Incidence of Multiple Revision Cervical Surgeries After Single-Level Anterior Cervical Discectomy and Fusion.

Abstract

Retrospective analysis using the MarketScan private insurance database from 2010-2020. Determine the incidence of multiple revision cervical surgeries at 5 years following primary, single-level ACDF and assess the risk of subsequent revisions after the first and second surgeries. The rate of revision surgery after ACDF is well-documented, but data on multiple revision surgeries is limited. Adult patients ≤65 years undergoing primary, single-level ACDF were identified. Patients with infectious, traumatic, or neoplastic etiologies were excluded. The primary endpoint was any revision cervical surgery with follow-up ending at 5 years. Kaplan-Meier and Cox proportional hazards regression were used, adjusting for sex, age, CCI, and region. A total of 42,845 patients undergoing primary, single-level ACDF (P) were included, with a mean age of 48.9 years (SD 9.0); 52.8% were female. The "first revision" (R1) group included 2,374 patients, and the "second revision" (R2) group had 195 patients. The mean revision-free follow-up was significantly different across the P, R1, and R2 groups, though with small absolute differences (P 2.2 years, R1 2.0 years, R2 2.0 years; P<0.0001). At 5-years post-operatively, the incidence of revision surgery was 10.8% after primary surgery, 24.1% after one revision, and 42.5% after two revisions. In multivariable Cox regression, the risk of subsequent revision surgery was significantly higher after one revision (HR 1.6 vs. primary, P<0.0001) and even more so after two revisions (HR 2.6 vs. primary, P<0.0001). Interval hazard analysis showed a significantly higher incidence of revision from 2-5 years with each subsequent revision (all P<0.05). After primary ACDF in patients <65 years, approximately 10% underwent revision at 5 years post-operatively. The occurrence of subsequent revision surgery was higher; >20% after one revision, and >40% after two revisions, which is critical for patient decision-making. III.