To evaluate the effect of anterior cervical fusion or non-fusion surgery for the treatment of Hirayama disease.

The patients with Hirayama disease who underwent anterior cervical surgery in our hospital from July 2008 to December 2015 were retrospectively enrolled in this study. These cases were divided into two groups: fusion group (11 cases) and non-fusion group (12 cases). In fusion group, there were 9 males and 2 females, with an average age of 17.92±0.61 years (range, 15-29). In the non-fusion group, there were 10 males and 1 female, with an average age of 19.18±1.14 years (range, 15-23). Muscle strength, muscular atrophy, and the Michigan score for hand function were compared between the two groups preoperatively and postoperatively. The indexes such as arc cervical physiological curvature index, the maximum flexion angle, the maximal range of motion (ROM), flexion ROM of adjacent segments of lower cervical vertebra were measured from the X-ray of the two groups, and the cross-sectional area of spinal cord was measured and compared on neutral position MRI.

There were no significant differences between the two groups in grip strength, preoperative hand function Michigan score, cervical physiological curvature index, the maximum flexion angle, the maximal ROM, flexion ROM of adjacent segments of lower cervical vertebra, and cross-sectional area of spinal cord on neutral position MRI (P>0.05). Patients in fusion group were followed up for 38-103 months with an average of 59.73±5.57 months, while patients in non-fusion group were followed up for 40-95 months with an average of 56.25±4.57 months. Clinical outcomes: At the last follow-up, the finger tremor and cold paralysis were alleviated in all patients, but the symptoms in fusion group and non-fusion group were not completely disappeared in 2 and 3 cases, respectively. The grip strength of fusion group (preoperative 17.14±6.09, postoperative 17.47±5.64) and the non-fusion group (preoperative 17.75±5.49, postoperative 17.40±4.58) were not significantly improved compared with preoperative (P>0.05). The Michigan score for hand function of fusion group (preoperative 10.27±0.41, postoperative 6.64±0.24) and the non-fusion group (preoperative 9.25± 0.35, postoperative 6.83±0.24) were significantly improved compared with preoperative (P<0.05). There was no significant difference in postoperative Michigan score between the two groups (P>0.05). Radiological outcomes: The cervical physiological curvature index and cross-sectional area of the spinal cord of the two groups increased after operation, while the maximum flexion angle, the maximal ROM decreased. There was no significant difference in these indexes between the two groups postoperatively (P>0.05). The spinal cord had no compression on flexion MRI. The cross-sectional area of C₄-C₇ spinal cord in Neutral MRI were significantly increased at C₆ and C₇ levels in both groups. There was no significant difference in the cross-sectional area of C₄-C₇ spinal cord between the two groups postoperatively (P>0.05). Two cases of internal fixation loosening occurred in non-fusion group 3 years after the operation, while there were no cases of loosening in the fusion group.

Anterior cervical fusion or non-fusion surgery can limit excessive cervical kyphosis in patients with Hirayama disease, reduce atrophy of compressed cervical spinal cord, and alleviate the disease progress of muscular atrophy of the distal upper limb. It can be selected as a treatment method for Hirayama disease. However, there is a risk of internal fixation loosening in non-fusion cases, and also a risk of adjacent segment degeneration in fusion cases. The appropriate cases should be selected carefully for surgery.

Cervical vertebrae; Spinal cord compression; Upper extremity; Muscular atrophy; Spinal fusion