Isolated SAVR Volumes Slide as TAVR Numbers Mount in the US

The observational findings highlight the growing questions as to what the threshold might be for case volume in relation to procedural outcomes.

Isolated SAVR Volumes Slide as TAVR Numbers Mount in the US

As TAVR volumes in the United States have grown, those of isolated surgical aortic valve replacement (SAVR) have dropped, as have the comorbidities of the patients treated surgically, according to observational data through 2014. Moreover, patient mortality following SAVR has decreased among the highest-volume TAVR centers.

The implications of these shifts are profound for the field, according to senior author Robert Yeh, MD (Beth Israel Deaconess Medical Center, Boston, MA). “TAVR growth will continue to change the SAVR population, both in volume and risk profile, and one can envision taking this out a few years and seeing hospitals with historically robust SAVR programs really see those volumes decline,” he told TCTMD in an email. “How the field thinks about maintaining high quality in the face of declining volumes for a procedure with a robust volume-outcome relationship will be an important challenge.”

Secondly, Yeh continued, in the data set “there were still a large number of low (or no) TAVR volume facilities whose SAVR populations were very different than those at high TAVR volume places. One wonders whether patients treated at those facilities have had access to TAVR. How do we ensure that a potentially life-saving procedure is equitably made available to all patients?”


For the study, Yeh along with lead author Harun Kundi, MD (Beth Israel Deaconess Medical Center), and colleagues looked at all TAVR and SAVR cases performed in the United States between 2011 and 2014. After excluding cases performed at hospitals which did not maintain average volume of at least one TAVR or SAVR annually throughout the study period, their final population included 37,705 SAVR procedures.

Patients undergoing SAVR in hospitals that did not offer TAVR saw a reduction in certain patient covariates over the study period including age, chronic heart failure, CAD, prior CABG, peripheral vascular disease, chronic kidney disease, and chronic pulmonary obstructive disease. Further, SAVR done in hospitals also performing TAVR was associated with a decline in additional covariates including diabetes, prior MI, prior PCI, cerebrovascular disease, and home oxygen in patients at baseline. By contrast, there was little in the way of change over time in comorbidities among patients undergoing SAVR patients at hospitals with the lowest TAVR volumes.

Isolated SAVR procedural volume decreased among centers performing the highest number of TAVR procedures, as defined by the top two quartiles by volume. Also, mean TAVR volume exceeded that of SAVR in all but the lowest TAVR-performing hospitals, with the researchers estimating the crossover point to have occurred as early as 2012 at the highest-volume institutions.

Interestingly, 30-day all-cause mortality remained the same for SAVR patients throughout the study in the lowest-volume TAVR centers but decreased substantially for the higher-volume TAVR hospitals, with the largest annual rate of decline observed in the top quartile of TAVR centers. Additionally, 1-year all-cause mortality following SAVR went down in all quartiles of TAVR volume (P < 0.001 for all).

A sensitivity analysis including all TAVR and SAVR cases regardless of annual hospital volume confirmed the primary results.

“These findings suggest that TAVR availability may have led to the selective use of SAVR in lower-risk patients. It may also partially reflect secular trends in mortality for AVR patients, as SAVR mortality declined even in hospitals not performing TAVR,” the authors write. “Our study suggests that in low TAVR volume facilities, the number of transcatheter procedures was not large enough to displace surgical volumes. As TAVR use continues to expand worldwide, these results suggest that the volume of SAVR may continue to decline.”

Kundi and colleagues add: “Whether this shift toward less complex surgical patients may mitigate the adverse effects of declining surgeon volumes for SAVR is unknown.”

The results were published this week ahead of print in JACC: Cardiovascular Interventions.

‘Intuitive’ Findings

Tom Cahill, MBBS (Oxford University Hospitals, England), said the headline finding of this study is what “a lot of us would expect and intuitively feel,” namely, that there has been a change such that the highest-risk surgical patients, who previously would have been offered a high-risk operation or no therapy, are now getting TAVR.

Indeed, the findings of the current study differ somewhat from 10-years’ worth of data Cahill himself presented at the recent London Valves 2018 meeting, showing that while the number of TAVR procedures have risen exponentially, SAVR procedures have stayed constant. Those numbers, he said at the time, hint that TAVR may have predominantly helped patients at the higher end of the risk spectrum, to whom surgery was not previously being offered.

Given the shift seen here, “we need to be really asking the question about whether a center that's only offering SAVR is really doing the best thing by patients,” he told TCTMD. “If you were a patient going forward, and there is an increasingly large proportion of patients who might be suitable for both procedures, I think you would want to know that you were being discussed in a heart team meeting where both strategies were available, both strategies were on the table.”

However, this doesn’t necessarily mean that TAVR should be offered at more hospitals today, Cahill clarified. “We know that there is a likely benefit to concentrating expertise in high-volume centers. So we need to make sure that the full spectrum of options is on the table, but equally that we don't dilute our expertise such that everywhere is doing small volumes of cases,” he said. “It's a balance because you don't want to have a patient have to travel halfway across the country, but equally you want centers of relatively high volume and expertise to build up.”

We need to make sure that the full spectrum of options is on the table, but equally that we don't dilute our expertise such that everywhere is doing small volumes of cases. Tom Cahill

In an accompanying editorial, Vinod Thourani, MD (MedStar Heart and Vascular Institute, Washington, DC), and colleagues argue for additional data to home in on the exact relationship between volume and outcomes in both TAVR and SAVR, as well as the need for on-site surgical backup for TAVR centers and optimal resource utilization.

“The ideal aortic valve disease program requires cardiology and cardiac surgical expertise, together with a support network that includes anesthesia, intensive care, specialist surgical and medical services (neurology and aged-care medicine), nursing, and allied health. This is most likely to be found in a high-volume center and lead to the best results, especially in high-risk patients,” they write, noting that the same could be said for SAVR.

“Striking a balance between patient conveniences with a large number of low-volume centers versus improved clinical outcomes with regionalization of services should be guided with evidence,” the editorialists conclude. “This is most likely to come from large population sets, and the importance of the STS/ACC Transcatheter Valve Therapy Registry in monitoring TAVR outcomes in community cannot be underestimated.”

Benefits From Regionalization

Commenting to TCTMD, Maral Ouzounian, MD, PhD (Toronto General Hospital, Canada), a cardiac surgeon who was not involved in the study, said she was surprised at how few US hospitals qualified for inclusion into the study based on their volume criteria. “We know that there is a volume-outcome relationship certainly in the surgical population, with any type of complex surgery,” she said. “If the risk profile is improving, then maybe the volume-outcome relationship is less important—maybe surgeons can do well with less risky patients having surgical AVR.”

This type of relationship remains to be demonstrated with TAVR, Ouzounian continued, but it likely exists. “The procedure might have been simplified enough that there isn't much of [a relationship between volume and outcomes],” she commented. “Having said that, I think there’s still going to be a minimum number of cases per year to maintain your skills and to maintain the quality of the procedure. If you’re not in a room doing TAVR at least a couple of days a month, you’re not going to keep that skill set up. If you're only doing a handful a year, there's just no way to maintain your skills and keep up with the literature and new techniques and the subtleties of valve choice and technical details.”

If the risk profile is improving, then maybe the volume-outcome relationship is less important—maybe surgeons can do well with less risky patients having surgical AVR. Maral Ouzounian

Perhaps the United States could stand to learn from Canada, where SAVR is only offered at 33 centers nationwide and TAVR only at 28 of them, she suggested. “I think that the US could stand to benefit from a little bit more concentration of expertise in centers, [because] centers doing higher volumes of whatever procedure it is tend to have better results,” Ouzounian said. However, “we need data before we can recommend that small centers should not offer surgical or percutaneous AVR. We need better data to actually look at the threshold and inflection points above which outcomes are improved, but I do think that data should be looked at and that some consideration should be given to minimum surgical and percutaneous volumes of valve implantation per year, per site.”

Both Cahill and Ouzounian said they would like to see this study extended out to today to see if the transition of TAVR to more intermediate surgical risk patients has had an effect on volumes and outcomes. In the current analysis, “it's probable that some patients did not have access to TAVR and probably should have been having a percutaneous valve instead of a surgical valve and just did not have access to it,” Ouzounian said. “It would be very interesting to see how these data project into current times. I would suspect that it’s just going to keep going in this direction.”

“This field is still in evolution. TAVR technology is improving. Trials are being conducted in lower and lower risk populations,” Yeh concluded. “So there has been a paradigm shift, but that shift is not yet complete. As the field matures, identifying and measuring TAVR hospital quality, promoting equitable patient access, and ensuring that we sufficiently concentrate expertise will be important challenges to tackle.”

  • Yeh reports receiving investigator-initiated grant funding from Abiomed and grant support from Boston Scientific, and serving as a consultant for Abbott, Medtronic, and Teleflex.
  • Kundi, Thourani, Cahill, and Ouzounian report no relevant conflicts of interest.