Twenty one patients, who had undergone three-level ACDF with a cage and plate construct for degenerative cervical spinal disorder from November 2001 to April 2007 and were followed-up for at least two years, were enrolled in this study (Table 1). There were 10 men and 11 women with a mean age of 71.7 years (range, 65 to 91 years). The mean follow-up period was 34.1 months (range, 25 to 70 months). Fourteen patients had radiculopathy and 7 patients had symptoms indicating accompanying myelopathy. In terms of the fusion level, there were nine C 3/4, 4/5, 5/6 cases, eight C 4/5, 5/6, 6/7 cases, three C 3/4, 5/6, 6/7 cases, and one C 5/6, 6/7, C7/T1 case. Magnetic resonance imaging (Magnetom Vision, Siemens, Erlagen, Germany) was performed on all patients and the computed tomography myelogram (Somatom Sensation 16, Siemens) was conducted selectively to accurately define the lesion. The bone mineral density was measured in all patients; their average T-score was -2.5 SD.

ACDF was performed at all levels to which the clinical signs and symptoms of the disease were referable, and at which neural element compression was demonstrated by a neuroradiographic study. Cancellous bone for cage insertion was collected using a trocar (diameter 7 mm; AO Synthes, Davos, Switzerland) in the left anterior superior iliac spine. The surgical procedure was performed using a standard anterior Smith-Robinson approach.17) For cage fusion, PEEK cages (Solis, Stryker Spine, South Allendale, NJ, USA) were used and fixed with a rigid metal plate (Peg Plate System, BK Meditech, Seoul, Korea; Cervical Spine Locking Plate, AO Synthes). The patients were advised to wear a Philadelphia cervical collar for 2-4 weeks after surgery, and a soft cervical collar for an additional 2 weeks. Plain lateral and flexion/extension lateral radiographs were used to evaluate bone union at 6 weeks, and 3, 6, 9, and 12 months postoperatively. A CT evaluation was performed if union was uncertain on the radiograph. All radiology evaluations were performed on a digital image using built-in software (Marosis 5.0 PACS viewer, Marotech, Seoul, Korea). Union was defined as less than 2° movement by lateral flexion/extension radiography, bridging bone between endplates, no signs of implant failure of the anterior plate system and < 50% radiolucency of the implant outer surface.18) Cage subsidence was defined as the difference in distance between the center points of the inferior borders of proximal vertebrae and the superior borders of the distal vertebrae at the fusion level, as determined by lateral radiographs postoperatively and at last follow-up. The incidence of cage subsidence and the frequencies at the respective fusion levels were compared. The overall cervical lordosis was defined as the angle between the extended line in the 2nd posterior cervical vertebra and that in the 7th posterior cervical vertebra. For lordosis of fusion segments, the Cobb angle where the extension line of its upper/lower margin meet was selected. Degenerative changes in the adjacent segments were deemed to be present when the degenerative cervical vertebrae were unstable, the height of a disc in the adjacent segment decreased or kyphosis had occurred based on the comparisons of the postoperative and last follow-up radiographs.19) A decrease in the disc height of an adjacent segment was deemed to have occurred when the disc height was < 25% in grade I, 25% to 50% in grade II, 50% to 75% in grade III and > 75% in grade IV.20) Cases with more than grade II were deemed to show degenerative changes in the adjacent segments. The clinical outcomes were analyzed using the SF-36 physical composite score (PCS) and neck disability index (NDI) (Table 2). In addition, the complications associated with screws, plates and cages, and neurological complications and postoperative infections were also documented.

All 21 patients showed bone union after an average of 12.3 weeks (range, 10 to 14 weeks) (Fig. 1). The mean cage subsidence was 2.8 mm (range, 1 to 4.8 mm) in 5 patients (23.8%), which was observed in the anterior and posterior regions at 2 weeks postoperatively. The subsidence observed at 3 months after surgery was similar to that observed at the final follow-up visits (Fig. 2). Subsidence at two-levels was noted in 3 patients and at one level in 2, but this did not cause any clinical problems. Cervical lordosis was 5° before surgery, 17.6° immediately after surgery and 16.5° at the final follow-up visits. Adjacent segment degeneration was detected in 3 patients (14.3%), which included 2 patients with disc space narrowing more than grade II, and 1 patient with instability. However, no clinical problems were encountered.

The SF-36 PCS before surgery, second postoperative week and at the last follow-up was 29.5, 43.1, and 66.2, respectively. The respective NDI was 55.3, 24.6 and 15.9 (Table 2).

Loosening of a plate and screw occurred in 3 patients (14.3%) (Fig. 3), in which a locking screw had not been inserted properly during surgery in 1 and the screws protruded due to cage subsidence in mainly the lower segments during the follow-up in 2. However, these 3 patients had no clinical problems, such as dysphagia or dysphonia at the last physical examination, and no neurological complications or postoperative infection were encountered.