Study design: Prospective multicenter cohort study.
Objective: To assess the: (1) agreement between surgeon and independent review of fusion after single-level anterior cervical decompression and fusion, and (2) influence of surgeon impression of patient status on agreement.
Summary of background data: Failure to achieve fusion can lead to poor functional outcome. Visual inspection of plain radiographs is used to assess fusion, but this assessment's reliability is not well understood.
Methods: Of 668 participants in the Cervical Spine Research Society Outcomes Study, 181 underwent single-level procedures. Three independent reviewers and each surgeon assessed fusion (i.e., radiographic trabecular bridging of the graft-vertebral body gap and absence of spinous process motion) on plain radiographs at 3 and 6 months after surgery. Agreement was evaluated with an intraclass correlation coefficient (ICC). The influence of surgeon impression of patient status on agreement was assessed with logistic regression analysis.
Results: Agreement was high among reviewers (ICC 0.822 to 0.892) but poor between reviewers and surgeons (ICC 0.308 to 0.484); disagreement was higher when the surgeon reported medical (odds ratio [OR] = 0.19, 95%; confidence interval [CI] 0.12, 0.30; P < 0.001), neurologic (OR = 0.13, 95% CI: 0.09, 0.21, P < 0.001), or functional (OR = 0.19, 95% CI: 0.12, 0.29, P < 0.001) improvement than when the surgeon did not report this improvement.
Conclusions: The finding that surgeons and independent reviewers disagreed on fusion assessment highlights the need for objective and reproducible measures of fusion.