Risk Model For 30-Day Mortality After TAVR Allows Benchmarking, Quality Assessment

A clinical risk model for 30-day death after TAVR may allow for benchmarking, quality assessment and comparison of short-term mortality across centers as TAVR expands to lower-risk patients and less experienced sites, according to research published March 19 in JACC: Cardiovascular Interventions.

Suzanne V. Arnold, MD, MHA, et al., collected data from 21,661 patients in the STS/ACC TVT Registry who underwent TAVR at 188 sites between June 2013 and May 2016. The primary outcome was all-cause mortality 30 days after TAVR, while the secondary outcome was in-hospital all-cause mortality. The researchers developed a model to estimate risk for 30-day mortality after TAVR based on pre-procedural factors and access site. The model included factors from the TVT Registry in-hospital mortality model and added patient-reported health status and gait speed.

Results showed that of the 21,661 TAVR patients, 1,025 (4.7 percent) died within 30 days. Independent predictors of 30-day mortality included older age, low body weight, poor renal function, peripheral artery disease, home oxygen use, prior myocardial infarction, left main coronary artery disease, tricuspid regurgitation, nonfemoral access, worse baseline health status and inability to walk. The predicted 30-day mortality risk ranged from 1.1 percent at the lowest decile of risk to 13.8 percent at the highest decile of risk. There were no significant differences in teaching status, bed size, or annual TAVR procedural volume between included and excluded sites. In addition, 30-day mortality rate death was similar between included and excluded sites (4.7 percent vs. 5.1 percent). There were few meaningful differences associated with in-hospital vs. 30-day mortality. Lower patient-reported health status and inability to complete a 5-m walking test were associated with an increased risk of in-hospital mortality.

In an accompanying editorial, Stuart J. Head, MD, PhD, and Nicolas M. Van Mieghem, MD, PhD, FACC, write that "the use of risk prediction for benchmarking TAVR outcomes has been somewhat neglected." They add that although the new study is "an important contribution," the development of a TAVR risk prediction model "remains a work under construction."

Keywords: Hospital Mortality, Benchmarking, Tricuspid Valve Insufficiency, Transcatheter Aortic Valve Replacement, Coronary Artery Disease, Registries, STS/ACC TVT Registry, National Cardiovascular Data Registries

< Back to Listings