Most surgical procedures performed on account of degenerative disease of the cervical spine involve a discectomy and interbody fixation. Bone fusion at the implant placement site is evaluated post-operatively. It is agreed that computed tomography is the best modality for assessing bone union. We evaluated the results obtained with various methods based solely on conventional radiographs in the same group of patients and compared them with results obtained using a method that is a combination of CT and conventional radiography, which we considered the most precise and a reference method. We operated on a total of 170 disc spaces in a group of 104 patients. Fusion was evaluated at 12 months after surgery with five different and popular classifications based on conventional radiographs and then compared with the reference method. Statistical analyses of test accuracy produced the following classification of fusion assessment methods with regard to the degree of consistency with the reference method, in descending order: (1) bone bridging is visible on the anterior and/or posterior edge of the operated disc space on a lateral radiograph; (2) change in the value of Cobb's angle for a motion segment on flexion vs. extension radiographs (threshold for fusion vs. pseudoarthrosis is 2°); (3) change in the interspinous distance between process tips on flexion vs. extension radiographs (threshold of 2 mm); (4) change in the value of Cobb's angle of a motion segment (threshold of 4°); (5) change in the interspinous distance between process bases on flexion vs. extension radiographs (threshold of 2 mm). When bone union is evaluated on the basis on radiographs, without CT evidence, we suggest using the "bone bridging" criterion as the most reliable commonly used approach to assessing bone union.
Keywords: ACDF; cervical spine; fusion assessment; radiological measurements.