Asymptomatic ACDF Nonunions Underestimate the True Prevalence of Radiographic Pseudarthrosis

Study Design. Secondary analysis of subjects in the control anterior cervical discectomy and fusion (ACDF) arm of Medtronic Investigational Device Exemption (IDE) trials for cervical disc arthroplasty (CDA). Objectives. To compare patient-reported outcomes (PROs) in patients with radiographic nonunion to patients with a solid fusion. Summary of Background Data. The true prevalence of nonunion after ACDF is unknown. Nonunion may be under-reported, as some patients are not symptomatic enough to justify radiographic evaluation. Methods. Three hundred forty-five subjects enrolled in the control arm of IDE trials for CDA who had single-level ACDF with allograft and plate with 24 month data formed the study cohort. Using the 24-month postop evaluation, subjects were divided into those who had radiographic fusion and those who did not using strict study criteria. SF-36, NDI, Neck and Arm pain NRS scores were reported at 12, 24, 36, 60, and 84 months postop. For subjects who had secondary surgery failures, the last observations before the secondary surgery were carried forward to future visits for the analysis. Results. Forty-four (13%) patients had radiographic nonunion and 301 (87%) were fused at 24 months postop. At 24 months, PROs were similar between the two groups. Seven patients in the Nonunion group (16%) and 10 (3%) in the Fused group had additional surgery at the index level prior to the 24 follow-up (P = 0.003). Over the 84-month follow-up a total of 9 patients in the Nonunion group (21%) and 22 (7%) in the Fused group had additional surgery at the index level (P = 0.009). Conclusion. While the radiographic nonunion rate at 24 months was 13%, PROs show that many of the radiographic nonunions were asymptomatic. Although the majority of patients with radiographic nonunion did not undergo additional surgery, the rate of secondary surgeries at the index level was significantly higher in the radiographic nonunion group.