Redo Surgery for Bioprosthetic Aortic Valve Failure on the Decline Since 2015

SAN DIEGO, CA—Patients who received redo surgical aortic valve replacement (SAVR) following bioprosthetic valve failure tended to be younger and lower risk following the 2015 approval of transcatheter devices for aortic valve-in-valve (ViV) procedures in the United States, according to new registry data.

“This may reflect the adoption of ViV transcatheter aortic valve replacement in older and in higher-risk patients,” said Ankur Kalra, MD (Cleveland Clinic, O H; formerly University Hospitals), who presented the research last week at the annual meeting of the Society of Thoracic Surgeons (STS). “Our results provide a national benchmark against which outcomes for valve-in-valve transcatheter aortic valve replacement can be compared in future studies.”

For the study, Kalra and colleagues included data from 4,239 patients undergoing isolated SAVR for a degenerated bioprosthesis from the STS database between 2012 and 2016. Between 2012 and 2014, the number of patients increased by 7.25%; however, following the approval of ViV TAVR in 2015, the number of patients decreased by 3.4% (P = 0.005). Additionally, the mean age of patients receiving redo SAVR fell from 65.80 to 64.45 years throughout the study period (P = 0.001).

Mean patient STS predicted mortality fell from 4.55% in 2012-2014 to 4.25% in 2015-2016 (P = 0.047). However, there was no difference in operative mortality in the earlier versus later years of the study period (3.5% vs 4.0%; P = 0.357).

There were no disparities in stroke, need for permanent pacemaker, or acute renal failure between the study groups, but A-fib was less likely (20.8% vs 23.7%; P = 0.024) and length of stay was longer (10.2 vs 9.7 days; P = 0.04) for the patients treated before compared with after 2015. Kalra suggested the possibility of A-fib being better diagnosed in the latter years of the study period being responsible for this finding.

Risk factors associated with mortality were identified as older age (OR 0.982; P = 0.041), male gender (OR 2.147; P = 0.001), and use of intra-aortic balloon pump (OR 0.119; P =0.0001), while PAD was associated with stroke (OR 2.584; P = 0.001).

‘Where’s the Beef?’

Discussing the study during the session, Robert Guyton, MD (Emory University School of Medicine, Atlanta, GA), said it is “a very nice verification of the expected decrease in redo surgical valve replacement as redo valve in valve became available.” While there are no surprises in the findings, “where's the beef?” he asked, inquiring whether there are more data on the apparent decrease in redo SAVR in higher-risk patients. “This vegan study would be better if it had a little bit more meat,” Guyton commented.

Kalra replied that the data do exist and his team plans to include this information in their manuscript.

Guyton also noted the numerical increase in mortality observed in the study, saying: “Was there not therefore necessarily a significant increase in the [observed-to-expected] outcomes from the earlier group to the latter group? Did you look at this and how would we explain our decreasing ability to get these patients through?”

“This finding is a play of chance, and I don’t think the data can fully explain it,” Kalra responded.

To TCTMD, session co-chair Derek Brinster, MD (Northwell Health, New York, NY), also said he is “not terribly surprised by [Kalra’s] results. In terms of the increased utilization of TAVR for surgical failure of bioprosthetic valves, I think that's going to be an increasing trend that we see throughout the nation. I think his results just reflect that.”

One concern Brinster had is the fact that the bioprosthetic failures happen at least 10 years after implantation and thus trends might not be the same today as they were, say, 10 to 20 years ago. “So if you're looking at 2012-2016, were the same amount of valves being put in 2002-2006? That we don't know,” he pointed out.

It also would be interesting to know the exact rate of valve implantation from when the failed study valves were first put in, Brinster continued. Without that information, “this is an assumption study that the valves that we're not treating by surgical re-op AVR are being treated by TAVR,” he said. “That's an assumption here that we don't have defined.”

This could be done perhaps by linking this study population to ViV TAVR registry data, Brinster suggested. “They should be kind of opposite graphs, [with] the surgical AVR graph going downward and the valve-in-valve tendency going upward. We know the valve-in-valve rate is going up, but we don't have a good understanding of the total ‘n.’ It's just a presumption.”