TAVR services have expanded significantly in the U.S., with regional clustering of sites, but expansion may have had unintended consequences on procedural quality, according to a study published Aug. 16 in the Journal of the American College of Cardiology.

Javier A. Valle, MD, MSCS, FACC, et al., used data from the STS/ACC TVT Registry to look at temporal and geographic dispersion of TAVR in the U.S. in relation to population density. The initial site cohort consisted of all TAVR centers participating in the registry between 2011 and 2017 after TAVR was commercially available. The researchers then used a more recent analytic cohort, between 2015 and 2017, to look at relationships between case volumes, site density, and changes in volume and density with patient risk profiles and outcomes in contemporary practice.

The number of TAVR sites participating in the registry increased from 198 in November 2011 to 556 in December 2017, corresponding to an increase from five TAVR sites per million Medicare beneficiaries in 2011 to 12 per million in 2017. Of 306 hospital referral regions (HRRs) analyzed, the number without a TAVR center grew from 178 (58.2%) in 2012 to 55 (17.9%) in 2017.

The study found a statistically significant association between case volume and patient risk based on STS risk for 30-day mortality and TVT risk for in-hospital mortality. Sites with higher case volumes were associated with higher estimates of patient risk (p=0.001 for STS risk; p=0.005 for TVT risk), and higher site density was associated with lower risk (p<0.001 for STS risk; p<0.001 for TVT risk). In addition, there was an association between higher annual case volume and lower hazard of 30-day and one-year mortality, as well as the composite endpoint of death, stroke, bleeding, acute kidney injury, or moderate or severe paravalvular leak at 30 days. There was also a relationship between higher site density and higher hazard of 30-day mortality but no association between site density and one-year mortality or the composite endpoint at 30 days.

According to the researchers, the findings "reinforce emerging data for a volume-outcome relationship in TAVR" and "raise concerns over unintended consequences of imbalanced dispersion and clustering." While TAVR has expanded, the results show geographic variation and regional clustering of sites. They note that lower annual site volumes and higher site densities were associated with lower estimates of patient and procedural risk and increased mortality risk, concluding that expansion have TAVR services "may have had unintended consequences on procedural quality and patient outcomes."

In an accompanying editorial comment, David R. Holmes Jr., MD, MACC, and D. Craig Miller, MD, FACC, write that although the study's findings "confirm that the well-intended rational dispersion of TAVR has indeed been problematic and has not lived up to expectations," the STS/ACC TVT Registry has made "multiple important observations" regarding TAVR. They conclude that an upcoming quality analysis on one-year TAVR outcomes by the Centers for Medicare and Medicaid Services "may ultimately reduce the number of TAVR centers and rectify the disparity in outcomes among sites."

Clinical Topics: Cardiac Surgery, Invasive Cardiovascular Angiography and Intervention

Keywords: Transcatheter Aortic Valve Replacement, Hospital Mortality, Centers for Medicare and Medicaid Services, U.S., Medicare, Registries, National Cardiovascular Data Registries, Cardiology, Acute Kidney Injury, Stroke, STS/ACC TVT Registry

< Back to Listings