Revision Rate and Opioid Use After Single-Level Versus Multi-Level Cervical Disc Arthroplasty: A Large Observational Study Using a National Database.

STUDY

DESIGN

Level 3 retrospective national database study.

OBJECTIVE

This study aims to assess the comparability of multilevel cervical disc arthroplasty (CDA) with single-level CDA in clinical outcomes.

SUMMARY OF BACKGROUND DATA: Despite multilevel CDA having been studied and utilized for some time, a well-powered, large investigation with a longitudinal follow-up period has not been adequately reported.

METHODS

A retrospective review of 15,705 patients who underwent either single- or multilevel CDA between 2017 and 2020 was performed using the PearlDiver Mariner Patient Claims Database (PearlDiver Technologies). Our primary outcomes were revision procedures, with a median follow-up period of 3 and a half years, and pain level at 90 days and 1 year, using opioid intake as a surrogate measure. Univariate and multivariate logistic regression models were used to compute the odds ratios (ORs) for revisions and opioid use. To ensure the follow-up period was not a mediating factor, we conducted a log-rank survival analysis using the Kaplan-Meier method.

RESULTS

Of the 15,705 patients, 10,456 (66.6%) had single-level CDA, while 5249 (33.4%) received multilevel CDA. The odds ratio and incidence rate of revision surgery were statistically similar across cohorts, with incidence rates of 5.27% and 4.95% for single- and multilevel observations, respectively. After adjustment for age, sex, obesity, diabetes, smoking history, and preprocedure opioid use, there was still no difference among the groups [OR=0.937 (0.803-1.09), P=0.405]. Similar results were observed in opioid usage at 1-year [OR=1.06 (0.988-1.14), P=0.104]. However, opioid cessation at 90 days significantly differs among the cohort, with patients undergoing multilevel CDA being less likely to stop using opioids at 90 days postprocedure [OR=0.903 (0.843-0.967), P<0.01].

CONCLUSIONS

Among patients who received single-level and multilevel CDA, we found that there was no difference in revision rate and long-term opioid usage, highlighting the comparability of multilevel CDA with single-level CDA and its ability to achieve similar success.