ACUTE AXIS FRACTURES - ANALYSIS OF MANAGEMENT AND OUTCOME IN 340 CONSECUTIVE CASES

Study Design. Retrospective review of acute axis fractures treated at a tertiary referral center. Objective. To determine the optimal treatm ent of axis fractures based on 340 cases from a single institution. Su mmary of Background Data. Axis fractures account for almost 20% of acu te cervical spine fractures. However, their management and the clinica l criteria predictive of nonoperative failure remain unclear. Methods. Admission imaging studies and clinical variables were obtained for 34 0 consecutive axis fracture patients. Fractures were classified as odo ntoid Type I, II, or III with dens displacement on admission roentgeno grams; hangman's fractures of Francis grade and Effendi type; and misc ellaneous fractures. Treatment methods were documented, and outcomes w ere based on dynamic lateral roentgenograms, clinical examination, or telephone interviews at last follow-up. Results. Follow-up data were a vailable in 92% of cases. Type II odontoid fractures comprised 35% of all axis fractures, were the most difficult to treat, and had the high est nonunion rate (28.4%). Odontoid displacement of 6 mm or more was a ssociated with Type II nonunion (chi-square = 33.74, P < 0.0001). Pati ents underwent surgical fusion if fracture alignment could not be main tained by an external orthosis, or if they had odontoid fractures with transverse ligament disruption, Type II odontoid fractures with dens displacement of at least 6 mm, or hangman's fractures of severe Franci s grade or Effendi type. Conclusions. Type II odontoid fractures have the highest nonunion rate and were associated with dens displacement o f 6 mm or greater. Early surgical fusion is recommended for acute frac ture instability despite external immobilization, transverse ligament disruption, Type II odontoid fractures with dens displacement of at le ast 6 mm on admission, or severe Francis grade or Effendi-type hangman 's fractures. Otherwise, nonoperative management is sufficient.