Ethics of high-tech, high-cost interventions

Advance planning needs improvement

It is important that the medical community understands that the cardiology and cardiac surgery communities are confronting end-of-life issues and the need to make appropriate treatment decisions based on an assessment of the likelihood that it will make a difference in a positive way, usually in a frail, sick, elderly patient’s life, argues Patrick O’Gara, MD, FACC, president-elect of the American College of Cardiology and executive medical director of the Carl J. and Ruth Shapiro Cardiovascular Center at Brigham and Women’s Hospital in Boston.

“There is a lot of ongoing research to determine the point at which patients become too frail and too sick to derive benefit from these new technologies,” he says. “We are very sensitive to this issue and determined to make sure we don’t travel headlong down the technological highway. We are very keen to make sure that these innovations are studied in the context of their appropriate application.”

It’s always easier to do something than to stop and ask whether it’s the right thing to do, says O’Gara. Heart failure teams have grappled with this for a long time when trying to determine if someone is a good candidate for transplant, relying on both physiological markers and an assessment of the family dynamic, he adds.

“Percutaneous aortic valve replacement looks like fabulous technology, but still anywhere from 30% to 50% of patients die within three years of having the procedure done,” he says. “There are people who die with aortic stenosis, and there are people who die because of aortic stenosis, and we wish to pick out the latter group.”

Many patients sent to cardiac surgeons are elderly, frail, and have multiple comorbidities for which an intervention like this is not likely to allow them to either live longer or live better, however. “We don’t offer the procedure under those circumstances,” says O’Gara. “We are trying to codify the decision making around it so that it’s done for the right reasons, and not just because it can be accomplished technically.”

Palliative care not always offered

High-tech, costly interventions also involve issues of distribution of resources, especially when considering these interventions at end of life or for patients who have limited life expectancy, are elderly, and/or have arguably reduced quality of life, says James N. Kirkpatrick, MD, an assistant professor of medicine in the Cardiovascular Division and in the Department of Medical Ethics and Health Policy at the Hospital of the University of Pennsylvania in Philadelphia.

“The expense involved in these technologies is thought to contribute to health care costs that do not have justified benefit. These issues can be compounded when the intervention is considered futile,” says Kirkpatrick. “Interventions which are inexpensive and futile, such as simple blood tests, do not raise the same questions as interventions which are costly and futile.” Of course, there is a question as to how to define futility, he acknowledges. “‘Won’t work’ can be defined in terms of not having any physiological effect, as when a patient is already dead, or probabilistically, as when an intervention has a chance to work but only a small one,” says Kirkpatrick.

A futile intervention could also be one in which there is a physiological effect, but the outcome cannot produce an acceptable quality of life. “In these situations, we begin to think about cost/benefit in terms of the cost and chance of different types of benefit, rather than just the magnitude of physiological benefit,” says Kirkpatrick. These considerations require more nuanced discussions prior to implant, casting the benefits of a device in terms of the patient’s overall values and goals, he adds. Palliative care practitioners are experts in these discussions, but need to partner with clinicians who are experts in the device technology, says Kirkpatrick.

“Although we don’t have data to inform us, I suspect palliative care is not routinely being offered in most centers, either as an alternative to [transcatheter aortic valve implantation] TAVI or [ventricular assist devices] VADs, or along with TAVI and VADS. Palliative care is not the same thing as hospice and does not necessarily preclude interventions,” says Kirkpatrick. Patients who are turned down for TAVI and VADS are probably sent back to referring clinicians with the presumption that they will receive appropriate palliative care, he explains, but this might not happen as much as it should.

Specialists and referral centers have the expertise in disease- and device-specific management, but primary clinicians and generalists usually have the longitudinal relationships, adds Kirkpatrick.

“We are left in a difficult spot trying to figure out whose responsibility it is to discuss palliative care,” he says. “Often, these folks are desperate to feel better and will grasp at whatever technology provides hope. In these cases, palliative care focused on symptom relief may be what they most need, whether or not they are candidates for high-cost interventions.” (See related story, below, on managing unrealistic expectations.)


  • James N. Kirkpatrick, MD, Assistant Professor of Medicine, The Hospital of the University of Pennsylvania, Philadelphia. Phone: (215) 662-7726. E-mail:
  • Paul S. Mueller, MD, MPH, Consultant and Chair, Division of General Internal Medicine/Professor of Biomedical Ethics, College of Medicine, Mayo Clinic, Rochester, MN. Phone: (507) 538-6341. E-mail:

Public expectations often unrealistic

A patient with severe, debilitating chronic illness who has a heart attack might be sent to a center miles away from his home and family for last-ditch efforts that aren’t likely to reverse the tide.

“We struggle at times with cardiac emergencies, to make the right decision in cases when it doesn’t seem to be appropriate, when the expectations of the family, the referring physicians, and everyone else are at odds with our sense of the utility of performing highly invasive procedures,” says Patrick O’Gara, MD, FACC, president-elect of the American College of Cardiology and executive medical director of the Carl J. and Ruth Shapiro Cardiovascular Center at Brigham and Women’s Hospital in Boston.

“We need to educate our aging population about what is realistic,” he urges. “We practice medicine in an era where there are often no limits in terms of what people expect. This is something we have to face as a society.”

This scenario raises pressing ethical issues with informed consent, argues O’Gara. “It is one thing to delineate the expected benefits and the potential complications of a procedure,” he says. “It is another thing to make sure one explains this in terms that can be understood by the patient and the family, and frame it in terms of what the long-term outcome might be, and allow people enough time to really understand the limitations and possible benefits of the procedure.”

Ideally, this decision-making process occurs well before patients become sick to the extent that is hard for them to understand the nuances of this discussion, says O’Gara. “If one is heading down the pathway of a ventricular assist device or a percutaneous aortic valve, the time to have this discussion is months before they become sicker, so the process can be iterative and people can be educated,” he says. “But that isn’t always possible. Sometimes people get sick overnight, and they come in and we have to make the best decisions we can.”

Paul S. Mueller, MD, MPH, consultant and chair of the Division of General Internal Medicine/Professor of Medicine and Biomedical Ethics at the College of Medicine, at Mayo Clinic in Rochester, MN, says the draw of the “treatment imperative” is very real for clinicians and patients. “This may differ by institution. But it’s hard to say ‘no’ to a promising technology,” says Mueller. “We offer palliative care when such treatments and technologies are not likely to benefit the patient, and when using them is not consistent with their healthcare-related goals.”

Mueller says clinicians should discuss end-of-life values and preferences with patients for whom they are considering life-extending technologies, before the technology is used. “Unfortunately, advance care planning is not where it should be and there is room to improve,” he says, adding that Mayo Clinic physicians talk about end-of-life planning with patients who have left ventricular assist devices (LVADs) or are being considered for them, rather than waiting until the last days of life when they may no longer have decision-making capacity.

“Such advance care planning ensures that clinicians carry out their wishes, rather than guessing,” says Mueller. “We are actually looking at the our TAVI patients to see which ones have engaged in advance care planning. We have already examined the same in our LVAD patients. All of our LVAD patients undergo palliative care consultation.”

Patients might not have thought about their values and preferences, says Mueller. “Discussing them, ideally along with loved ones, will clarify for clinicians what to do if the technology is unsuccessful and the patient loses capacity to make decisions on their own,” he explains.