39 - 20-year Clinical Outcomes of Cervical Disc Arthroplasty: A Prospective, Randomized, Controlled Trial

Introduction. Anterior cervical discectomy and fusion (ACDF) has long been the standard of treatment for degenerative cervical disc disease. Complications including loss of index level motion and adjacent level disease prompted development of alternative treatments, such as cervical disc arthroplasty (CDA). Previous comparison of the two treatments has proven noninferiority of CDA and suggested its superiority with long term follow up.

Objectives. This study is a prospective analysis of 20-year outcomes of patients with degenerative cervical disc disease treated with CDA compared to ACDF. Patients with single-level cervical degenerative disc disease and at least 6 weeks of failed conservative treatment for either cervical radiculopathy or myelopathy were eligible for inclusion. Quantitative patient outcomes include visual analog scales (VAS) for neck and arm pain and neck disability index (NDI). Reoperation rates at 20 years were also analyzed.

Methods. 47 eligible patients were randomized 1:1 to either ACDF or CDA as part of an FDA IDE trial. One patient was deceased, and 46 patients were eligible for analysis at 20 years. Postoperative data for 240-month follow up was collected and statistically analyzed using SPSS.

Results. At 20 years, VAS neck pain and VAS arm pain scores were available for 21/25 patients in the ACDF group (84%) and 17/21 patients in the CDA group (81%). Comparison of VAS arm scores at 20 years showed a significant difference between ACDF and CDA (2.48 vs 0.67, p=0.036). VAS neck scores were not significantly different (2.73 vs 0.98, p=0.068, power=0.448).

NDI scores were available for 18/25 (72%) ACDF patients and 16/21 (76.2%) CDA patients. The difference in NDI scores at 20 years was not significant (19.00 vs 9.25, p=0.053, power=0.497).

Both treatments demonstrated significant improvement compared to preoperative functional scores at 20 years in terms of NDI, VAS arm, and VAS neck scores (p < 0.001). Significant increases in VAS scores were seen when comparing 10-year to 20-year values in the ACDF cohort (VAS arm 0.42 vs 2.7 p=0.004, VAS neck 1.4 vs 3.0 p=0.01). Differences in ACDF NDI scores were not significantly different. No significant difference in pain scores was found for CDA patients between 10- and 20-year follow up.

Patients with reoperations at any timepoint were included in reoperation data even if they did not complete 20-year follow up. Data was available for 24/25 (96.00%) ACDF and 18/21 (85.71%) CDA patients. Reoperations occurred in 10/24 (41.67%) ACDF patients and 2/18 (11.11%) CDA patients (p=0.03). One ACDF patient had an index level posterior spinal fusion. No index level reoperations were performed in the course of this 20-year study for CDA patients. Of the two CDA reoperations, one was at an adjacent level to the index site. 7/10 (70%) ACDF reoperations occurred at an adjacent level.

Conclusions. CDA continues to be an effective means of treating degenerative cervical disc disease at 20-year follow up. The arthroplasty device proved to withstand 20 years of motion in all patients, as shown by the lack of index site reoperations, while offering continued radiculopathy relief.