Study design

Retrospective cohort study of prospectively accrued data.

Objective

To evaluate a large, prospective, multicentre dataset of surgically treated degenerative cervical myelopathy (DCM) cases on the contemporary risk of C5 palsy with surgical approach.

Summary of background data

The influence of surgical technique on postoperative C5 palsy after decompression for DCM is intensely debated. Comprehensive, covariate-adjusted analyses are needed using contemporary data.

Methods

Patients with moderate to severe DCM were prospectively enrolled in the multicenter, randomized, Phase III CSM-Protect clinical trial and underwent either anterior or posterior decompression between Jan 31, 2012 and May 16, 2017. The primary outcome was the incidence of postoperative C5 palsy, defined as the onset of muscle weakness by at least one grade in manual muscle test at the C5 myotome with slight or absent sensory disruption after cervical surgery. Two comparative cohorts were made based on the anterior or posterior surgical approach. Multivariate hierarchical mixed-effects logistic regression was used to estimate odds ratios (OR) with 95% confidence intervals (CI) for C5 palsy.

Results

A total of 283 patients were included, and 53.4% underwent posterior decompression. The total incidence of postoperative C5 palsy was 7.4% and was significantly higher in patients who underwent posterior decompression compared with anterior decompression (11.26% vs. 3.03%, P =0.008). After multivariable regression, the posterior approach was independently associated with greater than four times the likelihood of postoperative C5 palsy ( P =0.017). Rates of C5 palsy recovery were comparable between the two surgical approaches.

Conclusion

The odds of postoperative C5 palsy are significantly higher after posterior decompression compared to anterior decompression for DCM. This may influence surgical decision-making when there is equipoise in deciding between anterior and posterior treatment options for DCM.

Level of evidence

Therapeutic Level-II.