Worse TAVI and M-TEER Outcomes Seen With Low-Volume Operators

Even in the contemporary era, there is an association between operator volume and early clinical outcomes for patients undergoing transcatheter aortic and mitral valve procedures, data from the Society of Thoracic Surgeons/American College of Cardiology (STS/ACC) TVT Registry show.

The risks of in-hospital complications and mortality at 30 days were significantly higher in patients with severe aortic stenosis treated by low-volume TAVI operators, for example. For patients requiring transcatheter edge-to-edge repair for mitral regurgitation (M-TEER), there was no difference in 30-day mortality between those treated by low- and high-volume operators, but the risk of in-hospital complications was higher in those done by less-experienced interventionalists.  

“In general, it’s true for athletes, for musicians, for everything, that the more you do, the better you get,” Dharam Kumbhani, MD (UT Southwestern Medical Center, Dallas, TX), told TCTMD. “However, it was unclear if the differences were still relevant in the current era. If you look at PCI, it’s very streamlined, and yes, there is a need for experience and training, but at the same time, you’re not going to see a steep volume-outcome relationship. That’s kind of the why here: has the field [of TAVI and M-TEER] matured to a level where we’re not going to see those relationships anymore?”

The answer is no, but Kumbhani cautioned that the analysis only shows an association, not causation. However, there were notable differences in processes of care between low- and high-volume operators. For example, low-volume operators were less likely to use minimal or moderate sedation during TAVI, had longer procedure times, and used more contrast. Additionally, patients treated by low-volume operators had higher postprocedure mean aortic gradients and paravalvular regurgitation than those treated by high-volume doctors. Processes of care with M-TEER were also worse with low-volume interventionalists.

“To me, it speaks to this not just being a chance finding, but more that there is a plausible explanation why,” Kumbhani said. “There’s obviously a lot that we cannot measure, but the things that we can measure—the procedure takes longer, you have more leaks, you have higher gradients—suggests that there’s something systematic about the way these procedures are done by lower-volume operators.”

Currently, the national coverage determinations (NCDs) for TAVI and M-TEER from the US Centers for Medicare & Medicaid Services (CMS) mandate minimum volume requirements for hospitals but not individual operators. Professional societies also focus on hospital volume as a surrogate for quality of care.

In general, it’s true for athletes, for musicians, for everything, that the more you do, the better you get. Dharam Kumbhani

David Cohen, MD (St. Francis Hospital, Roslyn, NY, and Cardiovascular Research Foundation, New York, NY), who wasn’t involved in the study, said there are very few procedures in cardiology for which the volume-outcome relationship doesn’t hold up.

“It’s the old adage that practice makes perfect, but the challenge is we have tension, which [the authors] acknowledge, between quality and access,” he told TCTMD. “The United States is a big place and there are concerns that if you make strict volume requirements, many patients will have less access than they do now to these technologies which are very valuable. Any guidelines or policy that is made has to straddle that line very carefully.”

Cohen said that while there are relationships between operator volume and outcomes with TAVI and M-TEER, the absolute differences in risk are small. “I think many people would be willing to tolerate those kinds of differences for better access,” he said.

Ashish Pershad, MD (Dignity Health, Chandler, AZ), who wasn’t involved in the study, also emphasized the fine line between access and physician competency. However, the expansion of TAVI programs, and M-TEER behind it, to rural America hasn’t happened even though there is no mandatory annual volume requirement for operators.   

“Most of the concentration of these providers has all been in urban hospitals, so you have a clear dilution of experience, with groups having multiple operators,” Pershad told TCTMD. “People have gotten trained through fellowship pathways that have clearly surpassed the need. This has resulted in a flooding of the market with structural operators that have predominantly populated a lot of urban America.”  

Hospital vs Operator Volumes

While the TAVI and M-TEER NCDs specify a minimum hospital volume for coverage, contemporary studies have shown a limited correlation with outcomes. In a recent analysis of more than 400,000 patients undergoing TAVI or M-TEER, researchers saw no association between hospital volumes and in-hospital mortality with either procedure. There were small differences in 30-day mortality between low- and high-volume centers performing TAVI, but none were observed with M-TEER.  

“The hospital-level association between volume and outcomes, at least after 30 days, has started to plateau a little bit,” said Kumbhani. “As technology matures and the experience has been shared and disseminated, those outcomes have settled a little bit, but what about operators? There’s been a tremendous explosion in the number of people who want to do these procedures.”

The new analysis, which was published in JAMA Cardiology, spans a period that includes the fourth-generation MitraClip (Abbott), which became available for M-TEER in 2019, and the commercial availability of the Pascal device (Edwards Lifesciences). It also includes patients undergoing M-TEER for primary and secondary mitral regurgitation (MR).

In all, 358,943 patients (median age 79.0 years; 42.4% female) underwent TAVI performed by 7,524 operators at 827 hospitals. With M-TEER, 51,407 patients (median age 79.0 years; 45.5% female) were treated by 2,483 operators at 493 hospitals. The median annual operator volumes for TAVI and M-TEER were 24 and 12 cases, respectively. In all, 7.8% of TAVI operators did five or fewer cases each year and 21.8% did no more than 10. For M-TEER, those figures were 16.6% and 42.1%, respectively.

I think many people would be willing to tolerate those kinds of differences for better access. David Cohen

With TAVI, operators were classified as having low (< 15 cases), medium (15-37 cases), or high volume (> 37 cases). In the fully adjusted model, the 30-day mortality rate was higher among low-volume operators (2.4% vs 2.0% with high-volume operators; OR 1.13; 95% CI 1.02-1.26), as was the risk of in-hospital complications (7.1% vs 6.4%; OR 1.09; 95% CI 1.03-1.16). A composite endpoint that included 30-day death, stroke, major/life-threatening bleeding, stage 3 acute kidney injury, or paravalvular leak was also significantly higher when TAVI was performed by low-volume operators (3.7% vs 3.2%; OR 1.10; 95% CI 1.01-1.19).

With M-TEER, annual volumes were also stratified into low (< 8 cases), medium (8-16 cases), and high tertiles (> 16 cases). In the adjusted model, in-hospital complications were significantly higher in cases done by low- versus high-volume operators (4.0% vs 3.2%; OR 1.31; 95% CI 1.11-1.56). There was no difference in 30-day mortality risk, but the 30-day composite endpoint that included death, stroke, heart failure hospitalization, or residual MR of 2 or higher was worse with low-volume operators (39.7% vs 35.3%; OR 1.12; 95% CI 1.03-1.21).

The association between operator volumes and clinical outcomes was independent of hospital volume. Even if the low-volume operator practiced in a high-volume center, they had worse short-term outcomes than those with higher volumes.

Defining Competency

The researchers performed several sensitivity analyses, including those to account for a potential learning curve, and the results were similar. The analysis was complicated by the fact that CMS requires both a surgeon and interventional cardiologist to jointly perform the procedure, which makes teasing out the primary operator tricky. However, the group ran several simulations—designating the primary operator as the clinician with the lowest and highest lifetime experience—and the results were broadly similar to the main findings.

Pershad pointed out that there are now three commercially available TAVI devices in addition to Pascal and MitraClip for M-TEER. As a result, relatively new operators with limited lifelong experience encounter each device infrequently. No operator performing a procedure once per month—the median number of M-TEER cases in this analysis, for example—can expect to be proficient.

“Unless you do this once per week, there’s just no way you can be anywhere close to competent,” said Pershad.

One of the difficulties of imposing a minimum annual volume requirement for operators is that doing so would be arbitrary. “It’s a continuous relationship,” said Cohen of the volume-outcome link. Moreover, volume is only a proxy for quality, he said, noting that “high volume doesn’t equal high quality.” Many lower-volume operators deliver exceptional care.

Most of the concentration of these providers has all been in urban hospitals, so you have a clear dilution of experience. Ashish Pershad

He noted that while hospital volume didn’t mitigate the relationship between operator volume and outcomes, the numbers are small and the strength of that subgroup analysis is weak.

“If you squint, you can start to see that the relationship, especially on the TAVR side, is starting to blunt,” said Cohen. “I think there’s a suggestion that a higher-volume hospital helps the low-volume operators, which makes sense either because someone is around to rescue if something bad happens or because a high-volume hospital has someone else to take the tough cases.”   

For Cohen, the analysis points to the strength of the STS/ACC TVT Registry for monitoring and continually improving quality. CMS is currently considering changes to the NCD for TAVI based on a request from Edwards Lifesciences, and one of the proposed changes includes ending coverage with evidence development rules that require data to be gathered through clinical trials or registries.

“That would be a shame because you can’t learn this stuff and you can’t act on these types of insights without having good data collection,” said Cohen.