TAVR Boosts Health Status in Most Patients, but Many Still Suffer Postprocedure

In everyday practice, patients who undergo TAVR generally have gains in health status that are at least as good as those seen in the pivotal trials, according to an analysis of the TVT Registry. More than one-third of patients, however, continue to have poor outcomes after the procedure, suggesting that expectations should be tempered.

Among more than 7,000 patients with available data at 1 year, 62.3% had a favorable outcome, defined as a reasonable quality of life (QoL) without a meaningful decline from baseline, lead author Suzanne Arnold, MD (Saint Luke’s Mid America Heart Institute, Kansas City, MO), and colleagues report in a study published online February 1, 2017, ahead of print in JAMA Cardiology.

However, certain groups of patients—those with severe lung disease, a requirement for dialysis, or very poor baseline health status, for example—had only about a 50% chance of a favorable outcome at 1 year.

“The procedure generally does well for the majority of people, but there still are a number of patients who don’t meaningfully benefit from TAVR and [we need to] set realistic expectations prior to TAVR so that patients don’t get frustrated if they don’t have quality-of-life improvement,” Arnold told TCTMD. Nonetheless, she said the fact that most patients do well “is very reassuring.”

Greater Gains Than in Clinical Trials

The researchers examined health status data on patients undergoing TAVR between November 2011 and March 2016 who were included in the Society of Thoracic Surgeons and the American College of Cardiology TVT Registry. The analysis included information from 31,636 patients with Kansas City Cardiomyopathy Questionnaire (KCCQ) scores available at 30 days and 7,014 with scores at 1 year. The average summary score at baseline was 42.3, indicating substantial impairment.

There were large mean improvements in KCCQ scores of 27.6 points at 30 days and 31.9 points at 1 year. Patients who had information available at baseline, 30 days, and 1 year showed significant gains both between baseline and 30 days and between 30 days and 1 year.

By 1 year, most patients had a favorable outcome, defined as a KCCQ score of 60 or higher and a lack of a decline of 10 or more points from baseline. Poor outcomes were most commonly due to death (19.4%), persistently impaired quality of life (17.4%), and a decline in quality of life (4.9%).

The authors note that the mean changes in KCCQ score observed in the current study were about 5 to 10 points greater than those seen in analyses of the PARTNER trial and the CoreValve US Pivotal Trial. This likely has to do with the fact that real-world patients have lower risk than those treated in the trials, as well as refinements in patient selection, procedural techniques, and postprocedural care, say the researchers.

Missing Data an Issue

In an invited commentary, Karen Alexander, MD (Duke Clinical Research Institute, Durham, NC), points out that in very sick patients with severe aortic stenosis, symptomatic improvement is the primary goal of treatment.

The greater gains in quality of life seen in this analysis compared with the pivotal trials might suggest better patient selection informed by prior experience, she notes, adding, however, that KCCQ scores at 1 year were nonexistent for the majority of patients in the registry.

“This large amount of missing data should prompt renewed efforts to gather complete representative symptom and QoL assessments to continue understanding these important outcomes,” Alexander writes.

In an editor’s note, Robert Bonow, MD (Northwestern Medicine, Chicago, IL), and Patrick O’Gara, MD (Brigham and Women’s Hospital, Boston, MA), say the findings of the analysis are “comforting,” but they also highlight the holes in the data.

They note that patients without available QoL information at 1 year were more severely disabled and had lower baseline KCCQ scores. “And of course, QoL data are not captured in those who died before 1 year,” they write. “This suggests that the current TVT data may overestimate the true symptomatic benefit of TAVR in this elderly population.”

Bonow and O’Gara believe the data are insightful, “but we urge the TVT Registry to intensify its efforts for more comprehensive QoL outcome data collection so that the cardiology community has a fuller comprehension of the benefits of this expensive and resource-intensive technology,” they write.

In her commentary, Alexander states that it should be possible to achieve higher rates of symptomatic improvement with TAVR.

“The ability to anticipate symptom and QoL outcomes from TAVR with more informed selection is central to achieving goals of care,” she says. “This process will necessitate commitment to routine assessment of patient-reported outcomes as a starting point, understanding that competing risks are in play for symptoms as well as survival, and embracing palliation when treatment is unlikely to achieve goals of care.”