Effect of Ventral vs Dorsal Spinal Surgery on Patient-Reported Physical Functioning in Patients With Cervical Spondylotic Myelopathy: A Randomized Clinical Trial

Abstract

Importance: Cervical spondylotic myelopathy is the most common cause of spinal cord dysfunction worldwide. It remains unknown whether a ventral or dorsal surgical approach provides the best results.

Objective: To determine whether a ventral surgical approach compared with a dorsal surgical approach for treatment of cervical spondylotic myelopathy improves patient-reported physical functioning at 1 year.

Design, setting, and participants: Randomized clinical trial of patients aged 45 to 80 years with multilevel cervical spondylotic myelopathy enrolled at 15 large North American hospitals from April 1, 2014, to March 30, 2018; final follow-up was April 15, 2020.

Interventions: Patients were randomized to undergo ventral surgery (n = 63) or dorsal surgery (n = 100). Ventral surgery involved anterior cervical disk removal and instrumented fusion. Dorsal surgery involved laminectomy with instrumented fusion or open-door laminoplasty. Type of dorsal surgery (fusion or laminoplasty) was at surgeon's discretion.

Main outcomes and measures: The primary outcome was 1-year change in the Short Form 36 physical component summary (SF-36 PCS) score (range, 0 [worst] to 100 [best]; minimum clinically important difference = 5). Secondary outcomes included 1-year change in modified Japanese Orthopaedic Association scale score, complications, work status, sagittal vertical axis, health resource utilization, and 1- and 2-year changes in the Neck Disability Index and the EuroQol 5 Dimensions score.

Results: Among 163 patients who were randomized (mean age, 62 years; 80 [49%] women), 155 (95%) completed the trial at 1 year (80% at 2 years). All patients had surgery, but 5 patients did not receive their allocated surgery (ventral: n = 1; dorsal: n = 4). One-year SF-36 PCS mean improvement was not significantly different between ventral surgery (5.9 points) and dorsal surgery (6.2 points) (estimated mean difference, 0.3; 95% CI, -2.6 to 3.1; P = .86). Of 7 prespecified secondary outcomes, 6 showed no significant difference. Rates of complications in the ventral and dorsal surgery groups, respectively, were 48% vs 24% (difference, 24%; 95% CI, 8.7%-38.5%; P = .002) and included dysphagia (41% vs 0%), new neurological deficit (2% vs 9%), reoperations (6% vs 4%), and readmissions within 30 days (0% vs 7%).

Conclusions and relevance: Among patients with cervical spondylotic myelopathy undergoing cervical spinal surgery, a ventral surgical approach did not significantly improve patient-reported physical functioning at 1 year compared with outcomes after a dorsal surgical approach.

Trial registration: ClinicalTrials.gov Identifier: NCT02076113.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Kanter reported receipt of personal fees and royalties from Zimmer Biomet and NuVasive. Dr Mummaneni reported receipt of personal fees from DePuy Synthes, Globus, and Stryker; holding stock in Spinicity/ISD; receipt of grants from the Neurosurgery Research and Education Foundation, the International Spine Study Group, AO Spine, and the National Institutes of Health; and receipt of book royalties from Thieme and Springer. Dr Bisson reported receipt of personal fees from Stryker and MiRus and grants from the Neurosurgery Research and Education Foundation. Dr Schwartz reported receipt of personal fees from Allergan, Bayer, Blue Cross and Blue Shield Associations, and Pfizer. Dr Harrop reported receipt of personal consulting fees from DePuy Synthes and Ethicon. Dr Albert reported consultancy for NuVasive; board of directorship for Back Story, the American Orthopaedic Association (past relationship), and the Scoliosis Research Society; ownership interest in Augmedics, Bonovo Orthopedics, CytoDyn, HS2, Innovative Surgical Designs, InVivo Therapeutics, Morphogenesis, Paradigm Spine, Physician Recommended Nutriceuticals, Parvizi Surgical Innovations, Precision Orthopedics, Pulse Equity, Spinicity, Strathspey Crown, and Surg.io; and receipt of royalties from DePuy Synthes Spine, Zimmer Biomet, Elsevier, JP Medical Publishers, Springer, and Thieme. Dr Riew reported receipt of personal fees from Biomet and Medtronic and stock ownership in Amedica, Axiomed, Benvenue, Expanding Orthopedics, Osprey, Paradigm Spine, Spinal Kinetics, Spineology, and Vertiflex. Dr Steinmetz reported receipt of royalties from Zimmer Biomet and Elsevier and consultancy and receipt of an honorarium from Globus. Dr Wang reported consultancy for Zimmer Biomet, Medtronic, and Titan Biologics. Dr Whitmore reported receipt of personal fees from Intrinsic Therapeutics and DePuy Synthes and stock ownership in Theseus. Dr Heller reported receipt of personal fees and royalties from Medtronic and holding a patent for laminoplasty plates issued by Medtronic. No other disclosures were reported.

Figures

Figure 1.. Spinal Expert Review

Clinical vignettes for each patient accompanied by relevant imaging studies were presented to an expert review panel. Imaging typically included sagittal T2-weighted magnetic resonance (MRI) image of the cervical spine, flexion-extension cervical radiographs, and relevant axial MRI images at the point of compression of the spinal cord. Ventral decompression and fusion would typically involve the removal of disks that are compressing the spinal cord from the front of the neck (blue arrow shown on sagittal T2-weighted MRI image). Dorsal decompression would involve decompression of the spinal column from the back of the neck (red arrow). A summary of expert opinion showing clinical equipoise was generated for surgeons and patients to review before accepting randomization (eFigure 1 in Supplement 3).

Figure 2.. Flow of Participants in the Cervical Spondylotic Myelopathy Surgical Trial

The majority of the 189 patients screened who did not enroll were not eligible. Some were unwilling to complete study questionnaires or to consider randomization. A few ultimately did not choose to have surgery or had their surgery at another institution. aOne patient randomized to ventral surgery underwent dorsal fusion surgery. bFour patients randomized to dorsal surgery underwent ventral fusion surgery.

Figure 3.. Comparative Outcomes Assessment

MCID indicates minimum clinically important difference. A, Change in Short Form 36 physical component summary (SF-36 PCS) score for each patient in the trial from baseline to 1 year. Each bar extends from a patient’s baseline score to their 1-year score, with patients in each group ordered by baseline score. At baseline, mean SF-36 PCS scores for the dorsal and ventral groups were 37.3 (SD, 9.9) and 37.4 (SD, 8.8) points, respectively. The 1-year mean change from baseline in the SF-36 PCS score was 6.2 (SD, 10.2) points for dorsal surgery and 5.9 (SD, 8.2) points for ventral surgery. B, Trajectory of change in SF-36 PCS score by randomized group, with box plots showing distribution of change in SF-36 PCS scores at each time point. Box plots represent the distribution of SF-36 PCS scores: the center line of the box is the median, with the box tops and bottoms indicating the interquartile range (IQR). The upper whisker is the largest value that is less than or equal to the third quartile plus 1.5 times the IQR, and the lower whisker is the smallest value that is greater than or equal to the first quartile minus 1.5 times the IQR. Circles represent more extreme values.