Transcatheter aortic valve replacement (TAVR), which has rarely been described as a significant bargain economically, emerged as more expensive than previously reported in a cost-effectiveness analysis published July 9. A team from the PARTNER trial agreed with the researchers on one point: the need to define patient groups for whom TAVR provides good value.
The results were published online in Circulation: Cardiovascular Quality and Outcomes.
The PARTNER trial gave TAVR a boost with findings that it improved survival and quality of life in patients with severe aortic stenosis who are not suitable candidates for surgery. An economic analysis of this patient group, cohort B, presented at American College of Cardiology’s annual scientific sessions in 2011 placed the incremental cost-effectiveness ratio (ICER) at approximately $62,000 per quality-adjusted life-years (QALYs) gained. Later that year, the FDA approved the Sapien transcatheter aortic heart valve (Edwards Lifesciences) as a treatment for this patient population.
Cyrena T. Simons, MD, PhD, of the Veterans Affairs Palo Alto Healthcare System in California, and colleagues used a Markov model calibrated to the PARTNER trial to assess TAVR’s cost-effectiveness compared with medical management. They calculated that medical management had an average discounted life expectancy of 2.1 years and 1.2 QALYs compared with TAVR’s discounted life expectancy of 2.9 years and 1.9 QALYs.
TAVR reduced subsequent hospitalizations but had more complications, particularly from stroke. Lifetime discounted medical costs increased $85,600 with TAVR for a lifetime cost of approximately $169,100. Its ICER compared with medical management was $99,900 per life-year gained and $116,500 per quality-adjusted life-year gained.
“Our analysis suggests TAVR improves average health outcomes but is relatively expensive, particularly in patients with aortic stenosis who have high healthcare costs from other comorbid conditions,” Simons et al wrote. “Our results highlight the importance of patient selection for TAVR because patient-level characteristics (baseline healthcare costs, extent of comorbidity, and life expectancy) were more predictive of the cost-effectiveness of TAVR than procedure-level characteristics (such as procedure cost or initial effectiveness).”
Given the pent-up demand for TAVR, the first wave of patients may be sicker and older followed by newly symptomatic patients, they wrote. “Thus, TAVR may be more cost-effective in the future as it is applied to incident patient cohorts that are generally healthier.”
They noted several differences between their model and the PARTNER cohort B findings, including a smaller increase in quality-adjusted life expectancy with TAVR, higher lifetime costs with TAVR and a higher life expectancy in the medical management group.
In an accompanying editorial, PARTNER researchers Matthew R. Reynolds, MD, MSc, of the Harvard Clinical Research Institute in Boston, and David J. Cohen, MD, MSc, of Saint Luke’s Mid America Heart Institute in Kansas City, Mo., emphasized differences between PARTNER and current practice that could affect outcomes, including newer generation technologies and gains in expertise.
The survival advantage with TAVR adds to higher accumulated lifetime medical costs, which makes TAVR appear less cost-effective. “[A]t some point, if background medical costs exceed societal willingness to pay, any life-extending medical intervention will fail to be a good value in health economic terms,” Reynolds and Cohen wrote. “As Simons et al conclude, the cost-effectiveness equation at some point becomes much more about the patients than about the technology.”
The editorial writers reiterated the need to better define patient groups who gain little from TAVR to help patients and heart teams in their decision making.