Morbidity and Mortality After Vertebral Fractures: Comparison of Vertebral Augmentation and Nonoperative Management in the Medicare Population.

STUDY

DESIGN

Vertebral compression fracture (VCF) patients in the 100% US Medicare data set (2005-2009).

OBJECTIVE

To compare the mortality and morbidity risks for VCF patients undergoing conservative treatment (nonoperated), balloon kyphoplasty (BKP), and vertebroplasty (VP).

SUMMARY OF BACKGROUND DATA: Studies have reported lower mortality risk for BKP or VP cohorts than nonoperated cohorts, but it is uncertain whether there are any differences in morbidity risks.

METHODS

Survival and morbidity was estimated by the Kaplan-Meier method, and the differences in outcomes were assessed by Cox regression between BKP, VP, and nonoperated cohorts. A propensity matching analysis was used to account for potential bias.

RESULTS

A total of 1,038,956 VCF patients were identified, including 141,343 BKP patients and 75,364 VP patients. The nonoperated cohort was found to have a 55% higher adjusted risk of mortality (P < 0.001) than the BKP cohort and 25% higher adjusted risk of mortality (P < 0.001) than the VP cohort. The BKP cohort was also found to have a 19% lower adjusted risk of mortality (P < 0.001) than the VP cohort. The findings were similar for mortality with pneumonia diagnosed in the 90 days before death and also after propensity matching, as well as for subgroups of osteoporotic VCF patients, including those who survived at least 1 year and those with no cancer diagnosis. With propensity matching, the nonoperated cohort had significantly higher adjusted risks of pneumonia, myocardial infarction/cardiac complications, DVT, and urinary tract infection than the BKP cohort but lower adjusted risks of subsequent augmentation/fusion, subsequent augmentation, and pulmonary/respiratory complications. The BKP cohort also had significantly lower risks of morbidity than the VP cohort, except for deep venous thrombosis (DVT), infection, and myocardial infarction/cardiac complications, which were similar between both cohorts.

CONCLUSION

VCF patients in the Medicare population who received vertebral augmentation therapies, specifically BKP and VP, experienced lower mortality and overall morbidity than VCF patients who received conservative management.

LEVEL OF EVIDENCE: 3.