133. Mortality Following Operative and Nonoperative Treatment of Odontoid Fracture among Medicare Beneficiaries and the Influence of Dementia: a Matched Cohort Analysis

BACKGROUND CONTEXT Type II odontoid fractures are common among elderly patients due to the increased fragility of the upper cervical spine coupled with the greater incidence of low energy trauma. These fractures increase morbidity and limit physiologic reserve, leading to greater mortality. Current literature suggests lower mortality rates following surgical fixation of odontoid fractures within this population. However, it is unclear whether this reduction in mortality is confounded by the selection of healthier patients for surgery. Dementia represents age-related cognitive impairment that crosses a threshold of deterioration greater than what would be expected for a patient's functional status. Preoperative dementia has been shown to increase the complication profile following elective spine surgery. However, no previous studies have evaluated the association of dementia on outcomes following operative vs nonoperative management of Type II odontoid fracture. PURPOSE To compare mortality among operative and non-operatively treated Medicare beneficiaries with Type II odontoid fractures, and to describe the association of dementia on mortality in this cohort. STUDY DESIGN/SETTING Retrospective analysis of Medicare claims, 2016-2019. PATIENT SAMPLE A total of 22,145 beneficiaries (age 65+) with an International Classification of Diseases (ICD-10) claim with diagnosis code for Type II odontoid fracture from 2016 – 2019 were evaluated, including 3,166 (14.4%) treated surgically. Patients on Medicare HMO, under age 65, and dual eligible patients were excluded. The final cohort was 96.1% White, 58.6% female, 74% having any comorbidity, and 21.9% with dementia. OUTCOME MEASURES Mortality separately measured through 90 days and 1, 2, and 3 years. METHODS The number of days from fracture to date of death was derived by linking claims to the Master Beneficiary Summary File. Comorbidity, including dementia, was calculated with a 90-day lookback prior to Type II odontoid fracture. The Average Treatment Effect among operatively treated patients was reported in comparison to non-operatively treated patients using a 1:2 nearest-neighbor Mahalanobis matching by age, sex, race, and comorbidity. Robust logistic regressions were used to report the association of dementia on mortality, controlling for age, sex, race, and comorbidity. RESULTS Unadjusted mortality was significantly lower among operative patients compared to nonoperative patients: 10.8% vs 19.0% within 90 days, 17.9% vs 30.9% at 1 year, 25.9 vs 41.7 at 2 years, and 35.5% vs 50.3% at 3 years (p<0.001 at all time points). From the matched analysis, mortality remained significantly lower among operative patients with an average treatment effect of 5.9 percentage points lower than non-operative patients at 90 days (95%CI -7.5; -4.3), 8.5 percentage points lower at 1 year (95%CI -10.5; -6.4), 8.9 percentage points lower at 2 years (95%CI -11.7; -6.2), and 9.4 percentage points lower at 3 years (95%CI -13.6; -5.3). Dementia was strongly associated with mortality (OR 1.96; 95%CI 1.69 - 2.27; p<0.001). CONCLUSIONS While the risk of mortality is greater in patients with dementia, operatively treated Medicare beneficiaries with Type II odontoid fractures still have a lower mortality rate through four years compared to nonoperative patients in a matched analysis. FDA Device/Drug Status This abstract does not discuss or include any applicable devices or drugs. Type II odontoid fractures are common among elderly patients due to the increased fragility of the upper cervical spine coupled with the greater incidence of low energy trauma. These fractures increase morbidity and limit physiologic reserve, leading to greater mortality. Current literature suggests lower mortality rates following surgical fixation of odontoid fractures within this population. However, it is unclear whether this reduction in mortality is confounded by the selection of healthier patients for surgery. Dementia represents age-related cognitive impairment that crosses a threshold of deterioration greater than what would be expected for a patient's functional status. Preoperative dementia has been shown to increase the complication profile following elective spine surgery. However, no previous studies have evaluated the association of dementia on outcomes following operative vs nonoperative management of Type II odontoid fracture. To compare mortality among operative and non-operatively treated Medicare beneficiaries with Type II odontoid fractures, and to describe the association of dementia on mortality in this cohort. Retrospective analysis of Medicare claims, 2016-2019. A total of 22,145 beneficiaries (age 65+) with an International Classification of Diseases (ICD-10) claim with diagnosis code for Type II odontoid fracture from 2016 – 2019 were evaluated, including 3,166 (14.4%) treated surgically. Patients on Medicare HMO, under age 65, and dual eligible patients were excluded. The final cohort was 96.1% White, 58.6% female, 74% having any comorbidity, and 21.9% with dementia. Mortality separately measured through 90 days and 1, 2, and 3 years. The number of days from fracture to date of death was derived by linking claims to the Master Beneficiary Summary File. Comorbidity, including dementia, was calculated with a 90-day lookback prior to Type II odontoid fracture. The Average Treatment Effect among operatively treated patients was reported in comparison to non-operatively treated patients using a 1:2 nearest-neighbor Mahalanobis matching by age, sex, race, and comorbidity. Robust logistic regressions were used to report the association of dementia on mortality, controlling for age, sex, race, and comorbidity. Unadjusted mortality was significantly lower among operative patients compared to nonoperative patients: 10.8% vs 19.0% within 90 days, 17.9% vs 30.9% at 1 year, 25.9 vs 41.7 at 2 years, and 35.5% vs 50.3% at 3 years (p<0.001 at all time points). From the matched analysis, mortality remained significantly lower among operative patients with an average treatment effect of 5.9 percentage points lower than non-operative patients at 90 days (95%CI -7.5; -4.3), 8.5 percentage points lower at 1 year (95%CI -10.5; -6.4), 8.9 percentage points lower at 2 years (95%CI -11.7; -6.2), and 9.4 percentage points lower at 3 years (95%CI -13.6; -5.3). Dementia was strongly associated with mortality (OR 1.96; 95%CI 1.69 - 2.27; p<0.001). While the risk of mortality is greater in patients with dementia, operatively treated Medicare beneficiaries with Type II odontoid fractures still have a lower mortality rate through four years compared to nonoperative patients in a matched analysis.