Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Review - Intervention

Pharmacological interventions for heart failure in people with chronic kidney disease

Authors' declarations of interest

Version published: 27 February 2020 Version history

https://doi.org/10.1002/14651858.CD012466.pub2
Collapse all Expand all

Background

Approximately half of people with heart failure have chronic kidney disease (CKD). Pharmacological interventions for heart failure in people with CKD have the potential to reduce death (any cause) or hospitalisations for decompensated heart failure. However, these interventions are of uncertain benefit and may increase the risk of harm, such as hypotension and electrolyte abnormalities, in those with CKD.

Objectives

This review aims to look at the benefits and harms of pharmacological interventions for HF (i.e., antihypertensive agents, inotropes, and agents that may improve the heart performance indirectly) in people with HF and CKD.

Search methods

We searched the Cochrane Kidney and Transplant Register of Studies through 12 September 2019 in consultation with an Information Specialist and using search terms relevant to this review. Studies in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE, conference proceedings, the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.

Selection criteria

We included randomised controlled trials of any pharmacological intervention for acute or chronic heart failure, among people of any age with chronic kidney disease of at least three months duration.

Data collection and analysis

Two authors independently screened the records to identify eligible studies and extracted data on the following dichotomous outcomes: death, hospitalisations, worsening heart failure, worsening kidney function, hyperkalaemia, and hypotension. We used random effects meta‐analysis to estimate treatment effects, which we expressed as a risk ratio (RR) with 95% confidence intervals (CI). We assessed the risk of bias using the Cochrane tool. We applied the GRADE methodology to rate the certainty of evidence.

Main results

One hundred and twelve studies met our selection criteria: 15 were studies of adults with CKD; 16 studies were conducted in the general population but provided subgroup data for people with CKD; and 81 studies included individuals with CKD, however, data for this subgroup were not provided. The risk of bias in all 112 studies was frequently high or unclear. Of the 31 studies (23,762 participants) with data on CKD patients, follow‐up ranged from three months to five years, and study size ranged from 16 to 2916 participants. In total, 26 studies (19,612 participants) reported disaggregated and extractable data on at least one outcome of interest for our review and were included in our meta‐analyses.

In acute heart failure, the effects of adenosine A1‐receptor antagonists, dopamine, nesiritide, or serelaxin on death, hospitalisations, worsening heart failure or kidney function, hyperkalaemia, hypotension or quality of life were uncertain due to sparse data or were not reported.

In chronic heart failure, the effects of angiotensin‐converting enzyme inhibitors (ACEi) or angiotensin receptor blockers (ARB) (4 studies, 5003 participants: RR 0.85, 95% CI 0.70 to 1.02; I2 = 78%; low certainty evidence), aldosterone antagonists (2 studies, 34 participants: RR 0.61 95% CI 0.06 to 6.59; very low certainty evidence), and vasopressin receptor antagonists (RR 1.26, 95% CI 0.55 to 2.89; 2 studies, 1840 participants; low certainty evidence) on death (any cause) were uncertain. Treatment with beta‐blockers may reduce the risk of death (any cause) (4 studies, 3136 participants: RR 0.69, 95% CI 0.60 to 0.79; I2 = 0%; moderate certainty evidence).

Treatment with ACEi or ARB (2 studies, 1368 participants: RR 0.90, 95% CI 0.43 to 1.90; I2 = 97%; very low certainty evidence) had uncertain effects on hospitalisation for heart failure, as treatment estimates were consistent with either benefit or harm. Treatment with beta‐blockers may decrease hospitalisation for heart failure (3 studies, 2287 participants: RR 0.67, 95% CI 0.43 to 1.05; I2 = 87%; low certainty evidence).

Aldosterone antagonists may increase the risk of hyperkalaemia compared to placebo or no treatment (3 studies, 826 participants: RR 2.91, 95% CI 2.03 to 4.17; I2 = 0%; low certainty evidence). Renin inhibitors had uncertain risks of hyperkalaemia (2 studies, 142 participants: RR 0.86, 95% CI 0.49 to 1.49; I2 = 0%; very low certainty). We were unable to estimate whether treatment with sinus node inhibitors affects the risk of hyperkalaemia, as there were few studies and meta‐analysis was not possible. Hyperkalaemia was not reported for the CKD subgroup in studies investigating other therapies.

The effects of ACEi or ARB, or aldosterone antagonists on worsening heart failure or kidney function, hypotension, or quality of life were uncertain due to sparse data or were not reported.

Effects of anti‐arrhythmic agents, digoxin, phosphodiesterase inhibitors, renin inhibitors, sinus node inhibitors, vasodilators, and vasopressin receptor antagonists were very uncertain due to the paucity of studies.

Authors' conclusions

The effects of pharmacological interventions for heart failure in people with CKD are uncertain and there is insufficient evidence to inform clinical practice. Study data for treatment outcomes in patients with heart failure and CKD are sparse despite the potential impact of kidney impairment on the benefits and harms of treatment. Future research aimed at analysing existing data in general population HF studies to explore the effect in subgroups of patients with CKD, considering stage of disease, may yield valuable insights for the management of people with HF and CKD.

PICOs

Population
Intervention
Comparison
Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

Heart failure drugs in people with chronic kidney disease

What is the issue?

People with heart failure (when the heart doesn't pump blood properly) often need medicine to help with symptoms like fatigue, swelling, and breathing problems. Studies have looked at whether these medicines are harmful or helpful. However, but they do not clarify whether these treatments may help people with heart failure that also have kidney problems (when the kidneys don't remove waste and fluid from the body properly).

What did we do?

We searched for all research studies that assessed the different treatments for heart failure up to September 2019. We evaluated whether medicines prevent death or hospital admissions, or increase risk of harm, for people with kidney disease. We also measured how certain we could be about the effects of these medicines on the body using a system called "GRADE".

What did we find?

We found 31 studies involving 23,762 people with heart failure and chronic kidney disease. Patients were given either a heart failure medicine compared to standard care or a placebo. The treatment they received was decided by random chance. Although there were many different treatments studied, unfortunately, few of them looked at the same type of medicine. As well, there were many different ways that researchers measured what happened when patients took these medicines. As a result, we could not combine the studies together and clarify the benefits and harms of each treatment. Existing studies cannot really tell us whether medicines used to treat heart failure in the general population are effective or safe for people who have both heart failure and chronic kidney disease.

Conclusions

We are not able to recommend which heart failure medicines are best for people with heart failure and chronic kidney disease. We need more information from large clinical studies. Most of the heart failure studies did not report treatment effects separately based on levels of kidney function. Obtaining this information from existing studies may be helpful to learn more about how to treat heart failure in people with chronic kidney disease.

Authors' conclusions

Implications for practice

  • In accordance with current guidelines, pharmacological interventions for HF should be used in caution in patients with comorbid renal insufficiency. Systematic review of the existing literature didn't change the certainty of evidence around the using interventions for HF in people with CKD.

  • While most analyses did not suggest a reduced risk of death or hospitalisation, the limited certainty of this evidence suggests a lack of evidence, not a lack of benefit, on these interventions. No changes to clinical practice are advised at this time until further information becomes available.

  • Evidence on the appropriateness of the use of HF therapies is limited by: (1) the relatively small number of studies from which CKD data could be extracted to inform each comparison:outcome combination; and (2) the fact that people with advanced CKD are often under‐represented in general HF studies.

  • Available data suggest beta‐blockers may reduce risk of death. However, high‐quality studies enrolling patients with CKD are needed to increase the level of certainty for this effect.

  • There is low certainty that anti‐aldosterone interventions lead to hyperkalaemia. Information on the risk of other harms for most therapies is limited.

  • Studies assessing interventions for acute decompensated HF compared outcomes within the first six months only.

Implications for research

  • A better understanding of the risks of harms associated with HF interventions in people with CKD is needed.

  • While the evidence surrounding the efficacy and safety of HF interventions is strong in the general population, exploring whether the same applies to people with CKD is important because these people may potentially gain more benefits from these interventions but also suffer from more harms. Future studies comparing risk between non‐CKD patients with those with CKD will help guide this knowledge.

  • The stage of CKD should be considered when exploring the impact of kidney disease on the efficacy or safety of HF interventions. Including patients with a very low baseline eGFR (including those receiving kidney replacement therapy) will help inform practice for patients with advanced CKD and HF.

  • Although the majority of HF studies did not exclude participants with CKD, subgroup analyses by CKD were not available in most studies.

  • Future research aimed at analysing existing data in general population HF studies to explore the effect in sub‐groups of patients with CKD may yield valuable insights for the management of people with HF and CKD.

  • One of the limitations of this systematic review is the inability to access the data generated by RCTs of pharmacological interventions for HF that included people with CKD. Removing existing barriers to study data would promote more rapid advancement of new knowledge, improve patient care, and the use of resources in health systems.

Summary of findings

Open in table viewer
Summary of findings for the main comparison. ACUTE: Serelaxin versus placebo in patients with heart failure and chronic kidney disease

ACUTE: Serelaxin versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: acute, inpatient

Intervention: serelaxin

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

Serelaxin

Death (any cause)

42 per 1000

25 per 1000

(14 to 43)

RR 0.59

(0.34 to 1.02)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Cardiovascular death

97 per 1000

46 per 1000

(20 to 105)

RR 0.47 (0.21 to 1.08)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Hospitalisation (any cause)

not reported

not reported

‐‐

‐‐

‐‐

Heart failure hospitalisation

not reported

not reported

‐‐

‐‐

‐‐

Worsening heart failure

162 per 1000

115 per 1000

(87 to 149)

RR 0.71 (0.54 to 0.92)

1395 (2)

⊕⊕⊝⊝
low3,4

Worsening kidney function

86 per 1000

74 per 1000

(32 to 170)

RR 0.86 (0.37 to 1.98)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Hypotension

48 per 1000

60 per 1000

(39 to 94)

RR 1.26 (0.81 to 1.96)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Hyperkalaemia

48 per 1000

52 per 1000

(15 to 187)

RR 1.08 (0.30 to 3.87

234 (1)

⊕⊝⊝⊝
very low2,3,4

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

CI: Confidence interval; RR: Risk Ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Open in table viewer
Summary of findings 2. CHRONIC: ACEi or ARB versus placebo in patients with heart failure and chronic kidney disease

CHRONIC: ACEi or ARB versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: chronic, outpatient

Intervention: ACEi or ARB

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

ACEi or ARB

Death (any cause)

(follow‐up 2 to 3.5 years)

320 per 1000

272 per 1000
(224 to 326)

RR 0.85 (0.70 to 1.02)

5003 (4)

⊕⊕⊝⊝
low2,3,4

Cardiovascular death

(follow‐up 3 to 4 years)

411 per 1000

333 per 1000
(263 to 415)

RR 0.81 (0.64 to 1.01)

1368 (2)

⊕⊕⊝⊝
low2,3,4

Hospitalisation (any cause)

not reported

not reported

‐‐

‐‐

‐‐

Heart failure hospitalisation

(follow‐up 2 to 3.5 years)

546 per 1000

491 per 1000
(235 to 1037)

RR 0.90 (0.43 to 1.90)

1368 (2)

⊕⊝⊝⊝
very low2 ,3,4

Worsening heart failure

not reported

not reported

‐‐

‐‐

‐‐

Worsening kidney function

not reported

not reported

‐‐

‐‐

‐‐

Hypotension (ARB; withdrawn from study)

42 per 1000

109 per 1000

(47 to 254

RR 2.60 (1.12 to 6.07)

332 (1)

⊕⊝⊝⊝
very low3,4

Hyperkalaemia

not reported

not reported

‐‐

‐‐

‐‐

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

*The assumed risk (e.g. the median control group risk across studies) was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. The corresponding risk was calculated by multiplying the assumed risk with the relative effect from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

CI: Confidence interval; RR: Risk Ratio; ACEi: angiotensin‐converting enzyme inhibitor; ARB: angiotensin receptor blocker

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Open in table viewer
Summary of findings 3. CHRONIC: Beta‐blockers versus placebo in patients with heart failure and chronic kidney disease

CHRONIC: Beta‐blockers versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: chronic, outpatient

Intervention: beta‐blockers

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

Beta‐blockers

Death (any cause)

(follow‐up 1 to 2.5 years)

215 per 1000

148 per 1000
(129to 170)

RR 0.69

(0.60 to 0.79)

3136 (4)

⊕⊕⊝⊝
moderate3,4

Cardiovascular death

(follow‐up 1 to 2.5 years)

123 per 1000

65 per 1000
(43 to 100)

RR 0.53 (0.35 to 0.81)

2287 (3)

⊕⊕⊝⊝
low3,4

Hospitalisation (any cause)

432 per 1000

324 per 1000
(225 to 467)

RR 0.75 (0.52 to 1.08)

1583 (2)

⊕⊕⊝⊝
low2,3,4

Heart failure hospitalisation

(follow‐up 1 to 2.5 years)

318 per 1000

213 per 1000
(137 to 334)

RR 0.67 (0.43 to 1.05)

2287 (3)

⊕⊕⊝⊝
low2,3,4

Worsening heart failure

25 per 1000

34 per 1000

RR 1.36

(0.58 to 3.20)

704 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

no events

no events

‐‐

704 (1)

⊕⊝⊝⊝
very low3,4

Hypotension

no events

2/356**

RR 5.11

(0.25 to 106.15)

704 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

not reported

not reported

‐‐

‐‐

‐‐

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

*The assumed risk (e.g. the median control group risk across studies) was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. The corresponding risk was calculated by multiplying the assumed risk with the relative effect from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

** Event rate derived from the raw data. A 'per thousand' rate is non‐informative in view of the scarcity of evidence and zero events in the control group

CI: Confidence interval; RR: Risk Ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Open in table viewer
Summary of findings 4. CHRONIC: Aldosterone antagonists versus placebo in patients with heart failure and chronic kidney disease

CHRONIC: Aldosterone antagonists versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: chronic, outpatient

Intervention: aldosterone antagonists

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

Anti‐aldosterone

Death (any cause)

(follow‐up 6 to 21 months)

294 per 1000

179 per 1000
(17 to 1,938)

RR 0.61 (0.06 to 6.59)

34 (2)

⊕⊝⊝⊝
very low1,2,3,4

Cardiovascular death

not reported

not reported

‐‐

‐‐

‐‐

Hospitalisation (any cause)

not reported

not reported

‐‐

‐‐

‐‐

Heart failure hospitalisation

not reported

not reported

‐‐

‐‐

‐‐

Worsening heart failure

not reported

not reported

‐‐

‐‐

‐‐

Worsening kidney function

not reported

not reported

‐‐

‐‐

‐‐

Hypotension

not reported

not reported

‐‐

‐‐

‐‐

Hyperkalaemia

follow‐up 6‐24 months

84 per 1000

243 per 1000
(170 to 350)

RR 2.91 (2.03 to 4.17)

826 (3)

⊕⊕⊝⊝
Low1,4

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

*The assumed risk (e.g. the median control group risk across studies) was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. The corresponding risk was calculated by multiplying the assumed risk with the relative effect from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

CI: Confidence interval; RR: Risk Ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Background

Description of the condition

Chronic kidney disease (CKD) is defined by structural (presence of albuminuria) or functional (persistently reduced kidney function) alterations of the kidney that persists for three or more months. CKD is classified according to estimated glomerular filtration rate (eGFR) and albumin:creatinine ratio (ACR), using 'G' to denote the GFR category (G1 to G5, with cut‐offs at 90, 60, 30 and 15 mL/min/1.73 m2) and 'A' for the ACR category (A1 to A3, with cut‐offs at 3 and 30 mg/mol) (Levey 2011). CKD is usually clinically silent until end‐stage kidney disease (ESKD) requiring dialysis or kidney transplantation, although complications including fluid retention, hypertension or anaemia are often identified in earlier CKD stages.

Heart failure (HF) results from any structural or functional cardiac disorder that impairs the ability of the heart to function as a pump to support the circulation of blood. People with HF have symptoms of fatigue and breathlessness and signs of fluid retention such as ankle swelling, elevated jugular venous pressure, and pulmonary crackles. Guidelines (McKelvie 2012; Ponikowski 2016; Yancy 2016) restrict HF definition and management recommendations to stages at which symptoms are apparent.

HF can be classified by symptoms or based on measurement of the left ventricular ejection fraction (EF) as below:

Based on symptoms (Yancy 2016)

  • Stage I: no limitation of physical activity; ordinary physical activity does not cause symptoms of HF

  • Stage II: slight limitation of physical activity; comfortable at rest, but ordinary physical activity results in symptoms of HF

  • Stage III: marked limitation of physical activity; comfortable at rest, but less than ordinary activity causes symptoms of HF

  • Stage IV: unable to carry on any physical activity without symptoms of HF, or symptoms of HF at rest.

Based on left ventricular EF (Ponikowski 2016)

  • HF with preserved EF (≥ 50%; HFpEF)

  • HF with reduced EF (< 40%; HFrEF or systolic HF)

  • HF with mid‐range EF (40% to 49%; HFmrEF).

According to current guidelines (McKelvie 2012; Ponikowski 2016; Yancy 2016) most studies of pharmacological interventions for HF included people with HFrEF, partly because HFrEF is felt to be more severe and possibly because HFpEF is more difficult to diagnose. Because most people with HFrEF also have diastolic dysfunction, and subtle systolic abnormalities are present in people with HFpEF, the terminology based on left ventricle EF (HFrEF/HFpEF) is preferred over the terminology of preserved or reduced ‘systolic function’ (Ponikowski 2016).

Based on time course and clinical presentation (Ponikowski 2016)

  • Chronic HF identifies HF for some time; unchanged symptoms and signs for at least one month define ‘stable HF’

  • Acute HF could present as decompensation of chronic stable HF, occurrence of de novo HF (e.g. due to a myocardial infarction (MI)), or progression of HF in a subacute (gradual) fashion (e.g. due to progressive dilated cardiomyopathy)

  • Congestive HF is a term sometimes used to describe acute or chronic HF with significant evidence of volume overload.

CKD and HF are highly prevalent in the general population. In population‐based studies, the prevalence of HF ranges from 0.2% in those aged 18 to 44 years, to 24% in patients 85 years and older (Hemmelgarn 2012). The prevalence of CKD ranges from 1% to 33% in individuals 18 to 44 years and over age 60, respectively (Hemmelgarn 2012). The two conditions often co‐exist and 40% to 60% of patients with HF will have evidence of CKD (Shiba 2011). Furthermore, people with both HF and CKD have a greater risk of cardiovascular morbidity and death than people with HF alone (Santoro 2014). As compared to people with either condition alone, those with both HF and CKD have worse quality of life (QoL), a higher risk of morbidity and death, and often require complex multidisciplinary management (Shiba 2011). Criteria to classify cardio‐renal syndromes, according to the direction of the organ damage or whether the clinical manifestations are acute or chronic, have been defined by the Acute Dialysis Quality Initiative consensus group (Ronco 2010).

The prevalence of HF and concomitant CKD is increasing. Additionally, CKD may increase the risk of harms associated with HF medication (e.g. spironolactone (Quach 2016)). Therefore, a better understanding of appropriate pharmacological management of HF in this patient population is particularly important.

Description of the intervention

Several pharmacological interventions are used to treat HF. These include:

  1. Direct agents

    • Those that reduce peripheral resistance, heart rate or blood volume (anti‐hypertensive agents)

    • Those that increase the strength of the heart contraction (positive inotropes)

  2. Indirect agents

    • Those that may improve the heart performance indirectly (lipid lowering agents, antiplatelet agents and nutritional supplements)

Guidelines for the general population (Ponikowski 2016; Yancy 2016) make a distinction between:

  1. Interventions that are recommended for HFrEF: angiotensin‐converting enzyme inhibitors (ACEi), beta‐blockers and mineralocorticoid receptor antagonists

  2. Interventions that may be beneficial in HFrEF: diuretics, angiotensin receptor neprilysin inhibitor (a single molecule that combines the moieties of valsartan and sacubitril), sinus node If channel inhibitors (ivabradine), angiotensin receptor blockers (ARB), hydralazine and isosorbide dinitrate

  3. Interventions with uncertain benefits in HFrEF: digoxin and other digitalis glycosides, and n‐3 polyunsaturated fatty acids

  4. Interventions with unproven benefits in HFrEF: statins, oral anticoagulants and antiplatelet therapy, renin inhibitors

  5. Interventions potentially harmful in HFrEF: calcium‐channel blockers.

The same guidelines specify that no treatment has yet been shown, convincingly, to reduce morbidity or death in patients with HFpEF (Ponikowski 2016).

Interventions for acute HF are supported by a lower level of evidence, according to these guidelines (McKelvie 2012; Ponikowski 2016; Yancy 2016). These interventions include diuretics for congestive HF (loop diuretics, thiazides and mineral‐corticoid antagonists), vasodilators for hypertensive HF (nitroglycerin, isosorbide dinitrate, nitroprusside, nesiritide), inotropic agents for hypotensive HF (dopamine, dobutamine, levosimendan, milrinone, enoximone), vasopressors for hypotensive HF (norepinephrine, epinephrine), and digoxin, beta‐blockers or amiodarone for atrial fibrillation with high ventricular rate.

How the intervention might work

Pharmacological interventions improve the heart performance by reducing how quickly the heart beats, reducing its workload, or by increasing the strength of the cardiac muscle directly or indirectly. In the general population, there is strong evidence to support the use of renin‐angiotensin‐aldosterone system (RAAS) blockers, beta‐blockers and their combination, and moderate levels of evidence for diuretics and inotropes. In the CKD population, these agents are thought to work in the same way. However, most studies of interventions for HF excluded people with CKD because it was felt that these interventions may not have the same benefits in a CKD population, and they may be potentially harmful (Coca 2006; Konstantinidis 2016). For example, RAAS blockers may increase the risk of hyperkalaemia, excessive volume depletion with diuretics may worsen kidney function, and some inotropes may accumulate in CKD and adversely affect cardiac rhythm.

Why it is important to do this review

People with CKD and HF require coordinated care from CKD and HF specialists. Individuals with CKD are at high risk of cardiovascular morbidity and death. A narrative review including 1,076,104 patients found a significantly higher risk of death among HF patients with comorbid CKD, compared to those without baseline CKD (Damman 2014a). They may have the potential for significant benefit more from interventions for HF, but they may also be more vulnerable to the potential harms of these interventions. We do not know if benefits and harms of interventions for HF vary with the severity of CKD. For example, people with more advanced stages of CKD benefit less from interventions for ischaemic heart disease (e.g. statins (Palmer 2012)), and complications from HF dominate as the leading cause of hospitalisation. A systematic review of studies addressing questions about pharmacological interventions for HF in people with CKD is not available. Given the increasing prevalence of HF as eGFR declines, the complex and costly management of people affected by both conditions, and the uncertain benefits and harms of available therapies to treat HF in people who also have CKD, this information is important for patients, healthcare providers and policy makers.

Objectives

This review aims to look at the benefits and harms of pharmacological interventions for HF (i.e., antihypertensive agents, inotropes, and agents that may improve the heart performance indirectly) in people with HF and CKD.

Methods

Criteria for considering studies for this review

Types of studies

We included all randomised controlled trials (RCTs) and quasi‐RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) evaluating the benefits and harms attributable to any pharmacological interventions for HF in people with CKD.

Types of participants

We included studies in people with HF as defined by the authors. For example, people with HF or people with symptoms of HF in case the term HF or the NYHA classification were not explicitly stated. These included breathlessness, ankle swelling and fatigue that may be accompanied by signs (e.g. elevated jugular venous pressure, pulmonary crackles and peripheral oedema) caused by a structural and/or functional cardiac abnormality, resulting in a reduced cardiac output and/or elevated intracardiac pressures at rest or during stress (Ponikowski 2016). Eligible studies included those in any population age, sex, race, educational status or disease type/severity, and could be conducted in any setting (e.g. community, hospital, nursing home, chronic care institution, or outpatient setting). We included individuals with CKD of any stage (from persistent albuminuria without diminished eGFR to ESKD treated with dialysis or kidney transplantation).

Inclusion criteria

CKD is defined based on the presence of urinary albumin excretion of ≥ 30 mg/mol, or equivalent (marker of kidney damage), or as eGFR < 60 mL/min/1.73 m2 (marker of decreased kidney function) for three months or longer regardless of the primary cause of kidney injury. We considered studies that included people with any CKD stage (including stage V treated with dialysis or transplantation); however, we considered definitions used by study authors. HF is a common clinical syndrome resulting from any structural or functional cardiac disorder that impairs the ability of the ventricle to fill with (diastolic dysfunction) or eject blood (systolic dysfunction). Therefore, we considered studies of HF due to left ventricular dysfunction (left‐sided HF), with reduced ejection fraction (EF≤ 40%; systolic left‐sided HF (HFrEF)) or preserved ejection fraction (EF > 40%; diastolic left‐sided HF (HFpEF or HFmrEF)); or due to right ventricular dysfunction (right‐sided HF). We considered studies including people with stable HF or acute HF as described above, similarly, as defined by study authors. We only included studies of at least three months in duration, without publication year or language restriction.

Exclusion criteria

We excluded studies shorter than three months.

Types of interventions

Based on existing recommendations (McKelvie 2012; Ponikowski 2016; Yancy 2016), we considered the following pharmacological intervention for either acute or chronic HF.

Interventions for acute HF

  • Diuretics for congestive HF: loop diuretics, thiazides and mineral‐corticoid antagonists

  • Vasodilators for hypertensive HF: nitroglycerin, isosorbide dinitrate, nitroprusside, nesiritide

  • Inotropic agents for hypotensive HF: dopamine, dobutamine, levosimendan, milrinone, enoximone

  • Vasopressors for hypotensive HF: norepinephrine, epinephrine

  • Digoxin, beta‐blockers or amiodarone for atrial fibrillation with high ventricular rate.

Interventions for chronic HF

  • The combination of ACEi and beta‐blockers with and without mineral‐corticoid receptor antagonists

  • Diuretics, angiotensin receptor neprilysin inhibitor, sinus node If channel inhibitors, ARB, hydralazine and isosorbide dinitrate

  • Digoxin and other digitalis glycosides, n‐3 polyunsaturated fatty acids, coenzyme Q10, creatine analogues, L‐carnitine, thiamine, vitamin D supplementation

  • Statins, fibrates, ezetimibe, oral anticoagulants and antiplatelet therapy (cyclooxygenase inhibitors, ADP receptor blockers, PDE inhibitors, PAR‐1 antagonists, thromboxane inhibitors, adenosine re‐uptake inhibitors, others), renin inhibitors

  • Calcium‐channel blockers.

For all interventions, we considered any preparation, route of administration, dose, duration, frequency, and combinations.

Given the large number of research questions that this review could potentially address, we focused on the following interventions that are recommended in the non‐CKD population.

  • Use of ACEi or ARB, and beta‐blockers

  • Mineralocorticoid receptor antagonists

  • Use of diuretics in congestive HF.

The primary focus of this review was to assess whether there is evidence that these treatments for HF are effective and safe in people with concomitant CKD when compared to an active control, placebo or no intervention.

Types of outcome measures

Primary outcomes

  • Death (any cause)

  • Hospitalisation for decompensated HF

Secondary outcomes

  • Cardiovascular death

  • Hospitalisation (any cause)

  • Worsening HF (reported as a dichotomous outcome by authors)

  • Worsening kidney function, including acute kidney injury (reported as a dichotomous outcome by authors)

  • Hyperkalaemia

  • Hypotension

  • Other adverse events due to treatment, worsening of the disease course (syncope/pre‐syncope, shortness of breath, and peripheral oedema), and measures of quality of life (QoL) as reported by authors

Search methods for identification of studies

Electronic searches

We searched the Cochrane Kidney and Transplant Register of Studies up to 12 September 2019 through contact with the Information Specialist using search terms relevant to this review. The Register contains studies identified from the following sources.

  1. Monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL)

  2. Weekly searches of MEDLINE OVID SP

  3. Hand‐searching of kidney‐related journals and the proceedings of major kidney and transplant conferences

  4. Searching of the current year of EMBASE OVID SP

  5. Weekly current awareness alerts for selected kidney and transplant journals

  6. Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.

Studies contained in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE based on the scope of Cochrane Kidney and Transplant. Details of search strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available on the Cochrane Kidney and Transplant website.

See Appendix 1 for search terms used in strategies for this review.

Searching other resources

We searched the reference lists of review articles, relevant studies and clinical practice guidelines.

Data collection and analysis

Selection of studies

We used the search strategy to identify studies relevant to this review. Two authors screened the titles and abstracts independently and selected studies based on whether they met the patient, intervention, comparator, and design criteria. When necessary, two authors (ML, PR) independently assessed full texts to determine which studies satisfied the full inclusion criteria.

Data extraction and management

Four authors extracted data independently using standard forms. Studies reported in non‐English language journals were translated before assessment. Where more than one publication of a study existed, reports were grouped together and the publication with the most complete data was used in the analyses.

Assessment of risk of bias in included studies

Four authors independently assessed the following items using the risk of bias assessment tool (Higgins 2011) (see Appendix 2).

  • Was there adequate sequence generation (selection bias)?

  • Was allocation adequately concealed (selection bias)

  • Was knowledge of the allocated interventions adequately prevented during the study (blinding)?

    • Participants and personnel (performance bias)

    • Outcome assessors (detection bias)

  • Were incomplete outcome data adequately addressed (attrition bias)?

  • Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?

  • Was the study apparently free of other problems that could put it at a risk of bias (registration, funding, etc.)?

Measures of treatment effect

We used risk ratios (RR) with 95% confidence intervals (CI) to summarize dichotomous outcomes (e.g. death, hospitalisation, kidney failure). Where continuous scales of measurement were reported (e.g. QoL, symptom burden), we assessed the effects of treatment using the mean difference (MD) or the standardised mean difference (SMD) for different scales. We referred to the Cochrane Handbook for Systematic Reviews of Interventions to deal with specific issues (e.g. change scores or time‐to‐event data) (Higgins 2011). Summary intervention effect measures were estimated in comparisons with at least two studies.

Unit of analysis issues

We referred to the Cochrane Handbook for Systematic Reviews of Interventions to deal with non‐standard designs (e.g. as cross‐over studies or cluster‐randomised studies) (Higgins 2011).

Dealing with missing data

We requested any further information required from the original author by written correspondence (e.g. emailing corresponding authors) and included any relevant information obtained in this manner in the review. We also investigated attrition rates, for example drop‐outs, losses to follow‐up and withdrawals and critically appraised issues of missing data and imputation methods (for example, last‐observation‐carried‐forward) (Higgins 2011).

Assessment of heterogeneity

In analyses with at least two studies, we assessed heterogeneity by visual inspection of the forest plot. We then quantified statistical heterogeneity using the I2 statistic, which described the percentage of total variation across studies that is due to heterogeneity rather than sampling error (Higgins 2003). We followed this guide to interpret I2 values.

  • 0% to 40%: might not be important

  • 30% to 60%: may represent moderate heterogeneity

  • 50% to 90%: may represent substantial heterogeneity

  • 75% to 100%: considerable heterogeneity.

The importance of the observed value of I2 depended on both its magnitude and the strength of evidence for heterogeneity (e.g. P‐value from the chi‐squared test or a confidence interval for I2) (Higgins 2011).

Assessment of reporting biases

In analyses with at least 10 studies, we planned to use funnel plots to assess for the potential existence of small study bias (Higgins 2011).

Data synthesis

We summarized data using random‐effects models.

Subgroup analysis and investigation of heterogeneity

In analyses with at least three studies, we investigated potential sources of heterogeneity. We considered participant factors, including demographics (age, sex, race), comorbidity (diabetes, coronary heart disease), type of HF, and stage of CKD. Where possible, we considered agent subclasses (ACEi versus ARB) and dose and duration of therapy, as well as the following pre‐specified indicators for subgroup analyses (defining different types of cardiac disease or stages of CKD).

  1. The type of HF (right‐ versus left‐sided HF)

  2. The nature of the functional impairment of the left ventricle (HFrEF versus HFmrEF or HFpEF)

  3. Congestive versus non‐congestive acute HF

  4. Cardio‐renal (acute versus chronic) versus reno‐cardiac (acute versus chronic) HF

  5. The severity of CKD (defined by levels of urine albumin:creatinine ratio and eGFR; CKD stages 1‐5).

Sensitivity analysis

Where possible, we planned to perform sensitivity analyses to explore the influence of the following factors on effect size.

  • Repeating the analysis excluding unpublished studies

  • Repeating the analysis taking into account of the risk of bias, as specified

  • Repeating the analysis excluding studies of short duration or small studies (using the median as a cut‐off)

  • Repeating the analysis excluding studies using the following filters: diagnostic criteria, language of publication, source of funding (industry versus other), and country.

'Summary of findings' tables

We presented the main results of the review in 'Summary of findings' tables. These tables present key information concerning the quality of the evidence, the magnitude of the effects of the interventions examined, and the sum of the available data for the main outcomes (Schunemann 2011a). The 'Summary of findings' tables also include an overall grading of the evidence related to each of the main outcomes using the GRADE (Grades of Recommendation, Assessment, Development and Evaluation) approach (GRADE 2008; GRADE 2011). The GRADE approach defines the quality of a body of evidence as the extent to which one can be confident that an estimate of effect is close to the true quantity of specific interest. The quality of a body of evidence involves consideration of within‐trial risk of bias (methodological quality), directness of evidence, heterogeneity, precision of effect estimates and risk of publication bias (Schunemann 2011b). Assumed risk was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. Corresponding risk was calculated by multiplying the assumed risk with the risk ratio estimate from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

We included the following outcomes in the 'Summary of findings' tables.

  • Death (any cause)

  • Cardiovascular death

  • Hospitalisation (any cause)

  • Hospitalisation for decompensated HF

  • Worsening HF

  • Worsening kidney function

  • Hyperkalaemia

  • Hypotension

  • Quality of Life

Results

Description of studies

Results of the search

Our search of the Cochrane Kidney and Transplant specialised register identified 869 records. We identified an additional 78 records using other sources (reference lists of review articles, relevant studies, and clinical practice guidelines) ‐ therefore a total of 947 records (176 studies) were identified. We excluded 61 studies (252 records), either due to a population other than HF (38 studies), a non‐pharmacological intervention (5), follow‐up shorter than three months (16), or a study design other than a RCT (2) (see Characteristics of excluded studies).

Overall, 115 studies were eligible. Of these, three are ongoing and awaiting publication of primary data (PARAGON‐HF 2018; RELAX‐AHF‐2 2017; TMAC 2007) and will be included in a future update of this review. As a result, 112 studies were included in this review (Figure 1) and are categorised as follows.

Figure 1
Study flow diagram.

Study flow diagram.

CKD population studies:

General population studies, including people with CKD:

General population studies, potentially including people with CKD:

Detailed information on the design, participants, intervention, comparator and outcomes of these included studies are summarised in Characteristics of included studies.

Included studies

In total, 31 studies reported data for CKD patients and the characteristics of these 31 studies are reported here. Full details for all studies are summarised in the Characteristics of included studies table.

Study size, country and year

A total of 23,762 adult CKD participants were randomised; six studies (3892) were in the setting of acute HF and 25 (19,870) were in the setting of chronic HF. The number of CKD participants per group was not provided in two studies (A‐HeFT 2002; DIG 1996). Sample sizes ranged from 16 to 3157 participants and the reported follow‐up ranged from 12 weeks to five years.

Ten of 31 studies (32%) were conducted in single‐country settings. Three (10%) in Canada or the USA only, and 16 (52%) were in at least three countries. Two studies (6%) did not report this information. Study initiation ranged from 1986 to 2015.

Participants’ mean age was reported in 24 (77%) of the 31 studies. The proportion of men ranged between 64.9% and 82.1% and the proportion of people with diabetes ranged between 14.5% and 62.5%. Demographic data for CKD participants (age, sex, diabetes status) were not reported in seven studies (A‐HeFT 2002; ALOFT 2008; DIAMOND 1999; EMPHASIS‐HF 2010; EVEREST 2005; HEAAL 2008; PROMISE 1989).

Studies in acute HF (6) included patients who had been admitted for acute decompensated HF (Palazzuoli 2014; Pre‐RELAX‐AHF 2009; PROTECT 2008; RELAX‐AHF 2012; RENO‐DEFEND 1 2011; ROSE AHF 2013). Studies in chronic HF (25) included patients with HFrEF (A‐HeFT 2002; ARIANA‐CHF‐RD 2016; CIBIS II 1997; Cice 2001; Cice 2010; DIAMOND 1999; DIG 1996; EMPHASIS‐HF 2010; EVEREST 2005; FUSION II 2008; HEAAL 2008; MERIT‐HF 1997; PROMISE 1989; RALES 1995; SAVE 1991; SENIORS 2002; SHIFT 2010; SOLVD (Treatment) 1992; Taheri CAPD 2012; Taheri HD 2009; Val‐HeFT 1999), HFpEF (Cice 1999a), or with no prespecified ejection fraction criteria for inclusion (ALOFT 2008; AQUAMARINE 2014; Pita‐Fernandez 2015).

Interventions

Studies in acute HF evaluated adenosine A1‐receptor antagonists (PROTECT 2008; RENO‐DEFEND 1 2011), diuretics (Palazzuoli 2014), dopamine and nesiritide (ROSE AHF 2013), and serelaxin (Pre‐RELAX‐AHF 2009; RELAX‐AHF 2012).

In chronic settings, studies evaluated ARB (Cice 2010; HEAAL 2008; Pita‐Fernandez 2015; Val‐HeFT 1999), beta‐blockers (CIBIS II 1997; Cice 2001; MERIT‐HF 1997; SENIORS 2002), ACEi (Cice 2010; SAVE 1991; SOLVD (Treatment) 1992), aldosterone antagonists (EMPHASIS‐HF 2010; RALES 1995; Taheri CAPD 2012; Taheri HD 2009), nesiritide (FUSION II 2008), calcium‐channel blockers (Cice 1999a), renin inhibitors (ALOFT 2008; ARIANA‐CHF‐RD 2016), vasopressin receptor antagonists (AQUAMARINE 2014; EVEREST 2005), anti‐arrhythmics (DIAMOND 1999), digitalis glycoside (DIG 1996), phosphodiesterase inhibitors (PROMISE 1989), sinus node inhibitors (SHIFT 2010), and vasodilators (A‐HeFT 2002).

Of the 31 studies, 28 studies (90%) including a total of 22,893 CKD patients, the intervention was compared with placebo or standard care. Of the remaining studies, two compared different doses of the same drug (HEAAL 2008; Pita‐Fernandez 2015) and one compared different IV infusion strategies (continuous versus bolus) of furosemide (Palazzuoli 2014).

Outcomes

Study outcomes included death, hospitalisation, worsening HF, progression of kidney disease, quality of life, and adverse events such as hyperkalaemia and hypotension. Of the 31 studies, nine reported data in a form that could not be extracted for meta‐analysis. We contacted the authors of these nine studies and two (22%) were able to provide data (AQUAMARINE 2014; SHIFT 2010). One study investigating calcium‐channel blockers (Cice 1999a) did not report any outcomes of interest and was also not included the in meta‐analysis. At this time, we do not have extractable data for any outcomes for four studies (A‐HeFT 2002; DIG 1996; EMPHASIS‐HF 2010; HEAAL 2008) and did not include these studies in our meta‐analyses. Of the 81 studies in the general population, 39 had valid author contact information and the authors were emailed to request data on the subset of patients with CKD. No authors have responded to this request, to date.

Combining all intervention types:

  • 29 (94%) studies reported on death (any cause), of which 21 provided extractable data and were meta‐analysed

  • 25 (81%) on hospitalisations for HF, of which eight provided extractable data and were meta‐analysed

  • 18 (58%) on cardiovascular death, of which ten provided extractable data and were meta‐analysed

  • 10 (32%) on all‐cause hospitalisation, of which three provided extractable data and were meta‐analysed

  • 11 (35%) reported a dichotomous measure of worsening HF, of which seven were extractable and meta‐analysed

  • 17 (55%) reported a dichotomous measure of progression of kidney disease, of which eight were extractable and meta‐analysed

  • 10 (31%) reported a dichotomous measure of hyperkalaemia, of which seven were extractable and meta‐analysed

  • 15 (48%) reported a dichotomous measure of hypotension, of which eight were extractable and meta‐analysed

  • 3 (10%) reported a quality of life score, of which one was extractable and analysed

Other secondary outcomes (e.g. dyspnoea, oedema) in CKD populations were not extractable and excluded from analysis.

In summary, 26 studies (19,612 participants) reported disaggregated and extractable data on at least one outcome of interest for our review and were included in our meta‐analyses (ALOFT 2008; AQUAMARINE 2014; ARIANA‐CHF‐RD 2016; CIBIS II 1997; Cice 2001; Cice 2010; DIAMOND 1999; EVEREST 2005; FUSION II 2008; MERIT‐HF 1997; Palazzuoli 2014; Pita‐Fernandez 2015; Pre‐RELAX‐AHF 2009; PROMISE 1989; PROTECT 2008; RALES 1995; RELAX‐AHF 2012; RENO‐DEFEND 1 2011; ROSE AHF 2013; SAVE 1991; SENIORS 2002; SHIFT 2010; SOLVD (Treatment) 1992; Taheri HD 2009; Taheri CAPD 2012; Val‐HeFT 1999). Studies were either exclusively in patients with HF and CKD (14) or in the general HF population, but included patients with CKD and stratified results based on kidney function at baseline (12).

Excluded studies

Reasons for exclusion are summarised in the Characteristics of excluded studies table. In 38 studies, the populations studied had conditions other than HF, five evaluated a non‐pharmacological intervention, and 16 measured short‐term outcomes (less than three months follow‐up)

Risk of bias in included studies

Risk of bias for all 112 included studies is summarised in Figure 2 and Figure 3. In general, the risk of bias was unclear in most studies for many domains. Fifty‐four studies (48%) were published after 2005, the year when trial registration became mandatory. Of these, 37 had evidence of registration within a trial registry and 33 reported information on authorship and/or involvement of the study sponsor in data collection, analysis, and interpretation.

Figure 2
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Figure 3
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

Random sequence generation

Risk of bias was deemed low for 46 studies, high for one study, and unclear for 65 studies.

Allocation concealment

Risk of bias was deemed low for 28 studies and unclear for the remaining 84 studies.

Blinding

Participants and personnel (performance bias)

Eighty‐six studies were judged to be at low risk of bias, 13 did not blind participants or study personnel, and 13 were judged unclear.

Blinding of outcome assessment (detection bias)

Fifty‐four studies were judged to be at low risk of bias, either because outcome assessors were blinded or because the outcome measured was unlikely to be influenced by a lack of blinding. Seven studies were judged to be at high risk of bias, and the remaining 51 were judged to be at unclear risk of bias.

Incomplete outcome data

Seventy‐four studies were at low risk of attrition bias as < 10% of patients were lost to follow‐up and analysis followed an intention‐to‐treat design. Thirty‐four studies were at high risk of attrition bias and 14 studies were judged to be at unclear risk of bias.

Selective reporting

Ninety‐nine studies were at low risk of bias, either because a protocol was published or because it was clear that the published report included all expected outcomes. Three studies were at high risk of bias and 10 were judged to be at unclear risk of bias.

Other potential sources of bias

Twenty‐nine studies appeared to be free of other sources of bias. Forty‐three studies were at high risk of bias, as the study sponsor was involved in the design, data acquisition and/or analysis, or preparation of the manuscript. The remaining 40 studies were deemed to be at unclear risk of bias.

Effects of interventions

See: Summary of findings for the main comparison ACUTE: Serelaxin versus placebo in patients with heart failure and chronic kidney disease; Summary of findings 2 CHRONIC: ACEi or ARB versus placebo in patients with heart failure and chronic kidney disease; Summary of findings 3 CHRONIC: Beta‐blockers versus placebo in patients with heart failure and chronic kidney disease; Summary of findings 4 CHRONIC: Aldosterone antagonists versus placebo in patients with heart failure and chronic kidney disease

Effects in an acute heart failure setting

(See summary of findings Table for the main comparison and Table 1).

Open in table viewer
Table 1. Additional Summary of Findings

Comparison

Outcome

Relative effect (95% CI)

No. participants (studies)

Quality of the evidence
(GRADE)

Acute heart failure

Adenosine A1‐receptor antagonist versus placebo

Death (any cause)

RR 0.71 (0.39 to 1.26)

2078 (2)

⊕⊕⊝⊝

low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

RR 0.85 (0.43 to 1.68)

45 (1)

⊕⊝⊝⊝
very low2,3,4

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 0.93 (0.70 to 1.24)

2033 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

1.14 (0.89 to 1.48)

2033 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Diuretic bolus versus continuous infusion

Death (any cause)

Not reported

‐‐

Cardiovascular death

RR 0.56 (0.15 to 2.01)

57 (1)

⊕⊝⊝⊝
very low2,3,4

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

RR 0.83 (0.33 to 2.10)

57 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Dopamine versus placebo

Death (any cause)

RR 0.89 (0.41 to 1.95)

241 (1)

⊕⊝⊝⊝
very low1,2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 2.15 (0.77 to 5.99)

241 (1)

⊕⊝⊝⊝
very low1,2,3,4

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

RR 0.09 (0.01 to 0.65)

226 (1)

⊕⊝⊝⊝
very low1,2,3,4

Quality of life

Not reported

‐‐

Nesiritide versus placebo

Death (any cause)

RR 0.67 (0.28 to 1.57)

238 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 0.83 (0.26 to 2.66)

238 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

RR 1.80 (0.94 to 3.47)

232 (1)

⊕⊝⊝⊝
very low2,3,4

Quality of life

Not reported

‐‐

Chronic heart failure

Anti‐arrhythmics versus placebo

Death (any cause)

RR 0.95 (0.83 to 1.09)

755 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Nesiritide versus placebo

Death (any cause)

RR 0.98 (0.64 to 1.50)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

RR 0.88 (0.56 to 1.38)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

RR 1.80 (1.38 to 2.35)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Quality of Life

MD 1.20 (‐1.85 to 4.25)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Phosphodiesterase inhibitors versus placebo

Death (any cause)

RR 1.26 (0.98 to 1.61)

539 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Renin inhibitors versus placebo

Death (any cause)

Not reported

‐‐

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 0.26 (0.08 to 0.88)

41 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

RR 1.52 (0.22 to 10.33)

142 (2)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

RR 0.86 (0.49 to 1.49)

142 (2)

⊕⊝⊝⊝
very low2,3,4

Hypotension

RR 1.44 (0.62 to 3.31)

142 (2)

⊕⊝⊝⊝
very low2,3,4

Quality of life

Not reported

‐‐

Sinus node inhibitors versus placebo

Death (any cause)

Not reported

Cardiovascular death

RR 0.96 (0.48 to 1.93)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

RR 0.83 (0.48 to 1.45)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening heart failure

Not reported

‐‐

Worsening kidney function

RR 0.95 (0.71 to 1.27)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

RR 0.53 (0.28 to 1.01)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Vasopressin receptor antagonists versus placebo

Death (any cause)

RR 1.26 (0.55 to 2.89)

1840 (2)

⊕⊝⊝⊝
low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

RR 0.82 (0.37 to 1.81)

208 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening heart failure

Not reported

‐‐

Worsening kidney function

RR 0.87 (0.56 to 1.38)

217 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

CI: Confidence interval; RR: Risk Ratio; MD: Mean difference

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Diuretic bolus versus continuous infusion

Palazzuoli 2014 compared a bolus diuretic versus infusion. Evidence was very uncertain on whether administering furosemide as a bolus compared to a continuous infusion has any impact on cardiovascular death (Analysis 1.1) or hospitalisation for HF (Analysis 1.2).

Serelaxin versus placebo

Two studies (Pre‐RELAX‐AHF 2009; RELAX‐AHF 2012) compared serelaxin to placebo. Treatment with serelaxin had uncertain effects on the risk of death (any cause) (Analysis 2.1 (2 studies, 1395 participants): RR 0.59, 95% CI 0.34 to 1.02; I2 = 0%; low certainty evidence). Similarly, treatment with serelaxin had uncertain effects on the risk of cardiovascular death (Analysis 2.2 (2 studies, 1395 participants): RR 0.47, 95% CI 0.21 to 1.08; I2 = 55%; low certainty evidence). Treatment with serelaxin may reduce the risk of worsening HF in acute chronic HF (Analysis 2.3 (2 studies, 1395 participants): RR 0.71, 95% CI 0.54 to 0.92; I2 = 0%; low certainty evidence). Treatment with serelaxin had uncertain effects on worsening kidney function (Analysis 2.4 (2 studies, 1395 participants): RR 0.86, 95% CI 0.37 to 1.98; I2 = 58%; low certainty evidence). It is uncertain whether treatment with serelaxin increases the risk of hyperkalaemia (Analysis 2.5 (1 study, 234 participants): RR 1.08, 95% CI 0.30 to 3.87; very low certainty evidence) or hypotension in acute settings (Analysis 2.6 (2 studies, 1395 participants): RR 1.26, 95% CI 0.81 to 1.96; I2 = 0%; low certainty evidence).

Adenosine A1‐receptor antagonists versus placebo

Two studies (PROTECT 2008; RENO‐DEFEND 1 2011) compared adenosine A1 receptor antagonists to placebo. Treatment with adenosine A1‐receptor antagonists had uncertain effects on the risk of death (any cause) (Analysis 3.1 (2 studies, 2078 participants): RR 0.71, 95% CI 0.39 to 1.26; I2 = 0%; low certainty evidence).

Evidence was very uncertain on whether adenosine A1 receptor antagonists have any impact on hospitalisation (any cause) (Analysis 3.2), worsening HF (Analysis 3.3), or worsening kidney function (Analysis 3.4) as only single studies reported these three outcomes.

Dopamine versus placebo

ROSE AHF 2013 compared dopamine to placebo. Evidence was very uncertain on whether dopamine has any impact on death (any cause) (Analysis 4.1), worsening HF (Analysis 4.2), or hypotension (Analysis 4.3).

Nesiritide versus placebo

ROSE AHF 2013 compared nesiritide to placebo. Evidence was very uncertain on whether nesiritide has any impact on death (any cause) (Analysis 5.1), worsening HF (Analysis 5.2), or hypotension (Analysis 5.3).

Effects in a chronic heart failure setting

(See summary of findings Table 2;summary of findings Table 3;summary of findings Table 4; and Table 1).

Angiotensin‐converting enzyme inhibitors or angiotensin receptor blockers versus placebo

Four studies (Cice 2010; SAVE 1991; SOLVD (Treatment) 1992; Val‐HeFT 1999) compared ACEi or ARB with placebo. Treatment with ACEi or ARB had uncertain effects on death (any cause) (Analysis 6.1 (4 studies, 4003 participants): RR 0.85, 95% CI 0.70 to 1.02; I2 = 78%; low certainty evidence). In subgroup analysis, there was no evidence that ACEi or ARB had different effects (P for subgroup interaction = 0.91). Treatment with either ACEi or ARB may decrease the risk of cardiovascular death compared to placebo or no treatment (Analysis 6.2 (2 studies, 1368 participants): RR 0.81, 95% CI 0.64 to 1.01; I2 = 51%; low certainty evidence). Treatment with either ACEi or ARB had uncertain effects on hospitalisation for HF (Analysis 6.3 (2 studies, 1368 participants): RR 0.90, 95% CI 0.43 to 1.90; I2 = 97%; very low certainty evidence).

Angiotensin receptor blockers (ARB): 50% reduction versus full dose

It was not possible to ascertain the effect of different doses on death (any cause) as only Pita‐Fernandez 2015 was designed to examine drug dosing (Analysis 7.1).

Beta‐blockers versus placebo

Four studies (CIBIS II 1997; Cice 2001; MERIT‐HF 1997; SENIORS 2002) compared beta‐blockers with placebo. Treatment with beta‐blockers may reduce the risk of death (any cause) (Analysis 8.1 (4 studies, 3136 participants): RR 0.69, 95% CI 0.60 to 0.79; I2 = 0%; moderate certainty evidence) compared with placebo. Treatment with beta‐blockers may decrease cardiovascular death (Analysis 8.2 (3 studies, 2287 participants): RR 0.53, 95% CI 0.35 to 0.81; I2 = 61%; low certainty evidence) or hospitalisation (any cause) (Analysis 8.3 (2 studies, 1583 participants): RR 0.75, 95% CI 0.52 to 1.08; I2 = 67%; low certainty evidence). Treatment with beta‐blockers may decrease hospitalisation for HF (Analysis 8.4 (3 studies, 2287 participants): RR 0.67, 95% CI 0.43 to 1.05; I2= 87%; low certainty evidence). Potential causes of high heterogeneity include differences in study year, sample size, location, or use of contaminant ACEi.

Evidence was very uncertain on whether beta‐blockers have any impact on worsening HF (Analysis 8.5), worsening kidney function (Analysis 8.6), or hypotension (Analysis 8.7) as only SENIORS 2002 reported these three outcomes.

Aldosterone antagonists versus placebo

Three studies (RALES 1995; Taheri CAPD 2012; Taheri HD 2009) compared aldosterone antagonists to placebo. Aldosterone antagonists had uncertain effects on death (any cause) compared with placebo (Analysis 9.1 ( 2 studies, 34 participants): RR 0.61, 95% CI 0.06 to 6.59; I2 = 57%; very low certainty evidence). Aldosterone antagonists may increase the risk of hyperkalaemia in people with HF and CKD (Analysis 9.2 (3 studies, 826 participants): RR 2.91, 95% CI 2.03 to 4.17; I2= 0%, low certainty evidence).

Anti‐arrhythmics versus placebo

DIAMOND 1999 compared anti‐arrhythmics to placebo. Evidence was very uncertain on whether anti‐arrhythmics have any impact on death (any cause) (Analysis 10.1).

Nesiritide versus placebo

One study FUSION II 2008 compared nesiritide to placebo in a chronic setting. Evidence was very uncertain on whether nesiritide has any impact on death (any cause) (Analysis 11.1), cardiovascular death (Analysis 11.2), or quality of life (Analysis 11.4). There were 196/605 patients with at least one episode of hypotension in the nesiritide group and 55/306 in the placebo group (Analysis 11.3).

Phosphodiesterase inhibitors versus placebo

PROMISE 1989 compared phosphodiesterase inhibitors to placebo. Evidence was very uncertain on whether phosphodiesterase inhibitors have any impact on death (any cause) (Analysis 12.1).

Renin inhibitors versus placebo

Two studies (ALOFT 2008; ARIANA‐CHF‐RD 2016) compared renin inhibitors to placebo. Evidence was very uncertain on whether renin inhibitors have any impact on worsening HF (Analysis 13.1) as only ARIANA‐CHF‐RD 2016 reported this outcome. Treatment with renin inhibitors had uncertain effects on worsening kidney function (Analysis 13.2 (2 studies, 142 participants): RR 1.52, 95% CI 0.22 to 10.33; I2 = 36%; very low certainty evidence). The risk of hyperkalaemia was uncertain with renin inhibitors (Analysis 13.3 (2 studies, 142 participants): RR 0.86, 95% CI 0.49 to 1.49; I2 = 0%; very low certainty evidence). The risk of hypotension was uncertain with renin inhibitors (Analysis 13.4 (2 studies, 142 participants):RR 1.44, 95% CI 0.62 to 3.31; I2 = 0%; very low certainty evidence).

Sinus node inhibitors versus placebo

One study (SHIFT 2010) compared sinus node inhibitors to placebo. Evidence was very uncertain on whether sinus node inhibitors have any impact on cardiovascular death (Analysis 14.1), hospitalisation for HF (Analysis 14.2), or worsening kidney function (Analysis 14.3). There were 14/780 patients with at least one episode of hyperkalaemia in the sinus node inhibitor group and 27/799 in the placebo group (Analysis 14.4).

Vasopressin receptor antagonists versus placebo

Two studies (AQUAMARINE 2014; EVEREST 2005) compared vasopressin receptor antagonists to placebo. Vasopressin receptor antagonists had uncertain effects on death (any cause) (Analysis 15.1 (2 studies, 1840 participants): RR 1.26, 95% CI 0.55 to 2.89; I2 = 41%; low certainty evidence). Evidence was very uncertain on whether vasopressin receptor antagonists have any impact on HF hospitalisations (Analysis 15.2) or worsening kidney function (Analysis 15.3), as only AQUAMARINE 2014 reported these two outcomes.

Discussion

Summary of main results

In this systematic review of HF therapies in people with CKD, we identified 112 studies: 15 studies reported data on CKD patients only; 16 studies included both people with and without CKD, and provided stratified data for CKD patients; and 81 studies included people with and without CKD, however data for those with CKD was not available. Twenty‐six studies with disaggregated and extractable data in people with CKD were included in our meta‐analyses. In acute settings, six studies investigated dopamine, nesiritide, furosemide (dose reduction), serelaxin, and adenosine A1 receptor antagonists. Based on the level of evidence available, it is uncertain whether these therapies make any difference to death, hospitalisations, or adverse effects including worsening kidney function, hyperkalaemia, or hypotension. In chronic settings, 18 studies investigated ACEi, ARB, aldosterone antagonists, anti‐arrhythmic, beta‐blockers, digoxin, phosphodiesterase inhibitors, renin inhibitors, sinus node inhibitors, vasodilators, and vasopressin receptor antagonists. It is uncertain whether these interventions make any difference to death, hospitalisations, or harms, as the number of studies available was low. Due to few data observations, studies with zero events in both arms, and situations where only single studies were available for clinical outcomes, meta‐analysis was not possible for most of the review outcomes, particularly for harms.

Considering that CKD is a significant independent predictor of worse outcome in acute or chronic HF (Grande 2017; Smith 2006), knowledge about benefits and harms of recommended interventions for HF in this patient population is important. Insufficient evidence is available to inform decision‐making.

Overall completeness and applicability of evidence

Of the 112 included studies, quantitative data on CKD patients could be extracted in only 23% of studies. In total, 26 studies were eligible for meta‐analysis. Due to the diversity of drug classes and outcomes available, few summary estimates of the effects were possible. Meta‐analysis in acute settings was not performed. Meta‐analysis for death (any cause) was only possible for ACEi/ARB, anti‐aldosterone, beta‐blockers, and vasopressin receptor antagonists. Meta‐analysis for cardiovascular death was only possible for ACEi/ARs and beta‐blockers. Re‐hospitalisation for HF was meta‐analysed only for ACEi/ARB and beta‐blockers. Analyses for harms were only possible for renin inhibitors (worsening kidney function and hyperkalaemia) and aldosterone antagonists (hyperkalaemia). Overall, the certainty of the evidence across most interventions and outcomes was very low due to the limited number of studies. Treatment with beta‐blockers may reduce risk of all‐cause or cardiovascular death; however, the certainty was moderate and low, respectively. Furthermore, due to the few number of studies in each meta‐analysis, sub‐group analyses by type of HF (systolic or diastolic) or level of kidney failure were not possible.

Quality of the evidence

Of the 112 studies, the risks of bias were generally high, or unclear. The risk of selection bias was unclear in the majority of studies due to unclear randomisation methods or allocation concealment. Over three‐quarters of studies blinded participants and nearly half blinded outcome assessors. The majority reported on outcomes as per a pre‐specified protocol. Over 70% of studies followed an intention‐to‐treat approach. Nearly one‐third were at high risk of attrition bias. In most studies, the study sponsor was involved in the study or their role was unclear.

Potential biases in the review process

Although based on a peer‐reviewed protocol and conducted using methods developed by Cochrane, our review has limitations mostly related to lack of data availability. Data for CKD populations that could be extracted for meta‐analysis were only available in 23% of studies. Considering that our review was broad in scope and included a variety of pharmacological interventions and outcomes, availability of more data on interventions and outcome would be key to obtaining summary measures of effect. Of the studies in the CKD population (or subgroup analyses), the definition of CKD varied and ranged from having eGFR < 60 mL/min/1.73 m2 to being on dialysis. Many of the studies did not specify the stage of CKD among patients with an eGFR < 60 mL/min/1.73 m2, which may be problematic as the effect or safety of these interventions likely differs based on early versus advanced CKD, Further, many of these studies defined CKD by a baseline eGFR, which may not represent a true CKD diagnosis if not confirmed with a second measure three months apart. Generalisation of findings to people with HF and CKD requires additional data, some of which could be made available from studies already completed.

Agreements and disagreements with other studies or reviews

Our review only identified six studies that assessed interventions for acute HF with a follow‐up of at least three months: investigating dopamine or nesiritide (1 study), serelaxin (2), adenosine A1 receptor antagonists (2) and diuretics (1). We were unable to estimate risk of any outcome for studies investigating diuretics, nesiritide, or dopamine in an acute setting. No significant reduction in risk of death (any cause) was found for adenosine A1 receptor antagonists; however, the certainty of the evidence was low. No other analyses could be performed for this intervention due to limited data. Serelaxin did not significantly reduce risk of all‐cause or cardiovascular death nor increase the risk of worsening kidney function or hypotension. There appeared to be a decrease in risk of worsening HF; however, the certainty of evidence for all analyses in this intervention group was very low. Previous research has suggested that kidney impairment may affect the response to interventions for acute decompensated HF, such as tolvaptan (Ikeda 2017). Further data are needed to clarify whether reduced kidney function hampers the efficacy of the interventions for HF or leads to excess harm. While we acknowledge that MI and HF often are physiopathologically related and overlap, to address the specific question of this review, we excluded studies that examined pharmacological interventions in people with acute MI without HF.

The ACC/AHA Clinical Practice Guidelines for people with HF (Yancy 2017) recommend the use of renin‐angiotensin system inhibition (with ACEi or ARB) in conjunction with beta‐blockers and aldosterone antagonists, in selected patients with chronic HFrEF. Of note, the guidelines mention that treatment with ACEi and ARBs should be administered with caution to people with HFrEF renal insufficiency, and further, that aldosterone antagonists may be beneficial for people with HFpEF only if their eGFR > 30 mL/min/17.3 m2 (Yancy 2017). Further, a 2014 review exploring the efficacy of HF interventions suggested a benefit among people with moderate CKD (i.e. stage 3), but the evidence in those with more advanced CKD (stage 4 to 5) is unknown (Damman 2014b). Our review was unable to differentiate CKD stage due to limited data accessibility. However, we found that beta‐blockers may reduce risk of all‐cause and cardiovascular death in people with HF who also have CKD. Pooling studies administering ARB or ACEi found no significant risk reduction in the CKD population for any outcome; however, one study (Cice 2010) reported that treatment with ARB significantly reduced the risk of cardiovascular death as well as hospitalisation for HF. We searched the OVID database and found three reviews that evaluated the combination of ARB and ACEi versus a single agent (Heran 2012; Kuenzli 2010; Lee 2004). While reviews have shown a significant benefit of ARB versus placebo (Heran 2012; Lee 2004) and a similar effect of ARB versus ACE inhibitors (Heran 2012; Kuenzli 2010; Lee 2004), we were unable to find any published reviews of studies comparing the effects of ACEi versus placebo. The Cochrane Database of Systematic Reviews reports protocols currently investigating a number of interventions: ACEi or ARB (http://www.cochrane.org/CD003040/VASC_are‐angiotensin‐receptor‐blockers‐arbs‐an‐effective‐treatment‐for‐heart‐failure) and beta‐blockers (http://www.cochrane.org/CD012897/VASC_beta‐blockers‐heart‐failure). This research will be important to ensure guidelines are based on high quality systematic reviews of RCTs.

Following appropriate ACEi/ARB dose‐adjustment based on eGFR may be important for people with HF and impaired kidney function (Grande 2017). This was addressed in only one study (Pita‐Fernandez 2015), which found that reducing the dose of ACEi by 50% in people with HF and CKD resulted in an improvement in anaemia and kidney function, decreased protein C, and improved survival. Further research into dose‐adjustment for people with CKD may be important for preventing the progression of kidney failure.

The effect of beta‐blockers did not significantly reduce the risk of all‐cause or cardiovascular death in a study including people over 70 years of age (SENIORS 2002), which is consistent with findings from the general population (Grossman 2002). While that study appears to show a trend toward an increased risk of worsening HF with beta‐blockers, the estimate of this effect was imprecise. An observational study reported that beta‐blocker use in patients with HF and CKD was safe and effective in reducing risk of death or hospitalisations (Chang 2013), but RCTs in this population are warranted (Franco Peláez 2016). Insufficient data are available on potential harms, including hypotension and acute kidney injury in people with CKD.

Our review reported an uncertain effect of aldosterone antagonists on death (any cause). Other research investigating the use of aldosterone antagonists in people on dialysis (with or without HF) reported a protective effect with respect to all‐cause and cardiovascular death (Quach 2016). However, hyperkalaemia may be a concern in this patient population. The authors of the same review were unable to provide any recommendations due to insufficient sample size and the quality of the study. Due to the potential risk of hyperkalaemia in CKD patients who also have HF (as we found pooling the results of three studies), further work investigating the safety and efficacy of aldosterone antagonists, and other agents that block the renin‐angiotensin system, is warranted.

Our review found a significant protective effect of aliskiren on reducing the risk of worsening HF in one study (ARIANA‐CHF‐RD 2016). However, while not significant, a trend suggesting a potential harmful effect on kidney function with administration of aliskiren was reported in two studies (ALOFT 2008; ARIANA‐CHF‐RD 2016). Other research has suggested that aliskiren is safe for use in people with CKD due to its anti‐proteinuric effect (Persson 2013). However, much of the research done pertains to the treatment of hypertension and findings may not necessarily apply to HF populations. Furthermore, risks of hyperkalaemia with this intervention could be of concern for patients with CKD (Morishita 2013).

There are a number of additional therapies for the treatment of HF (Lother 2016). Our review identified studies investigating anti‐arrhythmics (dofetilide), digoxin, phosphodiesterase inhibitors (milrinone), sinus node inhibitors (ivabradine), and vasopressin receptor antagonists (tolvaptan). Evidence on the safety and efficacy of these interventions in people with CKD is limited and further research is needed before any conclusions may be drawn about their use in this patient population. As the co‐occurrence of HF and CKD appears to be increasing (Conrad 2018), a better understanding of how to manage this patient population, particularly with a focus on the potential harms of conventional therapies, is important.

Study flow diagram.
Figures and Tables -
Figure 1

Study flow diagram.

Navigate to figure in ReviewOpen in new tab

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Figures and Tables -
Figure 2

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Navigate to figure in ReviewOpen in new tab

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Figures and Tables -
Figure 3

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Navigate to figure in ReviewOpen in new tab

Comparison 1 ACUTE: Diuretic bolus versus continuous infusion, Outcome 1 Cardiovascular death.
Figures and Tables -
Analysis 1.1

Comparison 1 ACUTE: Diuretic bolus versus continuous infusion, Outcome 1 Cardiovascular death.

Navigate to figure in ReviewOpen in new tab

Comparison 1 ACUTE: Diuretic bolus versus continuous infusion, Outcome 2 Hospitalisation for heart failure.
Figures and Tables -
Analysis 1.2

Comparison 1 ACUTE: Diuretic bolus versus continuous infusion, Outcome 2 Hospitalisation for heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 2.1

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 2 Cardiovascular death.
Figures and Tables -
Analysis 2.2

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 2 Cardiovascular death.

Navigate to figure in ReviewOpen in new tab

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 3 Worsening heart failure.
Figures and Tables -
Analysis 2.3

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 3 Worsening heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 4 Worsening kidney function.
Figures and Tables -
Analysis 2.4

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 4 Worsening kidney function.

Navigate to figure in ReviewOpen in new tab

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 5 Hyperkalaemia.
Figures and Tables -
Analysis 2.5

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 5 Hyperkalaemia.

Navigate to figure in ReviewOpen in new tab

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 6 Hypotension.
Figures and Tables -
Analysis 2.6

Comparison 2 ACUTE: Serelaxin versus placebo, Outcome 6 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 3.1

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 2 Hospitalisation (any cause).
Figures and Tables -
Analysis 3.2

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 2 Hospitalisation (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 3 Worsening heart failure.
Figures and Tables -
Analysis 3.3

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 3 Worsening heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 4 Worsening kidney function.
Figures and Tables -
Analysis 3.4

Comparison 3 ACUTE: Adenosine A1‐receptor antagonist versus placebo, Outcome 4 Worsening kidney function.

Navigate to figure in ReviewOpen in new tab

Comparison 4 ACUTE: Dopamine versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 4.1

Comparison 4 ACUTE: Dopamine versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 4 ACUTE: Dopamine versus placebo, Outcome 2 Worsening heart failure.
Figures and Tables -
Analysis 4.2

Comparison 4 ACUTE: Dopamine versus placebo, Outcome 2 Worsening heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 4 ACUTE: Dopamine versus placebo, Outcome 3 Hypotension.
Figures and Tables -
Analysis 4.3

Comparison 4 ACUTE: Dopamine versus placebo, Outcome 3 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 5 ACUTE: Nesiritide versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 5.1

Comparison 5 ACUTE: Nesiritide versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 5 ACUTE: Nesiritide versus placebo, Outcome 2 Worsening heart failure.
Figures and Tables -
Analysis 5.2

Comparison 5 ACUTE: Nesiritide versus placebo, Outcome 2 Worsening heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 5 ACUTE: Nesiritide versus placebo, Outcome 3 Hypotension.
Figures and Tables -
Analysis 5.3

Comparison 5 ACUTE: Nesiritide versus placebo, Outcome 3 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 6.1

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 2 Cardiovascular death.
Figures and Tables -
Analysis 6.2

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 2 Cardiovascular death.

Navigate to figure in ReviewOpen in new tab

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 3 Hospitalisation for heart failure.
Figures and Tables -
Analysis 6.3

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 3 Hospitalisation for heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 4 Hypotension.
Figures and Tables -
Analysis 6.4

Comparison 6 CHRONIC: ACEi OR ARB versus placebo, Outcome 4 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 7 CHRONIC: 50% ARB dose reduction versus full dose, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 7.1

Comparison 7 CHRONIC: 50% ARB dose reduction versus full dose, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 8.1

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 2 Cardiovascular death.
Figures and Tables -
Analysis 8.2

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 2 Cardiovascular death.

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 3 Hospitalisation (any cause).
Figures and Tables -
Analysis 8.3

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 3 Hospitalisation (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 4 Hospitalisation for heart failure.
Figures and Tables -
Analysis 8.4

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 4 Hospitalisation for heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 5 Worsening heart failure.
Figures and Tables -
Analysis 8.5

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 5 Worsening heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 6 Worsening kidney function.
Figures and Tables -
Analysis 8.6

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 6 Worsening kidney function.

Navigate to figure in ReviewOpen in new tab

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 7 Hypotension.
Figures and Tables -
Analysis 8.7

Comparison 8 CHRONIC: Beta‐blockers versus placebo, Outcome 7 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 9 CHRONIC: Aldosterone antagonists versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 9.1

Comparison 9 CHRONIC: Aldosterone antagonists versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 9 CHRONIC: Aldosterone antagonists versus placebo, Outcome 2 Hyperkalaemia.
Figures and Tables -
Analysis 9.2

Comparison 9 CHRONIC: Aldosterone antagonists versus placebo, Outcome 2 Hyperkalaemia.

Navigate to figure in ReviewOpen in new tab

Comparison 10 CHRONIC: Anti‐arrhythmics versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 10.1

Comparison 10 CHRONIC: Anti‐arrhythmics versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 11.1

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 2 Cardiovascular death.
Figures and Tables -
Analysis 11.2

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 2 Cardiovascular death.

Navigate to figure in ReviewOpen in new tab

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 3 Hypotension.
Figures and Tables -
Analysis 11.3

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 3 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 4 Quality of Life.
Figures and Tables -
Analysis 11.4

Comparison 11 CHRONIC: Nesiritide versus placebo, Outcome 4 Quality of Life.

Navigate to figure in ReviewOpen in new tab

Comparison 12 CHRONIC: Phosphodiesterase inhibitors versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 12.1

Comparison 12 CHRONIC: Phosphodiesterase inhibitors versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 1 Worsening heart failure.
Figures and Tables -
Analysis 13.1

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 1 Worsening heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 2 Worsening kidney function.
Figures and Tables -
Analysis 13.2

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 2 Worsening kidney function.

Navigate to figure in ReviewOpen in new tab

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 3 Hyperkalaemia.
Figures and Tables -
Analysis 13.3

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 3 Hyperkalaemia.

Navigate to figure in ReviewOpen in new tab

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 4 Hypotension.
Figures and Tables -
Analysis 13.4

Comparison 13 CHRONIC: Renin inhibitors versus placebo, Outcome 4 Hypotension.

Navigate to figure in ReviewOpen in new tab

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 1 Cardiovascular death.
Figures and Tables -
Analysis 14.1

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 1 Cardiovascular death.

Navigate to figure in ReviewOpen in new tab

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 2 Hospitalisation for heart failure.
Figures and Tables -
Analysis 14.2

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 2 Hospitalisation for heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 3 Worsening kidney function.
Figures and Tables -
Analysis 14.3

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 3 Worsening kidney function.

Navigate to figure in ReviewOpen in new tab

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 4 Hyperkalaemia.
Figures and Tables -
Analysis 14.4

Comparison 14 CHRONIC: Sinus node inhibitors versus placebo, Outcome 4 Hyperkalaemia.

Navigate to figure in ReviewOpen in new tab

Comparison 15 CHRONIC: Vasopressin receptor antagonists versus placebo, Outcome 1 Death (any cause).
Figures and Tables -
Analysis 15.1

Comparison 15 CHRONIC: Vasopressin receptor antagonists versus placebo, Outcome 1 Death (any cause).

Navigate to figure in ReviewOpen in new tab

Comparison 15 CHRONIC: Vasopressin receptor antagonists versus placebo, Outcome 2 Hospitalisation for heart failure.
Figures and Tables -
Analysis 15.2

Comparison 15 CHRONIC: Vasopressin receptor antagonists versus placebo, Outcome 2 Hospitalisation for heart failure.

Navigate to figure in ReviewOpen in new tab

Comparison 15 CHRONIC: Vasopressin receptor antagonists versus placebo, Outcome 3 Worsening kidney function.
Figures and Tables -
Analysis 15.3

Comparison 15 CHRONIC: Vasopressin receptor antagonists versus placebo, Outcome 3 Worsening kidney function.

Navigate to figure in ReviewOpen in new tab
Summary of findings for the main comparison. ACUTE: Serelaxin versus placebo in patients with heart failure and chronic kidney disease

ACUTE: Serelaxin versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: acute, inpatient

Intervention: serelaxin

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

Serelaxin

Death (any cause)

42 per 1000

25 per 1000

(14 to 43)

RR 0.59

(0.34 to 1.02)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Cardiovascular death

97 per 1000

46 per 1000

(20 to 105)

RR 0.47 (0.21 to 1.08)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Hospitalisation (any cause)

not reported

not reported

‐‐

‐‐

‐‐

Heart failure hospitalisation

not reported

not reported

‐‐

‐‐

‐‐

Worsening heart failure

162 per 1000

115 per 1000

(87 to 149)

RR 0.71 (0.54 to 0.92)

1395 (2)

⊕⊕⊝⊝
low3,4

Worsening kidney function

86 per 1000

74 per 1000

(32 to 170)

RR 0.86 (0.37 to 1.98)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Hypotension

48 per 1000

60 per 1000

(39 to 94)

RR 1.26 (0.81 to 1.96)

1395 (2)

⊕⊕⊝⊝
low2,3,4

Hyperkalaemia

48 per 1000

52 per 1000

(15 to 187)

RR 1.08 (0.30 to 3.87

234 (1)

⊕⊝⊝⊝
very low2,3,4

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

CI: Confidence interval; RR: Risk Ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Figures and Tables -
Summary of findings for the main comparison. ACUTE: Serelaxin versus placebo in patients with heart failure and chronic kidney disease
Navigate to table in Review
Summary of findings 2. CHRONIC: ACEi or ARB versus placebo in patients with heart failure and chronic kidney disease

CHRONIC: ACEi or ARB versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: chronic, outpatient

Intervention: ACEi or ARB

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

ACEi or ARB

Death (any cause)

(follow‐up 2 to 3.5 years)

320 per 1000

272 per 1000
(224 to 326)

RR 0.85 (0.70 to 1.02)

5003 (4)

⊕⊕⊝⊝
low2,3,4

Cardiovascular death

(follow‐up 3 to 4 years)

411 per 1000

333 per 1000
(263 to 415)

RR 0.81 (0.64 to 1.01)

1368 (2)

⊕⊕⊝⊝
low2,3,4

Hospitalisation (any cause)

not reported

not reported

‐‐

‐‐

‐‐

Heart failure hospitalisation

(follow‐up 2 to 3.5 years)

546 per 1000

491 per 1000
(235 to 1037)

RR 0.90 (0.43 to 1.90)

1368 (2)

⊕⊝⊝⊝
very low2 ,3,4

Worsening heart failure

not reported

not reported

‐‐

‐‐

‐‐

Worsening kidney function

not reported

not reported

‐‐

‐‐

‐‐

Hypotension (ARB; withdrawn from study)

42 per 1000

109 per 1000

(47 to 254

RR 2.60 (1.12 to 6.07)

332 (1)

⊕⊝⊝⊝
very low3,4

Hyperkalaemia

not reported

not reported

‐‐

‐‐

‐‐

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

*The assumed risk (e.g. the median control group risk across studies) was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. The corresponding risk was calculated by multiplying the assumed risk with the relative effect from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

CI: Confidence interval; RR: Risk Ratio; ACEi: angiotensin‐converting enzyme inhibitor; ARB: angiotensin receptor blocker

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Figures and Tables -
Summary of findings 2. CHRONIC: ACEi or ARB versus placebo in patients with heart failure and chronic kidney disease
Navigate to table in Review
Summary of findings 3. CHRONIC: Beta‐blockers versus placebo in patients with heart failure and chronic kidney disease

CHRONIC: Beta‐blockers versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: chronic, outpatient

Intervention: beta‐blockers

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

Beta‐blockers

Death (any cause)

(follow‐up 1 to 2.5 years)

215 per 1000

148 per 1000
(129to 170)

RR 0.69

(0.60 to 0.79)

3136 (4)

⊕⊕⊝⊝
moderate3,4

Cardiovascular death

(follow‐up 1 to 2.5 years)

123 per 1000

65 per 1000
(43 to 100)

RR 0.53 (0.35 to 0.81)

2287 (3)

⊕⊕⊝⊝
low3,4

Hospitalisation (any cause)

432 per 1000

324 per 1000
(225 to 467)

RR 0.75 (0.52 to 1.08)

1583 (2)

⊕⊕⊝⊝
low2,3,4

Heart failure hospitalisation

(follow‐up 1 to 2.5 years)

318 per 1000

213 per 1000
(137 to 334)

RR 0.67 (0.43 to 1.05)

2287 (3)

⊕⊕⊝⊝
low2,3,4

Worsening heart failure

25 per 1000

34 per 1000

RR 1.36

(0.58 to 3.20)

704 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

no events

no events

‐‐

704 (1)

⊕⊝⊝⊝
very low3,4

Hypotension

no events

2/356**

RR 5.11

(0.25 to 106.15)

704 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

not reported

not reported

‐‐

‐‐

‐‐

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

*The assumed risk (e.g. the median control group risk across studies) was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. The corresponding risk was calculated by multiplying the assumed risk with the relative effect from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

** Event rate derived from the raw data. A 'per thousand' rate is non‐informative in view of the scarcity of evidence and zero events in the control group

CI: Confidence interval; RR: Risk Ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Figures and Tables -
Summary of findings 3. CHRONIC: Beta‐blockers versus placebo in patients with heart failure and chronic kidney disease
Navigate to table in Review
Summary of findings 4. CHRONIC: Aldosterone antagonists versus placebo in patients with heart failure and chronic kidney disease

CHRONIC: Aldosterone antagonists versus placebo in patients with heart failure and chronic kidney disease

Patient or population: heart failure with chronic kidney disease

Settings: chronic, outpatient

Intervention: aldosterone antagonists

Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No. of participants
(studies)

Quality of the evidence
(GRADE)

Assumed risk

Corresponding risk

Placebo

Anti‐aldosterone

Death (any cause)

(follow‐up 6 to 21 months)

294 per 1000

179 per 1000
(17 to 1,938)

RR 0.61 (0.06 to 6.59)

34 (2)

⊕⊝⊝⊝
very low1,2,3,4

Cardiovascular death

not reported

not reported

‐‐

‐‐

‐‐

Hospitalisation (any cause)

not reported

not reported

‐‐

‐‐

‐‐

Heart failure hospitalisation

not reported

not reported

‐‐

‐‐

‐‐

Worsening heart failure

not reported

not reported

‐‐

‐‐

‐‐

Worsening kidney function

not reported

not reported

‐‐

‐‐

‐‐

Hypotension

not reported

not reported

‐‐

‐‐

‐‐

Hyperkalaemia

follow‐up 6‐24 months

84 per 1000

243 per 1000
(170 to 350)

RR 2.91 (2.03 to 4.17)

826 (3)

⊕⊕⊝⊝
Low1,4

Quality of life

not reported

not reported

‐‐

‐‐

‐‐

*The assumed risk (e.g. the median control group risk across studies) was calculated by dividing the number of participants in the control group that experienced an event by the total number of participants in the control group. The corresponding risk was calculated by multiplying the assumed risk with the relative effect from the meta‐analysis. The 95% CI for the corresponding risk was calculated by multiplying the assumed risk with the lower and upper CI of the risk ratio estimate from the meta‐analysis.

CI: Confidence interval; RR: Risk Ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Figures and Tables -
Summary of findings 4. CHRONIC: Aldosterone antagonists versus placebo in patients with heart failure and chronic kidney disease
Navigate to table in Review
Table 1. Additional Summary of Findings

Comparison

Outcome

Relative effect (95% CI)

No. participants (studies)

Quality of the evidence
(GRADE)

Acute heart failure

Adenosine A1‐receptor antagonist versus placebo

Death (any cause)

RR 0.71 (0.39 to 1.26)

2078 (2)

⊕⊕⊝⊝

low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

RR 0.85 (0.43 to 1.68)

45 (1)

⊕⊝⊝⊝
very low2,3,4

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 0.93 (0.70 to 1.24)

2033 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

1.14 (0.89 to 1.48)

2033 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Diuretic bolus versus continuous infusion

Death (any cause)

Not reported

‐‐

Cardiovascular death

RR 0.56 (0.15 to 2.01)

57 (1)

⊕⊝⊝⊝
very low2,3,4

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

RR 0.83 (0.33 to 2.10)

57 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Dopamine versus placebo

Death (any cause)

RR 0.89 (0.41 to 1.95)

241 (1)

⊕⊝⊝⊝
very low1,2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 2.15 (0.77 to 5.99)

241 (1)

⊕⊝⊝⊝
very low1,2,3,4

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

RR 0.09 (0.01 to 0.65)

226 (1)

⊕⊝⊝⊝
very low1,2,3,4

Quality of life

Not reported

‐‐

Nesiritide versus placebo

Death (any cause)

RR 0.67 (0.28 to 1.57)

238 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 0.83 (0.26 to 2.66)

238 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

RR 1.80 (0.94 to 3.47)

232 (1)

⊕⊝⊝⊝
very low2,3,4

Quality of life

Not reported

‐‐

Chronic heart failure

Anti‐arrhythmics versus placebo

Death (any cause)

RR 0.95 (0.83 to 1.09)

755 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Nesiritide versus placebo

Death (any cause)

RR 0.98 (0.64 to 1.50)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

RR 0.88 (0.56 to 1.38)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

RR 1.80 (1.38 to 2.35)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Quality of Life

MD 1.20 (‐1.85 to 4.25)

911 (1)

⊕⊝⊝⊝
very low2,3,4

Phosphodiesterase inhibitors versus placebo

Death (any cause)

RR 1.26 (0.98 to 1.61)

539 (1)

⊕⊝⊝⊝
very low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

Not reported

‐‐

Worsening kidney function

Not reported

‐‐

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Renin inhibitors versus placebo

Death (any cause)

Not reported

‐‐

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

Not reported

‐‐

Worsening heart failure

RR 0.26 (0.08 to 0.88)

41 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening kidney function

RR 1.52 (0.22 to 10.33)

142 (2)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

RR 0.86 (0.49 to 1.49)

142 (2)

⊕⊝⊝⊝
very low2,3,4

Hypotension

RR 1.44 (0.62 to 3.31)

142 (2)

⊕⊝⊝⊝
very low2,3,4

Quality of life

Not reported

‐‐

Sinus node inhibitors versus placebo

Death (any cause)

Not reported

Cardiovascular death

RR 0.96 (0.48 to 1.93)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

RR 0.83 (0.48 to 1.45)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening heart failure

Not reported

‐‐

Worsening kidney function

RR 0.95 (0.71 to 1.27)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

RR 0.53 (0.28 to 1.01)

1576 (1)

⊕⊝⊝⊝
very low2,3,4

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

Vasopressin receptor antagonists versus placebo

Death (any cause)

RR 1.26 (0.55 to 2.89)

1840 (2)

⊕⊝⊝⊝
low2,3,4

Cardiovascular death

Not reported

‐‐

Hospitalisation (any cause)

Not reported

‐‐

Heart failure hospitalisation

RR 0.82 (0.37 to 1.81)

208 (1)

⊕⊝⊝⊝
very low2,3,4

Worsening heart failure

Not reported

‐‐

Worsening kidney function

RR 0.87 (0.56 to 1.38)

217 (1)

⊕⊝⊝⊝
very low2,3,4

Hyperkalaemia

Not reported

‐‐

Hypotension

Not reported

‐‐

Quality of life

Not reported

‐‐

CI: Confidence interval; RR: Risk Ratio; MD: Mean difference

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Important study limitations due to risks of bias

2 The confidence intervals include potential for important benefits and harms

3 Important and unexplained heterogeneity present (or insufficient data observations to evaluate)

4 Insufficient data observations to evaluate

Figures and Tables -
Table 1. Additional Summary of Findings
Navigate to table in Review
Comparison 1. ACUTE: Diuretic bolus versus continuous infusion

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Cardiovascular death Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

2 Hospitalisation for heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 1. ACUTE: Diuretic bolus versus continuous infusion
Navigate to table in Review
Comparison 2. ACUTE: Serelaxin versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

2

1395

Risk Ratio (M‐H, Random, 95% CI)

0.59 [0.34, 1.02]

2 Cardiovascular death Show forest plot

2

1395

Risk Ratio (M‐H, Random, 95% CI)

0.47 [0.21, 1.08]

3 Worsening heart failure Show forest plot

2

1395

Risk Ratio (M‐H, Random, 95% CI)

0.71 [0.54, 0.92]

4 Worsening kidney function Show forest plot

2

1395

Risk Ratio (M‐H, Random, 95% CI)

0.86 [0.37, 1.98]

5 Hyperkalaemia Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

6 Hypotension Show forest plot

2

1395

Risk Ratio (M‐H, Random, 95% CI)

1.26 [0.81, 1.96]
Figures and Tables -
Comparison 2. ACUTE: Serelaxin versus placebo
Navigate to table in Review
Comparison 3. ACUTE: Adenosine A1‐receptor antagonist versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

2

2078

Risk Ratio (M‐H, Random, 95% CI)

0.71 [0.39, 1.26]

2 Hospitalisation (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

3 Worsening heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

4 Worsening kidney function Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 3. ACUTE: Adenosine A1‐receptor antagonist versus placebo
Navigate to table in Review
Comparison 4. ACUTE: Dopamine versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

2 Worsening heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

3 Hypotension Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 4. ACUTE: Dopamine versus placebo
Navigate to table in Review
Comparison 5. ACUTE: Nesiritide versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

2 Worsening heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

3 Hypotension Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 5. ACUTE: Nesiritide versus placebo
Navigate to table in Review
Comparison 6. CHRONIC: ACEi OR ARB versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

4

5003

Risk Ratio (M‐H, Random, 95% CI)

0.85 [0.70, 1.02]

1.1 ACEi versus placebo

2

1755

Risk Ratio (M‐H, Random, 95% CI)

0.85 [0.70, 1.03]

1.2 ARB versus placebo

2

3248

Risk Ratio (M‐H, Random, 95% CI)

0.83 [0.52, 1.32]

2 Cardiovascular death Show forest plot

2

1368

Risk Ratio (M‐H, Random, 95% CI)

0.81 [0.64, 1.01]

2.1 ACEi versus placebo

1

1036

Risk Ratio (M‐H, Random, 95% CI)

0.88 [0.75, 1.03]

2.2 ARB versus placebo

1

332

Risk Ratio (M‐H, Random, 95% CI)

0.69 [0.52, 0.92]

3 Hospitalisation for heart failure Show forest plot

2

1368

Risk Ratio (M‐H, Random, 95% CI)

0.90 [0.43, 1.90]

3.1 ACEi versus placebo

1

1036

Risk Ratio (M‐H, Random, 95% CI)

1.30 [1.18, 1.43]

3.2 ARB versus placebo

1

332

Risk Ratio (M‐H, Random, 95% CI)

0.62 [0.48, 0.79]

4 Hypotension Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

4.1 ARB versus placebo

1

Risk Ratio (M‐H, Random, 95% CI)

0.0 [0.0, 0.0]
Figures and Tables -
Comparison 6. CHRONIC: ACEi OR ARB versus placebo
Navigate to table in Review
Comparison 7. CHRONIC: 50% ARB dose reduction versus full dose

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 7. CHRONIC: 50% ARB dose reduction versus full dose
Navigate to table in Review
Comparison 8. CHRONIC: Beta‐blockers versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

4

3136

Risk Ratio (M‐H, Random, 95% CI)

0.69 [0.60, 0.79]

2 Cardiovascular death Show forest plot

3

2287

Risk Ratio (M‐H, Random, 95% CI)

0.53 [0.35, 0.81]

3 Hospitalisation (any cause) Show forest plot

2

1583

Risk Ratio (M‐H, Random, 95% CI)

0.75 [0.52, 1.08]

4 Hospitalisation for heart failure Show forest plot

3

2287

Risk Ratio (M‐H, Random, 95% CI)

0.67 [0.43, 1.05]

5 Worsening heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

6 Worsening kidney function Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

7 Hypotension Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 8. CHRONIC: Beta‐blockers versus placebo
Navigate to table in Review
Comparison 9. CHRONIC: Aldosterone antagonists versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

2

34

Risk Ratio (M‐H, Random, 95% CI)

0.61 [0.06, 6.59]

2 Hyperkalaemia Show forest plot

3

826

Risk Ratio (M‐H, Random, 95% CI)

2.91 [2.03, 4.17]
Figures and Tables -
Comparison 9. CHRONIC: Aldosterone antagonists versus placebo
Navigate to table in Review
Comparison 10. CHRONIC: Anti‐arrhythmics versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 10. CHRONIC: Anti‐arrhythmics versus placebo
Navigate to table in Review
Comparison 11. CHRONIC: Nesiritide versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

2 Cardiovascular death Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

3 Hypotension Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

4 Quality of Life Show forest plot

1

Mean Difference (IV, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 11. CHRONIC: Nesiritide versus placebo
Navigate to table in Review
Comparison 12. CHRONIC: Phosphodiesterase inhibitors versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 12. CHRONIC: Phosphodiesterase inhibitors versus placebo
Navigate to table in Review
Comparison 13. CHRONIC: Renin inhibitors versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Worsening heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

2 Worsening kidney function Show forest plot

2

142

Risk Ratio (M‐H, Random, 95% CI)

1.52 [0.22, 10.33]

3 Hyperkalaemia Show forest plot

2

142

Risk Ratio (M‐H, Random, 95% CI)

0.86 [0.49, 1.49]

4 Hypotension Show forest plot

2

142

Risk Ratio (M‐H, Random, 95% CI)

1.44 [0.62, 3.31]
Figures and Tables -
Comparison 13. CHRONIC: Renin inhibitors versus placebo
Navigate to table in Review
Comparison 14. CHRONIC: Sinus node inhibitors versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Cardiovascular death Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

2 Hospitalisation for heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

3 Worsening kidney function Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

4 Hyperkalaemia Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 14. CHRONIC: Sinus node inhibitors versus placebo
Navigate to table in Review
Comparison 15. CHRONIC: Vasopressin receptor antagonists versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Death (any cause) Show forest plot

2

1840

Risk Ratio (M‐H, Random, 95% CI)

1.26 [0.55, 2.89]

2 Hospitalisation for heart failure Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected

3 Worsening kidney function Show forest plot

1

Risk Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 15. CHRONIC: Vasopressin receptor antagonists versus placebo
Navigate to table in Review