Late Subaxial Lesion after Overcorrected Occipitocervical Reconstruction in Patients with Rheumatoid Arthritis

Iwata, Akira; Abumi, Kuniyoshi; Takahata, Masahiko; Sudo, Hideki; Yamada, Katsuhisa; Endo, Tsutomu; Iwasaki, Norimasa

Abstract

STUDY DESIGN: Retrospective case-control study, level 4. PURPOSE: To clarify the risk factors for late subaxial lesion after occipitocervical (O-C) reconstruction. We examined cases requiring fusion-segment-extended (FE) reconstruction in addition to/after O-C reconstruction. OVERVIEW OF LITERATURE: Patients with rheumatoid arthritis (RA) frequently require O-C reconstruction surgery for cranio-cervical lesions. Acceptable outcomes are achieved via indirect decompression using cervical pedicle screws and occipital plate-rod systems. However, late subaxial lesions may develop occasionally following O-C reconstruction.
METHODS
O-C reconstruction using cervical pedicle screws and occipital plate-rod systems was performed between 1994 and 2007 in 113 patients with RA. Occipito-atlanto-axial (O-C2) reconstruction was performed for 89 patients, and occipito-subaxial cervical (O-under C2) reconstruction was performed for 24 patients. We reviewed the cases of patients requiring FE reconstruction (fusion extended group, FEG) and 26 consecutive patients who did not require FE reconstruction after a follow-up of >5 years (non-fusion extended group, NEG) as controls.
RESULTS
FE reconstructions were performed for nine patients at an average of 45 months (range, 24-180 months) after O-C reconstruction. Of the 89 patients, three (3%) underwent FE reconstruction in cases of O-C2 reconstruction. Of the 24 patients, five (21%) underwent FE reconstruction in cases of O-under C2 reconstruction (p=0.003, Fisher exact test). Age, sex, RA type, and neurological impairment stage were not significantly different between FEG and NEG. O-under C2 reconstruction, larger correction angle (4° per number of unfixed segment), and O-C7 angle change after O-C reconstruction were the risk factors for late subaxial lesions on radiographic assessment.
CONCLUSIONS
Overcorrection of angle at fusion segments requiring O-C7 angle change was a risk factor for late subaxial lesion in patients with RA with fragile bones and joints. Correction should be limited, considering the residual mobility of the cervical unfixed segments.