2024 年 31 巻 10 号 p. 1427-1442
Aim: Older patients with chronic kidney disease (CKD) are more likely to be excluded from clinical trials. This exclusion affects the quality of cardiovascular disease (CVD) prevention in this population.
Methods: Baseline data from the Fukuoka Kidney Disease Registry (FKR) cohort, which included 4476 adult patients with CKD stages G1–G5, were cross-sectionally analyzed to compare the use of recommended drugs for preventing CVD in each age group.
Results: Different prescribing patterns were observed according to age for the cardiovascular drug classes. Older patients with CKD were less likely to receive renin–angiotensin system (RAS) inhibitors and were more likely to receive calcium channel blockers. The proportion of anticoagulation prescriptions for patients with CKD and atrial fibrillation decreased in the older age group (≥ 75 years). However, the proportion of antiplatelet therapy in patients with ischemic CVD increased linearly with age, even in the very old group aged ≥ 85 years. These findings suggest a severe cardiovascular burden in patients with CKD. Notably, RAS inhibitor use was avoided in the older group despite a severe cardiovascular burden, such as a high prevalence of CVD history and massive albuminuria ＞300 mg/g creatinine. This finding indicates that an older age independently contributed to the non-use of RAS inhibitors, even after adjusting for other covariates.
Conclusions: This study suggests that age is a potential barrier to the treatment of patients with CKD and highlights the need to establish individualized treatment strategies for cardiovascular protection in this population.
Patients with chronic kidney disease (CKD) are a high-risk population for atherosclerotic disease, with a variety of risk factors that are frequently complicated by various forms of cardiovascular disease (CVD)1-5). The CVD guidelines advocate the same pharmaceutical options for patients with CKD as for non-CKD patients6-8). However, patients with CKD experience more adverse events with standard CVD treatment medications than patients without CKD. The Systolic Blood Pressure Intervention Trial (SPRINT), which examined the effectiveness of stringent blood pressure control in non-diabetic patients with CKD, showed an increase in adverse events, such as a low blood pressure and high potassium concentrations9, 10).
The CVD burden tends to increase with age. Despite the high CVD burden, older patients with CKD tend to be excluded from clinical trials. Additionally, age affects the pharmacokinetics of commonly used drugs and makes older individuals more susceptible to adverse drug events. Most clinical trials are conducted in younger, healthier patient populations, and there is limited evidence on the benefits and risks of prescribing cardiovascular drugs to fragile older patients. Therefore, age has been reported to be associated with the underuse of cardiovascular medications in non-CKD populations11-15). Moreover, there is a lack of large population-based studies that have extensively evaluated cardiovascular drugs prescribed to the older CKD population, thus raising the concern that older patients with CKD may not receive adequate recommended therapy for CVD prevention.
The Fukuoka Kidney Disease Registry (FKR) study is a prospective multicenter observational study of a cohort of non-dialysis-dependent CKD16). This cohort has unrestricted inclusion criteria, which allows the study of a broad CKD population, including patients aged ≥ 75 years and patients with advanced CKD with an estimated glomerular filtration rate (eGFR) ＜15 mL/min/1.73 m2. We hypothesized that older patients with CKD use cardiovascular medications less frequently than younger patients with CKD, despite their higher CVD burden. This study aimed to investigate the effects of age, the level of kidney function, and comorbidities on the pattern of use of recommended cardiovascular drugs in older patients with CKD using the FKR cohort data.
The FKR Study, which has previously been described in detail16, 17), enrolled 4,476 patients with CKD aged ≥ 16 years from 12 centers between January 2013 and March 2017. All patients were under the care of nephrologists and met the definition of CKD according to the KDIGO Clinical Practice guidelines18). The study protocol was approved by the Clinical Research Ethics Committee of the Institutional Review Board of Kyushu University (approval number: 469-09) and the ethics committees of all participating institutions. Written informed consent was obtained from all patients at the start of the study. This study was registered in the University Hospital Medical Information Network Clinical Trial Registry (UMIN000007988). The data analysis was conducted from December 2022 to January 2023.
Clinical ParametersDemographic and clinical information on each patient’s health status and medication were collected from the patients’ medical records using a structured data format by the clinical research coordinators in our research network. Data on alcohol consumption, smoking, comorbidities, and current medication use, such as antihypertensive and lipid-lowering medications, were collected through a self-administered questionnaire. Heart failure, coronary artery disease, stroke, and peripheral vascular disease were defined as CVDs. The systolic and diastolic blood pressure were measured with the patients in the sitting position using an automated device by the attending physician or a trained clinical research coordinator at the time of enrolment. Hypertension was defined as a systolic blood pressure of ≥ 140 mmHg, diastolic blood pressure of ≥ 90 mmHg, and/or current use of antihypertensive agents. Diabetes was identified in patients with an HbA1c value ≥ 6.5% (National Glycohemoglobin Standardization Program), diabetic nephropathy, diabetic complications, and/or current treatment with insulin or oral hypoglycemic agents. Diabetic nephropathy was classified according to clinical judgement based on the information in each patient’s medical record or histological diagnosis based on kidney biopsy. Dyslipidemia was defined as a low-density lipoprotein cholesterol concentration ≥ 140 mg/dL, high-density lipoprotein cholesterol concentration ＜40 mg/dL, triglyceride concentration ≥ 150 mg/dL, or current use of lipid-lowering drugs. Lipid-lowering drugs include statins and ezetimibe. Antithrombotic agents included antiplatelet agents and/or oral anticoagulants. Oral anticoagulants include warfarin and/or direct oral anticoagulants (DOACs). Cardiovascular drug classes included RAS inhibitors, calcium channel blockers (CCBs), beta-blockers, lipid-lowering drugs, antiplatelet agents, and oral anticoagulants.
Statistical AnalysisThe patients’ baseline characteristics are shown as the median and interquartile range for continuous variables and as number and percentage for categorical variables. Logistic regression models were used to compare categorical variables and linear regression models were used for continuous variables. To investigate prescribing trends by age category (by 10 years) and the recommended drug class by the glomerular filtration rate (GFR), logistic regression models were used to calculate the adjusted odds for each age category for each cardiovascular drug. Clinically or biologically validated risk factors for CVD (i.e., sex, hypertension, diabetes, dyslipidemia, prior CVD, body mass index, serum albumin, eGFR, and urine protein/creatinine [Cr] ratio) were initial candidate variables for multivariable adjustment models. Multivariable-adjusted odds ratios and 95% confidence intervals for factors associated with cardiovascular drug use were estimated using a final model in which covariates were selected using stepwise backward elimination with a P value ＜0.05, as the selection criterion to determine independent predictors of each cardiovascular drug use.
Several sensitivity analyses were performed. First, we performed subgroup analyses based on history of CVD to detect the potential modifying effects of CVD. Second, to explore the impact of aging on hypertension and dyslipidemia treatment independent of CVD, we performed subgroup analyses based on elevated blood pressure and elevated lipid levels, restricted to patients without a history of CVD. Elevated blood pressure was defined as systolic blood pressure ≥ 140 mmHg and diastolic blood pressure ≥ 90 mmHg, and elevated lipid levels were defined as low-density lipoprotein cholesterol ≥ 140 mg/dL, high-density lipoprotein cholesterol ＜40 mg/dL, and triglyceride ≥ 150 mg/dL. Third, we evaluated the association between antidiabetic medications and age in a population restricted to participants with comorbid diabetes.
Statistical analyses were performed using the SAS version 9.4 (SAS Institute, Cary, NC, USA) and STATA version 17 (StataCorp, College Station, TX, USA) software programs. Statistical significance was set at P＜0.05 (two-sided) were considered to be statistically significant.
The participants’ characteristics are presented in Table 1. The median patient age was 67 years. The study population consisted of 4,476 patients, of whom 56% were men. A total of 38.8% of the patients were non-smokers, 47.6% were current smokers, and 41.2% were former smokers. The primary disease was chronic glomerulonephritis without biopsy in 14.4% of patients, biopsy-proven primary chronic glomerulonephritis in 34.9%, diabetic nephropathy in 11.2%, hypertensive nephrosclerosis in 21.6%, and others in 17.9%. Hypertension, dyslipidemia, and diabetes mellitus were present in 83.3%, 45.9%, and 28.7% of the patients, respectively. The median systolic blood pressure was 130 mmHg, the median diastolic blood pressure was 74 mmHg, and 41.7 percentage of patients achieved ＜130/80 mmHg. The median eGFR was 40.0 mL/min/1.73 m2, and the median urine protein to Cr ratio and urine albumin to Cr ratio were 0.40 g/gCr and 206.5 mg/gCr, respectively. With increasing age, the proportion of biopsy-proven chronic glomerulonephritis decreased and hypertensive nephrosclerosis increased. The blood pressure levels significantly decreased with advancing age. eGFR showed a decreasing trend in the older age groups, and urinary protein and albumin excretion increased with age. The total cholesterol, low-density lipoprotein cholesterol, and triglyceride levels were lower in the older group.
| All patients | Age categories (years) | P value | |||||
|---|---|---|---|---|---|---|---|
| ＜55 | 55–64 | 65–74 | 75–84 | ≥ 85 | |||
| No. of participants | 4,476 | 1,115 | 798 | 1297 | 1,008 | 258 | |
| Age, years | 67 (55–76) | 43 (34–49) | 61 (58–63) | 69 (67–72) | 79 (77–81) | 87 (85–89) | ＜0.001 |
| Sex (male), % | 56.0 | 46.9 | 56.8 | 58.3 | 63.1 | 53.9 | ＜0.001 |
| Smoking history | |||||||
| Never smoker, % | 47.6 | 57.6 | 43.0 | 45.6 | 43.7 | 49.4 | ＜0.001 |
| Current smoker, % | 11.2 | 17.1 | 14.5 | 10.2 | 6.4 | 3.4 | 0.57 |
| Former smoker, % | 41.2 | 25.3 | 42.5 | 44.2 | 49.9 | 47.2 | ＜0.001 |
| Drinking history | |||||||
| Never drinker, % | 38.8 | 32.6 | 32.6 | 41.5 | 43.0 | 51.0 | 0.02 |
| Current drinker, % | 48.6 | 59.8 | 58.6 | 45.2 | 39.6 | 28.5 | ＜0.001 |
| Former drinker, % | 12.6 | 7.6 | 8.8 | 13.3 | 17.4 | 20.5 | ＜0.001 |
| Underlying kidney disease | |||||||
| CGN without biopsy, % | 14.4 | 10.6 | 13.3 | 16.9 | 15.4 | 17.4 | ＜0.001 |
| Biopsy-proven CGN, % | 34.9 | 60.2 | 39.6 | 28.6 | 17.6 | 9.7 | ＜0.001 |
| Diabetic nephropathy, % | 11.2 | 5.9 | 14.5 | 13.7 | 12.3 | 7.0 | ＜0.001 |
| Hypertensive nephrosclerosis, % | 21.6 | 4.7 | 13.5 | 22.9 | 37.8 | 50.8 | ＜0.001 |
| Others, % | 17.9 | 18.6 | 19.1 | 17.9 | 16.9 | 15.1 | 0.80 |
| Comorbidities | |||||||
| Hypertension, % | 83.3 | 68.1 | 86.4 | 88.9 | 92.2 | 91.0 | ＜0.001 |
| Diabetes mellitus, % | 28.7 | 13.4 | 30.1 | 35.5 | 35.5 | 30.2 | ＜0.001 |
| Dyslipidaemia, % | 45.9 | 45.2 | 50.9 | 47.3 | 45.2 | 31.3 | ＜0.001 |
| Cardiovascular disease, % | 23.3 | 7.1 | 14.5 | 24.5 | 39.9 | 49.8 | ＜0.002 |
| Ischaemic heart disease, % | 10.7 | 1.8 | 5.5 | 11.7 | 19.8 | 25.7 | ＜0.001 |
| Congestive heart failure, % | 2.9 | 1.4 | 1.8 | 2.9 | 4.4 | 8.2 | ＜0.001 |
| Ischaemic stroke, % | 9.1 | 2.5 | 6.7 | 9.4 | 15.2 | 20.2 | ＜0.001 |
| Haemorrhagic stroke, % | 2.0 | 1.6 | 1.5 | 1.9 | 2.6 | 3.9 | ＜0.001 |
| Peripheral artery disease, % | 3.2 | 0.5 | 1.8 | 3.6 | 6.2 | 6.2 | ＜0.005 |
| Atrial fibrillation, % | 5.5 | 0.6 | 1.5 | 6.6 | 10.2 | 14.3 | ＜0.001 |
| Cancer, % | 12.7 | 3.1 | 9.4 | 14.3 | 21.2 | 22.1 | ＜0.001 |
| Bone fracture, % | 6.3 | 2.5 | 4.3 | 5.7 | 9.9 | 17.9 | ＜0.001 |
| Blood pressure | |||||||
| Systolic BP, mmHg | 130 (119–142) | 122 (112–133) | 130 (120–141) | 132 (122–144) | 133 (122–145) | 132 (120–145) | ＜0.001 |
| Diastolic BP, mmHg | 74 (67–82) | 75 (67–83) | 78 (70–85) | 75 (69–81) | 70 (64–79) | 69 (60–76) | ＜0.001 |
| Patients with BP ＜130/80 mmHg, % | 41.7 | 53.1 | 42.9 | 28.7 | 40.2 | 60.5 | ＜0.001 |
| Kidney parameters | |||||||
| Serum Cr, mg/dL | 1.29 (0.88–2.09) | 0.90 (0.68–1.37) | 1.21 (0.87–1.92) | 1.34 (0.95–2.14) | 1.64 (1.20–2.40) | 1.91 (1.31–2.77) | ＜0.001 |
| Serum cystatin C, mg/L | 1.43 (0.96–2.26) | 0.91 (0.72–1.43) | 1.24 (0.94–2.01) | 1.48 (1.07–2.34) | 1.89 (1.40–2.61) | 2.37 (1.72–3.11) | ＜0.001 |
| eGFR, mL/min/1.73 m2 | 40.0 (23.4–59.2) | 64.2 (42.2–85.7) | 43.6 (26.8–59.5) | 38.2 (22.4–52.1) | 29.1 (19.6–41.9) | 23.1 (15.9–34.7) | ＜0.001 |
| UPCR, g/g Cr | 0.40 (0.11–1.30) | 0.34 (0.10–1.12) | 0.40 (0.10–1.30) | 0.40 (0.11–1.35) | 0.42 (0.13–1.38) | 0.55 (0.18–1.89) | ＜0.001 |
| UACR, mg/g Cr | 206.5 (35.7–794.5) | 206.7 (35.4–732.5) | 207.2 (35.1–801.2) | 205.3 (31.7–821.3) | 196.7 (37.1–804.4) | 232.2 (54.3–957.8) | ＜0.001 |
| Lipid parameters | |||||||
| Total cholesterol, mg/dL | 191 (167–217) | 199 (176–224) | 196 (173–223) | 190 (167–215) | 183 (158–208) | 180 (155–208) | ＜0.001 |
| LDL-cholesterol, mg/dL | 104.4 (84.7–125.8) | 107.8 (89.0–129) | 108.2 (88.8–129.8) | 103.8 (84.2–125.4) | 100.2 (80.8–121) | 98.4 (80.8–120.6) | ＜0.001 |
| HDL-cholesterol, mg/dL | 56 (45–70) | 61 (48–77) | 56 (46–69) | 56 (45–68) | 53 (43–66) | 56 (47–68) | 0.01 |
| Triglycerides, mg/dL | 120 (87–171) | 111 (77–178) | 131 (93–186) | 126 (92–170) | 119.5 (88.0–164) | 101.0 (80.5–139) | ＜0.001 |
| Cardiovascular drugs | |||||||
| Antihypertensive drugs, % | 78.8 | 65.3 | 82.0 | 84.5 | 87.9 | 85.9 | ＜0.001 |
| RAS inhibitors, % | 70.1 | 61.7 | 76.5 | 73.8 | 72.7 | 59.5 | 0.30 |
| CCBs, % | 48.7 | 25.0 | 47.3 | 58.0 | 61.6 | 59.1 | ＜0.001 |
| Beta-blockers, % | 15.2 | 8.2 | 13.4 | 17.5 | 20.5 | 19.8 | ＜0.001 |
| Lipid-lowering drugs, % | 42.1 | 29.6 | 46.9 | 49.2 | 45.0 | 37.0 | ＜0.001 |
| Statins, % | 40.2 | 28.3 | 44.5 | 46.5 | 43.0 | 36.6 | ＜0.001 |
| Ezetimibe, % | 4.0 | 2.9 | 5.8 | 5.0 | 3.4 | 1.2 | ＜0.001 |
| Antithrombotic drugs, % | 22.5 | 5.3 | 12.5 | 24.9 | 38.9 | 51.4 | ＜0.001 |
| Antiplatelets, % | 18.3 | 3.4 | 10.8 | 20.0 | 32.3 | 42.0 | ＜0.001 |
| Oral anticoagulants, % | 6.6 | 2.3 | 2.8 | 7.6 | 11.2 | 14.4 | ＜0.001 |
Quantitative data are presented as the median (interquartile range) and categorical data as the percentage.
Abbreviations: CKD, chronic kidney disease; CGN, chronic glomerulonephritis; BP, blood pressure; Cr, creatinine; eGFR, estimated glomerular filtration rate; UPCR, urine protein to creatinine ratio; UACR, urine albumin to creatinine ratio; LDL, low-density lipoprotein; HDL, high-density lipoprotein; RAS, renin–angiotensin system; CCBs, calcium channel blockers.
Lipid-lowering drugs included statins and/or ezetimibe. Antithrombotic agents comprised antiplatelet agents and/or oral anticoagulants. Oral anticoagulants comprised warfarin and/or direct oral anticoagulants.
Hypertension was defined as blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg, and/or current use of antihypertensive agents.
Diabetes was identified when haemoglobin A1c values (National Glycohemoglobin Standardization Program) were ≥ 6.5%, underlying kidney disease was diabetic nephropathy, and/or current treatment with insulin or oral antihyperglycemic drugs was ongoing.
Dyslipidaemia was defined as triglyceride concentrations ≥ 150 mg/dL, low-density lipoprotein cholesterol concentrations ≥ 140 mg/dL, high- density lipoprotein cholesterol concentrations ＜40 mg/dL, or the use of antihyperlipidaemic medication.
Cardiovascular disease was defined as a composite of ischaemic heart disease, congestive heart failure, ischaemic stroke, haemorrhagic stroke, peripheral vascular disease, thoracic aortic aneurysm, and abdominal aortic aneurysm.
Antihypertensive medications were prescribed to 3,527 (78.8%) patients, of whom 3,138 (70.1%) were prescribed RAS inhibitors, 2,179 (48.7%) were CCBs, and 680 (15.2%) were prescribed beta-blockers. Lipid-lowering drugs were prescribed to 1,884 (42.1%) patients, statins to 1,798 (40.2%), ezetimibe to 179 (4.0%), overall antithrombotic treatment to 1,005 (22.5%), antiplatelets to 817 (18.3%), and oral anticoagulants to 297 (6.6%). The percentage of patients receiving RAS inhibitors, beta-blockers, lipid-lowering agents, and hypoglycemic agents peaked at the ages of 55–64, 75–84, 65–74, and 65–74 years, respectively, with a downward trend in the older group. In contrast, the frequency of antithrombotic therapy increased linearly with age (Fig.1). The cumulative number of cardiovascular medications increased linearly with decreasing kidney function in all age groups (Fig.2). Notably, a linear upward trend was observed between CKD severity and the number of cardiovascular medications administered to patients aged ＜85 years, but no such trend was observed in those aged ≥ 85 years. Table 2 shows the multivariable-adjusted odds ratios for cardiovascular medication use according to age group. The odds of receiving RAS inhibitors were significantly lower in patients aged ≥ 75 years. In contrast, the odds of CCB use increased linearly with age. The odds of lipid-lowering drug use peaked at 65–74 years of age and then decreased with older age, thus indicating lower use in the older age group. The adjusted odds ratio of antithrombotic and antiplatelet use in patients with ischemic CVD increased linearly with age. The adjusted odds of anticoagulation in patients with atrial fibrillation peaked in patients aged 65–74 years and decreased in the older age categories.
Abbreviations: CKD, chronic kidney disease. RAS: renin–angiotensin system; CCBs: calcium channel blockers. Lipid-lowering drugs include statins and ezetimibe. Antithrombotic agents included antiplatelet agents and/or oral anticoagulants. Oral anticoagulants included warfarin and/or direct oral anticoagulants (DOACs).
Abbreviations: eGFR, estimated glomerular filtration rate; RAS, renin–angiotensin system. The cumulative number of cardiovascular agents (RAS inhibitors, beta blockers, lipid-lowering agents, and antithrombotic agents) increased with age.
| Age group (years) |
Adjusted OR (95% CI) a All patients (n= 4,476) |
Adjusted OR (95% CI) a Patients with a history of IHD, ischaemic stroke, and PAD (n= 853) |
Adjusted OR (95% CI) a Patients with atrial fibrillation (n= 243) |
|||||
|---|---|---|---|---|---|---|---|---|
| No. of participants |
Antihypertensive drugs b 3,527 (78.8%) |
RAS inhibitors c 3,138 (70.1%) |
CCBs d 2,179 (48.7%) |
Beta-blockers e 680 (15.2%) |
Lipid-lowering drugs f 1,884 (42.1%) |
Antithrombotic drugs g 645 (75.6%) |
Antiplatelet drugs h 574 (67.3%) |
Oral anticoagulants i 179 (73.7%) |
| ＜55 years | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55–64 years | 1.20 (0.93–1.55) | 1.26 (1.01–1.59) | 1.84 (1.48–2.29) | 1.01 (0.73–1.39) | 1.83 (1.50–2.25) | 2.03 (0.93–4.43) | 2.64 (1.23–5.81) | 1.39 (0.18–10.7) |
| 65–74 years | 1.17 (0.92–1.49) | 0.96 (0.78–1.17) | 2.64 (2.16–3.24) | 1.18 (0.89–1.57) | 1.99 (1.66–2.40) | 2.23 (1.16–4.29) | 2.45 (1.27–4.81) | 3.10 (0.54–16.7) |
| 75–84 years | 1.17 (0.92–1.49) | 0.77 (0.62–0.97) | 2.62 (2.10–3.28) | 1.00 (0.74–1.35) | 1.64 (1.33–2.00) | 2.07 (1.08–3.95) | 2.32 (1.21–4.53) | 2.79 (0.50–14.7) |
| ≥ 85 years | 0.91 (0.59–1.40) | 0.42 (0.30–0.58) | 2.44 (1.75–3.39) | 0.84 (0.55–1.28) | 1.16 (0.85–1.59) | 4.13 (1.85–9.43) | 4.11 (1.91–9.07) | 2.23 (0.36–13.2) |
| P value for trend | 0.579 | ＜0.001 | ＜0.001 | 0.631 | 0.001 | ＜0.001 | 0.011 | 0.602 |
Abbreviations: CKD, chronic kidney disease; CAD, coronary artery disease; OR, odds ratio; RAS, renin–angiotensin system; CCBs, calcium channel blockers; IHD, ischaemic heart disease; PAD, peripheral arterial disease; CI, confidence interval; estimated glomerular filtration rate; UPCR, Urinary protein to creatinine ratio.
Lipid-lowering drugs included statins and/or ezetimibe. Antithrombotic agents comprised antiplatelet agents and/or oral anticoagulants. Oral anticoagulants comprised warfarin and/or direct oral anticoagulants.
a Variables were selected by a logistic regression model using a stepwise backward method with P＜0.05 for the remaining variables to determine the independent predictors for each drug use.
b Final model adjusted by sex, diabetes, prior CVD, body mass index, serum albumin, eGFR, and UPCR.
c Final model adjusted by sex, dyslipidaemia, body mass index, serum albumin, eGFR, and UPCR.
d Final model adjusted by sex, diabetes, prior CVD, body mass index, serum albumin, eGFR, and UPCR.
e Final model adjusted by dyslipidaemia, prior CVD, body mass index, serum albumin, and eGFR.
f Final model adjusted by sex, hypertension, diabetes, prior CVD, body mass index, serum albumin, and UPCR.
g Final model adjusted by sex, diabetes, and eGFR.
h Final model adjusted by hypertension, diabetes, and eGFR.
i Final model adjusted by body mass index.
The clinical parameter values for each age category stratified by the presence or absence of RAS inhibitors are shown in Fig.3. The systolic blood pressure tended to be higher in the treated group and lower in the untreated group. Urinary albumin excretion tended to be higher in the older age group, even in the older age group (≥ 75 years). When RAS inhibitors were not actively used, the treated and untreated patients showed severe urinary albumin excretion (equivalent to A3). Supplementary Fig.1 shows the mean total cholesterol concentration and the rate of CVD history stratified by statin status. The mean total cholesterol concentration decreased with age and was higher in the untreated group than in the statin-treated group across all age groups. The prevalence of a cardiovascular history was higher in the statin-treated group than in the untreated group, and increased with increasing age. The serum potassium concentration increased with age and remained consistently higher in the group treated with RAS inhibitors than in the untreated group across all age groups. However, in patients aged ≥ 85 years, the serum potassium concentrations were similar between the treated and untreated groups (Supplementary Fig.2). The percentage of patients aged ≥ 85 years with advanced stages G4–5 was higher in the treated group (73.9%) than in the untreated group (56.7%) (Supplementary Fig.3).
Abbreviation: RAS, renin–angiotensin system.
Abbreviation: CVD, cardiovascular disease.
Abbreviations: RAS, renin–angiotensin system.
Abbreviations: CKD, chronic kidney disease; RAS, renin–angiotensin system.
A subgroup analysis stratified by history of CVD showed a downward trend in the treatment rates of hypertension and dyslipidemia in the elderly compared to younger subjects, regardless of CVD history (Supplementary Fig.4). Furthermore, even when the population was restricted to subjects without a history of CVD, the adjusted odds ratios for treatment for blood pressure and lipids tended to be lower in the older group than in the younger group (Supplementary Tables 1 and 2). As shown in Supplementary Table 3, the multivariate-adjusted odds for prescribing RAS inhibitors showed a downward trend in elderly patients, regardless of CVD status. There was no significant trend between overall antidiabetic drug use and age category, but only for insulin, as well as for cardiovascular drugs, there was a downward trend in the odds ratio for drug use with increasing age category (Supplementary Table 4).
Abbreviations: CVD, cardiovascular disease; RAS, renin–angiotensin system; CCBs, calcium channel blockers. Lipid-lowering drugs include statins and ezetimibe. Antithrombotic agents included antiplatelet agents and/or oral anticoagulants. Oral anticoagulants included warfarin and/or direct oral anticoagulants (DOACs).
| Elevated BP (-) | Adjusted OR (95% CI) a | ||||
|
No. of participants, N = 1,010 |
Antihypertensive drugs 851 (84.3%) | RAS inhibitors 764 (75.6%) | CCBs 995 (98.5%) | Beta-blocker 129 (12.8%) | Lipid lowering drugs 415 (41.1%) |
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 0.93 (0.51-1.71) | 1.12 (0.67-1.88) | 1.21 (0.79-1.87) | 0.59 (0.31-1.13) | 1.58 (1.04-2.40) |
| 65-74 | 0.85 (0.47-1.50) | 0.58 (0.37-0.93) | 1.79 (1.19-2.71) | 0.88 (0.51-1.52) | 1.77 (1.20-2.61) |
| 75-84 | 0.63 (0.33-1.18) | 0.59 (0.36-0.99) | 1.13 (0.73-1.77) | 0.77 (0.43-1.40) | 1.81 (1.19-2.75) |
| ≥ 85 | 0.54 (0.21-1.52) | 0.41 (0.20-0.87) | 1.46 (0.72-2.97) | 0.57 (0.21-1.53) | 1.35 (0.68-2.67) |
| P-value for trend | 0.115 | 0.001 | 0.262 | 0.633 | 0.021 |
| Elevated BP (+) | Adjusted OR (95% CI) a | ||||
|
No. of participants, N = 2,382 |
Antihypertensive drugs 1,680 (70.5%) | RAS inhibitors 1,548 (65.0%) | CCBs 883 (37.1%) | Beta-blocker 195 (8.2%) | Lipid lowering drugs 525 (50.4%) |
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 1.18 (0.87-1.59) | 1.10 (0.83-1.45) | 2.37 (1.77-3.16) | 1.29 (0.80-2.10) | 2.01 (1.55-2.63) |
| 65-74 | 1.23 (0.92-1.63) | 0.97 (0.75-1.26) | 3.30 (2.51-4.33) | 1.41 (0.90-2.20) | 2.35 (1.83-3.02) |
| 75-84 | 1.29 (0.89-1.87) | 0.75 (0.55-1.02) | 4.51 (3.27-6.22) | 1.73 (1.07-2.80) | 1.94 (1.43-2.63) |
| ≥ 85 | 0.97 (0.52-1.90) | 0.31 (0.18-0.53) | 2.82 (1.64-4.85) | 1.24 (0.53-2.87) | 1.25 (0.71-2.18) |
| P-value for trend | 0.349 | 0.001 | ＜0.001 | 0.042 | ＜0.001 |
Abbreviations: BP, blood pressure; CKD, chronic kidney disease; CAD, coronary artery disease; OR, odds ratio; RAS, renin–angiotensin system; CCBs, calcium channel blockers; IHD, ischaemic heart disease; PAD, peripheral arterial disease; CI, confidence interval; estimated glomerular filtration rate; UPCR, Urinary protein to creatinine ratio.
Elevated blood pressure was defined as systolic blood pressure ≥ 140 mmHg and diastolic blood pressure ≥ 90 mmHg.
Antihypertensives were defined by the use of RAS inhibitors and/or CCBs and/or beta-blockers and/or loop diuretics and/or thiazide diuretics and/ or aldosterone antagonists and/or alpha-blockers. Lipid-lowering drugs included statins and/or ezetimibe.
a Variables were selected by a logistic regression model using a stepwise backward method with P＜0.05 for the remaining variables to determine the independent predictors for each drug use.
| Elevated lipid (-) | Adjusted OR (95% CI) a | ||||
|
No. of participants, N = 1,550 |
Antihypertensive drugs 1,198 (77.3%) | RAS inhibitors 1,092 (70.5%) | CCBs 738 (47.6%) | Beta-blocker 176 (11.4%) | Lipid lowering drugs 636 (41.0%) |
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 1.05 (0.69-1.55) | 1.05 (0.74-1.48) | 1.87 (1.35-2.58) | 1.01 (0.62-1.65) | 1.25 (0.92-1.70) |
| 65-74 | 1.03 (0.70-1.51) | 0.87 (0.63-1.21) | 2.35 (1.72-3.22) | 1.00 (0.63-1.59) | 1.44 (1.08-1.93) |
| 75-84 | 1.08 (0.66-1.77) | 0.77 (0.52-1.15) | 2.38 (1.65-3.42) | 1.44 (0.88-2.35) | 1.68 (1.20-2.37) |
| ≥ 85 | 0.77 (0.29-2.07) | 0.33 (0.15-0.69) | 2.05 (0.97-4.33) | 0.93 (0.33-2.62) | 1.07 (0.51-2.25) |
| P-value for trend | 0.005 | ＜0.001 | ＜0.001 | 0.510 | 0.005 |
| Elevated lipid (+) | Adjusted OR (95% CI) a | ||||
|
No. of participants, N = 1,872 |
Antihypertensive drugs 1333 (71.2%) | RAS inhibitors 1,220 (65.2%) | CCBs 718 (38.4%) | Beta-blocker 148 (7.9%) | Lipid lowering drugs 679 (36.3%) |
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 1.29 (0.90-1.85) | 1.38 (1.00-1.93) | 2.35 (1.65-3.35) | 1.12 (0.60-2.09) | 2.80 (2.05-3.83) |
| 65-74 | 1.26 (0.90-1.77) | 0.96 (0.71-1.30) | 3.60 (2.60-4.99) | 1.79 (1.04-3.07) | 3.12 (2.35-4.15) |
| 75-84 | 1.10 (0.73-1.66) | 0.80 (0.57-1.14) | 3.70 (2.58-5.31) | 1.42 (0.78-2.57) | 2.21 (1.60-3.06) |
| ≥ 85 | 0.85 (0.45-1.61) | 0.41 (0.25-0.69) | 2.78 (1.64-4.70) | 1.27 (0.54-2.98) | 1.71 (1.02-2.88) |
| P-value for trend | 0.046 | ＜0.001 | ＜0.001 | 0.330 | ＜0.001 |
Abbreviations: CKD, chronic kidney disease; CAD, coronary artery disease; OR, odds ratio; RAS, renin–angiotensin system; CCBs, calcium channel blockers; IHD, ischaemic heart disease; PAD, peripheral arterial disease; CI, confidence interval; estimated glomerular filtration rate; UPCR, Urinary protein to creatinine ratio.
Elevated lipid levels were defined as low-density lipoprotein cholesterol ≥ 140 mg/dL, high-density lipoprotein cholesterol ＜40 mg/dL and triglyceride ≥ 150 mg/dL.
Antihypertensives were defined by the use of RAS inhibitors and/or CCBs and/or beta-blockers and/or loop diuretics and/or thiazide diuretics and/ or aldosterone antagonists and/or alpha-blockers. Lipid-lowering drugs included statins and/or ezetimibe.
a Variables were selected by a logistic regression model using a stepwise backward method with P＜0.05 for the remaining variables to determine the independent predictors for each drug use.
| CVD (-) | Adjusted OR (95% CI) a | ||||
|
No. of participants, N = 3,429 |
Antihypertensive drugs 2,531 (73.8%) | RAS inhibitors 2,312 (67.4%) | CCBs 1,456 (42.5%) | Beta-blocker 324 (9.4%) | Lipid lowering drugs 1,315 (38.3%) |
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 1.20 (0.92-1.56) | 1.22 (0.96-1.55) | 2.05 (1.62-2.60) | 1.06 (0.72-1.56) | 1.84 (1.48-2.29) |
| 65-74 | 1.20 (0.93-1.55) | 0.93 (0.75-1.16) | 2.90 (2.32-3.62) | 1.32 (0.94-1.87) | 2.10 (1.72-2.57) |
| 75-84 | 1.15 (0.84-1.56) | 0.79 (0.61-1.03) | 3.02 (2.34-3.90) | 1.38 (0.95-2.02) | 1.86 (1.47-2.35) |
| ≥ 85 | 0.92 (0.54-1.57) | 0.39 (0.26-0.60) | 2.50 (1.64-3.82) | 1.03 (0.52-1.92) | 1.32 (0.86-2.02) |
| P-value for trend | ＜0.001 | ＜0.001 | ＜0.001 | 0.168 | ＜0.001 |
| CVD (+) | Adjusted OR (95% CI) a | ||||
|
No. of participants, N = 1,041 |
Antihypertensive drugs 910 (87.4%) | RAS inhibitors 759 (72.9%) | CCBs 662 (63.6%) | Beta-blocker 343 (32.9%) | Lipid lowering drugs 525 (50.4%) |
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 1.40 (0.49-4.01) | 1.70 (0.80-3.65) | 0.71 (0.37-1.35) | 0.78 (0.42-1.45) | 1.94 (1.03-3.63) |
| 65-74 | 1.07 (0.48-2.40) | 1.13 (0.61-2.02) | 1.12 (0.65-1.95) | 0.74 (0.44-1.27) | 1.55 (0.91-2.63) |
| 75-84 | 1.20 (0.53-2.72) | 0.85 (0.46-1.51) | 1.03 (0.59-1.78) | 0.54 (0.32-0.92) | 1.24 (0.73-2.11) |
| ≥ 85 | 0.82 (0.31-2.11) | 0.54 (0.27-1.04) | 1.08 (0.57-2.06) | 0.52 (0.28-0.97) | 0.99 (0.53-1.84) |
| P-value for trend | 0.408 | 0.001 | 0.559 | 0.012 | 0.199 |
Abbreviations: CKD, chronic kidney disease; CAD, coronary artery disease; OR, odds ratio; RAS, renin–angiotensin system; CCBs, calcium channel blockers; IHD, ischaemic heart disease; PAD, peripheral arterial disease; CI, confidence interval; estimated glomerular filtration rate; UPCR, Urinary protein to creatinine ratio.
Antihypertensives were defined by the use of RAS inhibitors and/or CCBs and/or beta-blockers and/or loop diuretics and/or thiazide diuretics and/ or aldosterone antagonists and/or alpha-blockers. Lipid-lowering drugs included statins and/or ezetimibe.
a Variables were selected by a logistic regression model using a stepwise backward method with P＜0.05 for the remaining variables to determine the independent predictors for each drug use.
|
No. of participants, N = 1,285 |
Adjusted OR (95% CI) a | ||||
|
Antidiabetic drugs 924 (71.9%) |
Sulfonyl urea 132 (10.3%) |
α-glucosidase inhibitor 212 (16.5%) |
Dipeptidyl peptidase-4 inhibitors 555 (43.2%) |
Insulin 276 (21.5%) |
|
| Age group (years) | |||||
| ＜55 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 55-64 | 1.24 (0.78-1.97) | 1.05 (0.49-2.30) | 1.63 (0.87-3.03) | 1.25 (0.82-1.92) | 0.95 (0.58-1.55) |
| 65-74 | 1.65 (1.07-2.54) | 1.60 (0.81-3.16) | 1.66 (0.93-2.96) | 1.26 (0.85-1.86) | 0.81 (0.51-1.28) |
| 75-84 | 1.17 (0.75-1.85) | 1.68 (0.84-3.39) | 1.59 (0.87-2.89) | 1.33 (0.89-2.00) | 0.60 (0.37-0.98) |
| ≥ 85 | 1.15 (0.60-2.19) | 1.80 (0.71-4.58) | 1.31 (0.58-2.97) | 1.71 (0.96-3.04) | 0.47 (0.22-0.99) |
| P-value for trend | 0.421 | 0.047 | 0.479 | 0.092 | 0.004 |
Abbreviations: OR, odds ratio; RAS, renin–angiotensin system; CCBs, calcium channel blockers; IHD, ischemic heart disease; PAD, peripheral arterial disease; CI, confidence interval.
Antidiabetic drugs were defined by the use of insulin and/or sulfonyl ureas and/or biguanides and/or alpha-glucosidase inhibitors and/or Dipeptidyl peptidase-4 inhibitors and/or glinides, thiazolidinediones and/or Glucagon-Like Peptide-1 receptor agonists, and/or Sodium–glucose cotransporter 2 inhibitors.
a Variables were selected by a logistic regression model using a stepwise backward method with P＜0.05 for the remaining variables to determine the independent predictors for each drug use.
This study showed different prescribing patterns for the six cardiovascular drug classes, depending on age. Older patients with CKD were less frequently prescribed antihypertensive and lipid-lowering drugs, thus reflecting adherence to a guideline-based standard of care. Furthermore, the proportion of anticoagulation prescriptions in patients with CKD and atrial fibrillation was lower than that of other prescription types in patients aged ≥ 75 years. However, the rate of antiplatelet therapy increased linearly in the very old (≥ 85 years) group with ischemic CVD, thus highlighting the seriousness of cardiovascular burden in the CKD population. RAS inhibitors were avoided in the older group. This finding indicates that older age (≥ 75 years) is an independent contributor to non-use of RAS inhibitors, even after adjustment for other covariates. However, the prevalence of pre-existing CVD was higher in the older group with or without RAS inhibitors. Even in the RAS inhibitor-naive group, the mean urinary albumin excretion was ＞300 mg/gCr, thus indicating a high level of proteinuria. Therefore, in the cardiovascular treatment of older patients with CKD, prescription decisions tend to be based solely on the patient’s advanced age, irrespective of the risk profile of the individual case. This finding suggests that individualized treatment according to the cardiovascular burden of each patient was not considered. These findings indicate that concerns regarding the harm of treatment outweigh the benefits, or the lack of evidence for treating older patients with CKD may prevent the establishment of individualized treatment strategies for cardiovascular protection in this population.
The proportion of patients in the present study using cardiovascular medications such as RAS blockers, lipid-lowering agents, and antithrombotic agents is generally consistent with previous reports on older people. The proportion of beta-blocker use in our cohort was approximately half of that reported previously19, 20). Current Japanese CKD guidelines recommend statin therapy for CVD prevention in older patients with CKD21), but in our cohort, the percentage of lipid-lowering drug use decreased with age owing to lower lipid concentrations in the older group. Although oral anticoagulants are effective in older people, previous studies have shown conflicting results regarding the association between age at discharge after hospitalization for stroke due to atrial fibrillation and the use of oral anticoagulants13-15, 19). Adherence to long-term antiplatelet medications after stroke is better in older patients, whereas adherence to oral anticoagulants is worse in older patients22). In the present study, the frequency of antiplatelet drug use increased linearly with age in older adults with a history of atherosclerotic CVD. However, in patients aged ≥ 75 years with atrial fibrillation, the frequency of oral anticoagulant prescriptions was lower than that of other types of prescriptions.
RAS inhibitors are pivotal medications for patients with CKD and urinary proteinuria because they effectively lower the blood and intraglomerular pressures. This reduction then decreases urinary protein excretion and protects against kidney damage while also greatly reducing the risk of cardiovascular events. A sub-analysis of the Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients (JATOS) study in older Japanese patients with CKD showed that CCBs improved the prognosis of the kidney and increased the eGFR at 2 years, even in patients aged ≥ 75 years23). Therefore, the Japanese CKD guidelines recommend the use of CCBs for CKD stages G4 and G5 in patients ≥ 75 years of age21). In this study, a downward trend in RAS inhibitor use and prescription, and an upward trend in CCB use were observed in patients aged ≥ 75 years, thus reflecting compliance with the Japanese CKD guidelines. However, a stratified analysis of the patients with and without RAS inhibitors showed that even in the RAS inhibitor-naive group, the degree of cardiovascular burden, including cardiovascular history and urinary albumin excretion, was severe. Additionally, low blood pressure and high serum potassium concentrations, which could be contraindications for RAS inhibitor use, were absent. While these findings suggest appropriate care provided by Japanese nephrologists based on high adherence to the guidelines, they also indicate the possibility that uniform treatment is applied to older patients with CKD without consideration of their cardiovascular burden status. Indeed, our sensitivity analysis showed a lower prescribing trend for RAS inhibitors in the elderly, regardless of CVD comorbidities. This suggests that aging has a strong influence on clinical decisions, particularly the avoidance of RAS inhibitor induction, for both the primary and secondary prevention of CVD. These results thus appear to provide further evidence that aging may be a barrier to treatment in the management of CVD risk factors, including hypertension and dyslipidemia.
Several factors may explain the low prescription rates of some cardiovascular medications in the older patients with CKD in this study. First, many clinical trials exclude older patients with comorbid CKD and thus lack specific evidence of recommendations in this population. For example, the Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan (RENAAL) trial, which examined the efficacy of renin–angiotensin system (RAS) inhibitors in patients with diabetic nephropathy, set age 70 years as the upper limit of the inclusion criteria24). The Assessment of Clinical Usefulness in CKD Patients with Atorvastatin (ASUCA) trial, which examined the clinical benefit of statins in Japanese patients with CKD, also excluded subjects older than 70 years25). Second, the efficacy of cardiovascular drugs is more difficult to establish in older patients than in younger patients because they have a different disease profile (e.g., heart failure with preserved left ventricular ejection fraction is more common in older patients)26). In addition, the risk of adverse events is higher in older patients owing to pharmacokinetic differences, reduced kidney function, drug interactions, and polypharmacy, which often limits medication use27). Patient and physician preferences may also affect medication use28-30).
The present study is associated with several limitations. First, the cross-sectional design did not allow us to ascertain whether drugs were not used at all or discontinued due to side effects. Second, the ethical requirements for informed consent may have led to a selection bias by not including patients with dementia at baseline. Third, there is a risk of misclassification owing to the prescription of cardiovascular medications at multiple medical facilities. In this study, nephrologists did not control all cardiovascular protective medications; in some cases, they were managed through concurrent consultations with other clinics or specialists. Therefore, we cannot completely rule out the possibility that the prescription of cardiovascular protective drugs by physicians other than FKR collaborating physicians is not completely known. However, in our study, we conducted a thorough survey using self-administered questionnaires to the patients and their families, as well as telephone calls and letters to medical institutions and nursing homes. Prescriptions were also obtained whenever possible to accurately assess the medication status. The longitudinal follow-up rate of our cohort was extremely high at 97.6%; therefore, we believe that the medication status of our cohort of patients attending multiple medical facilities was relatively reliable. Finally, the patients were recruited exclusively from nephrology outpatients, which may not accurately reflect the prescription patterns in the overall CKD population.
In conclusion, this study distinguished between different CVD drug classes and showed different usage patterns depending on age. Robust evidence on the efficacy of cardiovascular medications in the elderly population with CKD comorbidities is still unclear. The older population with comorbid CKD has a higher incidence of cardiovascular events and a greater health economic burden of care. Therefore, further longitudinal and randomized controlled trials are required to assess the benefits and risks of cardiovascular medications in this population and the reasons for their underuse.
The authors thank the participants of the FKR Study, members of the FKR Study Group, and all personnel at the participating institutions involved in the study. The authors specifically thank Satoru Fujimi (Fukuoka Renal Clinic), Hideki Hirakata (Fukuoka Renal Clinic), Tadashi Hirano (Hakujyuji Hospital), Tetsuhiko Yoshida (Hamanomachi Hospital), Takashi Deguchi (Hamanomachi Hospital), Koji Mitsuiki (Harasanshin Hospital), Kiichiro Fujisaki (Iizuka Hospital), Keita Takae (Japanese Red Cross Fukuoka Hospital), Masanori Tokumoto (Japanese Red Cross Fukuoka Hospital), Akinori Nagashima (Japanese Red Cross Karatsu Hospital), Ritsuko Katafuchi (Kano Hospital), Hidetoshi Kanai (Kokura Memorial Hospital), Kenji Harada (Kokura Memorial Hospital), Tohru Mizumasa (Kyushu Central Hospital), Takanari Kitazono (Kyushu University), Toshiaki Nakano (Kyushu University), Toshiharu Ninomiya (Kyushu University), Kumiko Torisu (Kyushu University), Shigeru Tanaka (Kyushu University), Shunsuke Yamada (Kyushu University), Akihiro Tsuchimoto (Kyushu University), Yuta Matsukuma (Kyushu University), Sho Shimamoto (Kyushu University), Hiromasa Kitamura (Kyushu University), Hiroto Hiyamuta (Fukuoka University), Dai Matsuo (Munakata Medical Association Hospital), Yusuke Kuroki (National Fukuoka-Higashi Medical Center), Hiroshi Nagae (National Fukuoka-Higashi Medical Center), Masaru Nakayama (National Kyushu Medical Center), Kazuhiko Tsuruya (Nara Medical University), Masaharu Nagata (Shin-eikai Hospital), Taihei Yanagida (Steel Memorial Yawata Hospital), and Shotaro Ohnaka (Tagawa Municipal Hospital). The authors thank Ellen Knapp, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
S.T. contributed to the study design, data acquisition, statistical analysis, data interpretation, and manuscript drafting. H.K. contributed to the interpretation of the data and drafting of the manuscript. T.N. contributed to the study design, statistical analysis, data interpretation, and manuscript drafting. K.T. and T.K. contributed to critical revision of the manuscript and supervised the study. All authors provided a critical review of the manuscript and approved the final version.
The authors declare that they have no relevant financial interests.
The FKR Study did not receive any specific funding.
