Resuscitation Preferences in Community Patients With Heart Failure

Naveen L. Pereira, MD; Shannon M. Dunlay, MD


June 09, 2014

Editorial Collaboration

Medscape &

This feature requires the newest version of Flash. You can download it here.

Do-Not-Resuscitate Orders in Heart Failure

Naveen L Pereira, MD: Greetings. My name is Dr Naveen Pereira. I am an assistant professor in medicine and pharmacology. My area of specialty is advanced heart failure and heart transplantation. Today on Mayo Clinic Talks I have the privilege and pleasure of having Shannon Dunlay, who is an assistant professor in medicine in the division of cardiology and practices advanced heart failure and heart transplantation. She has a specific research interest in palliative care, quality of life, and frailty in heart failure, and we are going to be discussing a fascinating and very important paper that has just been published in Circulation: Cardiovascular Quality Outcomes.[1] Shannon was the lead author on this great paper. Welcome, Shannon.

Shannon M Dunlay, MD: Thank you very much. I am happy to be here.

Dr Pereira: Heart failure is a very morbid and chronic disease. It's progressive in many cases and has high mortality implications. It becomes very cumbersome, especially on the healthcare providers who deal with these patients, to pay careful attention to quality of life and outcomes as far as morbidities are concerned with these patients. We are often told that as part of decision making with these patients that we should get "do-not-resuscitate" (DNR) orders, and we should learn about the resuscitation preferences of these patients after explaining the morbidity and mortality associated with heart failure.

You have published an article titled "Resuscitation Preferences in Community Patients With Heart Failure." I was unaware that no systematic study had been done on this very important topic. Give me a little background as to what led you to do the study in the first place and then a brief overview of the population that you studied and what were you hoping to achieve.

Do Preferences Change Over Time?

Dr Dunlay: One thing that prompted me to do this study was the background that you mentioned, that heart failure is common and is associated with a high mortality rate. As providers, it behooves us to continually and iteratively assess our patients' preferences for care and discuss with them their personal goals, which can change over time.

In patients who are in the hospital, we always think about discussing what they would want to have done in the event of a cardiac arrest. We thought it would be important to evaluate and document resuscitation preferences in patients with heart failure and how these can change over the course of the disease and relate this to their health status and to their outcomes, including mortality.

Here in Olmsted County, we have the ability to do this through the Rochester Epidemiology Project . We have a well-established, large cohort of patients with heart failure. We looked at their documented resuscitation preferences and how they changed over the course of the disease and how these preferences related to outcomes.

Dr Pereira: Many great longitudinal studies about the prevalence of systolic and diastolic heart failure have come from the Rochester Epidemiology Project. You were specifically looking at the resuscitation preferences in this well-characterized cohort of patients. Can you describe how you went about doing this study? Was this an observational study, or did you intervene actively while trying to ascertain resuscitation preferences from these patients with heart failure?

Dr Dunlay: Yes, this was purely an observational study. We sought and documented their resuscitation preferences and how they changed over time in the course of the disease. Because patients were followed for several years on average, a significant proportion of patients died during follow-up, which allowed us to be able to look at changes over time until the time of death.

No Excess Mortality Risk With DNR

Dr Dunlay: The study had a few major findings. First, in patients with heart failure, resuscitation preferences changed over time. On initial evaluation, most patients actually prefer to be "full code," but closer to death, about three-fourths of patients prefer to be DNR.

This brings up an important point for providers. It's not a one-time decision. The resuscitation preferences for patients with heart failure can change over time. It's important for clinicians to have iterative discussions about the beliefs, goals, and wishes of patients with respect to resuscitation preferences.

We also found that a DNR preference is not an independent risk factor for death, and this is important because previous studies have had disparate results in this respect. Some studies[2,3] have suggested that patients who are DNR may have a higher risk for death that is independent of their comorbid conditions, age, and other factors.

We found that older patients and people with more comorbid conditions are certainly more likely to be a DNR. They are also more likely to die, but when you account for these other factors that could be confounding this relationship, no excess risk is associated with choosing to be DNR. This is important for patients to know, because sometimes patients will hesitate to switch their preference to DNR because they think that they won't receive care in other areas that could affect their mortality. We did not find this to be the case.

Finally, we found that in-hospital cardiac arrest in patients who are "full codes" is pretty rare in patients with heart failure. Only about 5% required resuscitation for an in-hospital cardiac arrest. However, when it did occur, the outcomes were very poor. Only 9% of the individuals who had a cardiac arrest and received cardiopulmonary resuscitation (CPR) in the hospital actually made a full neurologic recovery and were able to return home. Several studies[4,5,6] in the past have shown that this finding can influence patients' resuscitation preferences, because of an erroneous belief that they are highly likely to survive a resuscitation. However, we have found that outcomes are poor when people with heart failure receive CPR following a cardiac arrest.

The Right to Choose

Dr Pereira: You had 608 patients in this study—a fairly large number—in whom to ascertain resuscitation preferences. It appears that most of these patients were full codes at the time of enrollment into this study. What do you mean by enrollment?

Dr Dunlay: Patients were prospectively recruited into a longitudinal study under the auspices of the Rochester Epidemiology Project and Véronique Roger, who is the senior author on this manuscript. Those patients were followed over time through the Rochester Epidemiology Project for outcomes.

Dr Pereira: Patients were primarily recruited between 2007 and 2011, so this was the "modern era" in the treatment of heart failure, when discussing resuscitation preferences with patients has been well publicized in the guidelines. Was having most patients designated as full code at the time of enrollment a bad thing?

Dr Dunlay: No, I don't think it was bad. The individual chooses his or her resuscitation preferences, and it's our job to have the discussions with patients and to provide some sort of context for that decision. Many people with heart failure are doing quite well and are able to participate in all of their normal activities. It may be within their best interest and in alignment with their goals and personal beliefs to be full codes.

It is the job of the provider to have that conversation with patients, not just once but over and over again, and not just asking about resuscitation preferences but also providing them some context about their overall health status to help them make those kind of decisions. Providers should ask about their personal beliefs and their goals of care, to enable patients to make decisions about resuscitation, clinical treatment options, and many other things.

An important take-home point is that because goals and resuscitation preferences change over time, they can certainly change over the course of the disease as a person's health status worsens or changes. It's important for providers to talk with patients, not just once but on an iterative basis.

A nice paper came out a few years ago addressing this specifically. Larry Allen is the first author from a paper in Circulation[7] titled "Decision Making in Advanced Heart Failure." The authors point out is that it's important to revisit health status and resuscitation preferences and decisions, not just when patients are having an acute change in their health status (such as a hospitalization event) but also on a routine basis for patients who may be otherwise doing quite well. He and others have advocated for doing this as part of an annual heart-failure review, to discuss with patients where they are and where are they going and to ask them what they would want in the event of an unanticipated change in their health status.

The Right to Change One's Mind

Dr Pereira: You observed this cohort during follow-up. Over what period of time did you follow these patients?

Dr Dunlay: Follow-up was through April 1, 2013, so on average, most patients were followed for a few years.

Dr Pereira: What changes did you observe? At enrollment, 73% of patients were full codes. During follow-up, how many remained full codes?

Dr Dunlay: Over the course of the study, slightly more than one-third of patients changed their resuscitation preferences. At the time of death (and only a subset of patients actually died during follow-up) most were DNR. In general, however, about one-third of people (37%) changed their resuscitation preferences.

We also observed that some patients didn't just change their preferences once. Some would be DNR and then they would choose to be full code again, and then back to DNR. This underscores the fact that this is an iterative process, and patients have the right to change their minds.

Dr Pereira: Having looked at these data, a fairly substantial number of patients change their resuscitation preferences over time. Do you have any sense about what goes into making the decision as to being a full code vs DNR?

Dr Dunlay: That is a very interesting area worthy of further evaluation. I have an ongoing study that is going to look more specifically into some of the factors that affect end-of-life decision making and patients with heart failure. So I don't have an answer for you now, but hopefully we will know more in the future.

Dr Pereira: A large number of patients continued to be full codes during follow-up, and a large proportion of those did not suffer from a terminal event, according to your data. Was it the correct decision by the patient to remain a full code, on the basis of the physician's opinion that the patient was not going to experience a terminal event? How do you ascertain that?

Dr Dunlay: We should not place value preferences on patients' decisions about whether to be a full code or DNR. It's their personal decision. In patients with very poor health, who are unlikely to have a good outcome, it is still their decision about whether they want to maintain a full-code status.

In this study, it was rare for a patient who was a full code to undergo CPR, but it must be taken into consideration that many patients were DNR, so when they had a terminal event, they were not resuscitated. That was not captured in the 5% figure.

Take-Home Messages For Providers

Dr Pereira: That was an excellent summary of the data and the fascinating results of your study. Do you have any advice for healthcare providers about discussing this very important issue with patients and families?

Dr Dunlay: It is important to address these issues not just once but on an iterative basis. It is important to make this a part of the routine care of the patient. I am on board with the idea of annual heart-failure visits, during which you talk with patients, assess their goals of care, find out what they see as their health status, and where they see the trajectory of their heart failure going. You should provide as much input as you can about whether you agree with the patient's assessment and suggest modifications to the plan. Resuscitation preferences are only a small part of patient preferences.

The main take-home point from this study is that providers should talk with their patients. There is no right or wrong decision. There is only a right decision for an individual patient, so it is important to keep that conversation going.

Dr Pereira: You have answered my question about discussing resuscitation preferences. It should be an ongoing discussion.

Dr Dunlay: That's right, because preferences can change over time, as a person's health status changes. Don't ask just once—ask often.

Dr Pereira: Thank you, Dr Dunlay, for these great insights, and we would like to thank all of our listeners for tuning in to Mayo Clinic Talks.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.