An approach to treating older adults with chronic kidney disease

• Chronic kidney disease (CKD) in older adults can result from a combination of physiologic, age-related decline in kidney function, renocardiovascular risk factors and external insults.

• The management of risk factors for CKD should be individualized with consideration of less aggressive targets for frail patients at risk of polypharmacy.

• The Kidney Failure Risk Equation is a useful tool in the evaluation of and ongoing treatment for CKD for both general practitioners and nephrologists.

• Clinicians should have open discussions with their older patients with CKD regarding their goals of care and the potential benefits and harms of renal replacement therapy.

Chronic kidney disease (CKD) in older adults is a worldwide epidemic that affects nearly 40% of people aged 65 or older.1 In the United States, CKD in people aged 65 years and older is more common (38%) than in all other age groups.1 Over the last 4 decades, patients older than 75 years are the fastest-growing group to start dialysis.1 In Canada, more than half of patients starting dialysis are 65 years or older.2 Chronic kidney disease confers considerable morbidity on patients and consumes a substantial amount of health care resources.1 Few evidence-based guidelines specifically speak to management of CKD in older adults, many of whom are frail and have multiple comorbidities,3 and physicians should exercise caution when extrapolating general guidelines to older adults. We discuss the causes and consequences of CKD through a geriatric lens, and outlines principles of best practice. The evidence underpinning this review is summarized in Box 1.

What are the causes of kidney dysfunction in older adults?

Chronic kidney disease in patients older than 65 years4 is defined by an estimated glomerular filtration rate (eGFR) of less than 60 mL/min/m².4 However, a physiologic decline in GFR of about 8 mL/min per decade begins around the fourth decade of life.5 Physiologic CKD usually does not lead to end-stage kidney disease (ESKD), which is defined by the development of uremic symptoms or need for renal replacement therapy. A hallmark of age-related kidney decline is the absence of proteinuria,6 which portends a favourable prognosis. The rate of eGFR decline can also be instructive; most older patients have stable eGFR trends,7 and patients with slow eGFR decline (< 2 mL/min/yr) and minimal proteinuria (albumin-to-creatinine ratio [ACR] < 3 mg/mmol8) are unlikely to reach ESKD.7

Older adults are at higher risk of acute kidney injury (AKI) from nephrotoxic medications, volume depletion, urinary tract obstruction and acute illnesses including sepsis and acute cardiovascular events.9 Nonsteroidal anti-inflammatory drug (NSAID) use is a risk factor for AKI and CKD.10 Topical NSAIDs at lower potencies may be safer to use in patients with CKD, although systemic absorption in older frail patients with low muscle mass may occur.11 Finally, systemic diseases can cause kidney dysfunction, including plasma cell dyscrasias (e.g., multiple myeloma), drug-induced interstitial nephritis and anti-neutrophilic cytoplasmic autoantibody vasculitis, the most common glomerulonephritis in older adults.12

How should clinicians investigate kidney dysfunction?

A current (2018) Canadian guideline recommends ordering eGFR and ACR to screen for kidney disease annually in high-risk populations such as those with hypertension, type 2 diabetes mellitus or cardiovascular disease.13 Screening older patients who do not have risk factors is not recommended.13

In a patient presenting with new or worsening kidney dysfunction, additional tests should be ordered as indicated (Table 1). These may include a urinalysis and an abdominal ultrasound. Appropriate medications should be held (Table 2), particularly in the presence of relative hypotension.9 Nephrology consultation should be sought if a readily reversible cause of AKI or CKD is not discovered.

In older patients, physiologic CKD should be distinguished from pathologic causes of CKD, such as systemic diseases (cardiovascular disease, diabetes, hypertension, autoimmune diseases) or external insults. A sudden decline in eGFR (e.g., a more than 1.5-times increase from baseline creatinine) over a short period (days to weeks), especially when associated with hematuria and proteinuria, may be suggestive of glomerulonephritis. Low urine sodium is an excellent test of volume depletion,9 although it can be falsely negative in patients on diuretics. Patients with established CKD should have serum creatinine, eGFR, ACR and electrolytes measured routinely (Table 1). The frequency of investigations is dictated by the stage and stability of the CKD.

Can development of ESKD be predicted?

An important challenge facing physicians is to identify patients at high risk of CKD progression who would benefit from intensive risk factor control. Stage of CKD (Box 2) alone is a poor discriminant of progression. Creatinine-based eGFR using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) model15 has several drawbacks: it is affected by age, sex and muscle mass.15 It also uses a race coefficient, and will eventually be supplanted by a race-independent refitted model that is more accurate,16 a change based on race being recognized as a social rather than biological construct.16

Adding ACR to eGFR improves prognostication15 and has led to a paradigm shift, ushered in by the development of an internationally validated risk prediction model known as the Kidney Failure Risk Equation (KFRE).14 The KFRE predicts 2- and 5-year risk of ESKD using 4 variables: age, gender, eGFR and ACR (online calculator available at https://kidneyfailurerisk.com/). Generalists can use KFRE scores to identify patients with CKD who are at higher risk of progression and who should be referred to nephrology. For example, the Ontario Renal Network suggests that people with a 5-year KFRE score greater than 3%–5% may benefit from a nephrology referral.17

Kidney Failure Risk Equation scores help generalist physicians balance cardiovascular risk factor control with the risk of polypharmacy. The KFRE promotes an individualized approach in frail individuals; those with low-risk KFRE scores may not require strict blood pressure, lipid and glycemic control specifically for renoprotection, thus potentially decreasing polypharmacy, the risk of falls and medication-related adverse effects such as AKI.

The KFRE has limitations: it may be less accurately predictive in patients with glomerulonephritis, or hereditary or structural diseases such as polycystic kidney disease.14 However, these conditions are less prevalent in older populations. The KFRE should not be used in patients with AKI or fluctuating eGFR.14

What are the risk factors for progressive CKD?

Factors associated with an accelerated rate of GFR decline include hypertension, hyperglycemia, atherosclerotic cardiovascular disease, smoking and previous episodes of AKI.18

Blood pressure (BP), lipid and blood glucose control are the cornerstones of CKD management.18 Blood pressure control and albuminuria reduction are reno-protective; renin–angiotensin–aldosterone system (RAAS) blockade and sodium–glucose cotransporter 2 (SGLT2) inhibition are important for achieving these goals.19 However, guidelines do not agree regarding the optimal BP targets for older patients. Hypertension Canada (2020) recommends a target BP goal of less than 130/80 in patients with CKD and diabetes mellitus, and less than 120/80 in patients with CKD without diabetes mellitus, irrespective of age.19 These recommendations are based primarily on analysis of the Systolic Blood Pressure Intervention Trial (SPRINT), which included older patients, and the Hypertension in the Very Elderly Trial (HYVET), which included patients older than 80 years.20^,21 The Kidney Disease: Improving Global Outcomes (KDIGO) international working group’s 2021 guidelines on hypertension recommended intensive BP lowering (systolic < 120 mm Hg), irrespective of diabetes status and age.22 Caution must be exercised in older adults; KDIGO guidelines acknowledge the limitations of evidence for BP lowering in patients older than 75 years.22 Individuals with substantial frailty who have a higher risk of postural hypotension and falls may benefit from more liberal BP targets.20^,21 Post hoc analysis of the SPRINT trial showed a higher risk of adverse events in patients with lower diastolic BP.23

Similarly, glycemic control should be based on patients’ characteristics. The 2018 Diabetes Canada guideline recommends that glycosylated hemoglobin (HbA_1c) targets should be adjusted according to the level of frailty: aiming for HbA_1c 7.1%–8.5% in patients with dementia is a reasonable strategy, with a focus on avoiding extremes in blood glucose levels.24 For most other older patients, the target should be HbA_1c 7.5%–9.0%.25

Statin use in older patients reduces cardiovascular events and mortality, but this effect is attenuated or even lost in those older than 85 years.26 If myopathies are identified, lower statin dosing should be considered.26 Given the potential musculoskeletal adverse effects of statins, it may be prudent to defer statin therapy in patients with poor nutrition who are already at risk of sarcopenia and falls.26

Special considerations for older adults with CKD

Conclusion

Chronic kidney disease in the older-adult population is associated with substantial morbidity.1 A patient-centred approach is necessary to balance the benefits of cardiovascular optimization with the risks of adverse events and polypharmacy. The KFRE is a valuable prognostic tool that can help physicians identify high-risk patients who warrant more intense management.14 Management of older patients must incorporate geriatric competencies. By appreciating how CKD affects older patients, primary care providers and nephrologists can pursue a nuanced and holistic approach to care, while respecting patients’ goals and values.43 Future research should address gaps in knowledge about optimal blood pressure targets, symptom management and best care for older patients with end-stage renal disease who decide not to start dialysis (Box 4).