TAVR in the US: Regional Variations in Numbers, Outcomes

The number of transcatheter aortic valve replacement cases performed in the US increased over a recent 3-year period, but the risk of in-hospital mortality continues to vary across different geographic regions, according to the results of a new study.

Compared with TAVR procedures performed in the Northeast US, there was a statistically significant 26% increased risk of in-hospital mortality for cases performed in the Midwest and a 61% increased risk of in-hospital death for patients treated in the South.

The difference in mortality between patients treated in the Northeast and Midwest was evident only in 2012. The higher risk of in-hospital mortality in the South persisted over the 2012-2014 study period but narrowed over time. For example, TAVR patients treated in the South had a twofold higher risk of in-hospital mortality compared with the Northeast region in 2012, while the risk of in-hospital mortality was just 40% higher in 2014.

There was no difference in in-hospital mortality risk between patients treated in the Northeast and those in the West at any time point, although average hospital costs were higher in the West.

To TCTMD, senior investigator Deepak Bhatt, MD (Brigham and Women’s Hospital, Boston, MA), pointed out that data were only available up to 2014 and overall TAVR outcomes have improved significantly since then. “Any initial regional differences, if real, were likely due to the rate of diffusion of the technology and associated operator and institutional experience with TAVR,” he commented via email. “I suspect such regional differences would not be apparent in 2017, though it is certainly worth reexamining.”

Lead investigator Tanush Gupta, MD (Albert Einstein College of Medicine, New York, NY), told TCTMD the findings, including the different in-hospital mortality rates, utilization rates, and hospital lengths of stay across the country, should direct professional societies and hospitals with structural heart programs “to explore the underlying source of the variation, and once we’re able to identify what’s causing it, to implement expert decision pathways for streamlining patient care.”

The study was published August 10, 2017, in the American Journal of Cardiology.

TAVR Penetration Varies by Region

Based on data from the National Inpatient Sample, which was developed by the Agency of Healthcare Research and Quality, there were 41,025 TAVR procedures performed in the US between 2012 and 2014, with 10,390 done in the Northeast, 9,090 in the Midwest, 14,095 in the South, and 7,450 in the West.

Overall, the number of TAVR procedures per one million adults increased from 24.8 in 2012 to 63.2 in 2014, with an overall in-hospital rate of 4.2%. TAVR procedures increased in all four US regions, with the number of implants per one million adults highest in the Northeast, and then followed by the Midwest, South, and West. In 2014, there 90.8 TAVRs per one million adults in the Northeast region compared with 66.6, 55.7, and 51.1 TAVRs per one million adults in the Midwest, South, and West, respectively.

The regional disparity in TAVR utilization per one million adults may be related to differences in the number of TAVR-eligible patients in the four regions, specifically the number of elderly patients, although Gupta and colleagues don’t believe this to be the reason.

“The next question we asked was whether there was a differential distribution of elderly patients across the regions,” said Gupta. “When we looked at the number of TAVRs utilized for patients 75 years and older, we found similar results. It was higher in the Northeast in all 3 years.”  

Between 2012 and 2014, there was an increase in the number of hospitals performing TAVR in all four regions and an increase in the number of TAVR hospitals per one millions adults. By 2014, there were 1.38 TAVR-capable hospitals per one million adults in the Northeast, which was similar in the Midwest. In contrast there were 1.03 and 0.91 hospitals per one million adults performing TAVR in the southern and western US.  

A Public Health Issue

Christopher Meduri, MD (Piedmont Heart Institute, Atlanta, GA), who was not involved in the study, told TCTMD that the disparity in TAVR penetration is sobering and concerning. “It’s been shown in numerous studies that cardiac procedures are less likely to be utilized in people from lower socioeconomic strata, and I really think this is what we’re seeing,” he told TCTMD. “My hope is that societies and industry approach this as a public health issue.”

My hope is that societies and industry approach this as a public health issue. Christopher Meduri

In the US, it is estimated that 35% to 40% of patients eligible for TAVR or surgical aortic valve replacement for aortic stenosis remain untreated, said Meduri. Directing resources to educate healthcare providers in these underserved geographic regions, including impoverished regions in the southern US, would go a long way toward ensuring that all people have access to therapy for valvular heart disease.

“As interventionalists, we often don’t think that way,” said Meduri. “We get the patient as they come to us, and we treat them. Maybe we have more of a responsibility than we recognize to look at this as part of the bigger picture.”

Gupta noted that previous studies have shown cardiac procedures, such as PCI and ICD implantation, are performed less frequently in poorer regions. He added that many of the large TAVR trials were conducted by hospitals and physicians practicing in the Northeast and this “legacy effect” might explain why utilization is higher than other regions.

In addition to the difference in in-hospital mortality across the four regions, the analysis showed that average length of stay for TAVR was shorter in the Midwest, South, and West compared with the Northeast, while patients treated in the Northeast were more likely to be discharged to a skilled nursing facility or receive home health care.

Reasons for the Disparity?

In an attempt to explain the higher risk-adjusted mortality in regions outside the Northeast, the researchers point out that the number of TAVR procedures in the Midwest was lower but the number of TAVR-capable hospitals was nearly identical to the Northeast. This amounts to a lower average hospital TAVR volume in the Midwest compared with the Northeast. Similarly, the Southern region also had a lower average hospital TAVR volume compared with the Northeast.

“One of the possible reasons is that this could be related to the procedural volume of the Northeast compared with the South,” said Gupta, noting that higher procedural volume has been linked with more favorable in-hospital TAVR outcomes. He pointed out, however, they did not specifically analyze results based on annual hospital TAVR volumes. 

While procedural volume might partially explain the difference in mortality between the Northeast and other regions, Gupta and colleagues acknowledge that unmeasured confounding variables might have also influenced their analysis. They point out, though, that while the risk-adjusted mortality was highest in the South, these patients were younger than those treated in Northeast and had a similar prevalence of comorbidities.

Meduri told TCTMD that although patients in the South were younger, treating a younger patient for severe symptomatic aortic stenosis likely means they have major health issues, some of which might not be captured in a multivariable risk adjustment. “Even just from anecdotal experience, a 70-year-old coming in to see me in Atlanta is not the same as a 70-year-old I was seeing in Boston,” said Meduri. “Yet if you saw their notes, you wouldn’t see anything.”