Abstract
BACKGROUND CONTEXT:
While cervical disc replacement (CDR) has been emerging as a reliable and efficacious treatment option for degenerative cervical spine pathology, not all patients undergoing CDR will achieve minimal clinically important difference (MCID) in patient-reported outcome measures (PROMs) postoperatively—risk factors for failure to achieve MCID in PROMs following CDR have not been established.
PURPOSE:
To identify risk factors for failure to achieve MCID in Neck Disability Index (NDI, Visual Analog Scale (VAS) neck and arm following primary 1- or 2-level CDRs in the early and late postoperative periods.
STUDY DESIGN:
Retrospective review of prospectively collected data.
PATIENT SAMPLE:
Patients who had undergone primary 1- or 2-level CDR for the treatment of degenerative cervical pathology at a single institution with a minimum follow-up of 6 weeks between 2017 and 2022.
OUTCOME MEASURES:
Patient-reported outcomes: Neck disability index (NDI), Visual analog scale (VAS) neck and arm, MCID.
METHODS:
Minimal clinically important difference achievement rates for NDI, VAS-Neck, and VAS-Arm within early (within 3 months) and late (6 months to 2 years) postoperative periods were assessed based on previously established thresholds. Multivariate logistic regressions were performed for each PROM and evaluation period, with failure to achieve MCID assigned as the outcome variable, to establish models to identify risk factors for failure to achieve MCID and predictors for achievement of MCID. Predictor variables included in the analyses featured demographics, comorbidities, diagnoses/symptoms, and perioperative characteristics.
RESULTS:
A total of 154 patients met the inclusion criteria. The majority of patients achieved MCID for NDI, VAS-Neck, and VAS-Arm for both early and late postoperative periods—79% achieved MCID for at least one of the PROMs in the early postoperative period, while 80% achieved MCID for at least one of the PROMs in the late postoperative period. Predominant neck pain was identified as a risk factor for failure to achieve MCID for NDI in the early (OR: 3.13 [1.10–8.87], p-value: .032) and late (OR: 5.01 [1.31–19.12], p-value: .018) postoperative periods, and VAS-Arm for the late postoperative period (OR: 36.63 [3.78–354.56], p-value: .002). Myelopathy was identified as a risk factor for failure to achieve MCID for VAS-Neck in the early postoperative period (OR: 3.40 [1.08–10.66], p-value: .036). Anxiety was identified as a risk factor for failure to achieve MCID for VAS-Neck in the late postoperative period (OR: 6.51 [1.91–22.18], p-value: .003). CDR at levels C5C7 was identified as a risk factor for failure to achieve MCID in NDI for the late postoperative period (OR: 9.74 [1.43–66.34], p-value: .020).
CONCLUSIONS:
Our study identified several risk factors for failure to achieve MCID in common PROMs following CDR including predominant neck pain, myelopathy, anxiety, and CDR at levels C5–C7. These findings may help inform the approach to counseling patients on outcomes of CDR as the evidence suggests that those with the risk factors above may not improve as reliably after CDR.
Keywords: Cervical disc replacement, Minimal clinically important difference, Neck disability index, Patient reported outcome measures, Visual analog scale
Introduction
Degenerative pathology of the cervical spine contributes to significant disability and impairment of quality of life [1]. Today, several surgical management options are available for patients who experience symptoms that are unresponsive to conservative treatment. Traditionally, anterior cervical discectomy and fusion (ACDF) is considered an effective and reliable treatment option for cervical disc disease—however, within the past 2 decades, there has been growing interest and clinical application of the developing technique of cervical disc replacement (CDR) as a motion preserving treatment option for similar degenerative cervical pathologies [2]. Recent studies have demonstrated CDR as equal or superior to ACDF when comparing clinical outcomes [3−8]. A 2018 systematic review and meta-analysis found that CDR compared with ACDF leads to significantly higher rates of overall success, neurologic success, improvement of disability scores, and lower rates of repeat surgeries [8].
Patient-reported outcome measures (PROMs) are standardized, validated questionnaires that allow for patient-centered evaluation of quality of life, pain, and functional status pre- and postoperatively [9]. Common PROMs used in cervical spine surgery include Neck Disability Index (NDI), VAS-Neck, and VAS-Arm [10,11]. Neck Disability Index assesses the ways in which neck pain impacts a patient’s daily life, while VAS-Neck and VAS-Arm assign numerical values to pain in their respective anatomical regions [10,11]. Minimal clinically important differences (MCID) allow for the translation of PROMs scores into clinically relevant and meaningful information, allowing for the assessment of improvement postoperatively and surgical success [12]. Minimal clinically important differences are predetermined thresholds that compare preoperative and postoperative PROMs scores to assess clinical improvement, defined as the smallest change in treatment outcome that a patient would identify as important [12].
While CDR has been emerging as a reliable and efficacious treatment option for cervical spine pathology, not all patients undergoing CDR will achieve MCID in PROMs postoperatively. The risk factors associated with failure to achieve MCID following CDR are not well established. Previous studies have identified risk factors associated with failure to achieve MCID for PROMs after ACDF [13,14]. Other studies have investigated differences in MCID following CDR between different groups (ie, predominant pain symptom, type of compressive pathology) [15]. However, few studies have investigated risk factors for failure to achieve MCID in PROMs following CDR. The purpose of this study is to identify risk factors for failure to achieve MCID in NDI, VAS-Neck, and VAS-Arm in early and late postoperative periods following primary 1- or 2-level CDR.
Methods
Patient population
This was an Institutional Review Board-approved retrospective review of a prospectively maintained database of spine surgeries to identify patients who underwent primary, 1- or 2-level CDRs by four attending orthopedic spine surgeons at a single academic institution from April 2017 to November 2022. Patients with a minimum follow-up of 6 weeks were included in the study. Patients who were missing preoperative PROMs or failed to complete any postoperative PROM by 2 years in any of the 3 PROMs evaluated (NDI, VAS-Neck, VAS-Arm) were excluded from the study.
Surgical procedure
A left-sided anterior cervical approach is used for CDR. A complete discectomy is performed with posterior longitudinal ligament resection and neuroforamina decompression bilaterally. Since complete uncinate resection may result in intersegmental instability, partial resection is typically performed until adequate decompression of the corresponding foramina is achieved. Careful endplate preparation is performed using a high-speed burr. A trial implant is then placed using a tamp and mallet attachment, with proper positioning and size confirmed via fluoroscopy. The trial is removed, and the implant (Mobi-C Cervical Disc, ZimVie Spine, Inc., or Prestige LP Cervical Disc System, Medtronic or Simplify Cervical Disc, NuVasive, depending on surgeon preference) is impacted using both direct visualization and fluoroscopic imaging. Drain placement is at the discretion of the surgeon and is removed on postoperative day 1. A soft collar is worn for comfort postoperatively for 3 weeks.
PROMs and MCID analysis
Patient-reported outcome measures evaluated in this study include NDI, VAS-Neck, and VAS-Arm. Questionaries were completed preoperatively and at 6-week, 12-week, 6-month, 1-year, and 2-year timepoints postoperatively. NDI is scored as a percentage (max 100%) with higher scores associated with greater disability, while VAS-Neck and VAS-Arm measure pain on a scale of 0 (no pain) to 10 (maximum pain).
Achievement of MCID was assessed by calculating the difference between postoperative PROM values and their respective baseline (preoperative) values and compared with minimum difference in scores for achievement of MCID as established in literature: NDI: −15.0, VAS-Neck: −2.5, VAS-Arm: −2.5 [16−20]. For each PROM, achievement of MCID was assessed in an early and late postoperative period. Early postoperative period includes the 6-week and 12-week postoperative timepoint, and late postoperative period includes the 6-month, 1-year, and 2-year postoperative timepoints. Most recent available PROM scores for the early and late postoperative periods were used for the MCID calculation and achievement determination.
Statistical analysis
Risk factors for failure to achieve MCID were identified using multivariate logistic regression models with alternating backward stepwise elimination and forward entry of removed variables (entry: 0.05, removal: 0.10). Models were generated for early and late post-operative periods, for NDI, VAS-Neck, and VAS-Arm. The outcome variable for each model was the failure to achieve MCID for either NDI, VAS-Neck, or VAS-Arm within the early or late postoperative period. Univariate logistic regressions were performed for each predictor variable to determine which variables were to be included in the multivariate analyses. Predictor variables included patient demographics, comorbidities, diagnoses/symptoms, and perioperative factors (listed in Table 1). All statistical analyses were performed using IBM SPSS Statistics version 26.0 (IBM Corp). The p-values of <.05 were interpreted as statistically significant.
Table 1.
Patient baseline characteristics and perioperative factors
| Characteristic | Primary 1- and 2-level CDRs (N=154) |
|---|---|
| Age (mean±SD) | 42.8±9.3 |
| Sex | |
| Male | 57.1% (88) |
| Female | 42.9% (66) |
| BMI (kg/m2) | |
| <30 kg/m^2 | 83.8% (129) |
| ≥30 kg/m^2 | 16.2 % (25) |
| Insurance | |
| Private | 98.1% (151) |
| Medicare/Medicaid | 1.9% (3) |
| ASA score | |
| <2 | 17.5% (27) |
| ≥2 | 82.5% (127) |
| Smoking status | |
| Current smoker | 8.4% (13) |
| Former smoker | 17.5% (27) |
| Comorbidities | |
| Hypertension | 9.7% (15) |
| Diabetes | 1.3% (2) |
| Dyslipidemia | 14.3% (22) |
| Depression | 13.6% (21) |
| Anxiety | 20.1% (31) |
| CCI ageless | |
| ≥1 | 9.1% (14) |
| <1 | 90.9% (140) |
| Preoperative narcotic pain medication use | 20.8% (32) |
| Compressive pathology | |
| Radiculopathy only | 53.9% (83) |
| Myelopathy only | 15.6% (24) |
| Myeloradiculopathy | 30.5% (47) |
| Predominant pain symptom | |
| Predominant neck pain | 39.0% (60) |
| Predominant arm pain | 27.3% (42) |
| Equal neck and arm pain | 33.7% (52) |
| Number of levels replaced | |
| =1 | 67.5% (104) |
| =2 | 32.5% (50) |
| Level(s) replaced | |
| C3C4 only | 3.9% (6) |
| C4C5 only | 3.9% (6) |
| C5C6 only | 32.5% (50) |
| C6C7 only | 27.9% (43) |
| C4C6 | 8.4% (13) |
| C5C7 | 23.3% (36) |
| Operative time (mean ± SD) | 90.1 ± 35.5 |
| Estimated blood loss | 35.4 ± 23.6 |
| Length of stay | |
| ≥1d | 75.3% (116) |
| <1d | 24.7% (38) |
CDR, cervical disc replacement; BMI, body mass index; ASA, American Society of Anesthesiology; CCI, Charloson Comorbidity Index.
Note. All statistics reported as mean ± SD or number (%).
Results
Demographics and perioperative variables
The findings are summarized in Table 1. A total of 154 patients who underwent primary 1- or 2-level CDR with preoperative NDI, VAS-Neck, or VAS-Arm were included in this study. The mean age of this patient cohort was 42.8 years. Males comprised 57.1% of the cohort and 83.8% were classified as nonobese (BMI < 30 kg/m2). The majority of patients had private insurance (98.1%) compared with Medicaid/Medicare (1.9%). The most common comorbidity was anxiety (20.1%) followed by dyslipidemia (14.3%), depression (13.6%), hypertension (9.7%), and diabetes (1.3%). Former smokers comprised 17.5% of the cohort, while current smokers comprised 8.4%. Of patients, 20.8% use narcotic pain medication for symptomatic management preoperatively. Radiculopathy was the most common compressive pathology (51.3%) followed by, myeloradiculopathy (27.3%) and myelopathy alone (14.3%). Predominant neck pain was the most common pain symptom experienced by patients (39%) followed by equal neck and arm pain (33.7%), and predominant arm pain (27.3%). Most patients underwent 1-level CDR (67.5%), while the remaining underwent 2-level CDR (32.5%). The most common level(s) of disc replacement was C5C6 (32.5%), followed by C6C7 (27.9%), C5C7 (23.3%), C4C6 (8.4%), C3C4 (3.9%), and C4C5 (3.9%). The mean operative time was 90.1 (±35.5) minutes, with a mean estimated blood loss of 35.4 (±23.6) mL. Most cases required admission with discharge on postoperative day ≥ 1 (75.3%) compared with ambulatory cases (24.7%).
Achievement of MCID for NDI, VAS-Neck, and VAS-Arm in early and late postoperative periods
The findings are summarized in Tables 2 and 3. The early postoperative period for this study was defined as follow-up at either the 6-week or 12-week timepoint, while the late-postoperative periods were defined as follow-up at either the 6-month, 1-year, or 2-year timepoints. In the early postoperative period, 129 patients had completed either NDI, VAS-Neck, or VAS-Arm at the 6-week or 12-week timepoints. Within this early postoperative period, 68.2% of patients achieved MCID for NDI following CDR, while 70.5% and 83.5% of patients achieved MCID for VAS-Neck and VAS-Arm respectively. A total of 102 (79.1%) patients achieved MCID in at least one of the three PROMs included in this study within the early postoperative period. When broken down by diagnosis, patients in the myelopathy-only group had the lowest rates of achievement of at least one of the three PROMs (65%) compared with patients with radiculopathy only (80%) and myeloradiculopathy (88.1%). There were no statistically significant differences in mean PROMs scores for NDI, VAS-Neck, or VAS-ARM for the early follow-up timepoints of 6- and 12-weeks between patients with only short-term follow-up (only 6- or 12-week) and patients with both short- and long-term follow-up.
Table 2.
Achievement of MCID for NDI, VAS-Neck, and VAS-Arm
| Early postop period (within 3 months postop, % [n]) | |
|---|---|
| Achieved MCID for NDI (n=107) | 68.2% (73) |
| Achieved MCID for VAS-Neck (n=105) | 70.5% (74) |
| Achieved MCID for VAS-Arm (n=85) | 83.5% (71) |
| Achieved MCID for at least 1 of 3 PROMs (n=129) | 79.1% (102) |
| Late postop period (6 mo to 2 y postop, % [n]) | |
| Achieved MCID for NDI (n=88) | 71.6% (63) |
| Achieved MCID for VAS-Neck (n=91) | 75.8% (69) |
| Achieved MCID for VAS-Arm (n=77) | 81.8% (63) |
| Achieved MCID for at least 1 of 3 PROMs (n=109) | 79.8% (87) |
MCID, minimal clinically important difference; NDI, neck disability index; VAS, visual analog scale; PROM, patient reported outcome measure.
Table 3.
Achievement of MCID for NDI, VAS-Neck, and VAS-Arm by diagnosis
| Early postop period (within 3 months postop, % [n]) | |
|---|---|
| Radiculopathy only | |
| Achieved MCID for NDI (n=53) | 64.2% (34) |
| Achieved MCID for VAS-Neck (n=51) | 70.6% (36) |
| Achieved MCID for VAS-Arm (n=43) | 79.1% (34) |
| Achieved MCID for at least 1 of 3 PROMs (n=65) | 80.0% (52) |
| Myelopathy only | |
| Achieved MCID for NDI (n=17) | 58.8% (10) |
| Achieved MCID for VAS-Neck (n=17) | 41.2% (7) |
| Achieved MCID for VAS-Arm (n=12) | 75.0% (9) |
| Achieved MCID for at least 1 of 3 PROMs (n=20) | 65.0% (13) |
| Myeloradiculopathy | |
| Achieved MCID for NDI (n=37) | 78.4% (29) |
| Achieved MCID for VAS-Neck (n=37) | 83.8% (31) |
| Achieved MCID for VAS-Arm (n=30) | 93.3% (28) |
| Achieved MCID for at least 1 of 3 PROMs (n=42) | 88.1% (37) |
| Late postop period (6m to 2ys postop, % (n)) | |
| Radiculopathy only | |
| Achieved MCID for NDI (n=43) | 72.1% (31) |
| Achieved MCID for VAS-Neck (n=43) | 72.1% (31) |
| Achieved MCID for VAS-Arm (n=37) | 81.1% (30) |
| Achieved MCID for at least 1 of 3 PROMs (n=56) | 78.6% (44) |
| Myelopathy only | |
| Achieved MCID for NDI (n=13) | 53.8% (7) |
| Achieved MCID for VAS-Neck (n=14) | 71.4% (10) |
| Achieved MCID for VAS-Arm (n=11) | 63.6% (7) |
| Achieved MCID for at least 1 of 3 PROMs (n=16) | 68.8% (11) |
| Myeloradiculopathy | |
| Achieved MCID for NDI (n=32) | 78.1% (25) |
| Achieved MCID for VAS-Neck (n=34) | 82.4% (28) |
| Achieved MCID for VAS-Arm (n=29) | 89.7% (26) |
| Achieved MCID for at least 1 of 3 PROMs (n=38) | 84.2% (32) |
MCID, minimal clinically important difference; NDI, neck disability index; VAS, visual analog scale; PROM, patient reported outcome measure.
In the late postoperative period, 109 patients had completed either NDI, VAS-Neck, or VAS-Arm at the 6-month, 1-year, or 2-year timepoints. Within this late postoperative period, 71.6% of patients achieved MCID for NDI following CDR, while 75.8% and 81.8% of patients achieved MCID for VAS-Neck and VAS-Arm respectively. A total of 87 (79.8%) patients achieved MCID in at least one of the three PROMs included in this study in the late postoperative period. Similar to the early postoperative period, when broken down by diagnosis, patients with myelopathy only had the lowest rate of the achievement of MCID in at least one of three PROMs (68.8%) compared with patients with radiculopathy only (78.6%) and myeloradiculopathy (84.2%).
Risk factors for failure to achieve MCID in PROMs within the early postoperative period
The findings are summarized in Table 4. Multivariate logistic regressions were performed with failure to achieve MCID in NDI, VAS-Neck, or VAS-Arm assigned as the outcome variable with all models controlling for at least age and sex (in addition to other included predictor variables). Multivariate analysis identified predominant neck pain (p-value: .032, OR: 3.13, 95% CI: 1.10–8.87) as a risk factor failure to achieve MCID in NDI within the early postoperative period. For VAS-Neck, myelopathy (p-value: .036, OR: 3.40, 95% CI: 1.08–10.66) was identified as a risk factor for failure to achieve MCID within the early postoperative period. No risk factors for failure to achieve MCID in the early postoperative period were identified in multivariate analysis for VAS-Arm.
Table 4.
Risk factors for failure to achieve MCID within early postoperative period (within 3 months post CDR)
| Odds ratio | 95% CI | p-value | ||
|---|---|---|---|---|
| Neck Disability Index (NDI) | ||||
| Patient characteristics | ||||
| Age | 1.00 | 0.95 | 1.05 | .890 |
| Sex (ref: male) | 0.84 | 0.35 | 2.01 | .690 |
| Diagnosis/Symptoms | ||||
| Predominant arm pain | Reference | - | - | - |
| Predominant neck pain | 3.13 | 1.10 | 8.87 | .032 |
| Equal neck and arm pain | 0.75 | 0.22 | 2.60 | .655 |
| Perioperative factors | ||||
| Length of stay ≥1d | 2.14 | 0.78 | 5.85 | .138 |
| VAS-Neck | ||||
| Patient characteristics | ||||
| Age | 0.93 | 0.93 | 1.03 | .475 |
| Sex (ref: male) | 0.90 | 0.37 | 2.21 | .820 |
| Diagnosis/Symptoms | ||||
| Radiculopathy only | Reference | - | - | - |
| Myelopathy only | 3.40 | 1.08 | 10.66 | .036 |
| Myeloradiculopathy | 0.45 | 0.16 | 1.31 | .144 |
| VAS-Arm | ||||
| Patient characteristics | ||||
| Age | 1.01 | 0.93 | 1.10 | .799 |
| Sex (ref: male) | 1.40 | 0.38 | 5.07 | .612 |
| Dyslipidemia | 5.80 | 0.97 | 34.81 | .054 |
| Diagnosis/Symptoms | ||||
| Radiculopathy only | Reference | - | - | - |
| Myelopathy only | 0.85 | 0.15 | 5.00 | .860 |
| Myeloradiculopathy | 0.17 | 0.03 | 1.06 | .057 |
| Predominant arm pain | Reference | - | - | - |
| Predominant neck pain | 4.47 | 0.79 | 25.40 | .091 |
| Equal neck and arm pain | 1.82 | 0.38 | 8.66 | .454 |
MCID, minimal clinically important difference; NDI, neck disability index; VAS, visual analog scale; PROM, patient reported outcome measure. Bold values indicate statistically significant values.
Risk factors for failure to achieve MCID in PROMs in the late postoperative period
The findings are summarized in Table 5. Multivariate logistic regressions were performed for NDI, VAS-Neck, and VAS-Arm for the late postoperative period with minimum follow-up of 6 months, through 2 years postop—similar to NDI findings for the early postoperative period, multivariate analysis identified predominant neck pain (p-value: .018, OR: 5.01, 95% CI: 1.31–19.12) as a risk factor for failure to achieve MCID for NDI in the late post-operative period. Additionally, CDR at levels C5C7 (reference: CDR at C5C6) (p-value: .020, OR: 9.74, 95% CI: 1.43–66.34) was identified as a risk factor for failure to achieve MCID in NDI. Operative time (p-value: .046, OR: 0.98, 95% CI: 0.95 – 0.99) was identified as a predictor for the achievement of MCID in NDI in the late postoperative period. Anxiety (p-value .003, OR: 6.51, 95%CI 1.91 −22.18) was identified as a risk factor for failure to achieve MCID in VAS-Neck in the late postoperative period. For VAS-Arm, predominant neck pain (p-value: .002, OR: 36.63, 95% CI: 3.78–354.56) was identified as a predictor for failure to achieve MCID in the late postoperative period.
Table 5.
Risk factors for failure to achieve MCID early in late postoperative period (6 months to 2 years postCDR)
| Odds ratio | 95% CI | p-value | ||
|---|---|---|---|---|
| Neck Disability Index (NDI) | ||||
| Patient characteristics | ||||
| Age | 1.00 | 0.94 | 1.07 | .992 |
| Sex (ref: male) | 1.02 | 0.33 | 3.14 | .969 |
| Diagnosis/Symptoms | ||||
| Predominant arm pain | Reference | - | - | - |
| Predominant neck pain | 5.01 | 1.31 | 19.12 | .018 |
| Equal neck and arm pain | 0.88 | 0.17 | 4.58 | .879 |
| Perioperative factors | ||||
| Level C5C6 | Reference | - | - | - |
| Level C3C4 | 0.55 | 0.04 | 7.52 | .654 |
| Level C4C5 | 5.20 | 0.29 | 93.12 | .263 |
| Level C6C7 | 1.61 | 0.40 | 6.53 | .508 |
| Level C4C6 | <0.01 | <0.01 | - | .999 |
| Level C5C7 | 9.74 | 1.43 | 66.34 | .020 |
| Operative time | 0.98 | 0.95 | 0.99 | .046 |
| VAS-Neck | ||||
| Patient characteristics | ||||
| Age | 0.96 | 0.90 | 1.02 | .206 |
| Sex (ref: male) | 0.57 | 0.19 | 1.76 | .330 |
| Anxiety | 6.51 | 1.91 | 22.18 | .003 |
| Preoperative narcotic pain medication use | 0.25 | 0.06 | 1.12 | .069 |
| VAS-Arm | ||||
| Patient characteristics | ||||
| Age | 1.03 | 0.95 | 1.12 | .483 |
| Sex (ref: male) | 0.31 | 0.06 | 1.54 | .151 |
| Diagnosis/Symptoms | ||||
| Predominant arm pain | Reference | - | - | - |
| Predominant neck pain | 36.63 | 3.78 | 354.56 | .002 |
| Equal neck and arm pain | 4.23 | 0.43 | 41.38 | .215 |
MCID, minimal clinically important difference; NDI, neck disability index; VAS, visual analog scale; PROM, patient-reported outcome measure. Bold values indicate statistically significant values.
Discussion
Patient-reported outcome measures and MCID have emerged as effective and reliable parameters to assess post-operative improvement and surgical success through a patient-centered approach to evaluation [12]. Cervical spine procedures including CDR are commonly assessed through NDI, VAS-Neck, and VAS-Arm, which aim to characterize quality of life, functional status, and pain pre- and postoperatively [10,11]. As CDR continues to become more widely adopted as a reliable and efficacious treatment option for cervical spine pathology, sharing many similar indications to the traditional “gold-standard” ACDF, it is essential to understand the risk factors for failure to achieve MCID in PROMs following CDR to improve patient outcomes through addressing modifiable risk factors when appropriate, optimizing patients preoperatively, and counseling them on postoperative expectations based on these risk factors. Previous studies have investigated the risk factors for failure to achieve MCID in PROMs following ACDF, but no study has done so for CDR to date [13−15]. The purpose of this study was to investigate demographics, comorbidities, spinal pathology, and peri-operative factors that may be associated with either greater or lower risk of failure to achieve MCID in NDI, VAS-Neck, and VAS-Arm following primary 1- or 2-level CDR.
In the current study, the majority of patients achieved MCID in the NDI, VAS-Neck, and VAS-Arm following CDR in both the early- and late postoperative periods, with 79% achieving MCID in at least 1 of the 3 PROMs in the early postoperative period, and 80% in the late postoperative period. This is similar to findings by Alluri et al. which reported 89% of patients achieving MCID in at least 1 PROM with at least 6 months follow-up, compared with 74% of patients following ACDF [21]. For both the early and late postoperative periods, patients with only myelopathy had lower rates of achievement of MCID for all PROMs compared with those with radiculopathy or myeloradiculopathy. It is worth noting that patients with myelopathy alone comprised a relatively small subset of the total cohort (15.6%). Although several previous studies have demonstrated the successful treatment of radiculopathy and/or myelopathy with CDR, few studies have systematically compared outcomes between patients with radiculopathy and myelopathy [22−25]. One retrospective cohort study by Alluri et al. reported no significant differences in MCID between patients with radiculopathy and myeloradiculopathy for several PROMs including NDI, VAS-Neck, VAS-Arm, at 6 weeks to final follow-up (minimum 6 months) postoperative time-point [14]. However, there remains controversy regarding whether CDR for treatment in patients with myelopathy can achieve similar outcomes compared with use in patients with radiculopathy only, stemming from the concern that motion-preserving techniques like CDR in myelopathy can contribute to persistent micro-trauma of the spinal cord as static compressive pathology may be addressed, but dynamic compression may not be alleviated [26,39]. There is also potential for worsening of myelopathic symptoms following CDR due to segmental kyphosis and paravertebral ossification [27,28]. Future prospective studies may further elucidate any differences in clinical outcomes in patients with either myelopathy or radiculopathy treated with CDR.
Myelopathy was identified as a risk factor for failure to achieve MCID for VAS-Neck within 3 months of CDR but not for the late postoperative period—this may support the idea that the spinal cord is more sensitive to compressive insult compared with the nerve roots, at least within the early postoperative period, and that it may take longer for patients presenting with myelopathy to experience improvement of neck pain compared with those with radiculopathy. Many patients with myelopathy experience stepwise worsening of symptoms and decline in function often requiring surgery to prevent progression, compared with radiculopathy which is commonly a nonurgent indication for surgery [32]. Interestingly, Alluri et al. [14] compared MCID achievement rates for PROMs following CDR at multiple timepoints postoperatively between patients with radiculopathy versus myelopathy and found no significant difference in MCID achievement between the groups at each time-point. It is important to note that preoperative VAS-Neck scores for the failure to achieve MCID group were lower compared with those for the achievement of MCID group (mean of 5.6 compared with 7.0 respectively), which may partially explain the distribution of patients who failed to achieve versus achieved MCID in this group as patients with lower preop VAS scores may have less potential for improvement postoperatively. It is important to acknowledge the limitations of MCID with NDI and VAS in the setting of myelopathy. Classically, patients with radiculopathy may present with the major complaint of pain in nerve distributions, while those with myelopathy may present more insidiously with features such as loss of fine motor dexterity. While pain may be present with myelopathy, it is less of a hallmark finding in myelopathy than it is in the setting of radiculopathy. Considering PROMs like NDI and VAS attempt to characterize function and pain postoperatively, it is possible that patients who are more symptomatic have greater potential to experience greater improvement after surgery, therefore increasing their likelihood of achieving MCID. A minimal clinically important difference is only as valid as the PROM it describes, and while NDI and VAS scores are convenient in characterizing clinical outcomes, they may not fully capture patients with myelopathy alone for the reasons described above.
The multivariate analysis identified predominant neck pain as a risk factor for failure to achieve MCID for NDI in both early and late postoperative periods, and for VAS-Arm in the late postoperative period. Previous studies comparing MCID achievement rates following CDR in NDI and VAS-Arm have found that patients with predominant arm pain compared with predominant neck pain have statistically significant higher rates of MCID achievement [15,29]. Similarly, with regards to ACDF, Divi et al. [30] reported patients with predominant arm pain have significantly greater improvement in NDI scores compared with patients with predominant neck pain. These findings support the idea that axial neck pain may be harder to treat with cervical spine surgery compared with radicular pain, which importantly informs the approach to counseling patients with predominant neck pain about cervical spine surgery as the evidence suggests they may not improve as reliably compared with patients with predominant arm pain [31].
The current study identified preoperative anxiety as a risk factor for failure to achieve MCID for VAS-Neck in the late postoperative period. Anxiety in this study was extracted from patients’ charts under their medical histories —it is not specified whether this represents clinically diagnosed anxiety disorder or anxiety that was previously endorsed by the patient or by provider upon interview/exam. Pain is well-described in literature as a multifactorial construct with a significant mental health and affective component—mood disorders including anxiety and depression play major roles in the exacerbation of pain perception and contribute to the development of chronic postoperative pain [33–35]. Li et al. [36] investigated the impact of depression and anxiety on prognosis of CDR and identified a strong correlation between neck pain and quality of life, and depression and anxiety, which had negative effects on prognosis. In a prospective, single-institution study of 27 patients, Adogwa et al. [37] found that the pretreatment of anxiety before ACDF resulted in a significant reduction in postoperative neck pain at 6 months and 1 year postop. Together, these findings suggest that coexisting affective disorders like anxiety play an important role in cervical spine surgery outcomes, and that it may be beneficial to attempt to manage anxiety prior to surgery and counsel patients on postoperative expectations as those with anxiety may not improve as reliably as those without anxiety.
While the multivariate analysis did not identify two-level CDR (compared with single-level CDR) as an independent risk factor for failure to achieve MCID, CDR at level C5–C7 specifically (compared with C5–C6, the most common level replaced) was identified as a risk factor for failure to achieve MCID for NDI in the late postoperative period. Additionally, multivariate analysis did not identify CDR at C6–C7 as a risk factor for failure to achieve MCID. These findings may possibly be explained by the increased axial loading of prostheses at C5–C7 compared with only C5–C6 and supported by the thought that prosthesis at C6–C7 may demonstrate relative hypermobility compared with withher levels. Huppert et al. [38] found no major significant clinical difference between single and multi-level CDRs including NDI, VAS scores, improvement of mobility, and satisfaction after 2 years postCDR. Our findings suggest that two-level CDRs compared with single-level CDRs may not increase the risk for failure to achieve MCID, however, replacement at C5–C7 specifically compared with the more common C5–C6 may increase the risk for failure to achieve MCID.
There are several limitations of the present study that should be considered when interpreting our findings. While important for assessing postoperative outcomes through a clinically relevant approach, PROMs and MCID are based on self-reported data, presenting the inherent risk of various forms of bias, namely selection bias, which could affect MCID achievement rates and subsequently risk factors associated with failure to achieve MCID. Additionally, all cases evaluated in this study were performed by 4 surgeons at a single academic institution, which may limit the generalizability of our results to all CDRs. Follow-up rates varied for each of the three PROMs included across the two period of evaluation, with fewer patients completing PROMs in the late postoperative period compared with the earlier post-operative period, reducing the statistical power of the analyses. This also limited the ability to detect risk factors in subgroup analysis according to the diagnosis. Additionally, our group did not routinely collect mJOA scores during the study to assess the degree of myelopathy and determine whether neurological recovery was achieved postoperatively for myelopathic patients—future studies may include mJOA analyses to further characterize MCID and neurological recovery in patients with myelopathy undergoing CDR. Furthermore, MCID thresholds for NDI, VAS-Neck, and VAS-Neck used in this study were based on those established in previous studies, however, there currently does not seem to be a consensus for such values in the literature, and having different thresholds for these PROMs would affect MCID achievement rates and subsequent multivariate analyses. Additionally, it is possible that the PROMs result in this study may have been affected by the heterogeneity of biomechanical properties featured in the different prostheses used for CDR.
Conclusion
The results of our study demonstrate that the majority of patients undergoing primary 1- or 2-level CDRs achieve MCID in NDI, VAS-Neck, and VAS-Arm within 3 months and up to 2 years postop. Through multivariate analyses, our study identified several risk factors for failure to achieve MCID in PROMs following CDR including predominant neck pain, myelopathy, anxiety, and CDR at levels C5–C7. Operative time was identified as a predictor for achievement of MCID following CDR. These findings may help inform the approach to counseling patients on outcomes of CDR as the evidence suggests that those with the risk factors above may not improve as reliably after CDR.
Acknowledgments
No direct funding was received for this study. However, the study used REDCap (Research Electronic Data Capture) hosted at Weill Cornell Medicine Clinical and Translational Science Center supported by the National Center for Advancing Translational Science of the National Institute of Health (NIH) under award number: UL1 TR002384.
Footnotes
Declarations of Competing Interest
One or more of the authors declare financial or professional relationships on ICMJE-TSJ disclosure forms.
References
- [1].Theodore N Degenerative cervical spondylosis. N Engl J Med 2020;383(2):159–68. 10.1056/NEJMra2003558. [DOI] [PubMed] [Google Scholar]
- [2].Leven D, Meaike J, Radcliff K, Qureshi S. Cervical disc replacement surgery: indications, technique, and technical pearls. Curr Rev Musculoskelet Med 2017;10(2):160–9. 10.1007/s12178-017-9398-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Moatz B, Tortolani PJ. Cervical disc arthroplasty: pros and cons. Surg Neurol Int 2012;3(Suppl 3):S216–24. 10.4103/2152-7806.98582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Gao Y, Liu M, Li T, Huang F, Tang T, Xiang Z. A meta-analysis comparing the results of cervical disc arthroplasty with anterior cervical discectomy and fusion (ACDF) for the treatment of symptomatic cervical disc disease. J Bone Joint Surg Am 2013;95(6):555–61. 10.2106/JBJS.K.00599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Hu Y, Lv G, Ren S, Johansen D. Mid- to long-term outcomes of cervical disc arthroplasty versus anterior cervical discectomy and fusion for treatment of symptomatic cervical disc disease: a systematic review and meta-analysis of eight prospective randomized controlled trials. PLoS One 2016;11(2):e0149312. 10.1371/journal.pone.0149312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Davis RJ, Nunley PD, Kim KD, Hisey MS, Jackson RJ, Bae HW, et al. Two-level total disc replacement with Mobi-C cervical artificial disc versus anterior discectomy and fusion: a prospective, randomized, controlled multicenter clinical trial with 4-year follow-up results. J Neurosurg Spine 2015;22(1):15–25. 10.3171/2014.7.SPINE13953. [DOI] [PubMed] [Google Scholar]
- [7].Johansen TO, Sundseth J, Fredriksli OA, Andresen H, Zwart J, Kolstad F, et al. Effect of arthroplasty vs fusion for patients with cervical radiculopathy: a randomized clinical trial. JAMA Netw Open 2021;4 (8):e2119606. 10.1001/jamanetworkopen.2021.19606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Gutman G, Rosenzweig DH, Golan JD. Surgical treatment of cervical radiculopathy: meta-analysis of randomized controlled trials. Spine (Phila Pa 1976) 2018;43(6):E365–72. 10.1097/BRS.0000000000002324. [DOI] [PubMed] [Google Scholar]
- [9].McCormick JD, Werner BC, Shimer AL. Patient-reported outcome measures in spine surgery. J Am Acad Orthop Surg 2013;21(2):99–107. 10.5435/JAAOS-21-02-99. [DOI] [PubMed] [Google Scholar]
- [10].Vernon H The Neck Disability Index: state-of-the-art, 1991–2008. J Manipulative Physiol Ther 2008;31(7):491–502. 10.1016/j.jmpt.2008.08.006. [DOI] [PubMed] [Google Scholar]
- [11].MacDowall A, Skeppholm M, Robinson Y, Olerud C. Validation of the visual analog scale in the cervical spine. J Neurosurg Spine 2018;28(3):227–35. 10.3171/2017.5.SPINE1732. [DOI] [PubMed] [Google Scholar]
- [12].Sedaghat AR. Understanding the minimal clinically important difference (MCID) of patient-reported outcome measures. Otolaryngol Head Neck Surg 2019;161(4):551–60. 10.1177/0194599819852604. [DOI] [PubMed] [Google Scholar]
- [13].Narain AS, Hijji FY, Khechen B, Haws BE, Patel DV, Bohl DD, et al. Risk factors associated with failure to reach minimal clinically important difference in patient-reported outcomes following anterior cervical discectomy and fusion. Int J Spine Surg 2019;13(3):262–9. 10.14444/6035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Alluri RK, Vaishnav AS, Sivaganesan A, Albert TJ, Huang RC, Qureshi SA. Cervical disc replacement for radiculopathy versus myeloradiculopathy: an MCID analysis. Clin Spine Surg 2022;35(4):170–5. 10.1097/BSD.0000000000001313. [DOI] [PubMed] [Google Scholar]
- [15].Shahi P, Vaishnav AS, Lee R, Mai E, Steinhaus ME, Huang R, et al. Outcomes of cervical disc replacement in patients with neck pain greater than arm pain. Spine J 2022;22(9):1481–9. 10.1016/j.spinee.2022.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Rihn JA, Radcliff K, Hipp J, Vaccaro AR, Hilibrand AS, Anderson DG, et al. Radiographic variables that may predict clinical outcomes in cervical disk replacement surgery. J Spinal Disord Tech 2015;28(3):106–13. 10.1097/BSD.0b013e31826a0c84. [DOI] [PubMed] [Google Scholar]
- [17].Massel DH, Mayo BC, Bohl DD, Narain AS, Hijji FY, Fineberg SJ, et al. Improvements in neck and arm pain following an anterior cervical discectomy and fusion. Spine (Phila Pa 1976) 2017;42(14):E825–32. 10.1097/BRS.0000000000001979. [DOI] [PubMed] [Google Scholar]
- [18].Parker SL, Godil SS, Shau DN, Mendenhall SK, McGirt MJ. Assessment of the minimum clinically important difference in pain, disability, and quality of life after anterior cervical discectomy and fusion: clinical article. J Neurosurg Spine 2013;18(2):154–60. 10.3171/2012.10.SPINE12312. [DOI] [PubMed] [Google Scholar]
- [19].Andresen AK, Paulsen RT, Busch F, Isenberg-Jørgensen A, Carreon LY, Andersen MØ. Patient-reported outcomes and patient-reported satisfaction after surgical treatment for cervical radiculopathy. Global Spine J 2018;8(7):703–8. 10.1177/2192568218765398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Radcliff K, Davis RJ, Hisey MS, Nunley PD, Hoffman GA, Jackson RJ, et al. Long-term evaluation of cervical disc arthroplasty with the Mobi-CÓ Cervical disc: a randomized, prospective, multicenter clinical trial with seven-year follow-up. Int J Spine Surg 2017;11(4):31. 10.14444/4031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Alluri RK, Vaishnav AS, Fourman MS, Sivaganesan A, Lee R, Urakawa H, et al. Anterior cervical discectomy and fusion versus cervical disc replacement in patients with significant cervical spondylosis. Clin Spine Surg 2022;35(2):E327–32. 10.1097/BSD.0000000000001250. [DOI] [PubMed] [Google Scholar]
- [22].Heller JG, Sasso RC, Papadopoulos SM, Anderson PA, Fessler RG, Hacker RJ, et al. Comparison of BRYAN cervical disc arthroplasty with anterior cervical decompression and fusion: clinical and radiographic results of a randomized, controlled, clinical trial. Spine (Phila Pa 1976) 2009;34(2):101–7. 10.1097/BRS.0b013e31818ee263. [DOI] [PubMed] [Google Scholar]
- [23].Hisey MS, Bae HW, Davis R, Gaede S, Hoffman G, Kim K, et al. Multi-center, prospective, randomized, controlled investigational device exemption clinical trial comparing Mobi-C Cervical Artificial Disc to anterior discectomy and fusion in the treatment of symptomatic degenerative disc disease in the cervical spine. Int J Spine Surg 2014;8:7. 10.14444/1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Gornet MF, Lanman TH, Burkus JK, Hodges SD, McConnell JR, Dryer RF, et al. Cervical disc arthroplasty with the Prestige LP disc versus anterior cervical discectomy and fusion, at 2 levels: results of a prospective, multicenter randomized controlled clinical trial at 24 months. J Neurosurg Spine 2017;26(6):653–67. 10.3171/2016.10.SPINE16264. [DOI] [PubMed] [Google Scholar]
- [25].Phillips FM, Lee JY, Geisler FH, Cappuccino A, Chaput CD, DeVine JG, et al. A prospective, randomized, controlled clinical investigation comparing PCM cervical disc arthroplasty with anterior cervical discectomy and fusion. 2-year results from the US FDA IDE clinical trial. Spine (Phila Pa 1976) 2013;38(15):E907–18. 10.1097/BRS.0b013e318296232f. [DOI] [PubMed] [Google Scholar]
- [26].Riew KD, Buchowski JM, Sasso R, Zdeblick T, Metcalf NH, Anderson PA. Cervical disc arthroplasty compared with arthrodesis for the treatment of myelopathy. J Bone Joint Surg Am 2008;90(11):2354–64. 10.2106/JBJS.G.01608. [DOI] [PubMed] [Google Scholar]
- [27].Zeng C, Tian W, Liu B, Fan M. The assessment of paravertebral ossification progression after cervical disc arthroplasty based on CT images: a long-term follow-up. Orthop Surg 2020;12(6):1760–7. 10.1111/os.12743. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Cao JM, Zhang YZ, Shen Y, Ding WY. Complications of Bryan cervical disc replacement. Orthop Surg 2010;2(2):86–93. 10.1111/j.1757-7861.2010.00069.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Nie JW, Hartman TJ, MacGregor KR, Oyetayo OO, Zheng E, Singh K. Influence of predominant neck versus arm pain on clinical outcomes in cervical disc replacement. World Neurosurg 2023;169: e206–13. 10.1016/j.wneu.2022.10.107. [DOI] [PubMed] [Google Scholar]
- [30].Divi SN, Goyal DKC, Woods BI, Nicholson KJ, Salmons HI, Galetta MS, et al. How do patients with predominant neck pain improve after anterior cervical discectomy and fusion for cervical radiculopathy? Int J Spine Surg 2022;16(2):240–6. 10.14444/8212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Malcolm GP. Surgical disorders of the cervical spine: presentation and management of common disorders. J Neurol Neurosurg Psychiatry 2002;73(Suppl 1):i34–41. 10.1136/jnnp.73.suppl_1.i34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Rao R Neck pain, cervical radiculopathy, and cervical myelopathy: pathophysiology, natural history, and clinical evaluation. J Bone Joint Surg Am 2002;84(10):1872–81. 10.2106/00004623-200210000-00021. [DOI] [PubMed] [Google Scholar]
- [33].Woo AK. Depression and anxiety in pain. Rev Pain 2010;4(1):8–12. 10.1177/204946371000400103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Melzack R, Torgerson WS. On the language of pain. Anesthesiology 1971;34(1):50–9. 10.1097/00000542-197101000-00017. [DOI] [PubMed] [Google Scholar]
- [35].Strøm J, Bjerrum MB, Nielsen CV, Thisted CN, Nielsen TL, Laursen M, et al. Anxiety and depression in spine surgery-a systematic integrative review. Spine J 2018;18(7):1272–85. 10.1016/j.spinee.2018.03.017. [DOI] [PubMed] [Google Scholar]
- [36].Li S, Qi M, Yuan W, Chen H. The impact of the depression and anxiety on prognosis of cervical total disc replacement. Spine (Phila Pa 1976) 2015;40(5):E266–71. 10.1097/BRS.0000000000000743. [DOI] [PubMed] [Google Scholar]
- [37].Adogwa O, Elsamadicy AA, Cheng J, Bagley C. Pretreatment of anxiety before cervical spine surgery improves clinical outcomes: a prospective, single-institution experience. World Neurosurg 2016;88: 625–30. 10.1016/j.wneu.2015.11.014. [DOI] [PubMed] [Google Scholar]
- [38].Huppert J, Beaurain J, Steib JP, Bernard P, Dufour T, Hovorka I, et al. Comparison between single- and multi-level patients: clinical and radiological outcomes 2 years after cervical disc replacement. Eur Spine J 2011;20(9):1417–26. 10.1007/s00586-011-1722-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Goh GS, Liow MHL, Yeo W, Ling ZM, Guo C, Yue W, et al. Predictors of outcomes after single-level anterior cervical discectomy and fusion for cervical spondylotic myelopathy: a multivariate analysis. Clin Spine Surg 2020;33(10):E525–32. 10.1097/BSD.0000000000000997. [DOI] [PubMed] [Google Scholar]