Spot Urinary Sodium as a Biomarker of Diuretic Response in Acute Heart Failure

Nonstandard Abbreviations and Acronyms

AHF

acute heart failure

ENACT HF

Efficacy of a Standardized Diuretic Protocol in Acute Heart Failure Study

ESC

European Society of Cardiology

PUSH‐AHF

Pragmatic Urinary Sodium‐Based Treatment Algorithm in Acute Heart Failure

ROSE‐AHF

Renal Optimization Strategies Evaluation in Acute Heart Failure

UNa

urinary sodium

Activation of the renin‐angiotensin system in heart failure (HF), a critical adaptation in mammals for survival, is a classic example of maladaptation as it leads to sodium retention‐reabsorption coupled with hyperaldosteronism and increased sympathetic activity. ¹ In HF, especially in acute scenarios, neurohormonal activation enhances the kidney reabsorption of sodium, contributing to loop diuretic resistance. ² Persistent congestion due to an inadequate diuretic response, namely an inadequate quantity of natriuresis, is associated with more severe outcomes. ²

Urinary sodium (UNa) measured in a timed urine collection sample or spot urine measurement has emerged as a promising biomarker of diuretic response in patients with acute HF (AHF). European Society of Cardiology (ESC) HF guidelines ³ recommends an early evaluation of loop diuretic treatment response with a spot UNa analysis in patients with AHF after 2 hours of decongestive therapy initiation. According to this statement, a patient with fluid overload presents a UNa <50 to 70 mEq/L in a urine spot sample after a 2‐hour diuretic administration or an hourly urine output <100 to 150 mL during the first 6 hours, generally identifies those with an insufficient diuretic response, and recommends doubling the intravenous loop diuretic dose or adding another diuretic to obtain incremental diuresis/natriuresis. ³ This proactive approach for AHF management is based on findings from observational studies and expert opinion. ³ , ⁴ More controlled studies, especially randomized controlled, are underway to confirm the utility of UNa for monitoring and guiding the intensity of diuretic therapy in fluid overload patients with AHF syndromes. The paucity of randomized controlled trial findings may be one reason that UNa is not yet included in the 2022 American Heart Association/American Stroke Association guidelines.

ESC recommendations are attractive because the traditional assessment of diuretic response and diuretic therapy guided by fluid‐based metrics involves a combination of symptoms and signs, such as weight loss and urine volume output, to evaluate treatment. This response has shown to be inaccurate or logistically challenging to obtain and may, in part, delay the diuretic titration because these reevaluations are usually possible after 24 hours. ⁵ According to observational studies, ⁴ the early UNa‐guided assessment of the diuretic response in the first hours, if protocolized and correctly interpreted, may help to tailor the intensity of diuretic therapy. ⁶ Indeed, a recent systematic review and meta‐analysis concluded that high UNa after diuretic administration is associated with higher urinary output, greater weight loss, shorter length of stay, and lower odds of death. ⁴ However, there are several differences between studies, mainly including different time points and cutoffs for defining low UNa (between <50 and <89), or different units of measurement. The timing of UNa assessment has not been homogeneous between studies, varying most of them between 2 hours to 6 hours post‐diuretic treatment ⁴ , ⁶ and even in others measuring baseline values immediately after entering an emergency department before and not after diuretic infusion.

Recently, Martens et al. ⁷ reported in a post hoc analysis of the ROSE‐AHF (Renal Optimization Strategies Evaluation in Acute Heart Failure) trial that in patients with very early evaluation (before intravenous diuretic administration) a low spot sample UNa (range 19–40 mmol/L) was associated with lower 72‐hour natriuresis, weight loss, and diuretic response. ⁷ An admission spot urinary sodium at emergency might have pragmatic logistic advantages over 2‐hour post‐diuretic UNa for risk stratification. ⁸ However, a 2‐hour UNa assessment would capture the effect of early intervention (diuretic administration). Possibly, early changes in UNa from a baseline status to the first hours after diuretic administration may provide additional information rather than only 1 measurement. ⁹

A very recent study endorses the feasibility and efficacy of UNa‐guided diuretic therapy. ¹⁰ The ENACT HF (Efficacy of a Standardized Diuretic Protocol in Acute Heart Failure Study), a multicenter, open‐label, non‐randomized trial that compared the efficacy of 2 diuretic strategies in 2 chronological cohorts: standard of care versus UNa‐guided strategy in acute HF according to ESC consensus. ³ The investigators enrolled 401 patients without meaningful differences in baseline characteristics except for mineralocorticoid receptor antagonist and sodium‐glucose transport protein 2 inhibitors treatment, which were superior in the protocol group (59.2% versus 47.2%, P=0.021 and 25.2% versus 12.2%, P=0.001). The authors reported an increased natriuresis after day +1, noting that natriuresis was a 64% higher in the UNa‐guided group −282 mmol compared with the standard of care group (174 mmol)(mean ratio 1.64 [95% CI, 1.37–1.95], P<0.001). Secondary end points included diuresis (urine output) after day +2 with 5.8 L in the protocol group versus 4.4 L (mean ratio 1.33 [95% CI, 1.21–1.47], P<0.001). Interestingly, the UNa‐guided strategy group had a lower duration of hospitalization 5.8 days versus 7 days (mean ratio 0.87 [95% CI, 0.77–0.99], P<0.036) (Dauw J. ENACT‐HF: Efficacy of a Standardized Diuretic Protcol in Acute Heart Failure. Heart Failure; May, 20–23, 2023; Prague‐Czechia. 5th August 2023). Currently, one ongoing randomized controlled trial, PUSH‐AHF (Pragmatic Urinary Sodium‐Based Treatment Algorithm in Acute Heart Failure) will provide more definitive results. ¹¹ The PUSH‐AHF is evaluating whether natriuresis‐guided therapy using a prespecified stepwise diuretic treatment approach will improve natriuresis and clinical outcomes in patients with AHF. ¹¹

From a logistical point of view, using a spot baseline and a 2‐hour UNa post‐diuretic assessment instead of a 24‐hour UNa or a partial sample collection has several advantages: (1) early accurate prediction of poor natriuretic response (2) easy to implement in daily clinical practice, and (3) yields a more dynamic assessment, useful for short‐term monitoring, guiding therapy, and early recognition of complications.

In this article, we discuss some limitations, flaws, and aspects not previously discussed in the ESC guidelines that we consider critical for the correct clinical use and interpretation of UNa analysis as a biomarker of diuretic response in AHF.

1. The current recommendation did not account for the severity or distribution of fluid overload. We speculate that the current recommendation applies to patients with overt fluid overload but probably lacks usefulness in AHF and predominant fluid redistribution, a situation in which an aggressive diuretic strategy may be harmful. Further studies should focus on the influence of fluid overload status at baseline.

2. UNa as a biomarker loses strength and usefulness after day 1, as high urine output classically turns into hypotonic urine in patients with HF and as the overall urinary sodium excretion tends to decrease rapidly with decongestion. ⁶ Studies suggest that in AHF, UNa decreases after 24 hours for diverse reasons. The diuretic response, natriuresis, might be more effective in the first hours and after an initially favorable response clinicians might be tempted to decrease the diuretic dosage, thereby limiting natriuresis after 24 hours. ⁶ The neurohormonal activation and the enhanced tubular sodium reabsorption in other parts of the nephron, other than the loop of Henle, “braking phenomenon” after day 1 may play a role. ¹² This means that the high levels of natriuresis after a first dose of diuretics decline over time as a result of both hemodynamic changes at the glomerulus as well as adaptive changes in the distal nephron. Therefore, consecutive doses of diuretics are less effective. Additionally, a dose of loop diuretic increases urinary excretion of sodium chloride for several hours (influenced by its half‐life), but this is followed by a period of very low sodium excretion termed “post‐diuretic sodium retention.” ¹² Finally, after 24 hours from initiation of treatment, large hypotonic volumes of urine generated contain more free water and proportionally diluted sodium concentration, which is lower when evaluated in a spot urine sample. This might alter measurements of spot UNa concentrations and to a lesser degree in sample urine collections. This overproduction of free water urine with low electrolytes can be highly variable according to their kidney intrinsic response or resistance to diuretics and therefore can be earlier or later in different patients with AHF after diuretic infusion.

3. The current recommendation on UNa does not account for kidney function: Normal kidney function (glomerular and tubular) is crucial for the kidney management of sodium and its reabsorption. At a normal glomerular filtration rate of 180 L/day filtered with plasma water sodium of 150 mEq/L, the filtered sodium load is 27 000 mEq/day. UNa measured by spot samples can be affected by glomerular filtration rate, by the presence or absence of chronic kidney disease (CKD), salt intake, intravenous sodium infusion as hypertonic saline solution with furosemide or bicarbonate, and importantly by loop diuretics administered before admission. Concomitant acute kidney injury, particularly acute tubular necrosis, the leading cause of intrinsic renal acute kidney injury in hospitalized patients, is a common complication of admitted patients with AHF. It alters UNa measurements, leading to characteristically high UNa levels of >50 mEq/L, and patients usually develop anuria, which would be contrary to the ESC recommended approach. CKD is particularly relevant because it affects up to 40% to 70% of individuals with symptomatic HF. ¹³ The classically described inability of patients with CKD to reabsorb sodium salt wasting state occurs because of osmotic diuresis induced by increased urea excretion per nephron related in part to the decreased number of functioning nephrons, because of the tubular injury due to the underlying disease, and because of the inability to acutely activate the natriuretic mechanisms. ¹⁴ Proximal tubule reabsorption dysfunction is common in patients with CKD and HF. ¹⁵ Patients with advanced kidney disease and estimated glomerular filtration rate below 25 mL/min per 1.73 m² may have a quadruple obligatory sodium excretion when compared with a normal subject (10–40 mEq/day versus 5 mEq/day), ¹⁴ making patients with CKD classically less responsive to diuretics and confounding the diuretic titration guided by UNa. These important features, the presence or not of kidney disease, must be taken into account when interpreting UNa in patients with AHF. When the kidneys that, despite intrinsic disease, namely CKD, or diuretics before admission for AHF continue to avidly reabsorb sodium after loop diuretics administration indicates that they still have some preserved tubular function. This highlights the need to study early UNa's utility across CKD categories. Fractional excretion of sodium may be more accurate than spot UNa concentration and short time frame sodium excretion by sample collections because it directly measures sodium management by kidneys with no influence in the amount of urinary volume, unlike UNa concentration, which is affected by the rate of free water reabsorption. For example, large urinary volume output falsely dilutes sodium results if measured by spot urinalysis and not the whole urine sample. A patient who is water but not sodium avid with low urine volume may have high urine sodium concentration despite having little sodium in the urine. ¹⁶ Currently, new natriuretic equations are being developed and validated to predict loop diuretic response in patients with heart failure that may address this issue in part with spot urine measures. ¹⁷ Thus, we speculate that fractional excretion of sodium and new natriuretic equations may be less biased alternatives to UNa measured in spot sample urine (mmol/L or meq/L) as gold standard measurement of sodium 24‐hour urine is not practical for early titration of diuretic dose.

4. UNa and its relationship with other surrogates of decongestion: UNa trajectory requires parallel evaluation of signs of decongestion as urinary output or others (eg, weight loss). For example, a decreasing slope of UNa in a patient with adequate diuretic urine output and decongestion could be a favorable response to diuretics inviting the reduction of the diuretic dose. In contrast, a decreasing level of UNa in a patient with persistent low urine output and hydric overload may imply diuretic resistance. Therefore, a hourly urine output must prevail over UNa measures. Unfortunately, also, diuretic titration to urine output has minimal trial evidence and clear limitations, yet substantial clinical experience still exists.

5. The type of diuretics. Traditional diuretic enhances natriuretic response; however, data on the role of UNa for monitoring response to aquaretics (eg, tolvaptan) or sodium‐glucose transport protein 2 inhibitors are scarce. ¹⁸ Theoretically, in this latter setting, UNa will lose clinical usefulness. The osmotic diuresis due to glycosuria rather than natriuresis may also lead to lower spot urinary sodium concentration despite increased urinary output. Because sodium‐glucose transport protein 2 inhibitors have become key in AHF therapy, this interaction is important when using urine sodium‐based diuretic titration guidance. Subjects who have been treated previously with furosemide have a lesser increase in fractional excretion of sodium after the administration of furosemide but have a greater natriuretic response to the addition of chlorothiazide, suggesting increased tubular sodium reabsorption at the thiazide‐sensitive site in the distal tubule when distal sodium delivery is chronically increased by furosemide. ¹⁸ Also, both oral and intravenous thiazide diuretics generate higher spot urine sodium as compared with tolvaptan. ¹⁸ Thus, further studies are warranted to evaluate the clinical usefullness of UNa with different “combo” diuretic approaches.

6. Other limitations: UNa is potentially influenced by the amount of sodium of the diet and chronic use of diuretics. ¹² Unfortunately, accurate data about the magnitude of this influence are lacking in patients with HF. In patients with HF, other concomitant disorders characterized by reduced effective arterial blood perfusion, such as splanchnic vasodilatation and sequestration of fluid in the peritoneal cavity and arterial shunts (cirrhosis) or low plasma oncotic pressure (in some patients with nephrotic syndrome) can influence UNa. Cirrhotic, hepatorenal syndrome, with characteristic low UNa should be ruled out and excluded from this guided diuretic therapy because of its antagonistic diuretic management. ¹⁹

Overall, cumulative data support the promising role of urine sodium‐based diuretic titration in AHF, but high‐quality evidence and clinical experience remain limited. ²⁰ Prior issues should be clarified in further dedicated studies. We consider spot urinary sodium may be a useful measurement for monitoring and guiding diuretic therapy in patients with true fluid overload. However, further studies in a broad spectrum of patients and clinical situations are required to understand the pros and cons of this parameter. Before we have more evidence, we advocate for using early hospitalization of UNa for monitoring and guiding diuretic approaches as 1 more piece of the puzzle, as stated in the Figure. In addition, we should be aware of some clinical conditions (not overt fluid overload, acute kidney injury, cirrhosis, multiple diuretic therapy) in which the interpretation of urinary sodium excretion may be challenging.

Figure 1. Guided diuretic therapy according to early post‐diuretic spot urinary sodium in acute heart failure.

In AHF, the kidneys shows higher avidity of sodium reabsorption, and consequently low urinary sodium is frequently present and associated with more severe disease. Excluding cirrhotic, in fluid overload patients with AHF, a spot UNa <50 mEq/L at 2 hours after loop diuretic infusion is associated with worse loop diuretic response (diuretic resistance) using different metrics. Expert consensus recommends early intensification of diuretic therapy in case of low UNa (<50–70 mEq/L) at 2 hours after the first intravenous loop diuretic administration. We propose a comprehensive interpretation of early UNa after diuretic treatment accounting for other parameters of decongestion, especially in specific cases in which the evidence is scarce and pathophysiology of UNa complex, such as ambiguity about volume overload status, AKI, severe CKD, cirrhotic patients, and use of aquaretics. Equally as ESC guidelines declare and because most studies are observational and use different cutoffs from 50 to 89 mmol, the cutoff used in this figure is based on expert opinion and based in recent released results of the ENACT‐HF study. AHF indicates acute heart failure; AKI, acute kidney injury; CKD, chronic kidney disease; ENACT HF, Efficacy of a Standardized Diuretic Protocol in Acute Heart Failure Study; ESC, European Society of Cardiology; RAAS, renin‐angiotensin‐aldosterone system; and Una, urinary sodium.

Sources of Funding

Maria Jose Soler has funding from Investigation SANITARIA‐FEDER, Instituto de Salud Carlos III (ISCIII), PI21/01292. Julio Nuñez has funding from Centro de Investigación Biomédica en Red Enfermedades Cardiovasculares (CIBER Cardiovascular) (16/11/00420).

Disclosures

Julio Nuñez reports personal fees or advisory boards from Alleviant, AstraZeneca, Boehringer Ingelheim, Bayer, Novartis, NovoNordisk, Rovi, and Vifor Pharma (outside the submitted work). Maria Jose Soler reports honorarium for conferences, consulting fees and advisory boards from Astra Zeneca, NovoNordsik, Esteve, Vifor, Bayer, Mundipharma, Ingelheim Lilly, Jansen, ICU Medical, Fresenius, Travere Therapeutics, and Boehringer, and is ex‐Editor in Chief of Clinical Kidney Journal. Nestor Oliva‐Damaso has no disclosures to report.