Key points
-
•
Changes in muscle mass have a large impact on the most used equations for estimated glomerular filtration rate. This has a greater impact on older adults, particularly in those with functional limitations or clinical frailty, compared with the overall population.
-
•
Older individuals are at higher risk of being misclassified as having chronic kidney disease (CKD). Use of definitions that include age-modified thresholds to define CKD and/or the use of age-dependent formulae that include albuminuria assessments are being proposed.
-
•
Older individuals with CKD appear to have an unexpectedly high burden of geriatric syndromes such as frailty, cognitive impairment, functional limitations, depression, incontinence, sensory deficits, malnutrition, and falls.
-
•
A complete evaluation (known as a comprehensive geriatric assessment) is recommended for all older individuals. This should include a detailed medical history (including treatment targets), assessment of the home environment and social support structure, functional history, and assessment of personal values and lifestyle preferences.
-
•
Patients with CKD are at risk of polypharmacy. Polypharmacy in older adults is further complicated by altered pharmacokinetics and pharmacodynamics in the setting of advancing age and presence of CKD.
-
•
Growing evidence indicates that dialysis does not alleviate the signs and symptoms of advanced kidney disease in older adults.
-
•
Most studies have shown that dialysis in older adults is associated only with a modest survival advantage over comprehensive conservative kidney care.
-
•
Several studies have suggested that dialysis initiation may exacerbate or accelerate functional and cognitive decline, resulting in a 1.6-fold higher risk of serious fall injuries and a higher risk of requiring professional nursing care.
-
•
It is important to distinguish between younger and older individuals, robust or frail, and those with complex clustering of multiple chronic diseases while engaging in shared decision-making processes.
Although chronic kidney disease (CKD) is a significant public health issue for patients of all ages, there is a limited body of literature addressing the management of CKD in older adults. Most health care practitioners would agree, however, that caring for older adults with CKD presents unique challenges because of the substantial differences between older and younger patients with this condition. In this chapter, we provide an overview of a broad number of issues related to providing care for older adults with CKD. We have included discussions about the debate rejecting current definitions used to define CKD, detailed information about common geriatric syndromes (e.g., frailty, functional decline, cognitive impairment), and their relationship with kidney disease and treatment. We also introduce the role of palliative dialysis and comprehensive conservative renal care.
Chronic Kidney Disease and Age-Related Changes in Estimated Glomerular Filtration Rate
Evaluation of Kidney Function in Older Adults
Several studies have suggested that GFR declines steadily with healthy aging. This decline is often at a slow rate in the absence of albuminuria. Although normal aging is said to cause fibrotic changes in the kidney, there is insufficient evidence to suggest that pathologic findings parallel changes in kidney function. In an evaluation of 1203 healthy potential kidney donors, Rule and colleagues correlated the degree of kidney fibrosis on histology with GFR, measured using iothalamate clearance, across a number of age groups. This reduction in kidney function appeared to begin in the third decade but was most discernible by the sixth or seventh decade. Fibrosis was more common with advancing age, and the measured GFR (mGFR) fell with age (estimated at 4.6 mL/min/1.73 m 2 per decade in men and 7.1 mL/min/1.73 m 2 per decade in women). There was substantial heterogeneity in mGFR values among patients with similar degrees of fibrosis and of the same age, and the resulting correlation between mGFR and fibrosis was weak. (See Chapter 21 for further discussion on the physiology and pathophysiology of the aging kidney).
Estimates of the prevalence of CKD vary widely from study to study and appear to be highly dependent on the method used to evaluate kidney function and on the population under study. , Several observational studies report a high CKD burden among older individuals. In these studies, authors use traditional eGFR thresholds to define CKD (eGFR <60 mL/min/1.73 m 2 that persists longer than 3 months). Using this definition, CKD prevalence in those aged 65 years and older is approximately 10-fold higher than that seen in those younger than 50 years (40% vs. 4%, respectively). However, recent studies within living kidney donor populations have characterized age-related decline in kidney function seen in the absence of disease. On the basis of these data, it seems likely that CKD prevalence is overestimated in the older age group and increasingly the use of a single GFR threshold across all age groups is being contested.
Current KDIGO guidelines recommend assessment of kidney function using estimating equations (eGFR) in preference to formal assessments. Formal evaluation of kidney function, using inulin or iothalamate, is both impractical and costly in the clinical setting. Similarly, for older individuals, 24-hour urine collections can be cumbersome to collect and are more likely to be unreliable due to inappropriate collection, incontinence, or accidental spills. The widespread use of estimating equations to categorize kidney function is, however, also imperfect in older, frail populations and may contribute to mislabeling of those both with and without kidney “disease.” Current standards require estimating equations to fall within a 30% margin of the measured GFR in ≥90% of estimates. Thus even at baseline, eGFR estimates incorporate a relatively large error margin into the reported eGFR. This error margin is further amplified in older individuals because of non-GFR determinants resulting in a larger risk of both underestimation and overestimation of kidney function. Non-GFR determinants include, among other factors, a low dietary intake of protein and reduced muscle mass (both factors specific to creatinine levels), subclinical inflammation, and a change in body habitus with increased adiposity. Even in the absence of overt disease, older individuals are more likely to be frail or have chronic inflammation arising from normal age-related conditions (sometimes referred to as “inflammaging”). A number of different equations have been proposed to improve detection of kidney disease within an older population ( Table 60.1 ), but no one equation appears to be optimal. Consequently, the inaccuracies around eGFR estimation and the observation that older patients have less albuminuria and slower deterioration rates continue to fuel debate as to whether it is appropriate to label individuals, in their eighth and ninth decades, as having a chronic disease based solely on laboratory criteria.
Table 60.1
Estimated Glomerular Filtration Rate (eGFR) Equations
| Research Group | Endogenous Filtration Marker (Units) | Equation | Development Population | Limitations |
|---|---|---|---|---|
| Cockcroft & Gault, 1976 | Creatinine (mg/dL) | CrCl = ([140– age] × weight)/[72 × S Cr ]) × 0.85 (if female) |
N =
505 (236 for training sample)
whites, 4% women hospitalized pts Age range: 18–92 yr (only 17 patients > 80 yr) |
Values not adjusted for BSA
Developed to predict CrCl instead of GFR Underestimates GFR in older adults, especially at higher GFRs |
| MDRD, 1999 | Creatinine (μmol/L) | eGFR = 175 × (S Cr ) −1.154 × (age) −0.203 × (0.742 if female) × (1.212 if black) |
N =
1628 whites and blacks; 40% women
Nondiabetic pts with CKD Mean age: 51 ± 13 yr (age range, 18−70 yr) |
Tends to underestimate kidney function in those with a normal eGFR
Not validated in patients >70 yr, at extremes of body weight, and in some ethnic groups (e.g., Asian ethnic groups) |
| CKD-EPI, 2009, 2012 | CKD-EPIcr—creatinine (mg/dL) | eGFR = 141 × min(S cr/κ,1 ) α × max(S Cr/κ,1 ) −1.209 × 0.993 age (×1.018 if female) (×1.159 if black) where κ is 0.7 for females and 0.9 for males |
N =
8254 nonblacks and blacks, 44% women
Mean age: 47 ± 15 yr (only 216 patients > 70 yr and 41 > 80 yr) |
Developed in a population with few patients >
70
yr and almost none >
80
yr
Less accurate in men with BMI ≥35 kg/m 2 |
| CKD-EPIcys—cystatin C (mg/L) | eGFR = 133 × min (S cys/0.8,1 ) −0.499 × max (S cys/0.8,1 ) −1.328 × 0.996 age [×0.932 if female] where min indicates the minimum of S Cr/κ or 1 and max indicates the maximum of S Cr/κ or 1 |
N =
5352 nonblacks and blacks; 42% women
Mean age: 47 ± 15 yr (719 patients > 65 yr) |
||
| LMR | Creatinine (μmol/L) |
eGFR = e
X–0.0158 × age + 0.438 × ln(age)
Female: pCr < 150
μmol/L: X = 2.50 + 0.0121 × (150– pCr)
Female: pCr ≥ 150 μmol/L: X = 2.50–0.926 × ln (pCr/150) Male: pCr < 180 μmol/L: X = 2.56 + 0.00968 × (180– pCr) Male: pCr ≥ 180 μmol/L: X = 2.56–0.926 × ln (pCr/180) |
N =
3495 Swedish whites; 47% women
Mean age: 63 yr (age range, 21−86 yr; 756 pts 70−79 yr and 342 patients ≥80 yr) |
Difficult to apply
Overestimation in patients with BMI <20 kg/m 2 , especially males Not validated in other racial and ethnic groups (e.g., black, Asian) Lack of external validation |
| FAS, 2017 , 2016 | Creatinine (mg/dL) | eGFR = 107.3. ( SCr / Q crea ) × 0.988 ( age – 40 ) Equation for age >40 yr, where Q crea = 0.7 mg/dL for females and 0.9 mg/dL for males |
N =
6870 nonblacks; 47% women
Mean age: 74 yr (1764 patients ≥70 yr). |
May perform better in healthy and general population than in the CKD population
Needs further external validation (e.g., blacks and Asians) Clinical utility needs to be confirmed in further studies |
| Cystatin C (mg/L) | eGFR = 107.3. ( S cysC / Q cysC ) × 0.988 ( age – 40 ) where Q cysC = 0.82 mg/L when age 40−70 yr and 0.95 mg/L when age ≥70 yr |
N =
6132 nonblacks; 7% women
Mean age: 58 ± 18 yr (1469 patients ≥70 yr) |
||
| BIS, 2012 | BIS1—creatinine (mg/dL) | BIS1: 3736 × creatinine −0.87 × age −0.95 × 0.82 (if female) |
N =
570 (for equation development; whole cohort,
N =
2073)
White German participants; 43% women Community-dwelling older adults ≥70 yr Mean age: 78 yr (208 patients ≥80 yr) |
Needs further external validation (e.g., blacks and Asians) |
| BIS2—cystatin C (mg/L) | BIS2: 767 × cystatin C −0.61 × creatinine −0.40 × age −0.57 × 0.87 (if female) |
BIS, Berlin Initiative Study equation; BMI, body mass index; BSA, body surface area; CKD-EPI, Chronic Kidney Disease–Epidemiology Collaboration equation; FAS, Full Age Spectrum equation; LMR, revised Lund-Malmö equation ; MDRD, Modification of Diet in Renal Disease Study equation.
U.S. guidelines continue to recommend use of the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine-based equation and advise additional confirmatory testing with cystatin C or a clearance measurement in specific circumstances. European guidelines specific to the older population allow more flexibility and do not recommend one equation over the other. No guideline, at present, distinguishes among different age groups or among robust, vulnerable, and frail individuals. Revisions to U.S. guidelines are anticipated because most laboratories no longer include race in the computation. However, it is unclear if newer guidelines will also include age-related modifications as data about the correlation to clinically important outcomes including identification of those at risk of accelerated cardiovascular morbidity, progression to end-stage kidney failure (ESKF), and/or identification of those at increased risk of death emerge.
Prognostic and Clinical Value of Estimating Kidney Function
Contemporary definitions of CKD stage 1 to 5 do not differentiate between older and younger adults and likely result in numerous older individuals being mislabeled and potentially overtreated for CKD.
In a U.S.-based observational study of 4562 patients older than 65 years of age, Arora and colleagues found that 62% of patients with CKD stage 3b and 69% of patients with CKD stage 4 progressed at an annual rate of <1 mL/min/1.73 m 2 . In the Berlin Initiative Study, individuals aged 70 years or more living in Berlin were followed prospectively for a median of 6 years. They found eGFR decline, in men for example, fell from a median of 3 mL/min/year at age 70 years to 0.75 mL/min/year by age 92 years. Declines in renal function appeared to be nonlinear with slower decline in those reaching the eighth and ninth decades of life. , Similar findings have been reported elsewhere , and likely reflect the fact that relatively few older patients with an eGFR <45 mL/min have albuminuria. , Consequently, inclusion of information about albuminuria is important when interpreting results showing reduced eGFR in older subjects. Albuminuria rates are low among those aged older than 70 years. In the Berlin Initiative Study, the albuminuria rate among 2069 adults 70 years of age and older was 26%. Unlike younger individuals, only a small percentage had macroalbuminuria (3.6% had an albumin-to-creatinine ratio [ACR] ≥300 mg/g). Similar results were found in the National Health and Nutrition Examination Survey (NHANES) III population, where 21% presented with albuminuria at 70 years of age and older and 32.7% at 80 years and older.
The lower rate of kidney decline observed in older individuals, especially those older than 75 years, means that older adults are more likely to die than progress to ESKF, even when eGFR is severely reduced. In a large Veterans Administration cohort, the risk of death exceeded that of initiating dialysis in adults older than 85 years, regardless of eGFR ( Fig. 60.1 ). In a community-dwelling population of 1268 older individuals (mean age, 75 years) with an eGFR <60 mL/min/1.73 m 2 followed for 9.7 years, death was 13 times more likely than progression to ESKF, and 25 times more likely for those between 76 and 85 years of age. Even among those who had more advanced CKD (eGFR <45 mL/min/1.73 m 2 ), the risk of death remained more than sixfold higher than the risk of ESKF.
Baseline estimated glomerular filtration rate (eGFR) threshold below which the risk for end-stage kidney failure (ESKF) exceeded the risk of death for each age group.
From O’Hare AM, Choi AI, Bertenthal D, et al. Age affects outcomes in chronic kidney disease. J Am Soc Nephrol . 2007;18:2758−2765.
Notwithstanding the mortality data, there continues to be an association between the presence of CKD and increased vascular events extending across all age groups. In the Cardiovascular Health Study, a community-based cohort of U.S. adults aged 65 years and older, individuals with CKD had more than twice the prevalence of coronary artery disease compared with those without CKD. , In three large cohorts, the Kidney Early Evaluation Program (KEEP; N = 27,017), NHANES, 1999 to 2006 ( N = 5538), and the Medicare 5% sample ( N = 1,236,946), the prevalence of diabetes in those with CKD ranged from 21.4% to 46.2% and that of hypertension was 90% or higher. High rates of concomitant high cholesterol, coronary artery disease, congestive heart failure, cardiovascular disease, and cancer were also seen in the KEEP and NHANES cohorts.
Role of Clinical Prediction Models in Older Persons
Clinical prediction models to assess the risk of death specifically among older patients with CKD are rare and not widely used. Instead, prediction models that predict the likelihood that an individual will progress to ESKF have been widely incorporated into clinical practice. Using two large Canadian cohorts of patients with CKD stages 3 to 5 defined using eGFR, Tangri and colleagues , developed and subsequently validated in diverse global cohorts (N = 721,357), two risk equations, termed the eight- and four-variable kidney function risk equation (KFRE). Addition of other clinical data, such as comorbidity or previous rate of kidney decline, did not significantly improve the predictive power of the KFRE. The more popular four-variable KFRE uses age, gender, eGFR, and urinary ACR and performs sufficiently well for use in older patients with an eGFR <45 mL/min/1.73 m 2 . One criticism of the four-variable KFRE, particularly in older populations, is that it does not address the risk that an acute illness may be associated with an acute precipitous drop in kidney function. In fact, between 25% and 50% of older individuals starting dialysis do so after an episode of acute kidney injury (AKI). , It has been suggested that future modeling equations also include the patient’s competing risk of death before dialysis requirements, rate of kidney function decline, and information about the AKI risk.
Comorbidities and Geriatric Syndromes Among Older Individuals With Chronic Kidney Disease
“Geriatric syndromes” (GSs) is the term used to describe a group of common health conditions in older people that do not fit into discrete disease categories. These conditions include frailty, functional limitation, falls, depression, polypharmacy, malnutrition, and cognitive impairment. Collectively, the GSs arise from a complex interplay among age-related physiologic changes, chronic disease, and functional stressors in older adults. Each interacts with the other, and often the syndromes co-occur.
Emerging data in older individuals with CKD have suggested an unexpectedly high burden of GSs. It is important for nephrology teams to recognize these geriatric issues for several reasons. The appropriateness and effectiveness of medical interventions, such as dialysis care, are dependent on the overall patient well-being, GSs, and general health. Moreover, from the patient’s perspective, the presence of one or more GSs almost invariably affects their lifestyle, as well as that of their family and social circle. Timely recognition of these syndromes can help patients and families have more realistic expectations while participating in treatment decision making and trigger discussions around goals of care and advance care planning (see Chapter 61 ).
The formal evaluation of a patient, using a geriatric lens to recognize GSs and incorporate findings into clinical care, is referred to as a comprehensive geriatric assessment (CGA). Ideally, the CGA assessment is conducted over several visits, in the home and in the clinic, by a variety of health care professionals in addition to the physician including nursing, physiotherapy, occupational therapy, and social work. Evaluation includes a detailed medical history, such as treatment targets, assessment of the home environment and social support structure, functional history, and an assessment of personal values and lifestyle preferences. This information is then coupled with the medical issues and assessment of strength, balance, sensory function, cognition, depression, nutrition, and communication skills and used to identify potential areas of concern. Multidisciplinary teams use the results of the assessment to develop a coordinated plan that maximizes health while minimizing the impact of geriatric impairments, or the treatments being initiated, on the individual and his or her caregivers. This multidimensional and multidisciplinary approach to GSs has been shown to be effective in the general population and across multiple kidney clinical settings. Awareness of GSs can improve outcomes and well-being, , and routine CGA has been recommended by the American Society of Nephrology, Geriatric Advisory Group, for all older CKD patients.
Frailty
In the clinical realm, frailty is a multidimensional syndrome (or state) associated with increased vulnerability. It often results from growing late-life morbidity combining problems occurring across different domains of daily functioning including physical, sensory, psychological, and social. Frailty has multiple causes and contributors and is characterized by a diminished capacity of the physiologic function to appropriately respond to a stressor, whether physical or psychological. Physical frailty increases the risk of dependency and/or death. , Operational definitions for frailty and how to measure it vary, and a multitude of validated scales have been developed over the past decades. This contributes to much of the variability in estimated prevalence. The sole use of the physical domain as a surrogate measure for the global assessment of frailty, such as the classification by Fried and colleagues, in which definitions are based on five physical components—unintentional weight loss, decreased strength, decreased exercise tolerance, reduced gait speed, and fatigue—may be an inadequate metric for frailty, particularly in CKD. Social determinants of frailty, such as those associated with environmental and mental well-being, are also of importance and may be recognized through more holistic assessments. , The Clinical Frailty Scale is an example of a simple and clinically applicable validated tool that can be used to assess global frailty status. The scale uses several grades of frailty ( Fig. 60.2 ) based largely on the clinical judgment of the physician, making it an easily applied screening instrument to identify those who would most benefit from a formal CGA, which remains the gold standard for frailty assessment.
Clinical frailty scale.
IADLs, Instrumental activities of daily living.
From Rockwood K, Song X, MacKnight C, et al. A global clinical measure of fitness and frailty in elderly people. CMAJ . 2005;173(5):489−495.
Frailty is of clinical importance because it is associated with an increased risk of both morbidity (e.g., falls, hospitalization, dependence, need for long-term care) and mortality. Risks of mortality are doubled in those with frailty compared with nonfrail individuals across the complete spectrum of CKD. For example, in CKD patients enrolled in the MDRD study, physical frailty determined from self-reported questionnaires was associated with a 1.7-fold increase in mortality over time (hazard ratio [HR] 1.7; 95% confidence interval [CI], 1.3−2.3). A special U.S. Renal Data System study also found physically frail patients starting dialysis had a 1.8-fold higher mortality over a 3-year period (HR, 1.8; 95% CI, 1.4−2.2).
The prevalence of frailty appears to increase as kidney function declines , and is estimated to be twofold to fourfold higher in the kidney population, compared with those with normal kidney function. Recent prevalence estimates have suggested that between 40% and 60% of those on dialysis have frailty characteristics. , , Although more prevalent in those with increased age, there is also an unexpectedly high burden of frailty in younger individuals on dialysis. , This high prevalence may be partly explained by concomitant CKD-related conditions such as protein-energy wasting (PEW), anemia, inflammation, acidosis, and hormonal disturbances. There is also an increasing awareness of how kidney care treatments may inadvertently promote the sick role, create barriers to routine participation in exercise programs, and decrease participation in other positive health behaviors.
Frailty is a dynamic process in which transitions across states of frailty are common. While transitions can be bidirectional, changes to a worse state are more common than to a better state. Although characterized by a decline in physical function, including physical activity and exercise capacity, frailty is often also associated with reduced social participation and self-care abilities, depression, and cognitive impairment and should trigger open discussions reviewing the current goals of care, prognosis, and more frequent evaluation of symptoms. To date, there are no simple one-step interventions that reverse frailty, but multidisciplinary care, nutritional supplementation, exercise, and rehabilitation may attenuate the morbidity associated with frailty, particularly when applied early. ,
Functional Impairment
Functional status describes an individual’s ability to perform tasks associated with personal care and maintaining a household. In contrast to leisure activities (e.g., gardening and sports), functional tasks are often regarded as fundamental for daily life, and loss of independence is associated with reduced quality of life (QOL) for patients and their caregivers. Although patients with frailty characteristics have a high burden of functional dependence, both frailty and functional loss may occur independently of each other.
Previous estimates have shown that individuals with CKD stage 3b have a threefold increased risk of developing dependence in daily activities, such as bathing, dressing, and personal care, compared with individuals without kidney impairment. Although causation cannot be proven, several studies have suggested that dialysis initiation may exacerbate or accelerate functional decline ( Fig. 60.3 ) and increase caregiver burden. Most at risk are those with multiple comorbidities, those of older age, and those with functional loss at the time of dialysis initiation. In the Dialysis Outcomes and Practice Patterns Study (DOPPS), 77% of noninstitutionalized patients on maintenance hemodialysis (HD) aged 70 years or older reported dependency, in contrast to 15% of older adults without CKD (≥65 years) in the Cardiovascular Health Study. Similar levels of functional dependence are seen in those maintained on both HD and peritoneal dialysis. ,
Survival and functional status after the initiation of dialysis among nursing home residents.
From Kurella Tamura M, Covinsky KE, Chertow GM, et al. Functional status of elderly adults before and after initiation of dialysis. N Engl J Med . 2009;361:1539.
In the general geriatric literature, functional dependence is recognized as a contributor to subsequent disability, recurrent hospitalization, and increased mortality. Several studies have shown an association between poor mobility status (i.e., requiring assistance with or total dependency for transfers and ambulation) and mortality risk among incident dialysis patients ( Table 60.2 ) and predialysis patients. Functional dependency (increased need for nursing care, less participation in social and societal roles) is associated with a twofold increase in mortality (HR, 2.4; 95% CI, 1.9−2.9) for those with the highest degree of functional loss.
Table 60.2
Association of Functional Impairment at Time of Dialysis Initiation with Adverse Health Outcomes
| Study, Date of Study | Patients | Study Setting | Functional Assessment | Outcome | Association |
|---|---|---|---|---|---|
| Arai et al., 2014 |
N =
202
Mean age, 80 yr |
Single-center study, incident to dialysis
Age ≥75 yr |
Mobility (ability or lack of ability to walk without assistance) | 6-mo mortality | Higher mortality if unable to walk without assistance; 6-mo mortality risk: aHR, 4.9; 95% CI, 1.4-17.1 |
| Couchoud et al., 2009 |
N =
2500
Mean age, 81 yr |
French REIN Registry data, incident to dialysis
Age ≥75 yr |
Mobility (walk without help, assistance with or total dependency for transfers) | 6-mo mortality | Higher mortality if fully dependent for transfers; 6-mo mortality risk: aOR, 1.7; 95% CI, 1.4-2.0 |
| Couchoud et al., 2015 |
N =
12,500
Age (median), 81 yr |
French REIN Registry data, incident to dialysis
Age ≥75 yr |
Mobility (walk without help, assistance with or total dependency for transfers) | 3-mo mortality |
Higher mortality if requiring assistance for transfers; 3-mo mortality risk: aOR, 2.5; 95% CI, 2.1-2.9
Higher mortality if totally dependent for transfers; 3-mo mortality risk: aOR, 6.5; 95% CI, 5.4-7.9 |
| Dusseux et al., 2015 |
N =
8955
Age (median), 78 yr |
French REIN Registry data, incident to dialysis
Age ≥70 yr |
Mobility (walk without help, assistance with or total dependency for transfers) | 3-yr mortality |
Higher mortality if requiring assistance for transfers; 3-yr mortality risk: aOR, 1.7; 95% CI, 1.5-1.9
Higher mortality if total dependency for transfers; 3-yr mortality risk: aOR, 3.0; 95% CI, 2.3-3.8 |
| Glaudet et al., 2013 |
N =
557
Mean age, 71 yr |
French REIN Registry data, incident to dialysis | Mobility (walk without help, assistance with or total dependency for transfers) | 4-yr mortality |
Higher mortality if requiring assistance for transfers; 4-yr mortality risk: aHR, 1.4; 95% CI, 1.2-1.7
Higher mortality if total dependency for transfers; 4-yr mortality risk: aHR, 1.4; 95% CI, 1.0-2.0 |
| Kurella et al., 2007 |
N =
83,996
Mean age, 84 yr |
USRDS registry data, incident to dialysis
Age ≥80 yr |
Mobility (inability to walk or transfer) | Mortality |
Higher mortality if unable to walk or transfer
Overall mortality risk: aRR, 1.5; 95% CI, 1.5-.6 |
| Thamer et al., 2015 |
N =
52,796
Mean age, 77 yr |
USRDS registry data, incident to dialysis
Age ≥67 yr |
Assistance with daily living (requiring assistance with daily living, inability to ambulate, or has an amputation) | 3-mo mortality | Higher mortality if requiring assistance with daily living or for walking; 3-mo mortality risk: aOR, 1.4; 95% CI, 1.3-1.5 |
aHR, Adjusted hazard ratio; aOR, adjusted odds ratio; aRR, adjusted relative risk; USRDS, U.S. Renal Data System.
Many factors contribute to the high rate of functional dependencies among older patients with advanced CKD. High rates of depression, cognitive impairment, poor physical functioning, multimorbidity, and the need for repeated hospitalizations may all contribute to the patient’s reduced overall health and self-care ability. Early identification of functional impairment in CKD patients may provide an opportunity to ensure adequate management of modifiable contributing factors, provide support and referral to geriatric rehabilitation services, and trigger discussions, with the family and patient, around the prognosis and identification of the trajectory of functional decline in older patients starting dialysis with limitations.
Falls
As noted previously, several GSs tend to occur concomitantly. Accidental falls are more common in individuals with other GSs, such as concomitant frailty and functional decline, although this relationship is not absolute. Falls are common among older adults and are a leading cause of injury-related hospitalizations, often with prolonged hospital stays, functional decline, long-term care admission, and mortality. Up to one-third of community-dwelling individuals older than 65 years fall annually, often recurrently. Injurious falls, such as those causing fractures, head injuries, or cuts, occur in one out of five falls, , although usually one fall will trigger a cycle of fear of falling (>25% of cases) and a reduction in activity levels, furthering the risk of subsequent falls and the ability to care for oneself.
Falls are almost twofold more common in the dialysis population, in both those maintained on HD and those on peritoneal dialysis (PD). , Although there are multiple common risk factors for falls ( Table 60.3 ), one of the less appreciated complications of HD that may contribute to falls is the effect of forced immobility. The long time during which patients are sitting relatively still in the dialysis chair may lead to changes in balance, lack of sensory stimulation, and, over time, muscle atrophy. More concerning, however, is the 1.6-fold higher risk of serious fall injuries (i.e., resulting in fractures, head injury, or joint dislocation) that occur in the year after dialysis initiation, compared with the year before dialysis care. Clinical data also support the suggestion that dialysis itself may be causative. In a small single-center study of community-dwelling seniors, falls were more common on dialysis days than on nondialysis days (1.45 vs. 1.35 falls/person, respectively) and, if occurring on dialysis days, threefold more common after dialysis than before dialysis. The association between falls and earlier stages of kidney disease has been less well studied, but some authors have suggested that gait abnormalities, such as shorter stride length and greater time in the stance and double-support phases, are more common with a reduction in eGFR and with acidosis.
Table 60.3
Common Risk Factors for Falls in Older Patients with Chronic Kidney Disease
| Risks | Details | References |
|---|---|---|
| Previous falls | One or more falls during the previous 12 mo | , , |
| Fear of falling | Fearful anticipation of a fall; detectable on screening using the Falls Efficacy Scale (high scores) | |
| Frailty | Predisposed to falls for multiple reasons (see text) | , |
| Balance and gait problems | Increased postural sway, failed walking test; can be exacerbated by arthritic changes or neuropathy | , , |
| Pain | Pain may lead to guarding and abnormal posture due particularly to lower limb and foot pain | |
| Depression and antidepressants | Fall risk worsened by insomnia and sleep fragmentation, impaired attention and psychomotor slowing, and/or exacerbated using sedatives and antidepressants | , |
| Number of prescribed oral drugs | Increased fall risk, particularly with benzodiazepines, antidepressants, β-blockers, other antihypertensive agents, opiate derivatives | , , |
| Cognitive impairment | Falls precipitated by decreased awareness of environmental risks; altered judgment and reduced executive functioning result in less effective compensation for age- or disease-associated changes in gait and balance | , |
| Sensory deficits | Poor vision and/or hearing affect environmental awareness, balance, and compensation strategies | |
| Diabetes | Impaired vision due to retinopathy, peripheral neuropathy, and hypoglycemic episodes increases the risk of falls | , |
| Anemia | May result in low energy, poor exercise tolerance, and postural hypotension | |
| Protein-energy wasting and sarcopenia | Loss of muscle mass and strength results in less effective strategies to compensate and prevent falling; reduced exercise capacity | , |
| HD therapy | Increased risk of transient or rapid fluid and electrolyte shifts, hypotension, arrhythmias; stiffness and muscle weakness exacerbated by forced mobility | |
| PD therapy | Environmental risks—connected to a cycler machine overnight, tubing around the bed may pose trip risk; mechanical factors—changes in the center of gravity due to large volumes of intraabdominal fluid |
HD, Hemodialysis; PD, peritoneal dialysis.
Routine clinical questioning about previous falls and conducting simple tests for gait and balance may be a key strategy to identify at-risk individuals who warrant a more detailed evaluation and referral for a CGA. Multidisciplinary risk assessment, followed by tailored multifactorial interventions, such as exercise, treatment of pain, assistive devices, medication review, and home hazard modifications, may be effective in reducing fall risk.
Depression
Depressive symptoms and clinical depression are common in patients across the complete spectrum of CKD. The prevalence of depression varies, depending on the assessment tools used and the population studied, but, using self-report or screening questionnaires, it is estimated that close to 40% of those undergoing maintenance dialysis have some depressive symptomatology, even if they do not fulfill all criteria for a major depressive episode. Prevalence estimates of major depressive episodes, using formal psychiatrist interview assessments, are lower, ranging from 21% in earlier stages of CKD to 23% in those undergoing maintenance dialysis. These estimates may, however, be influenced by the observation, from a number of studies, that patients on dialysis are often unwilling to undergo formal psychiatric evaluation or treatment. Age-specific data are not available, but depression is often more common in socially isolated individuals and those with functional or cognitive impairment placing frail older individuals at increased risk. In addition, often co-occurring with other GSs, depression can be associated with a high burden of pain, fatigue, poor sleep, pruritus, and nausea.
The recognition of depression in the older individuals is important for several reasons. In addition to the well-recognized impact on QOL, individuals who are maintained on dialysis and are older than 75 years are among those at highest risk of suicide. Furthermore, depression is associated with increased hospitalization and mortality in both CKD and ESKF.
The approach to treating depression in older patients is the same as for younger patients. The mainstay treatments are depression-specific psychotherapies, such as cognitive behavioral therapy or interpersonal therapy, and pharmacologic treatment with an antidepressant.
Several studies have shown cognitive behavioral therapy to be successful in patients with ESKF and, where possible, it should be considered the first-line treatment for patients with mild to moderate depression. However, little is known about the influence of physical diseases, frailty, and cognitive impairment on the efficacy or feasibility of psychotherapy. Despite the high prevalence and clinical implications of depression, many patients remain untreated. Several studies have shown low treatment acceptance and, on average, only one third of patients who may benefit actually receive antidepressant medication.
Pharmacologic therapies have been used in patients with CKD or ESKF; however, there is little information on safety, efficacy, and optimal dosing. , Most antidepressant medications are highly protein bound and are not removed by dialysis. Although they commonly undergo hepatic metabolism, the active metabolites are excreted via the kidney, and consequently dosing should be adjusted to the eGFR level ( Table 60.4 ). On the basis of current data, selective serotonin reuptake inhibitors (SSRIs), such as sertraline, escitalopram, and citalopram, should be considered first-line for pharmacologic treatment for depression in patients with kidney disease. , Tricyclic antidepressants and monoamine oxidase inhibitors (MAOIs) should be avoided in older patients with (or without) kidney disease because of cardiac and central nervous system side effects, including prolonged QTc, arrhythmia, anticholinergic effects, and orthostatic hypotension. Paroxetine may be useful when treating a depressed patient with concomitant intractable pruritus because of its antipruritic actions; drugs such as duloxetine may be prescribed for those with concomitant pain. Treatment doses of duloxetine need to be adjusted and, based on data from a small pharmacokinetic study in HD patients, ideally given at 48-hour intervals. It is important to note that in many regions of the world, this is an off-label prescription, with current U.S. product labeling recommending that the drug not be administered if creatinine clearance (CrCl) is <30 mL/min. Mirtazapine, a noradrenergic and specific serotonergic antidepressant medication, has hypnotic, appetite stimulant, and antipruritic effects and can be used at reduced doses in older kidney patients with concomitant anorexia, sleep disturbance, and/or pruritus.
Table 60.4
Safety, Dosing, and Efficacy of Common Antidepressant Medications used in Chronic Kidney Disease (CKD) and End-Stage Kidney Disease
| Drug Class, Generic | Dose Adjustment in CKD | Comment | Efficacy Studies | Class Adverse Effects , , |
|---|---|---|---|---|
| Selective serotonin reuptake inhibitor (SSRI) | ||||
| Sertraline |
CKD stages 1-4: no dosage adjustment required; 50-200
mg/day
CKD stages 5-5D: start at 25 mg/day, consider decreasing max dose |
<1% excreted unchanged in urine
Pharmacokinetics in CKD are unchanged in single-dose studies, but no published data on multiple dosing Likely not dialyzed |
Three prospective efficacy studies , , : sertraline significantly improved BDI scores |
Common: insomnia, restlessness, nausea, headache, GI upset, sexual dysfunction, activation (mainly with fluoxetine and sertraline)
Rare: SIADH, increased bleeding risk, extrapyramidal symptoms, serotonin syndrome (in combination with other serotonergic drugs), and QT prolongation (seen with dose >40 mg of citalopram) |
| Paroxetine |
IR: 10-40
mg/day
CR: 12.5-50 mg/day |
<2% excreted in urine
Increased plasma concentration found when GFR <30 mL/min If GFR <30 mL/min, start at 10-20 mg/day, increase slowly Likely not dialyzed, decrease total daily dose by 50% |
One study among HD patients ( N = 34) : paroxetine significantly improved depressive symptoms and nutritional markers | |
| Citalopram | No dosage adjustment required |
<15% excreted in urine
Manufacturer does not recommend use if GFR <20 mL/min Dose adjustment normally not required in kidney impairment, but use with caution when GFR <10 mL/min Only 1% of drug removed by HD |
One randomized study (citalopram vs. psychological training) in HD population ( N = 44) : both citalopram and psychological training significantly reduced HADS scores at the end of 3 mo | |
| Escitalopram | No dosage adjustment required |
<10% excreted in urine
Manufacturer does not recommend use if GFR <20 mL/min Likely not dialyzed |
One randomized study (escitalopram vs. placebo) in 58 ESKFESKF patients—escitalopram significantly improved HADS scores compared with placebo | |
| Fluoxetine | No dosage adjustment requiredUse with caution. |
5%-10% excreted in urine
Long half-life. If GFR < 20 mL/min, consider using on alternate days or low dose Likely not dialyzed |
Two small prospective efficacy studies—fluoxetine improved depressive symptoms by more than 25% in 6 HD patients and, compared with placebo, improved depression at 4 wk but not 8 wk in 14 HD patients | |
| Serotonin-Norepinephrine Reuptake Inhibitor (SNRI) | ||||
| Venlafaxine |
Normal dosage: 75-225
mg/day
eGFR 10-70: consider reducing dose by 25%-50%.Dialysis: reduce total daily dose by 50% |
5% excreted in urine,
clearance decrease and half-life prolonged in kidney impairment Manufacturer advises to avoid use if GFR <10 mL/min Accumulation of toxic metabolite can occur, rhabdomyolysis and kidney failure have been reported but rare |
No efficacy data |
Like SSRIs plus increased blood pressure
Liver toxicity seen with duloxetine |
| Duloxetine | No adjustment required if eGFR >30 Recommend starting at 30 mg every other day, titrating slowly to a max 60 mg daily; use not recommended with eGFR <30 | Off-label use in patients with advanced CKD and on maintenance dialysis | No efficacy data | |
| Noradrenergic and Specific Serotonergic Antidepressant | ||||
| Mirtazapine |
No dosage adjustment recommended for eGFR 10-50Start at 3.75 mg and uptitrate to effect, max 45 mg daily
GFR <10: consider dose reduction |
75% excreted unchanged in urine
Clearance reduced by 50% in those with GFR <10 mL/min Has been used to treat pruritus caused by kidney failure |
No efficacy data | Appetite stimulation, weight gain, sedation |
| Norepinephrine-Dopamine Reuptake Inhibitor | ||||
| Bupropion | Consider reduced dose and/or frequency | No efficacy data | Increased risk of seizures, insomnia, anxiety, decreased appetite | |
| Tricyclic Antidepressant | ||||
| Nortriptyline |
No dosage adjustment required
Not recommended for older adults, especially those with CKD |
<5% excreted in urine
All metabolites are highly lipophilic. |
No efficacy data | Anticholinergic effects, orthostatic hypotension, sedation, cardiotoxicity |
| Amitriptyline |
No dosage adjustment required
Not recommended for older adults, especially those with CKD |
<2% excreted in urine | No efficacy data | |
BDI, Beck Depression Inventory; CKD, chronic kidney disease; CR, controlled release; ESKF, end-stage kidney failure; HADS, Hospital Anxiety and Depression Scale; HD, hemodialysis; IR, immediate release; SIADH, syndrome of inappropriate antidiuretic hormone.
Polypharmacy
Polypharmacy is often defined as a medication count of five or more medications. , Because patients with CKD take on average 8 medications, and those on dialysis take between 10 and 12 medications, polypharmacy is possibly one of the largest issues in kidney medicine. In older populations, polypharmacy is associated with increased mortality, adverse drug events and drug interactions, falls and other GSs, low medication adherence, and greater health care costs. Safe medication prescribing in older adults with CKD is complicated by not only the number of medications and presence of comorbidities but also the altered pharmacokinetics and pharmacodynamics in the setting of advancing age and presence of CKD. Pharmacist-led deprescribing interventions are effective and recommended. The combination of age-related changes and CKD places older patients at increased risk of drug accumulation and higher potential for adverse drug effects. , Evidence to guide prescribing is severely limited because a large proportion of clinical trials excluded those older than 65 years, and therefore treatment decisions were often based on evidence extrapolated from other patient groups. As a general guide, clinicians are advised to consider multiple factors listed in Table 60.5 . Drugs that have a narrow therapeutic index and/or high toxicity levels should only be used when other alternatives are not available, and individuals should be monitored carefully. Dialysis patients prescribed opioids, benzodiazepine, gabapentinoids, and baclofen are at risk of dose-dependent increased risks and caution is strongly recommended see Chapter 61 ).
Table 60.5
Key Elements in Considering Risks and Benefits when Prescribing for Older Adults with Kidney Disease
From Rifkin DE, Winkelmayer WC. Medication issues in older individuals with CKD. Adv Chronic Kidney Dis . 2010;17:320−328.
| Risk Considerations |
| Medication-Associated risk |
|
Is the medication cleared in whole or in part by the kidney?
Does the medication have a narrow therapeutic window? Is the medication thought to be of high risk in the general older population or in individuals with comorbidities, similar to those of the patient in question? Does the medication have potential central nervous system effects? Are data available to guide dosing in patients with kidney disease (e.g., pharmacokinetic studies, drug level monitoring)? |
| Patient-associated risk |
|
Is the patient already taking multiple medications?
Does the patient have cognitive dysfunction, poor vision, frailty? Risk of additional medications may be higher in this group. Does the patient have a history of adherence problems, and what would be the consequences of patient-related erratic dosing? |
| Benefit Considerations |
|
What is the population in which this medication has been studied? Does it include older adults or those with kidney disease? Is there observational or clinical trial evidence that benefits extend to those with kidney disease?
If patients with kidney disease have not been studied, does the medication have a track record of safety in postapproval studies? Would benefits of the medication be due to improvement in symptoms or in decrease in risk from asymptomatic disease? What are the patient’s preferences in adding new medications? Does the medication address a problem for which the patient is at significant risk (e.g., cardiovascular disease in patients with kidney disease)? Is the patient likely to accrue significant absolute risk reduction from the new medication? |
Several tools can identify medications in which the risks of use in older adults often outweigh the benefits or identify medications that have been omitted but are likely to benefit the individual. The most widely used tools include The American Geriatrics Society Beers Criteria and the STOPP/START criteria for Potentially Inappropriate Prescribing in Older People. The 2023 AGS Beers Criteria features a list of medications to avoid or adjust dose in older adults, specifically for those with reduced kidney function. Of importance, they recommend increased monitoring for older adults starting SGLT2 inhibitors because of the clinical impact of urogenital infections and ketoacidosis.
Malnutrition and Protein-Energy Wasting
Depending on the definitions used, anywhere between 25% and 75% of older patients maintained on dialysis have evidence of uremic malnutrition, muscle wasting, and body fat loss. In the kidney literature, the term “protein-energy wasting” (PEW) is used in preference to malnutrition because CKD-related factors contribute to the development of wasting, independent of the nutrient intake. A full review of PEW is beyond the scope of this chapter (see Chapter 55); however, it is important to recognize PEW in the older patients because it is associated with weakness, falls, and increased mortality and remains largely unrecognized and undertreated. ,
Appetite loss is relatively common in older patients with ESKF, with an estimated one-third to one-half of patients experiencing a progressive, spontaneous decrease in food intake as CKD progresses. , This change in appetite often drives the initial changes in nutrient intake ; however, the spiral of decline in overall nutrition is compounded by multiple factors common among older individuals ( Fig. 60.4 ). In kidney medicine, the introduction of dietary restrictions affects variety, flavor, and palatability of food and further increases undernutrition. In many cases, the loss of the social pleasures of eating leads to a low QOL, in addition to PEW.
Common factors related to aging and dialysis leading to malnutrition and protein-energy wasting (PEW).
CKD, Chronic kidney disease; IADL, instrumental activities of daily living; QOL, quality of life.
Special attention should be given to nutritional status in older individuals, and personalized support from a kidney dietitian is imperative when PEW is noted—both the subjective global assessment (SGA) and malnutrition inflammation score (MIS) are reliable screening tools.
Interventions to improve PEW are often limited to reducing dietary restrictions and adding oral supplements, with little evidence to support dialytic interventions. Protein intake should not be restricted for elderly patients with CKD because significant weight loss is seen in older people on low-protein diets. Appetite stimulants such as mirtazapine can be tried. Constipation and/or dry mouth should be aggressively treated. Where possible, drugs known to reduce appetite, such as calcium binders and semaglutide, should be avoided.
Cognitive Impairment
Cognitive impairment affects patients negatively by contributing to functional dependence and behavioral symptoms, which in turn have important effects on patient well-being and QOL. Impaired cognition potentially affects multiple areas of patient care including patient adherence to treatment plans and medication and may lead to premature institutionalization and death. Pathophysiology, prevalence, and treatments of neurologic conditions in kidney disease are discussed elsewhere, but it is important to note that the HD itself is associated with metabolic changes, consistent with transient ischemia, and may contribute to cognitive decline. Screening is important for all individuals, and relatively quick tests such as the Mini-Cog (three-word recall and clock drawing) or the Mini-Mental State Examination may be useful in a busy practice. In the dialysis population, administration of the Montreal Cognitive Assessment, preferably within the first 2 hours of starting dialysis, appears to be the most reliable cognitive screening tool. Because cognitive impairment deteriorates over time, particularly after dialysis initiation, it is important to recognize and discuss any features of cognitive impairment that are present and introduce the patient and family to the high likelihood of further progression to dementia. Implementation of adequate social support and planning for future patient care is a critical part of these discussions.
Controversies and Complexity in the Management of Older Patients With Chronic Kidney Disease
Over the past decade, multiple clinical practice guidelines have emerged, creating an evidence-based structure that is currently used to guide clinical CKD care. Clinical practice guidelines help standardize care and ensure a consistent approach, across all patients, that includes the interventions most likely to reduce mortality and cardiovascular events and to slow disease progression. Guidelines incorporate data from large population studies and consequently their strength—the ability to create a simplified model to guide consistent high-quality care—is also, in some respects, their greatest weakness. Individual patients bring their own complexities, values, and goals and, particularly when applied to frail older individuals approaching the end of life, guidelines may lead to care that is poorly aligned with the individual’s needs. Clinicians are encouraged to solicit individual patient goals and values; distinguish between younger and older individuals, robust or frail; and distinguish between those with kidney-limited CKD and those with complex clustering of multiple chronic diseases.
Within the kidney sphere, guidelines have given rise to several discussions around whether adaptation is required when treating an older frail patient. These controversial issues are introduced in the subsequent sections.
Hypertension in the Frail Older Patient with Chronic Kidney Disease: Targeting Blood Pressure
Decades of research have unequivocally shown that antihypertensive treatment reduces the risk of cardiovascular (CV) outcomes in several populations including older and very old patients. Currently, however, there is uncertainty about how strict blood pressure (BP) control should be in older individuals with advanced kidney disease. , Although a discussion of all hypertensive trials is beyond the scope of this chapter and is well covered elsewhere (see Chapter 48 ), data pertinent specifically to older individuals with CKD are included later.
Newer guidelines have incorporated data from the Systolic Blood Pressure Intervention Trial (SPRINT) published in 2015 promoting systolic BP targets <120 mm Hg. SPRINT is a prospective randomized controlled trial, funded by the National Institutes of Health, that has randomized 9361 patients with high CV risk to standard treatment targeting systolic BP <140 mm Hg or to intensive BP-lowering targeting systolic BP <120 mm Hg. SPRINT included adults aged 50 years or older who were at modest risk of vascular disease but who were free of diabetes mellitus and did not have a recent stroke, CV event, or heart failure. SPRINT showed that in this select population, lowering systolic BP to a goal of <120 mm Hg resulted in significantly lower rates of fatal and nonfatal CV events and death from any cause. The benefits of stricter BP goals were also confirmed in a prespecified subgroup analysis of 2636 subjects aged 75 years or older. Of interest, although the overall rate of serious adverse events was no different between groups (38.3% in intensive vs. 37.1% in the control group; P =.25), more patients in the intensive arm developed AKI, syncope, electrolyte abnormalities, and hypotension (but had no injurious falls). However, there is concern that the data from SPRINT may not be applicable to the older, frail patient population with advanced CKD. , , Although a proportion of those aged 75 years or older recruited in SPRINT had elevated creatinine levels (16% had an eGFR <45 mL/min/1.73 m 2 ), few had other features of progressive kidney disease (see earlier discussion on controversies around defining older adults as having a disease based on eGFR thresholds alone). Patients with proteinuria >1 g/day and those with polycystic kidney disease were actively excluded. Unlike many CKD patients, BP was easily controlled with a minimal number of medications. Furthermore, patients with dementia, unintentional weight loss >10%, or residence in a nursing home were excluded, as were individuals with medical conditions limiting life expectancy to <3 years (including cancer).
The benefits of treating hypertension have been confirmed in a meta-analysis incorporating data from almost 45,000 patients across all ages. The evidence for which BP targets to use is less clear. Weiss and colleagues have shown robust evidence that BP targets of less than 150/90 mm Hg led to a reduction in mortality, cardiac events, and stroke but only low to moderate evidence that a BP lower than 140/85 mm Hg was beneficial in adults aged 60 years or older. In addition, they found lower BP targets to be associated with an increase in syncope and medication burden but not higher rates of falls or cognitive decline. Although the rate of adverse effects seemed similar across all BP targets, the impact of an adverse event is likely higher in older patients, particularly those residing in a nursing home, and consequently caution is advised. In 2021, the STEP study group published a trial targeting specifically older patients with hypertension. The group enrolled 8511 patients in either an intensive treatment (systolic BP target of 110 to <130 mm Hg) or a standard treatment (systolic BP target of 130 to <150 mm Hg). The cardiovascular primary outcomes occurred less in the intensive group, but safety and renal outcomes did not differ. In all previously mentioned trials, comorbid patients including those with advanced CKD, frailty, and impaired cognition were excluded. Thus the reader is best advised to integrate data, as they emerge, with the results of a CGA into decision making and adapt treatments to align with individual patient complexity.
The Role of Statins in Octogenarians
CKD is associated with an increased risk of major CV events and mortality and is sometimes considered a coronary risk equivalent to diabetes and hypertension. Although dyslipidemia is an important modifiable risk factor for CV disease in the general population, particularly in those older than 50 years, its role and relevance in the older CKD population are less clear. Current KDIGO guidelines recommend the initiation of statin therapy, either alone or in combination with ezetimibe, for all nondialysis patients aged 50 years or older with an eGFR <60 mL/min/1.73 m 2 , irrespective of age or baseline lipid profile. Much of the evidence is based on high-quality data from the Study of Heart and Renal Protection (SHARP). The SHARP trial, published in 2011, randomized patients with CKD to 20 mg simvastatin plus 10 mg ezetimibe or placebo. Patients were followed prospectively for a median of 5 years, and results showed a lower incidence of cardiac death, stroke, nonfatal myocardial infarction, and cardiac revascularization (11.3% vs. 13.4%, respectively). Numerous participants were aged 70 years and older (28% of SHARP participants), and the effectiveness of the interventions, simvastatin and ezetimibe, was similar. There was little evidence from the data presented that suggested any increased risk of adverse events in these older participants. The dilemma, however, arises from what and who were not included in the study. Many patients were excluded because their physician did not think that participation was appropriate. Many of these may have been older individuals with social or other characteristics not captured in the study data. In an eloquent piece, Butler and colleagues have argued that application of current KDIGO guidelines advocating for the use of statins in those of any age with CKD may not be appropriate. They suggested that the older patients recruited into the study were likely not representative of many of those attending our clinics; they referred specifically to patients with a history of myocardial infarction, cognitive impairment, or functional limitations who would have been excluded but also allude to those with social or psychological frailty. Also, they remind us of the need to elicit treatment preferences and goals, especially because both pill burden and cost are often more important to older individuals than to younger patients, and the impacts of any adverse drug effects are higher.
Contemporary guidelines for the general population, published by the American College of Cardiology and American Heart Association, include age-based criteria while the European Society of Cardiology guidelines recommended use of risk scores that incorporate competing risk. However, KDIGO kidney guidelines do not specify if age adaptations are warranted, and it may be appropriate to take a more cautious approach with both the use of statins and the peak dosage prescribed in older or frail persons in whom life expectancy is short. This issue remains debatable because some authorities still advocate for more aggressive lipid lowering.
While the benefits of lipid lowering have been shown for patients with nondialysis CKD, the benefits do not appear to extend to dialysis patients. Several large randomized trials and high-quality meta-analyses have suggested that statin therapy has little or no effect on cardiovascular outcomes among adults undergoing maintenance dialysis. , Furthermore, emerging evidence has suggested that statins may accelerate vascular calcification in the dialysis population. At present the 2014 KDIGO Lipid Guidelines and 2021 Canadian Cardiovascular Society Guidelines recommend statins not be started in adults receiving dialysis but rather be continued in patients already receiving them at the time of dialysis initiation. Although not supported by evidence or by guidelines, discontinuation of statins may have a role in older individuals maintained on dialysis who present with increasing frailty, multimorbidity, polypharmacy, or malnutrition and those entering the end of life.
Anticoagulation in Atrial Fibrillation
The prevalence of atrial fibrillation (AF) in patients with impaired kidney function is considerably higher than that seen in the general population. , Adults aged 65 to 74 years who are undergoing maintenance dialysis have a prevalence of AF of 13%. This increases to 23% among those 85 years or older. A large proportion of these patients are treated with oral anticoagulants with the intention to reduce the risk of stroke and systemic embolism.
There is increasing evidence that the benefit-risk ratio of anticoagulation with warfarin worsens as kidney disease progresses, and some estimate bleeding risk to be 2 to 3 times higher in those with eGFR <30 mL/min/1.73 m 2 compared with those with normal kidney function. Ongoing physician uncertainty has led to warfarin no longer being the preferred anticoagulant of choice. Direct oral anticoagulants (DOACs) are playing an increasing role, particularly in those with CKD stage 3/4a. According to a cohort of 2453 patients with nonvalvular atrial fibrillation and CKD 3 to 5 (dialysis patients were excluded), bleeding risk is reduced by a third in those taking DOACs compared with warfarin. In a large cohort study of 115,564 patients with incident AF, 3% of 2940 patients with CKD had bleeding with DOAC use and after adjustment for additional comorbidity, bleeding risk approximated that of non-CKD individuals. Current KDIGO guidelines suggest DOAC use in those with early CKD and lower-dose apixaban (2.5 mg orally twice daily) or rivaroxaban (15 mg daily) for those with eGFR <30 mL/min/1.73 m 2 until clinical safety data are available. Apixaban remains the preferred drug for older adults due to its lower risk of major bleeding and gastrointestinal bleeding when compared with other DOACs. Studies comparing outcomes in the dialysis population, with and without anticoagulation, are awaited ( www.clinicaltrials.gov NCT03987711 and NCT03862859), and therefore no recommendation can be made for those maintained on dialysis. Nevertheless, the use of these drugs has become more prevalent, with an estimated 12% of dialysis patients on anticoagulation being treated with DOAC. ,
Antiresorptive Treatment: Bone Health
Bone health remains a challenging aspect of kidney care, amplified in the older population by the attendant high risk of fragility fractures, adynamic bone disease, and polypharmacy. In this section, we review the data (or lack thereof) supporting the use of antiresorptive therapies in older patients who have recently had a fracture.
Fragility fractures, fractures that occur as a result of low-energy trauma such as a fall from standing height or lower, are common in older people and are associated with excess mortality, prolonged rehabilitation, long-term pain, and high medical costs. This is particularly the case with hip fractures, which have a substantial impact on an older individual’s health, QOL, and abilities, with only 50% of patients regaining their prefracture level of mobility and independence. In 2019 there were an estimated 178 million new fractures worldwide (an increase of 33% since 1990). The risk of fracture increases with CKD progression, resulting in risks much higher in older CKD individuals versus those with normal kidney function. Hip fractures occurring in dialysis patients are particularly concerning because the 1-year mortality after hip fracture in the dialysis population is close to 65% versus 20% in the general U.S. population.
Despite these alarming numbers, there are sparse data to guide clinicians providing postfracture care to older patients with CKD. Older patients with CKD stages 4 to 5D have been uniformly excluded from clinical trials evaluating pharmacotherapies. The uniqueness and complexity of CKD in the pathogenesis of fractures, mixed with age-related and postmenopausal causes of bone loss, render this population challenging in terms of diagnosis and management of osteoporosis. Active management of hyperparathyroidism may have an effect on fracture risk but, together with phosphate-lowering therapies, may also negatively affect other predisposing risks such as polypharmacy, malnutrition, and aging itself.
There have been few observational or retrospective studies regarding the use of bisphosphonates or denosumab in patients with CKD stages 4 to 5D. Clinical experience is limited, and the theoretic risks of administering antiresorptives must be carefully weighed against unknown benefits, particularly because bone pathology can be difficult to determine accurately. Widespread use of osteoporosis therapies in patients with advanced CKD and/or evidence of CKD-mineral and bone disorder (MBD) was previously discouraged, especially in frail older individuals. However, there are a few studies proving their harm or benefit and recent data suggest that bisphosphonates might be safe in patients with eGFR of 15-59 mL/min/1.73 m 2 due to age-related decline in kidney function without CKD-MBD. ,
Antiresorptives and other osteoporosis therapies can be safely used in patients with modest reductions in eGFR (≥30 mL/min/1.73 m 2 ), particularly if the parathyroid hormone level lies within the normal to high range. However, for individuals with severe CKD (stages 4−5D) there are limited therapeutic interventions known to reduce fracture recurrence. Furthermore, the theoretic risks of adverse events remain of concern; for example, antiresorptives may worsen low bone turnover, osteomalacia, and mixed uremic osteodystrophy and exacerbate hyperparathyroidism, and denosumab may induce significant hypocalcemia. In the 2017 KDIGO guideline update, treatment recommendations are discretionary and recommended only after “taking into account the magnitude and reversibility of the biochemical abnormalities and the progression of CKD, with consideration of a bone biopsy.” Nonpharmacologic intervention with proven efficacy should always be part of the overall management of osteoporosis and includes vitamin D 3 supplementation, a reduction in alcohol intake, smoking cessation, fall prevention, improved nutrition, and weight-bearing exercise.
Nephrologists should continue to pay careful attention to emerging information in this relatively new field while prospective studies, specifically designed for CKD patients, are performed.
Vascular Access Creation
Vascular access creation in older patients undergoing HD remains a topic of ongoing discussion, particularly because of the need to balance concerns about infection-related morbidity and mortality against any potential residual confounding arising from the observational nature of most studies. There is a clear association with poor functional status and a lower use of AVF, suggesting earlier studies comparing AVFs with arteriovenous grafts (AVGs) or central venous catheters (CVCs) may have overestimated the survival benefits for those dialyzed via an AVF. By comparing outcomes of patients who underwent an unsuccessful fistula attempt with those who did not have fistula creation attempted, it appears that the observed survival gains are attributable to residual confounding factors, arising from the inherent bias that individuals undergoing surgery are likely healthier than those for whom surgery is not attempted. , Furthermore, although the relative gains may be similar regardless of age, the absolute gains, in terms of survival days, are likely to be small because overall survival on dialysis is lower in older individuals. One argument in favor of avoiding CVC use is that Staphylococcal aureus bloodstream infections are known to be 10-fold higher in individuals undergoing HD with central vein catheter use (adjusted rate ratio 6.2; 95% CI = 5.7–6.7 for dialysis catheter vs. fistula). This has been challenged by DeSilva and colleagues, who analyzed data from 115,425 incident U.S. patients on HD aged 67 years and older. They assessed mortality outcomes on the basis of the first vascular access placed but found no difference in mortality for patients with an AVG as the first access compared with those who started with an AVF. Similar results were reported by Drew and colleagues. They used decision analysis to apply current knowledge of the risks and benefits to different access strategies. Although their study has faced criticism for assuming a base where patients initiate dialysis with a CVC, the methodology used is robust, valid, and generalizable. Results show a modest survival benefit for those aged 80 years when patients are dialyzed using a fistula strategy. Maximal life expectancy gains were seen in males and nondiabetic patients. For example, an 80-year-old nondiabetic male had an estimated life prolongation of 8 months with a fistula rather than CVC and 6 months when compared with AVG. Survival, however, for an 80-year-old diabetic female was best with an AVG and, when comparing AVF with CVC, survival differed by only 3 weeks.
Access-related morbidity is also known to be high. Observational studies suggest that up to one-third to one-half of octogenarians undergoing predialysis fistula creation die before they need kidney replacement therapy (RRT). , Fistula failure rates are high in older patients, with some studies suggesting up to 60% of older adults will have a fistula that fails to mature. Of those that do function, several will require multiple interventions, particularly early in the clinical course, causing a high clinical and economic burden when overall survival and gains are limited. Patients undergoing dialysis with an AVF or an AVG are more likely to report pain, bleeding, bruising, and swelling than those with a CVC. Cannulation-related complications are more common, in part due to fragile skin and in part related to the higher use of antiplatelet or anticoagulant medications. Access creation is also associated with reduced grip strength and impaired arm and hand function. In patients with marginal functional status, creation of AVFs or AVGs may affect personal independence and limit dressing, feeding, and other activities of daily living, although bathing and showering were considered more difficult with CVC. Most clinicians would agree it is ultimately important to align the choice of dialysis access to individual patient expectations.
The ideal clinical scenario is when results of the individual’s CGA help inform the choice of vascular access. AVGs in predialysis patients placed closer to the time of HD initiation have been shown to be an effective CVC-sparing strategy and to be less costly than placing an AVF. Because of this, we continue to recommend a patient-first approach that identifies “the right access for the right patient” based on patient preferences, priorities, life expectancy, and QOL.
Care for Older Adults With End-Stage Kidney Disease
Dialysis
The demographics of the dialysis population continue to change with time. Compared with the early 2000s, patients are older and more medically complex with 42% being older than 65 years and 17% older than 75 years. Until recently, the adjusted ESKF prevalence has also been rising, with steeper increases among older age groups compared with those younger than 65 years. These trends, however, have started to change, and data from the U.S. Renal Data System have suggested more conservative dialysis practice patterns, with pronounced declines in the number of older people starting dialysis resulting in 2020 incidence rates being the lowest among those aged 65 years and older since the late 1990s.
Although it is often assumed that treatment with dialysis will extend life and alleviate the signs and symptoms of advanced kidney disease, there is growing evidence that benefits may not always accrue in older adults. Longitudinal studies, across much of Europe, have shown that the symptom burden increases dramatically in the 6 months preceding dialysis initiation, with a corresponding decline in QOL ( Fig. 60.5 ). , The recovery for QOL scores, after dialysis initiation, is, however, modest. Likewise, few individuals report a return to predialysis symptom burden.
Figure showing change in quality of life score in older adults before, and after starting dialysis.
From de Rooij ENM, Meuleman Y, de Fijter JW, et al. Quality of life before and after the start of dialysis in older patients. Clin J Am Soc Nephrol. 2022;17(8):1159–1167.
Life extension is also modest. The median life expectancy after the initiation of chronic dialysis in the United States is comparable with (or worse than) many cancers and estimated to be less than 2 years for patients aged 75 years or older, although considerable heterogeneity exists. The data suggest that up to 25% of those aged 70 years or older die within the first 6 months. Conversely, a similar proportion survive beyond 4 years. , For all ages, the relative median life expectancy after starting dialysis is around 25% to 30% that of age-matched individuals from the general population, and therefore older adults are most likely to have the smallest absolute gain measured in months or years.
Among factors that predict early mortality are the presence of frailty, impaired functional status, low body weight or serum albumin concentration, and need for nursing care at the time of dialysis initiation. , , There are several prognostic tools, , , , , but most are limited to use at the time individuals initiate dialysis and do not help prognosticate survival in those who are already established on dialysis or who opt for a comprehensive conservative kidney care approach. The limits and deficiencies of the most common prognostic tools have been reported elsewhere.
Dialysis Modality Selection
Among patients 65 years of age and older initiating dialysis in the United States, most of them (87%) will start in-center HD as their initial modality choice, followed by PD (11%) and preemptive transplantation (1.6%). Until recently, one of the main arguments favoring HD was based on the overall mortality rates seen in the older patient population treated with PD. In a meta-analysis involving 631,421 older patients, the mortality was higher in those undergoing PD compared with HD (HR, 1.10; 95% CI, 1.01−1.20). However, particularly within the field of geriatric nephrology, it is important to recognize that the flexibility and gentle nature of PD offers much to an older individual, and many individuals forfeit a longer time alive but feeling unwell for a shorter time feeling well. Patients may be attracted to several other aspects of PD, such as increased dietary freedom, decreased hospital visits, control over one’s own care, and a feeling of personal autonomy. In those with a guarded prognosis but high symptom levels from fluid accumulation, PD may be an ideal alternative to comprehensive conservative kidney care because it can be adapted to patients’ needs as they transition along their disease course until they reach end of life. For example, in a patient experiencing increasing pain or deterioration from non–kidney-related comorbidity, dialysis exchange frequency can be reduced and the solution volume and concentration adapted to improve comfort, prevent shortness of breath, and enhance the feeling of control and freedom.
Frail older patients who are unable to perform their own PD exchanges can be successfully supported by assisted PD. , With assisted PD, home care workers, family members, or staff in long-term care facilities are trained to perform the exchanges and supervise the dialysis. The level and tasks for assistance vary and assistance tends to be required long term ( Fig. 60.6 ). , Different models have been used across different countries, and currently neither functional dependence nor cognitive impairment are considered absolute barriers to PD. ,
(A and B) The change in need for assistance over time in older patients maintained on dialysis.
The different shading patterns represent the degree of assistance provided.
From Fonseca-Correa JI, Farragher JF, Tomlinson G, et al. Longitudinal changes in the use of PD assistance for patients maintained on peritoneal dialysis. Kidney360. 2021;2(3):469–476.
In-center HD may be preferred over home-based PD treatment by some individuals because their day can be structured around scheduled and fixed times, facilitating coordination for family members or staff providing support. Others benefit from socialization with staff and other patients at the dialysis facility and from having frequent contact with the dialysis staff to provide both medical and psychosocial support. , HD in frail older patients is, however, more likely to precipitate episodic postural hypotension and possibly accelerate cognitive decline from myocardial and cerebral stunning. Additional challenges include the creation and maintenance of vascular access; problems associated with postdialysis recovery time, such as fatigue; and the risk of falls after dialysis.
As a result, there is no one dialysis modality of choice for the frail older patient. Clinicians are advised to use information from the CGA to build a framework of total care, focusing on the needs of the individual patient and the goals of care in the context of her or his life experience.
Kidney Transplantation
In the United States, patients aged 65 years or older currently comprise 26% of all kidney transplant wait-listed patients; 21% of all recipients are older adults. While in general the number of transplants occurring per year has remained relatively static, there has been a progressive increase in the number of individuals aged 65 years or more undergoing transplant surgery, with the absolute number of older patients increasing almost twofold between 2008 and 2021. Although advanced age is no longer a barrier to kidney transplantation, age remains an important surrogate factor when assessing suitability for transplantation. Chronologic age is less important than the patient’s level of frailty, burden of coexisting conditions, and overall prognosis. , Frailty assessment is increasingly recognized as an essential component of pretransplant assessment to identify and exclude high-risk candidates and allow transplantation to elderly patients regardless of age. , However, the best modality for frailty assessment in this context has not been yet defined. Transplantation assessments are also commonly more rigorous for older candidates when evaluating cardiac and cancer risk.
Transplantation offers survival benefits across all ages, although due to the attendant risks of the surgery itself and side effects associated with immunosuppressive drugs, patients have an increased mortality risk in the initial few months. Survival equivalence, when compared with maintenance dialysis, is seen approximately 8 months after the surgery is performed in younger patients but not seen in those aged 60 years and older until more than 12 months have elapsed from the date of transplantation. In those aged 70 years and older at the time of transplantation, survival advantage may be delayed beyond 21 months post transplantation. Several strategies have been used to maximize kidney transplantation accessibility including the use of expanded criteria donor (ECD) kidneys and pairing organs harvested from older donors with older recipients. These systems offer advantages and disadvantages, but overall the most beneficial and commonly used strategy in North America is to offer ECD kidneys to the older patient.
The most common cause of graft loss in older kidney transplant recipients remains death with a functioning graft. , Death-censored graft failure risk remains lowest, across all age groups, in patients aged 65 years or older likely because of a shorter life span and lower risk of acute rejection. Older patients are at higher risk of medication-related adverse events and higher rates of opportunistic infections and neoplasia. ,
Comprehensive Conservative Kidney Care
Although the change in dialysis utilization is captured across several regions of the world through dialysis registries, little is known about how often patients with advanced stages of CKD choose not to undergo dialysis or are not offered dialysis. Single-center studies from Europe and Australia have suggested that a substantial number of older patients with advanced kidney disease are treated conservatively and do not receive dialysis. Other studies, using administrative data, also found that a notable number of older patients do not undergo dialysis. These studies identified patients with advanced CKD who were likely at risk of dialysis by selecting those who had two or more creatinine levels determined at least 3 months apart that met criteria for a low eGFR. In one study, from Canada, the clinical course of all individuals with creatinine values consistent with CKD were tracked over a 2-year period. In the second study, from the Veterans Administration, detailed data were extracted from electronic chart data and used to identify whether patients either underwent dialysis or were being prepared for dialysis. In both studies, older adults were found to be less likely to undergo dialysis than their younger counterparts, implying that there may be a relatively large “reservoir” of older adults with high creatinine levels (and, by implication, low levels of eGFR) who, for a number of reasons, do not get started on dialysis ( Fig. 60.7 ). Similar patterns have been reported for the incidence of hospitalized patients with AKI treated with dialysis. Although provocative, these data cannot distinguish between those who did not meet clinical criteria or a need for dialysis from those who did not get offered or who chose not to undergo dialysis.
Age differences in treatment decisions and practices for advanced kidney disease.
Based on a chart review of a random sample of patients not enrolled in the U.S. Renal Data System and who did not have a dialysis procedure code in Medicare or Department of Veteran Affairs (VA) administrative files during follow-up.
From Wong SP, Hebert PL, Laundry RJ, et al. Decisions about kidney replacement therapy in patients with advanced kidney disease in the US Department of Veterans Affairs, 2000-2011. Clin J Am Soc Nephrol . 2016;11(10):1825−1833.
Although registry data serve as a valuable source of information about outcomes in older adults with ESKF who are treated with dialysis, less is known about outcomes among those who are not treated with dialysis. What little is known about this group comes from several small, single-center studies that have examined outcomes among patients with advanced kidney disease who were managed with comprehensive conservative kidney care (CCRC) (see also Chapter 61 ).
Most studies have shown that dialysis is associated with a modest survival advantage over CCRC; however, the magnitude of this effect varies across studies, and the potential advantage seems to be lost for a subgroup of older patients with a high burden of comorbidity. The comparisons, however, between patients managed with CCRC and those initiating dialysis are challenging because there is no clear starting point for analysis. It is often difficult to know when dialysis would have been started in those who chose a CCRC pathway, and it is possible that studies suffer from a lead time bias. Robust patients and those with rapid declines in kidney function are more likely to undergo dialysis; those who are frail or have slowly progressing kidney declines are more likely to undergo CCRC. Thus the use of time points determined from the eGFR, to set a time when dialysis may have been otherwise started, is likely to lead to further systematic biases.
Evolving literature has recognized that many older patients may benefit equally from dialysis as from CCRC. Particularly in older adults with a high burden of GSs, small gains in longevity that are potentially achieved with dialysis care need to be balanced against the high risk of increased symptom burden, worsening dependency, and often a negative effect on QOL. , , Overall gains are modest, particularly in those who have recently undergone one or more life-prolonging procedures, such as intubation, feeding tube placement, and cardiopulmonary resuscitation.
Patients who opt for a CCRC pathway tend to fare well for several months. Their symptom burden and QOL remain stable until the last 1 or 2 months of life, at which time palliative care interventions are often beneficial. Functional status is often maintained until the last months of life and, overall, patients are more likely to die in their home environment or in hospice compared with those starting RRT. Further information on the trajectories seen in those initiating dialysis have been emerging, although insufficient information is currently available to make clear distinctions between patient groups.
Importance of Shared Decision Making in End-Stage Kidney Disease
Shared decision making is a critical component of kidney care, particularly for older patients and their families. Shared decision making incorporates discussions about realistic treatment options with patients and their families or caregivers in open, collaborative, and structured environments (see Chapter 61 ). Discussions should be specific to the individual and his or her lifestyle and values, as well as limited to medically appropriate interventions. Ideally, a multidisciplinary kidney team, patient, and family should be present and information should be shared about the potential benefits and harms of various treatment options. Prognosis should be integrated with the results of the CGA, allowing clinicians to share information about possible positive and negative effects on daily living and symptom burden. Information about longevity, the expected changes in symptom burden, and improvement in social functioning can be customized to the individual ( Fig. 60.8 ). This approach, in which patients and families are involved in an individualized management plan, will maximize treatment benefits.



