Effect of industry funding on outcome reporting in cervical disc arthroplasty: a systematic review and meta-analysis of randomized controlled trials.

Abstract

Background context: Cervical Disc Arthroplasty (CDA) has been shown to be an effective and safe alternative to Anterior Cervical Discectomy and Fusion (ACDF), with randomized controlled trials (RCTs) reporting non-inferior or even favorable outcomes to ACDF. However, the current literature of large RCTs reporting long-term outcomes of CDA primarily comprises of the industry sponsored Food and Drug Administration (FDA) Investigational Device Exemption (IDE) trials. As a result, CDA has yet to be universally accepted by surgeons due to concerns of bias in the current literature.

Purpose: To compare the outcomes of single-level CDA and ACDF by conducting a meta-analysis of RCTs, with a subgroup comparison of IDE and non-IDE trial results.

Study design: Systematic review and meta-analysis.

Patient sample: 19 studies (9 IDE, 10 non-IDE) reporting the outcomes of 18 RCTs were included with a total of 3054 patients (1691 CDA and 1363 ACDF). Among CDA patients, 1229 (72.7%) were enrolled in an FDA IDE trial, while 462 (27.3%) were involved in RCTs that were not funded by industry. Minimum follow-up among the RCTs included ranged from 2 to 10 years.

Outcome measures: Outcomes of interest included index and adjacent segment reoperation rates, postoperative disability as reported by the Neck Disability Index (NDI), and the Visual Analog Scale (VAS) for Neck and Arm pain.

Methods: A random effects meta-analysis was performed comparing CDA and ACDF by pooling the outcomes of all RCTs for each outcome of interest. A subgroup analysis was then performed comparing the pooled outcomes of the FDA IDE trials and non-IDE RCTs. Standardized mean differences (SMD) and log relative risk (RR) were used to analyze continuous and categorical variables with corresponding 95% confidence intervals (CI).

Results: Among all RCTs, there was a significantly lower risk for all secondary surgical interventions with CDA relative to ACDF (RR: 0.91, 95% CI: 0.55-1.28; p < 0.0001) in addition to lower risk for adjacent segment surgery (RR: 1.06, 95% CI: 0.66-1.45; p < 0.0001), and index segment reoperation (RR: 0.48, 95% CI: 0.005-0.96; p=.048). No significant differences in NDI, VAS Neck, or VAS Arm were found in the analyses comparing ACDF and CDA (p>.05). When comparing between the IDE and non-IDE trial subgroups, there were no significant differences noted in any assessed outcome (p>.05).

Conclusion: Cervical disc arthroplasty appears to be equivalent to ACDF in reducing postoperative pain and disability, while also potentially decreasing the risk for subsequent surgical intervention, as demonstrated by the FDA IDE trials and non-IDE RCTs without industry ties. While a large number of high-quality trials for CDA do pose a risk for bias due to industry sponsorship, the current literature of high-quality RCTs without industry affiliations corroborates similar findings.