Radiographic Predictors of Subaxial Subluxation After Atlantoaxial Fusion.

Abstract

A retrospective study. The aim of this study was to clarify preoperative radiographic predictors associated with the development of subaxial subluxation (SAS) after surgery. The incidence of atlantoaxial fusion for atlantoaxial instability has been increasing. SAS can develop after surgery despite atlantoaxial fusion with the optimal C1-C2 angle. We hypothesized that preoperative discordant angular contribution in the upper and subaxial cervical spine is associated with the occurrence of postoperative SAS. Patients who underwent surgery for atlantoaxial instability with a minimum 5-year follow-up and control participants were included. The O-C2 angle, C2 slope (C2S), C2-C7 cervical lordosis (CL), and T1 slope (T1S) were measured. We focused on the angular contribution ratio in the upper cervical spine to the whole CL, and the preoperative C2/T1S ratio was defined as the ratio of C2S to T1S. Twenty-seven patients (SAS=11, no-SAS=16; mean age, 60.7y old; 77.8% female; mean follow-up duration, 6.8y) and 23 demographically matched control participants were enrolled. The SAS onset was at 4.7 postoperative years. Preoperatively, the O-C2 angle, C2-C7 CL, and T1S were comparable between the SAS, no-SAS, and control groups. The preoperative C2S and C2/T1S ratio were smaller in the SAS group than in the no-SAS or control group (C2S, 11.0 vs. 18.4 vs. 18.7 degrees; C2/T1S ratio, 0.49 vs. 0.77 vs. 0.78, P <0.05). The receiver operating characteristic curve analysis demonstrated that the C2/T1S ratio had higher specificity and similar sensitivity as a predictor of postoperative SAS than C2S (specificity: 0.90 vs. 0.87; sensitivity: 0.73 vs. 0.73). The estimated cutoff values of the C2S and C2/T1S ratio were 14 degrees and 0.58, respectively. The preoperative C2/T1S ratio was closely associated with postoperative SAS. Patients with a C2/T1S ratio <0.58 were at a high risk of SAS after atlantoaxial fusion. Level 4.