The “Soft” Code

EMS notification. 115 year old female in cardiac arrest. ACLS in progress. Estimated time of arrival: 5 minutes. Your stomach groans.

Upon EMS arrival, you learn that she comes from a nursing facility and has end-stage dementia; she is bed-bound, contracted and non-verbal. Maybe she’s got a PEG tube or a trach. This is going to be grotesque. Painful for you, for the family, and most importantly, for the patient. You press EMS for advance directives. They shrug.   Meanwhile, osteoporotic rib-crunching CPR is underway. Let’s say, for the sake of argument, you get family on the phone: they tell you “she’s a fighter, she wanted to be resuscitated.”

So the teams runs ACLS. Maybe less feverishly than normal, maybe with fewer cycles, maybe with gentler compressions. A “soft code,” it’s called.

But no matter how “soft,” this resuscitation is nauseating.  Who is driving this out-of-control resuscitation train? Why is CPR the de-facto treatment for cardiac arrest? Isn’t this a treatment like any other? Doesn’t it have risks and benefits? Indications and contraindications?


Nowadays we talk about patient-centered care and autonomy. How the pendulum has swung from the olden days of beneficent docs practicing enlightened paternalism. Autonomy has its perks. It empowers patients and families. It absolves us in those iffy situations where maybe, just maybe, an admission, procedure or further testing is indicated, but may come with risks and pitfalls. We hear the language of autonomy everywhere.   In the recently published and contentious ACEP guidelines for the management of acute ischemic stroke, the panel suggests that tPA should be “offered” to those meeting NINDS inclusion/exclusion criteria (strength of recommendation: level 1A. No comment).

But autonomy doesn’t just happen. Autonomy depends on an informed decision maker. And so it falls on us, the medical professionals, to be the informers. It is a sacred responsibility; all autonomy depends on it. Conversely, if we allow patients and/or their surrogates to make uninformed decisions by withholding our knowledge and experience, we have undermined their autonomy.[1]

How do patients and families feel regarding physician influence on end-of-life decision-making? A prospective, cohort of seriously ill hospitalized patients were surveyed on how they wanted to come to their treatment decisions. 16% wanted to make their choices alone; the rest preferred varied degrees of physician-involvement.[2]

The weight and urgency of decision-making during end-of-life care can be overwhelming for anyone, even for those of us only remotely engaged with the course of illness. For families, end-of-life decision-making can be complicated and obscured by guilt, denial, misinformation, among other confounders. Into this un-informed and emotionally high-stakes milieu, we say, “what would your loved one want?” or, even worse, “would they want ‘everything’ done?”And we hope for a rational, outcomes-based decision.

Looking again at the example of tPA: arguably, the therapy with the worst risk profile of anything we can give. We are so cautious in trying to select the correct patient for its application because of the risks and shaky evidence-based footing. Do no harm, right? So, we insulate our patients (and ourselves) from the dangers of tPA with rigid indications and contraindications. Why should CPR be any different?

CPR has risks and benefits, too. And we need to better clarify its indications and contraindications.   CPR should not be the de-facto treatment for cardiac arrest; it should be applied to the correct patient after an assessment of the pre-morbid state, circumstances of arrest, and likelihood of good outcome.


Blinderman et al., in JAMA 2012, describe a rational approach to determining the appropriateness of CPR in cardiac arrest. They divided patients into 3 categories.[3]

Patients for whom CPR is a plausible option. The relative risks and benefits of CPR in these patients are unclear. They may have chronic diseases but have not yet reached the terminus of their disease state. Care should be taken to inform patients and/or surrogates about what is known about outcomes after CPR/ACLS and about the likelihood of harm caused by resuscitation.

Patients for whom CPR is not recommended. These patients have advanced, terminal diseases and are unlikely to benefit from CPR. The harms of CPR outweigh the benefits, and, in the event of a “successful” resuscitation are likely to succumb to their chronic disease shortly thereafter or spend their remaining time on life support.

Patients in whom CPR should not “offered.” These patients are actively or imminently dying and stand no chance at a meaningful recovery. The likelihood of harm greatly outweighs that of benefits. The authors assert that in this patient population, withholding CPR is not only medically sensible, but is also ethical.


The concept of medical futility is absent from these categorizations. “Undue harm and suffering” has taken its place. A seismic shift in semantics. But it does seem easier to tell families that a full-on resuscitation will likely cause “undue suffering” rather than imply that their loved one’s care is “futile,” which is certainly not what we mean.

It’s easier to pontificate on an abstract and extreme case than in the resus bay. In real-time and in the absence of clarified advance directives, the practice of resuscitating those who are unlikely to benefit from CPR and may be harmed will continue. The “soft code” will go on. The low-hanging fruit here is how we approach end-of-life decision-making and the conversations we have patients and families.




  1. Billings J, Krakauer EL. On Patient Autonomy and Physician Responsibility in End-of-Life Care. Arch Intern Med. 2011;171(9):849-853.
  2. Heyland DK, Tranmer J, O’Callaghan CJ, Gafni A. The seriously ill hospitalized patient: preferred role in end-of-life decision making? J Crit Care 2003;18:3-10
  3. Blinderman CD, Krakauer EL, Solomon MZ. Time to revise the approach to determining cardiopulmonary resuscitation status. JAMA: The Journal of the American Medical Association. 2012 Mar. 7;307(9):917–918.
  4. Lamas D, Rosenbaum L. Freedom from the tyranny of choice–teaching the end-of-life conversation. N Engl J Med. 2012 May 3;366(18):1655–1657.
  5. Anderson D. It’s Ethically, Morally, and Legally OK to not do CPR (sometimes.) The Expensive Care Blog. April 29, 2013.
The following two tabs change content below.

Jay Khadpe MD

  • Editor in Chief of "The Original Kings of County"
  • Assistant Professor of Emergency Medicine
  • Assistant Residency Director
  • SUNY Downstate / Kings County Hospital

Latest posts by Jay Khadpe MD (see all)

3 comments for “The “Soft” Code

  1. Mark Silverberg
    June 23, 2014 at 6:50 pm

    Interesting post Joey. Most of us know who will and who will not do well with CPR, yet we do not voice that opinion in a code situation for one reason. That reason is we are afraid of being sued for “not doing everything” for the patient that has just died in front of us. I honestly think that very few physicians use their medical knowledge to decide who will undergo ACLS. Physicians who practice are just too scared that someone will find out they did not “try” and save Grandma’s life and they are afraid of litigation.

  2. Nathan
    June 23, 2014 at 7:12 pm

    I like the change in terminology. Families are much less willing to agree to terminate futile efforts if you ask them “should we do EVERYTHING” and pretend they have some sort of clue about what that actually means.

    If you take the time to explain that resuscitation is more likely to cause harm and suffering than to extend life meaningfully the families generally are more willing to allow their family member to die peacefully. Semantics matter a lot especially in these cases which are always more emotional than rational.

  3. atan
    June 24, 2014 at 11:00 pm

    Great discussion Dr. Freedman. I would really love to hear what other people’s thoughts on this subject are.

    In essence what Dr. Blinderman proposes is swinging the pendulum away from the notion of “patient autonomy” and back towards paternalistic medicine when it comes to providing CPR. The problem with this when it comes to making these decisions in the ED is three-fold.

    – First is the one that Dr. Silverberg is referring to. In today’s litigious world, it is “safer”, for lack of a better term, for doctors to do more rather than less. As much as we attempt to avoid this, the underlying fear of legal repercussion often affects our decisions in the ED.

    – The second problem is the lack of information available to us regarding our patients. We are meeting our patients for the first time and gathering information as we stabilize, resuscitate, diagnose, treat. Medical records are scarce. Patient and family knowledge is often limited. It is very difficult to be making these critical decisions with limited data.

    – And finally, there is the lack of good data available on accurate prognostication. The 115 year old. Yes, it is not so hard. The 67 yo septic patient with diabetes who codes from septic shock. It’s not so clear how they will do following resuscitation.

    This is when I start advocating for ED docs to foster and develop their skills in palliative care. It is a much “easier” position for a physician if the family now instead tells you that their grandmother is a strong lady who lived a good life and who felt her independence was sacred. She had seen Uncle Bob spend his last days in the ICU and had mentioned repeatedly that she would never have wanted to die that way. You as the physician now have a glimpse into this lady’s values. And as you inquire more with such questions as “based on what you are telling me, it sounds to me that she would not have wanted to be artificially sustained on machines and to be kept in an intensive care unit, is that correct?,” you are absolved of that decision. The family has made it for you. This conversation ideally happens long before her arrival to the ED at a clinic visit with her PMD. But more often than not, it may happen in the ED before the patient codes.

    Don’t get me wrong. Every decision we make in the Emergency Department is based on both medical appropriateness and patient goals of care. If a patient with stage IV lung CA codes from respiratory failure, you must question the appropriateness of coding and resuscitating this patient. But if the decision regarding resuscitation is made before they die, you can avoid the code, soft or not, all together.

Leave a Reply

Your email address will not be published.