Course Authors
Frieda Wolf, M.D.
Dr. Wolf is Assistant Professor of Medicine, Department of Medicine, Downstate Medical Center, Brooklyn, New York.
Within the past 12 months, Dr.Wolf reports no commercial conflicts of interest.
Estimated course time: 1 hour(s).
Albert Einstein College of Medicine – Montefiore Medical Center designates this enduring material activity for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
In support of improving patient care, this activity has been planned and implemented by Albert Einstein College of Medicine-Montefiore Medical Center and InterMDnet. Albert Einstein College of Medicine – Montefiore Medical Center is jointly accredited by the Accreditation Council for Continuing Medical Education (ACCME), the Accreditation Council for Pharmacy Education (ACPE), and the American Nurses Credentialing Center (ANCC), to provide continuing education for the healthcare team.
Upon completion of this Cyberounds®, you should be able to:
Describe the increasing burden of comorbidities faced by geriatric patients with ESRD
Discuss the withdrawal/withholding of therapy and its consequences -- many symptoms and ultimately death for ESRD patients
Incorporate principles of end-of-life care into the care of the dialysis patient in order to improve their quality of life and quality of death.
Every nephrologist has to eventually come to terms with the fact that, for ESRD patients, mortality is high with an annual 23% mortality in dialysis units.(1) This overall high mortality in individuals with ESRD is driven by the mortality of dialysis patients, as shown in Figure 1.
Figure 1. USRDS 2006 ADRD: All Cause Mortality Rates By Modality, 2004.
As age increases, mortality increases. Note the higher rates for dialysis patients compared to the general CMS population.
Thus a dialysis patient has a five-year life expectancy similar to a cancer patient's. Figure 2 shows the expected remaining lifetime for people with renal replacement therapy, compared with the general aging population.
Figure 2. USRDS 2006.
Expected remaining lifetime years of the general U.S. population by age gender and race. The life expectancy of a patient on dialysis is much shorter than that of the general population.
Survival plummets for ESRD patients with comorbidities such as diabetes mellitus, hypertension and congestive heart failure.(1) U.S. ESRD patients are most likely to die from cardiovascular disease, infections and a third cause, not usually mentioned, the withholding of or withdrawal from dialytic therapy -- 21% of patients die because of withholding/withdrawal. By far, the principal causes of death, as seen in Figure 3, are cardiovascular disease and infections.
Figure 3. Causes of Death in Dialysis.
USRDS 2005.
Withdrawal Is Common
For U.S. dialysis patients, withdrawal from dialysis is the third most common cause of death.(1) Figure 3 is a composite, based on data from the American Hospice Association and the USRDS. Nearly 22% of dialysis patients died secondary to withdrawal from dialysis in 2003; however, only 13% of patients with ESRD who died that year did so in a hospice setting. Death was attributed to withdrawal from dialysis in only 9.8% of patients with ESRD who died in 2003 in a hospice program. This means that the majority of deaths that are attributed to withdrawal/withholding takes place at home or the hospital without end of life care, although these deaths are predictable to some extent (see Figure 4).
Figure 4. Withdrawal of HD (WHD) and Hospice.
USRDS 2005.
Death as a consequence of withdrawal/withholding may also be common around the world - it is the second most reported cause of death in Canada(2) and may well be the most common cause of death in Europe, especially in France(3), Finland(4) and Australia.(5)
The precise incidence of withdrawal from hemodialysis is not known because cause of death is often categorized imprecisely: sudden cardiac death or ventricular arrhythmia from hyperkalemia may simply be reported as "cardiac arrest." Or the cause of death might be listed as uremia, which is more accurate than to state that the patient died because of withdrawal of therapy. Thus, this method of reporting death masks the true incidence of withdrawal as the precipitant cause of death.
We also don't know whether death after withdrawal from dialysis is similar to death from uremia and what the symptoms are.
Palliative Care Measures to Be Implemented in Dialysis
The shorter life expectancy and high rate of withdrawal from dialysis for ESRD patients indicate that there are many medical, psychological, legal and spiritual issues associated with end-of-life that need to be addressed. As Figure 5 shows, in 2004, nearly 335,000 Americans were receiving dialytic therapy. Who better to address these issues than the dialysis staff and the nephrologist specifically? These are the people that require knowledge and skills in palliative care.
Figure 5. USRDS 2006 Annual Data Report.
Prevalent patient counts by modality, 2004.
Dialysis patients routinely visit their centers three times a week, making possible a close relationship between them and their health care team. The health care team in dialysis includes the nephrologist, the nurse, the dietician, the technician, the social worker and, of course, the patient. This close relationship provides an opportunity to discuss deteriorating quality of life, symptoms such as fatigue, pain, depression, sexual dysfunction or even loss of income.
The quality of one's death is also part of the 'quality of life.' Some patients may need immediate attention to end-of-life issues because death is more imminent. The dialysis health care team is equipped to identify the individuals most likely to die sooner.
Is there a tool to identify patients at greater risk for dying? Alvin Moss and Sanjay Sharma reported at the 2006 American Society of Nephrology (ASN) meeting that the 'surprise question,' specifically "Would I be surprised if this patient died in the next year?" may be a tool in identifying those patients. The question helps spotlight those patients with greater morbidity, for whom immediate discussion of end-of-life care would be appropriate.
All patients need and deserve repeated discussions on end-of-life issues and scenarios; however, the patient who fails to thrive or has a higher comorbid burden, in particular an untreatable disease such as dementia or cancer and a higher symptom burden such as uncontrollable pain, may warrant more pointed factual discussion. The discussion with these patients should not only focus on advanced directives but needs to address withdrawal from dialysis, especially if suffering is related to technical issues of dialysis which outweigh its benefits.
Identifying patients who may benefit from withdrawal or withholding of dialysis is by no means easy, and most certainly should involve the patients and their support systems or surrogates. A survey of 400 dialysis patients by Hines and Moss demonstrated that patients prefer to involve family members in advanced decision-making, and they are more comfortable with legislation which allows their family, not an advanced directive, to decide on their behalf in the event they are incapacitated.(6) This may reflect the need for the nephrologists to lead the discussion on advanced directives or may perhaps reflect the mistrust patients harbor toward the system. However, since a discussion of advanced directives should address the medical issues, who more naturally than nephrologists should lead the discussion?
Leading an end-of-life discussion requires certain skills; surprisingly, nephrologists are not well trained in these.(7) Formal training is not the only, or even necessarily the best, way to acquire these end-of-life skills. Some tools can be acquired during training, especially if shown by example and not only formal training. Although end-of-life care has not been included in all curricula, it has been taught in what Susan Block characterized as the "hidden" curriculum,(8) -- the implicit values of our teachers and peers expressed by their behavior rather than explicit language.
In Dr. Block's course, called "Living With a Life-threatening Illness", taught at Harvard Medical School, students develop strong bonds with volunteer dying patients. This experience teaches important lessons, including the tools of listening and bearing with suffering. Students learn how to communicate in a manner that is more accessible for patients. They learn to avoid medical jargon, to be active listeners, and to be sensitive to cultural issues that may affect the patient's experience. These are basic core skills that are helpful to every physician, not only those practicing palliative care.
Balancing control and compassion is also required. In their effort to maintain professionalism and control, those attending the dying may come to be viewed by patients as lacking compassion or too distant. Skill in eliciting the patient's perspective gives the physician and nursing staff a chance to show they value the patient's feelings and experience and replaces avoidance with approachability. At the same time, the palliative care staff needs to be ready to exert control in the form of, for example, making sure that the setting and the individuals who attend the meetings are carefully chosen. Family meetings should be conducted in a private quiet location, without people passing through the meeting room. No unfamiliar individuals should attend.
The Quality of Death
The quality of one's death is part of the quality of life in the same way that a person's work or ability to travel or what they eat defines their life. What is considered a 'good death'? Uremia has been assumed to lead to a 'good death' which, for most people, implies a symptom-free and dignified death. Although death from uremia is perceived to be painless and with a significant degree of somnolence, many patients have symptoms which mimic death from cancer (Figure 6).
Figure 6. Symptoms During Last 24 Hours.
| Symptom (79 patients) | % present |
|---|---|
| Pain | 42 |
| Agitation | 30 |
| Myoclonus/twitching | 28 |
| Dyspnea/agonal breathing | 25 |
| Fever | 20 |
| Diarrhea | 14 |
| Dysphagia | 14 |
| Nausea | 13 |
Cohen et al. AJKD, 2000;36:140-144.
In a study by Cohen(9) of 79 patients who were prospectively studied until their death after discontinuation of dialysis therapy, 42% of patients were in pain in the last 24 hours and 30% were agitated. One quarter had shortness of breath. This was confirmed by a retrospective chart review of 35 patients who withdrew from dialysis in Ottawa(10) and who were treated by a palliative care team for symptoms after withdrawal. The most common symptoms were pain, agitation and shortness of breath. Symptoms were successfully treated with opioids and benzodiazepines.
These two studies support the role of palliative care medicine within nephrology. Specific skills should be taught to nephrologists in training and to the health care teams in dialysis units. Hemodialysis patients are a captive audience, present in their seats for 12 hours a week. During these hours, we should teach the proper 'dialysis diet,'we should educate patients about the optimal kt/v and we should take the opportunity to discuss advanced directives. It should be made clear, furthermore, that every decision is reversible, even the decision to discontinue dialysis, so dialysis could be restarted, for example, to alleviate symptoms such as congestion or shortness of breath.
Discontinuation of dialysis is not synonymous with patient abandonment. It is the realization that patients have autonomy and that our job is that of healers. It is an opportunity to show the patient and the family compassionate care.
EOL: The Process
Is everyone on dialysis dying? Of course not. But eventually we all die, regardless of kidney function. It is not only whether we live or die that matters but how we do it (Figures 7-9).
Figure 7.
Figure 8.
Figure 9.
Discussion of end-of-life (EOL) issues should not be limited to those whose death is more imminent but should be discussed with everyone at the nearest opportunity. Those who care for patients dying of cancer can tell us that there is no worse time to discuss "Do Not Resuscitate" (DNR) with a patient or family than when they are actively dying or are admitted to a hospital with an acute illness. However, when it is not imminent, death may seem as a remote event and few may want to talk about EOL issues. Advanced directives are one example of palliative care measures. Unfortunately, very often when advanced directives are discussed, even with dialysis patients, the issue of withdrawal from dialysis is not addressed.(11) Advanced directives in our ethical framework enable autonomous decision-making. Living wills and healthcare proxies are commonly used in order to clarify a patient's wishes when he is otherwise unable to make them known.
Although this discussion and crafting of these initiatives should ideally take place as part of general medical care or in preparation for renal replacement therapy, more often than not it is omitted completely. Dialysis units often 'screen' for advanced directives with the use of a questionnaire upon entry into a dialysis unit. An advantage of this approach is that the patients may perhaps feel they are not singled out and that everyone gets the same treatment. This screening question, however, is useless if there is no follow-up discussion and continuous vigilance for related concerns.
The Price of Success
Most nephrologists are well versed with USRDS data regarding incidence and cause of end stage renal disease. They know that nearly one and a half million people have ESRD. This surely underestimates the true prevalence of ESRD because not all countries with the ability to dialyze have registries and not all registries report all renal replacement therapy. When Willem Kolff invented the artificial kidney in 1943, no one dreamed how successful this therapy would become. Never have the number of patients and the cost of their medical care been greater.
According to the Centers for Medicare and Medicaid Services (CMS), uremia therapy cost $18 billion in 2004, about $64,000 per patient year, according to the USRDS annual data report. One might call this the price of success. Although more nephrologists are in training, they cannot match the rising prevalence of patients with chronic kidney disease, estimated at nearly 20 million people by NHANES III.(12) The high mortality may also reflect the increasing enrollment of an older population (Figure 10) into the ESRD program, a population with more significant comorbidities. Since the fastest growing population segment in the U.S. is in its eighth decade or older, it's no surprise that this oldest group has the worst survival, which is confirmed by another study that looked at survival worsening with increasing age (Figure 11).(13)
Figure 10. ESRD Enrollment By Age.
Increasingly older patients are enrolled in the CMS ESRD program.
USRDS, 2005.
Figure 11. Survival Is Worse for Older Patients.
Munshi SK, Nephrology Dialysis Transplantation, Volume 16, Number 1, January 2001, pp. 128-133(6).
Renal Palliative Care Initiatives
There have been two major initiatives for palliative care in dialysis, both funded by the Robert Wood Johnson Foundation. One is the RPCI or renal palliative care initiative. The second is the ESRD peer workgroup, established by the National Kidney Foundation, with its subgroups for quality of life, quality of death and education. Rather than starting palliative care/end-of-life care interventions when curative or therapeutic efforts stop, these initiatives should be given concurrently with therapy in order to address concerns and symptoms appropriately. Prevention and acceptance of the emotional, psychological, physical and existential suffering associated with end-of-life are part of healing. Dialysis need not stop when patients enter hospice care. Although controversial, there are hospices that accept patients on hemodialysis. It should be used as palliative treatment when congestion, shortness of breath or even hyperkalemia cause symptoms.
The Renal Physicians Association and the American Society of Nephrology developed guidelines for end-of-life decision-making in 1999. They recommended that CMS and ESRD networks work together to develop guidelines for patients and for training clinicians in end-of-life care, and that dialysis and hospice care be linked and better coordinated.
One challenge to overcome is to identify those patients most likely to benefit from withdrawal or withholding dialysis. The guidelines suggest that those patients with very poor prognosis because of untreatable disease or very poor quality of life secondary to various reasons be approached first. These may include patients with cancer, dementia or pain that is unresponsive to therapy, or patients in whom technical difficulties make dialysis intolerable. Five years after dialysis discontinuation most families reported becoming more comfortable with the decision to hasten death.(14) Good deaths presuppose a mentally alert patient and occur at home, take place while asleep, are painless (or largely pain-free), peaceful and take place in the company of loved ones.
Nephrologists and EOL: Changing Attitudes
There is some evidence, as reported by Holley and colleagues, that attitudes towards end-of-life care held by nephrologists are changing.(15) A survey of 318 nephrologists about current practice revealed that nephrologists are much more likely today, as compared to 1990, to withhold dialysis from a permanently unconscious patient; they are much more likely to honor a patient's DNR status and to consider consultation with the Network ethics committee; and dialysis units were far more likely to have a policy on CPR or withdrawal of dialysis.
In support of patient-centered end-of-life care, Network 5 formed the Kidney End-of-Life Coalition, made up of representatives from RPA, CMS, the National Hospice and Palliative Care Organization and other experts, to enable better coordination of end-of-life care for patients with chronic kidney disease. Their website is http://www.kidneyeol.org.
EOL Training and Nephrology Training
In a national survey of second-year nephrology fellows,(7) respondents were asked about the amount of training they had and how prepared they felt to manage various nephrological diseases and conditions. Asked whether they were taught how to respond to a request to stop dialysis notice, 50% responded yes, though more than 70% indicated that they were not explicitly taught to help with reconciliation and goodbyes, or to tell their patient that they are dying. If fellows were exposed to palliative/end-of-life specialists, they were more likely to have been explicitly taught appropriate skills.
Fellows also felt they got far less feedback on palliative/EOL skills, than on technical procedures such as kidney biopsy. All kidney biopsies were performed under observation and all fellows received direct feedback from senior faculty. In contrast, for family meetings if conducted, about half of the fellows were observed, while only a third of these received any feedback from supervisors.
Improving Renal Replacement Therapy
Multiple efforts are being made to improve the quality of life for dialysis patients. Historically, these strategies include tools to improve adequacy of dialysis such as kt/v, the use of recombinant erythropoietin to treat anemia and the control of bone disease.
In the Congress, the frequency and reimbursement of dialysis have recently been addressed by a just introduced bill -- Access to a Better Choice in Dialysis. The bill proposes a staggered payment rate for individuals who dialyze more than three times a week. The first three treatments per week would be reimbursed at the standard composite rate, the fourth session would be reimbursed at 70%, the fifth session at 40% and so on. There would be a cap on the amount of money Medicare would spend on the pilot project each year; CMS would study the overall health of the patient and the use of other separately billable services such as erythropoietin, iron and hospitalizations. The government would also look into patient work status by analyzing Social Security Disability Insurance payments, federal and state tax payments and the impact on Medicare payments. These analyses may inevitably lead to "daily dialysis," which may be the next step in the continuing improvement of patient quality of life. We will discuss daily dialysis in an upcoming Cyberounds®.
Conclusions
The integration of palliative care into dialysis affords an opportunity to improve end-of-life care and thus the quality death as well. The skills acquired by end-of-life training enable better patient care, maximize the patient's autonomy and allow the patient to die in optimal comfort and with dignity. It also affords the health care team an opportunity to help and learn from their patients, to value their beliefs and respect their autonomy. One model for the incorporation of end-of-life care to dialysis is suggested in Figure 12 -- attention should be paid simultaneously to the medical problems associated with CKD, while concurrently addressing the emotional and psychological needs of the patient.
Figure 12. Model for End of Life Care in Dialysis.
Brown University EOL toolkit.
Communication skills and compassion are the key features of this effort. It affords us all the opportunity to address death and dying, to explore our own values and beliefs, and to show our patients that we are human. This way we can become better healers.