Interobserver agreement using computed tomography to assess radiographic fusion criteria with a unique titanium interbody device.

The accuracy of using computed tomography (CT) to assess interbody fusion in patients with titanium implants has been questioned in the past. Radiologists have reported difficulty assessing fusion bone quality because of metal artifact and small graft windows. A new titanium interbody implant with a large footprint and a wide graft aperture has been developed. We conducted a study to determine the interobserver reliability of using CT to assess radiographic fusion variables with the new titanium interbody device. Patients underwent anterior lumbar interbody fusion with the same titanium interbody implant. Reconstructed CT images were obtained randomly at 6, 9, or 12 months. Two independent radiologists reviewed the scans. Interobserver reliability was calculated using the κ statistic. Fifty-six spinal fusion levels (33 patients) were analyzed. The radiologists agreed on 345 of the 392 fusion data points reviewed (κ = .88). Agreement for solid fusion formation was 0.77. This interbody device demonstrated minimal artifact and minimal subsidence, and trabecular bone was easily identified throughout the implant in the vast majority of cases reviewed. High interobserver agreement was noted across all radiographic variables assessed.