Objective: To evaluate the influence of carpentry of Bryan disc, including artificial disc coverage ratio and shell angular, on long-term efficacy and complications cervical artificial disc replacement. Methods: To retrospectively analyze the clinical data of 86 patients who underwent Bryan disc replacement in Beijing Jishuitan Hospital between 2003 and 2007 (including 101 operated segments). All the patients were followed-up for more than 10 years. Mean age of the patients was (53±9) years, including 51 males and 35 females, and 71 patients underwent single-level and 15 patients underwent two-level artificial disc replacement. Artificial disc coverage ratio and shell angular were measured on postoperative natural X-ray. According to the perfectness of carpentry including 2 parameters (postoperative shell kyphosis and artificial disc coverage ratio smaller than 95%), operated levels were divided into 2 groups: 43 levels in the optimal group and 58 levels in the suboptimal group. Radiographic indexes were evaluated preoperatively and at the last follow-up, included the segmental range of motion(ROM), segmental COBB's angle, the occurrence of segmental mobility lost, segmental kyphosis and the high grade paravertebral ossification(PO). Clinical indexes including Japanese Orthopaedic Association (JOA) score, neck disability index (NDI) and Odom's criteria were evaluated preoperatively and at the last follow-up. The improvement of JOA score, NDI% decline and the satisfaction rate of Odom's criteria were calculated too. The continuous data were analyzed by independent sample t test. The grading data were analyzed by Mann-Whitney test. Results: The mean age of the optimal group was (52±7) years, including 23 males and 14 females; the mean age of the suboptimal group was (54±10) years, including 28 males and 21 females.The baseline data of two groups were comparable. There were significant differences in the segmental ROM, the segmental COBB's angle, the occurrence of segmental mobility lost, segmental kyphosis and the high grade PO between the two groups at the last follow-up (t=5.608, 3.812, χ(2)=15.097, 18.334, 27.141, all P<0.05). The improvement rate of JOA score was 69%±39% and 66%±44%, the NDI% decline was 14%±8% and 11%±8%, the satisfaction rate of Odom's criteria was 97.3% and 83.7% in optimal group and suboptimal group, respectively; there was no significant difference in the up-mentioned indexes between the two groups(t=0.307, 1.483, Z=0.989, all P>0.05). Conclusion: The carpentry of prosthesis will influence the occurrence of segmental mobility lost, segmental kyphosis, and high grade PO with cervical artificial disc replacement in the long run. It is suggested that the artificial disc coverage smaller than 95% and the shell kyphosis should be avoided.