2. End-of-life issues
16. Surgical interventions near the end of life: "therapeutic trials"
Carl C. Hug Jr.
An 86-year-old retired accountant was experiencing worsening symptoms and signs of congestive heart failure. He was particularly frustrated by generalized fatigue and dyspnea on excretion, which severely limited his functionality. Cardiac catherization and echocardiography revealed calcific aortic stenosis (0.8 cm2), significant occlusions of his coronary arteries (>95% LAD, >75% circumflex and RCA), global hypokinesis (EF 0.25), 2+ mitral regurgitation, dilated left ventricle, and pulmonary artery hypertension (70/35 mmH)g. Co-morbidities included non-insulin diabetes mellitus and right ocular blindness. The cardiac surgeon estimated a 20%–30% operative mortality and a 30%–40% chance of reducing his debilitating symptoms. The patient declined surgery initially, but returned 3 months later stating he would “rather die than go on in his present condition.”
He underwent aortic valve replacement and coronary artery bypass grafting during 3 hours of cardiopulmonary bypass after which an intraaortic balloon pump and multiple inotropic drugs were required to sustain marginal cardiac function. Platelets were required to treat his coagulopathy.
His 26-day stay in the intensive care unit required treatment for bacteremia and pneumonia, tracheotomy, and peritoneal dialysis. Fluctuating mentation, persistent low-grade fever, and two failed attempts of tracheal extubation indicated insertion of a percutaneous endoscopic gastric (PEG) feeding tube and a percutaneous intravenous central catheter. On the 27th postoperative day he was transferred to a long-term acute care facility.
On the 33rd postoperative day he suffered a massive GI bleed and systolic blood pressures in the 50’s. He was transferred to the hospital ER, then to the ICU with continuing requirements for transfusions, inotropic vasopressors, and mechanical ventilation to stabilize his vital functions. Two days later, profuse bleeding recurred and after 45 minutes of cardiopulmonary resuscitation (CPR) death was declared.
Ethical considerations: should the operation have been offered?
The patient clearly understood and accepted the balance of benefits versus risks, and gave informed consent to the procedure. Offering the operation is respectful of the ethical principle of autonomy. On the other hand, a surgeon might consider the balance of the principles of beneficence and nonmaleficence to be unacceptable, and as an independent moral agent, could justifiably refuse to perform the operation.1
Patients should carefully consider three specific questions before undergoing an intervention with substantial risks of bad outcomes:
(1) What are my goals for the intervention?
(2) What disabilities are unacceptable to me?
(3) What are my alternative options?
The patient’s primary goals were to improve his functionality and to reduce his symptoms even in the face of a substantial risk of dying. It is not clear that he fully understood that he might incur disabilities beyond those he was experiencing preoperatively. It is almost certain that the alternative option to surgery was palliative care and hospice either at home or in a hospice facility.
Medicine has been described as the science of uncertainty and the art of probability. There is never 0% or 100% benefit or risk. Sir William Osler said, “Every treatment is an experiment.”2 An interventional trial with both potential benefits and recognizable risks of death is neither euthanasia nor physician-assisted suicide. Statements by ethicists include: “only after starting treatments will it be possible in many cases … to balance prospective benefits and burdens”;3 “… trial interventions, coupled with hard-nosed clinical realism, may appropriately balance the possibilities for good or ill …”;4 “… The moral burden of proof often should be heavier when the decision is to withhold than it is to withdraw treatments.”3
However, an interventional trial has important requirements in the case of surgical operations with substantial risks of serious disabilities and death. Some time will be required to allow recovery from the operation. With evidence of progressive improvement over days (with transient ups and downs) support of vital functions should be continued. But there should also be an understanding and willingness to discontinue vital function support if:
(a) It is unlikely that the patient’s goals will be achieved,
(b) There is worsening of the patient’s condition despite continuing or even increased supportive therapy, and most assuredly,
(c) If multiple organ systems fail.
Ideally, all these possibilities should be discussed with the patient and his/her family well in advance of undertaking the operation.
Does this case represent surgical “euthanasia?”
On the basis of the principle of double effect, the answer to that question depends on the intentions of the patient and the surgeon.3 Although the patient and the surgeon recognized death as a possible outcome of surgery, it is clear that they accepted this possibility and had the intention of improving the patient’s status. The patient never stated a wish for euthanasia or assisted suicide.
What roles should the patient’s living will and durable power of attorney for healthcare (DPAHC) decisions have played?
The restrictions on interventions expressed in his living will (dated several years earlier) would have to be suspended for any patient undergoing cardiac surgery, because resuscitation measures such as mechanical ventilation, defibrillation, etc. are routinely used in the conduct of cardiac operations. His specific instructions to his primary (wife) and secondary (son who is an obstetrian/gynecologist) surrogates, if any, are unknown. Both were supportative of his decision to have the operation. It is doubtful that the issues of suspending his living will and the timing of reinstating it were ever discussed.
Several questions need to be addressed when the outcome of the intervention is poor and the possibilities of achieving the patient’s stated goals are virtually nil:
(1) When should the primary emphasis of care be changed from restorative to palliative? Both are intimately part of patient care throughout the intervention,5 but the restorative measures may become ineffective and burdensome to the patient and even to his family and friends. The burdens may include pain and discomfort as analgesic and sedative drug doses are reduced in attempts to wean the patient from mechanical ventilation. Examples of burdens to his family and friends include perceiving the patient as suffering, the uncertainty about the degree and duration of his dependency on continuing interventions and life supporting measures, the escalation of continuing costs of care, and the possibility that the patient would never be able to return to his home.
(2) What are the options for decelerating vital function support? First of all, the family should receive regular and realistic updates on the patient’s progress, or lack thereof, toward achieving the patient’s stated goals for the operation. The focus should be on the overall trend and not the minor ups and downs of vital signs, drug infusion rates, ventilatory settings, etc. It is important to recognize that it will take time, perhaps days, for the family to grasp the lack of improvement or deterioration in the patient’s condition.
Decelerating and discontinuing vital function support 6
When it becomes clear to the physicians that the patient is failing to respond to maximum therapy, and deterioration of one or more vital organ functions is occurring, it is appropriate to approach the patient (if able to comprehend) and family members on the issues of decelerating vital function support while continuing to keep the patient comfortable. Never say: “There is nothing more we can do.” Rather, assure the family that “we can keep the patient comfortable.”
Recognizing the family’s need for time to accept and adjust to the circumstances, deceleration may take a stepwise course. Often, the first step is to establish a do-not-resuscitate (DNR) order, while reassuring the family that all other therapies will be continued for the present time. Assure the family that the patient’s needs will not be ignored. The patient will not be abandoned. To the author, DNR has two meanings: Do Not Resuscitate. Do Not Relax!
Subsequent discussions may include discontinuation of one or more vital function supporting measures (e.g., dialysis). Sometimes a family will agree to discontinue all vital function supports with the exception of one particular measure which may be symbolic to them. When the ultimate decision is to discontinue all vital function supports, the family should be reassured that the patient will not be allowed to suffer. For that reason, an intravenous infusion of fluids should be continued, and drugs administered to prevent agonal respiration, which is most distressful to the family (e.g., morphine infusion at 2–5 mg per hour), and a drying agent (e.g. scopolamine 0.2 mg iv) along with oral pharyngeal and tracheal suction to prevent the “death rattle.” Intensivists differ on the maintenance or removal of the endotracheal tube to prevent these distressing signs.
Transferring the patient from the ICU
If the patient is stable but in need of continuing life-supporting measures, discussions with the family about transferring the patient from the ICU to another facility (e.g., long-term acute care) should be outlined in some detail and assurances given that there will continuity of care by virture of direct communication with the physicians and nurses at that facility.
Social Services can be very helpful in locating the appropriate facility that is satisfactory to the family. Most importantly, there should be assurances that communications with the physicians in that facility will include specific information about any limitations on further interventions (e.g., DNR order) to which the patient and family have agreed.
If the patient requires only moderate assistance with the activities of daily living, adhering to the prescribed schedule of medications, and a proper diet, the transfer sequence may be from the ICU to a private room in the hospital and then to a rehabilitation facility and then to a nursing home-type facility.
Given the stress of caring for such a patient in a private home the family should think long and hard about the burdens of providing 24/7 continuous care, especially if that burden will fall primarily on one person. 7
Dealing with demands that “everything” be done
This case is typical of the cases most commonly brought to an ethics committee of an acute care hospital today. Acute care hospitals are primarily involved with surgical and other types of invasive interventions sometimes followed by intensive care. Fortunately, most patients are successfully discharged from the hospital in stable condition, but some have varying degrees of disability and continuing requirements for assistance with their convalescence.
Aging draws on physiologic reserves (see Fig 16.1) to maintain independent function, and for some, the functional level is amazingly good as they compensate for accumulating chronic diseases. But when a significant hurdle is imposed, even apparently very fit patients may have too little additional reserves to clear the hurdle.8 This fact of life is not recognized or accepted by the general public in the US, especially by those “baby boomers” who strive to live “forever.” And discussions of death are taboo in most families and among friends.
In addition, the ideas of “medical miracles” in the news media coupled with television situations in which resuscitation virtually never fails, the common expectations are that “doing everything” will successfully restore any patient. Obviously, the truth is otherwise for many cases.
Suggestions of what physicians and nurses can do when faced with a family who demands that everything be done to save their loved one who is deteriorating include:
(1) Consistently state the truth about the patient’s condition in a compassionate manner and do not suggest there is hope based on minor, usually transient, changes in vital functions.
(2) Ideally, agree to have a single physician communicate with the family to avoid their impressions that the care team is divided because different physicians use different words and have different body language that are interpreted as different conclusions.
(3) Constantly refer to what the goals of the patient were in consenting to the intervention. Of course it is important to have the goals explicitly stated and recorded in advance.
Figure 16.1 Functional reserve is needed when a person faces a stressful situation. As persons age, they draw on their functional reserve to maintain relatively normal function, and as a consequence, they have less reserve to meet a new, stressful challenge such as a major surgical operation and the demands of recovery.
(4) Remind all involved that it is one thing to keep a heart beating, but quite another to restore a meaningful life as defined by the patient’s previous statements.
(5) Distinguish prolonging death from extending life.
(6) Involve pastoral services to help the family deal with their dread and grief. For many people, there is belief that death is a transition to another (better) existence.
As a therapeutic trial, the lengthy and complex operation initially had limited success in so far as the patient made slow progress. When it became evident that his situation remained tentative with virtually no chance that the patient could tolerate any major set back, consideration of initiating DNR would have been appropriate for two purposes: (a) it was in line with the patient’s advance directives; and (b) it would have been a clear signal to his surrogate-family members that expectations for reaching his stated goals were diminishing day by day.
Most importantly, the failure to have a DNR order in place at the outside facility resulted in an inappropriate transfer back to the hospital and futile attempts to salvage what clearly was a fatal event given his already extremely limited reserves.
• In end-of-life care, focus on the patient’s goals, not merely the individual procedures involved.
• Be willing to discontinue aggressive care if the goals cannot be achieved.
• Recognize that “comfort care” is an important and continuous form of caring.
• Deal with the demand that “everything be done” sympathetically but realistically.
1* Code of Medical Ethics. Termination of the physician–patient relationship. American Medical Association, Chicago, IL 2008–2009 edition, Section 8.115, p. 261–263; Section 8.20, p. 289, and Section 10.05, p. 354.
2* Jonsen, A.R. (2000). A Short History of Medical Ethics. New York: Oxford University Press, p. 89.
3* Beauchamp, T.L. and Childress, J.F. (2001). Non-maleficence. In Principles of Biomedical Ethics, 5th edn, New York: Oxford University Press, pp. 113–64.
4* McCullough, L.B., Jones, J.W., and Brody, B.A. (1998). Surgical Ethics. New York: Oxford University Press, pp. 152–70.
5* Statement on clinical practice guidelines for quality palliative care. (2006). American Academy of Hospice and Palliative Care. Glenview, IL. www.nationalconsensusproject.org.
6* Truog, R.D., Cist, A.F.M., Brackett, S.E. et al. (2001). Recommendations for end-of-life care in the intensive care unit: the Ethics Committee of the Society of Critical Care Medicine. Crit Care Med,29(12), 2232–48.
7* Mittelman, M. (2005). Taking care of the caregivers. Curr Opin Psychiatry, 18 (6), 633–9.
8* Muravchick S: (2000). Anesthesia for the elderly. In Anesthesia 5th edn. Miller, R.D., ed. Philadelphia, PA: Churchill Livingstone, p. 2141.
Ethical guidelines for the anesthesia care of patients with do-not-resuscitate orders or other directives that limit treatment (2008). ASA Standards, Guidelines and Statements, American Society of Anesthesiologists. Park Ridge, IL.
McCullough, L.B., Jones, J.W., and Brody, B.A. (1998). Surgical Ethics. New York: Oxford University Press, p. 182.
Troug, R.D., Campbell M.L, Curtis J.R. et al. (2008) Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College of Critical Care Medicine. Crit Care Med, 36(3), 953–63.