Thanks to the many of you who shared your thoughts on last week’s ethical dilemma. Your sophistication and focus on the needs of the patient and family inspired me.
As a reminder, we were wrestling with how to think about and care for a patient on five vasoactive drugs, who is about to code after several hours of severe hypotension. The key questions were:
- If we withhold chest compressions when the patient loses her pulse, should her death be classified as “full code” or “DNR?”
- What role, if any, should the family play in the decision to withhold chest compressions?
- If we decide to withhold chest compressions, should we enter a DNR order?
Question #1: You may be surprised to learn, as I was twenty years ago, that my colleagues in San Francisco would classify this death as “full code” (or “full ICU care” as they called it), even if chest compressions were withheld. The rationale was that persistent treatment with multiple vasopressors constituted CPR, even without chest compressions. The decision to withhold compressions didn’t constitute a treatment limitation, given that pushing on the chest couldn’t save the patient once vasopressors had failed. In contrast, the investigators labeled deaths as “DNR” (i.e., “withholding of resuscitation”) only if resuscitation wasn’t occurring yet (e.g., if a stable patient went into v fib). What struck me then and now is how our arbitrary use of terminology drives the way we think about resuscitation- as well as how we present treatment options to families. If our notion of CPR includes treating persistent hypotension with vasopressors, then there is no need to discuss adding chest compressions when patients lose their pulse, just as we would feel no obligation ask families if we should push amiodarone or turn up the joules on the defibrillator. In contrast, if we equate withholding chest compressions with forgoing CPR, then we may feel obligated to negotiate this decision with the family. Even now, I don’t think we’ve settled this question, but I do believe we should acknowledge that we have endowed treatment with chest compressions with a moral weight and symbolic meaning that greatly outstrip its clinical effectiveness.
Question #2: All letter writers opined that the MICU team should discuss decision making with the family, even if the actual decisions were left to the physicians. Here’s what they said:
- An attending said the family should be told why it didn’t make sense to do chest compressions.
- A resident said the family should be told that the patient was declining despite maximal interventions, that the patients’ body had “made the decision,” and that the team “would not recommend chest compressions,” implying that the ultimate say on whether to do chest compressions should remain with the family.
- Another resident stated that chest compressions would provide minimal benefit to the patient and could cause harm, such as fractures and prolonging the time to an inevitable death. This writer believed the family should be told that further interventions would be futile and that chest compressions should not be offered.
- One resident cited a similar case in the MICU, where the patient was coded six times despite running 5 pressors to maintain a systolic blood pressure of 40. The resident felt that doing so respected the families wish to know that “the doctors did everything they could.” Similar views were expressed in a provocative NEJM perspective several years ago.
- Another resident described a similar scenario of a patient on five pressors. The family was told that the patient would not be resuscitated again after being coded previously. The writer felt that the decision protected the family from having to feel responsible for “withholding care” from their loved one. The writer believed the family felt grateful and understood that the team had done its best. However, she also related an experience with another patient on five pressors whose family was adamant about maintaining a “full code” status, after which the patient was coded for 45 minutes. In this case, the family may have benefited from seeing the code, which, the writer noted, was consistent with an NEJM article that suggested that families have better psychological outcomes after witnessing codes. Still, the writer felt more comfortable with her role, and with the family’s response, in the first case than in the second.
- One resident noted that the important question should not be how likely it is that the patient would regain circulation on five pressors but, rather, what the goals should be, which she expressed as “goals of life” (a lovely term that I had not considered before). She felt that once goals were articulated, the medical team could align these goals with care decisions. This resident would not offer or even mention resuscitation unless it became clear that the patient’s goal was to “be alive no matter what.” Otherwise, she felt resuscitation would just cause suffering.
- Another resident noted that this patient was already receiving ACLS for irreversible vasodilatory shock. She felt chest compressions and possible shocks would be futile and, therefore, should not be provided. She felt that physicians should play a more directive role when making technical decisions, for example in the use of pressors. She also shared the story of a case where the team struggled with end-of-life decision making with a family that was demanding aggressive medical care. The patient subsequently deteriorated and was sent to the MICU, where morphine was quickly given, ending the patient’s suffering. The writer was struck by the dramatic difference in approach taken by the MICU and the Stepdown team.
- One resident questioned whether it was fair or appropriate to involve families in code status discussions when the prognosis is so poor. However, she noted that some families may welcome the opportunity to participate in decision making and advocate for their loved one. This resident appropriately noted that different families have a wide range of needs and preferences when it comes to decision making.
- Finally, a resident noted that this discussion reminded him of a superb Medical Grand Rounds given by one of our former Chief Residents, Sam Gelfand, who explored the history of CPR and its evolution from an innovative treatment for cardiac arrest to a rite of passage for countless deaths, particularly in the US. This resident felt that the patient should not receive chest compressions, nor should the family be burdened with having to make this decision. He felt we should treat CPR decisions the way surgeons decide to withhold surgery when it isn’t indicated, or the way nephrologists decide not to offer dialysis when it’s inappropriate. He said he would communicate with the family and seek their thoughts, all the while updating them on the patient’s approaching death despite the team’s efforts. He would give the family ample opportunity ask questions, let them cry if they needed to, or remain silent if that was what the situation called for. He would stay by their side no matter what. If the family asked him to “do more,” he would respond with “I wish” statements, such as “I wish we could do more” or “I wish medicine was stronger” or “I wish this wasn’t happening.”
Question #3: This question was dealt with technically. One resident felt that there was no need to write a “DNR” if death were imminent but that it would be advisable to enter an order if coverage had to be passed off to another team. He noted that continuity in the ICU made it easier to communicate between teams, even without bracelets and orders.
So what happened in our case? As expected, the patient lost her pulse after several hours of hypotension despite five vasopressors. Based on prior interactions with the family, we felt we did not need not push a decision to stop the vasopressors or extubate the patient since death was approaching rapidly. The family was made aware that we would not perform chest compressions and we entered a DNR at the end of the day shift so there would be no chest compressions when the patient lost her pulse that night. This communication ensured that the night nurses and physicians would be able to implement our plan without confusion. The family was comfortable with the team’s approach and grateful for the care provided.
So in the end, I believe we agree that we should always discuss code status with families, just as we discuss all treatment plans. However, we should recognize that some of the treatments we discuss may be ineffective, and even harmful, and we must always share our opinions honestly and thoughtfully. We should also recognize that the medical community has endowed certain ineffective treatments with a symbolic significance, which our practices only serve to perpetuate. While I don’t think we’ve fully resolved the question of how to classify these deaths, or how much decision making responsibility should rest with the family, I do believe we have achieved consensus that commitment to good care—and to the wellbeing of our patients and families—can take many forms. When thoughtfulness and compassion drive our response to difficult questions, we will inevitably reach the right answer.
A special thanks to Michael Bennick, Brienne Miner, Brian Persaud, Fahad Faruqi, Ethan Bernstein, Sandra Williams, Zehra Omer, Lindsay Vendetta, Mike Breen, and Jadry Gruen for sharing their wisdom.
Enjoy your Sunday, everyone,
Signs of Spring from my front lawn!