Having to decide whether to attempt resuscitation when there is incomplete or conflicting evidence of the patient's wishes is challenging, but it is distressingly common in the emergency department and all too frequent in inpatient settings.
Many if not most of these dilemmas involve shortcomings in preventive ethics, reflecting either inadequate communication with the patient, family member or other appropriate surrogate, or inadequate or unavailable documentation of that communication.
In this case, a literal interpretation of the available documentation would allow the physician only to intubate the patient, which would be pointless without accompanying CPR, which the written order in the medical record explicitly forbids. There is no question, however, that the right thing to do (while urgently continuing to try to reach the primary care physician and/or daughter) is to institute CPR.
Additional clarifying discussion could confirm that the elderly patient would not have wanted resuscitation to be attempted. But the phrase “may intubate” raises at least a reasonable doubt in the physician's mind about the patient's wishes, especially since it is possible that the acute event that precipitated the patient's cardiopulmonary arrest is rapidly reversible with a potential return to her baseline condition.
Resuscitating a patient inappropriately is an error that is usually at least partially reversible. If additional information makes it clear that the patient would have preferred to be allowed to die, she can be promptly extubated and provided with whatever medication is necessary to ensure comfort. In contrast, allowing a patient to die when resuscitation would have been both wanted and plausibly successful is a grave, irreversible error.
From a preventive ethics perspective, the attending hospitalist should never have had to face this ethical dilemma. It is, however, too easy to say that entering a DNR order but giving permission to intubate is contradictory. For any clinical intervention, the central patient-centered question is, “Do the expected benefits outweigh the expected burdens from the patient's perspective?”
In the initial stage of serious pneumonia with progressive respiratory compromise, short-term intubation while waiting for antibiotics to take effect may have enormous benefit with acceptable burdens, but if the illness progresses to the point of full cardiopulmonary arrest, the benefit/burden ratio of CPR with intubation and subsequent ICU care may be far less favorable. At a minimum, information about the reasoning process should have been provided.
For example, if the PCP's note had explained that “for life-threatening events that may be rapidly reversible, the patient would be willing to undergo short-term aggressive life-saving measures …,” the hospitalist likely would not have agonized at all about the right thing to do.
The patient was successfully resuscitated and transferred to the ICU. When reached, the daughter's first question was, “But why did you do CPR?” She explained that she had been very conflicted about her mother's code status. She knew that her mother would never have wanted prolonged ICU life support, or to live permanently in a nursing home or as “a vegetable.”
However, she also knew that her mother “savored even small sweetnesses in life and would want a chance to live if she could get back home.” She said that she had never meant to suggest that short-term, life-saving treatment would be withheld. She agreed that her mother should be “full code” for the next few days while her prognosis became clearer. Palliative care was consulted to explore future treatment options.
The patient was successfully extubated several days later, and over the ensuing two weeks it became clear that, while she was unlikely to regain her prior level of semi-independent functioning, with adequate support she might still enjoy weeks or months of time at home.
One month after her cardiac arrest she was discharged home with hospice, with a clear DNR order and no plans to re-hospitalize.