Individualizing TAVR to Avoid Adverse Events

While incidence of adverse event rates remain low following transcatheter aortic valve replacement (TAVR), especially among experienced operators, researchers are looking to drive them down further through a greater understanding of how certain patient characteristics, procedural approach or even which device can impact outcomes.

Lessons from PARTNER

To avoid adverse events with TAVR, clinicians must first recognize that complications do occur. “While the PARTNER trial showed us that TAVR had a remarkable impact on survival in patients with high-operative risk with severely symptomatic aortic stenosis compared with medical therapy, there were also some important complications with the procedure,” says Prateek J. Khatri, MD, of the University of Toronto’s Institute for Clinical Evaluative Sciences, and Schulich Heart Centre, Sunnybrook Health Sciences Centre. 

Learning Curve & Survival: Kaplan-Meier survival in the first 25 vs. last 50 patients treated with transcatheter aortic valve replacement. Source: Elizabeth A. Magnuson, PhD

The prospective, randomized, controlled trial has revealed a great deal about this patient population, especially when assessing both cohort B—patients with severe aortic stenosis (AS) who were deemed inoperable—and cohort A—patients with severe AS who were deemed high-risk for surgery.

In late 2012, Philippe Généreux, MD, of Columbia University Medical Center/New York-Presbyterian Hospital in New York City, and colleagues combined both cohorts to identify incidence, predictors and impact of vascular complications (VCs) after transfemoral TAVR, using the first-generation Sapien valves, via a 22- or 24-Fr sheath (Edwards Lifesciences) (J Am Coll Cardiol 2012;60[12]:1043-1052).

The researchers included 419 patients (177 from cohort B and 242 from cohort A) from PARTNER. Overall, 15.3 percent had major VCs and 11.9 percent had minor VCs within 30 days of the procedure. Among patients with major VCs, vascular dissection (62.8 percent), perforation (31.3 percent) and access-site hematoma (22.9 percent) were the most frequent modes of presentation. Major VCs, not minor VCs, were associated with significantly higher 30-day rates of major bleeding, transfusions and renal failure requiring dialysis, and with a significantly higher rate of 30-day and one-year mortality. Major VC and renal disease at baseline were identified as independent predictors of one-year mortality.

In distinguishing between the two cohorts, the researchers found that VCs were more frequent and less well tolerated in the cohort B population, while in the cohort A population, the rate of VCs decreased in frequency and had less impact on mortality, explains Généreux. This finding is in spite of the fact that the larger device was typically used in cohort A.

“This is an important distinction because people often associate TAVR with higher rates of vascular complications, because the findings from cohort B were released and scrutinized first,” he says. Of course, because the cohort A patients are slightly less sick, this finding would be expected.

Also, the study authors found that the only identifiable independent predictor of major VC was female gender. “This is probably due to several factors; one being a smaller vessel size, but women also may have more fragile vessels,” says Généreux.

“An intrinsic increased vulnerability to peri-procedural complication during and early after an invasive procedure may be present in women,” the study authors wrote. “Exact mechanisms for this finding remain elusive and warrant further investigation.”

However, longer term, TAVR is still found to be beneficial for women. While this analysis of PARTNER demonstrated that although the female sex is a strong predictor of peri-procedural vascular complications, the female sex also was associated with a reduction in all-cause mortality at one year.

While not identified as independent predictors of adverse events, diabetes and small vessel size showed a trend toward experiencing vascular complications.

Lessons from practice 

“In clinical trials, we look at individual endpoints as independent entities, such as stroke or death; however, in practice, things are a little more complicated,” says Stephan Windecker, MD, head of interventional cardiology at the Swiss Cardiovascular Center Bern. “In clinical reality, one adverse event may trigger another one, as many events are interrrelated.”

To get an assessment of how these patients fare in real-life clinical practice with experienced operators, Windecker et al examined a cohort of 389 patients with a mean age of 82.4 years and a Society of Thoracic Surgeons score of 6.8, undergoing TAVR between 2007 and 2011 at the Swiss Cardiovascular Center Bern at Bern University Hospital (Circ Cardiovasc Interv 2012;5:856-861). The researchers performed a multistate analysis to study the sequence of adverse events during long-term follow-up in patients who underwent TAVR.

“A multistate analysis is a statistical approach that allows researchers to not only look at isolated, individual endpoints, but also take into consideration time-dependent transitions from one adverse event to another over time, which can ultimately lead to death,” Windecker explains. 

Transapical access emerged as a predictor of kidney injury and access site and bleeding complications, but this had no impact on the risk of subsequent stroke or death in this single-center study. There were no differences in clinical outcomes based on gender. 

Most often, patients complain about the pain associated with the transapical approach, says Généreux. “Especially in patients with small cavity and those with a lot of ventricle hypertrophy, the transapical approach can be very challenging because the sheath has to pass through the apex of the heart.” Particularly in the hands of lesser experienced operators, this complexity can result in complications leading to stroke.

Despite the complexity of the procedure and its painful nature, Généreux recommends the transapical approach for patients with low ejection fraction or chronic obstructive pulmonary disease. He adds that the newer iterations of devices, which continue to get smaller, may negate some of the negative outcomes with this approach.

To evaluate mortality outcomes of the two approaches, Himbert et al assessed 75 patients at high risk or with contraindications to surgery—51 patients were treated transfemorally and 24 transapically (J Am Coll Cardiol 2009;54[4]:303-311). Mean one-year survivals were 78 percent in the whole cohort, 81 percent in the transfemoral group and 74 percent in the transapical group. They also noted that mean one-year survival was 60 percent in the first 25 patients vs. 93 percent in the last 50 patients treated, causing the authors to conclude that “results are strongly influenced by experience.”

As clinical outcomes are better understood and operator experience increases, physicians can make better-informed decisions for unique patient considerations. For example, there are some advantages with the transapical approach, especially with the potentially fatal complication of paravalvular leaks. “[D]espite the current drawback of requiring a minithoracotomy … larger sheath diameters can be used with the transapical access, thus leading to less need for crimping of the valves, which may translate into better longevity,” wrote Thomas Walther, MD, PhD, and Jörg Kempfert, MD, of the Kerckhoff Heart Center in Bad Nauheim, Germany. “Solutions for improving paravalvular leakage may also be implemented into clinical practice through the TA [transapical] approach (theoretically these solutions require the larger diameters afforded by the TA route)” (Ann Cardiothorac Surg 2012;1[2]:216-219).

In the Windecker et al study, body mass index (BMI) of 20 kg/m2 or less emerged as a marker that increased the risk of stroke or death. “These patients are really underweight, so this may be a marker or indicator for frailty—which is a complicated biological condition, in which there is a process of accelerated decline of multiple bodily functions,” he says. Windecker discourages physicians from interpreting this risk factor as a reason not to treat patients with TAVR, which is “highly therapeutic in the overall cohort, including frail patients, even though they may be at increased risk.”

Also, the multistate analysis identified several concomitant risk factors. For instance, patients older than 80, with a BMI of 20 kg/m2 or less, prior stroke and a presence of atrial fibrillation at baseline have an increased risk of stroke and death after an inter-current event of access site and bleeding complications.

Based on these and other findings Windecker suggests that the differences between real-life clinical practice and the PARTNER trial aren’t that disparate, especially when TAVR is compared with conventional surgery or medical therapy. “Our experience outside of randomized clinical trials continues to mirror the results of PARTNER A and B.”

Despite knowledge gained through multiple studies and European experience since 2005, Windecker notes that there are not any clear-cut exclusion criteria for AS patients who are deemed appropriate for TAVR. “Obviously, patients who have a very low quality of life, or are bed-ridden or have a poor prognosis for longevity should not be evaluated for TAVR,” he says. The guidelines recommend that patients have estimated life expectancy of one year. “However, in our practice, we assess elderly patients with multiple comorbidities or who appear frail. We have implemented a systematic geriatric evaluation.”

Physicians perform functional, mental and intellectual evaluations of patients before and after TAVR. The researchers are currently studying the effectiveness of this evaluation. 

Are all valves created equal?

In a large meta-analysis, Khatri et al sought to amalgamate the complication rates from multiple, different studies to determine if PARTNER trial’s complications rates were similar to those in real-world, observational studies. The other objective was to compare complication rates between the Sapien valve and the CoreValve (Medtronic).

The researchers assessed a total of 49 studies enrolling 16,063 patients who met the inclusion criteria. All the included studies enrolled at least 100 patients who had TAVR, and reported at least one outcome of interest (Ann Intern Med 2013;158:35-46). Overall 30-day and one-year survival rates after TAVR were 91.9 percent and 79.2 percent, respectively.

Heart block requiring permanent pacemaker implantation was the most common adverse outcome (13.1 percent) and was five times more common with the CoreValve than the Sapien valve implanted using the transarterial route (25.2 percent vs. 5 percent, respectively). Among the devices used, the most common were the third-generation CoreValve, and the first-generation Sapien valve, but not the newer-generation Sapien XT, Khatri says.

However, assessing the more important outcome of mortality in a larger population, the CoreValve and the Sapien valves are more aligned. In a study of the Belgian national registry including 328 patients, Bosmans et al found that the one-month survival was 88 percent for Sapien- and 89 percent for CoreValve-treated patients (Interact CardioVasc Thorac Surg 2011;12[5]:762-767). Overall one-year survival was 78 percent in the CoreValve transfemoral-treated patients, 100 percent in the CoreValve subclavian-treated patients, 82 percent in the Edwards transfemoral-treated patients and 63 percent in the Sapien transapical-treated patients.

In the Khatri et al study, the overall rate of vascular complications was 10.4 percent—the second most common complication rate—and was highest with transarterial implantation of the Sapien valve (22.3 percent). “This is where the generation of the valves is important,” says Khatri. “The first-generation Sapien valve has a larger sheath size, compared with the third-generation CoreValve.” Previous studies have shown that the largest predictor of major vascular complications is the ratio of the outer diameter of the vascular access sheath to the femoral artery diameter, he says.

“Thus, the difference in major vascular complication rates between the two valves is likely a reflection between the two sheath sizes,” Khatri says. Acute renal failure requiring renal replacement therapy was the third most common complication, occurring in 4.9 percent of patients in this study.

TAVR for appropriately selected patients with severe AS is associated with a relatively low event rate in a high-risk population; yet, further progress in improved patient selection, device iterations and evaluation of the most optimal access approach are essential to improve actual results.