DNR: Not the Total Picture

Introduction Although Do Not Resuscitate (DNR) designates that in the event of a cardiopulmonary arrest there will be no cardiopulmonary resuscitation, the designation of DNR is often extrapolated and inferred as a less aggressive care plan. Although more than ten years have passed since the American Medical Association (AMA) updated position statement emphasizing that DNR orders “should not influence other therapeutic interventions that may be appropriate,”1 the exercise in extrapolating care plans from DNR continues to occur in the course of our daily patient encounters. Do Not Resuscitate is a designation requested by the patient or the patient’s proxy. As a designation, it is a static declaration to forgo cardiopulmonary resuscitation (CPR) in the event of a cardiopulmonary arrest. The degree of diagnostic workup and intensity of treatment desired by the patient should not be inferred from the DNR designation. A more meaningful and profound discussion with the patient or the patient’s proxy is needed to come to an understanding of the expectations for end-of-life medical care. Following is the case of an elderly woman with a DNR order, in a not uncommon situation. Case Presentation A 92-year-old Japanese woman residing in a nursing home developed a large amount of hematemesis with hypotension, 94/57, in the middle of the night when a covering physician was on duty. She had just been transferred to the nursing home from acute care the day prior to experiencing hematemesis. Her medical history was significant for moderately differentiated adenocarcinoma of the colon, for which she had just undergone open transverse colectomy of the proximal transverse colon, with placement of a gastrostomy tube for progressive dysphagia associated with profound dementia. The patient was sent to the emergency room for further evaluation, management, and consideration for readmission to acute care. In the emergency room, she was given intravenous fluids of approximately 300 cc, with her blood pressure rising to 116/50. She was also noted to have a hemoglobin of 11.2, amylase of 206, and lipase of 250. Her nursing home transfer information indicated Do Not Resuscitate, and with the presumption that she also wanted total comfort care, she was sent back to the nursing home with a diagnosis of hematemesis and pancreatitis. At the nursing home, she experienced continued bleeding and abdominal pain, and within five hours of returning there the patient was admitted to acute care for further workup and intervention. The Gray Area The not uncommon case described above illustrates how a designation of DNR can also be interpreted to imply a desire for less aggressive intervention, or sometimes equated with “total comfort care” in spite of lacking other documentation to this effect. The patient’s blood pressure was temporarily stabilized with fluids, and influenced by her DNR status, the emergency room physician sent her back to the nursing home without any further diagnostic evaluation or intervention. In today’s medical world, many different physicians take care of a given patient over different time periods and at different institutions. Emergency, hospital, outpatient, and long-term care physicians are often involved at one time or another in a patient’s care. The situation is further complicated by a newly and often transiently formed physician-patient relationship in which there is not an overall understanding of a patient’s wishes that comes with years of nurturing a venerable physician-patient relationship. Although it may seem natural that the physician who has established the longest physician-patient relationship should take the lead in DNR and end-of-life care discussions, such is not always the case. Even with a DNR designation, DNR in and of itself is not sufficient to communicate health care directives and care plans. In taking the liberty of carving out care plans regarding interventions based solely on DNR status, the risk of rendering treatment that may be incongruent with a patient’s wishes abounds. The History of Do Not Resuscitate Prior to the mid-1970s, when DNR orders first began to appear in the medical literature, CPR was universally applied to those who experienced a cardiopulmonary arrest.2 Developed in the 1960s, CPR was designed to restore basic circulation and breathing for anesthesia-induced arrests.2,3 Universally applied, however, ethical questions arose regarding unintentionally prolonging the agony of terminally ill patients.2,4 Formal DNR orders written in the medical records soon became acceptable in the 1970s with the support of the AMA. The AMA affirmed that “CPR is not indicated in certain situations, such as in cases of terminal irreversible illness where death is not unexpected.”2,5 Today, it is not uncommon to see DNR orders. The Meaning of Do Not Resuscitate Do Not Resuscitate in its pure form refers to forgoing the technical procedure of CPR in the event of a cardiopulmonary arrest. The President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, in 1983, stated that “any DNR policy should ensure that the order not to resuscitate has no implications for any other treatment decisions.”6 The AMA’s Council on Ethical and Judicial Affairs published its updated position on DNR orders in 1991 in The Journal of the American Medical Association. The article reemphasized that “DNR orders only preclude resuscitative efforts and should not influence other therapeutic interventions that may be appropriate.”1 In addition, other professional organizations such as the American College of Physicians, National Hospice Organization, American Thoracic Society, and Society of Critical Care Medicine, have formal statements on DNR which affirm that a DNR order does not imply declining of other therapeutic interventions that may be appropriate.2 The American Nurses Association’s position statement on DNR states: “There should be no implied or actual withdrawal of other types of care for patients with DNR orders.”7 Thus, barring the actual cardiopulmonary arrest, a DNR order provides no directive on diagnostic and treatment issues experienced prior to an actual cardiopulmonary arrest. As in the previous case, an elderly bedridden nursing home resident with dementia and a recent history of colon resection may still have wanted diagnostic and therapeutic interventions in conditions such as hematemesis with associated hypotension, regardless of the DNR designation. In one study of nursing home residents’ preferences for life-sustaining treatments, it was found that in the residents capable of decision making, the majority chose life-sustaining treatments and specifically indicated a preference to be hospitalized in the event of a serious illness.8 The Unsolicited Exercise in Extrapolating from DNR In the nursing home, acute care hospital, or rehabilitation center, many elderly patients often ride on a continuum of ever increasing frailty. Clinical situations often require decision making with respect to the degree of aggressiveness regarding diagnostic studies, interventions, and treatments. Although a DNR status does not explicitly imply a desire of refraining from further diagnostic or therapeutic interventions, there has been confusion over this issue.9,10 Part of the problem may be related to the current lack of documentation of health care directives or care plans that accompanies a DNR order. Health care directives and care plans should generally reflect the reason behind a DNR order. A study by Uhlmann et al10 showed that physicians writing DNR orders intended to also withhold other interventions. However, 43% of the patients’ charts lacked documentation as to the complete care plan.10 Another part of the problem may be related to the historical development of DNR. When DNR orders first arrived on the scene in the 1970s, the reason behind the order was to withhold CPR in cases such as those with “terminal irreversible illnesses where death is not unexpected or where prolonged cardiac arrest dictates the futility of resuscitation efforts.”5 In these cases, either death was expected or resuscitative efforts were felt to be futile. With the arrival of DNR on the scene, some hospitals developed “policy statements regarding the care of patients with DNR orders.”11 One hospital developed a four-tiered patient classification system for critically ill patients with respect to “therapeutic effort.”11,12 DNR patients fell into the last two classes and were deemed inappropriate candidates for the ICU.11,12 Today, the reasons for a DNR order are varied and may not be related to any terminal illness. Some patients may request it on a philosophical or religious basis, feeling that “if the time has come, it is time to go naturally.” They may be in excellent health but request DNR with the understanding that a DNR order has no implication for lessening other diagnostic or therapeutic interventions. In a study by Beach and Morrison,13 physicians were found to be less likely to initiate diagnostic and intervention procedures such as blood cultures and blood transfusions in those patients with DNR orders. The methodology used in this study was a survey of hypothetical identical case scenarios of patients with and without a DNR order.13 In another study by Shepardson et al,14 increased risk of death in patients with DNR orders was found. In this retrospective study, to decrease selection bias, stroke patients were given propensity scores that reflected the likelihood of a DNR order.14 After adjustment for these scores and other variables, it was found that having a DNR order increased the risk of death.14 The question was raised whether the result of the study to some extent may reflect the possibility of “omissions of care (eg, withholding of indicated treatment).”14 Also raised was the possibility that “practitioners may monitor patients with DNR orders less aggressively, respond less promptly to changes in patients’ conditions, or pay less attention to patients’ needs for nutrition and other supportive measures.”14 Wenger et al15 in their retrospective study found that mortality was higher in hospitalized patients with a DNR order, even after adjustments for illness and patient and hospital characteristics. One possible explanation was that “patients with DNR orders may receive different care than patients without DNR orders.”15 Comfort care, which may be appropriate in certain patients with DNR, may be substituted for a more aggressive care plan in all patients with a designated DNR status. If the intensity or quality of care is diminished because of a DNR order, then writing a DNR order would have important ramifications.15 Do Not Resuscitate in and of itself should not allude to a care plan that is less aggressive in monitoring, caring, or intervening. Tomlinson and Brody, in an article in The New England Journal of Medicine, described the issue aptly: “The ambiguity arises from the existence of two meanings–the explicit ‘no CPR’ and the often inferred ‘no extraordinary measures.’”16 Clarification: DNR and End-of-Life Care Plans or Healthcare Directives May Have Common Interwoven Threads, But Are Not the Same Do Not Resuscitate refers to forgoing the technical procedure of cardiopulmonary resuscitation. End-of-life care plans or health care directives refer to a more customized care plan tailored to the individual in consideration of his or her unique personhood and his or her current health status. End-of-life care plans and health care directives often include, among the many documents available, living wills, durable health care proxies, and Commission on Aging with Dignity’s “Five Wishes” form.17 However, it is difficult to capture a patient’s wishes a priori in their entirety that would cover and apply to all the variable and unexpected circumstances in the clinical setting. Care plans and directives can often be fluid, changing with different circumstances. As long as the patient retains decision-making capacity, care plans may be redefined by the patient. A highly functional, healthy, elderly resident may request DNR and at the same time expect fully aggressive diagnostic studies and treatment as the need arises. Similarly, a wheelchair-bound elderly person with Parkinson’s disease and dysphagia may request DNR but at the same time fully expect any diagnostic studies and treatment as the need arises. An elderly resident with atrial fibrillation may request CPR as long as he is ambulatory but customize his care plans to forgo any chemotherapy for his colon cancer. There may be situations where the choice to request CPR is seemingly incongruent to the end-of-life care plans requested by the patient. An example is an elderly patient with endometrial cancer and progressive bleeding who has declined any intervention whatsoever but requests CPR. In attempting to integrate DNR with other treatment decisions, order forms such as the “Physician Orders for Life-Sustaining Treatment” (POLST) have been used in long-term care to communicate preferences during transfers from other facilities.18, 19 In a recent article, Berger20 discussed a treatment guide for life-threatening conditions in adult patients with DNR orders. This guide would support a care plan to be used in life-threatening conditions.20 As such, DNR orders integrated with a care plan in one document have been proposed, used, and discussed but have yet to be universally applied, studied, and tested. Summary In an updated position statement issued in 1991, the American Medical Association emphasized that “DNR orders only preclude resuscitative efforts and should not influence other therapeutic interventions that may be appropriate.”1 More than ten years have passed since this updated position statement, but the exercise in extrapolating care plans from DNR continues to occur in the course of our daily patient encounters. Further research, education, and work in this area would be of benefit to all who care for the elderly. Please submit manuscripts on ethical issues in long-term care to Fred Feinsod, MD, MPH, CMD, Department Editor, at Feinsod@fmfmd.com