INTRODUCTION — This topic reviews major dialysis concerns that are specific to older patients, including decisions regarding initiation, modality, vascular access, and targets for the amount of dialysis that should be provided. We also discuss the prognosis of older patients on dialysis.
Issues surrounding kidney transplantation among older adults are discussed elsewhere. (See "Kidney transplantation and the older adult patient".)
Conservative care among older adults with end-stage kidney disease (ESKD) is discussed elsewhere. (See "Kidney palliative care: Conservative kidney management".)
OVERVIEW — Issues that are addressed in all patients with progressive chronic kidney disease (CKD) include:
●Whether to pursue any form of kidney replacement therapy (KRT; ie, including dialysis and transplantation)
●What modality of KRT is most suitable and preferred by the patient and caregivers
●When to start dialysis
●If starting hemodialysis, what type of vascular access is most suitable and when to place it
These fundamental issues are different for some older individuals, particularly those with extensive comorbidities for whom dialysis may not provide a significant benefit; may be inconsistent with their goals; and may be more difficult to implement without compromising of quality of life.
EPIDEMIOLOGY
Overall incidence — Information from the United States Renal Data System (USRDS) database shows that the adjusted incidence of dialysis initiation among older adults (≥75 years of age) declined for several years to approximately 1450 per million persons in 2020 but increased slightly to approximately 1580 per million persons in 2021; this is compared with a peak of approximately 1900 per million persons in 2009 [1].
Older adults may be less likely to be started on dialysis compared with younger patients with a similar degree of kidney dysfunction. This was suggested by a community-based cohort study of over 1.8 million adults in Alberta, Canada [2]. At a follow-up of 4.4 years, among patients who had an estimated glomerular filtration rate (eGFR) of 15 to 29 mL/min/1.73 m2 at the beginning of the study, the rate of untreated kidney failure (defined as eGFR <15 mL/min/1.73 m2) was over fivefold greater for patients ≥85 years of age compared with those 18 to 44 years. The individual reasons for withholding dialysis cannot be determined from this study [3]. On the other hand, in the United States in 2015, the mean eGFR at initiation of dialysis is very similar for those ≥75 years of age (10.4 mL/min/1.73 m2) compared with those in the 65 to 74 and 45 to 64 age groups (10.2 and 10.0 mL/min/1.73 m2, respectively) [4]. In 2018, patients starting dialysis in the United States aged 75 years and older were only slightly more likely to have an eGFR 10 mL/min/1.73 m2 or above compared with those 65 to 74 years.
Selected modality — In the United States, most older patients (≥75 years of age) with end-stage kidney disease (ESKD) use in-center hemodialysis [4]. In 2021, the percentages of incident dialysis patients ≥75 years of age who started in-center hemodialysis and home dialysis were 89 percent and 11 percent, respectively [1]. The vast majority of older patients (97 percent) starting on a home dialysis modality used peritoneal dialysis rather than home hemodialysis.
MANAGEMENT
Decision to initiate maintenance dialysis — The decision to initiate maintenance dialysis should be discussed with all patients with progressive chronic kidney disease (CKD) regardless of age, although these discussions with older adults often have a different focus than with younger patients. The decision to initiate dialysis should be shared between the clinician and the patient (and caregivers, if appropriate). (See "Kidney palliative care: Ethics", section on 'Education and communication with the patient'.)
The timing of the discussions regarding dialysis depends on the rate of decline in kidney function. Dialysis should be discussed early in the course of CKD in patients with rapidly deteriorating kidney function so that an appropriate access for dialysis may be placed in time to be ready for use. For patients who choose hemodialysis, a arteriovenous vascular access should be placed at least three to six months before anticipated use (see 'Timing of access placement' below). For patients who choose peritoneal dialysis, a peritoneal dialysis catheter should be placed at least 10 to 14 days before use, unless the dialysis center is equipped to perform urgent-start peritoneal dialysis. (See "Placement of the peritoneal dialysis catheter" and "Urgent-start peritoneal dialysis".)
Among patients with a slower decline in estimated glomerular filtration rate (eGFR), we generally initiate the discussion when the eGFR declines below 20 to 30 mL/min/1.73 m2. Initial early discussions should simply introduce the possible need for dialysis (or transplantation) at some point in the future. As the eGFR declines, more focused discussions should occur over multiple office visits. These discussions should include dialysis modalities, timing, and the option of conservative care without dialysis. Options for kidney transplantation may also be discussed. (See "Kidney transplantation and the older adult patient".)
In order for the patient to make a decision regarding dialysis, they need to know what dialysis may provide in terms of quantity and quality of life. The clinician should realistically characterize (to the extent possible) the prognosis, including risk of hospitalization, life expectancy, and quality-of-life expectations with and without dialysis, as well as details of the extensive time commitment required for dialysis. Prognostic models have been developed to estimate both short-term (three- to six-month) and longer-term (more than one year) mortality in various populations [5-13]. These may be useful for guiding discussions with some patients who desire rough estimates of mortality risk over specific timeframes, although the imprecise predictions of these models must be acknowledged [5-10] (see 'Prognosis' below). Newer models using machine learning have been developed; whether these prove to be clinically useful remains to be seen [14].
Not all older patients are good candidates for dialysis. Dialysis is generally less likely to provide the same benefit for older patients as it does for younger individuals. Any anticipated extension of life expectancy among older adults needs to be balanced with the need for access-related procedures and a likely increase in hospitalization risk. (See 'Prognosis' below.)
Quality of life may deteriorate after initiation of dialysis. Hospitalizations tend to increase among older patients who start dialysis, and, even in the absence of hospitalizations, the hemodialysis procedure itself may cause fatigue and malaise that is accentuated and lasts longer among older patients compared with younger individuals.
However, many older patients benefit from dialysis. Such patients tend to have fewer comorbidities and good quality of life at baseline.
Patients generally respond to questions regarding dialysis in one of three ways:
●Some will choose to go ahead with dialysis.
●Some will choose not to be dialyzed and elect conservative care. (See "Kidney palliative care: Conservative kidney management".)
●Many older patients are ambivalent toward dialysis and will ask that they be dialyzed only if and when they have no alternative but death.
It is important to remember that the patient may change their mind at subsequent visits, and we almost always discuss dialysis during multiple visits over time regardless of the patient's initial decision.
Regardless of the decision regarding dialysis, all patients should be followed closely by the nephrologist. In addition, particularly among patients who do not wish to be dialyzed or who are ambivalent about starting dialysis, we make every effort to maintain a good nutritional status and preserve kidney function by carefully controlling blood pressure, appropriately using angiotensin-converting enzyme (ACE) inhibitors (or angiotensin receptor blockers [ARBs]) and sodium-glucose cotransporter 2 inhibitors, and avoiding nonsteroidal antiinflammatory drugs (NSAIDs) and contrast agents, if possible. Blood pressure control goals should be individualized (see "Goal blood pressure in adults with hypertension", section on 'Older adults'). Some clinicians stop or reduce the dose of ACE inhibitors or ARBs at the very late stages of CKD to allow improvement of GFR, particularly in patients with relatively minimal proteinuria. Specific recommendations for conservative care are provided elsewhere. (See "Kidney palliative care: Conservative kidney management", section on 'Medical management'.)
Choice of modality — Dialysis modality (ie, hemodialysis versus peritoneal dialysis) should be selected based on comorbidities, patient preference and capability (as well as social support), and occasionally goals of care but not on outcomes, which have not been shown to be different between modalities [15,16]. Most older patients initiate dialysis with in-center hemodialysis. (See 'Selected modality' above.)
●Based on comorbidities – Peritoneal dialysis may be preferred in older patients with significant heart failure or pulmonary hypertension because it allows daily ultrafiltration and avoids risks associated with vascular access required for hemodialysis. However, peritoneal dialysis may not be possible for some older patients, such as those who have had extensive abdominal surgery. Patients should be individually evaluated for medical indications or contraindications based on comorbidities and levels of social support.
In general, the medical indications or contraindications to hemodialysis or peritoneal dialysis are not different for older adults as for younger patients with CKD, although comorbidities tend to be more common among older adults. (See "Evaluation and management of heart failure caused by hemodialysis arteriovenous access", section on 'Prevention' and "Evaluating patients for chronic peritoneal dialysis and selection of modality", section on 'Physical, cognitive, or psychological impairment' and "Pulmonary hypertension in patients with end-stage kidney disease", section on 'Management'.)
●Patient preference, capability, and goals of care – If comorbidities allow for either modality and both hemodialysis and peritoneal dialysis are available, then the modality should be selected based on patient preference (and, when appropriate, preference of cohabiting individuals if the patient is likely to need help performing peritoneal dialysis at home).
Some older patients benefit from the social nature of being at a hemodialysis facility, while others prefer to avoid the need to travel to and from a facility and prefer to dialyze at home.
Furthermore, if goals of care include minimizing the impact of dialysis on lifestyle, then peritoneal dialysis may be preferred. Patients who initiate dialysis with significant residual kidney function often require less dialysis. Among individual patients, it is easier to factor in residual kidney function to the prescription of peritoneal dialysis (also called incremental dialysis) than it is for in-center hemodialysis, which can be constrained by scheduling of shifts. For many patients, less dialysis is less burdensome.
However, peritoneal dialysis requires reasonable cognitive function and patient mobility and may require significant support at home. Assisted peritoneal dialysis, if available, may be an option for older patients who prefer peritoneal dialysis but are unable to perform the procedure without assistance and who lack sufficient support at home. With assisted peritoneal dialysis, a hired caregiver provides support. While some studies have suggested that assisted peritoneal dialysis may provide better quality of life than in-center hemodialysis for older frail patients, others have not [17-22].
As discussed below, incremental peritoneal dialysis and twice weekly hemodialysis have also been used in older individuals to reduce the burden of dialysis while providing some solute clearance and volume removal to help manage symptoms. (See 'Treatment targets and incremental dialysis' below.)
●Outcomes – Modality is not selected on the basis of improved survival. There are no conclusive data that suggest benefit in terms of mortality or technique survival with either peritoneal dialysis or hemodialysis among older patients [23]. Retrospective studies including adult patients of all ages with CKD have shown mortality to be better, worse, or similar with peritoneal dialysis compared with hemodialysis [24-34]. The interpretation of these studies is limited by selection bias; more functional patients tend to be selected for peritoneal dialysis.
In addition, few studies have focused on quality of life [15] or controlled for frailty, which is an important predictor of outcome with any dialysis modality [35]. (See 'Prognosis' below.)
Timing of initiation — As for all patients, dialysis should be initiated based on clinical indications (ie, refractory volume overload, symptoms of uremia) rather than a specific level of kidney function. (See "Indications for initiation of dialysis in chronic kidney disease".)
Among older patients, it is important to try to distinguish the symptoms of uremia from those of aging (particularly fatigue, depression, and decreased appetite) and medication side effects. This is particularly important for patients who are ambivalent about starting dialysis since such symptoms related to aging will not improve with dialysis.
Vascular access
Selection of access — Endogenous arteriovenous fistulas (AVFs), arteriovenous grafts (AVGs), and tunneled catheters are all used in older patients for permanent vascular access. The selection of access should be individualized based on life expectancy and comorbidities and in consultation with a vascular surgeon with experience in the creation of hemodialysis vascular access [36].
●For older patients on dialysis who are thought to have a good chance of AVF maturation prior to initiation of hemodialysis and a reasonable life expectancy, AVFs should be the first choice for chronic hemodialysis access [37,38]. Among all patients with CKD and compared with AVGs and with catheters, AVFs are associated with fewer complications and lower mortality [39]. (See "Approach to the adult patient needing vascular access for chronic hemodialysis", section on 'Strategy for lifelong hemodialysis access'.)
●For older patients who have little chance of successful AVF maturation but have a reasonable life expectancy, an AVG is a reasonable choice. (See "Arteriovenous fistula creation for hemodialysis and its complications", section on 'Evaluation and planning'.)
●For older patients with extensive peripheral vascular disease, exhausted access sites, short life expectancy from other comorbidities, or chronic hypotension, a tunneled dialysis catheter is a reasonable choice [40]. AVGs are less likely to be successful in such patients and may be associated with higher risk. (See "Arteriovenous graft creation for hemodialysis and its complications", section on 'Access flow-related problems'.)
One study found that among incident patients on hemodialysis aged 80 years or older, initial placement of a tunneled hemodialysis catheter followed by placement of an AVF within the first year of dialysis was associated with similar mortality compared with initial use of an AVF [39].
It is not clear that the benefits associated with AVFs extend to older patients with a shorter life expectancy [38,41,42]. In an analysis of 115,425 patients ≥67 years of age who were identified from the United States Renal Data System (USRDS), there was no difference in overall mortality between patients who received an AVF for the first predialysis dialysis access (n = 21,436) and those who received an AVG (n = 3472) [43]. When data were stratified by age, AVGs were associated with decreased survival compared with AVFs among patients ages 67 to ≤79 years but not among those ≥80 years.
The success rates of AVFs among older patients vary between centers but are generally lower than adults overall [44-47]. In one report from two teaching hospitals in Canada including 152 accesses in patients older than 80 years, the primary patency rate (ie, with no intervention) was 40 percent at 12 months and 12 percent 36 months [46]. The reported primary patency rate among adults overall is approximately 40 to 80 percent. (See "Primary failure of the hemodialysis arteriovenous fistula", section on 'Incidence'.)
Older adults are less likely than younger people to have blood vessels suitable for successful creation of an endogenous AVF; this is particularly true in females, especially those with diabetes and hypertensive vascular disease [44,45,48-51]. Because AVFs are less likely to be successful, older patients who have an AVF placed as a first access are more likely to initiate dialysis with a catheter than had an AVG been placed [43,52]. In one study of 3418 older patients, those who underwent predialysis AVF creation were more likely to have undergone placement of a catheter for hemodialysis initiation compared with those who underwent AVG creation (46 versus 29 percent) [52]. AVGs are generally preferred to catheters. In the general CKD population, compared with AVFs and AVGs, dialysis catheters are associated with the highest complication and mortality. (See "Approach to the adult patient needing vascular access for chronic hemodialysis", section on 'Strategy for lifelong hemodialysis access'.)
Site of access — The selection of a site for AV access must be individualized using physical exam assessment and ultrasound. (See "Patient evaluation prior to placement of hemodialysis arteriovenous access", section on 'Patient evaluation' and "Patient evaluation prior to placement of hemodialysis arteriovenous access".)
Although a radiocephalic AVF is considered the optimal first choice for most end-stage kidney disease (ESKD) patients, a radiocephalic AVF should not attempted if there is little likelihood of timely, adequate maturation. (See "Approach to the adult patient needing vascular access for chronic hemodialysis", section on 'Chronic hemodialysis access'.)
Among older patients, an upper-arm brachiocephalic AVF may be more likely to mature and remain patent than a forearm radiocephalic AVF [37,46,47,53-55]. In a meta-analysis of dialysis outcomes in older patients, brachiocephalic fistulas were found to have a 12 percent higher one-year patency rate compared with radiocephalic fistulas [47]. A more recent meta-analysis also supported use of brachiocephalic fistulas in older patients reporting higher 12-month primary and secondary patency rates for brachiocephalic compared with radiocephalic AVF. Pooled primary patency rates were 50 and 59 percent for radiocephalic and brachiocephalic AVF, respectively, with secondary patency rates of 65 and 73 percent, respectively [53].
However, vascular steal may be more common following placement of brachiocephalic AVF, with one study reporting an incidence of 10 percent versus 2 percent with radiocephalic AVF in older hemodialysis patients [56]. (See "Arteriovenous fistula creation for hemodialysis and its complications", section on 'Ischemia and other systemic problems'.)
Timing of access placement — We generally try to place an AVF at least three to six months prior to the anticipated start of hemodialysis. This is slightly different than the approach for most patients with ESKD, for whom the minimum time for AVF maturation is one month, but a lead time of 6 to 12 months is recommended. (See "Approach to the adult patient needing vascular access for chronic hemodialysis", section on 'Chronic hemodialysis access'.)
A minimum of three months is often required for maturation among older patients. A study of 17,111 patients older than 67 years of age demonstrated a progressively higher success rate associated with AVFs placed six to nine months before initiation compared with those placed one to three months before initiation [57]. However, there was no further increase in the success rate associated with AVFs placed more than six to nine months prior to hemodialysis. Furthermore, there was a persistent and progressive increase in the number of interventions required to keep the fistula patent associated with increasing time between AVF placement and hemodialysis initiation.
AVGs can usually be cannulated within three to six weeks, and tunneled catheters may be used immediately. (See "Approach to the adult patient needing vascular access for chronic hemodialysis", section on 'Strategy for lifelong hemodialysis access' and "Central venous catheters for acute and chronic hemodialysis access and their management".)
Treatment targets and incremental dialysis — For most older patients, we target standard Kt/V thresholds as are used in the more general dialysis population (ie, 1.4 in order to achieve a minimum of ≥1.2 per session for hemodialysis and 1.8 in order to achieve a minimum of ≥1.7 per week for peritoneal dialysis). (See "Prescribing and assessing adequate hemodialysis", section on 'The optimal amount of dialysis'.)
However, the relationship between Kt/V and mortality has not been well studied among older patients. It is possible that this relationship is different among older patients than the general dialysis population.
We believe that there is a role for a more individualized approach in selected older patients. For instance, we may provide incremental dialysis (ie, shorter or less frequent treatments that do not meet standard Kt/V targets) to patients whose quality of life is impaired by standard hemodialysis or peritoneal dialysis schedules, particularly those who still have significant residual kidney function. The risks and benefits of such an approach should be discussed with the patient if such an approach is taken. In one study, utilization of incremental dialysis allowed a delay in the start of full dialysis for approximately one year, without an increase in mortality [58]. On the other hand, another study found a higher mortality among patients starting incremental hemodialysis among those with the lowest residual kidney function at the start of dialysis [59]. Incremental dialysis may be easier to implement with peritoneal dialysis compared with hemodialysis because of logistic factors around scheduling hemodialysis at outpatient dialysis facilities. (See "Incorporating residual kidney function into the dosing of intermittent hemodialysis", section on 'Which patients may be candidates for incremental hemodialysis?'.)
PROGNOSIS — For many older patients, the prognosis on dialysis is poor [60-64]. In a meta-analysis that included 89 studies and 294,921 older patients (mean age 76.5 years) with end-stage kidney disease (ESKD), one-year survival was only 73 percent; this was just slightly better than the survival for patients who underwent conservative care (71 percent) [63]. However, there was marked heterogeneity between studies that could not be explained by study design, size, the definition of older adult, or the cohort era. Many mortality analyses from the United States exclude patients who did not survive the first 60 to 90 days after dialysis initiation. An analysis of 391 United States Medicare patients (aged 65 years or older) initiating dialysis reported a one-year survival rate of only 45 percent, after taking into account the deaths of patients in the hospital before they established care in an outpatient dialysis center [64]. Of the 391 patients, 23 percent died within the first 30 days of dialysis initiation, and 44 percent died within six months.
The comparison of survival between patients who choose dialysis compared with conservative care has yielded mixed results. The meta-analysis discussed above reported comparable survival between these two groups [63]. However, a subsequent retrospective study compared survival outcomes among patients >70 years who chose conservative care (n = 107) versus dialysis (n = 204) [65]. All patients participated in in-depth discussions regarding kidney replacement therapies (KRTs; including dialysis and transplantation) and conservative care when their estimated glomerular filtration rate (eGFR) fell to <20 mL/min/1.73 m2. Patients who chose conservative care continued to receive full medical treatment and multidisciplinary care, including specialist nurses, dieticians, and social workers. The overall median survival was higher for patients who chose dialysis over conservative care (3.1 versus 1.5 years). The survival advantage conferred by dialysis was substantially reduced, though still statistically significant, among patients >70 years who had cardiovascular or other severe comorbidity. However, there was no difference between groups in survival among patients 80 years or older.
Independent predictors of poor survival include age over 85 years, poor nutritional status, activity of life dependence, late referral for dialysis initiation, inpatient dialysis initiation, and the presence of significant comorbidities, particularly cardiovascular disease [64,66,67].
Prognostic models have been developed to estimate both short-term (three- to six-month) and longer-term (more than one year) mortality in various populations [5-10,68], although most performed more poorly when tested in an external validation cohort [10]. Perhaps the most accurate model was developed using data from the AROii European database, including 11,508 incident dialysis patients, and validated using the Dialysis Outcomes and Practice Patterns Survey [5,10]. Using this risk score, patients older than 80 years start with a 10 and 20 percent one- and two-year mortality. The addition of smoking history, diabetes, cardiovascular disease, body mass index <18.5, and initiation with a catheter increased the one- and two-year mortality risk to 25 and 65 percent, respectively. The addition of a low hemoglobin, high C-reactive protein, and low serum albumin increased the mortality risk further. This and other risk scores may overestimate mortality as they did not include patients who transition to transplantation or peritoneal dialysis, both of which select for healthier individuals [69].
The utility of these risk scores for the care of individual patients remains uncertain. In addition, frailty, which may have prognostic implications that are as important as age, was not fully considered in any of the models [5].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis".)
SUMMARY AND RECOMMENDATIONS
●Management
•Decision to initiate dialysis – The decision to perform maintenance dialysis should be discussed with all older patients with progressive chronic kidney disease (CKD). Among most patients, we generally initiate the discussion when the estimated glomerular filtration rate (eGFR) declines below 20 mL/min/1.73 m2. Our approach takes into consideration the prognosis for individual patients and the benefit, if any, provided by dialysis. (See 'Decision to initiate maintenance dialysis' above.)
•Choice of modality – The selection of chronic hemodialysis access is individualized based on life expectancy and comorbidities:
-For older dialysis patients who are thought to have a good chance of arteriovenous fistula (AVF) maturation prior to initiation of hemodialysis and a reasonable life expectancy, AVFs should be the first choice for access. Among all patients with CKD and compared with arteriovenous grafts (AVGs) and with catheters, AVFs are associated with fewer complications and lower mortality.
-For older patients who have little chance of successful AVF maturation but have a reasonable life expectancy, an AVG is a reasonable choice.
-For older patients with extensive peripheral vascular disease, exhausted access sites, short life expectancy from other comorbidities, or chronic hypotension, a tunneled dialysis catheter is a reasonable choice. In addition, we may use a tunneled catheter in patients who are ambivalent about starting dialysis and are reluctant to undergo any procedures until absolutely necessary to prevent death. (See 'Vascular access' above.)
•Treatment targets – For most older patients, we provide sufficient dialysis to achieve standard minimum Kt/V thresholds. However, we accept lower targets among older patients for whom quality-of-life benefits of shorter or less frequent treatments are more important than the limited mortality benefit resulting from a higher Kt/V. (See 'Treatment targets and incremental dialysis' above.)
●Prognosis – For many older patients, but not all, the prognosis on dialysis is poor. The prognosis is worse for patients who have multiple comorbidities, for whom dialysis may not provide any benefit over that of conservative care. (See 'Prognosis' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟