The field of nephrology is shifting from an exclusive focus on increasing survival to one that provides greater attention to quality of life. There is an opportunity to integrate many of the advances of palliative medicine into the comprehensive treatment of these patients.
—J Palliat Med. 2006;9:977-992
The statement above was advanced nearly four years ago, yet as far back as 1991, the Institute of Medicine was calling for a clinical practice guideline that would help practitioners evaluate patients who might be better off without dialysis treatment.
Nine years later, the Renal Physicians Association/American Society of Nephrology (RPA/ASN) Working Group in Washington, DC, obliged, issuing a "Clinical Practice Guideline on Shared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis" (J Am Soc Nephrol. 2000;11:1340-1342).
In general, the RPA/ASN concluded that withholding or withdrawing dialysis in cases of end-stage renal disease (ESRD) or acute renal failure (ARF) was appropriate for:
Patients with decision-making capacity who, being fully informed and making voluntary choices, refused dialysis or requested that dialysis
be discontinued.
Patients who no longer possess
decision-making capacity but had previously indicated refusal of dialysis in an oral or written advance directive.
Patients who no longer possess decision-making capacity and whose properly appointed legal agents refused dialysis or requested that it be discontinued.
Patients with irreversible, profound neurologic impairment.
The guideline further instructed that patients who were taken off dialysis receive continued palliative care.
Today, the guideline is being revised, "and it's going to have even more
recommendations with regard to palliative dialysis," said Alvin H. "Woody" Moss, MD, a coauthor of the Journal
of Palliative Medicine report cited
above and chair of the RPA/ASN Working Group that introduced the clinical practice guideline a decade ago and is overhauling it now.
Medical evidence grows
Although a 2003 review turned up too little evidence to justify making changes to the RPA/ASN guideline, by 2008
the research landscape had changed considerably. "We determined that there was sufficiently new medical evidence with regard to this whole area of withholding and withdrawing dialysis, and palliative care to make it worthwhile to revise the guideline," said Dr. Moss, Professor of Medicine and Director of the Center for Health Ethics &
Law at West Virginia University's Robert C. Byrd Health Sciences Center in Morgantown. He is board-certified in internal medicine, nephrology, and hospice and palliative medicine.
Dr. Moss told Renal & Urology News that the working group hopes to have the revised document ready by autumn of this year. "And I can tell you the recommendations are even going to be much stronger now than they were a decade ago."
Palliative care defined
Rehabilitative dialysis is for individ-uals in whom clinicians are trying to maintain or improve function, whereas palliative dialysis is for people "who we would not be surprised to see die in the next year," Dr. Moss explained. One major catalyst in the decision to rework the dialysis-withholding/withdrawal guideline is the fact that more than half a dozen studies demonstrate that select patients may live just as
long with medical management alone as with dialysis.
"Our research shows that these are patients who are over age 75; have high symptom burden and multiple other medical problems, particularly heart disease; and are often very functionally impaired—for example, they have difficulty transferring from a wheelchair to a dialysis chair,"
Dr. Moss said. "I would propose that we use this definition of palliative dialysis and that we actually incorporate it into our thinking about treating
dialysis patients." Studies conducted by
Dr. Moss and others indicate that 15% to 20% of patients in any given dialysis unit would probably fit this palliative-care profile. As the renal population grows older and sicker, so grows an appreciation that palliative care is more appropriate for many of these individuals—either in combination with dialysis, or in lieu of it.
"In the past the thinking always was, 'You have renal failure; dialysis can save your life,' and then once you were on dialysis, you were on it until the end," commented nephrologist Michael J. Germain, MD, Professor of Medicine at Boston's Tufts University School of Medicine and Medical Director of the transplantation services at Baystate Medical Center in Springfield, Mass. "And once you were on dialysis, [that decision implied] that you wanted to be kept alive at all costs."
A shift in attitude
Dr. Germain co-edited a book on palliative services in kidney disease called Supportive Care for the Renal Patient (Oxford University Press). A second edition is due out this year, updating the 2004 version. "There have been not only a lot of...studies done over the past six years, but I think there have been changes in the attitudes of nephrologists and renal care providers over those six years, too," Dr. Germain said. In updating the text with co-editors Edwina Brown, MD, and E. Joanna Chambers, MD, Dr. Germain came across data that reinforced his belief that dialysis is not always the best course of action. "Recent studies have shown the very, very poor outcomes for patients with renal failure once they're already in long-term care or go into long-term care," he said. "Very few of them ever get out or go and live independently. Dialysis may not provide the best quality of life for whatever time some of those patients have left."
In many of those cases, "probably the better thing would be to not start dialysis at all."
Like Dr. Moss, Dr. Germain pointed to the somewhat startling findings of the past few years that have showcased the benefits of conservative management in patients who choose not to undergo dialysis. "What's really surprising about some of these studies is they show that a certain subset of patients will live just as long if they don't get dialysis as if they did get dialysis."
He acknowledged the counter-intuitive nature of this concept. "If you have a glomerular filtration rate (GFR) of 10 and you choose not to have dialysis, well, you'll probably die sooner," Dr. Germain said. "But we see that while that might be true for 'all' patients, in a certain subset of people over 75 who have comorbidities, evidence suggests these people might live just as long if they choose not to have dialysis, and may have a better quality of life in their remaining time. They may have a low GFR, but it stays there—their renal impairment doesn't progress, and they'll die of something else, but not uremia or renal failure."
Lewis M. Cohen, MD, agrees with Dr. Germain, his Baystate colleague and frequent research collaborator. "The tradition in nephrology—as in medicine in general—has been to view death as an enemy to overcome by any means," said Dr. Cohen, Director of the Renal Palliative Care Initiative, a collaboration between eight dialysis units and Baystate Medical Center. "A generational shift has taken place, with some balance now being provided by interest in quality of life."
Dr. Cohen practices psychosomatic medicine and psychiatry with a special focus on end-of-life issues. He addresses these controversial matters in his newly released book, No Good Deed: A Story of Medicine, Murder Accusations, and the Debate Over How We Die (HarperCollins). The book is an account of two Massachusetts nurses who were accused of murder when they helped alleviate a dying patient's pain and suffering.
Dr. Cohen also was lead author on the Journal of Palliative Medicine report cited at the start of this article. Asked what progress has been made in integrating palliative dialysis or other palliative care into the typical dialysis setting since that study was published, he responded: "Each year has seen an increase in the number of ESRD deaths preceded by a decision to stop all treatments. Each year has seen more research into the symptoms experienced by dialysis patients. Each year has seen more emphasis from CMS [Centers for Medicare & Medicaid Services] and other organizations in encouraging completion of health-care proxies and living wills."
According to Dr. Cohen, the most common mistake people make when discussing palliative dialysis is to assume that it's only about providing end-of-life care. "It certainly attends to those issues, but it is also about focusing on symptom amelioration for all dialysis patients," he said. "This is something that nephrology has paid less attention to until recently."
Kenneth Covinsky, MD, MPH, concurred. "One thing to keep in mind when thinking about palliative care in ESRD [is that] there is often a mistaken belief that one must choose either palliative care or life-prolonging care," said Dr. Covinsky, who holds the Edmund G. Brown, Sr., Distinguished Professorship in Geriatrics at University of California-San Francisco (UCSF). "But palliative care and life-prolonging care can be delivered simultaneously."
In the case of dialysis, then, it is quite possible that a patient would choose to start or continue dialysis, but would still benefit from palliative care, observed Dr. Covinsky, who also serves as a staff physician at the San Francisco VA Medical Center and is a prominent contributor to GeriPal, an online forum dedicated to geriatrics and palliative care news, research, and discussion (www.GeriPal.org).
Target nursing home patients
"Because the rates of mortality and functional decline are so high in nursing home patients receiving dialysis, nursing home patients with ESRD should be targets for palliative care services regardless of whether they choose to initiate dialysis," he said.
Like the other thought leaders interviewed for this article, Dr. Covinsky references the research of his colleague, Manjula Kurella Tamura, MD, MPH. Dr. Kurella Tamura is Assistant Professor of Nephrology at the Stanford University School of Medicine as well as a coinvestigator for the Frequent Hemodialysis Network study, a randomized trial of short-daily or long-nocturnal dialysis versus conventional thrice-weekly dialysis.
Earlier this year, Dr. Kurella Tamura led a team that analyzed patients'
preferences for dialysis withdrawal
and engagement in advance care planning (Nephrol Dial Transplant. 2010;25:237-242). Her group found that preferences for dialysis withdrawal varied by race and ethnicity, with black patients less likely than whites to opt for withdrawal. Older age was not associated with withdrawal preferences, but the pattern of engagement in advance care planning varied by age as well as race and ethnicity. (Younger participants and minorities were generally less likely to document treatment preferences than older patients and white patients, but were equally likely to engage in end-of-life discussions.)
Advance planning is crucial
In a 2009 study on the functional status of elderly adults before and after dialysis initiation, Dr. Kurella Tamura, Dr. Covinsky, and other investigators used a national registry to identify 3,702 U.S. nursing home residents who began dialysis between June 1998 and October 2000 and who had at least one available pre-dialysis measure of functional status. They found that in nursing home residents with ESRD, dialysis initiation was associated with "a substantial and sustained decline" in functional status (N Engl J Med. 2009;361:1539-1547).
"I think the area of palliative care is one of several issues the nephrology community needs to address in order to better serve the needs of our aging and increasingly frail population,"
Dr. Kurella Tamura said. "Ideally, advance care planning should take place before starting dialysis, and these discussions should be continued after evaluating the response to dialysis."
In Dr. Kurella Tamura's opinion, palliative ESRD care (a term she prefers to "palliative dialysis") should be considered when an ESRD patient has a very high symptom burden and a limited life expectancy despite the provision of dialysis—for example, less than three months.
In addition, she stated that symptom management should be a priority for all patients—whether or not they are receiving palliative ESRD care—and more attention should be focused on preserving independence. "Some might not consider these areas to be the domain of the nephrologist or the dialysis unit, but because we have regular contact with ESRD patients, we are well-placed to implement these measures. Perhaps implementing a geriatric allied health professional into the dialysis unit, in the same way as the unit dietitian, may be a useful way to address these issues regardless of their goals of care."
Integrating palliative care
For the most part, Dr. Kurella Tamura said, there appears to be no coordination at the national level to address the integration of palliative care into the typical dialysis setting other than the upcoming guidelines from
Dr. Moss's group.
What steps, then, does Dr. Moss recommend that an individual dialysis unit take to better integrate palliative care into its offerings? "What they can do is a very complicated process—it's called talking to one another!"
Dr. Moss remarked. "It's called communication, and advance care planning, and explaining to patients and families that they need to talk about these issues and put their wishes in writing."
As Dr. Moss pointed out, the re-vamped conditions for ESRD coverage issued by CMS in 2008 require that dialysis units have a policy on advance directives. "If patients are adequately informed about their options and how sick they are, they can make informed choices that will lead to less pain and suffering and more death with dignity in a setting of their choice as opposed to the ICU," he said. "I think that very slowly, physicians, nurses, and social workers are beginning to understand that perhaps our default in our
health-care system—which is the patient dying in the ICU on a breathing machine, restrained so they don't pull out their tubes—is probably not the most humane way to go."
Tough sell to the public
Nevertheless the trend toward backing off dialysis as an automatic course of treatment could be a tough sell to the public, which could feel shortchanged. "Patients and their families might think that dialysis is their right, but they might be demanding unrealistic care that's not likely to benefit them," Dr. Germain said. "They have to understand that our job is to do what's best for them, and that doesn't always mean providing the most aggressive, expensive care."
Nephrologists themselves may be an equally tricky audience to win over, as their focus has traditionally been on prolonging life and helping patients avoid death at all costs. "I can tell you that the leadership in nephrology and large dialysis organizations already agree with the viewpoint [of pursuing more palliative services for ESRD patients]," Dr. Moss said, "but I think it's going to take a while for nephrol-ogists throughout the country to subscribe to the concept. It's not the way we're used to doing things."
One barrier, Dr. Germain noted, is that physicians may not feel comfortable discussing palliative and hospice care if they don't feel they have a good handle on a patient's prognosis. To that end, he and Drs. Moss and Cohen (along with biostatistician Robin Ruthazer, MPH) once again teamed up, this time to develop a "mortality tool"—an instrument that enhances the accuracy of predicting six-month survival for a person on maintenance hemodialysis (Clin J Am Soc Nephrol. 2010;5:72-79).
Nephrologist training needed
Even if a nephrologist is confident that dialysis is more likely to interfere with a terminal patient's quality of life than improve it, he or she simply may not feel comfortable discussing
end-of-life care with the patient
and/or family. "That's where we need to improve training and education of the doctors, so they're more comfortable having these discussions,"
Dr. Germain said. "Unfortunately, it's not really happening anywhere routinely
at the present time."
As a medical student at the University of Calgary in Alberta, Canada,
Dr. Germain saw patients and communicated with families as part of the curriculum "from day 1," but that is usually not the case in the States. "For a lot of medical schools, you do your basic science work for two or three years, and then you see patients,"
Dr. Germain said. "But [learning to interact with patients and families] really needs to start from the beginning of medical school, and carry on in residency and then through fellowship."
Dr. Cohen recalled that when he and Dr. Germain began studying palliative renal care, they hoped nephrologists could be easily trained in these techniques. Now, however, "we have come to appreciate that while [palliative-care education] should be incorporated into fellowship programs, most practicing nephrologists are already saturated with their existing skills and responsibilities. The idea of them, for example, participating in a week-long course to communicate better with patients and families on end-of-life issues is simply not practical."
Strategies for improvement
Instead, Dr. Cohen said, one of the best courses of action would be to invite local palliative-medicine and hospice resources to participate in dialysis care. "One of our recent research efforts entailed encouraging staff from both the local hospices and the dialysis clinics to become better acquainted with each other," he said. "We were able to demonstrate a significant increase in dying ESRD patients receiving hospice services, especially among the elderly, following this simple process."
In addition, Dr. Cohen recommended providing more training to the interdisciplinary team that already exists in the dialysis unit. For example, nephrologists and dialysis social workers can work together to improve communications with patients, their families, other health-care personnel, and representatives from relevant community resources.
The lack of formal education in palliative care for nephrology trainees is just one obstacle to the integration of palliative services into a dialysis unit. Inadequate funding for palliative-care research, poorly aligned financial incentives, and lack of access to palliative-care specialists in some regions are all significant barriers as well. But the picture is far from dismal.
"I can tell you we are making progress," Dr. Moss said, pointing to a recent jump in hospice referral rates for dialysis patients as proof.
"When patients with end-stage
renal disease stop dialysis, they
usually die in about eight days," he explained. "So you'd think that these people would be perfect candidates to be referred to hospice, but we
found that according to the USRDS [United States Renal Data System]
data on dialysis patients who had died in 2001 and 2002, only 41% of those who stopped dialysis were being referred to hospice."
At the time of that analysis, Dr. Moss was chairing a Robert Wood Johnson Foundation working group that
was looking at end-of-life care for dialysis patients. "So one of the recommendations we made was that
CMS should track hospice referral
for dying dialysis patients," he re-
counted. "And in October 2004, CMS revised its death notification form—which must be completed by dialysis unit personnel and submitted to the End-Stage Renal Disease Networks whenever a patient dies—so that it now asked whether the person had been referred to hospice prior to death. This serves as a sort of reminder to the dialysis unit staff
[that], 'Hey, maybe we should be referring patients to hospice if CMS
is asking about it.'"
Now, Dr. Moss said, 60% of patients stopping dialysis are being referred to hospice, according to 2008 USRDS data. However, further research by various ESRD networks indicates some gaps in knowledge among social workers, nurses, and nursing directors in dialysis units. "There is clearly a need for more education to be done,"
Dr. Moss said.
Among other nationwide efforts aimed to educate the nephrology and hospice communities about palliative care for dialysis patients, Dr. Moss,
Dr. Germain, and others are presenting a "webinar" series on behalf of the Kidney End-of-Life Coalition
(www.kidneyeol.org). The coalition has also recently posted on its Web site an algorithm for pain management
of dialysis patients.
Marked progress foreseen
All in all, Dr. Moss expects that in
the next five years or so, marked
progress will be made in the incorporation of palliative renal services. This is due, in large part, to grow-
ing public awareness of end-of-
life issues, the pending completion of the RPA working group's newly revised clinical practice guideline, and the increasing number of nephrology fellowship programs that are incorporating palliative-care training into their agenda.
"It may take a few more years, but I'm optimistic," he said. "We are not used to considering palliative services in renal care, and it takes a little bit of time to turn the direction on an
aircraft carrier." n