Editor's Note: Below, see the video discussion of this article with Dr. Kern and Dr. George Vetrovec.
I received a query from a cath lab director in the southeastern U.S., who asked, “How should we manage the ST-elevation myocardial infarction (STEMI) patient coming to the emergency department (ED) who appears to be so critically ill that further care would be futile?” The lab director works with physicians who, over concern for overlooking salvageable cases, are aggressive regardless of the high likelihood of death and low likelihood of survival. The expense of inserting an Impella (Abiomed) or other mechanical support pump for little or no clinical gain is of concern. For many operators, the easiest thing to treat all shock STEMI patients the same way; that is, do everything for everybody. His questions:
1. Do you have or know of guidelines or suggestions on medical futility to help guide physicians in the STEMI patient? Is there a scoring system?
2. For out-of-hospital cardiac arrest with an unknown time to return of spontaneous circulation entering the emergency department with STEMI, do you go directly to the lab, or stabilize and assess neurologic function first? Would such a patient in shock get an Impella?
3. For late-arriving STEMI in Society for Cardiovascular Angiography and Interventions (SCAI) shock stage D or E, do you proceed with ‘the whole enchilada’?
Before viewing the experts’ responses, let’s talk about medical futility, a concept ill-defined and often meaning different things to different people. Here are 3 definitions I found on the internet.1
1. Futile treatment: Treatment evaluated by the healthcare team, family, or both, as being non-beneficial or harmful to a dying patient.
2. Futility means any treatment that, within a reasonable degree of medical certainty, is seen to be without benefit to the patient, as when the treatment at issue is seen as ineffective regarding a clinical problem that it would ordinarily be used to treat (this one sounds lawyerly to me).
3. Medical futility refers to interventions that are unlikely to produce any significant benefit to the patient: a) quantitative futility, where the likelihood that an intervention will benefit the patient is exceedingly poor, and b) qualitative futility, where the quality of benefit that an intervention will produce is exceedingly poor.
In the real world, these ethical questions must be considered in a practical sense. For example, “should we use a device or a technique that likely will not change the outcome in a patient with known survival of less than 10%?” What about 25%? 50%? Do we have data to support our conjecture? Because we have little real data to help us, the survival statistic is probably the biggest point of contention. As we will see below, there is little data, some consensus, and several proposals for addressing the critically ill patient and guiding the physician.
Mort Kern, Long Beach, California: Each patient’s clinical presentation and personal scenario must be investigated and joint decision making for the best approach — one acceptable to the patient, family, and co-physicians — comes after the operator personally speaks with the patient. This is not the situation to delegate to a fellow or resident for a decision. In 2013, my CLD editor’s page2 described just such a patient, with a late STEMI presentation complicated by a terminal illness and pneumonia. The decision to treat was the most challenging I’ve ever made, since the patient was critically ill yet fully aware, intelligently communicating, and well informed. Although we did go to the lab, we had agreed in advance of what we (the patient and care providers) would and would not do depending on the findings. After the diagnostic angiograms were obtained, percutaneous coronary intervention (PCI) was deemed to be futile. On return to the intensive care unit (ICU), the patient requested to be taken off oxygen and expired hours later.
Medical futility for the STEMI/shock patient has been addressed for those arriving after out-of-hospital cardiac arrest,3 but rarely for those arriving in shock with non-STEMI. As advocated below, more data are needed, and I believe the SCAI Interventional Consensus writing group should take up this problem after a good look at the data and then provide recommendations.
While guidelines do not specifically address our southeastern cath lab director’s query, I suggested that he review his lab’s data on each operator’s use of mechanical circulatory support (MCS)/Impella/intra-aortic balloon pump (IABP), and types of STEMI patients in whom these devices are used and their outcomes. Then, in a blinded fashion for the lab, let everyone see if there was a pattern or if there was a physician outlier. Analyze which patients are surviving with the full-court press and which are not. While this may be an impossible task, to begin to change behavior, some data is needed to persuade the ‘do-all-for-all’ operator to think differently.
Here are some thoughts from my expert cath lab colleagues on this issue.
Kirk Garratt, Wilmington, Delaware: I can’t answer all your questions, but I can tell you that a Cardiac Arrest Hospital Prognosis (CAHP) score (Figure 1) was developed from a French registry a few years ago.4 It was validated in a European population; we have a paper under review right now showing good performance of this scoring system in a U.S. population. We certainly need clarity about when to say no, since current multi-society guidelines recommend early cath for all out-of-hospital arrest (OHA) patients who achieve return of spontaneous circulation (ROSC).5 SCAI’s recent expert consensus paper6 notes the lack of randomized, controlled trials (RCTs) and underscores that we shouldn’t expect to see any in this population. The authors also note that observational studies tell us that when patients look salvageable to an interventionalist (STEMI as cause of OHA, limited down time, nothing to indicate irreversible brain injury), we take them to the cath lab, and they generally do better than the rest. The bias here isn’t a bad thing but it has to be recognized, and we shouldn’t look at observational reports and conclude that a trip to the cath lab is good for everybody. The overall mortality for OHA patients is less than 10% in most reports.
Michael Ragosta, MD, Charlottesville, Virginia: This is a great question and something we discuss regularly at our mortality and morbidity (M and M) conferences. The out-of-hospital cardiac arrest patients are probably the most common scenario where the issue of futility is often in play. I always found the obsession with “getting the artery open” in a patient who had a prolonged resuscitation and whose prognosis is clearly defined more by their brain than their heart as particularly misplaced.
There are scoring systems for the out-of-hospital arrest patients based on neuro status that are helpful. We use the CCGApH4 score in decision making at UVA and do not take all out-of-hospital arrest patients to the cath lab. I wrote an editorial for JACC Intervention7 a few years ago on this topic, based on a study they published from France.
There is not much written about futility in other scenarios in patients with other complex medical problems or illnesses such as advanced malignancies or dementia that might make intervention futile. Sound medical judgment goes a long way. We often discuss these cases with each other for second opinions. More importantly, we have a healthy M and M process, with regular review of cases regarding these issues. This process has been very helpful in giving physicians “permission” to use judgement in these hard decisions.
Richard Chazal, MD, Fort Meyers, Florida: The tension between wishing to save as many patients as possible and providing care in the setting of unlikely benefit is highly problematic for clinicians. This is particularly true in the setting of emergency care requiring an immediate decision. The opportunity to leverage second opinions, ethics committees, or shared decision making at such a time is virtually nil. Some guidance is provided for non-STEMI from the COACT trial,8 but less information is available in post-arrest patients with STEMI. In the absence of data, expert opinion/guidance from trusted sources such as SCAI and the American College of Cardiology (ACC) would be welcomed in augmenting on-the-spot clinical judgement.
Steven L. Goldberg, MD, Monterey, California: This is such a timely question. With ICU beds at a significant premium due to the pandemic, many of us are forced into triage decisions like never before. We have less ability to accommodate futile or near-futile management cases than many of us have ever had to deal with before. Guidance is not only needed, but urgently so.
Bonnie H. Weiner, MD, Worcester, Massachusetts: This is an issue that goes beyond STEMI patients. When reviewing cases from a variety of facilities, we see practices where patients with little to no expected benefit are being taken to the lab and even undergoing PCI. Very few hospitals have, as part of their internal review process, a mechanism for reviewing these cases and even fewer have a process of assessing futility prior to making the decision. We also see what appears to be arbitrary suspension of pre-existing do not resuscitate (DNR) orders with little documentation to understand the rationale.
I agree that that it is unlikely that we would see a randomized trial in these patients, even if limited to STEMI, nor do I think we should, since there are so many variables that feed into this that understanding the results would be difficult. Part of a quality culture would certainly include review of these cases and should, if possible, include cases both with and without invasive procedures in order to better understand how those decisions are being made. This would also allow for the identification of outlier physicians with practice patterns in both directions. Using any available scores, such as Kirk suggested, is at least a place to start and these can then be further refined as data are accumulated.
Paul S. Teirstein, MD, La Jolla, California: We have an ethics committee to deal with elective cases like this and it has been especially helpful for some of our questionable transcatheter aortic valve replacement (TAVR) cases. Perhaps the local ethics committee could come up with a standard…i.e., something like, if time does not permit a meeting of the ethics committee, and 2 or maybe 3 physicians agree it is futile, the procedure should not be undertaken.
Bonnie H. Weiner, MD, Worcester, Massachusetts: I wish every hospital had a working ethics committee. My observation is that this is not the case. Particularly when looking at smaller institutions, the local environment, pressures, and politics tend to avoid such discussions.
George W. Vetrovec, MD, Richmond, Virginia: This is a terribly difficult problem. No one wants to “shortchange” a possibly salvageable person and the initial information is often sketchy, and even I am sure unintentionally misleading. I have been often been surprised after being told in the ED that a person had immediate cardiopulmonary resuscitation (CPR), but 48 hours later when the patient is not waking, a visitor who witnessed the arrest says, “I’m not surprised; he (or she) was down for 25 minutes before anyone started CPR.” Then I realize the numbers are an estimate, but the first report always seems more encouraging than a later one. And if we make the rules too stringent, usually by penalizing physicians, we run the risk of creating the New York State Cardiogenic Shock scenario, where survival improved by removing reporting. So, while I agree physicians need to be judicious, these are some slippery slopes that must be avoided.
Spencer King, MD, Atlanta, Georgia: We will never be able to write a formula for this. Gene Stead said: “What this patient needs, is a Doctor.” The data we have and what more we may learn should inform decision making, with the family if possible.
Timothy D. Henry, MD, Cincinnati, Ohio: Clearly a challenging issue and the balance between less aggressive-futility and more aggressive-NO ROSC patients, for example, will always depend on local expertise and resources. From my perspective, out-of-hospital cardiac arrest (OHCA) and cardiopulmonary support (CS) programs using standardized protocols and algorithms developed by multidisciplinary teams and with comprehensive registries to carefully follow your results work best. For example, in an aggressive regional STEMI system with an OHCA program, we knew that the survival to hospital DC neurologically intact for a STEMI patient with an asystole/PEA arrest was 20% and then the committee evaluated the potential predictors/implications, including ethical, and we elected to continue.
J. Jeffrey Marshall, MD, Atlanta, Georgia: I agree with George about the OHCA patients and the variable duration of “down time”. We finally developed a protocol that required the paramedics, who ran the code, to stay in the ED until the interventional cardiologist came down to see the patient briefly in the ED. The paramedics, and more importantly, the dispatchers, are the only ones that know the exact times. We would briefly go to our communication hub and get the actual 911 call time, on-site time, departure site time, and travel times that are recorded (as 10-codes) by the county dispatcher. This ensured that we knew all the “real” times, as the ED physicians do not have the time or motivation to know these critical intervals. The one time that is difficult to truly discern is from the family — how long did they wait or was the patient down before they called 911. While this is only one aspect of a careful decision regarding futility of OHCA, I found it to be one of the most important data points to making a thoughtful clinical plan.
Lloyd W. Klein, MD, Sonoma, California: There is no formulaic definitive answer to this question. One must use judgment and be at the bedside. You must speak with the family and fully evaluate the alternatives. There are algorithms; one I helped to create has been cited by others,6 which are intended as a tool to help everyone begin to weigh pros and cons. There are no simple solutions. Nothing in this area predicts anything with 100% certainty. I know that if my own prognosis were so adverse as suggested by these algorithms, I would not want a procedure. Others may take a different approach and that is also fine. I think the main question is whether the data that went into these predictive models are still operative as our experience grows and our techniques improve.
Stephen R. Ramee, MD, New Orleans, Louisiana: Very interesting discussion. While I always agree with Spencer, it would be helpful in this information age if we could quickly pre-screen the futility patients using the data that is available on arrival to the ED. As co-author of one of the papers cited, I still struggle with how to determine quickly and ethically who to take to the cath lab and who not to take.
The hardest decision is not to take a cardiac arrest patient to the lab, and the various scoring systems4,9,10 and algorithms11 are attempts to help us with this decision. I fully agree with getting a second opinion from an interventional colleague before declining to take a patient to the lab, since the outcome is almost always bad.
Steve Bailey, MD, Shreveport, Louisiana: This question seems full on imponderables, influenced by patient, family, facility, and physician. While we have guidance from several scoring systems and society documents, they really address population outcomes, not individual patients. The suggestions about more complete data from the field and discussion in the ED is important, and an area we should investigate and hopefully apply in using these algorithms.
I believe that having a “Out of Hospital Team” approach with more than one cardiologist arriving at that decision is helpful. It will share that difficult burden of not taking the patient to the cath lab, and give solace to family and other team members that there has been due consideration. We see this in teaching hospitals already with fellows and faculty having this interaction.
Do we have the resources after hours for this? It would be interesting to compare what happens with individual versus team-based therapeutic decisions. Ultimately, it comes down to the decision of the clinician evaluating the patient, as Dr. King has stated. For now, strengthening our individual quality improvement programs, reviewing practitioners’ approach, using evidence-based care when available, and discussions with active M and M reviews in STEMI programs can assist in developing a consistent approach to OHCA.
Sam Butman, MD, Cottonwood, Arizona: Mort, over the years, most of us who have been in a position of oversight have seen the variability in care provided by colleagues across the spectrum. That any provider would have a one-size-fits-all approach cannot be considered the practice or the art of medicine. Such a provider should have any egregious procedures reviewed and discussed.
Regarding the questions you have posed, I really have little to add to the above, but what is noteworthy and unexpected is the record number of respondents. With that in evidence, you have touched on something that rises above what we have typically discussed previously. To your questions,
1. “Do you have or know of guidelines or suggestions on medical futility to help guide the physicians in the STEMI patient? Is there a scoring system?” No.
2. “For out-of-hospital cardiac arrest with an unknown time to return of spontaneous circulation entering in with STEMI, do you go directly to lab or stabilize and assess neurologic function first?” I believe there have been a series of recent publications that speak to who benefits most from early diagnostic angiography and in whom this can be deferred. Specifically, patients with clear STEMI evidence seem to benefit when compared to those patients with non-STEMI.
3. “For late-arriving STEMI in shock stage D or E, do you proceed with ‘the whole enchilada’?” I never proceed in any patient with the whole enchilada, but prefer reviewing the menu, so to speak, before proceeding.
Having read all the previous comments, the following comments, now quotes, struck me as most cogent and the underlying theme, which is obvious, is key:
“Prognosis is clearly defined more by their brain than their heart...”
“We see practices where patients with little to no expected benefit are being taken to the lab and even undergoing PCI.”
“This is a terribly difficult problem.”
“What these patient needs, is a Doctor.”
“One has to use judgment and be at the bedside.”
Lastly, the subject and the discussion by so many speaks to its importance for those on the front lines.
Carl Tommaso, MD, Chicago, Illinois: One of the issues that has always bothered me is when the ED doc leverages a cath in a futile patient because “the family wants everything done.” The ED should be reminded that “everything” should be “everything appropriate.” The ED should be encouraged to make necessary assessments and discussion with my family members to arrive at what is “appropriate” and not shift the responsibility of these discussions to us.
Bonnie H. Weiner, MD, Worcester, Massachusetts: Well said. This may explain why some feel obliged to be unrealistically aggressive, since frequently that horse has already left the barn, or at least they believe that is the case and making the more difficult decision is difficult to explain and for families to understand.
Mitchell W. Krucoff, MD, Durham, North Carolina: The length and number of these responses from my colleagues illuminate how charged this topic is with challenges and emotions in putting together intuitive clinical wisdom with an acute time clock ticking. Two comments:
1. First, each of us as an operator would do well to decide how many failed salvage procedures is worth one save. We do not highlight this inner dialogue; at least, most fellows do not hear this during training. Having population-based statistics and scores are a good probabilistic guide to social perspectives, and even professional society guidelines, but as an operator at the bedside in this setting, the only numbers that reach into your soul are 0 and 100%...and the easier out of “must have brain injury by now” is long gone with hypothermia’s impact on the certainty of what lives and what dies above the neck.
2. From comments above, it is clear we are still barely limping when it comes to collecting truly informative evidence in shock settings, especially where salvage vs futility hangs in the balance. There is now a collaborative effort across academic leaders, the FDA and other federal agencies, the American Heart Association, and device manufacturers to change this landscape of uncertainty.
The Bottom Line
How should we make good decisions and understand what is futile and what is not? Several comments hold true for every situation.
The patient needs a doctor. Knowing the patient, especially their state of mind and their wishes for their future, is always key to making good decisions. Take the time to talk with the patient and family about what is known and what is not, what you can do and what you cannot.
You cannot and should not do everything for everybody. Understand the metrics associated with the worst survival and use the best information that permits the best decision.
Finally, decisions are easy when the patient says “no”, but it’s hard for the physician and his team to say “no” to the patient and family. I want to repeat a phrase that my colleague Dr. Arnold Seto provided to us in 2013: “Only with thoughtfulness and wisdom can we resist the pressures to act rapidly.”2 We ended our discussion of futility on that particular editor’s page by saying that “when there is nothing more we can do for the patient, we must make them comfortable and keep them company until the end.” I hope this was helpful as we think about our critically ill patients.
Disclosures: Dr. Morton Kern reports he is a consultant for Abiomed, Abbott Vascular, Philips Volcano, ACIST Medical, and Opsens Inc.
Dr. Kern can be contacted at firstname.lastname@example.org.
On Twitter @drmortkern
- Angelucci PA. What is medical futility? Nursing Critical Care. 2007 Jan; 2(1); 20-21.
- Kern M. When do you decide not to take a STEMI patient to the cath lab? Addressing an ethical dilemma. Cath Lab Digest. 2013 Jan; 21(1): 4-8. Accessed February 19, 2021. Available online at https://www.cathlabdigest.com/articles/When-Do-You-Decide-Not-Take-STEMI-Patient-Cath-Lab-Addressing-Ethical-Dilemma
- Yannopoulos D, Bartos J, Raveendran G, et al. Advanced reperfusion strategies for patients with out-of-hospital cardiac arrest and refractory ventricular fibrillation (ARREST): a phase 2, single centre, open-label, randomized controlled trial. Lancet. 2020 Dec 5; 396(10265): 1807-1816. doi: 10.1016/S0140-6736(20)32338-2
- Maupain C, Bougouin W, Lamhaut L, et al. The CAHP (Cardiac Arrest Hospital Prognosis) score: a tool for risk stratification after out-of-hospital cardiac arrest. Eur Heart J. 2016 Nov 7; 37(42): 3222-3228. doi: 10.1093/eurheartj/ehv556
- Yannopoulos D, Bartos JA, Aufderheide TP, et al; American Heart Association Emergency Cardiovascular Care Committee. The evolving role of the cardiac catheterization laboratory in the management of patients with out-of-hospital cardiac arrest: a scientific statement from the American Heart Association. Circulation. 2019 Mar 19; 139(12): e530-e552. doi: 10.1161/CIR.0000000000000630
- Lotfi A, Klein LW, Hira RS, et al. SCAI expert consensus statement on out of hospital cardiac arrest. Catheter Cardiovasc Interv. 2020 Oct 1; 96(4): 844-861. doi: 10.1002/ccd.28990
- Ragosta M. The heart or the brain?: Which takes priority after cardiac arrest and can we identify patients in whom aggressive cardiac care is futile? JACC Cardiovasc Interv. 2018 Feb 12; 11(3): 257-259. doi: 10.1016/j.jcin.2017.10.009
- Lemkes JS, Janssens GN, van der Hoeven NW, et al. Coronary angiography after cardiac arrest without ST-segment elevation. N Engl J Med. 2019 Apr 11; 380(15): 1397-1407. doi: 10.1056/NEJMoa1816897
- Kiehl EL, Parker AM, Matar RM, et al. C-GRApH: a validated scoring system for early stratification of neurologic outcome after out-of-hospital cardiac arrest treated with targeted temperature management. J Am Heart Assoc. 2017; 6: e003821.
- Bougouin W, Dumas F, Karam N, et al. Should we perform an immediate coronary angiogram in all patients after cardiac arrest? Insights from a large French registry. JACC Cardiovasc Interv. 2018 Feb 12; 11(3): 249-256.
- Rab T, Kern KB, Tamis-Holland JE, Henry TD, et al; Interventional Council, American College of Cardiology. Cardiac arrest: a treatment algorithm for emergent invasive cardiac procedures in the resuscitated comatose patient. J Am Coll Cardiol. 2015 Jul 7; 66(1): 62-73. doi: 10.1016/j.jacc.2015.05.009