Functional Outcome of Elderly Patients Treated for Odontoid Fracture: A Multicenter Study

STUDY DESIGN: Retrospective multicenter study. OBJECTIVE: Analysis of impact of conservative and surgical treatments on functional outcome of geriatric odontoid fractures. SUMMARY OF BACKGROUND DATA: Treatment of odontoid fractures in aged population is still debatable. METHODS: One hundred fourty-seven consecutive odontoid fractures in elderly patients were classified according to Anderson-D'Alonzo and Roy-Camille classifications. Philadelphia type collar was always positioned and kept as a treatment whenever acceptable. Halo-vest, anterior screw fixation, C1-C2 posterior arthrodesis, and occipito-cervical fixation were the other treatments adopted. Conservative or surgical treatment strategy was more significantly influenced by antero-posterior displacement (< or >5 mm) and by surgeon decision. On admission ASA, modified Rankin scale (mRS-pre) and Charlson Comorbidity Index (CCI) were assessed. Modified Rankin scale (mRS-post), Neck Disability Index (NDI), and Smiley Webster Pain Scale (SWPS) were administered 12 to 15 months after treatment to estimate functional outcome in terms of general disability, neck-related disability, and ability to return to work/former activity. Risk of treatment crossover was calculated considering factors affecting outcome. Fracture healing process in terms of fusion-stability, no fusion-stability, no fusion-no stability was evaluated at 12 months through a cervical computed tomography (CT) scan. Dynamic cervical spine x-rays were obtained whether necessary. No fusion-stability was considered an adequate treatment goal in our geriatric population. Chi square/Fisher exact test and logistic regression were performed for statistical anal. RESULTS: Overall 67 patients were treated conservatively whereas 80 underwent surgery. Collar was adopted in 45 patients, while anterior odontoid fixation and C1-C2 posterior arthrodesis were preferred for 30 patients each. 79.8% of patients showed good outcomes according to NDI. No significant differences were observed between patients of 65 to 79 years and more than or equal to 80 years (P = 0.81). CCI greatly correlated with mRS-post, with higher indexes in 68.8% of cases characterized by good outcomes (P = 0.05). mRS-pre correlated with NDI (P < 0.000001) and mRS-post (P = 0.04). CCI, mRS-pre, and surgery were associated with worse NDI, while both C1-C2 posterior arthrodesis and occipito-cervical stabilization were associated with worse mRS-post, respectively in 40% and 30% of cases. Younger patients had a higher risk of treatment crossover. CONCLUSION: mRS-pre and CCI provided two independent predictive values respectively for functional outcome and post-treatment disability. Compared with conservative immobilizations, surgery revealed no advantages in the elderly in terms of functional outcome.