Comparison of revision surgeries for one- to two-level cervical TDR and ACDF from 2002 to 2011

Background context: Cervical total disc replacement (TDR) and anterior cervical discectomy and fusion (ACDF) provide comparable outcomes for degenerative cervical pathology. However, revisions of these procedures are not well characterized.

Purpose: The purpose of this study is to examine the rates, epidemiology, perioperative complications, and costs between the revision procedures and to compare these outcomes with those of primary cases.

Study design: This study is a retrospective database analysis.

Patient sample: A total of 3,792 revision and 183,430 primary cases from the Nationwide Inpatient Sample (NIS) database from 2002 to 2011 were included.

Outcome measures: Incidence of revision cases, patient demographics, length of stay (LOS), in-hospital costs, mortality, and perioperative complications.

Methods: Patients who underwent revision for either one- to two-level cervical TDR or ACDF were identified. SPSS v.20 was used for statistical analysis with χ(2) test for categorical data and independent sample t test for continuous data. The relative risk for perioperative complications with revisions was calculated in comparison with primary cases using a 95% confidence interval. An alpha level of less than 0.05 denoted statistical significance.

Results: There were 3,536 revision one- to two-level ACDFs and 256 revision cervical TDRs recorded in the NIS database from 2002 to 2011. The revision cervical TDR cohort demonstrated a significantly greater LOS (3.18 vs. 2.25, p<.001), cost ($16,998 vs. $15,222, p=.03), and incidence of perioperative wound infections (13.6 vs. 5.3 per 1,000, p<.001) compared with the ACDF revision cohort (p<.001). There were no differences in mortality between the revision surgical cohorts. Compared with primary cases, both revision cohorts demonstrated a significantly greater LOS and cost. Furthermore, patients who underwent revision demonstrated a greater incidence and risk for perioperative wound infections, hematomas, dysphagia, and neurologic complications relative to the primary procedures.

Conclusions: This study demonstrated a significantly greater incidence of perioperative wound infection, LOS, and costs associated with a TDR revision compared with a revision ACDF. We propose that these differences are by virtue of the inherently more invasive nature of revising TDRs. In addition, compared with primary cases, revision procedures are associated with greater costs, LOS, and complications including wound infections, dysphagia, hematomas, and neurologic events. These additional risks must be considered before opting for a revision procedure.