Trends in Vascular Complications in High-Risk Patients Following Transcatheter Aortic Valve Replacement in the United States

Vascular complications (VC) following transcatheter aortic valve replacement (TAVR) are associated with worse outcomes. The trend of VC incidence in patients considered high risk is unclear. We sought to assess the trend of VC after TAVR in patients at high risk. We investigated the VC trend in female, diabetes mellitus, and peripheral vascular disease (PVD) patients. Patients who underwent TAVR from 2011 to 2014 in the United States were identified using the International Classification of Diseases-Ninth Revision code 35.05 from the Nationwide Inpatient Sample database. Frequency of any VC (per 100 transcatheter aortic valve implantation procedure or hospital discharges) for each year from 2011 to 2014 was assessed for the overall population as well as within each category of high-risk cohorts. The overall VC rate was 6.0% (2,044/33,790). Patients who had VC were more likely to be female and had higher rates of PVD at baseline. The annual rate of VC in the overall population from 2011 to 2014 was 4.6%, 9.4%, 6.8%, and 4.4%, respectively. There was a significant increase in VC rate from 2011 to 2012 (p = 0.03), whereas there was a significant decrease in VC rate from 2012 to 2014 (p <0.001). The rate of VC between 2011 and 2014 was similar (p = 0.82). The rate of VC did not increase in any of the high-risk groups from 2011 to 2012. However, the rate of VC from 2012 to 2014 decreased significantly in all the high-risk groups. The VC rate was similar for groups between 2011 and 2014. The overall VC rate among TAVR patients initially increased from 2011 to 2012 but decreased thereafter. Similar trend in VC rate was found among high-risk patients except that the initial increase in rates from 2011 to 2012 did not reach statistical significance. Whether further reduction in VC with improvement in devices and operator/center experience for both overall and high-risk groups in TAVR occurs will require continuous longitudinal monitoring.