Medical Management of Heart Failure with Reduced Ejection Fraction in Patients with Advanced Renal Disease

Abstract

Large randomized clinical trials (RCT) supporting guidelines for the management of HFrEF have typically excluded patients with advanced CKD. Patients with concomitant advanced CKD and HFrEF experience poor cardiovascular outcomes and mortality relative to either disease in isolation and have been shown to consistently receive lower rates of HFrEF guideline-directed medical therapy (GDMT). In this review, we evaluate recent evidence for the use of GDMT in patients with HFrEF and advanced CKD approaching dialysis from RCTs and observational cohorts. We also discuss the limitations and challenges inherent in the evidence for GDMT in this population and offer guidance to clinicians for proper clinical use and future research directions.

Introduction

Patients with HF experience significant morbidity and mortality associated with CKD. In patients with HFrEF, the prevalence of CKD stage ≥G4 (Table 1) is approximately 10%.(1,2) Pre-existing HF predisposes patients to acute kidney injury (AKI) and the development of CKD and end-stage renal disease (ESRD) due to impaired renal hemodynamics.(3,4) Similarly, patients with impaired renal function are prone to sodium and fluid retention and thus, more likely to develop HF.(5) Co-morbid CKD is an independent predictor of both short-term and long-term cardiovascular outcomes and death in patients with HF, with more advanced renal disease conferring a worse prognosis.(6,7) Patients with HFrEF in particular experience a greater increase in morbidity and mortality when advanced CKD is present relative to other HF subtypes.(6)

Table 1.

Chronic Kidney Disease Categories.

eGFR Categories	eGFR (mL/min/1.73m2)	Terms
G1	>90	Normal or high
G2	60–89	Mildly decreased
G3a	45–59	Mildly to moderately decreased
G3b	30–44	Moderately to severely decreased
G4	15–29	Severely decreased
G5	<15	Kidney failure
Albuminuria Categories	Albumin Excretion Rate (mg/24 hours)	Albumin-to-Creatinine Ratio		Terms
		(mg/mmol)	(mg/g)
A1	<30	<3	<30	Normal to mildly increased
A2	30–300	3–30	30–300	Moderately increased
A3	>300	>30	>300	Severely increased

Adapted with permission from the Kidney Disease Improving Global Outcomes Guidelines.(2) In the absence of evidence of kidney damage, neither eGFR category G1 nor G2 fulfill the criteria for CKD. G2 and A2 terms are relative to young adult level.

^*Abbreviations: eGFR, estimated glomerular filtration rate.

Despite the high prevalence of concomitant HFrEF and CKD, routine GDMT of this cohort is often lower than in the general HFrEF population.(8,9) There remains an unmet need to characterize challenges and recent therapeutic advances from clinical trials in the HF population including those with CKD, and to evaluate real-world evidence from registries and observational cohorts that can inform routine practice. We evaluated contemporary literature on pharmacologic management of patients with HFrEF and advanced CKD, defined as those with CKD≥G4 not on renal replacement therapy. We searched MEDLINE (via Pubmed) from January 1985 to August 2018 using Medical Subject Headings and key words, focusing on relevant terms for this topic (see Supplementary Appendix). We also manually searched reference lists of pertinent reviews and studies to find any relevant citations. All citations were screened and evaluated by one reviewer (A.M.H.) to select relevant studies.

We discuss the limitations of the current evidence, future research directions, and offer clinicians guidance regarding the safe use of GDMT in this high-risk population. We focus on pharmacotherapy with renin-angiotensin system inhibitors (RAS-I), angiotensin receptor blockers/neprilysin inhibitors (ARNIs), mineralocorticoid receptor antagonists (MRAs), and beta-blockers, and briefly discuss device therapy in this population. There is little to no evidence for other HF pharmacologic therapies, such as isosorbide dinitrate and ivabradine, in advanced CKD and thus, these therapies were not included in this review. Detailed characteristics and outcomes for relevant studies are outlined in Table 2, with studies of note highlighted in the body of the text.

Table 2.

Outcomes for Heart Failure Therapies in Patients with HFrEF and CKD.

Study	Design	N, Follow-Up	Population and Subgroup	Renal Criteria	Adjustment	Intervention, Comparison	Primary Outcome	Outcome(s) of Interest (95% CI)
Renin-Angiotensin System Inhibitors
Swedberg, et. al 1990(24)	PC-RCT* SG analysis	253, 6 months	NYHA IV HFrEF, SG: sCr >1.39 mg/dL	sCr <3.4 mg/dL; 12% estimated with CKD G4	--	Enalapril vs placebo	ACM	30% vs. 55%, p=0.004
Hillege et. al 2006(25)	PC-RCT SG analysis	2680, 37.7 months median	Age >18, NYHA II-IV, (LVEF≤40%, n=1656), SG: eGFR ≤45 mL/min/1.73m2	sCr <3.0 mg/dL	--	Candesartan vs placebo	Cardiovascular death, unplanned readmission, or ACM	No significant interaction between candesartan, sCr, and primary outcome (P=0.84)
Edner et. al 2015(26)	CE-OC	1204, 1 year	LVEF ≤39%	CKD G4–5 or sCr >2.5 mg/dL	Propensity-matched	RAS-I vs no RAS-I prescription at discharge	ACM	HR=0.76 (0.67-0.86)
Masoudi et. al 2004	CE-OC	1258, 1 year	Age ≥65, LVEF <40%, after HF hospitalization	sCr >2.5 mg/dL	Multivariable regression	ACE-I vs no ACE-I prescription at discharge	ACM	RR=0.65 (0.51-0.80)
Berger et. al	CE-OC	381, 1 year	Age 35–84, with HF	CKD G4, n=238;	Risk adjustment	RAS -I vs no RAS-I	30-day ACM	CKD G4: 9.4% vs.
2007(19)			exacerbation (n=137 with LVEF <35%)	CKD G5, n=143	models	prescription in-hospital		18.5%, p=0.008 CKD G5: 11.9% vs. 22.8%, p=0.03
McAlister et. al 2004	CE-OC	754, 2.5 years median	Outpatients with HF (57% with LVEF ≤35%)	CKD ≥G3; n=118 with CKD G4–5	Multiple logistic regression	ACE-I vs. no ACE-I at discharge	1-year ACM	OR=0.46 (0.26-0.82)
Ahmed et. al 2012	CE-OC	541, 8 years	Medicare cohort, LVEF ≤45% after HF hospitalization	eGFR ≤38.5 mL/min/1.73m2 n=541	Propensity-matched	RAS-I at target dose vs. no RAS-I prescription at discharge	ACM	HR=0.82 (0.70-1.00); reduced ACM for those at target doses
Angiotensin Receptor Blocker/Neprilysin Inhibitors
Damman et. al 2018(31)	CER-RCT SG analysis	2745	NYHA II-IV, LVEF ≤40%	CKD G3–4	--	Valsartan/ sacubitril vs enalapril	Cardiovascular death and HF hospitalization	HR=0.79 (0.69–0.90)
Solomon et. al 2016(32)	Meta-analysis	14,472	IMPRESS, OVERTURE, PARADIGM-HF trial participants	--		ARNI vs single RAS-I therapy	Adverse events	Increased symptomatic hypotension, but decreased serum potassium elevation and renal impairment
Mineralocorticoid Receptor Antagonists
Lu et. al 2016(41)	Meta-analysis	4935	Adults with CKD	CKD G1–5	--	MRA vs non-MRA treatment	ACM, MACE, hyperkalemia	ACM: RR=0.78 (0.62–0.97) MACE: RR=0.65 (0.50–0.83) Hyperkalemia: RR=2.32 (1.83-2.94) Not significant in patients with HF
Inampudi et. al 2014(42)	CE-OC	1140, 1 year	Hospitalized patients with HF and LVEF <45%	eGFR <45 mL/min/1.73m2	Propensity-matched	Spironolactone vs. no spironolactone at discharge	1-year allcause readmission	HR=1.36 (1.13–1.63) eGFR <15 mL/min/1.73m2: HR=4.75 (1.84–12.28)
Cooper et. al 2017(43)	CE-OC	16,848, 3 years	Age>65, HF hospitalization	sCr >2.0 mg/dL or type 2 diabetes mellitus (mean 1.9 mg/dL)	Inverse probabilityweighted proportional models	MRA vs. no MRA prescription at discharge	ACM (30-day, 1-year, 3-year)	No differences in any ACM; 3-year all-cause readmission: HR=0.94 (0.89-0.98) Long-term hyperkalemia readmission: HR=1.30 (1.11-1.53)
Beta-Blockers
Ghali et. al 2009(49)	PC-RCT SG analysis	493	Age 40–80 NYHA II-IV with LVEF <40%	CKD ≥G3b (mean eGFR 36.6 mL/min/1.73m2)	--	Metoprolol succinate vs placebo	ACM, HF mortality, all-cause hospitalization	ACM: HR=0.41 (0.25–0.68) HF mortality: HR=0.25 (0.10-0.62) All-cause hospitalization: HR=0.61 (0.47-0.79)
Erdmann et. al 2001	PC-RCT SG analysis	63, 1.3 years	NYHA III-IV, LVEF ≤35% on ACE-I therapy	CKD G4–5	--	Bisoprolol vs placebo	ACM	RR=0.59 (0.301.18) More permanent treatment withdrawals in eGFR <30 mL/min/1.73m2
Castagno et. al 2010(50)	PC-RCT SG analysis	450, 1.3 years	NYHA III-IV, LVEF ≤35% on ACE-I therapy	CKD ≥G3b (mean 38.4 mL/min/1.73m2)	--	Bisoprolol vs placebo	ACM, Composite of ACM or HF hospitalization	HR=0.71 (0.48-1.05) ACM or HF hospitalization: HR=0.72 (0.53-0.99)
Badve et. al 2011(51)	Meta-analysis	5972	Patients with chronic HFrEF	CKD ≥G3	--	Beta-blocker vs placebo	ACM	RR=0.72 (0.64-0.80) Cardiovascular mortality: RR=0.66 (0.49–0.89)
Wali et. al 2011	Meta-analysis	1116	Patients with systolic left ventricular dysfunction	CKD ≥G3b	--	Carvedilol vs placebo	ACM	HR=0.94 (0.72-1.23) HF mortality: HR=0.86 (0.61-1.21)
Nagatomo et. al 2017(52)	CE-RCT	360, 3.8 years	Age 20–80, NYHA II-III, LVEF <40%	Stratification at eGFR =60mL/min/1.73m2	--	2.5 mg vs 5 mg vs 20 mg carvedilol	Change in LVEF	No difference in LVEF increase between eGFR >60 mL/min/1.73m2 and eGFR <60 mL/min/1.73m2 for all groups
Chang et. al 2013(9)	CE-OC, 2.4 years median	668	Age >18, newly diagnosed HFrEF, LVEF <40%	CKD ≥3 or proteinuria, nondialysis; 6.3% CKD G4–5	Nested covariable models	New-user of betablocker vs. no betablocker	ACM	HR=0.75 (0.51-1.12) Death or HF hospitalization: HR=0.67 (0.51-0.88)
McAlister et. al	CE-OC	419, 2.5 years	Outpatients with HF	CKD ≥G3; n=118	Multiple logistic	Beta-blocker vs. no	1-year ACM	OR=0.40 (0.23-
2004		median	(57% with LVEF≤35%)	with CKD G4–5	regression	beta-blocker prescription at discharge		0.70)

Description of trials and observational studies investigating medical therapies in patients with HFrEF and CKD. Studies not cited in the body of the text are referenced in order of appearance in the online supplement.

^*Abbreviations: ACE-I, angiotensin-converting-enzyme inhibitor; ACM, all-cause mortality; ARNI, angiotensin receptor blocker/neprilysin inhibitor; CE-OC, comparative effectiveness observational cohort; CE-RCT, comparative effectiveness randomized controlled trial; CI, confidence interval; eGFR, estimated glomerular filtration rate; HR, hazard ratio; LVEF, left ventricular ejection fraction; MACE, major adverse cardiac events; MRA, mineralocorticoid receptor antagonist; NYHA, New York Heart Association class; OR, odds ratio; PC-RCT, placebo-controlled randomized controlled trial; RR, risk ratio; RAS-I, renin-angiotensin system inhibitor; sCr, serum creatinine; SG, subgroup.

Limitations of Current Evidence in Patients with HFrEF and Advanced CKD

Many pivotal clinical trials for guideline-directed HFrEF therapies excluded patients with severe kidney disease, resulting in limited evidence for contemporary GDMT in those with advanced CKD and HFrEF (see Central Illustration).(10) Dedicated clinical trials in advanced CKD and HFrEF are uncommon and thus, recommendations often rely upon extrapolation from populations with low numbers or without advanced CKD or observational studies with inherent limitations as reviewed below.

Central Illustration. Limitations in the Evidence for Patients with HFrEF and Advanced CKD.

Current challenges affecting the paucity of evidence in patients with HFrEF and advanced CKD contributing to reduced utilization of medical therapy in this vulnerable population.

Many completed trials, subgroup analyses, and observational studies in advanced CKD and HFrEF used serum creatinine (sCr) cut-offs to define levels of renal insufficiency. While a common clinical approach, sCr alone is an imperfect measure of kidney function influenced by demographic and patient-specific characteristics such as age, sex, race and muscle mass, used to derive direct estimates of kidney function as estimated glomerular filtration rate (eGFR). With widespread adoption of eGFR reporting and increased awareness, recent observational studies and trials largely incorporate eGFR, but challenges remain in the included older studies.

Similarly, there is increasing recognition of the importance of proteinuria/albuminuria in the evaluation of kidney dysfunction in patients with HFrEF. Recent studies have identified even modest increases in proteinuria/albuminuria are independently associated with the development of HFrEF, and with morbidity and mortality in those with HFrEF.(11–13) Unfortunately, many previous studies evaluating GDMT in patients with advanced CKD, who experience increased proteinuria relative to the general population, do not account for or report data on proteinuria. Further research should determine the risk and benefits of GDMT when proteinuria is present in those with advanced CKD.

Lack of universal outcome definitions represent another key limitation. For instance, AKI is a common feature of decompensated HFrEF often evaluated as an adverse effect or study outcome. Definitions of the degree of kidney function decline, acuity, or timeframe of the change vary, and whether the change must be sustained or confirmed on repeat testing. Recent international guidelines in kidney disease may provide standardized outcome criteria that could reduce this variation in future studies.(14) Furthermore, worsening renal function suggestive of AKI may not reflect true kidney injury in decompensated HFrEF or during initiation of RAS-I therapy.(15,16) Conversely, improving measures of function during decompensated HFrEF have been associated with poor outcomes, reflecting the uncertainty surrounding renal function measurement in this clinical setting.(17)

Importantly, many results from observational studies are limited by potential confounding, as the sickest, most fragile patients, already at increased risk for poor outcomes, may be less likely to receive GDMT due to concerns for adverse effects. This reality may result in attribution of morbidity and mortality benefit to GDMT in an observational cohort that is truly due to unaccounted factors in participant baseline status. Circumventing this limitation is challenging and requires high-quality data sources with capture of relevant clinical variables and rigorous analytic techniques including new-user designs, propensity scoring, marginal structural models, among others.

Several medications can cause adverse effects in advanced CKD that affect the risk-benefit calculation. For instance, RAS-I therapies may result in or exacerbate hyperkalemia or precipitate AKI in this vulnerable population. Furthermore, due to renal clearance of many medications used in HFrEF, dose adjustments and monitoring may vary greatly based on the degree of renal impairment, as outlined in Figure 1. Concerns for adverse events and polypharmacy often shift the cost/benefit ratio of medication prescription for these patients.

Figure 1. Recommended Dose Adjustment and Monitoring Considerations in Patients with HFrEF and CKD G4–5.

Dose adjustments and monitoring in patients with HFrEF and CKD G4–5. Adjustments refer to initial dosage, with titration as tolerated based on safety.

^†Reduced rate of titration recommended for captopril.

*Abbreviations: ACE-I, angiotensin-converting-enzyme inhibitor; ARNI, angiotensin receptor blocker/neprilysin inhibitor; eGFR, estimated glomerular filtration rate; MRA, mineralocorticoid receptor antagonist.

Thus, while it is known those with CKD and HFrEF experience worse outcomes on average, unsurprisingly they receive lower rates of GDMT. A recent national registry study of ESRD patients found 40% patients approaching dialysis were receiving an angiotensin-converting-enzyme inhibitor (ACE-I) or angiotensin receptor blocker (ARB), regardless of HFrEF history, and 60–75% of those with HFrEF were receiving a beta-blocker.(9) Other studies in outpatients with advanced CKD and HF have found prescription rates of 50%−70%, often 20% lower than their non-CKD counterparts.(18,19) Studies have shown that GDMT improves after dialysis initiation.(19) Given the limited life-expectancy and co-morbid conditions often experienced by these patients, in combination with limited clinical trial evidence for efficacy in this group, clinicians may be dissuaded from initiating GDMT.

Renin-Angiotensin System Inhibitors

ACE-Is and ARBs in large randomized controlled trials (RCT) have decreased cardiovascular events, HF hospitalizations, and mortality while improving quality of life in those with HFrEF.(20,21) However, many RCTs establishing ACE-Is and ARBs as integral therapies excluded patients with advanced CKD.(10) While subgroup analyses of many major trials suggested benefit in those with CKD ≥G3 similar or greater to that in the larger trial,(22,23) data for those with advanced CKD are scarce.

Of the RAS-I RCTs, only a small proportion of two trials included patients with advanced CKD. In a subgroup analysis of the CONSENSUS trial, enalapril significantly reduced all-cause mortality in patients with sCr >1.39 mg/dL compared to placebo, but did not have a significant effect in the those with sCr <1.39 mg/dL.(24) A post-hoc CHARM analysis found no interaction between renal function and the cardiovascular benefits of candesartan in those with HF and CKD ≥G3b. Candesartan resulted in a greater absolute risk reduction in those with significant renal dysfunction who experienced overall worse outcomes.(25)

Only observational data have been reported for the population with advanced CKD and HFrEF. In a propensity-matched cohort study of patients with HFrEF and CKD ≥G4, RAS-I therapy was associated with increased 1-year survival compared to non-users, with similar efficacy to that seen in patients without CKD, particularly in those with NYHA-III symptoms.(26) Another cohort study found RAS-I therapy during HF hospitalization was associated with decreased 30-day all-cause mortality in patients with CKD G4–5. This study did not independently evaluate patients with HFrEF, however.(19) Other studies have demonstrated improved outcomes and a potentially higher absolute risk reduction in this population with RAS-I therapy (see Table 2).

Though the aforementioned studies suggest therapeutic benefit for RAS-I therapy in those with HFrEF and advanced CKD, concern for increased adverse renal outcomes and electrolyte disturbances are substantial. In a subgroup of patients with unspecified HF from a population-based study, those with “moderate-to-severe” CKD had a higher risk of a >30% increase in sCr relative to other groups, and this adverse event conferred an increased risk of death in those with co-morbid HF.(27) Further, a meta-analysis of patients with CKD ≥G3, in which HF history was unspecified, found a roughly doubled risk of hyperkalemia with RAS-I monotherapy relative to placebo or active controls, though RAS-I monotherapy also reduced all-cause mortality.(28) However, in an outpatient cohort with HFrEF and CKD G3–4, chronic RAS-I therapy or up-titration was not associated with worsened renal outcomes or hyperkalemia after one year, though this study may have excluded participants with previous adverse effects related to RAS-I administration.(29)

Recently, ARNIs have been shown to effectively reduce HFrEF hospitalizations and mortality.(30) However, the tolerability and safety of this new class of medications in patients with co-morbid advanced CKD is unclear. A PARADIGM trial subgroup analysis found patients with CKD G3 receiving sacubitril/valsartan had decreased rates of cardiovascular mortality and HF hospitalization compared to those on enalapril in addition to reductions in all-cause mortality, noting a higher absolute risk reduction than those without CKD.(30,31) Further, the rate of CKD progression was reduced in those treated with sacubitril/valsartan compared with enalapril, though these patients also experienced an increase in urinary albumin/creatinine ratio. Notably, the increase in albuminuria did not affect the 30-day risk of cardiovascular mortality or HF hospitalization.(31) Despite these encouraging findings, the study excluded patients with CKD G4–5, leaving its applicability to patients with advanced CKD untested. Regarding adverse events, one meta-analysis found decreased rates of renal impairment and potassium elevation with ARNI therapy relative to ACE-I therapy for patients with HFrEF, though ARNI therapy resulted in increased rates of symptomatic hypotension.(32)

Recommendations and Future Research Directions

Given the weak evidence base for ACE-I/ARBs in this population, current ACC/AHA guidelines suggest caution in using RAS-I therapy in those with co-morbid CKD ≥G4.(33)

Kidney Disease:

Improving Global Outcomes (KDIGO) guidelines recommend specialist supervision for RAS-I use in patients with CKD ≥G4, a reduced starting dose for those with CKD ≥G3b, eGFR monitoring and serum potassium measurement within one week of initiation, while advising against routine discontinuation once eGFR reaches <30 mL/min/1.73m².(2,34) However, many of the included studies suggest a survival benefit with the use of RAS-I therapy in this population, and the introduction of therapies to prevent hyperkalemia may lessen safety concerns.(35) Robust trials with adequate safeguards are needed to confirm the potential benefits in patients with HFrEF and advanced CKD.

Current guidelines do not comment on the use of ARNI therapy in those with CKD, but suggests the potential for renal insufficiency with this medication.(33) While the present evidence is reassuring for ARNI therapy in those with CKD G3, further studies are need to elucidate if these beneficial effects extend to those with co-morbid CKD ≥G4.

Mineralocorticoid Receptor Antagonists

In major RCTs, MRAs have consistently improved symptoms while reducing hospitalizations and mortality in patients with HFrEF.(36,37) While sub-analyses of major RCTs showed cardiovascular benefits in those with mild-moderate CKD and HFrEF(10,38,39), concern for adverse effects, notably the heightened risk of hyperkalemia observed after the completion of the Randomized Aldactone Evaluation Study(40), has resulted in exclusion of patients with advanced CKD and HFrEF from major trials. Though a meta-analysis of 12 studies reported MRA use in a general population of patients with CKD ≥G3 was associated with reduced all-cause mortality, these mortality benefits were not significant in studies incorporating a HF sub-analysis.(41)

Observational studies of MRA use in patients with HFrEF and advanced CKD have also been scant, but mainly raise concerns about adverse effects without clear evidence of benefit. A propensity-matched analysis of the Alabama Heart Failure registry found MRA prescription at discharge for hospitalized patients with HFrEF and CKD ≥G3b was associated with increased readmission rates, as well as the combined endpoint of 1-year all-cause mortality or readmission; the risk of 1-year all-cause readmission was greatly increased in those with CKD G5 not on dialysis.(42)

A study incorporating the Get With The Guidelines-HF registry and Medicare claims found MRA discharge prescription, after adjustment for patient characteristics and concomitant medication use, was not associated with differences in 30-day, 1-year, or 3-year mortality for hospitalized patients with HF and co-morbid diabetes mellitus type II or a sCr >2.0 mg/dL. In contrast, MRA therapy was associated with modest decreases in 1-year and 3-year all-cause readmissions. Increased risk of short- and long-term hyperkalemia readmissions was also observed.(43)

Recommendations and Future Research Directions

ACC/AHA guidelines recommend against utilizing MRA therapy in men with sCr >2.5 mg/dL and in women with sCr >2.0 mg/dL due to concern for hyperkalemia, AKI, and inadequate evidence from RCTs.(44) KDIGO guidelines suggest a reduced starting dose for those with CKD ≥G3b, eGFR monitoring and serum potassium measurement within one week of initiation, but advise against routine discontinuation once eGFR reaches <30 mL/min/1.73m².(2) The reviewed studies suggest MRA therapy in those with HFrEF and advanced CKD raises valid safety concerns without clear benefit independent of concomitant use of RAS-I therapy.

Novel Potassium Binders as Adjunct Therapy

The emergence of novel potassium binders such as patiromer and sodium zirconium cyclosilicate may mitigate the hyperkalemia associated with RAS-I agents, particularly in vulnerable CKD populations.(35) In a recent trial, patients with CKD G3–4 receiving RAS-Is who experienced elevated serum potassium levels received either patiromer or placebo; in the HF subgroup, patiromer therapy resulted in a significant decrease in serum potassium levels and hyperkalemia recurrence.(45) Patiromer therapy may also decrease the risk of life-threatening hyperkalemia in those treated with MRAs, and recently has been tolerated in CKD patients up-titrated on spironolactone.(46) Adjunct use of these therapies may help shift the feasibility of RAS-I use in this group. However, these therapies also require consideration of potential adverse events, such as binding of concurrent oral medications, gastrointestinal upset, hypomagnesemia with patiromer, and worsening edema from an increased sodium load with sodium zirconium cyclosilicate.(35)

Beta-Blockers

Beta-blockers have consistently reduced morbidity and mortality in patients with HFrEF.(47,48) Large RCTs evaluating beta-blocker use in HFrEF often included more patients with advanced CKD compared to RAS-I trials, likely due to decreased safety concerns with initiation.(10) Multiple subgroup analyses support the benefits of beta-blockers in those with advanced CKD. A MERIT-HF analysis found participants with CKD ≥G3b receiving metoprolol had significant decreases in all-cause mortality, HF mortality, and all-cause hospitalization.(49) One CIBIS-II analysis likewise found CKD ≥G3b patients had decreased risk of either all-cause mortality or HF hospitalization, with a lower number needed to treat due to poor outcomes in this subgroup.(50)

In a large meta-analysis, beta-blocker therapy in patients with CKD ≥G3a reduced all-cause and cardiovascular mortality risk but increased the risk of bradycardia and hypotension; no difference in hyperkalemia was noted.(51) The effect of increased LVEF with carvedilol treatment seen overall after 56 weeks was not blunted for patients with CKD ≥G3a in the Japanese-CHF trial, which randomized patients with HFrEF to receive three different doses of carvedilol.(52) Each of these studies suffered from few included participants with CKD G4–5, some below 10%, which limits the generalizability to those with advanced CKD.

Observational studies involving beta-blocker therapy in patients with CKD and HFrEF have also been hindered by small proportions of patients with advanced CKD and have reported mixed results for mortality benefits. For example, in a retrospective study of 668 patients with HFrEF and CKD ≥G3, beta-blocker therapy initiation was not associated with decreased all-cause mortality, but it was associated with a decreased risk of the composite outcome of death or HF hospitalization. There were only 47 patients with CKD G4–5 in this study, however.(53)

Recommendations and Future Research Directions

Beta-blockers in HFrEF patients have often shown beneficial effects on morbidity and mortality in those with CKD. AHA/ACC guidelines support beta-blocker therapy in patients with comorbid CKD and HF, and KDIGO guidelines, while recommending beta-blocker therapy, suggest dose reduction by 50% for those with CKD G4–5.(2,34,44) However, though many findings from subgroup analyses of RCTs are promising, there is conflicting observational evidence for efficacy in those with advanced CKD, primarily due to a low number of participants included. Careful monitoring for side effects such as bradyarrhythmias and hypotension are suggested with beta-blocker initiation in this population.

Implantable Device Therapy

Device therapy in patients with HFrEF and advanced CKD is another area of continued controversy, as multiple trials have shown benefit in patient populations with HFrEF, but application to the advanced CKD population is less clear. The risk of sudden cardiac death increases with worsening kidney dysfunction(54), suggesting patients with advanced CKD may benefit from device therapy with an implantable cardioverter-defibrillator (ICD). However, a meta-analysis of 3 primary prevention ICD RCTs found no survival benefit from ICD therapy in patients with CKD ≥G3.(55) Also, a secondary analysis of the Multicenter Automatic Defibrillator Implantation Trial-II found ICD therapy was not associated with decreased mortality in patients with eGFR <35 mL/min/1.73 m².(54) Another meta-analysis of 4 primary prevention ICD trials found the therapeutic benefit was attenuated with increasing co-morbidity burden.(56) While one study showed eGFR improvement with cardiac resynchronization therapy (CRT) in those with CKD G3(57), no studies specifically evaluated the benefit of CRT in advanced CKD.(58) In this population, factors such as life expectancy and periprocedural risk must be balanced when considering the potential benefit of device therapy.

Conclusions

The lack of high-quality evidence for GDMT in patients with HFrEF and advanced CKD leaves little guidance for the management of this vulnerable population. GDMT in this population can yield dangerous side effects, but overall poor outcomes may result in a greater absolute therapeutic benefit. While RAS-I therapy in patients with HFrEF and advanced CKD has generally shown positive outcomes, the role of MRAs is less clear. ARNI therapy requires thorough evaluation in CKD but may promise decreased safety concerns compared to ACEI/ARB therapy. Potassium binding therapies may prove beneficial as adjunct therapy to RAS-I therapies in patients with CKD experiencing hyperkalemia. Beta-blockers, while often beneficial in populations with co-morbid CKD-HFrEF, require careful monitoring for side effects, particularly in patients nearing dialysis. Finally, consideration of benefit from device therapy in this population must weigh the elevated risks of device placement, life expectancy, and comorbidity burden. Clearly, adequately-powered clinical trials with careful adverse event monitoring and follow-up are required to evaluate the benefits of these therapies

Abbreviations

HF

heart failure

CKD

chronic kidney disease

HFrEF

heart failure with reduced ejection fraction

GDMT

guideline-directed medical therapy

ACC/AHA

American College of Cardiology/American Heart Association

CONSENSUS

Cooperative North Scandinavian Enalapril Survival Study

CHARM

Candesartan in Heart Failure-Assessment of Reduction in Mortality and Morbidity

PARADIGM

Prospective Comparison of ARNI with ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure

MERIT-HF

Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure

CIBIS-II

Cardiac Insufficiency Bisoprolol Study II