A Step Forward Toward Precision Diuretic Management in Patients with Acute Decompensated Heart Failure and Diuretic Resistance

Intravenous (IV) loop diuretics are first-line therapy for decongestion in patients with acute decompensated heart failure (ADHF); however, patients with severe heart failure and/or kidney dysfunction frequently develop resistance to loop diuretics, requiring escalation of loop diuretic dosing and/or addition of adjuvant diuretics. Although there is no widely accepted definition for diuretic resistance, it is estimated that approximately 20%–50% of patients hospitalized for ADHF have an inadequate response to initial loop diuretic dosing and have some degree of diuretic resistance.¹ There are several proposed mechanisms by which patients may develop resistance to loop diuretics, including impaired diuretic delivery to the tubular lumen, increased proximal tubule sodium reabsorption, reduced efficacy of sodium-potassium-chloride cotransporter 2 (NKCC2) inhibition, and increased reabsorption of sodium in the distal tubule.^1–3 To combat diuretic resistance, adjunct diuretics can be used to reduce the compensatory sodium reabsorption in the proximal and distal tubules and enhance natriuresis. Several clinical trials have shown that thiazide diuretics (Safety and Efficacy of the Combination of Loop With Thiazide-type Diuretics in Patients With Decompensated Heart Failure),⁴ sodium-glucose cotransporter 2 inhibitors (SGLT2is; Empagliflozin in Heart Failure, EMPagliflozin in Patients Hospitalized for Acute Heart Failure, and Cox et al.),^5–7 and carbonic anhydrase inhibitors (Acetazolamide in Decompensated Heart Failure with Volume OveRload)⁸ are all effective adjuvant diuretics enhancing natriuresis in comparison with placebo control in patients with ADHF. By contrast, although mineralocorticoid antagonists are a pillar of guideline-directed medical therapy in patients with heart failure with reduced ejection fraction, high-dose spironolactone has not been shown to be an effective adjuvant diuretic in patients with ADHF (ATHENA-HF).⁹ However, there have been no randomized trials, until now, comparing the efficacy of an adjunct thiazide versus escalation of loop diuretic dosing, which are both pragmatic approaches used frequently in clinical practice. Furthermore, few studies have tested diuretic strategies on the basis of a patient's mechanism of diuretic resistance (i.e., loop versus distal nephron–mediated diuretic resistance).

A recent mechanistic randomized clinical trial published by Rao et al. in this issue of JASN was designed to address these important questions.¹⁰ The authors studied patients hospitalized for ADHF with diuretic resistance defined as a 6-hour cumulative urine sodium output <100 mmol to an IV loop diuretic dose chosen by the clinical team. The predominant mechanism of diuretic resistance in each patient was assessed on the basis of changes in the fractional excretion of lithium (FELi) and fractional excretion of sodium (FENa) with loop diuretic administration and determined to be either due to impaired response at the loop of Henle (DR-Loop) or due to a compensatory increase in distal sodium reabsorption (DR-Distal). Patients with DR-Loop were those who had a small increase in FELi and FENa after loop diuretic administration, indicating only a minimal decrease in sodium reabsorption in the proximal tubule and loop of Henle with loop diuretic administration. By contrast, patients with DR-Distal were those who had a large increase FELi but a small increase in FENa after loop diuretic administration, indicating significant reduction of sodium reabsorption in the proximal tubule and loop of Henle, but a compensatory increase in distal sodium reabsorption. Patients were then randomly assigned to receive either a single 2.5-fold increased dose of loop diuretic versus IV chlorothiazide (500 mg for patients with eGFR >20 ml/min per 1.73 m² and 1000 mg for patients with eGFR <20 ml/min per 1.73 m²). The primary outcome was change in 6-hour natriuresis. The authors hypothesized that patients with DR-Loop would have a greater increase in natriuresis with an increased dose of loop diuretic and that patients with DR-Distal would have a greater increase in natriuresis with the addition of a thiazide diuretic. Surprisingly, the authors found that patients with either mechanism of diuretic resistance (DR-Loop or DR-Distal) who received IV chlorothiazide, in comparison with an increased loop diuretic dose, had increased 6-hour natriuresis (107 mmol increase [95% confidence interval, 81 to 132] with IV chlorothiazide versus 49 mmol increase [95% confidence interval, 31 to 66] with loop diuretic intensification; P < 0.001). Furthermore, patients with DR-Loop, in comparison to patients with DR-Distal, exhibited equivalent increases in natriuresis in response to loop diuretic intensification. Proposed explanations for this unexpected finding include that patients with DR-Loop, in comparison with patients with DR-Distal, exhibit decreased baseline loop of Henle sodium reabsorption before diuretic receipt and thereby may have reduced expression of NKCC2, and that patients with DR-Loop may express a splice variant of NKCC2 that is less responsive to loop diuretics. The authors did find, however, that only patients with a high degree of baseline prediuretic distal sodium reabsorption exhibited a prolonged natriuretic effect with chlorothiazide in the subsequent 18 hours after the first 6 hours post thiazide administration. The authors propose that the prolonged natriuretic effect of chlorothiazide may be due to blockade of pendrin via inhibition of carbonic anhydrase, which regulates pendrin expression.

Although the main results of the study were different than expected, this study nevertheless identifies a novel diuretic strategy for patients with diuretic resistance, showing that addition of a thiazide diuretic results in superior short-term natriuresis in comparison with escalation of loop diuretics, independent of diuretic resistance mechanism. These findings support combination diuretic therapy in patients with ADHF with diuretic resistance, an important finding with immediate clinical translation.

Also importantly, this study uses an innovative design and precision medicine–based approach to attempt to identify personalized diuretic management strategies on the basis of the predominant mechanism of diuretic resistance. Although independent of loop diuretic resistance mechanism, the authors report that only patients with baseline prediuretic high distal sodium reabsorption experienced a prolonged natriuresis with chlorothiazide. Addition of thiazide diuretics in this subset of patients could reduce postdiuretic compensatory sodium reabsorption and result in more effective natriuresis. Thus far, in clinical practice selection of adjunct diuretics for patients with ADHF and diuretic resistance, it has largely been governed by medication side effect profiles and limiting negative effects of adjunct diuretics including serum electrolyte derangements (i.e., limiting hyponatremia, metabolic acidosis, hypokalemia, etc.). This study represents a significant step forward toward identifying personalized diuretic strategies that will preferentially benefit certain subsets of patients and should serve as an example for the design of future studies.

We applaud the authors for this innovative and ambitious clinical trial. However, we do recognize some limitations. Mainly, this study was performed over one 24-hour period with a one-time administration of an IV loop±thiazide diuretic with the primary outcome being 6-hour natriuresis. As emphasized by the findings from this study, renal sodium handling and mechanisms of diuretic resistance are complex. As such, renal sodium handling is likely dynamic as well and will change with decongestion, administration of multiple diuretic doses, and administration of other medications that affect glomerular hemodynamics and serum electrolyte concentrations. Subsequent longitudinal studies are needed to assess the temporal changes in renal sodium handling with decongestion and administration of multiple doses of different diuretics over days, as occurs in clinical practice. In addition, as stated by the authors, differences in FELi and FENa in response to loop diuretics can only identify differences in sodium reabsorption/excretion in the proximal tubule and loop of Henle combined versus the distal tubule. Proximal tubule sodium reabsorption was assumed to be constant throughout the study and no patients in the study were prescribed a SGLT2i. Assays to stratify proximal tubule versus loop of Henle sodium excretion/reabsorption would be useful and subsequent studies with patients receiving SGLT2is are warranted. Finally, although the authors propose possible molecular mechanisms to explain the clinical natriuretic findings, these mechanisms are hypothesis-generating and additional studies are needed to confirm these hypotheses.

In conclusion, this study by Rao et al. identifies a pragmatic diuretic strategy, addition of a thiazide to loop diuretics, that could immediately affect patients with ADHF by enhancing their short-term rate of natriuresis. The authors also use an innovative precision medicine–based approach to attempt to identify subsets of patients that may benefit from differing diuretic strategies. As we enter a new phase of precision medicine in diuretic management, standardized assays with identified cutoff values will need to be developed to allow for widespread implementation of strategies to identify different subsets of patients with different patterns of renal sodium handling. Clinical trials with longer follow-up will need to be performed to identify changes in renal sodium handling over time and will inform how frequently a patient's diuretic management should be reassessed. Whether important clinical outcomes including length of stay, time to euvolemia, readmission rate, and other kidney/cardiac outcomes are superior with a precision-diuretic approach requires additional investigation. Ultimately, the era of one-size-fits-all diuretic therapy is finally at a crossroads. Precision medicine approaches in the design of diuretic clinical trials can provide clinicians with the framework to choose the right diuretic strategy, for the right patient, at the right moment.

Supplementary Material

jasn-37-436-s001.pdf ^{(1.7MB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/F630.

Author Contributions

Conceptualization: Nisha Bansal, Sarah Haeger.

Supervision: Nisha Bansal.

Writing – original draft: Nisha Bansal, Sarah Haeger.

Writing – review & editing: Nisha Bansal, Sarah Haeger.

Funding

S Haeger: National Institute of Diabetes and Digestive and Kidney Diseases. N. Bansal: NIDDK.