Rates and Outcomes of Vertebral Augmentation for the Treatment of Osteoporotic Vertebral Fractures Among the Commercially Insured

Abstract

Introduction: Approximately 750,000 people per year in the U.S. sustain osteoporotic vertebral fractures (OVF). Vertebroplasty and kyphoplasty, collectively termed vertebral augmentation are commonly used to treat OVF yet the strength of evidence supporting the use of vertebroplasty is weak. Kyphoplasty lacks a robust evidence base to support or reject its use. Prior studies of vertebral augmentation have focused primarily on patients >65 years old. Our aims were twofold; to determine the temporal changes in vertebral augmentation rates over the past decade in a commercially insured, working-age (under 65) American population, and to compare the rates of major medical complications, resource utilization and medication use among a cohort of patients with OVF treated with vertebral augmentation as compared to propensity matched patients not treated with vertebral augmentation .Materials and Methods: Analysis of patients with OVF in the IBM MarketScan® Commercial Claims and Encounters Databases of Americans with employer-provided health insurance for the 2008-2017 period. To determine changes in OVF and vertebral augmentation rates over time we used ICD-CM 9/10 codes among a total of 149 million individual patients with an age range of 18-65 while excluding those with alternative fracture etiologies such as cancer, infection or transport accidents. We also 1:1 matched augmented patients to controls using propensity scores based on age, gender, region, fracture year, comorbidities, hospital admission at index fracture, and prior pain medication use, with further exact matching on use of opioids and advanced imaging in the time period until augmentation. We assessed rates of major medical complications, opioid use and all cause gross covered payments after augmentation and the analogous number of days from OVF in the control group (baseline day). Conditional logistic regression models were used to calculate odds ratios (ORs) and 95% confidence intervals (95% CI) for categorical variables, with median regression used for the continuous variables. Results: We identified 19,944 patients with OVF with an average incidence of 5.44 OVF per 100,000 MarketScan enrollees per year from 2008 - 2017. A total of 3424 patients (17.2%) with OVF underwent vertebral augmentation. Specifically, 708 (3.5%) were treated with vertebroplasty, 2571 (12.9%) received kyphoplasty and 145 (0.7%) had both. There was a trend of decreased utilisation of vertebroplasty, with a concomitant increase in kyphoplasty procedures performed over the time period. The percentage of patients with OVF treated with vertebral augmentation remained relatively stable. In our outcome analysis, which required 1 year post fracture enrollment we identified 14,995 patients with OVF of whom 2363 (16%) were augmented and 2304 (98%) were matched 1:1 to non-augmented controls, giving a total matched analysis subset of 4608 patients. We combined vertebroplasty and kyphoplasty subjects into a single augmented group for our primary analysis. Approximately 75% of the matched subset were female with a median age of 58 in each group. We did not observe meaningful differences in the odds of major medical complications within 30 days (adjusted OR= 0.89; 95%CI= 0.55, 1.43) comparing augmented to non-augmented patients, or all cause median costs between groups from 3 days post baseline to 1 year post OVF (adjusted difference between medians = $1285; 95%CI= -312, 2703). However, augmented patients were more likely to have filled opioid medications in the 7 to 30 days after the procedure with 33.9% filling a prescription compared to 28.4% among non-augmented subjects (adjusted OR= 1.40; 95%CI= 1.19, 1.64). We noted significant differences in the rates of major spine surgeries, with the augmented group much less likely to receive surgery than the non-augmented control group in the year after OVF with annual rates of 4.86% vs 8.55% (adjusted OR= 0.43; 95%CI= 0.31, 0.60). Conclusions: In our analysis of a large administrative claims database cohort of those under the age of 65 from 2008-2017 we found that 17% of patients with OVF were observed to have undergone vertebral augmentation. The ratio of those treated with kyphoplasty increased overtime and represented 87% of all augmentations performed in 2017. Contrary to previously published findings (in older adults), this analysis found that patients less than 65 years old who underwent vertebral augmentation did not have decreased major medical complications, opioid fills, or all cause payments in the year following their OVF. We found that all cause costs (with cost of augmentation excluded) to be similar in the year following OVF. Augmented patients were significantly less likely to undergo major spine surgery in the year following OVF, suggesting augmentation may diminish the ultimate need for surgery in this younger patient cohort. Clinical Relevance Statement: Vertebral augmentation used to treat osteoporotic fractures in those <65 was not meaningfully associated with decreased major medical complications or opioid use compared to controls.