The study looks at a treatment called left atrial catheter ablation or LACA, which in the last few years has attracted tremendous attention as a new option for the 2.3 million Americans who have a heart-rhythm condition called atrial fibrillation.
LACA delivers tiny "zaps" of radiofrequency energy directly to the heart muscle to stop the electrical circuits that cause irregular heartbeats. The treatment is seen as an exciting option for atrial fibrillation, which interferes with daily life and greatly increases the risk of stroke. But like any new therapy, LACA carries costs and risks as well as benefits -- and until now, no one has assessed the economics of this treatment, or compared its costs and benefits with those of drug-based treatment.
The new research, published in the Journal of the American College of Cardiology by a team from the University of Michigan Cardiovascular Center and the VA Ann Arbor Healthcare System, takes a hard look at the dollars and cents of the issue.
Using sophisticated computer modeling, the researchers tallied the cost and benefits of LACA and drugs for people of different ages and health statuses. They ran calculations based on a range of assumptions about LACA's ability to reduce stroke risk, since the procedure is still new enough that its stroke-preventing power isn't clear.
In general, the researchers found, it's far more cost-effective to provide LACA to atrial fibrillation patients who are relatively younger (around 55 years of age), and who have one or more risk factors for stroke besides atrial fibrillation, compared with keeping them on traditional drug therapy to control heart rate. The advantage isn't so clear in older patients, nor in those who don't have other risk factors such as high blood pressure, diabetes, heart failure or prior strokes.
"Before LACA is more generally adopted as a treatment option, it's important to ask from a societal perspective if it's cost effective, and under what conditions," says first author Paul Chan, M.D., M.Sc., a fellow in cardiovascular medicine at the U-M Medical School and member of the VA Health Services Research & Development Center. "This analysis provides a conservative model that could be used to assess LACA's cost-effectiveness as more is learned about the impact of LACA on stroke risk."
Adds senior author and cardiovascular medicine associate professor Hakan Oral, M.D., who is a member of a U-M team that has performed LACA on more than 2,000 atrial fibrillation patients and published extensive research on the procedure, "Through rigorous modeling with conservative estimates, this study provides the groundwork for the cost-effectiveness of catheter ablation in treatment of atrial fibrillation."
He continues, "It appears that the longevity of the patient population and the ability of catheter ablation to maintain sinus rhythm and prevent future complications -- primarily stroke -- will be the key factors in determining cost-effectiveness of catheter ablation. However, besides cost issues, the value of improvement in the quality of life of individual patients should also be carefully considered."
Chan, Oral and their colleagues performed the study using a computer model that took into account life expectancy at age 55 or 65, the cost of the LACA procedure, the annual cost of drugs to control heart rate or heart rhythm for the rest of a patient's life, the potential cost of LACA complications, the potential cost of caring for a patient who had a stroke or cerebral hemorrhage, and knowledge from previous research about the stroke risk faced by atrial fibrillation patients based on age and health.
They also factored in results from research on the efficacy of rate-control and rhythm-control drugs, using atenolol and digoxin for rate control and amiodarone for rhythm control. They assumed that all patients would take blood-thinning drugs (aspirin or warfarin) to reduce the risk that blood clots would form and cause a stroke, and included the cost of monitoring for patients on warfarin or digoxin.
Most of the assumptions made in building the model, and the parameters for the different variables, were conservative, meaning they were set to give drug therapy a cost-effectiveness advantage over LACA. The model assumed that 80 percent of patients who had the LACA procedure would experience a complete return to normal heart rhythm after the first year, taking into account that some of these patients may need to receive repeat procedures in that time. The model also accounted for the fact that patients who were initially cured with LACA may re-develop atrial fibrillation over time. The researchers ran the computer simulation ten thousand times to more precisely define how much stroke-risk reduction LACA would have to provide in order to be cost-effective compared with drugs.
In the end, LACA's cost-effectiveness prevailed over either kind of drug therapy -- and rate-control drug therapy was more cost-effective than rhythm control by drug therapy in all cases.
In patients who had no other stroke risk factors besides atrial fibrillation, the model showed LACA would not be cost effective, costing an extra $98,900 per quality-adjusted life year (QALY), or year of life adjusted for quality of life during that year. The standard threshold for considering a therapy cost-effective is $50,000 per QALY. In 55-year-old patients with one or two risk factors besides atrial fibrillation, LACA's added cost per QALY was $28,700, meaning it would be cost-effective. In 65-year-old patients with one or two more risk factors, the figure was $51,800 -- not quite cost-effective.
Since the stroke-preventing power of LACA isn't known, the researchers calculated what percentage risk reduction LACA would have to achieve in order to make it a cost-effective option for patients in different situations. For example, if LACA has an initial "cure" rate of 80 percent for patients with atrial fibrillation, it would have to result in a 42 percent reduction in annual stroke risk to be cost-effective for 65-year-olds with a moderate stroke risk -- but only an 11 percent reduction in stroke risk for 55-year-olds. "This means that LACA doesn't have to have as big an impact on stroke risk to be cost-effective for younger patients, because the benefits appreciate over their lifetime," says Chan.
The researchers also repeated their analysis using less conservative assumptions that were not biased in favor of medical therapy. For example, when they assumed that rate-control drugs would restore only 20 percent of patients to regular rhythm at first, rather than 38 percent, LACA didn't need to achieve as powerful a stroke-reducing effect in order to be more cost-effective than drugs. (Recent research has suggested that rate-control drugs only help about 10 to 20 percent of patients return to regular rhythm.) Similarly, when they changed the assumption about how much rate-control drugs would cost per year, from $400 to $800, LACA was more likely to be cost-effective.
In the end, says Chan, more research is needed on LACA's long-term effect on stroke risk, quality of life and maintenance of normal heart rhythm. But as more data emerge, he hopes the U-M/VA model can be used to revisit LACA's cost-effectiveness and guide decisions about reimbursement.
In addition to Chan and Oral, the study's authors include VA/U-M assistant professor Sandeep Vijan, M.D., M.Sc., and Fred Morady, M.D., the McKay Professor of Cardiovascular Diseases and director of the U-M Clinical Electrophysiology Laboratory.
Dr. Chan is supported by an NIH Cardiovascular Multidisciplinary Research Training Grant and by the Ruth L. Kirchstein Research Service Award. Oral and Morady are founders and stockholders of Ablation Frontiers, Inc, and have consulted to Biosense-Webster and Ablation Frontiers. There was no grant or financial support from any party and no party had any involvement in the design, collection, management, or analysis of the study or in manuscript preparation.
For more information, contact: Kara Gavin, kegavin@umich.edu, or Mary Beth Reilly, reillymb@umich.edu, 734-764-2220
Journal
Journal of the American College of Cardiology