Dr William T Abraham (Source: Ohio State University)

"I believe the data at present most strongly support the use of CRT, with or without a defibrillator, in patients with class-3 heart failure and in ambulatory patients with class-4 heart failure," said Dr William T Abraham (Ohio State University, Columbus). But not in patients with more severe class-4 disease. "There is a spectrum of class 4, from the patient who is symptomatic at rest or with minimal exertion but still capable of performing some activities of daily living, to the class-4 patient who is more advanced or end-stage and can't do those activities, who might depend on intravenous medications or is currently hospitalized." The latter group, he said, may be inappropriate for either CRT or an ICD.

Since many of those patients get CRT and then they become class 2 or class 3, it is probably reasonable to think about putting in an ICD.

"That raises the question of what to do about the ambulatory class-4 patient who is going to get CRT. Should they get CRT with an ICD?" There is good rationale, Abraham said, for adding an ICD in both NYHA class 3 and ambulatory class 4.

"It's a difficult question, but since many of those patients get CRT and then they become class 2 or class 3, it is probably reasonable to think about putting in an ICD," Abraham said. "Otherwise you might end up having to subject them to a second procedure to upgrade their device....It's not cost effective from the public-health standpoint, and it does subject patients to the risk of two procedures rather than one."

A class-4 patient who is a candidate for CRT should also be a candidate for an ICD.

Konstam agrees that a CRT/ICD indication in NYHA class-4 patients can be problematic. The COMPANION trial, he noted, showed benefits in both class 3 and 4 for the combination device. And since CRT can improve functional status, perhaps converting class 4 to class 2 or 3, a class-4 patient could become eligible for an ICD. "I think a class-4 patient who is a candidate for CRT should also be a candidate for an ICD....You may as well consider inserting them both at the same time." Except, he said, "I don't think an ICD should be implanted in a patients with a poor prognosis, for ethical reasons as well as the fact that it has not clearly been shown to benefit this group."

Linde, noting the 10- to 15-year gulf between the typical age of patients in the device-therapy HF trials and those in everyday practice, said age per se shouldn't necessarily decide whether someone receives a device. "I do not use chronological age to make this judgment. I assess individual patients and their lifestyle." But she may be more inclined to use CRT than an ICD in a very old person. "Most patients are positive with regard to CRT devicesthey want to feel better and 60% to 80% of them will with one of these devices....I would give CRT to elderly people with CHF even though evidence is lacking. Yes, their susceptibility to complications is greater, but I would still recommend it on an individual basis." However, she said, she would be hesitant to implant a device if the patient had multiple-organ failure or other comorbid conditions. Beyond the infection risk, both CRT and ICD therapy are associated with a risk of lead displacement and other complications.

Dr Anthony Steimle (Source: Kaiser Permanente Medical Center)

The recipient's overall health is another consideration. "The patients' comorbidities and whether they are likely to die from pump failure or some other condition in the near future affects how likely they are to benefit from the device," said Dr Anthony Steimle (Kaiser Permanente Medical Center, Santa Clara, CA). "So that is part of the criteria for whether we offer them a device. There's certainly a lot of the art of medicine there."

Clark said, "When CRT is offered, patients almost invariably want it as it makes them feel better. If they are symptomatic after maximum medical therapy, they are very grateful to try something that might make those symptoms go away. But ICDs are very different. They don't make you feel better, they just keep you alive."

He wouldn't offer an ICD to a severely ill patient and says many elderly infirm patients don't want one.

If I think patients are 'old and crumbly,' I don't even open the subject of an ICD.

"Sometimes they don't want to be kept alive," Clark said. "But this must be an individual decision in each patient. It is not one based on chronological age. If I think patients are 'old and crumbly,' I don't even open the subject of an ICD. But I would still discuss CRT if they were eligible."

"We should not be thinking about ICDs as salvage therapy for end-stage disease," Abraham agreed. "In end-stage patients, there is a greater likelihood of death [from pump failure], and the defibrillator is probably less likely to alter the natural history of the disease and their outcomes."

Dr Mandeep R Mehra (Source: Ochsner Clinical Foundation)

"In looking at the quality of life, one also has to look at the quality of death, which is often left unaccounted for in the trials," observed Dr Mandeep R Mehra (Ochsner Clinical Foundation, New Orleans, LA).

"If I were considering an ICD, even with resynchronization therapy, for very elderly patients with advanced, late-stage heart failure, I would sit down and talk to them about their expectations for a meaningful prolongation in life," Mehra said. He'd be reluctant to offer the device to such patients. "Most of these patients may not want that, frankly. They may not want the misery of transforming their modality of death, which is what we'd actually be doing for them."

Abraham would "try to have an honest discussion with patients about what their goals are." From the clinician's perspective, the primary goals are twofold: "One is to make the patients feel better, and the second is to help them live longer. But it's interesting when you talk to patients, sometimes the relative weight they place on one goal over the other is different from what the clinician would place," he said.

"In all honesty, what we should do is to provide adequate information for patients to be active participants in the discussion," according to Abraham. "Some people will take quality of life over duration, and others will want duration over quality. And some will want to achieve it all if they can, and in many instances that's a reasonable goal."

Some people will take quality of life over duration, and others will want duration over quality.

Noted Steimle, "I've had patients who are doing very well, whom I believe meet the criteria to benefit from an ICD, and I've offered them the device." Often the reply is "I'm not interested in a device whose full purpose is to prolong my life but isn't going to make me feel better now."

"Sometimes that surprises me, but it's a highly personal choice. Some people find the prospect of going through a surgeryeven a minor one, and then having a device under their skin that requires monitoringunappealing if it's not going to make them feel better," Steimle said. Most agree to have the devices, but "I've had a number of people turn it down."

Make me feel better or leave me alone.

Konstam agrees that patients' views vary widely. "Some may say, 'Make me feel better or leave me alone.' But in general most people do want to live longer. This stops only when patients are severely debilitated with no joy left in life....I have seen patients who are only modestly limited but don't want to go on. Others are severely debilitated but still get joy from their families and they do want to go on. All patients have to be counseled and what they really feel has to be solicited. But I would not give an ICD to someone near the end of life who has no reasonable quality of life."

"These are complex issues for physicians. Imagine how complex they are for patients," observed Konstam. "How much does a patient really understand about what it is like to be shocked? How many patients really understand the difference between CRT and ICD or the combination of both?"

You are not conferring immortality with a piece of hardware.

There are key points to cover when discussing the possibility of ICD implantation with a potential candidate, noted Dr Lynne Warner Stevenson (Brigham and Women's Hospital, Boston, MA). First, the device won't reduce the risk of death as much as patients often think. "I have considerable concerns about presenting patients with information on the percent of risk reduction, which is a complex concept," she said. "Patients hear 30% risk reduction or 23% risk reductionthat sounds like a really big number." But the absolute numbers are much smaller. "I like to use this example: I can decrease your risk of dying from lightning by 50% if I prescribe thick rubber boots and a rubber hat to wear all the time. On the other hand, your risk of dying from lightning is very small to begin with."

Some patients...can experience posttraumatic stress syndrome.

"The second thing they need to understand is that not all deaths are preventable by an ICD. So you are not conferring immortality with a piece of hardware," Stevenson said. "Third, they should understand that there are undesirable aspects to ICD therapy as well." Inappropriate shocks are a foremost example. Up to one third of HF patients have atrial fibrillation, which does not necessarily elicit shocks "but is the most common time when inappropriate shocks occur." Reprogramming the device can eliminate recurrence of the problem, Stevenson noted, "but some patients are quite distressed by this and can experience posttraumatic stress syndrome."

'Don't you want to see your grandchildren graduate?' That's doing everyone a disservice.

"We have not been sufficiently responsible in presenting these issues to patients so they can really decide in a rational way," Stevenson said. "Too often they may see an advertisement along the lines of, 'Don't you want to see your grandchildren graduate?' That's doing everyone a disservice."

Although some patients have unrealistic expectations for what device therapy may accomplish, "it probably cuts both ways. I think some practitioners give patients an overly pessimistic view," according to Abraham. But, in fact, he noted, HF is no longer the relentlessly progressive syndrome it was before the age of ACE inhibitors, beta blockers, and other available drug and device therapies.

I've got patients who came to me with the sense that they'd be dead in six months, and they're alive 10 years later still enjoying a pretty good quality of life.

Still, "we routinely have patients referred to our clinic after being managed by primary-care physicians and others along the way," Abraham said, "and they come with an advanced degree of disability with the sense that they may have only a few months to live, with a sense of hopelessness that nothing can be done to make them feel better. After taking them through a protocol of evidenced-based treatment, many of these patients improve to a better functional status and look forward to a much brighter outcome....I've got patients who came to me with the sense that they'd be dead in six months, and they're alive 10 years later still enjoying a pretty good quality of life."

Finally, what could be the final challenge in managing the patient's care: "There will be some changes that come about such that it becomes time to die," Stevenson said. "It's important for patients to be explicitly be told that circumstances may arise in which the defibrillator feature should be turned off. Too often, this is never discussed with the patient or family until end-stage disease is upon them, and then it's a very difficult concept to embrace and often confused with discontinuing life support."

Should I reach a stage in which I'm terminal and irreversible, please shut off my device.

Patients still in manageable stages of their disease will sometimes request deactivation of their ICDs in their living wills, Mehra noted. "They say, should I reach a stage in which I'm terminal and irreversible, please shut off my device. But that's usually a very informed patient of a very informed physician who takes the time to speak to the patient." His sense is that patients do not learn about such options from their primary-care physicians. "I've been doing this for over a decade now, and I've almost never had this conversation with a referring physician."

There are some cardiologists who believe that it is ethically inappropriate to turn off the device.

But according to Abraham, "There are some cardiologists who believe that it is ethically inappropriate to turn off the device, that we have an ethical obligation once the device is in to maintain it." It is true that such physicians would be less inclined to put ICDs in patients with advanced disease, he noted, "but of course it's a progressive disease. So if they put it in a class-2 or class-3 patient who a couple of years later becomes an end-stage class 4that's where the issue sometimes arises."

He added, "Our view has been, in patients who are in end-stage disease and define their own quality of life as being extremely poor, that if they want to avoid what in some instances become incessant shocks and to accept sudden cardiac death as the mechanism of their death, then that is fully appropriate."