Considerations in Starting a Patient with Advanced Frailty on Dialysis: Complex Biology Meets Challenging Ethics

Transcription

1 CJASN epress. Published on June 20, 2013 as doi: /CJN Ethis Series Considerations in Starting a Patient with Advaned Frailty on Dialysis: Complex Biology Meets Challenging Ethis Mark Swidler Summary Nephrologists have foused on the uremi syndrome as an indiation for dialysis. The elderly frail renal patient approahing ESRD represents a omplex biologi system that is already failing. This patient phenotype exhibits progressive geriatri disabilities and dependene interspersed with shrinking periods of stability regardless of whether dialysis is started. Consequently, the frail renal patient faes hallenging treatment hoies underpinned by ethial tensions. Identifying the advaned frail renal patient and optimizing the shared deision-making proess will enable him or her to make well informed hoies based on an understanding of his or her overall ondition and personal values and preferenes. This approah will also permit nephrologists to fulfill their ethial obligations to respet patient autonomy, promote patient benefit, and minimize patient harm. Clin J Am So Nephrol :, doi: /CJN Introdution The renal ommunity is grappling with an explosion of CKD stage 4 5 patients over the age of 80 years with a spetrum of omorbidities and geriatri syndromes, inluding frailty, who bring expetations and misoneptions about what dialysis as a life-sustaining and quality of life therapy may ahieve. Dialysis has beome routine pratie in the urrent environment of proeduredriven medial are and biomedialization of aging, whih has made it hallenging for patients, families, and health professionals to make true hoies (1). There are guidelines for situations where it is appropriate to forgo dialysis, like irreversible oma, terminal aner, or inability to tolerate the proedure (2). However, there is also a growing group of elderly frail renal patients who have no absolute ontraindiations to renal replaement but are at risk for early mortality, inreased hospitalizations, aeleration of geriatri syndromes, and signifiant symptom burden. Beause medial are of the seriously ill patient by tehnology has beome a major fous for ethial judgments about life, longevity, and how love and aring are expressed (1,3), patients and families struggle with saying no to dialysis therapy and yes to nondialysis medial therapy, although medial indiations, big piture goals, and quality of life reasons for hoosing the latter may be ompelling. Shared deision-making, whih inludes informing the patient about the prognosis, dialysis trajetory, and nondialysis medial therapy; exploring patient values and preferenes; and making an appropriate reommendation, ould improve ESRD treatment hoies and uphold the ethial priniples that support the pratie of mediine. The purpose of this artile is to examine the onsequenes of frailty for the elderly patient with renal disease, desribe the frail renal phenotype as a sreening tool to identify the advaned frail patient, and review ethial onsiderations involved in helping the advaned frail patient faing dialysis make good hoies. Frailty in Older ESRD Patients A reent study (4) using Mediare data from the US Renal Data System (USRDS) looked at the intensity of are in the last month of life in 93,329 patients 65 years and older who initiated hroni dialysis and then died over a 5-year period. Compared with similar data in aner and heart failure patients, those patients on dialysis had more hospitalizations (76 versus 61.3, 64.2%), intensive are unit admissions (48.9 versus 24, 19%), and hospital deaths (44.8 versus 29, 35.2%) and less hospie use (20 versus 55, 39%). The mean hospital stay was 9.8 days, and 29% underwent at least one life-sustaining intervention, inluding mehanial ventilation (22.3%), ardiopulmonary resusitation (11.9%), and feeding tube plaement (3.9%), before death. Use of these proedures did not differ signifiantly by sex, ause of ESRD, omorbid illness, or duration of dialysis and were more ommon in Afrian Amerians and those patients who died from ardiovasular auses. The average first-year mortality in dialysis patients over the age of 80 years an approah 46%, and it an be 58% in nursing home patients who initiate dialysis while in long-term are (5,6). Up to 34% of elderly patients will withdraw from renal replaement therapy ompared with 20% in the general dialysis population (7). Beause frailty is ommon in the CKD and dialysis (8 10) populations, a signifiant proportion of older ESRD patients may have advaned frailty at the end of life (EOL). Frailty starts early in CKD and is an independent risk fator for death and hospitalization Departments of Mediine and Geriatris/Palliative Mediine, Iahn Shool of Mediine at Mount Sinai, New York, New York Correspondene: Dr. Mark Swidler, Mount Sinai Medial Center, Renal Division, Annenberg 23-95, 1 Gustave L. Levy Plae, New York, NY mark.swidler@mssm. edu Vol, 2013 Copyright 2013 by the Amerian Soiety of Nephrology 1

2 2 Clinial Journal of the Amerian Soiety of Nephrology (8 10). The prevalene of frailty in the CKD population is approximately two times the prevalene in a general geriatri outpatient ommunity (14% 15% versus 6% 7%) (9,10). Frailty inreases as estimated GFR (egfr) delines, with an adjusted prevalene that is 2.1- and 2.8-fold greater for egfr values of and,30, respetively, ompared with an egfr of.60 ml/min per 1.73 m 2,andit is assoiated with an estimated 2.5-fold (95% onfidene interval [95% CI]51.4 to 4.4) greater risk of death or dialysis therapy (10). This result might explain the findings in a Veterans Administration study of CKD patients $85 years who were more likely to die than be treated with dialysis for ESRD (11). Frailty inreases as muh as fivefold in dialysis patients and is independently assoiated with a higher risk of death (adjusted hazard ratio [HR]52.24, 95% CI51.60 to 3.15) and ombined outome of death or hospitalization (adjusted HR51.63, 95% CI51.41 to 1.87) (8). A reent study found that frailty was assoiated with starting dialysis at a higher egfr as well as inreased mortality, whih was attenuated when this latter finding was orreted for frailty (12). This finding suggests that the symptoms and signs of frailty may be mistaken for or overlap with the signs of uremia and would then ontribute to the poor outomes in subgroups of elderly dialysis patients (Table 1), whereas traditional kidney quality indiators meet performane targets. Like the unreognized symptoms and palliative are needs (13) of ESRD patients, frailty is underdiagnosed and undertreated, in part beause of the lak of a uniform definition and diagnosti riteria (14). Beause this biologi entity has signifiant ramifiations and provides a formidable opponent to quality of life, it is an ethial imperative for the renal ommunity to understand, reognize, and be transparent about the Cyle of Frailty (see below and Figure 1) that impels ertain subgroups of advaned CKD and dialysis patients to the EOL experienes desribed above with a high potential for suffering. This sober view must frame information sharing and reommendations in the dialysis deision-making proess. The revised 2010 Renal Physiians Assoiation guideline on the appropriate initiation and withdrawal of dialysis (2,15) provides guidane in shared deision-making, advane are planning, prognosis evaluation, and palliative are and has been shown to be effetive in improving nephrologists preparedness in EOL deision-making (16). Consequenes of Frailty for the Elderly Patient with Renal Disease The geriatri syndrome of frailty is a biologi wasting syndrome of older adults that spans multiple physiologi systems, is haraterized by dereased reserves, resistane to stressors, saropenia, protein energy malnutrition, and atheroslerosis, and is preditive of disability, hospitalization, and mortality in ommunity elders (17,18). Frailty is assoiated with inreased inflammatory biomarkers (18,19). Reently, a frail mouse model has been haraterized (20,21) that onfirms the role of inflammatory pathway ativation in this syndrome. Uremia and the dialysis proedure provide miroinflammatory and oxidant stress environments (22,23) that may aelerate the expression and progression of frailty in predisposed subsets of the geriatri renal population. Frailty is a risk fator for disability and omorbidity but may exist independently of both (17). A useful oneptual model is the Cyle of Frailty (24) (Figure 1) that is ativated through trigger points of entry by intervening health events and ontinues to reyle after the event has terminated. At 6 months after an episode of septiemia or long bone frature, dialysis patients have adjusted relative risks of death of 7.1 and 3.2, respetively, ompared with a referene nonevent (no septiemia or long bone frature) dialysis population (25). As the number of aute health events inreases, the frailty yle will aelerate to its end points of dependene, disability, and death. The Cardiovasular Health Study, a longitudinal observational study in a geriatri outpatient $65-years-old patients ommunity, operationalized a frailty linial tool or frailty phenotype omprising five omponents: unintentional weight loss, exhaustion, low physial ativity, slow gait, and weakness; a positive test requires the presene of at least three of five omponents, and one to two of five omponents signifies a prefrail state (17). The frailty phenotype had an overall prevalene of 6.9% (16.3% in those patients years old and 25.7% in those patients years old), with a 4-year inidene of 7.2%. It was an independent preditor (HRs estimated over 3 years in parentheses; all signifiane levels at P,0.05) for inident falls (1.29), worsening mobility (1.50), Ativities of Daily Living (ADL) disability (1.98), hospitalization (1.29), and death (2.24). Of those patients who tested positive for frailty, 27% had neither ADL disability nor omorbidity (two or more omorbidities). Additionally, those patients whowereprefrailatbaseline(46.6%)hadanadjusted odds ratio of 2.63 of beoming frail in the next 3 4 years ompared with those patients who had no frailty omponents (17). It must be emphasized that the operational definition of frailty varies widely aording to the oneptual framework, and no gold standard exists (14,26). At what point does frailty beome advaned or irreversible? Like the EOL onept (27,28), frailty an be viewed Table 1. Mean life expetany by quartile following dialysis initiation aording to age and renal phenotype Renal Phenotype Quartiles Life Expetany by Age Group (yrs) th Perentile (0 25; frail) th Perentile (25 75; vulnerable) th Perentile (75 100; healthy) Adapted from referene 33, with permission.

3 Clin J Am So Nephrol :,, 2013 Ethial Considerations for Frail Patients Faing Dialysis, Swidler 3 Figure 1. Cyle of Frailty with trigger entry point health events. Reprinted from referene 28, with permission. as a spetrum diagnosed by liniian assessment and prognosti tools (see below). The Canadian Study of Health and Aging Clinial Frailty Sale (29) defines moderately frail as needing help with both ADLs and instrumental ADLs and severely frail as being ompletely dependent on others for ADLs or terminally ill. Alternatively, the presene of reurrent falls, inreasing disability, and snowballing episodes of aute illness with inomplete reovery might suggest the presene of end stage frailty (14). Although there are no urative options for the frailty syndrome, targeted exerise and ongoing geriatri evaluation may improve linial outomes (14). A positive effet on ADL and instrumental ADL disability has usually required relatively long-lasting and intensive multiomponent exerise programs (30). There are no speifi frailty intervention studies in ESRD patients, although geriatri renal rehabilitation programs have had suess in improving funtional disability (31). Elderly patients who initiate traditional dialysis will experiene loss of independene at home and the need for inreasing assistane (32). Dialysis also does not prevent funtional deline in nursing home patients (6). These onsequenes may reflet aspets of progressing frailty, and if the patient is at an advaned stage of frailty, then interventions, inluding renal replaement, will not be suessful in any signifiant and long-lasting way. Frail Renal Phenotype: A Global Sreening Tool to Identify the Frail Advaned Renal Patient The frailty phenotype measure by Fried et al. (17) is a useful sreening and researh tool for physial frailty, but it does not inorporate other ontributors to frailty, suh as ognitive deline or omorbidities (26), whih an drive the Cyle of Frailty. For any partiular age group, dialysis patients an be divided into quartiles with differenes in life expetany (33) (Table 1). Parallel healthy (75th to 100th perentile), vulnerable (25th to 75th perentile), and frail (0 to 25th perentile) renal phenotypes have been desribed (Table 2) to improve deision-making and supportive/palliative renal management (28,34). The frail renal phenotype (Table 3) is a useful global onstrut that ombines geriatri suseptibility fators (dementia, inability to ambulate, positive physial frailty testing, hypoalbuminemia, and signifiant symptom burden), survival data, and omorbidity information (2,34 38). Identifying this phenotype provides useful information relevant to the shared deision-making ESRD treatment disussion. Can this phenotype be onsidered an EOL diagnosis? Qualitatively, it an. Although the quantitative prognosis may be unertain, it is important to implement an advane are plan (ACP) (34). The geriatri omponent of the frail renal phenotype inludes funtional disability, dementia, and frailty. Funtional disability refers to limitations in mobility, ADLs, and/ or instrumental ADLs, and it is assoiated with inreased mortality, hospitalization, and long-term are independent of its ause (39). Inability to transfer is assoiated with a relative risk for death of 1.54 (5) and reeives a high-ranking number on omorbidity sores (37). Dementia diagnosed predialysis in advaned CKD is an independent risk fator for subsequent death and

4 4 Clinial Journal of the Amerian Soiety of Nephrology Table 2. Healthy, vulnerable, and frail renal phenotypes with assessment tools Renal phenotypes Healthy/usual Most optimal dialysis patient Might also be a transplant andidate Vulnerable More typial dialysis andidate Charaterized by inreasing hospitalizations and unpreditable outomes Frail Most suseptible to poor near-term outome (6 12 months) High risk of multiple and prolonged hospitalizations Some are nursing home patients with marked funtional disability, ognitive impairment, and dementia Medial are deisions depend more on patient preferenes and quality of life issues Assessment tools Geriatri suseptibility fators Presene of dementia Presene of frailty Funtional disability Comorbidity Modified Charleson omorbidity sore (36) Hemodialysis mortality preditor (35) Frenh Renal Epidemiology and Information Network 6-Month Prognosis Clinial Sore (37) Surprise question (38) Nursing home patient syndrome (6) Symptom burden assessment (13) Modified from referene 34, with permission. funtional deline after dialysis is initiated. In a retrospetive ohort study using USRDS data (40), the average time to death of patients with dementia before renal replaement who then started dialysis was 1.09 versus 2.7 years (P,0.001) for those patients without dementia, with a 2- year respetive survival of 24% versus 66% (P,0.001) and an adjusted HR for death of 1.87 (95% CI51.77 to 1.98). The dementia patients also experiened a threefold inrease in the loss of ambulation and a greater than fourfold inrease in the loss of transfer ability (both P,0.05). A marked inrease in mortality in inident dialysis patients who have oexistent dementia and annot ambulate has also been doumented (7). Ethial Considerations: Making Good Choies with the Advaned Frail Patient Faing Dialysis The ethial linial senarios faing frail patients with advaned renal disease and patients on dialysis are framed by the priniples of respet for autonomy, benefiene, nonmalefiene, and justie (41). Besides uring disease, the fundamental goals of mediine inlude the relief of suffering, treatment of symptoms, espeially pain, maintenane of quality of life, ommuniation about prognosis, and avoidane of harm (41). When patient goals are lear and one therapy option is superior, Table 3. Frail renal phenotype Karnofsky sore,50 (disabled; requires speial are and assistane) Older age ompared with years: years, RR51.22 (95% CI51.20 to 1.24); $90 years, RR51.56 (95% CI51.51 to 1.61) Presene of geriatri suseptibility fators (syndromes) Dementia Nonambulatory status (RR51.54, 95% CI51.49 to 1.58) Positive frailty testing Serum albumin,35 g/l (RR51.28, 95% CI51.25 to 1.30) Signifiant symptom burden Would you be surprised if this patient died in the next year? No Low survival probabilities by Comorbidity sores: CCI$8 (36); FREIN linial sore$9 (37) Hemodialysis mortality preditor (35) Four hroni onditions (RR51.68, 95% CI51.64 to 1.72) Nursing home patient (6) RR, relative risk for death (5); CCI, Charleson omorbidity sore; FREIN, Frenh Renal Epidemiology and Information Network 6-Month Prognosis Clinial Sore. Modified from referene 34, with permission. with signifiant benefits and small risks, no ethial dilemma exists; however, a onflit may arise when goals are not realisti or ahievable and when benefits are outweighed by burdens (42). Dialysis has beome one of many routine tehnology-based medial interventions ommonly used in the elderly, and it exemplifies the suesses of life extension but also, the responsibilities and burdens of medial hoie plaed on health pratitioners, elderly patients, and their families (43). When ounseling advaned CKD patients with a positive frail renal phenotype about dialysis versus nondialysis therapy, nephrologists must emphasize not only benefits,butalso,they must speak learly about possible negative onsequenes and offer reommendations for therapy that will either reflet patient values and preferenes or serve the patient s best interest if deisional apaity or prior wishes are absent (42). Best interests are the set of elements that make up quality of life and involve the balane of the benefits to the burdens assoiated with the proposed treatment. They must be examined from the patient s personal viewpoint and values, and they must take into aount not only the disease onditions but also nonmedial fators, suh as interpersonal relationships, resoures, and soial irumstanes (41) (Table 4). It should be emphasized that the ethial priniple of benefiene that underpins the best interests onept assumes that patients are vulnerable and medially uninformed, whereas patient autonomy implies an informed deision (42). Beause both an oexist, physiians must advise patients and guide them away from unwise deisions (42), whih means that the nephrologist should make a therapy reommendation but must also be prepared if the patient and family deide differently.

5 Clin J Am So Nephrol :,, 2013 Ethial Considerations for Frail Patients Faing Dialysis, Swidler 5 Table 4. Promoting best interests of the frail renal phenotype Understand patient perspetive in ontext of geriatri priniples, personal history, values, and quality of life needs Address misunderstandings and onerns What do dialysis and nondialysis therapies involve? What is the time and resoure ommitment? What are the assoiated symptoms? What will get better or ontinue to deteriorate? Make reommendations based on prognosti data and goals of are Negotiate a mutually aeptable plan Modified from referene 42, with permission. The patient preferene issue is partiularly important, beause older patients may hange their preferenes for life-sustaining therapy depending on the partiular linial situation (44). Family members and physiians may also pereive patient wishes that are disordant with what the patient atually desires. In this regard, patients preferenes for dialysis in different irumstanes were inorretly predited by surrogates, family members, and their physiians up to one third of the time (45,46), with families onsistently overestimating patients desires to ontinue dialysis aross hypothetial health onditions. This disordane may lead to tension between patient preferenes and best interests in the ontext of advaned frailty if deision-making apaity is absent and surrogates or health are agents are direting deisions. Does dialysis or nondialysis medial therapy have value or reate harm for a patient with advaned frailty? Does either enhane or diminish quality of life in these patients? These questions are omplex questions that involve assessment of both physial and psyhologial levels framed by the patient s personal narrative, ultural and religious bakground, support system, and exploration of what value and quality of life mean for that individual patient. Renal replaement therapy will address uremi syndrome and fluid overload-related issues, and it generally but not always (47) extends survival. In addition, the dialysis trajetory is assoiated with potential ompliations that should be dislosed and disussed, inluding (34) sudden death, whih may our while on the dialysis mahine, ardiovasular events, reurrent and prolonged hospitalizations, infetions like atheter-related bateremia, need for long-term are after hospitalizations, intensive are unit admissions (4), hroni ritial illness (48), inreasing frailty, funtional deline, loss of independene, and dialysis disontinuation with an average survival of 8 10 days (7,49,50). There is no evidene that dialysis an reverse geriatri syndromes like frailty, funtional disability, and dementia. Additional onerns relate to symptom and quality of life measures. Both advaned CKD and dialysis patients an experiene a signifiant symptom burden that may persist with renal replaement therapy (13,51). This burden may be, in part, beause of inadequate symptom assessment and treatment and underuse of palliative mediine expertise (13). Although quality of life is ultimately a personal value judgment, many health-related quality of life indiators do not appear to improve over the ourse of dialysis (52). CKD patients exhibit ompromised health-related quality of life that worsens as GFR dereases and is orrelated with inreasing age, dereasing serum albumin, and inreasing omorbidities (53). One study revealed similar impairments in both advaned CKD and maintenane dialysis patients groups with regard to symptom load, quality of life, and depression (54). Alternatively, nondialysis therapy may be assoiated with a stable funtional status until late in the trajetory, fewer hospitalizations, and more deaths at home (55,56). A range of issues and attitudes regarding dialysis hoies has been studied (43,57 60), and they reflet the ethial hallenges that providers and patients fae. In a series of open-ended interviews and partiipant observation (43) involving 43 dialysis patients and their hoies, only two patients were desribed as proative. Desriptors of patient experienes inluded there was no deision it just happens, I had no hoie, I wanted to live, and when you will need to start dialysis and not if. This finding suggests that primary emphasis is on mortality (survival and life expetany) rather than the inreasing geriatri, omorbid, and quality of life issues related to the dialysis trajetory. Sharing more information about this latter aspet is needed when onsidering quality of life preferenes, beause the ourse in frail patients is one of deline, regardless of whether dialysis or nondialysis therapy is hosen. Frail advaned CKD patients require inreasing support and palliative are with ongoing reassessment and heightened ommuniation. The need for better ommuniation was illustrated in a small study looking at kidney disease trajetory disussions between nephrologists and their older advaned CKD and dialysis patients (60). Patientbased themes inluded unertainty about the disease trajetory and lak of preparation for living with dialysis, whereas nephrologist-related issues inluded diffiulty explaining illness omplexity, diffiulty managing a disease over whih they have little ontrol, and the tendeny to avoid disussions about the future. These findings suggest barriers to optimal patient hoies by limiting the ability to weigh benefits, burdens, and best interests onerns. This information may have relevane to another study, where 62% of patients regretted their deision to start dialysis; the majority hose dialysis over supportive are, beause it was their physiian s hoie (52%) or family s wishes (15%) (61). Interventions to guide patients faing ESRD treatment hoies inlude shared deision-making and deision aids (59). As part of the Choosing Wisely ampaign to help health are providers optimize the medial deision-making proess and empower patients, the 2012 Amerian Soiety of Nephrology Quality and Patient Safety Task Fore reommends to not initiate hroni dialysis without ensuring a shared deision-making proess between patients, their families, and their physiian (62). Individual patient goals and preferenes underlie this shared deision-making proess, and therefore, information on prognosis and expeted benefits and harms of dialysis must be analyzed within the ontext of these goals and preferenes. Fousing on outomes with empatheti statements and questions that eliit big piture goals (34) first and then integrating them into a

6 6 Clinial Journal of the Amerian Soiety of Nephrology realisti ESRD treatment plan is required to make an informed deision. This information is the domain of ACP (63), whih improves the hanes of implementing patient wishes and inreases both patient and family satisfation at EOL (64,65). Also of interest is the inreasing use of deision aids that provide information on options in omplex medial deisions and help patients larify and ommuniate the personal values that they assoiate with different features of the options; however, these aids do not advise people to hoose one over another option (66). These deision tools improve patient knowledge, risk pereption, and realisti expetations of treatment options, help inorporate patient values in deision-making, and redue deisional onflit (59,67 69). Two CKD-speifi patient deision aids (what type of dialysis should I have? and should I stop kidney dialysis?) are available (70). Although studies need to be done to evaluate these kinds of tools, the development of CKD-speifi deision support best praties for renal patients along the CKD trajetory (59) will improve the quality of hoies for the frail renal phenotype. Conlusion Renal physiians an no longer onsider dialysis de fato treatment for all ESRD patients. The realities and ahievable end points of dialysis therapy for frail renal patients must be linked to their goals moving beyond not only living quantitatively but also exploration into what life means on a daily basis. Patients and their physiians must determine what quality of life is desirable and attainable, how it is to be ahieved, and what risks and disadvantages are assoiated with the desired quality target (41). Taken together, in evaluating a patient with advaned frailty for dialysis, the nephrologist should review ethial priniples, the proess of shared deisionmaking, and fundamental goals of mediine; disuss frailty as a biologi syndrome with vulnerability to adverse outomes; use the frail renal phenotype to supplement other prognosti information about what life with dialysis onstitutes for an elderly frail patient; desribe nondialysis medial therapy as an ative multidisiplinary treatment option (71); and explore big piture goals and math those goals with an appropriate and realisti ESRD treatment plan. Fortunately, palliative or supportive renal are, with its emphasis on deision-making, the ACP proess, and mathing medial therapy to patient goals, is beoming a part of the management of renal patients (2,34,63,72,73). Its usage in CKD patients with advaned frailty will lead to more realisti disussions about likely outomes with and without dialysis, resulting in better informed patient hoies that will optimize quality of life and entail less suffering. Dislosures None. Referenes 1. Kaufman SR, Shim JK, Russ AJ: Revisiting the biomedialization of aging: Clinial trends and ethial hallenges. Gerontologist 44: , Renal Physiians Assoiation: Shared Deision-Making in the Appropriate Initiation of and Withdrawal from Dialysis, 2nd Ed., Rokville, MD, Renal Physiians Assoiation, Kaufman SR: Making longevity in an aging soiety: Linking ethial sensibility and Mediare spending. Med Anthropol 28: , Wong SP, Kreuter W, O Hare AM: Treatment intensity at the end of life in older adults reeiving long-term dialysis. Arh Intern Med 172: , Kurella M, Covinsky KE, Collins AJ, Chertow GM: Otogenarians and nonagenarians starting dialysis in the United States. Ann Intern Med 146: , Kurella Tamura M, Covinsky KE, Chertow GM, Yaffe K, Landefeld CS, MCulloh CE: Funtional status of elderly adults before and after initiation of dialysis. N Engl J Med 361: , US Renal Data System (USRDS): Atlas of End-Stage Renal Disease in the United States. Bethesda, MD, National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases, 2008, p Johansen KL, Chertow GM, Jin C, Kutner NG: Signifiane of frailty among dialysis patients. J Am So Nephrol 18: , Shlipak MG, Stehman-Breen C, Fried LF, Song X, Sisovik D, Fried LP, Psaty BM, Newman AB: The presene of frailty in elderly persons with hroni renal insuffiieny. Am J Kidney Dis 43: , Roshanravan B, Khatri M, Robinson-Cohen C, Levin G, Patel KV, de Boer IH, Seliger S, Ruzinski J, Himmelfarb J, Kestenbaum B: A prospetive study of frailty in nephrology-referred patients with CKD. Am J Kidney Dis 60: , O Hare AM, Choi AI, Bertenthal D, Bahetti P, Garg AX, Kaufman JS, Walter LC, Mehta KM, Steinman MA, Allon M, MClellan WM, Landefeld CS: Age affets outomes in hroni kidney disease. J Am So Nephrol 18: , Bao Y, Dalrymple L, Chertow GM, Kaysen GA, Johansen KL: Frailty, dialysis initiation, and mortality in end-stage renal disease. Arh Intern Med 172: , Weisbord SD, Carmody SS, Bruns FJ, Rotondi AJ, Cohen LM, Zeidel ML, Arnold RM: Symptom burden, quality of life, advane are planning and the potential value of palliative are in severely ill haemodialysis patients. Nephrol Dial Transplant 18: , Ko FC: The linial are of frail, older adults. Clin Geriatr Med 27: , Moss AH: Revised dialysis linial pratie guideline promotes more informed deision-making. Clin J Am So Nephrol 5: , Davison SN, Jhangri GS, Holley JL, Moss AH: Nephrologists reported preparedness for end-of-life deision-making. Clin J Am So Nephrol 1: , Fried LP, Tangen CM, Walston J, Newman AB, Hirsh C, Gottdiener J, Seeman T, Tray R, Kop WJ, Burke G, MBurnie MA; Cardiovasular Health Study Collaborative Researh Group: Frailty in older adults: Evidene for a phenotype. J Gerontol A Biol Si Med Si 56: M146 M156, Walston J, MBurnie MA, Newman A, Tray RP, Kop WJ, Hirsh CH, Gottdiener J, Fried LP; Cardiovasular Health Study: Frailty and ativation of the inflammation and oagulation systems with and without linial omorbidities: Results from the Cardiovasular Health Study. Arh Intern Med 162: , Leng S, Chaves P, Koenig K, Walston J: Serum interleukin-6 and hemoglobin as physiologial orrelates in the geriatri syndrome of frailty: A pilot study. J Am Geriatr So 50: , Walston J, Fedarko N, Yang H, Leng S, Beamer B, Espinoza S, Lipton A, Zheng H, Beker K: The physial and biologial haraterization of a frail mouse model. J Gerontol A Biol Si Med Si 63: , Ko F, Yu Q, Xue QL, Yao W, Brayton C, Yang H, Fedarko N, Walston J: Inflammation and mortality in a frail mouse model. Age (Dordr) 34: , Himmelfarb J, Stenvinkel P, Ikizler TA, Hakim RM: The elephant in uremia: Oxidant stress as a unifying onept of ardiovasular disease in uremia. Kidney Int 62: , 2002

7 Clin J Am So Nephrol :,, 2013 Ethial Considerations for Frail Patients Faing Dialysis, Swidler Stenvinkel P: Interations between inflammation, oxidative stress, and endothelial dysfuntion in end-stage renal disease. J Ren Nutr 13: , Fried LP, Hadley EC, Walston JD, Newman AB, Guralnik JM, Studenski S, Harris TB, Ershler WB, Ferrui L: From bedside to benh: Researh agenda for frailty. Si SAGE KE 2005: pe24, US Renal Data System (USRDS): Atlas of End-Stage Renal Disease in the United States. Bethesda, MD, National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases, 2004, p Painter P, Kuskowski M: A loser look at frailty in ESRD: Getting the measure right. Hemodial Int 17: 41 49, Lorenz KA, Lynn J, Dy SM, Shugarman LR, Wilkinson A, Mularski RA, Morton SC, Hughes RG, Hilton LK, Maglione M, Rhodes SL, Rolon C, Sun VC, Shekelle PG: Evidene for improving palliative are at the end of life: A systemati review. Ann Intern Med 148: , Swidler MA: Palliative Care and Geriatri Treatment of Patients with Advaned Chroni Kidney Disease. NephSAP Geriatri Nephrology, Washington, DC, Amerian Soiety of Nephrology, Martohia A, Frugoni P, Indiano I, Tafaro L, Comite F, Amii A, Caiafesta M, Marigliano V, Falashi P: Sreening of frailty in elderly patients with disability by the means of Marigliano- Caiafesta polypathology sale (MCPS) and Canadian Study of Health and Aging (CSHA) sales. Arh Gerontol Geriatr 56: , Daniels R, van Rossum E, de Witte L, Kempen GI, van den Heuvel W: Interventions to prevent disability in frail ommunity-dwelling elderly: A systemati review. BMC Health Serv Res 8: 278, Jassal SV, Chiu E, Li M: Geriatri hemodialysis rehabilitation are. Adv Chroni Kidney Dis 15: , Jassal SV, Chiu E, Hladunewih M: Loss of independene in patients starting dialysis at 80 years of age or older. NEnglJMed 361: , Tamura MK, Tan JC, O Hare AM: Optimizing renal replaement therapy in older adults: A framework for making individualized deisions. Kidney Int 82: , Swidler MA: Geriatri renal palliative are. J Gerontol A Biol Si Med Si 67: , HD Mortality Preditor. Available at: alulators/sq. Aessed November 29, Beddhu S, Bruns FJ, Saul M, Seddon P, Zeidel ML: A simple omorbidity sale predits linial outomes and osts in dialysis patients. Am J Med 108: , Couhoud C, Labeeuw M, Moranne O, Allot V, Esnault V, Frimat L, Stengel B; Frenh Renal Epidemiology and Information Network (REIN) registry: A linial sore to predit 6-month prognosis in elderly patients starting dialysis for end-stage renal disease. Nephrol Dial Transplant 24: , Moss AH, Ganjoo J, Sharma S, Gansor J, Senft S, Weaner B, Dalton C, MaKay K, Pellegrino B, Anantharaman P, Shmidt R: Utility of the surprise question to identify dialysis patients with high mortality. Clin J Am So Nephrol 3: , Fried LP, Ferrui L, Darer J, Williamson JD, Anderson G: Untangling the onepts of disability, frailty, and omorbidity: Impliations for improved targeting and are. J Gerontol A Biol Si Med Si 59: , Rakowski DA, Caillard S, Agodoa LY, Abbott KC: Dementia as a preditor of mortality in dialysis patients. Clin J Am So Nephrol 1: , Jonsen AR, Siegler M, Winslade WJ: Clinial Ethis, New York, MGraw-Hill Medial, Lo B: Resolving Ethial Dilemmas: A Guide for Cliniians, Philadelphia, Wolters Kluwer Health/Lippinott Williams & Wilkins, Kaufman SR, Shim JK, Russ AJ: Old age, life extension, and the harater of medial hoie. J Gerontol B Psyhol Si So Si 61: S175 S184, Fried TR, O Leary J, Van Ness P, Fraenkel L: Inonsisteny over time in the preferenes of older persons with advaned illness for life-sustaining treatment. J Am Geriatr So 55: , Miura Y, Asai A, Matsushima M, Nagata S, Onishi M, Shimbo T, Hosoya T, Fukuhara S: Families and physiians preditions of dialysis patients preferenes regarding life-sustaining treatments in Japan. Am J Kidney Dis 47: , Shalowitz DI, Garrett-Mayer E, Wendler D: The auray of surrogate deision makers: A systemati review. Arh Intern Med 166: , Murtagh FE, Marsh JE, Donohoe P, Ekbal NJ, Sheerin NS, Harris FE: Dialysis or not? A omparative survival study of patients over 75 years with hroni kidney disease stage 5. Nephrol Dial Transplant 22: , Nelson JE, Cox CE, Hope AA, Carson SS: Chroni ritial illness. Am J Respir Crit Care Med 182: , Cohen LM, Germain M, Poppel DM, Woods A, Kjellstrand CM: Dialysis disontinuation and palliative are. Am J Kidney Dis 36: , Murtagh F, Cohen LM, Germain MJ: Dialysis disontinuation: Quo vadis? Adv Chroni Kidney Dis 14: , Murtagh FE, Addington-Hall JM, Edmonds PM, Donohoe P, Carey I, Jenkins K, Higginson IJ: Symptoms in advaned renal disease: A ross-setional survey of symptom prevalene in stage 5 hroni kidney disease managed without dialysis. J Palliat Med 10: , Gabbay E, Meyer KB, Griffith JL, Rihardson MM, Miskulin DC: Temporal trends in health-related quality of life among hemodialysis patients in the United States. Clin J Am So Nephrol 5: , Mujais SK, Story K, Brouillette J, Takano T, Soroka S, Franek C, Mendelssohn D, Finkelstein FO: Health-related quality of life in CKD Patients: Correlates and evolution over time. Clin J Am So Nephrol 4: , Abdel-Kader K, Unruh ML, Weisbord SD: Symptom burden, depression, and quality of life in hroni and end-stage kidney disease. Clin J Am So Nephrol 4: , Murtagh FE, Sheerin NS, Addington-Hall J, Higginson IJ: Trajetories of illness in stage 5 hroni kidney disease: A longitudinal study of patient symptoms and onerns in the last year of life. Clin J Am So Nephrol 6: , Carson RC, Juszzak M, Davenport A, Burns A: Is maximum onservative management an equivalent treatment option to dialysis for elderly patients with signifiant omorbid disease? Clin J Am So Nephrol 4: , Morton RL, Snelling P, Webster AC, Rose J, Masterson R, Johnson DW, Howard K: Fators influening patient hoie of dialysis versus onservative are to treat end-stage kidney disease. CMAJ 184: E277 E283, Morton RL, Tong A, Howard K, Snelling P, Webster AC: The views of patients and arers in treatment deision making for hroni kidney disease: Systemati review and themati synthesis of qualitative studies. BMJ 340: 112, Murray MA, Brunier G, Chung JO, Craig LA, Mills C, Thomas A, Staey D: A systemati review of fators influening deisionmaking in adults living with hroni kidney disease. Patient Edu Couns 76: , Shell JO, Patel UD, Steinhauser KE, Ammarell N, Tulsky JA: Disussions of the kidney disease trajetory by elderly patients and nephrologists: A qualitative study. Am J Kidney Dis 59: , Davison SN: End-of-life are preferenes and needs: Pereptions of patients with hroni kidney disease. Clin J Am So Nephrol 5: , Williams AW, Dwyer AC, Eddy AA, Fink JC, Jaber BL, Linas SL, Mihael B, O Hare AM, Shaefer HM, Shaffer RN, Trahtman H, Weiner DE, Falk AR; Amerian Soiety of Nephrology Quality, and Patient Safety Task Fore: Critial and honest onversations: The evidene behind the Choosing Wisely ampaign reommendations by the Amerian Soiety of Nephrology. Clin J Am So Nephrol 7: , Holley JL: Advane are planning in CKD/ESRD: An evolving proess. Clin J Am So Nephrol 7: , Detering KM, Hanok AD, Reade MC, Silvester W: The impat of advane are planning on end of life are in elderly patients: Randomised ontrolled trial. BMJ 340: 1345, Kirhhoff KT, Hammes BJ, Kehl KA, Briggs LA, Brown RL: Effet of a disease-speifi advane are planning intervention on endof-life are. J Am Geriatr So 60: , International Patient Deision Aid Standards (IPDAS). Available at: Aessed November 26, Barry MJ, Edgman-Levitan S: Shared deision making pinnale of patient-entered are. N Engl J Med 366: , 2012

8 8 Clinial Journal of the Amerian Soiety of Nephrology 68. Staey D, Bennett CL, Barry MJ, Col NF, Eden KB, Holmes-Rovner M, Llewellyn-Thomas H, Lyddiatt A, Légaré F, Thomson R: Deision aids for people faing health treatment or sreening deisions. Cohrane Database Syst Rev 10: CD001431, Murray MA, Thomas A, Wald R, Martiorena R, Donnelly S, Jeffs L: Exploring the impat of a deision support intervention on vasular aess deisions in hroni hemodialysis patients: Study protool. BMC Nephrol 12: 7, Ottawa Hospital Researh Institute: Ottawa Health Researh Institutes A to Z Inventory of Deision Aids. Available at: deisionaid.ohri.a/index.html 71. Murtagh FE, Murphy E, Shepherd KA, Donohoe P, Edmonds PM: End-of-life are in end-stage renal disease: Renal and palliative are. Br J Nurs 15: 8 11, Davison SN: Integrating palliative are for patients with advaned hroni kidney disease: Reent advanes, remaining hallenges. J Palliat Care 27: 53 61, Swidler M: Geriatri renal palliative are is oming of age. Int Urol Nephrol 42: , 2010 Published online ahead of print. Publiation date available at www. jasn.org.