Efficacy of anterior odontoid screw fixation in the elderly patient: a CT-based biometrical analysis of odontoid fractures

Abstract

In the elderly population, reported union rates with anterior odontoid screw fixation (AOSF) for odontoid fracture (OF) treatment vary between 23 and 93% when using plain radiographs. However, recent research revealed poor interobserver reliability for fusion assessment using plain radiographs compared to CT scans. Therefore, union rates in patients aged ≥60 years treated with AOSF have to be revisited using CT scans and factors for non-union to be analysed. Prospectively gathered consecutively treated patients using AOSF for odontoid fracture with age ≥60 years were reviewed. Medical charts were assessed for demographics, clinical outcomes and complications. Patients' preoperative radiographs and CT scans were analysed to characterize fracture morphology and type, fracture displacement, presence of atlanto-dental osteoarthritis as well as a detailed morphometric assessment of fracture surfaces (in mm(2)). CT scans performed after a minimum of 3 months postoperatively were analysed for fracture union. Those patients not showing CT-based evidence of completely fused odontoid fracture were invited for radiographic follow-up at a minimum of 6 months follow-up. Follow-up CT-scan were studied for odontoid union as well as the number of screws used and the square surface of screws used for AOSF and the related corticocancellous osseous healing surface of the odontoid fragment (in %) were calculated. Patients were stratified whether they achieved osseous union or fibrous non-union. Patients with a non-union were subjected to flexion-extension lateral radiographs and the non-union defined as stable if no motion was detected. The sample included 13 male (72%) and 5 female (18%) patients. The interval from injury to AOSF was 4.1 ± 5.3 days (0-16 days). Age at injury was 78.1 ± 7.6 years (60-87 years) and follow-up was 75.7 ± 50.8 months (4.2-150.2 months). 10 patients had dislocated fractures, 14 had Type II and 4 "shallow" Type III fractures according to the Anderson classification, 2 had stable C1-ring fractures, 8 had displayed atlanto-dental osteoarthritis. Fracture square surface was 127.1 ± 50.9 mm(2) (56.3-215.9 mm(2)) and osseous healing surface was 84.0 ± 6.8% (67.6-91.1%). CT-based analysis revealed osseous union in 9 (50%) and non-union in 9 patients (50%). Union rates correlated with increased fracture surface (P = 0.02). Statistical analysis revealed a trend that the usage of two screws with AOSF correlates with increased fusion rates (P = 0.06). Stability at C1-2 was achieved in 89% of patients. CT scans are accepted as the standard of reference to assess osseous union. The current study offers an objective insight into the union rates of odontoid fractures treated with AOSF using CT scans in consecutive series of 18 patients ≥60 years. Literature serves evidence that elderly patients with unstable OF benefit from early surgical stabilization. However, although using AOSF for unstable OF yields segmental stability at C1-2 in a high number of patients as echoed in the current study, our analysis stressed that using follow-up CT scans in comparison to biplanar radiographs dramatically reduces osseous union rates compared to those previously reported for AOSF.