Optimal Decision Making in Aortic Valve Replacement

Saint Luke'S Hospital, Kansas City MO

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Aortic stenosis is a highly prevalent disease among elderly patients and causes reduced life expectancy, poor quality of life (QoL), and increased healthcare costs. In the setting of severe, symptomatic aortic stenosis, valve replacement is the mainstay of treatment, which has traditionally meant open-heart surgery. Recently, transcatheter aortic valve replacement (TAVR) has emerged as a less-invasive approach to valve replacement, and is particularly attractive in elderly patients with multiple comorbidities. n the Placement of AoRTic TraNscathetER Valve (PARTNER) Trial, which randomized patients too ill to undergo surgery to medical therapy or TAVR, TAVR patients had improved survival and better QoL than those receiving medical therapy alone. Despite the benefits of TAVR, nearly 1/3 were dead within 1 year of treatment, and approximately half did not benefit from TAVR (either dead or no QoL improvement at 1 year). Given the upfront risks and costs of TAVR, identifying patients, prior to the procedure, who are unlikely to benefit can enable patients and practitioners to make a more informed decision about whether or not to undergo the procedure. Using data from the PARTNER trial and other ongoing prospective studies, we will build economic and QoL prediction models to support the most efficient use of this emerging technology. In order to accomplish these goals, we plan to use both multivariable statistical and decision analytic models of survival, QoL and costs try to clarify the potential risks and benefits of particular patients undergoing TAVR, thus quantifying the heterogeneity of treatment benefits and enabling these estimates to be calculated on a patient-by-patient basis. We then plan to feed this information back to patients and practitioners at the time when the treatment decision is being made using a novel web-based technology that can generate individualized estimates of patients' predicted risks and outcomes. These estimates of clinical outcomes (e.g. QoL) can then be incorporated into patient-specific shared decision-making tools. Providing these data prospectively to patients and practitioners will support a novel dialogue, based on the evidence-based, projected outcomes of the individual patient. In addition, the economic models can support policy decisions that allocate TAVR in the most cost-effective manner. Altogether, these studies will allow for the most effective and efficient application of this exciting and innovative medical technology.