A Meta-Analysis of Comparative Outcomes Following Cervical Arthroplasty or Anterior Cervical Fusion: Results from Four Prospective Multi-center Randomized Clinical Trials and up to 1226 Patients.

By London Spine

A Meta-Analysis of Comparative Outcomes Following Cervical Arthroplasty or Anterior Cervical Fusion: Results from Four Prospective Multi-center Randomized Clinical Trials and up to 1226 Patients.

Spine (Phila Pa 1976). 2011 Oct 27;

Authors: McAfee PC, Reah C, Gilder K, Eisermann L, Cunningham B

Abstract
ABSTRACT: Study Design: Meta-analysis of four prospective randomized controlled FDA IDE clinical trials.Objective. To maximize the information available from four IDE studies by analyzing the combined outcomes of cervical arthroplasty versus fusion at 24-month follow-up.Summary of Background Data. To date, four RCTs have been completed in the US under FDA IDE protocols to study cervical arthroplasty. Each trial reported arthroplasty to be at least as successful as fusion controls, based on non-inferiority trial designs. However, sample sizes in any given trial may not be sufficient to demonstrate superiority of treatment effect. Meta-analysis enables pooling of results from comparable trials which may lead to more precise and statistically significant estimates of treatment effect.Methods. Four cervical arthroplasty RCTs with comparable enrollment criteria and outcome measures were conducted independently by three separate sponsors to study the following devices: Bryan, Prestige, ProDisc-CÔ, and PCM cervical disc replacements. A total of 1,608 patients were treated across 98 investigative sites. Data was available for 1,352 treated patients, of which 1,226 were evaluable at 24 months. Assessments included clinical success definitions based on NDI, maintenance or improvement of neurological status, subsequent surgery or intervention at the index level (survivorship), and a composite score comprised of these as well as serious device-related adverse events. Trial endpoint comparisons were made at 24 months postoperatively. For each endpoint a random-effects meta-analysis was performed to compare the success rates of cervical arthroplasty versus ACDF. Also, supportive frequentist and Bayesian analyses were performed.Results. The pooled primary overall success results indicated a statistically significant treatment effect favoring arthroplasty over ACDF. Overall success was achieved by 77.6% of the arthroplasty patients and by 70.8% of the ACDF patients (pooled odds ratio [OR] 0.699, 95% confidence interval [CI]: 0.539-0.908, p = 0.007). The results of the individual subcomponent meta-analyses, all of which favored arthroplasty, were NDI success (OR 0.786, 95% CI: 0.589-1.050, p = 0.103), neurological status (OR 0.552, 95% CI: 0.364-0.835, p = 0.005), and survivorship (OR 0.510, 95% CI: 0.275-0.946, p = 0.033). Only the survivorship endpoint suggested low heterogeneity.Conclusion. These findings suggest that cervical arthroplasty is superior to ACDF in overall success, neurological success, and survivorship outcomes 24 months postoperatively.

PMID: 22037535 [PubMed – as supplied by publisher]