Defining the threshold of permissive renal function decline in heart failure

When managing patients with heart failure, clinicians, cardiologists and nephrologists face a daily dilemma: how much kidney function decline can be tolerated in pursuit of decongestion? After all, it is not uncommon for patients with seemingly chronic, stable trajectories to experience sudden deterioration, culminating in acute kidney injury (AKI) requiring dialysis, intensive care admission or even death. In the outpatient setting, this often leads to an overly cautious approach to diuresis, which can prolong symptom status and premature admission to hospital for closer monitoring of renal function. In the UK, renal dysfunction secondary to diuretics remains the second most common cause of hospitalisation due to an adverse drug reaction.¹

Although earlier studies^2–6 suggest that achieving decongestion, even at the expense of renal function, may be associated with improved survival, it remains uncertain whether there exists a threshold of kidney function beyond which this benefit is negated. While existing guidelines recognise the high prevalence of kidney impairment and caution against continued diuretic use in the presence of renal decline or dehydration, they stop short of defining serum creatinine or estimated glomerular filtration rate (eGFR) thresholds to guide these decisions, largely due to a paucity of data. 

Oka and colleagues⁷ attempt to address this in a post hoc analysis of the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST) trial, which was a randomised controlled trial that investigated the efficacy of tolvaptan in patients with acute decompensated heart failure. The study included patients with heart failure with reduced ejection fraction (HFrEF; left-ventricular ejection fraction ≤40%) with symptoms and signs of congestion, and excluded those with a baseline serum creatinine of >3.5 mg/dL (>309.5 µmol/L). 3404 patients from both arms of the trial who survived hospitalisation and had relevant follow-up data were included and assessed regularly over a median follow-up period of 44.0 (24.7–70.6) weeks.

Outcomes were evaluated in relation to longitudinal changes in volume status and kidney function, with patients stratified by congestion category (improved vs worsened) and by thresholds of renal function decline (no decline, 1%–20%, 21%–40% and >41%), yielding eight comparator groups. To minimise the effects of time-varying confounding, the authors employed advanced statistical methods, such as marginal structural models (MSMs) with inverse probability weighting. The primary outcome was all-cause mortality, and secondary outcomes included cardiovascular mortality and a composite of cardiovascular (CV) mortality and first rehospitalisation for heart failure.

Patients with worsened congestion had an approximately two-fold higher risk of all-cause mortality and CV mortality, and a nearly three-fold higher risk of the composite outcome of CV mortality and heart failure rehospitalisation, compared to those without congestion. For those with renal function decline, an eGFR reduction of 21%–40% was associated with an approximately 1.5-fold risk of all outcomes, whereas a reduction of ≥41% was associated with a 3-fold risk of all-cause and CV mortality and 1.7-fold risk of the composite outcome relative to those with preserved renal function. There were no significant differences in the risk of outcome events when renal function decline was within 20% of the baseline. 

In the analysis of the eight exposure groups, progressive renal function decline was associated with corresponding higher mortality across the congestion-worsened subgroups. By contrast, among those with improved congestion, mortality risk exceeded that of the reference group (worsened congestion but no eGFR decline) only when eGFR decline surpassed 40%. These findings suggest that decongestion generally confers a survival benefit and should take precedence even in the presence of modest kidney function impairment, but that a threshold of >40% eGFR reduction marks a point of substantially increased risk, regardless of congestion status, that warrants extra caution. 

Strengths of the study include the extensive efforts to adjust for confounders and minimise bias, use of longitudinal data from several time intervals, the randomised controlled study setting, large sample size and sensitivity analyses to validate the study results. Notably, the study addresses a gap in the evidence and offers clinicians a much-anticipated safety threshold that may be used to guide clinical practice. Nonetheless, there are important considerations that must be taken into account when interpreting the results. 

In this study, data on important confounders such as atrial fibrillation and chronic kidney disease, including presence of proteinuria and albuminuria which are prognostic in heart failure, were not available and it was not possible to determine which diuretics were used owing to their broad categorisation into a single group. 

Outcomes were analysed according to initial group assignment, meaning subsequent changes in congestion or kidney function were not accounted for, and this may have led to associations not reflective of patients’ actual clinical trajectories. As this study was limited to outpatients with HFrEF, the findings may not be applicable to other heart failure phenotypes or acute inpatient settings where intravenous diuretics may be used instead of oral diuretics and comorbidities, such as infection, may drive AKI via different pathways. 

Renal dysfunction in heart failure encompasses several pathophysiological mechanisms, including kidney hypoperfusion and rising interstitial pressures due to renal venous congestion. AKI resulting from such haemodynamic insults is potentially reversible and is even associated with improved survival, unlike AKI from sepsis or nephrotoxic drugs which carries a worse prognosis.⁸ Urinary biomarker studies indicate that although intensive volume removal is associated with worsening renal tubular injury biomarkers, this is paradoxically associated with better renal function recovery at 60 days compared to when there is no increase in these markers, likely due to the alleviation of congestive nephropathy.³ If this is the case, one could then question whether there is any value to having a rigid renal function threshold in the first place. Rather, it may be more prudent to identify patients, such as those who are older with baseline renal dysfunction, who are more likely to deteriorate even at lower thresholds of renal function decline. 

Lastly, it is important to note that the study was based on data from over 20 years ago. Heart failure management over the last decade has transformed significantly, with the incorporation of drugs such as angiotensin receptor–neprilysin inhibitors and sodium–glucose cotransporter 2 (SGLT2) inhibitors. The latter have been associated with preservation of kidney function and reduction in CV and all-cause mortality in the large SGLT2i randomised controlled trials and numerous studies.^9–14 Thus, the results of the study, if based on a contemporary cohort following the four pillars of heart failure therapy might differ considerably. 

Nonetheless, two important messages emerge from this work. First, clinicians should persevere with diuretic therapy to achieve decongestion even in the presence of mild to moderate renal decline; and, second, there appears to be a threshold of renal function decline beyond which outcomes worsen, underscoring the need for vigilant monitoring. Future prospective studies are needed to validate these observations in the modern treatment era.

A digest of the study can be read here.

References

1. Al-Naher A, Wright D, Devonald MAJ, Pirmohamed M (2017) Renal function monitoring in heart failure – what is the optimal frequency? A narrative review. Br J Clin Pharmacol 84: 5–17
2. Yamada T, Ueyama H, Chopra N et al (2020) Systematic review of the association between worsening renal function and mortality in patients with acute decompensated heart failure. Kidney Int Rep 5: 1486–94
3. Rao VS, Ahmad T, Brisco-Bacik MA et al (2019) Renal effects of intensive volume removal in heart failure patients with preexisting worsening renal function. Circ Heart Fail 2: e005552
4. Testani JM, Chen J, McCauley BD et al (2010) Potential effects of aggressive decongestion during the treatment of decompensated heart failure on renal function and survival. Circulation 122: 265–72
5. Nohria A, Hasselblad V, Stebbins A et al (2008) Cardiorenal interactions. Insights from the ESCAPE trial. J Am Coll Cardiol 51: 1268–74
6. Metra M, Davison B, Bettari L et al (2012) Is worsening renal function an ominous prognostic sign in patients with acute heart failure? The role of congestion and its interaction with renal function. Circ Heart Fail 5: 54–62
7. Oka T, Tighiouart H, McCallum W et al (2025) Thresholds of kidney function decline and congestion status and their relation with outcomes among discharged heart failure patients. Clin J Am Soc Nephrol 11 Jul [Epub ahead of print]
8. Banerjee D, Ali MA, Wang AYM, Jha V (2024) Acute kidney injury in acute heart failure – when to worry and when not to worry? Nephrol Dial Transplant 40: 10–18
9. Wiviott SD, Raz I, Bonaca MP, Mosenzon O et al; DECLARE-TIMI 58 Investigators (2019) Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med 380: 347–57
10. McMurray JJV, Solomon SD, Inzucchi SE et al; DAPA-HF Trial Committees and Investigators (2019) Dapagliflozin in patients with heart failure and reduced ejection fraction. N Engl J Med 381: 1995–2008
11. Zinman B, Wanner C, Lachin JM et al; EMPA-REG OUTCOME Investigators (2015) Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 373: 2117–28 
12. Packer M, Anker SD, Butler J et al; EMPEROR-Reduced Trial Investigators (2020) Cardiovascular and renal outcomes with empagliflozin in heart failure. N Engl J Med 383: 1413–24
13. Anker SD, Butler J, Filippatos G et al; EMPEROR-Preserved Trial Investigators (2021) Empagliflozin in heart failure with a preserved ejection fraction. N Engl J Med 385: 1451–61
14. Neal B, Perkovic V, Mahaffey KW et al; CANVAS Program Collaborative Group (2017) Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 377: 644–57