminoplasty in Patients With Degenerative Cervical Myelopathy

Supplemental Material

Supplemental Material - Accuracy of Realigned K-line for Predicting Surgical Outcomes After Laminoplasty in Patients With Degenerative Cervical Myelopathy

Footnotes

Acknowledgments

We thank Ke’er Wang for contributing to data collection.

Declaration of conflicting interests

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

Funding

Dr. Guoyan Liang has received funding from the Guangdong Provincial Traditional Chinese Medicine Research Project (20221001) and Guangdong Provincial Medical Science and Technology Research Fund Project (B2022059). Dr. Chong Chen has received funding from the National Natural Science Foundation of China (82102636), Basic and Applied Basic Research Foundation of Guangdong Province (2020A1515110545), Guangzhou Municipal Science and Technology Project (202102020100) and Guangdong Medical Research Foundation (A2021301). Dr. Yunbing Chang has received funding from the Natural Science Foundation of Guangdong Province (2022A1515012557) and Guangzhou Science and Technology Program Key Projects (202103000053).

Ethical Approval

The Research Ethics Committee of the institute approved this investigation [No.GDREC2017293H] and all investigations were conducted in conformity with ethical principles of research.

Availability of Data

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

ORCID iD

Qifei Duan

Supplemental Material

Supplemental material for this article is available online.

References

Cho

Kim

Overley

Merrill

. Cervical Laminoplasty: Indications, Surgical Considerations, and Clinical Outcomes. J Am Acad Orthop Surg. 2018;26(7):e142-e152. doi:10.5435/JAAOS-D-16-00242.

Kawaguchi

Kanamori

Ishihara

Ohmori

Nakamura

Kimura

. Minimum 10-year followup after en bloc cervical laminoplasty. Clin Orthop Relat Res. 2003;411:129-139. doi:10.1097/01.blo.0000069889.31220.62.

Hirabayashi

Watanabe

Wakano

Suzuki

Satomi

Ishii

. Expansive open-door laminoplasty for cervical spinal stenotic myelopathy. Spine. 1983;8(7):693-699. doi:10.1097/00007632-198310000-00003.

Kurokawa

Kim

. Cervical Laminoplasty: The History and the Future. Neurol Med -Chir. 2015;55(7):529-539. doi:10.2176/nmc.ra.2014-0387.

Chiba

Ogawa

Ishii

, et al. Long-term results of expansive open-door laminoplasty for cervical myelopathy--average 14-year follow-up study. Spine. 2006;31(26):2998-3005. doi:10.1097/01.brs.0000250307.78987.6b.

Seichi

Takeshita

Ohishi

, et al. Long-term results of double-door laminoplasty for cervical stenotic myelopathy. Spine. 2001;26(5):479-487. doi:10.1097/00007632-200103010-00010.

Matz

Anderson

Groff

, et al. Cervical laminoplasty for the treatment of cervical degenerative myelopathy. J Neurosurg Spine. 2009;11(2):157-169. doi:10.3171/2009.1.SPINE08726.

Weinberg

Rhee

. Cervical laminoplasty: indication, technique, complications. J Spine Surg. 2020;6(1):290-301. doi:10.21037/jss.2020.01.05.

Hirano

Ohara

Mizuno

Itoh

. History and Evolution of Laminoplasty. Neurosurg Clin. 2018;29(1):107-113. doi:10.1016/j.nec.2017.09.019.

10.

Iwasaki

Okuda

Miyauchi

, et al. Surgical strategy for cervical myelopathy due to ossification of the posterior longitudinal ligament: Part 1: Clinical results and limitations of laminoplasty. Spine. 2007;32(6):647-653. doi:10.1097/01.brs.0000257560.91147.86.

11.

Hirai

Kawabata

Enomoto

, et al. Presence of anterior compression of the spinal cord after laminoplasty inhibits upper extremity motor recovery in patients with cervical spondylotic myelopathy. Spine. 2012;37(5):377-384. doi:10.1097/BRS.0b013e31821fd396.

12.

Holly

Matz

Anderson

, et al. Clinical prognostic indicators of surgical outcome in cervical spondylotic myelopathy. J Neurosurg Spine. 2009;11(2):112-118. doi:10.3171/2009.1.SPINE08718.

13.

Uchida

Nakajima

Sato

, et al. Cervical spondylotic myelopathy associated with kyphosis or sagittal sigmoid alignment: outcome after anterior or posterior decompression. J Neurosurg Spine. 2009;11(5):521-528. doi:10.3171/2009.2.SPINE08385.

14.

Fujiyoshi

Yamazaki

Kawabe

, et al. A new concept for making decisions regarding the surgical approach for cervical ossification of the posterior longitudinal ligament: the K-line. Spine. 2008;33(26):E990-E993. doi:10.1097/BRS.0b013e318188b300.

15.

Taniyama

Hirai

Yoshii

, et al. Modified K-line in magnetic resonance imaging predicts clinical outcome in patients with nonlordotic alignment after laminoplasty for cervical spondylotic myelopathy. Spine. 2014;39(21):E1261-E1268. doi:10.1097/BRS.0000000000000531.

16.

Xing

Zhou

Chen

Liang

Dong

. MRI to measure cervical sagittal parameters: a comparison with plain radiographs. Arch Orthop Trauma Surg. 2017;137(4):451-455. doi:10.1007/s00402-017-2639-5.

17.

Akbar

Terran

Ames

Lafage

Schwab

. Use of Surgimap Spine in sagittal plane analysis, osteotomy planning, and correction calculation. Neurosurg Clin. 2013;24(2):163-172. doi:10.1016/j.nec.2012.12.007.

18.

Langella

Villafañe

Damilano

, et al. Predictive Accuracy of Surgimap Surgical Planning for Sagittal Imbalance: A Cohort Study. Spine. 2017;42(22):E1297-E1304. doi:10.1097/BRS.0000000000002230.

19.

Chan

A-W

Tetzlaff

Gøtzsche

, et al. SPIRIT 2013 explanation and elaboration: guidance for protocols of clinical trials. BMJ. 2013;346:e7586. doi:10.1136/bmj.e7586.

20.

Vitarbo

Sheth

Levi

. Open-door expansile cervical laminoplasty. Neurosurgery. 2007;60(1 suppl 1 1):S154-S159. doi:10.1227/01.NEU.0000215353.94448.16.

21.

Mehdain

Stokes

. Cervical laminoplasty. Eur Spine J. 2014;23(12):2759-2762. doi:10.1007/s00586-014-3634-y.

22.

Liang

Zheng

, et al.

Sagittal Alignment Outcomes in Lordotic Cervical Spine: Does Three-Level Anterior Cervical Discectomy and Fusion Outperform Laminoplasty?

Spine. 2019;44(15):E882-E888. doi:10.1097/BRS.0000000000003016.

23.

Taniyama

Hirai

Yamada

, et al. Modified K-line in magnetic resonance imaging predicts insufficient decompression of cervical laminoplasty. Spine. 2013;38(6):496-501. doi:10.1097/BRS.0b013e318273a4f7.

24.

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.

25.

Fukui

Chiba

Kawakami

, et al. JOA Back Pain Evaluation Questionnaire (JOABPEQ)/JOA Cervical Myelopathy Evaluation Questionnaire (JOACMEQ). The report on the development of revised versions. April 16, 2007. The Subcommittee of the Clinical Outcome Committee of the Japanese Orthopaedic Association on Low Back Pain and Cervical Myelopathy Evaluation. J Orthop Sci. 2009;14(3):348-365. doi:10.1007/s00776-009-1337-8.

26.

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.

27.

Flahault

Cadilhac

Thomas

. Sample size calculation should be performed for design accuracy in diagnostic test studies. J Clin Epidemiol. 2005;58(8):859-862. doi:10.1016/j.jclinepi.2004.12.009.

28.

Yamazaki

Homma

Uchiyama

Katsumi

Okumura

. Morphologic limitations of posterior decompression by midsagittal splitting method for myelopathy caused by ossification of the posterior longitudinal ligament in the cervical spine. Spine. 1999;24(1):32-34. doi:10.1097/00007632-199901010-00008.

29.

Baba

Uchida

Maezawa

Furusawa

Azuchi

Imura

. Lordotic alignment and posterior migration of the spinal cord following en bloc open-door laminoplasty for cervical myelopathy: a magnetic resonance imaging study. J Neurol. 1996;243(9):626-632. doi:10.1007/BF00878657.

30.

Morio

Yamamoto

Kuranobu

Murata

Tuda

. Does increased signal intensity of the spinal cord on MR images due to cervical myelopathy predict prognosis? Arch Orthop Trauma Surg. 1994;113(5):254-259. doi:10.1007/BF00443813.

31.

Kato

Mihara

Niimura

, et al. Impact of focal apex angle on postoperative decompression status of the spinal cord and neurological recovery after cervical laminoplasty. J Neurosurg Spine. 2021;35(4):410-418. doi:10.3171/2020.12.SPINE201831.

32.

Lau

Winkler

Than

Chou

Mummaneni

. Laminoplasty versus laminectomy with posterior spinal fusion for multilevel cervical spondylotic myelopathy: influence of cervical alignment on outcomes. J Neurosurg Spine. 2017;27(5):508-517. doi:10.3171/2017.4.SPINE16831.

33.

Boudreau

Carrondo Cottin

Ruel-Laliberté

Mercier

Gélinas-Phaneuf

Paquet

. Correlation of supine MRI and standing radiographs for cervical sagittal balance in myelopathy patients: a cross-sectional study. Eur Spine J. 2021;30(6):1521-1528. doi:10.1007/s00586-021-06833-0.

34.

Wang

Sun

Mao

, et al. MRI May Serve as a Valid Alternative to Standing Radiography in Evaluating the Sagittal Alignment of the Upper Thoracic Spine. Clinical spine surgery. 2017;30(3):124-128. doi:10.1097/BSD.0000000000000027.

35.

Qiao

Zhu

Liu

, et al. Measurement of Thoracic Inlet alignment on MRI: Reliability and the Influence of Body Position. Clinical spine surgery. 2017;30(4):E377-E380. doi:10.1097/BSD.0000000000000306.

36.

Sun

. Prediction of incomplete decompression after cervical laminoplasty on magnetic resonance imaging: The modified K-line. Clin Neurol Neurosurg. 2016;146:12-17. doi:10.1016/j.clineuro.2016.04.013.

37.

Carter

Pan

Rai

Galandiuk

. ROC-ing along: Evaluation and interpretation of receiver operating characteristic curves. Surgery. 2016;159(6):1638-1645. doi:10.1016/j.surg.2015.12.029.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB

0.05 MB

0.23 MB