Moving From Resuscitative To End-Of-Life or Palliative Treatment
Our case scenario concerns a young woman who suffers life-limiting intracerebral hemorrhage (ICH) during sports; arrives comatose, requiring mechanical ventilation; and is near death. Her distraught family hopes that transfer to your tertiary care facility could offer functional survival. However, because of the patient's grim prognosis, the neurosurgeon has signed off the case. You already have several admitted patients boarding.
The patient lacks decision-making capacity, and we have identified the need for a surrogate decision-maker, in this case, the patient's husband. Next, we must ensure that the surrogate understands the treatment options available from the medical perspective. When the patient meets criteria for brain death, the medical community generally withdraws resuscitative measures; however, some states legally permit certain religious exceptions to this practice. Since our patient is not brain dead, the surrogate needs to know that while we can predict with a high degree of medical certainty his wife will die imminently, we cannot know exactly when. In such cases, many physicians favor initiating aggressive life-sustaining medical treatments while medical prognosis remains uncertain and then withdrawing those interventions when grave prognosis clarifies. Nevertheless, both law and ethics consider life-sustaining medical treatment withdrawal or withholding morally equivalent when medically indicated.
Although emergency medical treatments possibly may accomplish short-term personal goals of the patient or surrogate, the patient's husband needs to understand clearly that his wife will not survive this injury. We need to help the surrogate to recognize that a decision to withdraw life-sustaining medical treatment is not a decision to end the patient's life—her life with him as the person he knows essentially already has ended. In our scenario, even persistent vegetative state is unlikely, and our interventions will only prolong her dying.
From a medical perspective, we face the imminent death of this patient and a very busy ED in a full hospital. Since we understand that her prognosis is imminent death, rapidly withdrawing life-sustaining medical treatment in the ED avoids needless suffering and continued use of resources that will not help the patient recover. However, immediate withdrawal does not allow time for the family to understand and accept the patient's pending death, nor time to discuss with the surrogate potential organ procurement and donation, as well as complete the appropriate protocols.
From a humanitarian perspective, the family or surrogate does need time to accept the imminent death of their loved one. They may even need time for other family members to arrive. Our EP ethical values of beneficence and nonmalfeasance extend to the family of our patients, as well as to our patients, particularly in the face of a fatal outcome. Under the circumstances of an acute, unexpected fatal event, should we expect a grieving family or surrogate to be able immediately to make a medically rational decision? Whether or not the family can reach final decisions, EPs should start these discussions; moreover, a systematic approach may facilitate shared decision-making.
Although our primary responsibility as clinicians is to this patient and her family, her condition also includes the possibility of organ or tissue donation. Our patient clearly has a near-terminal brain injury and requires ventilation. This situation usually triggers notification of the local organ procurement organization before end-of-life discussions begin. Hospital and medical staff members should leave discussion of organ donation to a trained designated requester, the organ procurement organization representative, so early notification helps to assure this presence. Discussion of possible organ donation options should occur only after the surrogate agrees with the recommendation to withdraw life-sustaining treatment, unless he or she initiates that conversation.
Organ donation historically follows the “dead donor rule” (Table 1): prior to organ retrieval a prospective donor must meet criteria for brain or cardiorespiratory death in accord with the Uniform Determination of Death Act[5-13]; however, nonemergent living donation now occurs. Although our patient does not meet brain death criteria (Table 2),[15, 16] her condition opens the possibility of donation after cardiac death (DCD).
Table 1. Criteria for Organ, Tissue, and Eye Donation
|Donation follows determination of death|
|Brain death (see Table 2)||Maximizes donation.|
|Controlled||Near maximal donation (limited or no recovery of hearts).|
|Uncontrolled||Warm ischemia time often permits only recovery of kidneys, tissue, or corneas.|
|Donation from living donor||Permits partial donation of single organ or one of paired organs (even donation of normal heart by heart–lung recipient).|
Table 2. Determination of Brain Death in Adults
|Preconditions (all needed)|
|• Coma, irreversible (neuroimaging may explain)—requiring mechanical ventilation|
|• No central nervous system depressant drug effect or residual paralytics|
|• No severe acid–base, electrolyte, or endocrine abnormality|
|• Normothermia or mild hypothermia and systolic blood pressure ≥ 100 mm Hg |
|Examination (all needed)|
|• Absent ocular reflexes (pupils; corneal, oculocephalic, oculovestibular)|
|• No facial movement to supraorbital or temporomandibular stimulation|
|• No motor response to noxious stimuli in all four limbs|
|• No gag and cough reflexes|
|• Apnea testing—confirms no breathing drive|
|• Cerebral angiogram or scintigraphy|
|• Electroencephalogram (EEG)|
|• Transcranial Doppler ultrasound |
|• Time of positive apnea test or ancillary test interpretation (requires prior notice to organ procurement organization)|
Following several reports by the Institute of Medicine supporting organ DCD, the Joint Commission in 2007 implemented an accreditation standard requiring that hospitals have a policy addressing this organ donation option; moreover, the Organ Procurement and Transplantation Network/United Network for Organ Sharing mandates that all transplant hospitals and organ procurement organizations in the United States comply with its rules regarding DCD. DCD procedures apply to potential donors who do not meet brain death criteria and may occur as a controlled or uncontrolled process. During controlled DCD, cardiac circulation of a critically ill donor stops following consented removal of mechanical ventilation and hemodynamic support. After the donor meets criteria for cardiac death, organ retrieval may begin. Uncontrolled DCD involves procuring organs from patients after termination of unsuccessful resuscitation outside the hospital or in the ED. DCD respects the ethics of withdrawal in the context of the patient's best interests.
The Uniform Anatomical Gift Act (UAGA), passed revised in 2006, permits EPs to initiate uncontrolled DCD, which includes preplanned insertion of central venous catheters for infusion of organ-preserving agents by an on-call trained team not involved in treatment of the patient. The organ preservation team, however, acts only after the treating team has determined that cardiac circulation has stopped and has decided to stop resuscitative efforts. This action by some presumes consent for organ donation and would permit reperfusion efforts to start prior to arrival of family or other surrogate decision-maker. A symbol on the vehicle driver license of a prospective donor may indicate desire to donate organs, especially in states that have adopted the revised 2006 UAGA (Table 3).[20-23]
Table 3. Adoption of UAGA (2006)
|Enacted—46 states, DC, US Virgin Islands|
|Introduced—Puerto Rico in 2009, PA in 2012|
|Alternate action—FL, IL, NY|
|FL: enacted Florida Statutes, XLIV§ 765, Part 5: Anatomical Gifts|
|NY: enacted New York Code - Article 43: Anatomical Gifts|
|IL: enacted Illinois Anatomical Gift Act, 755 ILCS 50|
In states with donor registries, a particular vehicle driver license symbol indicates prior authorization for organ donation whenever possible; moreover, this authorization has legal priority for donation during any subsequent discussions concerning measures for organ preservation with a surrogate for the patient. If a person has not made a document of gift during life, one interpretation of the revised 2006 UAGA presumes intent to donate and expands the list of persons who can consent to organ donation on behalf of that individual. If a person prefers not to donate, this must be documented in a signed explicit refusal.
Whenever controlled DCD may occur, the revised 2006 UAGA requires that life support systems in place at the end of life continue until a notified organ procurement organization completes evaluation of the prospective donor in an ED or intensive care unit (ICU) for suitability to donate. However, if the prospective donor has no signed refusal of donation, UAGA (revised in 2009) gives the surrogate authority to resolve with the attending physician any conflict between presumed intent to donate and an advance directive not to use life support systems merely to prolong life.
Despite passage of the revised 2006 UAGA in most states, the logistics of recovering any viable organs by uncontrolled DCD have proven difficult in the United States,[26, 27] where prevailing opinion still favors surrogate consent for organ preservation or donation, especially if the prospective donor has not registered. In Europe, where presumed consent for uncontrolled DCD is more widely accepted, EDs succeed much more often, especially with kidney recovery.[28-32] Nevertheless, EPs in the United States can successfully identify potential donors for support elsewhere until controlled DCD or donation after brain death may occur.
During controlled DCD, the treatment team, with family consent, maintains organ perfusion with needed medications without otherwise harming the potential donor. After an organ retrieval team assembles, the treatment team withdraws ventilation in the operating room from a donor candidate while providing palliative care. The treatment team determines death 2 to 5 minutes after monitored cardiac activity stops, and organ retrieval may begin.
In our scenario, the hospital has no monitored location available for admission to permit more favorable conditions for organ donation to develop. Whenever withdrawal of resuscitative interventions begins, procedures should include prior understanding that a patient may not die immediately. If this occurs in the ED, EPs should continue palliative care and plan for eventual transfer of the patient to an unmonitored floor bed, a family grief room, or a palliative care unit, when available, while continuing family support.[35-38] Respect for person and family calls for using the most private available space to minimize interruptions during withdrawal of resuscitative interventions in the ED.
Despite all of the advances in emergency care, EPs cannot save all patients. Cases like ours challenge emergency providers to transition care from lifesaving to maximizing the quality of end-of-life care. Death and dying affect the whole family, not just the person who is dying. EPs must remember that we play an important role in providing care at the end of life. The end-of-life policy of the American College of Emergency Physicians states: “helping patients and their families achieve greater control over the dying process will improve end-of-life care.”
In this situation, Iserson calls attention to the often-forgotten patient, namely, the family survivor or surrogate. He states that most resuscitations end with patient-survivors, rather than surviving patients. These patient-survivors remind EPs that our obligations to family match our obligations to our critically ill or injured patient.
Transitioning care from active resuscitation to palliative treatment requires understanding of the concept of futility. Futile actions provide no useful result; moreover, the inability to achieve shared goals of medical treatment justifies withholding or withdrawing resuscitative efforts. Medical futility, however, has no uniform definition. Only six states define medically ineffective care or use the term futile in statutes (Table 4),[42-48] even though most states provide for physicians and health care institutions to decline treatment that is contrary to generally accepted health care standards. Moreover, some states permit physicians to decline to comply with a health care decision for reasons of conscience or personal belief or even to withhold or withdraw medical treatment based on moral, religious, or philosophical beliefs. If physicians intend not to follow patient or surrogate health care directions, most states require the providers to notify the patient or surrogate and to facilitate patient transfer to another physician or institution willing to comply with the patient or surrogate request. However, the statutes of five states (Idaho, Indiana, Michigan, North Carolina, and South Dakota) do not even address whether providers may decline to comply with a patient's (or surrogate's) request for treatment for any reason. Although Maryland, West Virginia, and Texas provide a process for resolving patient–provider disputes about medically ineffective treatment, only Texas offers legal immunity for providers concluding this process by removing life-sustaining interventions over surrogate objection.
Table 4. State Statutes Defining Medically Ineffective Care or Using the Term Futile
|Alaska Stat. § 13.52.060(f)|
|Delaware Code Ann. tit. 16, § 2508(f)|
|Maryland Code Ann. [Health-Gen.] §5–611(b)(1)|
|New Mexico Stat. Ann. § 24–7A-7(F)|
|Idaho Code 39-4514(5)|
|New Jersey Stat. Ann §26:2H-67 |
Life-limiting emergencies, however, allow no time to resolve disputes about futile interventions. Ethically acceptable reasons for limiting or withholding life-sustaining medical treatment with wide consensus include that treatment will not work or will only prolong the dying process, the patient with decision-making capacity refuses treatment, or the patient's surrogate refuses treatment based on the patient's wishes, including that treatment will leave an unacceptable quality of life as determined by the patient. More controversial reasons include that treatment will leave an unacceptable quality of life as determined by persons other than the patient not based on the patient's wishes or the treatment is too resource-consuming. For the Society for Critical Care Medicine, interventions failing to accomplish intended goals would be medically futile. The American Society of Clinical Oncology affirms that, when treatments directed at cure no longer offer likely benefit to the patient, clinicians should modify management goals to provide care directed at symptom management and sensitive to a patient's spiritual and psychological needs.[52, 53] The ACEP Code of Ethics affirms similar statements, especially with regard to dying patients. In EM, unless the patient rejects treatment or an advance directive is in place, full resuscitative efforts occur during the immediate crisis. If treatment does not benefit the patient, the medical professional standards of care can determine when further interventions are futile. When resuscitative treatments cannot accomplish intended goals and offer virtually no chance of clinical benefit, prolonged survival, or quality of life improvement acceptable to the patient, physicians and family can decide whether to transition from curative to palliative treatment. The challenge to EPs becomes management of this transition to palliative care.[55-57]
In our case, the husband who is the surrogate barely comprehends the bad news about his wife's prognosis; he is shaking and temporarily unable to speak. Additional family members, including the patient's two children and her sister, are also in a state of shock and leave the room in tears. We recognize that this family lacks preparation for receiving this news and cannot fully process understanding of this event. Attending to the needs of the family should immediately follow attending needs of the patient. Vulnerable families struggle to make difficult decisions while simultaneously dealing with emotional grief and often anger over the grave situations of their loved ones.
Vig et al. demonstrated that the stress of making medical decisions for loved ones evoked symptoms of posttraumatic stress disorder in one-third of ICU surrogates. Azoulay et al. found that 82% of surrogates who had to make end-of-life decisions had symptoms of posttraumatic stress disorder. The emergency team must adopt a patient- and family-centered approach to end-of-life care with the goal of promoting psychological well-being and reducing traumatic stress. The SUPPORT study (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment) showed that hospitals need to improve the quality of end-of-life care. Since that time, many studies have focused on factors and interventions that improve the care of the dying patient and his or her family. In “Means to a Better End: A Report on Dying in America,” the Robert Wood Johnson Foundation identified the care of families as a major deficit in end-of-life care. Mularski et al. further defined seven domains of quality indicators in palliative and end-of-life care. Four of these domains focused upon improved care for families. These four domains include patient/family-centered decision-making, communication between the health care team and patients/families, emotional/practical support for families, and spiritual/cultural support of patients/families. Although designed for the ICU, these indicators also apply in the ED when EPs share decision-making and stress the importance of caring for patients' families.
Decision-making is not a one-time event; we cannot ignore that people in crisis initially do not completely register information and that emotions interfere with understanding. End-of-life decision-making takes time and may require more than one conversation. Even in the ED, a series of several brief conversations with the health care team grants families time to process information and prepare for discussions about end-of-life care. To avoid causing further pain and suffering to the family, EPs must consider the emotional and social aspects to decision-making. In this case involving sudden devastating illness, the family has had no time to prepare for the role of making end-of-life decisions. Stepwise communication helps the family to accept that the patient is dying. Family-witnessed resuscitation efforts may facilitate this understanding. Open visitation may also enhance emotional/practical support of families by giving families time with their loved ones to grieve, to gain information, and to strengthen bonds with emergency caregivers. Time helps build trusting relationships especially in an environment like the ED, where we have had no prior relationship with the patient and family. Using this time to learn who the patient was before she was dying and getting to know the family more personally facilitates this bond and conveys respect for patient dignity.
Assuring privacy and attention to care is challenging in the ED where we are often communicating in hallways and have multiple patients competing for our attention. Incorporating a team of health care providers (nurses, residents, consultants, social workers, and chaplains) into the patient's care is the key to success in the ED, but all team members must give accurate and consistent information as coordinated by the team leader—the attending EP. Since no ICU will take over care of the patient in our case, the EP must assure continuity of care by communicating detailed information at change of shift and introducing family members to the new care team. Consulting a palliative care team may promote a more comprehensive individualized care plan. Moreover, when circumstances permit, referral to palliative care programs benefits patients, reduces hospital costs, and leads to higher patient and family satisfaction than conventional care. Although palliative care teams are not universally available in all hospitals, evidence supports 24/7 palliative care access, even within an ED.[66-70]
Critical care literature reports that ineffective communication is the most significant barrier to quality end-of-life care. Families often rank communication skills as having equal or greater importance compared to clinical skills. Specific problems identified include inadequate time spent communicating (family meetings averaged 10 minutes), lack of private quiet space (noise and interruptions can interfere with communication), inadequate training and education in palliative care communication (Table 5), more time spent talking than listening, and lack of active communication (eye contact, open-ended questions, pausing and allowing time for responses, using words patients/families can understand, restating and summarizing patient/family responses). Families report not understanding diagnosis, prognosis, or treatment 54% of the time.
Table 5. Elements of Palliative Care Communication
|1. Initiates relationship-centered dialogs about care including determination of decision-making capacity and identifying legal surrogate|
|2. Demonstrates empathic communication by using:|
|A. Verbal behaviors (e.g., naming, affirmation, reflection, listening, self-disclosure)|
|B. Nonverbal behaviors (e.g., touch, open posture, eye-level approach)|
|3. Reacts appropriately to own emotions and those of patients, families, and colleagues|
|4. Educates patients/families about medical, social, and psychological concerns related to life-limiting illness|
|5. Uses age, sex, and culturally appropriate terminology when talking with families and patients|
|6. Demonstrates above skills in common patient–family situations, such as:|
|A. Giving bad news|
|B. Transitioning treatment goals from curative and/or life-prolonging focus to palliative care|
|C. Discussing advance care planning and resuscitation status|
|D. Identifying appropriate care settings, including hospice|
|E. Discussing end-of-life care needs related to dying child or dying adult|
|F. Discussing patient or family requests for care unable to meet patient's goals for treatment at end of life|
|G. Withholding or withdrawing of life-sustaining therapy|
|H. Discussing palliative sedation|
|I.Discussing artificial hydration and nutrition|
|J. Addressing severe spiritual or existential suffering|
|7. Organizes and conducts a family meeting|
|8. Facilitates effective interdisciplinary palliative care team meetings|
|9. Develops effective relationships with medical care team by:|
|A. Summarizing active palliative care issues and treatment|
|B. Building consensus about treatment plans and goals of care|
|10. Documents in medical records all relevant domains of palliative care, including death determination and completion of death certificate|
Prognostic information is often the single most important piece of information needed urgently to make informed decisions. Although accurate prognosis is often difficult, using an evidence-based approach makes disclosure as objective as possible. Coma, ICH volume (>60 mL), and advanced age independently predict high mortality.[77, 78] Thirty-day mortality post ICH decreased from 88% to 58% following introduction of computerized tomography. This contrasts early clinical diagnoses of severe ICH compared to a patient population including less severe hemorrhages. The American Heart Association, therefore, recommends early aggressive treatment of ICH for 2 days before considering “allow natural death” orders. However, Rost et al. validated an ICH clinical score predicting functional independence at 90 days. This tool may greatly aid decision-making regarding withdrawal of resuscitative interventions for patients who value more highly the likelihood of functional survival, rather than likelihood of survival alone. These tools apply to prognosis following a acute devastating ICH. Although palliative care physicians employ several other performance-based prognostic scoring models for patients with indolent life-limiting illness (Table 6),[82-86] Chan describes their limitations in emergency practice. Physicians may offer specifics on prognosis reluctantly due to uncertainty; however, providing clinical information within context of patient values and prognosis determined as best possible using clinical tools does not destroy hope. Families may continue to hope for the best while preparing for the worst outcomes.
Table 6. Prognostic Indicators
|Palliative.info: Guideline for Estimating Length of Survival in Palliative Patients||Prognosis by diagnosis, circumstances, or functional decline|
|Karnofsky Performance Scale||Performance status as predictor of survival|
|Palliative Performance Scale (PPS)||Predict approximate 6-month mortality|
|Palliative Prognostic Score (PaP)||Predict 30-day mortality|
|Palliative Prognostic Index (PPI)||Predict < 3-week mortality|
Making patient- and family-centered decisions regarding end-of-life care in the ED focuses upon the patient's values and goals. Such values include spiritual concerns and cultural beliefs; however, the EP should avoid cultural stereotypes, since beliefs may vary from patient to patient. If the patient has lost decision-making capacity, the emergency provider should determine with the family surrogate what treatment the patient would have wanted in these circumstance. Wishes of the patient should substitute for those of the family; however, if surrogates do not know what treatment the patient would choose, they should set, together with the attending clinicians, goals of treatment that reflect a positive balance of benefits over burdens.
A good death has no single definition. Empowering the family in end-of-life care decision-making should include the following parameters: 1) assess the family's understanding of what is happening to their family member; 2) give an honest view of the patient's prognosis; 3) elicit “substituted judgment” based on the patient's wishes so that end-of-life care honors the patient's values; 4) communicate medical information in small increments and use understandable language; and 5) build trust and make decisions with family that incorporate patient's values, beliefs, and preferences in determining goals of care. Clinicians should discuss these issues with all families of critically ill patients and not just those families of patients expected to die. When prognosis is poor, focus not on whether a patient will die but how.