Anterior cervical discectomy and fusion with and without plating versus laminectomy with and without fusion for multilevel cervical spondylotic myelopathy: a prospective observational study

Background Cervical spondylotic myelopathy (CSM) is a chronic progressive spinal compression that usually accompanies age-related degeneration of the cervical spine and represents one of the most common causes of spinal cord dysfunction. Surgical intervention is the foundation of management in symptomatic cases, but the approach of choice is constantly contentious. The aim of this study is to evaluate and appraise different surgical approaches for multilevel cervical spondylotic myelopathy (anterior cervical discectomy and fusion (ACDF) with or without plating and laminectomy with and without fusion) regarding the perioperative data, clinical outcomes, complications rates and radiographic parameters. Results In this prospective study, sixty patients in four matched cohorts were followed up. The first group ( n = 15)


Background
Cervical spondylotic myelopathy (CSM) stands as the prevailing cause for spinal cord dysfunction globally.It is a degenerative condition characterized by nontraumatic, gradual and persistent compression of the cervical spinal cord [1].The prevalence of the disease is higher in males and in the Asian population compared to females and Caucasians, respectively [1].Evidence from the natural history of CSM indicates that 20-60% of patients may experience neurological deterioration.Therefore, upon diagnosing CSM, decompression surgery should be undertaken to prevent worsening of neural function [2,3].
Certain factors must be considered when deciding the best approach for a given patient, namely source and location of the compression, number of segments, sagittal alignment, specific anatomy, patient's complaints and/or history of previous neck surgeries [4].
Anterior approaches are generally preferred when the compression involves three or fewer levels.It allows for direct decompression of anterior pathologies and correction of cervical kyphosis [5].Plating increases fusion rates with anterior approaches, especially if operating on more than two levels [6].
Posterior approaches are generally reserved for three or more compression levels.Decompression is achieved either directly if the compressive element is posteriorly located or indirectly by allowing the cervical cord to drift posteriorly away from the compressive anterior pathologies [7].Posterior approaches are most beneficial in the absence of kyphosis; posterior surgery in the presence of kyphotic alignment is associated with worse neurological outcomes [8].In addition to avoiding kyphotic malalignment, fusion has the added benefit of alleviating axial neck pain [7].
The aim of this study is to evaluate and appraise different surgical approaches for multilevel cervical spondylotic myelopathy (anterior cervical discectomy and fusion with or without plating and Laminectomy with and without fusion) regarding the perioperative data, clinical outcomes, complications rates and radiographic parameters.

Methods
This prospective observational study included a total of sixty patients with multilevel CSM.A written informed consent was obtained from all patients participating in this study.
A comprehensive analysis of all patients undergoing cervical spine surgery for CSM in the included hospitals was conducted.Patients fulfilling our inclusion criteria were included in the study.Our inclusion and exclusion criteria were designed to ensure that all patients included in the study are matched preoperatively regarding demographic characteristics, preoperative pain and myelopathy scores and preoperative sagittal alignment values.Eighteen patients have been excluded either for not meeting our inclusion criteria or for refusal to participate in the study.The enrolled sixty cases have been assigned to their groups.Patients undergoing anterior approach procedures have been assigned to Group A and have been further divided into Group A1 for ACDF only patients and Group A2 for ACDF with plating.No patients were lost during follow-up in Group A1 while one patient was lost from Group A2.Patients undergoing posterior approach procedures have been assigned to Group B and have been further divided into Group B1 for laminectomy alone and Group B2 for laminectomy with fusion.A single patient has been lost during follow-up in Group B1 while two patients have been lost in Group B2.All groups completed 12-month follow-up.A flow chart is shown in Fig. 1.
Inclusion criteria included: a clear picture of CSM by history and examination of degenerative origin in the subaxial cervical spine.MRI (magnetic resonance imaging) findings confirm the presence of cord compression at two or more anatomical levels.Age ≥ 18 years.
Exclusion criteria included: patients whose primary symptom is axial or radicular pain, but not myelopathy, patients with other non-degenerative pathologies such as cervical trauma, tumors, infections or rheumatoid arthritis, or pathologies involving the craniocervical junction, patients with preoperative cervical kyphosis and patients who underwent previous cervical spine surgery.
Preoperatively, the following data were extracted from the patients themselves, their sheets and/or preoperative images: history and physical examination, preoperative pain (axial or radicular) using the visual analogue score (VAS) [9], functional scoring, using the modified Japanese orthopedic association score (mJOA score) [10] and neck disability index (NDI) [11].If not available, they were extracted using structured telephone interviews, using a pre-designed interview protocol.Quality of life was assessed using the medical outcomes study short form health survey (SF-36) [12].
MRI studies, plain X-ray, including A-P (antero-posterior) and flexion/extension lateral views, and/or CT (computed tomography) scan (to exclude OPLL (ossification of posterior longitudinal ligament)) were done for all patients.Assessment of cervical sagittal alignment from C2 to C7 is carried out using Cobb method [13].
Postoperatively, follow-up visits were done at 3, 6 and 12 months postoperatively.An independent physician not related to the surgical procedures oversaw patient's evaluation.The following parameters were examined: long-term complications, postoperative mJOA, NDI, VAS and SF-36 as well as recovery ratio (RR), which is calculated by the equation: RR (%) = postoperative mJOA score−preoperative mJOA score/(18−Preoperative mJOA score) × 100.
Radiological outcomes assessment was done utilizing first, the change in lordosis using Cobb method on Plain X-ray lateral view, and second, radiological evaluation of fusion to detect pseudoarthrosis and/or instability; all postoperative X-rays and CT scans if available were independently evaluated by a neuroradiologist.Pseudoarthrosis was diagnosed if one of the following was observed: subluxation across the previous fusion site, as demonstrated on flexion-extension X-rays, or lucency across the previous fusion level.Statistical analysis: The collected data were coded, tabulated, and statistically analyzed using IBM SPSS statistics (Statistical Package for Social Sciences) software version 28.0,IBM Corp., Chicago, USA, 2021.Quantitative data were tested for normality using Shapiro-Wilk test, then described as mean ± SD (standard deviation) as well as minimum and maximum of the range and then compared using analysis of variance (ANOVA) test (three independent groups).Qualitative data were described as number and percentage and compared using Chisquare test as well as Fisher's exact test for variables with small, expected numbers.Bonferroni test was used for post hoc comparisons.The level of significance was taken at p-value < 0.050 was significant, otherwise was non-significant.

Results
A total of fifty-six patients were enrolled in this study and were assigned to four groups: group A1 (n = 15) patients underwent ACDF only, group A2 (n = 14) underwent ACDF with plating, group B1 (n = 14) underwent laminectomy only, and group B2 (n = 13) underwent laminectomy with lateral mass fusion.Regarding the demographics of the patients, there were no statistically significant differences between the four groups regarding age (p value = 0.273), sex (p value = 0.807) or smocking (p value = 0.999) (Table 1).
The baseline clinical characteristics between the studied groups show no significant differences as well.These parameters included presenting symptoms and signs, duration of illness and number of levels as shown in Table 2.
Regarding the outcome parameters, all the four studied groups showed clinical improvement with no statistically significant differences between them in both the change in mJOA score (p value = 0.973) and recovery ratio (p value = 0.913).Posterior approach groups scored non-significantly higher in postoperative NDI change (p value = 0.209) and scored non-significantly lower in postoperative VAS change (p value = 0.618) (Figs. 2, 3, 4 and 5).
Anterior approach groups scored higher in postoperative SF-35 values and its change, with no statistically significant difference between plated and non-plated groups.
There was no statistically significant difference between the studied groups regarding preoperative lordosis (p value = 0.852).Postoperative lordosis and its change, however, were significantly higher in anterior approach groups with no difference between either technique.
Intraoperative blood loss was significantly higher in laminectomy and laminectomy with fusion groups with no difference between them.Similarly, hospital stay was longer in posterior approach groups, but it was not statistically significant (p value = 0.127).Plated anterior approach and laminectomy with fusion groups both had the highest operative duration (p value < 0.001) with no statistically significant difference between them (Table 4).

Discussion
Surgical decompression is the only effective treatment option for multilevel cervical spondylotic myelopathy [14,15].Decompression can be performed either from the anterior approach, from the posterior approach, or via combined approaches.In this study, we compared two anterior decompression techniques: ACDF with and without anterior plating, versus two posterior decompression options; laminectomy with and without fusion.Every option has been evaluated regarding their functional and radiological outcomes.There was no statistically significant difference between the studied groups regarding their demographic  characteristics, symptoms, or signs.The operated levels were either three or four levels in all groups with almost 50:50 ratio except for two cases with only two levels in groups A1 and A2 one case for each group.These data are similar to Asher and Fehlings who both had relatively similar duration of symptoms (26 months) in both anterior and posterior groups [16,17].The mean operated levels were relatively fewer for the anterior group (2-3 levels) in Fehlings study compared to posterior group (3-4 levels) [16] While most of Asher's patients had 3-4 levels on both groups, which was similar to our distribution [17].
Many past studies comparing anterior and posterior approaches have been criticized for carrying out their neurological evaluation and outcome assessment using only tools like the JOA score (Japanese orthopedic association score) only and neglecting postoperative pain and quality of life assessment.In our study, we used the mJOA score as well as the NDI which are better at assessing neurological functions [19].We also used the VAS score and SF-36 form to measure the postoperative pain and quality of life.
Our study found no statistically significant difference between the studied groups regarding the preoperative, postoperative or change in mJOA as well as recovery ratio.Likewise, there was no significant difference between the studied groups regarding the preoperative and postoperative values for NDI score; however, the change in NDI was non-significantly higher in laminectomy only and laminectomy with fusion than in ACDF only and ACDF with plating.These results are in agreement with previous systematic reviews and meta-analyses [20][21][22][23].In a recently published study by Kato and colleagues [24] that analyzed 435 patients enrolled in two international AOSpine cervical spondylotic myelopathy studies, the authors found that anterior and posterior groups did not differ significantly in terms of the postoperative mJOA, or NDI scores.More recently, Asher and coworkers [17] performed a retrospective analysis of prospectively collected data from multiple surgical practices across the USA with over 250 patients undergoing three to five level fusion surgery for degenerative cervical myelopathy.They concluded that mJOA and NDI did not differ significantly between anterior and posterior groups.To investigate patterns and speed of neurological recovery after anterior versus posterior decompression, Ren and colleagues [25] using mJOA and recovery ratio as functional outcome assessment tools reached a conclusion that neurological recovery in both groups improved similarly, reached statistical significance at three months, recovered rapidly in the first nine months after surgical decompression and tended to be stable from then on.Accordingly, they recommended that rehabilitation should start as early as possible.
No statistically significant difference was found between the studied groups regarding preoperative and postoperative VAS; however, postoperative VAS change was non-significantly lower in laminectomy only and laminectomy with fusion (more postoperative pain) than in ACDF only and ACDF with plate (less postoperative pain).Traditionally, axial pain has been reported mainly after posterior cervical surgery, but there has been research evidence that patients with anterior surgery may also experience postoperative long-term axial pain [26].In a recent study by Wang and colleagues comparing ACDF with plating against laminectomy with fusion in managing 4-levels CSM, they observed no difference in postoperative VAS scores of axial pain between both groups [27].In their systematic review conducted to examine factors predisposing to postoperative axial pain following laminectomy, Wang and colleagues recommended the following to minimize axial pain after posterior surgery: early postoperative ROM (range of motion) exercise, shorter or no application of external immobilization, less surgical exposure, avoiding detachment of semispinalis cervicis muscle from C2 spinous process and reconstructing the extensor musculature as anatomical as possible [28].
Regarding quality of life assessment, there were no differences between the compared groups regarding preoperative SF-36 scores.However, postoperative SF-36 scores and its change were highest in anterior groups (with no significant difference between them) than in posterior groups.In accordance with our results, Ghogawala and colleagues reported better SF-36 scores associated with anterior surgery [29].In contrast, Fehlings [16], Kato [24] and Hitchon [30] found no difference in the  The extensive muscle dissection associated with posterior approaches usually reflects into more postoperative discomfort and accordingly longer hospital stays.In our research, hospital stay was non-significantly longer in posterior groups.There was no difference between anterior and posterior groups regarding operative duration.However, intraoperative blood loss was highest in posterior groups.These findings match most of the published literature [17,24,[29][30][31][32][33][34][35].Owing to the longer hospital days related to posterior surgery, Ghogawala [29] among other authors [16,36] consider anterior approaches to be highly cost-effective compared to posterior approaches.
In our study, both anterior and posterior cohorts were associated with complications, both short-and long-term with no statistically significant difference between either approach which is the same conclusion reached by previous authors like Ghogawala [29], Asher [17], Kato [24] and Fehlings.[16].
Previous studies in the literature had similar complications rate and distribution to ours; regarding hoarseness, El-Ghandour [35] rate was (6.3%), which is nearly like that reported in the series by Liu [34] (8%) and Kristof [32] (7.1%), all three values were lower than ours, and Edwards [31] series rate was (15.6%).
Posterior cervical approaches with or without instrumentations are often associated with loss of lordosis [39,40].Both Roguski and Tang in a separate research proved an inverse relationship between cervical lordosis and patients quality of life values even if those patients have improved regarding myelopathy scores [41,42].On the contrary, anterior approaches have been shown to result in more effective correction of kyphosis and the maintenance of postoperative lordosis than laminoplasty and laminectomy with instrumentation [42].Our results are in agreement with the literature in this as we found no statistically significant difference between the studied groups regarding preoperative lordosis; however, postoperative lordosis and its change were highest in both anterior groups (with no significant difference between them) than in posterior groups.
Our study did not find statistically significant differences between group A1 and A2 (ACDF only and ACDF with plating) regarding functional outcomes, lordosis or complications.This might be attributed to the relatively short follow-up period.The main advantages of adding an anterior cervical plate often cited in the literature are higher fusion rates, less incidence of graft subsidence and greater lordosis angles [6,[43][44][45][46].On the other hand, higher rates of adjacent segment disease (ASD) and subsequent reoperations in addition to greater incidence of postoperative dysphagia were often quoted as drawbacks to cervical plating [47][48][49].
Other than longer operative duration associated with laminectomy with fusion, no statistically significant differences have been found in our study between groups B1 and B2 (laminectomy only and laminectomy with fusion) regarding functional outcomes, lordosis or complications.Controversy exists concerning the use of prophylactic instrumented fusion as an adjunct to laminectomy for CSM as long-term effects on functional status remain uncertain [50].Laminectomy alone may be safe and effective in highly selected patients who have a stiff cervical spine, preserved cervical lordosis, and no radiographic evidence of spinal instability.Due to higher cost of instrumentations, and its association with more operative time, blood loss and perioperative complications, instrumented fusion as an adjunct to laminectomy should be reserved in cases when laminectomy alone is contraindicated.

Conclusions
Both anterior and posterior approaches addressing CSM were associated with significant improvements in most health-related outcomes with no differences in complications severity or frequency.Anterior surgery was associated with a shorter hospital stay, less blood loss with better quality of life scores and more successful restoration of cervical lordosis than posterior surgery.Despite that no difference has been demonstrated between plated and non-plated anterior fusion, plated techniques may enhance fusion and further restore lordosis.Likewise, although no difference has been detected between laminectomy alone and with lateral mass fusion, it is advised, however, to add fusion to laminectomy whenever there is evident instability or lack of lordosis.

Fig. 4 Fig. 5
Fig. 4 Postoperative lordosis change scores between the groups treated with anterior or posterior surgical techniques.

Fig. 6 A
Fig. 6 A-D Male 55 years old from group A1, A pre-op sagittal T2 MRI; B pre-op axial T2 MRI; C pre-op plain X-ray lateral view; D post-op plain X-ray lateral view

Fig. 7 A
Fig. 7 A-F Female 49 years old from group A2, A pre-op sagittal T2 MRI; B pre-op axial T2 MRI; C pre-op plain X-ray lateral view; D post-op plain X-ray lateral view; E post-op sagittal T2 MRI; F post-op axial T2 MRI

Fig. 8 A
Fig. 8 A-F Female 69 years old from group B1, A pre-op sagittal T2 MRI; B pre-op axial T2 MRI; C pre-op plain X-ray lateral view; D post-op plain X-ray lateral view; E post-op sagittal T2 MRI; F post-op axial T2 MRI

Fig. 9 A
Fig. 9 A-D Female 53 years old from group B2, A pre-op sagittal MRI; B pre-op axial MRI; C pre-op plain X-ray lateral view; D post-op plain X-ray lateral view; E post-op CT scan axial view

Table 1
Groups demographics Data presented as n (%) unless mentioned otherwise ACDF Anterior cervical discectomy and fusion; SD Standard deviation ^ ANOVA test # Chi-square test § Fisher's exact testFigures 6, 7, 8 and 9 show pre-and postoperative images for selected cases, one case from each group.

Table 2
Baseline clinical characteristics Data presented as n (%) unless mentioned otherwise ACDF Anterior cervical discectomy and fusion; SD Standard deviation ^ ANOVA test # Chi-square test § Fisher's exact test

Table 3
Complications among studied groups

Table 4
Operative findings among studied groups ACDF anterior cervical discectomy and fusion; SD Standard deviation ^ ANOVA test.* Significant.Homogenous groups had the same symbol "a,b" based on post hoc test