Risk factors for development of subaxial subluxations following atlantoaxial arthrodesis for atlantoaxial subluxations in rheumatoid arthritis

Study Design: Retrospective radiographic/imaging study. Objective: To evaluate preoperative and sequential postoperative radiographs following C1-C2 arthrodesis for atlantoaxial subluxation in patients with rheumatoid arthritis (RA) to determine risk factors for the development of subaxial subluxations (SAS). Summary of Background Data: The development of SAS has often been observed after C1-C2 arthrodesis. However, there have been no previous reports on the correlation between radiographic parameters and the incidence of postoperative SAS. Methods: The study group comprised of 58 patients with RA who underwent C1-C2 arthrodesis due to atlantoaxial subluxation. There were 5 men and 53 women with a mean age of 55.8 years. The mean follow-up period was 137 months. Nineteen patients with a postoperative SAS after C1-C2 arthrodesis were classified as the SAS+ group. Other 39 pat ents without a postoperative SAS were included in the SAS- group. Clinical outcomes and plain radiographs were reviewed retrospectively and compared between the 2 groups. Results: The difference between pre- and postoperative atlantoaxial (AA) angles in the SAS+ group was significantly greater than those in the SAS- group (P = 0.039). The C2-C7 angles changed significantly between pre- and postoperative periods in the SAS+ group (P = 0.039), but not in the SAS- group (P = 0.897). It was suggested that a large AA angle and a small C2-C7 angle observed at the early postoperative period were the risk factors for the development of SAS. We also demonstrated that a high incidence of the C3-C4 SAS resulted from excessive bone fusion at the C2-C3. Conclusion: Excessive correction of AA angle is likely to cause loss of cervical lordosis, resulting in the development of postoperative SAS. In addition, extensive bony union at C2-C3 following C1-C2 arthrodesis frequently leads to the development of extensive SAS at the C3-C4.