The benefits of transcatheter aortic valve-in-valve implantation (VIV) do not justify the risk in patients with degenerated stenotic or regurgitant bioprosthetic valves who are candidates for reoperation. But it may be an acceptable approach in carefully selected high-risk patients and in those considered “no option,” according to a study published online Oct. 10 in Circulation.
Danny Dvir, MD, of Washington Hospital Center in Washington, D.C., and colleagues evaluated the efficacy of the VIV technique by analyzing clinical information in the Global Valve-in-Valve Registry, which collects data on patients with degenerated bioprosthetic valves. The patients came from 38 cardiac centers on four continents.
Dvir and colleagues analyzed data from 202 patients who had undergone the VIV procedure; the CoreValve (Medtronic) device was used on 124 patients and the Sapien (Edwards Lifesciences) device on 78 patients. Their mean age was 77.7 years and 52.5 percent were men. The majority (85.6 percent) of the patients had undergone one surgical aortic valve replacement (SAVR); the remainder had undergone between two and four SAVR. The median time between last SAVR and VIV was nine years.
The patients received VIV because they were experiencing stenosis (42.1 percent), regurgitation (33.7 percent) or combined stenosis and regurgitation (24.3 percent). The authors discerned no differences in the mechanism of failure between the CoreValve and Sapien groups.
The majority of the VIV procedures were performed using the smallest available size device: 79.8 percent of the CoreValve group used the 26 mm device and 79.5 percent of the Sapien group used the 23-mm device. In the CoreValve group, the main access route was femoral (91.1 percent) and in the Sapien group the main access route was apical (69.2 percent).
The authors found 14 procedural failures (6.9 percent) and 83 device failures (41.1 percent). Both procedural and device successes were higher in the CoreValve group.
Initial device malposition occurred in 31 cases (15.3 percent; 16.9 percent CoreValve and 12.8 percent Sapien). Post-implantation valvuloplasty was used in 25 cases (16.9 percent CoreValve vs. 5.1 percent Sapien). Seven patients experienced ostial left main coronary obstruction; in-hospital mortality for those patients was 57.1 percent.
At 30 days, 8.4 percent of patients died and one was lost to follow-up. The authors noted no differences in the rates of mortality, major vascular complications or stroke between the CoreValve group and the Sapien group.
Mean post-procedure gradients were 5 mmHg greater in the Sapien group and there was a significantly higher rate of high post-procedural gradients in the Sapien group (58.8 percent) vs. the CoreValve group (20 percent) for VIV performed inside surgical bioprosthesis with internal diameters of less than 20mm. Severity of bioprosthetic stenosis and use of a Sapien device were independent predictors for high post-procedural gradients.
There were 23 deaths within the first post-procedure year; the one-year survival rate was 85.8 percent, with no significant differences in survival rate between the two groups.
"The VIV approach results in considerable hemodynamic improvement," the authors wrote. They also pointed out that "significantly elevated post-procedural gradients are common after VIV." The authors postulated that this may be due to the reduced area available for the functioning valve when implanted inside a surgical bioprosthesis or patient-prosthetic mismatch.
In an accompanying editorial, E. Murat Tuzcu, MD, Samir Kapadia, MD, and Lars G. Svenson, MD, PhD, of the Cleveland Clinic wrote that the "30-day rates of mortality (8.4 percent) and stroke (2 percent) were not much different from what was observed in the early experience of TAVR [transcatheter aortic valve replacement] for inoperable patients." But they expressed concern about the high rate of elevated gradients post-VIV.
Tuzcu and colleagues noted that determining the internal diameter of a particular valve is difficult, because there is often a discrepancy between the internal diameter of a valve and the manufacturer's label size. They wrote that pannus formation, degenerated and calcified prosthetic valve leaflets and stenosis also have an impact on the internal diameter of the valve.
"Creating a relatively small effective orifice area and consequently high transvalvular gradients will frequently lead to patient prosthesis mismatch. Whether this ... translates into higher long-term mortality or functional impairment remains to be seen ... VIV for a 'no option' patient with severely dysfunctional 19 mm or 21 mm prothesis may not be a perfect solution but it may offer an acceptable result with improvement in symptoms," they wrote.
The authors pointed out that in future devices designed specifically for VIV may become available, making the technique more feasible. They also suggested that surgeons should implant a large valve in the first surgery if they envision that a TAVR procedure may be necessary in the future.