Remember me
Chronic kidney disease (CKD) is a major public health issue affecting the burden of morbidity and mortality worldwide [1]. Current therapies for CKD focus mainly on slowing CKD progression through mitigation of hypertension and proteinuria with emphasis on evidence-based implementation of angiotensin-converting enzyme inhibitors (ACEi) or angiotensin II receptor blockers (ARBs) and, more recently, sodium-glucose transporter 2 inhibition [2,3]. Mineralocorticoid receptor antagonist (MRA) treatment imparts further inhibition of the renin-angiotensin-aldosterone system (RAAS) through direct antagonism of aldosterone leading to lowering of blood pressure and possible benefit on renal inflammation and fibrosis [3–5]. The efficacy of MRA treatment on survival in patients with symptomatic heart failure is well recognized [3,6,7]. Patients with CKD have a much higher incidence of heart failure compared with the general population and heart failure is an independent predictor of death in CKD [8]. A recent study has demonstrated a beneficial effect of MRA treatment on morbidity and mortality in patients with CKD and type 2 diabetes [9]. However, MRAs are potassium-sparing diuretics with associated increased risk of hyperkalemia, resulting in lower employment in patients with CKD due to a perceived increase in risk of attributable cardiac arrhythmia and death [10,11]. Consequently, MRA treatment is often not initiated, paused, or discontinued in patients with CKD, regardless of indication [6,12,13].
Although studies have investigated the use of MRA and associated risks of hyperkalemia in patients with heart failure and diabetes, information on risks associated with MRA treatment in CKD remains limited. On the basis of data from multiple nationwide Danish healthcare registers, this study examines the impact of MRA treatment on risk of hyperkalemia and subsequent mortality in patients with CKD.
MATERIALS AND METHODS Data sourcesAll Danish residents are assigned a unique personal identifier at birth or immigration enabling cross-referencing and linkage of individual-level data across administrative healthcare registers [14]. Universal healthcare is provided through tax-subsidization to all Danish residents. As such, wide-ranging data on provided healthcare and outcomes is accessible in Denmark on a national level. Specifically, information related to hospital admission, in-hospital procedures, and discharge is recorded in the Danish National Patient Register based on administrative code registered in accordance with 10th edition of the International Classification of Diseases (ICD-10) code [15]. Information on all redeemed prescription medication is recorded in the Danish National Database of Reimbursed Prescriptions based on Anatomical Therapeutic Chemical Classification System (ATC) code [16–18]. The National laboratory database records information related to laboratory results including renal function and plasma potassium levels from all central laboratories on a national level based on using the International System of Nomenclature, Properties, and Units (the NPU system) [18–19]. Mortality is recorded in the Cause of Death Register [20]. An overview of all employed administrative codes are provided in the supplementary materials (Table S1, https://links.lww.com/HJH/C361).
Study design Two distinct study designs were employedSusceptibility for hyperkalemia was assessed in a nested case–control framework. All Danish residents with available plasma creatinine work-up between 2011 and 2021 were followed until either measurement of hyperkalemia at least 6.0 mmol/l, death, or end of follow-up (October 5, 2021). Cases were identified based on incident hyperkalemia and subsequently matched with four controls without hyperkalemia on age, sex, diabetes, and hypertension.
Thirty-day all-cause mortality after hyperkalemia at least 6.0 mmol/l was assessed in a retrospective cohort design with inclusion of all Danish residents with available plasma creatinine between 2011 and 2021 and confirmed hyperkalemia at least 6.0 mmol/l. Index was defined as the date of hyperkalemia, with follow-up until either death or end of follow-up (October 5, 2021).
Study populationIn the case–control study, all Danish residents over 18 years of age with a plasma creatinine measured between January 1, 2011, and October 5, 2021, were included in the study population. If plasma creatinine was not measured within 2 years preceding the index for the cases and the pseudo index for the controls, patients were excluded from the study.
In the cohort study, patients over 18 years of age with a potassium at least 6.0 mmol/l and a plasma creatinine measured in Denmark between January 1, 2011, and October 5th, 2021, within 2 years preceding the index date or the pseudo index date were included.
Study exposures and covariatesNational healthcare registers were used to identify patient demographics, preexisting comorbidities, and concomitant medications. Comorbidities, characterizing the population, consisted of following diagnoses to be present before index date: hypertension, diabetes, ischemic heart disease, heart failure, stroke, and cancer. Data pertaining to prescription medication were employed to augment identification of hypertension and diabetes, because if exclusively using diagnosis codes approach, a substantial sample of patients would have been lost as treatment and monitoring of hypertension and diabetes often takes place in a primary care setting [15]. Hypertension was defined as the redemption of at least two antihypertensive drugs in two consecutive quarters. This definition was employed as monotherapy of many antihypertensive drugs are often used in treatment of other cardiovascular diseases, for example, heart failure, myocardial infarct, and atrial fibrillation. The definition has previously been validated with a positive predictive value of 80% and specificity of 94.7% [21]. Identification of diabetes by prescription of glucose-lowering medication has previously been validated with a positive predictive value of 96.9% [95% confidence interval (95% CI): 89.5–99.2] [15]. Baseline concomitant medication included redeemed prescription of antihypertensive and glucose-lowering treatment within 6 months prior to index. The identified medication is provided in the supplemental materials (Table S1B, https://links.lww.com/HJH/C361). Estimated glomerular filtration rate (eGFR) was computed based on age, sex, and the last recorded plasma creatinine until 30 days prior to index using the EPI-CKD equation [22].
Study outcomesThe association of MRA treatment with rate of hyperkalemia was analyzed using a nested case–control design with hyperkalemia defined as plasma potassium at least 6.0 mmol/l. Hazard rates were compared across strata of eGFR using patients without MRA treatment as reference.
The association of MRA treatment on subsequent 30-day mortality after hyperkalemia was analyzed across strata of eGFR based on a retrospective cohort study design.
Statistical analysesEvaluation of hyperkalemia susceptibility was computed in a multiple Cox regression model comparing rate of detected plasma potassium at least 6.0 mmol/l with stratification of baseline hazard rate by age, gender, diabetes, and hypertension. The association of MRA and hyperkalemia was evaluated across strata of eGFR (≥ 60, 45–59, 30–44, and < 30 ml/min/1.73 m2). Furthermore, based on same multiple Cox regression, hyperkalemia against eGFR was splined in patients on MRA and non-MRA treated patients, respectively, with restricted cubic spline of eGFR. Model fitting was accomplished using a nested case–control design with 1 : 4 risk-set matching of cases with age, sex, diabetes, and hypertension-matched controls. Cases, that is, individuals with confirmed plasma potassium at least 6.0 mmol/l, and controls were identified in national registers based on available laboratory results enabling estimation of renal function with use of CKD-EPI. All cases were, if possible, matched with four controls from the national cohort providing controls remained ‘at-risk’, that is alive and without hyperkalemia at index. Subgroup analyses were performed with stratification on sex, age, and specific risk factors (diabetes, hypertension, heart failure, and ACEi/ARB treatment).
Subsequent mortality after hyperkalemia was compared across strata of eGFR between patients treated with MRA and untreated patients based on the Kaplan–Meier estimator with adjusted comparison of risk computed based on multiple Cox regression. Furthermore, 30-day standardized mortality against eGFR was splined in patients on MRA and no MRA treatment, respectively, with restricted cubic spline of eGFR. Models were adjusted for age, sex, comorbidities (ischemic heart disease, diabetes, heart failure, stroke, malignancy, and hypertension), and concomitant medication (ACEi, ARBs, diuretics, calcium channel antagonists, metformin, insulin, and SGLT2 inhibitor). On the basis of the reported hazard ratios, 30-day risks of mortality were computed standardized to the distribution of risk factors of all patients in the sample. Subgroup analyses were performed with stratification on sex, age, and specific risk factors. To assess the underlying cause of hyperkalemia, we evaluated diagnosis codes in patients hospitalized within 7 days following measurement of hyperkalemia.
All statistical analyses were conducted using the SAS statistical software (version 9.4; SAS Institute, Cary, North Carolina, USA) and R [Version 4.0.1; R Core Team (2019)]. The level of statistical significance was set at 5%, and all statistical tests were two-tailed.
EthicsIn Denmark, register-based studies do not require preexisting ethical approval. All pseudo-anonymized data were linked, stored, and analyzed securely within a research platform administered through Statistics Denmark. All code is shared openly for review and re-use under the Statistics Denmark license. As detailed patient data hold potential for re-identification, full data sharing is not possible. Ethical approval is not required for retrospective registry-based studies in Denmark. Employment of study data is covered by a Data Processing Agreement between The Capital Region of Denmark and The Danish Data Protection Agency (ref. P-2019–191) [23].
RESULTSA total of 47 830 patients with hyperkalemia at least 6.0 mmol/l were identified in the Danish National Patient register. Plasma creatinine permitting estimation of renal function was available in 32 426 (67.8%) patients. In the subset of patients with available plasma creatinine, gender distribution was 57.5% male, median age was 73.7 [IQR 63.9–81.9] years, median eGFR was 56 [IQR 34–82] ml/min/1.73 m2 with an eGFR less than 60 ml/min/1.73 m2 in 53.5% of patients, and a prevalence of hypertension and diabetes of 19.0 and 34.1%, respectively.
Nested case--control studyPatients with confirmed hyperkalemia and known renal function (n = 32 426) were matched 1 : 4 with age, sex, diabetes, and hypertension-matched controls with known renal function without hyperkalemia from the general population (n = 127 038). Baseline characteristics are provided in Table 1. Median eGFR was 56 [IQR 34--82] ml/min/1.73 m2 and 77 [IQR 60–90] ml/min/1.73 m2 in cases and controls, respectively. Six thousand six hundred seventy-eight (20.6%) of the cases and five thousand two hundred forty-six (4.2%) of the controls were treated with MRA.
TABLE 1 - Nested case--control: Baseline characteristics of patients with hyperkalemia and matched controls Variables Patients with hyperkalemiaMRA treatment was associated with increased rate of hyperkalemia; hazard ratio 6.13 (95% CI 5.88–6.40, P < 0.001) (No MRA treatment as reference). In patients with diabetes, hazard ratio was 4.35 (95% CI 4.09–4.63, P < 0.001) as compared with hazard ratio 8.13 (95% CI 7.67–8.62, P < 0.001) in patients without diabetes, and in patients with hypertension, hazard ratio was 4.59 (95% CI: 4.25–4.96) P < 0.001, compared with 6.90 (95% CI: 6.56–7.25) P < 0.001 in patients without hypertension. Notably, the association of MRA treatment with increased rate of hyperkalemia persisted across all strata of renal function. eGFR-stratified results, as well as sex-specific and age-specific results are summarized in Table 2. The associated adjusted hazards for hyperkalemia in patients on MRA and no treatment, respectively, are illustrated for eGFR as a continuous variable in Figure S2, https://links.lww.com/HJH/C361. Furthermore, results for subgroups with specific risk factors stratified on eGFR are provided in the supplemental material (Table S2, https://links.lww.com/HJH/C361). Principal results remained unchanged in subgroup analyses stratified on age, sex, and specific risk factors.
TABLE 2 - Nested case--control: Hazard ratios for rate of hyperkalemia in mineralocorticoid receptor antagonist treated patients compared to patients without mineralocorticoid receptor antagonist treatment, stratified on eGFR All patients eGFR strata Hazard ratio P ≥ 60 ml/min/1.732 8.28 (7.78--8.81) < 0.001 45–59 ml/min/1.732 5.12 (4.67--5.62) < 0.001 30–44 ml/min/1.732 3.58 (3.23--3.97) < 0.001 <30 ml/min/1.732 1.89 (1.60--2.23) < 0.001 Sex-specific strataPatients without MRA treatment with corresponding eGFR.
All 32 426 patients with verified hyperkalemia were subsequently followed for assessment of 30-day mortality. Baseline characteristics stratified by MRA treatment are provided in Table 3. Overall, unadjusted 30-day mortality was 30.3% (n = 9825). eGFR-stratified Kaplan--Meier curves in patients with and without MRA treatment, respectively, are provided in Figure S1, https://links.lww.com/HJH/C361 in the supplemental materials, https://links.lww.com/HJH/C361.
TABLE 3 - Retrospective cohort of patients with hyperkalemia: Baseline characteristics of patients with hyperkalemia (plasma potassium ≥6 mmol/l) stratified by mineralocorticoid receptor antagonist Variables No MRAFor patients with hyperkalemia, MRA treatment was associated with the following hazard ratio of 30-day mortality: eGFR at least 60 ml/min/1.73 m2: hazard ratio 0.67 (95% CI: 0.62–0.72), P < 0.001; eGFR 45–59 ml/min/1.73 m2: hazard ratio 0.58 (95% CI: 0.51–0.65), P < 0.001, eGFR 30–44 ml/min/1.73 m2: hazard ratio 0.63 (95% CI: 0.56–0.70), P < 0.001; and eGFR less than 30 ml/min/1.73 m2: hazard ratio 0.86 (95% CI: 0.74–1.00), P = 0.047. Standardized 30-day risk of mortality and risk differences between patients with and without MRA treatment, stratified by eGFR, are provided in Table 4. The risk differences are furthermore illustrated in Fig. 1, and the associated adjusted hazards for 30-day standardized mortality in patients on MRA and no treatment, respectively, are illustrated for eGFR as a continuous variable in Figure S3, https://links.lww.com/HJH/C361. Across all strata of eGFR, MRA treatment was associated with a lower risk of subsequent 30-day mortality. Hazard ratios and standardized risks of 30-day mortality for the predefined subgroups are reported in the supplemental material (Table S3a-b and S4a-f, https://links.lww.com/HJH/C361). Of note, principal results remained unchanged in subgroup analyses stratified on age, sex, and specific risk factors.
TABLE 4 - Retrospective cohort study: Standardized 30-day risk of death following hyperkalemia for patients with and without mineralocorticoid receptor antagonist, respectively eGFR (ml/min/ 1.73 m2) 30-day risk % (95% CI) for non-MRA 30-day risk % (95% CI) for MRA Risk difference (%) (95% CI) P ≥60 39.8 (38.8–40.8) 29.3 (27.8–31.1) 10.4 (8.4–12.4) < 0.001 45–59 32.0 (30.7–33.1) 20.3 (18.7–22.4) 11.6 (8.8–14.2) < 0.001 30–44 28.8 (27.5–31.2) 19.5 (17.9–21.7) 9.3 (6.6–11.5) < 0.001 < 30 22.5 (21.4–23.4) 19.7 (17.4–22.5) 2.7 (0.2–5.3) 0.0695% CI, 95% confidence interval.
FIGURE 1: For each estimated glomerular filtration rate stratum, the standardized risk of 30-day mortality has been compared between patients with and without MRA treatment, and the difference is depicted in the graph of the respective renal function.
An overview of diagnoses in patients hospitalized for 7 days or less following measurement of hyperkalemia is provided in the supplemental materials (Table S5, https://links.lww.com/HJH/C361). Acute kidney injury including dehydration was demonstrated for 7.6% of hospitalized patients, with infections, cardiovascular disease, and diabetic dysregulation representing 34.2% of hospitalizations.
DISCUSSIONOn the basis of data from a nationwide population sample, treatment with MRA was associated with an increased rate of hyperkalemia in patients with renal insufficiency. However, subsequent 30-day mortality was lower in patients treated with MRA across all stages of CKD.
The risk of hyperkalemia associated with MRA treatment has previously predominantly been studied in patients with heart failure, with reported treatment benefit on mortality despite increased rates of hyperkalemia [11,24]. MRAs have recently gained broader indications beyond heart failure and are now furthermore recommended to treat patients with resistant hypertension and type 2 diabetes accompanied by albuminuria. Therefore, an increasing number of patients must be expected to initiate MRA treatment. This underscores the importance of conducting a study that does not primarily focus on patients with heart failure [13,25]. Moreover, as the new and upcoming nonsteroid MRAs, currently undergoing phase II and III clinical trials, have an equal risk of hyperkalemia, the clinical significance of this study adds to the field of kidney-protective treatment. Furthermore, CKD remains common in patients with heart failure, leading to increased discontinuation due to uncertainties related to a perceived increased risk of hyperkalemia [8,26]. However, evidence suggests that overactivation of the mineralocorticoid receptor causes progression in CKD through inflammation and fibrosis [5]. Hence, patients with CKD are, in addition to the cardiac benefits of MRA, suggested further to have renal benefits from MRA treatment [3,5]. Hyperkalemia is particularly feared due to its proarrhythmic properties [11]. Yet, the Fidelio study, investigating the effects of the nonsteroid MRA Finerenone on renal outcomes in patients with CKD and type 2 diabetes, demonstrated survival benefit despite a doubling of the rate of hyperkalemia [3]. Apart from mortality, studies have furthermore shown MRA to be associated with reduced risk of all-cause hospitalization in elderly patients with heart failure and concomitant diabetes or CKD, despite observed increase in risk of hospitalization due to AKI and hyperkalemia [27]. Existing studies have suggested possible divergent effects of hyperkalemia on risk of cardiac dysrhythmia dependent on context and concurrent medications [26]. In a recent study, hyperkalemia at least 5.5 mmol/l was found to be associated with an increased risk of 1-year mortality in patients with heart failure; however following adjustment for discontinuation of RAAS inhibitors, the association dissipated, most significantly for MRAs, suggesting a possible greater risk attributable to discontinuation of MRA as opposed to risk attributable to the consequences of hyperkalemia itself [11].
The present study found the rate of hyperkalemia to be higher in patients treated with MRA for all strata of eGFR, with comparable differences between patients treated with and not treated with MRA gradually decreasing with aggravation of renal impairment consistent with an overall greater propensity for hyperkalemia in advanced CKD regardless of MRA exposure. Despite the increased rate of hyperkalemia, subsequent 30-day mortality was comparably lower for patients treated with MRA across all strata of eGFR. Nonetheless, overall unadjusted 30-day mortality was substantial 30.3% (n = 9825), as such underscoring the severity associated with hyperkalemic disturbances.
That patients with higher renal function were found to have higher mortality, which might partly be explained by the smaller proportion of this group having hyperkalemia caused by renal insufficiency and medication thus giving more share to more severe and lethal causes of hyperkalemia, including critical illness, rhabdomyolysis, and necrosis [28]. Another phenomenon that might contribute to the present findings is physiological adaption in CKD. Because patients with CKD are more likely to be exposed to hyperkalemia for a longer period of time, it is suggested that their normal range of potassium levels should be considered higher than in other patients as they adapt to potassium imbalance [4]. This is further supported by studies suggesting that rapid increases in serum potassium, for example, in association to intravenous potassium loading, massive cell turnover or critical illness is more lethal [26]. On the contrary, patients with CKD could be more sensitive to toxicity due to hyperkalemia because of their, a priori, higher risk of other metabolic derangement including hypocalcemia, acidosis, and elevated uremic solutes [26]. However, other studies have shown that despite an increased risk of hyperkalemia, patients with CKD have a reduced 1-day mortality if measured serum potassium at least 5.5 mmol/l compared with patients without CKD, suggesting a better tolerance of hyperkalemia [29].
With advanced nephropathy follows loss of renal capacity for potassium excretion and increased risk of hyperkalemia. Guidelines overall advocate dietary restriction only in patients with end-stage kidney disease (ESKD) and in following hyperkalemia also patients with advanced nephropathy based on expert opinion [30,31]. However, few studies have investigated the effects of dietary potassium restriction [32], and retrospective studies have generally demonstrated a weak association of dietary potassium intake and potassium concentration in patients with CKD and ESKD [33–36]. Furthermore, several compensatory mechanisms counteract potassium accumulation in CKD including increased intestinal excretion [37]. Finally, trials evaluating the impact of modification of potassium intake in CKD remain nonconclusive [38–40].
Overall, concerns related to MRA use in patients with CKD may be associated with unfavorable benefit-harm; particularly due to an overriding tendency for discontinuation of treatment in patients with hyperkalemia, which might be unfavorable as discontinuation of treatment may increase mortality more than hyperkalemia itself [11]. In this study, hyperkalemia in patients exposed to MRA was associated with significantly lower risk of 30-day mortality compared with hyperkalemia in absence of MRA. It is possible that the mortality risk of hyperkalemia in MRA-treated patients may be lower due to the possibility of discontinuing the medication in these patients and thus the triggering cause, which is not an option in those patients whose hyperkalemia is caused by more severe, underlying disease. Nevertheless, the risk of 30-day mortality after hyperkalemia was of a substantial magnitude, and the definitive benefit of reinstating MRA treatment in patients with prior hyperkalemia remains untested in clinical trials.
Although models were adjusted for multiple relevant confounders, retrospective analyses remain prone to residual confounding, and important clinical parameters may remain unaddressed. Overall, a major limitation in terms of investigating MRAs in nonrandomized data is the lack of a nonbiased comparator, particularly considering the explicit indication biases associated with MRA treatment. However, comparisons remain of interest, particularly in light of the increasing advocacy of MRA treatment in patients with advanced nephropathy. Detection of hyperkalemia is only possible in patients with laboratory work-up and does not account for the occurrence of the metabolic disturbances at unobserved times leading to possible ascertainment biases. Furthermore, eGFR was used as a surrogate estimate for progression in CKD, and both eGFR and hyperkalemia were evaluated based on only one off measure. This study has several implications. Data pertaining to incidence and prevalence rates of hyperkalemia in general populations remain limited, with estimates based on admitted patients in North America suggesting prevalent rates of 2–4% [29,41]. However, rates plausibly vary dependent on case mix, with retrospective data supporting increased prevalence rates in CKD [42,43]. Accurate identification of hyperkalemia remains uncertain due to risk measurement error, with falsely elevated potassium demonstrated for a substantial number of patients following repeat measurement in observational studies [44]. Incidence of true hyperkalemia has been shown to increase with declining eGFR [45]. Nonetheless, the impact of pseudohyperkalemia or falsely elevated potassium remains unaddressed, with potential for overestimation of hyperkalemia overall. However, spurious hyperkalemia is plausibly unrelated to treatment with MRA, as such limiting bias on the observed results in our cohort. To reduce the risk of a temporary renal impairment, the eGFR was only applicable if measured between 2 years and 30 days before the employed plasma potassium; assessment of eGFR based on the last recorded plasma creatinine recorded until 7 days before index in data clusters has been appraised as a reliable rule for estimating baseline kidney function [46]. The impact of diet remains unaddressed in our results due to lack of access to data surveying diet in our cohort. Furthermore, duration of MRA treatment and dosages were not accounted for, and the effects of treatment pausation and discontinuation remain unaddressed.
CONCLUSIONMRA treatment was associated with an increased rate of hyperkalemia across all strata of renal impairment. However, subsequent 30-day mortality after hyperkalemia was lower in patients on MRA treatment in comparison to patients without MRA treatment.
ACKNOWLEDGEMENTSNo acknowledgments.
Conflicts of interestD.H.K. has received speaker honoraria from Bayer A/S outside the submitted work.
N.C. has received lecture fees from Bristol Squibb Myers and AstraZeneca outside the submitted work.
M.S. has received lecture fees from Novo, Novartis, Astra, and Bohringer outside the submitted work.
C.T.P. reports grants for studies from Bayer and Novo Nordisk not related to the current study.
M.L. has received research support from Boehringer Ingelheim, Bayer and Merck Sharp & Dohme, and lecture fees from AstraZeneca, Bayer and Boehringer Ingelheim, and consultancy fees from Bayer, Boehringer Ingelheim, and AstraZeneca. All fees are given to Holbaek Hospital.
REFERENCES 1. Banerjee D, Rosano G, Herzog CA. Management of heart failure patient with CKD. Clin J Am Soc Nephrol 2021; 16:1131–1139. 2. Bomback AS, Klemmer PJ. Mineralocorticoid receptor blockade in chronic kidney disease. Blood Purif 2012; 33:119–124. 3. Bakris GL, Agarwal R, Anker SD, Pitt B, Ruilope LM, Rossing P, et al. Effect of finerenone on chronic kidney disease outcomes in Type 2 diabetes. N Engl J med 2020; 383:2219–2229. 4. An JN, Lee JP, Jeon HJ, Kim DH, Oh YK, Kim YS, et al. Severe hyperkalemia requiring hospitalization: predictors of mortality. Crit Care 2012; 16:R225. 5. Barrera-Chimal J, Girerd S, Jaisser F. Mineralocorticoid receptor antagonists and kidney diseases: pathophysiological basis. Kidney Int 2019; 96:302–319. 6. Rossignol P, Dobre D, McMurray JJV, Swedberg K, Krum H, van Veldhuisen DJ, et al. Incidence, determinants, and prognostic significance of hyperkalemia and worsening renal function in patients with heart failure receiving the mineralocorticoid receptor antagonist eplerenone or placebo in addition to optimal medical therapy: results from the Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF). Circ Heart Fail 2014; 7:51–58. 7. Vardeny O, Claggett B, Anand I, Rossignol P, Desai AS, Zannad F, et al. Incidence, predictors, and outcomes related to hypo- and hyperkalemia in patients with severe heart failure treated with a mineralocorticoid receptor antagonist. Circ Heart Fail 2014; 7:573–579. 8. Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998; 32: (5 Suppl 3): S112–S119. 9. Agarwal R, Filippatos G, Pitt B, Anker SD, Rossing P, Joseph A, et al. Cardiovascular and kidney outcomes with finerenone in patients with type 2 diabetes and chronic kidney disease: the FIDELITY pooled analysis.
Comments (0)