In most patients with cervical degenerative radiculopathy, the identification of the nerve root compression site can be accomplished by the careful evaluation of clinical symptoms, physical examination findings, and imaging data.1 multilevel decompression may be necessary.2–4 5 6
When there is no diagnostic confidence regarding the site of the responsible lesion, electromyography (EMG) or selective nerve root block (SNRB) is sometimes used. However, these invasive techniques do not always provide an accurate diagnosis.7,8 cervical radiculopathy , whose physical and imaging findings do not match those of intervertebral stenosis, multilevel decompression may be required. This would be more invasive but is expected to comprehensively decompress all suspected nerve root compression areas, including those with polyradiculopathy. However, whether this approach provides the same surgical outcome as that of a single-level decompression remains unknown. This study aimed to compare the clinical outcomes between single-level and multilevel intervertebral decompression for cervical degenerative radiculopathy.
MATERIALS AND METHODS
Patient Samples
We retrospectively extracted data from 509 patients who underwent cervical spinal surgery for cervical radiculopathy registered in 11 hospitals participating in this prospective multicenter study between 2017 and 2020. The inclusion criteria were degenerative cervical radiculopathy without cervical myelopathy and myeloradiculopathy. Each patient had a cervical radiculopathy diagnosis on their preoperative surveillance record, which included other demographic data. Among the aforementioned patients, those with complete demographic data and patient-oriented questionnaires preoperatively and one year postoperatively were included. Patients with a diagnosis of spinal tumors, rheumatoid arthritis, congenital abnormalities, infection, or trauma were excluded. Patients were divided into two groups single-level (SLDG) or multilevel (MLDG) intervertebral decompression between C3/C4, C4/C5, C5/C6, C6/C7, and C7/T1.
Data Collection
Demographic data, including age, sex, body mass index (BMI), American Society of Anesthesiologists (ASA) physical status classification, diabetes mellitus (DM), smoking status, history of disk herniation or osteoarthritis, surgical procedure, and number of intervertebral decompression levels, were collected. Radiculopathy diagnosis, surgical indication, surgical procedure, and number of decompression levels were decided by surgeons at each institute. Surgical procedures for patients were classified into anterior decompression and fusion, posterior decompression, and posterior decompression and fusion.
Clinical Outcomes
Patient-reported outcome (PRO) scores were collected preoperatively with questionnaires, including the Neck Disability Index (NDI ) and Numerical Rating Scale (NRS ) for pain and numbness in the neck, upper back, and arms (Figure 1 ). At the one-year follow-up, patients were encouraged to complete the same questionnaires for postoperative PRO. In addition, the NDI improvement rate, defined as (postoperative NDI score−preoperative NDI score)/(preoperative NDI score)×100, and changes in NRS scores, defined as preoperative NRS score−postoperative NRS score, were analyzed.
Figure 1: Areas of pain and numbness were divided into the neck (1), upper back (2), and arms (3). Pain and numbness intensity were measured using Numerical Rating Scale.
Statistical Analyses
Demographic data and preoperative and postoperative PRO scores were compared between SLDG and MLDG using an unpaired t test or Mann-Whitney U test for continuous variables and a χ2 test for categorical variables, as appropriate. To adjust for preoperative background factors, propensity score matching was performed. Propensity scores were calculated from logistic regression models. In the present study, demographic data, including age, sex, BMI, ASA classification, DM, smoking status, history of disk herniation or osteoarthritis, surgical procedure, and the number of decompressed spinal levels, as well as preoperative PRO scores, including NDI scores, NRS scores for pain (neck, upper back, and arms), and NRS scores for numbness in these sites, were used for one-to-one propensity score matching between SLDG and MLDG. R statistical software, version 2.8.1 (The R Foundation for Statistical Computing) was used for statistical analyses. A P value <0.05 was considered statistically significant.
RESULTS
Comparison of Preoperative Characteristics
Of the 509 patients who underwent surgery for pure cervical radiculopathy and did not have myelopathy or myeloradiculopathy diagnoses, 357 patients (288 males and 69 females; mean age at surgery, 55.4 yr) with complete demographic data as well as preoperative and one-year postoperative PROs were included. Among these patients, 231 and 126 received single-level and multilevel intervertebral decompressions between C3/C4 and C7/T1, respectively.
Multilevel decompression was performed between two intervertebral segments in 89 cases, between three intervertebral segments in 24 cases, between four intervertebral segments in 12 cases, and between five intervertebral segments in one case (Table 1 ). For single-level decompression , the anterior method with 100% fusion surgery was used in 28% of cases, while the posterior method with 2% fusion surgery was used in 72% of cases. For multilevel decompression , the anterior method with 100% fusion surgery was used in 23% of cases, while the posterior method with 5% fusion surgery was used in 77% of cases.
TABLE 1 -
Preoperative Demographic Data
Mean (SD)
P
Total
Single-level
Multilevel
N (2, 3, 4, 5 levels)
357
231
126 (89, 24, 12, 1)
Age (yr)
55.4 (10.8)
54.4 (10.6)
57.2 (11.0)
0.020
Sex: male (%)
80.7
80.1
81.7
0.780
BMI (kg/m2 )
24.1 (3.6)
23.86 (3.6)
24.55 (3.7)
0.084
ASA grade (grade 1:2:3:4) (%)
1.7 (0.5)
36:60:4:0
21:72:6:0
0.012
Diabetes mellitus (%)
12.9
11.7
15.1
0.409
Current smoker (%)
16.8
13.9
22.2
0.054
Disk herniation (%)
27.7
34.6
15.1
<0.001
Surgical procedure (PD:ADF:PDF) (%)
71:26:3
70:28:2
72:23:5
0.256
Bold values indicate P <0.05.
ADF indicates anterior decompression and fusion; ASA, American Society of Anesthesiologists Classification; BMI, body mass index; PD, posterior decompression; PDF, posterior decompression and fusion.
Comparison of Unmatched Preoperative and Postoperative Data
MLDG included patients who were older (P= 0.02), those who had a higher ASA grade (P= 0.012), and fewer patients with a history of disk herniation (P <0.001) than SLDG. The preoperative NRS scores for pain in the neck (P= 0.002) and arms (P= 0.041) were significantly higher in SLDG than in MLDG. Other parameters, including the NDI score; NRS scores for upper back pain; and NRS scores for numbness in the neck, upper back, and arms, tended to be worse in the SLDG than in MLDG; however, the difference was not statistically significant.
Before propensity matching, the postoperative NDI improvement rate (P= 0.001); changes in NRS scores for pain in the neck (P =0.003), upper back (P= 0.040), and arms (P= 0.028); and changes in NRS scores for arm numbness (P= 0.025) were better in the SLDG than in MLDG. The postoperative NRS scores for arm numbness (P= 0.006) were lower in SLDG than in MLDG (Table 2 ).
TABLE 2 -
Comparison of Unmatched Preoperative and Postoperative Data
Mean (SD)
P
Single-level
Multilevel
Preoperative PRO (NDI and NRS scores)
NDI scores
16.9 (8.0)
15.2 (7.9)
0.065
Neck pain
4.3 (2.9)
3.3 (3.1)
0.002
Upper back pain
3.3 (3.3)
2.7 (3.1)
0.116
Arm pain
5.3 (3.1)
4.6 (3.4)
0.041
Neck numbness
1.6 (2.7)
1.6 (2.7)
0.840
Upper back numbness
1.3 (3.5)
1.1 (2.2)
0.605
Arm numbness
5.5 (3.1)
5.4 (3.1)
0.997
Postoperative PRO (NDI and NRS scores)
NDI scores
8.0 (7.5)
8.3 (6.6)
0.721
NDI improvement rate (%)
50.0 (44.6)
26.4 (91.2)
0.001
Neck pain
1.8 (2.4)
1.8 (2.2)
0.909
Change in NRS scores of neck pain
2.5 (3.0)
1.5 (3.0)
0.003
Upper back pain
1.2 (2.0)
1.5 (2.4)
0.351
Change in NRS scores of upper back pain
2.1 (3.3)
1.3 (3.7)
0.040
Arm pain
1.9 (2.5)
2.1 (2.5)
0.559
Change in NRS scores of arm pain
3.4 (3.67)
2.5 (3.5)
0.028
Neck numbness
0.7 (1.9)
0.8 (1.8)
0.910
Change in NRS scores of neck numbness
0.9 (2.8)
0.8 (2.8)
0.785
Upper back numbness
0.4 (1.2)
0.6 (1.6)
0.332
Change in NRS scores of upper back numbness
0.9 (3.6)
0.5 (2.7)
0.375
Arm numbness
2.1 (2.6)
2.9 (2.8)
0.006
Change in NRS scores of arm numbness
3.3 (3.4)
2.5 (3.2)
0.025
Bold values indicate P <0.05.
NDI indicates Neck Disability Index; NRS , Numerical Rating Scale; PRO, patient-reported outcome.
Comparison of Demographic Data and Clinical Outcomes With Propensity Score Matching
Preoperative age, sex, BMI, ASA classification, DM, smoking status, history of disk herniation or osteoarthritis, surgical technique, NDI scores, and NRS scores for pain and numbness in the neck, upper back, and arms were matched by propensity scores, resulting in 112 matched pairs of patients in SLDG and MLDG, respectively. Postoperatively, SLDG had a better NDI improvement rate (P= 0.029) and lower NRS score for arm numbness (P =0.037) than the MLDG. The postoperative NDI scores and other NRS scores tended to be better in SLDG than in MLDG; however, there was no statistically significant difference (Table 3 ).
TABLE 3 -
Comparison of Demographic Data and Clinical Outcomes With Propensity Score–matched Analysis
Mean (SD)
P
Single-level
Multilevel
N
112
112
Age (yr)
57.5 (11.2)
56.6 (10.9)
0.628
Sex: male (%)
80.4
81.2
>0.999
BMI (kg/m2 )
24.5 (3.6)
24.3 (3.5)
0.688
ASA grade (grade 1:2:3:4)
22:73:5:0
24:70:6:0
0.768
Diabetes mellitus (%)
15.2
15.2
>0.999
Current smoker (%)
17.0
19.6
0.730
Disk herniation (%)
15.2
17.0
0.856
Surgical procedure (PD:ADF:PDF) (%)
71:27:3
73:22:5
0.613
Preoperative PRO (NDI and NRS scores)
NDI scores
15.6 (8.2)
15.3 (7.9)
0.804
Neck pain
3.7 (3.1)
3.5 (3.1)
0.589
Upper back pain
3.1 (3.2)
2.8 (3.1)
0.485
Arm pain
4.7 (3.2)
4.8 (3.3)
0.745
Neck numbness
1.8 (2.8)
1.6 (2.8)
0.681
Upper back numbness
1.5 (4.4)
1.1 (2.2)
0.475
Arm numbness
5.3 (3.3)
5.5 (3.1)
0.691
Postoperative PRO (NDI and NRS scores)
NDI scores
7.2 (7.4)
8.0 (6.4)
0.411
NDI improvement rate (%)
50.5 (47.1)
28.8 (93.0)
0.029
Neck pain
1.5 (2.2)
1.8 (2.2)
0.214
Change in NRS scores of neck pain
2.2 (3.0)
1.7 (3.1)
0.149
Upper back pain
1.2 (2.1)
1.5 (2.5)
0.296
Change in NRS scores of upper back pain
1.9 (3.2)
1.3 (3.7)
0.185
Arm pain
1.7 (2.4)
2.2 (2.5)
0.187
Change in NRS scores of arm pain
3.0 (3.6)
2.7 (3.6)
0.540
Neck numbness
0.7 (1.7)
0.7 (1.8)
0.731
Change in NRS scores of neck numbness
1.1 (3.0)
0.9 (2.8)
0.547
Upper back numbness
0.5 (1.3)
0.6 (1.6)
0.614
Change in NRS scores of upper back numbness
1.0 (4.3)
0.5 (2.6)
0.373
Arm numbness
2.1 (2.7)
2.9 (2.8)
0.037
Change in NRS scores of arm numbness
3.2 (3.5)
2.6 (3.3)
0.185
Bold values indicate P <0.05.
ADF indicates anterior decompression and fusion; ASA, American Society of Anesthesiologists Classification; BMI, body mass index; NDI , Neck Disability Index; NRS , Numerical Rating Scale; PD, posterior decompression; PDF, posterior decompression and fusion; PRO, patient-reported outcome.
DISCUSSION
Despite propensity score matching for preoperative physical status and disability, the surgical outcomes in SLDG regarding the NDI improvement rate instead of the NDI score itself and the postoperative NRS scores for arm numbness were better than those undergoing MLDG. To the best of our knowledge, this study is the first to assess the effect of the number of decompression levels on the clinical outcomes of pure cervical radiculopathy in a multicenter cohort using propensity score matching .
In pure cervical radiculopathy , outcome comparison based on different surgical levels has not been previously performed. However, in a previous systematic review, the clinical outcomes of different surgical procedures for the limited symptoms of cervical radiculopathy showed no significant differences.9 et al 10 NDI and Visual Analog Scale scores for pain in the neck and arms between one or two and three or four levels of intervertebral decompression. Similar results were reported by Choi et al 11 et al .12
In contrast to these prior reports, multilevel decompression for cervical degenerative radiculopathy did not result in equivalent clinical outcome to that of single-level decompression in our study. This discrepancy could be explained by more severe degenerative process in the multilevel group and consequent difficulties decompressing the intervertebral foramen. In addition, since it takes many years for degeneration to extend to multiple intervertebral levels or distal to the intervertebral foramen, irreversible damage to the nerve roots may occur due to chronic nerve root compression. Furthermore, since the imaging evaluation of the intervertebral foramen for cervical radiculopathy is sometimes mismatched with physical findings,6,14
The significant difference in residual numbness in the distal body regions (eg, the arms), rather than the proximal (neck and upper back) areas suggests that residual nerve damage symptoms might be present distally. In fact, the length-dependent pattern of nerve axonopathy suggested that numbness in the distal area was more likely to occur in the distal part of the nerve,19 cervical radiculopathy surgery are not infrequent, with estimations as high as 26%.20
Our study has several limitations. First, determinations regarding the diagnosis, the surgical technique, the surgical site of the intervertebral level, and the number of decompressed intervertebral levels were surgeon-dependent, and anesthetic care as well as rehabilitation were center dependent, with no uniformity in treatment criteria. The decision to include SNRBs or EMGs as diagnostic tools was based on the surgeon’s discretion, and these diagnostic tools may contribute to diagnostic accuracy. Second, while the duration of preoperative symptoms might be associated with surgical outcomes, this parameter was not matched in this study among the 11 centers because the data were from a prospective multicenter study group, and the retrospective addition of data on morbidity duration from medical records may reduce data reliability, including data interpretation and handling of recurrent or additional symptoms. However, there were no significant differences in surgical outcomes in our subset analysis of patients with different symptom durations. Further, given that there are some conflicting reports regarding the association between surgical outcomes and symptom duration, we decided that matching on this parameter was not ideal. Third, the follow-up period of one year postoperatively was relatively short. This time window was used not only to compare clinical outcomes but also to determine whether the diagnosis at the intervertebral level was correct. However, it is conceivable that long-term outcomes might differ from the present findings. Finally, the study was a surveillance-based retrospective multicenter study, with a decline in the survey collection rate due to missing data and potentially a surveillance bias.
CONCLUSIONS
In patients with cervical radiculopathy , those receiving single-level intervertebral decompression demonstrated greater improvement in clinical outcomes than those undergoing multilevel decompressions. Numbness persisted to a greater degree in distal body regions (eg, the arms; as opposed to the neck and upper back) in patients receiving multilevel intervertebral decompression. In cases of widespread degeneration requiring multiple intervertebral decompressions, the possibility of poor postoperative symptomatic improvement and residual numbness in the upper extremities should be considered.
Key Points
A retrospective multicenter cohort study with propensity score matching was performed in patients with pure cervical radiculopathy to compare the clinical outcomes of single-level (SLD) and multilevel (MLD) intervertebral decompression after adjusting for baseline characteristics, including the preoperative scores of the NDI and NRS for pain and numbness in the neck, upper back, and arms.
The surgical outcomes of SLD showed greater improvement in clinical outcomes than those of MLD. SLD demonstrated worse preoperative NDI scores and higher preoperative NRS scores than MLD. However, SLD yielded a higher postoperative NDI improvement rate (P =0.029) and lower postoperative NRS score for arm numbness (P =0.037) compared with MLD.
Other outcomes, including postoperative NDI scores, tended to be more favorable in the SLD than in the MLDG, yet no statistically significant difference was detected.
The numbness remained in the distal (arms) areas rather than the central (neck and upper back) areas in the MLDG.
Acknowledgments
The authors thank Editage (http://www.editage.com) for editing and reviewing this manuscript for English language.
References
1. Radhakrishnan K, Litchy WJ, O’Fallon WM, et al. Epidemiology of
cervical radiculopathy . A population-based study from Rochester, Minnesota, 1976 through 1990. Brain. 1994;117:325–335.
2. Matsumoto M, Fujimura Y, Suzuki N, et al. MRI of cervical intervertebral discs in asymptomatic subjects. J Bone Joint Surg Br. 1998;80:19–24.
3. Siivola SM, Levoska S, Tervonen O, et al. MRI changes of cervical
spine in asymptomatic and symptomatic young adults. Eur
Spine J. 2002;11:358–363.
4. Boden SD, McCowin PR, Davis DO, et al. Abnormal magnetic-resonance scans of the cervical
spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am. 1990;72:1178–1184.
5. Slipman CW, Plastaras CT, Palmitier RA, et al. Symptom provocation of fluoroscopically guided cervical nerve root stimulation. Are dynatomal maps identical to dermatomal maps?
Spine (Phila Pa 1976). 1998;23:2235–2242.
6. Redebrandt HN, Brandt C, Hawran S, et al. Clinical evaluation versus magnetic resonance imaging findings in patients with radicular arm pain—a pragmatic study. Health Sci Rep. 2022;5:e589.
7. Anderberg L, Annertz M, Rydholm U, et al. Selective diagnostic nerve root block for the evaluation of radicular pain in the multilevel degenerated cervical
spine . Eur
Spine J. 2006;15:794–801.
8. Partanen J, Partanen K, Oikarinen H, et al. Preoperative electroneuromyography and myelography in cervical root compression. Electromyogr Clin Neurophysiol. 1991;31:21–26.
9. Broekema AEH, Groen RJM, Simões de Souza NF, et al. Surgical interventions for
cervical radiculopathy without myelopathy: a systematic review and meta-analysis. J Bone Joint Surg Am. 2020;102:2182–2196.
10. Basques BA, Louie PK, Mormol J, et al. Multi- versus single-level anterior cervical discectomy and fusion: comparing sagittal alignment, early adjacent segment degeneration, and clinical outcomes. Eur
Spine J. 2018;27:2745–53.
11. Choi MK, Kim SB, Park CK, et al. Comparison of the clinical and radiologic outcomes obtained with single- versus two-level anterior cervical decompression and fusion using stand-alone PEEK cages filled with allograft. Acta Neurochir (Wien). 2016;158:481–487.
12. Hatta Y, Shiraishi T, Hase H, et al. Is posterior spinal cord shifting by extensive posterior decompression clinically significant for multisegmental cervical spondylotic myelopathy?
Spine (Phila Pa 1976). 2005;30:2414–2419.
13. Gu BS, Park JH, Seong HY, et al. Feasibility of posterior cervical foraminotomy in cervical
foraminal stenosis : prediction of surgical outcomes by the foraminal shape on preoperative computed tomography.
Spine (Phila Pa 1976). 2017;42:E267–E271.
14. Kuijper B, Beelen A, van der Kallen BF, et al. Interobserver agreement on MRI evaluation of patients with
cervical radiculopathy . Clin Radiol. 2011;66:25–29.
15. Murphy DR, Hurwitz EL, Gerrard JK, et al. Pain patterns and descriptions in patients with radicular pain: does the pain necessarily follow a specific dermatome? Chiropr Osteopat. 2009;17:9.
16. Rainville J, Laxer E, Keel J, et al. Exploration of sensory impairments associated with C6 and C7 radiculopathies.
Spine J. 2016;16:49–54.
17. Guday E, Bekele A, Muche A. Anatomical study of prefixed versus postfixed brachial plexuses in adult human cadaver. ANZ J Surg. 2017;87:399–403.
18. Sasso RC, Macadaeg K, Nordmann D, et al. Selective nerve root injections can predict surgical outcome for lumbar and
cervical radiculopathy : comparison to magnetic resonance imaging. J Spinal Disord Tech. 2005;18:471–478.
19. Schaumburg HH, Zotova E, Raine CS, et al. The rat caudal nerves: a model for experimental neuropathies. J Peripher Nerv Syst. 2010;15:128–139.
20. Sampath P, Bendebba M, Davis JD, et al. Outcome in patients with
cervical radiculopathy . Prospective, multicenter study with independent clinical review.
Spine (Phila Pa 1976). 1999;24:591–597.
