In “Will We Code for Default ECMO?” Brauner and Zimmerman draw parallels between the history of cardiopulmonary resuscitation (CPR) and current developments in extracorporeal membrane oxygenation (ECMO). They fear that, as occurred with CPR, indications for ECMO will expand until cardiac arrest becomes a “blanket indication” for ECMO as an adjunct to CPR. If ECMO becomes a default treatment for patients experiencing cardiac arrest, patients and surrogates will likely need a mechanism to opt out of this default. As Klugman, a clinical ethicist, recently blogged: “Is It Time for the DNE: Do Not ECMO?”1 This question has also been raised in the bioethics and critical care literature.2,3 We agree with Brauner and Zimmerman that the best course of action would be to prevent ECMO from becoming a default treatment.
However, we should also consider how to proceed if ECMO does, in fact, become part of the default treatment for cardiac arrest. Such considerations include implementation challenges that would likely arise if do-not-ECMO (DNE) orders were to be incorporated into hospital code status systems. Specifically, we are concerned with implementation challenges related to the scope of DNE orders. We can gain insight into these challenges by comparing DNE orders with do-not-resuscitate (DNR) orders, which have faced scope-related implementation challenges since their adoption in the 1970s.4,5 DNR orders allow patients and surrogates to refuse CPR that would otherwise be provided by default.4 DNE orders could function similarly by allowing patients and surrogates to refuse ECMO that would otherwise be provided by default. By examining the scope-related implementation challenges associated with DNR orders, we can predict some of the challenges likely to arise when incorporating DNE orders into hospital code status systems.
First, clinicians sometimes erroneously infer patient preferences for treatments outside of cardiac arrest on the basis of a DNR order.5,6,7 For example, a clinician might assume that a patient with a DNR order would not want other life-sustaining interventions, such as dialysis. As Yuen et al explain, these erroneous inferences “may be due to misunderstanding the scope of DNR orders [italics added].”5 Despite decades of efforts to clearly define the scope of DNR orders in national guidelines,8,9 DNR orders have continued to shape clinical management decisions for treatments other than CPR.6,7 To prevent clinicians from misinterpreting DNR orders, some hospitals have implemented broadened DNR orders that explicitly communicate patient preferences for treatments other than CPR.10,11,12,13,14 However, there is limited data on whether this strategy is effective.10,11 We have little reason to believe that DNE orders will not also be subject to misinterpretation; clinicians may assume that patients with DNE orders do not want other life-sustaining interventions.
Second, the scope of DNR orders is unclear because many of the components of CPR, such as intubation and mechanical ventilation or intermittent mandatory ventilation (IMV), can be indicated in other contexts.15,16 For example, a patient who refuses CPR in the event of a cardiac arrest (and thus refuses IMV in this context) could want IMV for chronic obstructive pulmonary disease exacerbations. This contextual variation creates challenges in understanding the scope of DNR orders. For example, does a DNR order imply a do-not-intubate order and, if so, in what clinical circumstances? Or does a DNR order preclude intubation entirely? Some organizations and clinicians have navigated these questions by implementing “partial” code orders, although these are controversial.17,18 Similar to IMV, ECMO can be a component of CPR but can also be indicated in other contexts. Thus, ECMO would likely be subject to similar questions: Should a DNR order be interpreted as implying a DNE order and, if so, in what clinical circumstances? Or should a DNR order preclude ECMO entirely?
To address these questions, clinicians and bioethicists should proactively consider how to limit the scope of DNE orders before ECMO emerges as a default treatment for patients experiencing cardiac arrest. In particular, code status systems that incorporate DNE orders should prevent physicians from acting on erroneous inferences about patient preferences and should clearly define the conceptual and practical relationships between DNR and DNE orders.
Klugman C. Is it time for the DNE: do not ECMO? Bioethics.net. May 1, 2019. https://www.bioethics.net/2019/05/is-it-time-for-the-dne-do-not-ecmo. Accessed June 13, 2019.
Brodie D, Curtis JR, Vincent JL, et al; participants in the Round Table Conference. Treatment limitations in the era of ECMO. Lancet Respir Med. 2017;5(10):769-770.
Lantos JD, Meadow WL. Should the “slow code” be resuscitated? Am J Bioeth. 2011;11(11):8-12.
- Rosoff PM, Schneiderman LJ. Irrational exuberance: cardiopulmonary resuscitation as fetish. Am J Bioeth. 2017;17(2):26-34.
- Yuen JK, Reid MC, Fetters MD. Hospital do-not-resuscitate orders: why they have failed and how to fix them. J Gen Intern Med. 2011;26(7):791-797.
- Beach MC, Morrison RS. The effect of do-not-resuscitate orders on physician decision-making. J Am Geriatr Soc. 2002;50(12):2057-2061.
- Stevenson EK, Mehter HM, Walkey AJ, Wiener RS. Association between do not resuscitate/do not intubate status and resident physician decision-making. A national survey. Ann Am Thorac Soc. 2017;14(4):536-542.
- Mancini ME, Diekema DS, Hoadley TA, et al. Part 3: ethical issues: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2015;132(18)(suppl 2):S383-S396.
- Council on Ethical and Judicial Affairs, American Medical Association. Guidelines for the appropriate use of do-not-resuscitate orders. JAMA. 1991;265(14):1868-1871.
Chen YY, Gordon NH, Connors AF, Garland A, Chang SC, Youngner SJ. Two distinct do-not-resuscitate protocols leaving less to the imagination: an observational study using propensity score matching. BMC Med. 2014;12(1):146.
- Davila F, Boisaubin EV, Sears DA. Patient care categories: an approach to do-not-resuscitate decisions in a public teaching hospital. Crit Care Med. 1986;14(12):1066-1067.
- Uhlmann RF, Cassel CK, McDonald WJ. Some treatment-withholding implications of no-code orders in an academic hospital. Crit Care Med. 1984;12(10):879-881.
- Ells C. Levels of intervention: communicating with more precision about planned use of critical interventions. Am J Bioeth. 2010;10(1):78-79.
- Vanpee D, Swine C. Scale of levels of care versus DNR orders. J Med Ethics. 2004;30(4):351-352.
- Jesus JE, Allen MB, Michael GE, et al. Preferences for resuscitation and intubation among patients with do-not-resuscitate/do-not-intubate orders. Mayo Clin Proc. 2013;88(7):658-665.
- Breu AC, Herzig SJ. Differentiating DNI from DNR: combating code status conflation. J Hosp Med. 2014;9(10):669-670.
- Berger JT. Ethical challenges of partial do-not-resuscitate (DNR) orders: placing DNR orders in the context of a life-threatening conditions care plan. Arch Intern Med. 2003;163(19):2270-2275.
Sanders A, Schepp M, Baird M. Partial do-not-resuscitate orders: a hazard to patient safety and clinical outcomes? Crit Care Med. 2011;39(1):14-18.