Library › Pharmacology › Diuretics

Diuretics in heart failure and hypertension

Loop, thiazide-type, potassium-sparing, and carbonic anhydrase inhibitor diuretics in adult cardiovascular practice, from nephron pharmacology and dose conversion through diuretic resistance, sequential nephron blockade, hypertension, electrolyte complications, and special populations.

Diuretic classes and their place in cardiovascular practice

A diuretic increases urinary excretion of sodium and water. In cardiovascular medicine, diuretics serve two different purposes. In heart failure and other edematous states they remove sodium that the kidney has retained, relieving congestion; the dose is titrated to a clinical end point, the disappearance of signs and symptoms of volume overload. In hypertension, a low, fixed dose of a thiazide-type diuretic lowers blood pressure through a small initial reduction in plasma volume followed by a sustained fall in peripheral vascular resistance. The end point is the blood pressure goal. The classes differ in where they act along the nephron, how much natriuresis they can produce, and which electrolytes they disturb.

Diuretic classes by nephron segment. Natriuretic capacity is the approximate maximal fraction of filtered sodium excreted when the class is given alone. ENaC, epithelial sodium channel; NHE3, sodium-hydrogen exchanger 3.

Sodium handling along the nephron and sites of diuretic action

The glomeruli filter about 180 liters of plasma and some 25,000 mEq of sodium each day, and more than 99% of that sodium is reabsorbed. The proximal tubule reclaims about 60% to 70%, the thick ascending limb about 20% to 30%, the distal convoluted tubule about 5% to 10%, and the collecting duct about 2% to 5%. Two consequences follow. First, a drug acting on a segment that handles a large share of filtered sodium has a high natriuretic ceiling, which is why loop diuretics are the decongestive agents. Second, any segment downstream of the blocked one can increase its own reabsorption, so the natriuresis that reaches the urine is always smaller than the amount of sodium released at the site of action. Acetazolamide releases a great deal of sodium in the proximal tubule but produces little natriuresis alone, because the thick ascending limb reclaims most of it. This downstream compensation is also the basis of sequential nephron blockade.

Proximal tubule. Sodium enters the proximal tubular cell mainly through the sodium-hydrogen exchanger NHE3, which depends on hydrogen ions generated by carbonic anhydrase, and through sodium-coupled transporters for glucose (SGLT2), amino acids, and phosphate. Angiotensin II and renal sympathetic activity stimulate NHE3, so proximal reabsorption rises in heart failure and reduces the sodium delivered to the loop. Acetazolamide and SGLT2 inhibitors act here.

Thick ascending limb. NKCC2 moves one sodium, one potassium, and two chloride ions into the cell. Potassium recycles back into the lumen through the renal outer medullary potassium (ROMK) channel, creating a lumen-positive voltage that drives paracellular reabsorption of calcium and magnesium. Loop diuretics block NKCC2 from the luminal side, abolish the medullary concentration gradient, and remove the voltage that drives calcium and magnesium reabsorption; hypercalciuria and hypomagnesemia are therefore class effects. The macula densa senses luminal chloride through NKCC2, so loop diuretics also stimulate renin release.

Distal convoluted tubule. NCC reabsorbs sodium and chloride in a segment that is impermeable to water and dilutes the urine. Thiazide-type diuretics block NCC, which impairs urinary dilution (the mechanism of thiazide-induced hyponatremia) and increases calcium reabsorption. The distal tubule is the segment that hypertrophies after long-term loop diuretic exposure.

Collecting duct. In principal cells, aldosterone acting through the mineralocorticoid receptor increases ENaC activity, and sodium entry through ENaC generates the lumen-negative voltage that drives potassium secretion through ROMK and hydrogen ion secretion by intercalated cells. Any diuretic that increases sodium delivery to this segment (loop and thiazide-type diuretics) increases potassium and hydrogen loss, and any drug that blocks ENaC or the mineralocorticoid receptor retains potassium. Vasopressin acting on V2 receptors inserts aquaporin-2 water channels; tolvaptan blocks this step.

How a loop diuretic reaches its target: absorption, protein binding, and tubular secretion

Loop diuretics act from the tubular lumen, not from the blood. Each dose must pass through four steps: absorption from the gut (for oral doses), delivery to the kidney, secretion into the proximal tubular lumen, and binding to NKCC2 in the thick ascending limb. Each step can fail, and diuretic resistance is often a failure of delivery.

Absorption. Furosemide has variable oral bioavailability, from about 10% to 90% with a mean of about 50%, and its absorption is reduced by food. Bumetanide and torsemide are absorbed more completely and more predictably, with bioavailability of 80% to 100%. In decompensated heart failure, bowel wall edema and reduced splanchnic perfusion slow absorption and blunt the peak concentration, so an oral dose that kept a patient euvolemic can fail during decompensation. A brisk response to the same drug given intravenously after an inadequate oral response identifies an absorption problem.

Protein binding and secretion. All loop diuretics are more than 90% bound to albumin, so very little is filtered at the glomerulus. Albumin keeps the drug in the vascular space and carries it to the proximal tubule, where organic anion transporters (OAT1 and OAT3) on the basolateral membrane secrete it into the lumen. The amount of drug secreted, not the plasma concentration, determines the effect. Delivery falls when renal blood flow is low, when hypoalbuminemia enlarges the volume of distribution, and when drugs such as probenecid, or organic anions that accumulate in kidney dysfunction, compete for the transporters. In nephrotic syndrome, filtered albumin binds furosemide within the tubular lumen and further reduces free drug at NKCC2.

Elimination. About half of a dose of furosemide is excreted unchanged in the urine and the remainder is conjugated to a glucuronide in the kidney, so its half-life lengthens in kidney failure. Torsemide is cleared mainly by hepatic metabolism through CYP2C9 (about 80% of clearance), so its half-life is unchanged by kidney failure and roughly doubles in hepatic dysfunction. Bumetanide is cleared by both routes.

Loop diuretics: mechanism, threshold, and ceiling

The natriuretic response to a loop diuretic follows a sigmoid curve when plotted against the logarithm of the amount of drug delivered to the tubule. Below a threshold, no natriuresis occurs. Above it, natriuresis rises steeply to a ceiling, the dose beyond which further increases add little. Three clinical rules follow from this shape.

First, an ineffective dose is not repeated; it is increased, usually doubled, because the curve is logarithmic and small increments do not move the patient along it. Second, once the effective single dose has been found, further natriuresis is obtained by giving that dose more often, because each dose produces natriuresis only while the tubular concentration exceeds the threshold. Third, the threshold and the ceiling move to the right in disease. In chronic kidney disease less drug reaches the lumen, so a larger dose is needed to reach the threshold. In heart failure delivery is near normal, but the tubule responds less, so the maximal natriuresis is lower. Doses above the nominal ceiling can still add natriuresis because they prolong the time during which the tubular concentration remains above the threshold.

Duration and postdiuretic sodium retention. Natriuresis after an oral dose of furosemide lasts about 6 to 8 hours, and after an intravenous dose about 2 hours. During the remaining hours of the day, the kidney reabsorbs sodium avidly, driven by the renin-angiotensin-aldosterone system, sympathetic activity, and the distal nephron. With unrestricted sodium intake, this postdiuretic retention can cancel the natriuresis of a single daily dose of a short-acting loop diuretic, leaving the patient in neutral sodium balance despite substantial urine output after each dose. Twice-daily or three-times-daily dosing, a continuous infusion, a longer-acting agent, and dietary sodium moderation each reduce the retention window.

The braking phenomenon. Even after the first dose, the natriuretic response to each subsequent dose declines. This short-term tolerance, called braking, reflects volume contraction, neurohormonal activation, and increased distal reabsorption. It is present in healthy subjects, is accentuated in heart failure, and is offset when sodium losses are replaced. Over weeks, a second form of tolerance develops: chronic loop diuretic exposure increases sodium delivery to the distal convoluted tubule, which hypertrophies and increases its NCC transport capacity, as first shown in rats given continuous furosemide. The hypertrophied distal tubule reclaims sodium released from the loop and is the target of thiazide-type add-on therapy.

Effects beyond natriuresis. Intravenous furosemide produces venodilation within minutes, before natriuresis begins, which contributes to early relief of pulmonary congestion; the effect depends on prostaglandins and is blunted by nonsteroidal anti-inflammatory drugs (NSAIDs).

Pharmacokinetics of furosemide, bumetanide, torsemide, and ethacrynic acid

Loop diuretic pharmacology. Initial and maximum doses for furosemide, bumetanide, and torsemide, and durations of action, are from the 2022 AHA/ACC/HFSA guideline. Ethacrynic acid doses are from the prescribing information. Bioavailability, half-life, and clearance are from the 2020 Journal of the American College of Cardiology (JACC) state-of-the-art review and the 1998 and 2015 reviews of diuretic therapy.

Furosemide is the most widely used loop diuretic and the only one available as oral, intravenous, and subcutaneous preparations in the United States. Its erratic absorption is the main reason an oral regimen that seems adequate can fail, and the main reason the intravenous dose is about half the oral dose.

Bumetanide is about 40 times as potent as furosemide by weight and is well absorbed. It is short acting and is usually given twice daily. Its oral and intravenous doses are similar because absorption is nearly complete.

Torsemide combines reliable absorption with the longest duration of action of the loop diuretics, which allows once-daily dosing in many patients and shortens the window for postdiuretic sodium retention. Experimental studies suggested an additional antialdosterone action and an antifibrotic effect on myocardium; in a small open-label pilot study, torsemide reduced markers of collagen type I synthesis and myocardial collagen volume fraction more than furosemide. These effects have not translated into a clinical outcome benefit in TRANSFORM-HF, described below.

Ethacrynic acid is the only loop diuretic without a sulfonamide group. It is more ototoxic than the others and costs more, and its use is limited to patients with a documented severe reaction to sulfonamide-containing diuretics. The furosemide injection label advises against combining it with ethacrynic acid because of additive ototoxicity.

Loop diuretic dose equivalence and conversion

Conversion among loop diuretics rests on the approximate equivalence of oral furosemide 40 mg, intravenous furosemide 20 mg, oral torsemide 20 mg, and bumetanide 1 mg. The ratios come from small pharmacodynamic studies and summarize two properties at once: intrinsic potency and oral bioavailability. They are starting points. Because furosemide bioavailability ranges from about 10% to 90%, the oral dose that matches a given intravenous dose varies several-fold among patients, and published torsemide equivalents for 40 mg of oral furosemide range from 10 to 20 mg.

Approximate loop diuretic equivalence by total daily dose. Torsemide equivalents use the 1:2 ratio (20 mg per 40 mg of oral furosemide) except where the published 10–20 mg range is shown. Ethacrynic acid is about 0.8 times as potent as furosemide and is supplied as 25-mg tablets. Above the maximum labeled or guideline daily dose of an agent (furosemide 600 mg, torsemide 200 mg, bumetanide 10 mg), equivalence is not extrapolated; a second diuretic class is added.

Oral furosemide equivalents per milligram: oral furosemide 1, intravenous furosemide 2, oral torsemide 2, oral or intravenous bumetanide 40, oral ethacrynic acid 0.8. Doses are rounded to available steps (furosemide 10 mg, torsemide 5 mg, bumetanide 0.5 mg, ethacrynic acid 25 mg). Maximum daily doses: furosemide 600 mg, torsemide 200 mg, and bumetanide 10 mg (2022 AHA/ACC/HFSA guideline); intravenous furosemide 400 to 600 mg per the 2019 Heart Failure Association (HFA) position statement. The calculator converts between maintenance regimens; the initial intravenous dose in acute heart failure is set differently, as described below.

Why the conversion is an estimate. A conversion from oral furosemide to torsemide or bumetanide often produces more diuresis than predicted, because the patient's true furosemide bioavailability may have been well below 50%. A conversion in the opposite direction can produce congestion. After any conversion in heart failure, weight and symptoms are reassessed within days, and electrolytes and creatinine within 1 to 2 weeks. When intravenous furosemide is converted to an oral regimen before discharge, the effective oral dose is usually lower than the arithmetic equivalent of the intravenous dose that decongested the patient. The intravenous dose was chosen to remove excess fluid, and guideline-directed medical therapy (GDMT) reduces diuretic requirements over the following weeks.

Ceiling doses by clinical condition

The ceiling dose is the single dose that produces the maximal fractional sodium excretion in a given condition. Repeating a ceiling dose is more effective than giving one larger dose, and doses well beyond the ceiling mainly add toxicity.

Ceiling doses from the 2015 review of loop diuretics in clinical practice, which follows the 1998 review of diuretic therapy, and the maximal daily intravenous dose from the 2019 HFA position statement. Equivalent doses of bumetanide and torsemide follow the ratios in the conversion table.

Torsemide versus furosemide: from TORIC to TRANSFORM-HF

The pharmacologic advantages of torsemide, complete absorption and longer action, led to the hypothesis that it would improve outcomes. Early studies were small, open label, or nonrandomized, and their results were favorable. The pragmatic TRANSFORM-HF trial tested the hypothesis in a randomized comparison and found no difference in mortality.

Comparative trials of torsemide and furosemide in heart failure. HR, hazard ratio; CI, confidence interval.

TRANSFORM-HF tested a strategy of prescribing one or the other drug at discharge, with dosing left to clinicians, and it was designed to detect a 20% reduction in mortality. In a post hoc analysis of 2,421 participants, the 21% at high risk of diuretic resistance by the BAN-ADHF score had higher mortality, but risk status did not modify the effect of torsemide compared with furosemide. The choice of loop diuretic is therefore individualized. Torsemide or bumetanide is used when oral furosemide absorption is unreliable, as in right heart failure with gut edema, or when once-daily dosing helps adherence. Furosemide is used when cost, availability, or established response favors it. Statements that torsemide improves survival rest on the open-label and nonrandomized studies that preceded TRANSFORM-HF.

Intravenous loop diuretics in acute heart failure: summary of hospital decongestion

The protocol for decongestion in hospital, including timing of the first dose, response monitoring, escalation, and the transition to oral therapy, is covered in detail on the acute decompensated heart failure page. Patients admitted with significant fluid overload receive intravenous loop diuretics promptly (Class 1), and diuretics and GDMT are titrated to resolve clinical congestion (Class 1).

Summary of intravenous loop diuretic therapy in acute heart failure, from the 2022 AHA/ACC/HFSA guideline, the 2024 ACC expert consensus decision pathway, the 2019 HFA position statement, and the DOSE and ADVOR protocols. The full stepped protocol and an intravenous dosing and response calculator are on the acute decompensated heart failure page.

DOSE. The only large randomized trial of loop diuretic dosing in acute heart failure assigned 308 patients in a 2-by-2 factorial design. One comparison was bolus every 12 hours with continuous infusion. The other was a low dose (equal to the home oral dose) with a high dose (2.5 times the home oral dose). Neither comparison changed the coprimary end points: global symptom assessment over 72 hours (bolus 4,236 and infusion 4,373, P = 0.47; high dose 4,430 and low dose 4,171, P = 0.06) and change in creatinine (high dose 0.08 and low dose 0.04 mg/dL, P = 0.21). The high-dose strategy produced more net fluid loss and weight loss. It did not worsen the coprimary kidney end point, the change in creatinine at 72 hours. A creatinine rise of more than 0.3 mg/dL within 72 hours, a secondary end point, was more frequent with the high-dose strategy (23% versus 14%; P = 0.04), but creatinine and cystatin C did not differ at 60 days, and 60-day clinical outcomes were not worse. Bolus and continuous infusion did not differ significantly in symptoms or kidney function; the trial was not designed to test equivalence.

Natriuresis-guided therapy. A spot urine sample 1 or 2 hours after a loop dose predicts cumulative 6-hour sodium output. A predicted output below 50 mEq defines a poor natriuretic response, because twice-daily dosing at that level leaves the patient in positive sodium balance. In PUSH-AHF (310 patients), intensifying therapy whenever spot urine sodium was below 70 mEq/L increased 24-hour natriuresis (409 versus 345 mEq) without changing death or heart failure rehospitalization at 180 days (31% in both groups). In the nonrandomized ENACT-HF study (401 patients), a standardized natriuresis-guided protocol increased day 1 natriuresis (282 versus 174 mEq) and shortened hospital stay (5.8 versus 7.0 days). Natriuresis-guided escalation increases sodium excretion in the first days of treatment. A shorter hospital stay has been shown only in the nonrandomized ENACT-HF comparison, not in PUSH-AHF, and no effect on outcomes after discharge has been shown.

Bolus, infusion, subcutaneous, and outpatient parenteral delivery

Bolus and continuous infusion. In DOSE, continuous infusion showed no advantage over twice-daily boluses. An infusion keeps the tubular concentration above threshold and avoids the troughs of bolus dosing, but it reaches a natriuretic concentration slowly unless a loading bolus is given first. Infusion rates of 10 to 20 mg/h of furosemide, increased to as much as 40 mg/h, have been used after a loading dose of 40 to 200 mg selected by kidney function. The furosemide injection label states that for high-dose parenteral therapy a controlled infusion is advisable and that adult infusion rates not exceeding 4 mg/min have been used. Ototoxicity is associated with rapid injection, severe kidney impairment, doses above those recommended, hypoproteinemia, and concomitant aminoglycosides.

Subcutaneous furosemide. A buffered, neutral-pH furosemide solution delivered by a single-use on-body infusor provides 80 mg subcutaneously over 5 hours, 30 mg in the first hour and 12.5 mg/h for 4 hours. Its bioavailability relative to intravenous furosemide was 99.6% in a pharmacokinetic study, with equivalent diuresis. It was approved by the US Food and Drug Administration in 2022 for congestion in adults with chronic heart failure (NYHA class II–III) and in 2025 for edema in chronic kidney disease, including nephrotic syndrome. The label states that it is not for chronic use and that patients return to oral diuretics as soon as practical. It is contraindicated in anuria.

Outpatient intravenous therapy. Infusion clinics, heart failure day units, and hospital-at-home programs give intravenous loop diuretics to stable patients with worsening congestion, with same-day laboratory testing. They avoid admission for selected patients; randomized comparisons with hospitalization are lacking. A diuretic intensification in the clinic, whether intravenous or oral, marks worsening heart failure and carries an adverse prognosis.

Defining and measuring diuretic resistance

Diuretic resistance is an inadequate quantity of natriuresis despite an adequate diuretic regimen. Older definitions used persistent congestion despite a fixed dose, such as furosemide 80 mg or more daily, but a dose threshold ignores absorption, kidney function, and sodium intake, and no single quantitative definition is accepted. Contemporary studies measure the diuretic response itself.

Measures of diuretic response. Thresholds from the 2019 HFA position statement, the 2021 ESC guideline, and the 2016 validation study of the spot urine prediction equation.

Prediction before the first dose. The BAN-ADHF score combines blood urea nitrogen, creatinine, natriuretic peptide level, atrial fibrillation, diastolic blood pressure, hypertension and home diuretic dose, and prior heart failure hospitalization. It was derived from pooled trial cohorts (ROSE-AHF, CARRESS-HF, and ATHENA-HF; 794 patients) to identify low diuretic efficiency and validated in DOSE and ESCAPE. For patients with a high score, an early generous dose, early response measurement, and early combination therapy are planned.

Prognosis. Poor diuretic response is associated with longer stays, persistent congestion at discharge, rehospitalization, and death, independent of kidney function. Part of this association reflects the severity of the underlying heart failure, so improving the response does not necessarily change the prognosis, as PUSH-AHF and ADVOR illustrate.

Mechanisms of diuretic resistance

Resistance has several mechanisms that often coexist. Identifying the dominant one points to the countermeasure.

Mechanisms of loop diuretic resistance and the corresponding countermeasures, after the 2020 JACC state-of-the-art review, the 2017 review of diuretic treatment in heart failure, and the 2019 HFA position statement.

Neurohormonal activation. Each natriuresis lowers effective arterial volume and stimulates renin release, and loop diuretics stimulate renin at the macula densa. Angiotensin II increases proximal NHE3 activity, aldosterone increases distal ENaC activity, and renal sympathetic nerves do both and reduce renal blood flow. Renin-angiotensin inhibitors, MRAs, and beta-blockers counter these effects, which is one reason diuretic needs often fall as GDMT is established.

Hypochloremia. Chloride, more than sodium, is the ion sensed by the macula densa. In a cohort of 162 patients taking loop diuretics, 31.5% had a serum chloride of 96 mEq/L or less; hypochloremia was associated with higher renin and with poor diuretic response (odds ratio 7.3; 95% CI 3.3 to 16.1). In a pilot study of 10 patients, oral lysine chloride raised serum chloride and was followed by hemoconcentration and weight loss in most participants. Hypochloremic metabolic alkalosis therefore marks resistance and may sustain it. It is corrected with potassium chloride or acetazolamide.

Kidney congestion and perfusion. A high central venous pressure raises renal venous and interstitial pressure and lowers glomerular filtration, and in most congested patients the kidney improves as venous pressure falls. Low cardiac output with hypotension is a different and less common cause, identified by examination, lactate, and if needed invasive hemodynamics. The interpretation of a rising creatinine during decongestion is covered on the acute decompensated heart failure page.

Reversible causes: sodium intake, NSAIDs, adherence, and absorption

Before diuretics are escalated or combined, four reversible causes are excluded.

Dietary sodium. A patient eating large amounts of sodium can excrete a large natriuresis and still gain weight. A high spot or 24-hour urine sodium in a patient who is gaining weight points to intake, not resistance. How far sodium should be restricted is uncertain: in SODIUM-HF, 806 outpatients with NYHA class II–III heart failure were randomized to a diet with less than 1,500 mg (about 65 mEq) of sodium daily or usual care; the trial's name refers to 100 mmol, but the protocol defined the target as less than 1,500 mg. Median intake fell from 2,286 to 1,658 mg/day in the low-sodium group, and the composite of cardiovascular hospitalization, emergency visit, or death at 12 months did not differ (15% and 17%; hazard ratio 0.89; 95% CI 0.63 to 1.26). Avoidance of excess sodium is reasonable in symptomatic heart failure (2022 guideline, Class 2a); severe restriction has no proven benefit.

NSAIDs. NSAIDs inhibit renal prostaglandin synthesis, reduce renal blood flow and natriuresis, blunt the venodilator effect of intravenous furosemide, raise blood pressure, and with renin-angiotensin inhibitors and diuretics precipitate acute kidney injury. They are avoided in heart failure, and over-the-counter use is reviewed.

Adherence. Diuretics are among the drugs most often omitted, because of urinary urgency, incontinence, travel, or cost. Timing doses around daily activities and using a once-daily long-acting agent address nonadherence, and pharmacy refill data identify it.

Absorption. Gut edema, delayed gastric emptying, and food reduce furosemide absorption. Taking furosemide before meals, switching to torsemide or bumetanide, or using a period of intravenous or subcutaneous therapy addresses the problem.

Sequential nephron blockade: principles and choice of agent

Sequential nephron blockade combines diuretics that act at different nephron segments, so that sodium released at one site is not reclaimed at another. It is used when an adequate loop dose, given often enough, does not produce enough natriuresis. When diuresis in hospital is inadequate to relieve congestion, the 2022 AHA/ACC/HFSA guideline considers it reasonable to intensify the regimen with higher intravenous loop doses or a second diuretic (Class 2a). For outpatients and inpatients with congestive symptoms, the guideline reserves the addition of a thiazide such as metolazone for patients who do not respond to moderate- or high-dose loop diuretics, to minimize electrolyte abnormalities (Class 1).

Agents added to loop diuretics. 3T randomized 60 patients with intravenous loop diuretic resistance; CLOROTIC 230; ADVOR 519; DICTATE-AHF 240; DAPA-RESIST 61; ATHENA-HF 360. Intravenous loop diuretic therapy is covered on the acute decompensated heart failure page.

The choice among add-on agents rests on the dominant mechanism of resistance, the electrolyte and acid-base state, kidney function, route, and cost. A thiazide-type diuretic addresses distal hypertrophy and produces the largest increment in natriuresis but the greatest electrolyte losses. Acetazolamide addresses proximal reabsorption and corrects metabolic alkalosis. An SGLT2 inhibitor adds a modest natriuresis with the least electrolyte disturbance and is indicated on its own merits in most patients with heart failure. Three-segment blockade (acetazolamide, loop, and thiazide-type) is used in refractory congestion; published data on it are limited to small series.

Thiazide-type add-on therapy: metolazone, chlorothiazide, and hydrochlorothiazide

Metolazone is a quinazoline thiazide-like diuretic that is well absorbed and long acting (12 to 24 hours), and unlike hydrochlorothiazide it retains natriuretic activity when the glomerular filtration rate is low. It is the most widely used add-on in US practice. Doses start at 2.5 mg and rarely exceed 10 mg daily as an add-on; the guideline maximum is 20 mg. It is often given 30 to 60 minutes before the loop dose, a timing based on pharmacologic reasoning that has not been tested in trials. Because its effect persists, metolazone is given for a limited number of days in hospital or intermittently (for example, two or three times a week) in outpatients, with electrolytes and creatinine measured within days of each start or dose change. In an observational propensity-matched analysis cited by the 2022 guideline, adding metolazone in hospitalized patients was associated with more hypokalemia, hyponatremia, worsening kidney function, and mortality than an equivalent increase in loop dose. Confounding by severity limits interpretation, but the finding supports reserving the combination for true resistance.

Intravenous chlorothiazide acts within minutes and is used when oral absorption is uncertain. Its half-life is short (about 1.5 to 2 hours), so it is given once or twice daily. In the 3T trial, 60 patients with resistance to high-dose intravenous furosemide (mean 612 mg/day) received metolazone, intravenous chlorothiazide, or tolvaptan. All three improved diuresis. Weight loss at 48 hours was 4.6 kg with metolazone, 5.8 kg with chlorothiazide, and 4.1 kg with tolvaptan, without significant differences; serum sodium fell less with tolvaptan, but spot urine sodium was also lower. Intravenous chlorothiazide costs considerably more than oral metolazone.

Hydrochlorothiazide was tested in CLOROTIC, the largest placebo-controlled trial of thiazide add-on therapy. Among 230 patients with acute heart failure (mean age 83 years), hydrochlorothiazide added to intravenous furosemide increased weight loss at 72 hours (2.3 and 1.5 kg; P = 0.002), 24-hour diuresis (1,775 and 1,400 mL), and weight loss per 40 mg of furosemide. Dyspnea, mortality, and rehospitalization did not differ, and impaired kidney function occurred in 46.5% and 17.2%. CLOROTIC confirmed the natriuretic power of distal blockade and its cost in kidney function.

Acetazolamide: proximal blockade and the ADVOR trial

Acetazolamide inhibits carbonic anhydrase in the proximal tubule, which reduces bicarbonate, sodium, and water reabsorption through NHE3 and delivers more sodium to the loop of Henle. Alone it is a weak diuretic, because the loop and distal tubule reclaim the extra sodium, and its effect wanes within days as metabolic acidosis develops. With a loop diuretic it increases natriuresis, and it corrects the metabolic alkalosis that vigorous loop and thiazide diuresis produce.

In ADVOR, 519 patients with acute decompensated heart failure and clinical volume overload (edema, pleural effusion, or ascites) were enrolled if NT-proBNP was above 1,000 pg/mL or B-type natriuretic peptide was above 250 pg/mL. All received intravenous loop diuretics at twice the oral maintenance dose, with either intravenous acetazolamide 500 mg once daily or placebo. Successful decongestion within 3 days, without escalation of therapy, occurred in 42.2% and 30.5% (risk ratio 1.46; 95% CI 1.17 to 1.82). Cumulative urine output and natriuresis were higher and hospital stay was shorter with acetazolamide. Death or heart failure rehospitalization at 3 months did not differ (29.7% and 27.8%; hazard ratio 1.07; 95% CI 0.78 to 1.48), and worsening kidney function, hypokalemia, and hypotension were similar. Patients taking SGLT2 inhibitors were excluded, so the added effect of acetazolamide on top of an SGLT2 inhibitor is untested.

Practical use. Acetazolamide 500 mg intravenously once daily is given with the loop diuretic for the first days of decongestion, usually up to 3 days, with bicarbonate and potassium measured daily. Oral acetazolamide (250 to 500 mg) is used when intravenous drug is unavailable. Contraindications include severe hypokalemia or hyponatremia, hyperchloremic acidosis, marked hepatic disease (risk of hepatic encephalopathy), and adrenal insufficiency. Acetazolamide is a sulfonamide derivative, and the cross-reactivity considerations below apply. In hypercapnic patients with metabolic alkalosis that suppresses ventilatory drive, acetazolamide lowers bicarbonate and can improve ventilation.

SGLT2 inhibitors as diuretic adjuncts

SGLT2 inhibitors block glucose and sodium reabsorption in the proximal tubule, producing glycosuria with an osmotic diuresis and a natriuresis that is largest in the first days and attenuates thereafter. They reduce heart failure hospitalization and cardiovascular death across the range of ejection fraction, a benefit not explained by diuresis alone. Their natriuretic effect does not activate the renin-angiotensin system in the way loop diuretics do, and they cause little hypokalemia.

In EMPULSE, 530 patients hospitalized for acute de novo or decompensated chronic heart failure were randomized, once clinically stable (median 3 days after admission), to empagliflozin 10 mg or placebo for 90 days. Empagliflozin increased the hierarchical clinical benefit of death, heart failure events, and symptom change (win ratio 1.36; 95% CI 1.09 to 1.68), regardless of ejection fraction or diabetes. In DICTATE-AHF, dapagliflozin started within 24 hours of admission in 240 patients did not significantly improve diuretic efficiency, but patients needed less loop diuretic and fewer escalations for the same weight loss. In DAPA-RESIST, 61 patients resistant to intravenous furosemide were randomized to dapagliflozin 10 mg or metolazone 5 to 10 mg daily for 3 days. Weight loss at 96 hours was 3.0 and 3.6 kg (difference not significant); patients assigned to dapagliflozin received more furosemide (977 and 704 mg) but had smaller falls in sodium and potassium and smaller rises in urea and creatinine.

An SGLT2 inhibitor is started during hospitalization in eligible patients for its long-term benefit, and its diuretic contribution is secondary. When it is started in a euvolemic outpatient already taking a loop diuretic, the loop dose may need reduction to avoid volume depletion, particularly in older adults.

Mineralocorticoid receptor antagonists, amiloride, and triamterene

Potassium-sparing diuretics act in the collecting duct, which reabsorbs only 2% to 5% of filtered sodium, so their natriuretic effect is small. Their value lies in potassium conservation, in blocking aldosterone-driven sodium retention, and, for MRAs, in disease modification. The trial evidence for spironolactone, eplerenone, and finerenone in HFrEF and HFpEF, their dosing, and hyperkalemia management are covered on the HFrEF and HFpEF pages.

Potassium-sparing diuretics. Heart failure and resistant hypertension doses from the 2022 AHA/ACC/HFSA and 2025 AHA/ACC guidelines; ascites doses from the 2021 American Association for the Study of Liver Diseases (AASLD) practice guidance.

MRAs as diuretics. At standard heart failure doses, MRAs add little natriuresis. In ATHENA-HF, 360 patients with acute heart failure received spironolactone 100 mg daily or usual care (placebo or 25 mg) for 96 hours; the fall in NT-proBNP did not differ, and neither did secondary measures of congestion and urine output. Their role in diuretic resistance is indirect: they block aldosterone-driven distal reabsorption, offset the potassium losses of loop and thiazide diuretics, and are part of GDMT in HFrEF. In cirrhosis, in which secondary hyperaldosteronism dominates, spironolactone is the primary diuretic.

Amiloride. Amiloride blocks ENaC independently of aldosterone. In the PATHWAY-2 substudy, amiloride 10 to 20 mg daily lowered clinic systolic pressure in resistant hypertension by about as much as spironolactone. In a randomized open-label trial in South Korea, 118 patients whose home systolic pressure remained uncontrolled on an angiotensin receptor blocker (ARB), a calcium channel blocker, and a thiazide were assigned to amiloride or spironolactone. Amiloride 5 to 10 mg was noninferior to spironolactone 12.5 to 25 mg for the change in home systolic pressure at 12 weeks (−13.6 and −14.7 mm Hg; difference −0.68 mm Hg; 90% CI −3.50 to 2.14). The 2025 AHA/ACC guideline lists amiloride among the options when an MRA is not tolerated or is contraindicated (Class 2a).

Vasopressin antagonists, hypertonic saline, albumin, and ultrafiltration

These strategies are used in hospitalized patients with refractory congestion and are discussed in full on the acute decompensated heart failure page.

Tolvaptan blocks vasopressin V2 receptors and produces free water excretion without natriuresis. In EVEREST, tolvaptan started in hospital improved weight, dyspnea, and serum sodium early without changing mortality or heart failure readmission. Its role is the treatment of clinically significant hypervolemic hyponatremia, discussed below.

Hypertonic saline in small volumes with high-dose furosemide has been proposed to draw interstitial fluid into the vascular space and increase sodium delivery to the loop. The randomized trials come from a single center in Palermo, Italy, and are single-blind. The largest, SMAC-HF (1,771 patients with NYHA class III heart failure), reported shorter hospital stay and fewer readmissions and deaths with hypertonic saline. However, the control group was also assigned a stricter sodium diet, so the effects of saline and diet cannot be separated. The results have not been reproduced in a multicenter trial. In a retrospective US series of 58 administrations in 40 patients with refractory acute decompensated heart failure, hypertonic saline added to loop diuretics was followed by greater urine output, weight loss, and diuretic efficiency and by improvement in serum sodium, chloride, and creatinine. No respiratory or neurologic complications were observed. US guidelines make no recommendation, and the approach remains investigational.

Albumin with furosemide. Co-administration has been proposed in hypoalbuminemia to increase delivery to the tubule. In a meta-analysis of 13 studies with 422 participants, co-administration increased urine output by about 31 mL/h and sodium excretion by about 1.8 mEq/h over furosemide alone, with high heterogeneity. The effect was greater with serum albumin below 2.5 g/dL and albumin doses above 30 g and was confined to the first 12 hours. It is not routine in heart failure.

Ultrafiltration. Venovenous ultrafiltration removes isotonic fluid at a set rate. In CARRESS-HF, 188 patients with worsening kidney function and persistent congestion were randomized to ultrafiltration or a stepped pharmacologic algorithm that included sequential nephron blockade. Ultrafiltration was inferior for the combined end point of change in creatinine and weight at 96 hours, because creatinine rose more (+0.23 and −0.04 mg/dL) with similar weight loss (5.7 and 5.5 kg). Serious adverse events were more frequent (72% and 57%). A well-executed stepped diuretic protocol is therefore the standard before mechanical fluid removal, which is reserved for congestion refractory to pharmacologic therapy.

Thiazide and thiazide-like diuretics in hypertension

Thiazide-type diuretics are one of three first-line antihypertensive classes in the 2025 AHA/ACC guideline, with angiotensin-converting enzyme (ACE) inhibitors or ARBs and long-acting dihydropyridine calcium channel blockers. They lower blood pressure and reduce cardiovascular events in outcome trials. Initial therapy with two first-line classes of different mechanism, preferably in a single pill, is recommended in stage 2 hypertension (Class 1), and a thiazide-type diuretic is part of the standard three-drug regimen. Blood pressure thresholds, goals, and drug selection for compelling indications are covered on the hypertension page.

Thiazide-type and thiazide-like agents. Hydrochlorothiazide and chlorothiazide have the benzothiadiazine structure of the original thiazides. Chlorthalidone, indapamide, and metolazone lack it but block the same transporter and are called thiazide-like. The distinction that matters clinically is duration of action and potency per milligram, not chemistry.

Thiazide and thiazide-like diuretics. Antihypertensive doses from the 2025 AHA/ACC guideline (Table 13); durations from the 2022 AHA/ACC/HFSA guideline; half-lives from pharmacokinetic reviews and prescribing information.

Chlorthalidone and hydrochlorothiazide. Chlorthalidone was the diuretic in ALLHAT and in most trials that established diuretic-based treatment, and it is about 1.5 to 2 times as potent as hydrochlorothiazide per milligram. In ALLHAT, among 33,357 high-risk hypertensive adults, chlorthalidone was not inferior to amlodipine or lisinopril for coronary events and prevented heart failure more effectively than either. The Diuretic Comparison Project then compared them in 13,523 veterans aged 65 years or older taking hydrochlorothiazide 25 or 50 mg, randomized to continue it or switch to chlorthalidone 12.5 or 25 mg. Over a median of 2.4 years, the composite of myocardial infarction, stroke, heart failure, urgent revascularization, and non-cancer death did not differ (10.4% and 10.0%), and hypokalemia was more frequent with chlorthalidone (6.0% and 4.4%). Patients with prior myocardial infarction or stroke appeared to benefit from chlorthalidone. The 2025 guideline therefore groups the agents as thiazide-type diuretics for routine use but prefers chlorthalidone or indapamide when hypertension is resistant.

Older adults. In HYVET, 3,845 patients aged 80 years or older with systolic pressure of 160 mm Hg or more received sustained-release indapamide 1.5 mg, with perindopril added if needed, or placebo. Treatment reduced death from any cause (hazard ratio 0.79) and heart failure (hazard ratio 0.36), and the trial was stopped early for benefit.

Low eGFR. Thiazide-type diuretics were long considered ineffective below an eGFR of about 30 mL/min/1.73 m². In CLICK, 160 patients with stage 4 chronic kidney disease (mean eGFR 23 mL/min/1.73 m²) and poorly controlled hypertension were randomized to chlorthalidone 12.5 mg titrated to 50 mg or placebo; 60% were already taking loop diuretics. Chlorthalidone lowered 24-hour ambulatory systolic pressure by 10.5 mm Hg more than placebo at 12 weeks, at the cost of more hypokalemia, hyperglycemia, dizziness, and reversible rises in creatinine. Loop diuretics remain the agents for volume control in advanced kidney disease.

Dose and response. The antihypertensive effect of hydrochlorothiazide rises little above 25 mg, whereas hypokalemia, hyperuricemia, and hyperglycemia continue to increase with dose. Low doses in combination with a renin-angiotensin inhibitor, which reduces potassium loss, or with a potassium-sparing diuretic are the usual regimens. Electrolytes and creatinine are measured before and 2 to 4 weeks after a thiazide is started or the dose increased, and sodium is checked early in older adults, in whom thiazide-induced hyponatremia occurs early and can be severe.

Resistant hypertension. Resistant hypertension requires failure of three agents at optimal doses including a diuretic. The diuretic is optimized first: chlorthalidone or indapamide replaces hydrochlorothiazide, and a loop diuretic replaces the thiazide when eGFR is below about 30 mL/min/1.73 m² or volume overload is present. An MRA is then added when eGFR is 45 mL/min/1.73 m² or higher (Class 1), with amiloride among the alternatives when an MRA is not tolerated (Class 2a). Loop diuretics with a short action, such as once-daily furosemide, are poor antihypertensives; when a loop diuretic is needed for hypertension, torsemide or twice-daily furosemide is used.

Metabolic effects of loop and thiazide diuretics: calcium, uric acid, glucose, and lipids

Metabolic effects of loop and thiazide-type diuretics.

Hypokalemia and hypomagnesemia

Loop and thiazide-type diuretics increase potassium excretion by delivering more sodium to the collecting duct, where its reabsorption through ENaC drives potassium secretion, and by stimulating aldosterone through volume contraction. Hypokalemia is more frequent with long-acting agents (chlorthalidone more than hydrochlorothiazide), with combination loop and thiazide therapy, and with high sodium intake, which increases distal delivery. In heart failure and coronary disease, hypokalemia increases the risk of ventricular arrhythmias and sudden death, potentiates digoxin toxicity, and increases the risk of torsades de pointes with QT-prolonging drugs. In an individual patient data analysis of 12 heart failure trials, the risk of death and heart failure events was lowest at a serum potassium of 4.2 to 5.0 mEq/L, and a potassium below 3.5 mEq/L was associated with higher mortality in HFrEF.

Prevention. Renin-angiotensin inhibitors and MRAs reduce potassium loss, and most patients with HFrEF taking them need no supplement; some develop hyperkalemia. Potassium chloride is the preferred supplement because chloride depletion accompanies diuretic-induced hypokalemia and potassium cannot be retained without it. Amiloride or triamterene added to a thiazide prevents hypokalemia in hypertension.

Sequential blockade. The combination of a loop diuretic with metolazone or another thiazide can lower potassium by more than 1 mEq/L within days. Potassium, magnesium, sodium, and creatinine are measured at least daily in hospital during combination therapy and within a few days of starting it in outpatients.

Magnesium. Loop diuretics reduce paracellular magnesium reabsorption in the thick ascending limb, and thiazides increase distal magnesium loss. Hypomagnesemia increases renal potassium wasting, so hypokalemia does not correct until magnesium is replaced, and it independently promotes ventricular arrhythmias and digoxin toxicity. Magnesium is measured with potassium in patients taking high-dose or combination diuretic therapy.

Hyponatremia and hypochloremia

Hyponatremia in heart failure is usually hypervolemic: nonosmotic vasopressin release and low distal delivery limit free water excretion. It predicts mortality. Thiazide-type diuretics worsen it because they block NCC in the diluting segment; loop diuretics impair the medullary gradient and the concentrating ability of the kidney, so they cause less hyponatremia. Thiazide-induced hyponatremia occurs most often in older women and patients with low body weight or high water intake, usually within the first weeks of treatment, and recurs on rechallenge.

Management in heart failure. A thiazide is stopped or limited, and loop diuretic therapy continues to remove sodium and water. Fluid restriction modestly raises sodium; the 2022 guideline considers its benefit for congestive symptoms in advanced heart failure with hyponatremia uncertain (Class 2b). Tolvaptan is indicated for clinically significant hypervolemic or euvolemic hyponatremia (sodium below 125 mEq/L, or less marked hyponatremia that is symptomatic and has resisted fluid restriction). Treatment is started or restarted in hospital at 15 mg once daily and limited to 30 days because of liver injury; correction faster than 12 mEq/L in 24 hours risks osmotic demyelination. Severe symptomatic hyponatremia is treated with hypertonic saline under nephrology guidance.

Hypochloremia. Serum chloride falls with loop and thiazide therapy and with water retention. Epidemiologic studies implicate low chloride, more than low sodium, in the poor survival associated with hyponatremia in heart failure, and hypochloremia is associated with diuretic resistance and neurohormonal activation, as described above.

Acid-base disturbances: contraction alkalosis and acetazolamide acidosis

Vigorous loop and thiazide diuresis produces a hypochloremic metabolic alkalosis through contraction of the extracellular volume around a fixed amount of bicarbonate, chloride depletion, and increased distal hydrogen ion secretion stimulated by aldosterone and hypokalemia. The alkalosis maintains itself because chloride depletion prevents bicarbonate excretion. It suppresses ventilatory drive, which matters in patients with chronic lung disease or obesity hypoventilation, and it is associated with reduced loop diuretic efficiency. Correction requires chloride: potassium chloride and, where volume allows, sodium chloride, or acetazolamide, which increases bicarbonate excretion and allows decongestion to continue. Acetazolamide in turn produces a hyperchloremic non-anion-gap metabolic acidosis that limits its own effect after a few days; bicarbonate is measured daily during treatment. MRAs and amiloride cause mild hyperkalemic metabolic acidosis, particularly in kidney disease.

Volume depletion, kidney function, and assessment of congestion

Volume depletion. Overdiuresis causes orthostatic hypotension, prerenal azotemia with a disproportionate rise in blood urea nitrogen, fatigue, and falls, and it limits the doses of renin-angiotensin inhibitors and beta-blockers that patients tolerate. The risk is highest in small, stiff ventricles that depend on filling pressure (HFpEF, cardiac amyloidosis, hypertrophic cardiomyopathy with obstruction), in older adults, during acute illness with vomiting or diarrhea, and after SGLT2 inhibitors or angiotensin receptor-neprilysin inhibitors (ARNIs) are added.

Creatinine during decongestion. A rise in creatinine is common during active diuresis and does not by itself indicate harm. Rises accompanied by hemoconcentration (rising hemoglobin and albumin) and continuing effective decongestion reflect a hemodynamic change in filtration and are associated with better survival. The 2022 guideline recommends that diuresis and GDMT not be stopped routinely for a mild decrease in kidney function or an asymptomatic fall in blood pressure (Class 1). A rising creatinine with persistent congestion and poor diuretic response, or with hypotension and hypoperfusion, requires reassessment of the hemodynamic state. These patterns are detailed on the acute decompensated heart failure page.

Assessment of congestion. The end point of diuretic therapy is the disappearance of congestion, assessed by orthopnea, jugular venous pressure, hepatojugular reflux, edema, and weight, supported by natriuretic peptide trends. Hemoconcentration indicates that diuresis is outpacing interstitial refill and suggests that intravascular volume is no longer expanded. Point-of-care ultrasound adds objective measures: inferior vena cava size and collapse, B-lines on lung ultrasound, and the venous excess ultrasound (VExUS) grade, which combines inferior vena cava size with hepatic, portal, and intrarenal venous Doppler patterns. Right heart catheterization measures filling pressures in patients whose volume status remains uncertain or whose response is unexpected. The right heart catheterization page covers these measurements.

Ototoxicity, sulfonamide sensitivity, and other adverse effects

Ototoxicity. Loop diuretics alter the ionic composition of endolymph through the Na⁺-K⁺-2Cl⁻ cotransporter isoform 1 (NKCC1) in the stria vascularis and can cause tinnitus, vertigo, and reversible or permanent hearing loss. The risk rises with rapid intravenous injection, high doses, severe kidney impairment, hypoproteinemia, and concomitant aminoglycosides or other ototoxic drugs. Ethacrynic acid is the most ototoxic.

Sulfonamide sensitivity. Furosemide, bumetanide, torsemide, the thiazides, metolazone, and acetazolamide contain a sulfonamide group, but not the arylamine group at the N4 position and the N1 heterocyclic ring implicated in reactions to sulfonamide antibiotics. In a cohort study from the UK General Practice Research Database, 9.9% of patients with an allergic reaction to a sulfonamide antibiotic reacted to a later sulfonamide nonantibiotic, compared with 1.6% of those without such a history. However, the same patients had an even higher risk of reacting to a penicillin, which indicates a general predisposition to drug allergy and not cross-reactivity. A history of a nonsevere reaction to a sulfonamide antibiotic is therefore not a contraindication to a loop or thiazide diuretic. After a severe reaction to a sulfonamide diuretic, such as Stevens-Johnson syndrome, ethacrynic acid is the alternative.

Nonmelanoma skin cancer. Hydrochlorothiazide is photosensitizing. In a Danish case-control study, high cumulative use (50,000 mg or more) was associated with odds ratios of 1.29 for basal cell carcinoma and 3.98 for squamous cell carcinoma, with a dose-response relation. US labeling was updated in 2020 to describe a small increased risk of nonmelanoma skin cancer, mainly squamous cell carcinoma, with advice on sun protection and skin examination. Patients with a history of skin cancer or heavy sun exposure may be treated with another first-line antihypertensive agent.

Other adverse effects. Erectile dysfunction is more frequent with thiazides than with most other antihypertensive classes. Uncommon effects include interstitial nephritis, pancreatitis, thrombocytopenia, photosensitive rash, and cholestatic jaundice. Loop diuretics in high doses can cause muscle cramps from volume depletion and hypokalemia. Gynecomastia occurs with spironolactone but not eplerenone or finerenone. Triamterene can precipitate in urine and form stones.

Drug interactions of diuretics

Clinically important interactions of diuretics, from the prescribing information, the 2025 AHA/ACC hypertension guideline, and the 2022 AHA/ACC/HFSA guideline.

Outpatient diuretic adjustment, dry weight, and maintenance dosing

The maintenance goal. In chronic heart failure the loop diuretic is given at the lowest dose that maintains euvolemia, because excess dosing causes hypotension and kidney dysfunction that limit GDMT, and too little allows recurrent congestion. The target, often called the dry weight, is the weight at which the patient has no clinical congestion. It is an estimate and changes with nutrition, muscle loss, and illness, so it is revised at each visit.

After discharge. Most patients who required intravenous diuretics need a loop diuretic at discharge, and dose increases are common in the following weeks. The discharge dose is estimated from the intravenous dose that achieved euvolemia, the bioavailability of the oral agent, and the expected reduction in requirement as GDMT takes effect. A plan for diuretic adjustment at discharge is a Class 1 recommendation, and laboratory tests are repeated within 1 to 2 weeks.

Flexible dosing. Patients who weigh themselves daily can adjust the diuretic within limits set by the clinician. A written plan specifies the weight gain or symptoms that trigger an extra dose, the maximum extra dosing before the team is contacted, and when to hold a dose (dizziness, poor oral intake, vomiting, or diarrhea). Self-adjustment is taught with teach-back and works best in patients with reliable scales and stable cognition.

Outpatient diuretic adjustment in chronic heart failure, after the 2022 AHA/ACC/HFSA guideline, the 2020 JACC state-of-the-art review, and the ReBIC-1 trial.

Withdrawal. In ReBIC-1, 188 stable outpatients with heart failure (mean ejection fraction 32%), NYHA class I–II symptoms, no fluid retention, and low-dose furosemide were randomized double-blind to furosemide withdrawal or continuation. Dyspnea over 90 days did not differ (P = 0.94), and 75.3% and 83.7% remained free of furosemide reuse. In selected euvolemic patients on full GDMT, a trial of withdrawal is reasonable; ReBIC-1 was short and its result does not apply to patients with prior congestion requiring higher doses.

Tolerance and diuretic holidays. Short-term tolerance (braking) and long-term distal adaptation develop during continuous therapy. No trial supports scheduled interruption of diuretics to restore responsiveness in heart failure; interruption risks rebound sodium retention. Restoring responsiveness depends on the countermeasures to the mechanism of resistance.

Diuretics in HFpEF, right heart failure, and cirrhosis

HFpEF. Diuretics relieve congestion in HFpEF and are recommended for fluid retention (Class 1), without a mortality benefit. Patients with HFpEF often have a steep pressure-volume relation, so small volume changes produce large changes in filling pressure: pulmonary edema with modest volume gain and hypotension or kidney dysfunction with modest overdiuresis. SGLT2 inhibitors and, in selected patients, MRAs reduce heart failure events in HFpEF and lower diuretic needs. The HFpEF page covers these therapies. Cardiac amyloidosis is a common cause of diuretic-dependent HFpEF in older adults, and its management is covered on the cardiac amyloidosis page.

Right heart failure and severe tricuspid regurgitation. Systemic venous congestion causes gut edema, hepatic congestion, and ascites, which impair oral furosemide absorption and raise renal venous pressure. Torsemide or bumetanide, intravenous or subcutaneous dosing during exacerbations, and an MRA are commonly used. In right ventricular infarction or pulmonary embolism without systemic congestion, diuretics can lower right ventricular preload and output and are withheld.

Cirrhosis and ascites. Secondary hyperaldosteronism is the dominant mechanism of sodium retention, so an MRA is the base diuretic. The 2021 AASLD practice guidance recommends moderate dietary sodium restriction (2 g, or 90 mEq, daily) with spironolactone started at 100 mg daily and increased in steps at least 72 hours apart to a maximum of 400 mg daily. Spironolactone alone usually suffices for a first episode; in long-standing ascites, furosemide is added at 40 mg daily and increased toward 160 mg daily. Hyponatremia, hepatic encephalopathy, and acute kidney injury limit diuresis.

Diuretics in chronic kidney disease, nephrotic syndrome, and dialysis

Chronic kidney disease. As glomerular filtration rate falls, fewer nephrons secrete loop diuretic into the lumen, and endogenous organic anions compete for secretion, so the threshold rises and a larger single dose is needed. Frequency is increased once the effective dose is found. Furosemide accumulates in kidney failure because its renal glucuronidation is impaired, which increases the risk of ototoxicity with very high doses; torsemide, cleared mainly by the liver, does not accumulate. Thiazide-type diuretics are less natriuretic at low eGFR, but metolazone and chlorthalidone retain useful effects, and in combination with a loop diuretic they produce large natriuresis in advanced kidney disease, with a high risk of electrolyte disturbance. SGLT2 inhibitors are indicated for kidney protection in chronic kidney disease down to low eGFR and add a small natriuresis.

Nephrotic syndrome. Hypoalbuminemia expands the volume of distribution of loop diuretics and reduces delivery to the proximal tubule, and filtered albumin binds the drug in the tubular lumen. Loop doses two to three times higher than usual are needed, sequential blockade with a thiazide-type diuretic or amiloride is often required, and renin-angiotensin inhibition reduces proteinuria. Co-administration of albumin with furosemide produces a modest, short-lived increase in urine output and natriuresis, greatest when serum albumin is below 2.5 g/dL; it is reserved for refractory edema.

Dialysis. In patients receiving dialysis who retain some kidney function, high-dose loop diuretics increase urine volume and reduce interdialytic weight gain. They do not substitute for ultrafiltration when residual function is minimal.

Pregnancy, older adults, and gout

Pregnancy. Loop diuretics are used for pulmonary congestion in heart failure, peripartum cardiomyopathy, and valve disease during pregnancy, at the lowest effective dose, because volume contraction can reduce uteroplacental perfusion. Diuretics are not used to prevent or treat preeclampsia. Spironolactone is avoided because of antiandrogenic effects on the male fetus in animal studies. Thiazide-type diuretics are not first-line antihypertensive agents in pregnancy, and the choice of agents for hypertension in pregnancy is covered on the hypertension page.

Older adults. Older adults are more susceptible to volume depletion, orthostatic hypotension, falls, thiazide-induced hyponatremia, hypokalemia, and urinary incontinence, which also reduces adherence. Lower starting doses, measurement of sodium and potassium within 1 to 2 weeks of a change, a sick-day plan, and review of diuretic need when weight, appetite, or kidney function changes reduce these risks.

Gout. Loop and thiazide-type diuretics raise serum urate and can precipitate gout. When a diuretic is required for heart failure, it is continued and gout is managed with urate-lowering therapy. In hypertension, a patient with gout can be treated with other first-line classes; losartan is uricosuric, and SGLT2 inhibitors lower serum urate.

Outcome evidence and guideline recommendations for diuretic therapy

Loop diuretics are prescribed to most patients with heart failure, yet their effect on mortality has never been tested in an adequately powered placebo-controlled trial, and such a trial is not feasible in congested patients. Their benefit on symptoms, exercise capacity, and congestion is not in doubt. In hypertension, by contrast, diuretic-based regimens reduced stroke, heart failure, and cardiovascular events in large outcome trials such as ALLHAT and HYVET.

US guideline recommendations on diuretic therapy, from the 2022 AHA/ACC/HFSA heart failure guideline (corrections published in 2022 and 2023) and the 2025 AHA/ACC hypertension guideline (corrections published in 2025 and 2026).

How European guidance differs. The 2021 ESC heart failure guideline and the 2019 HFA position statement build early natriuresis assessment (spot urine sodium at 2 hours and urine output over 6 hours) into a formal decongestion algorithm. The algorithm specifies dose doubling and combination therapy at defined steps. US guidance describes response monitoring and intensification without a fixed protocol.

Sources of error in diuretic prescribing

• Repeating an ineffective loop dose: below the natriuretic threshold, a second identical dose produces no more natriuresis than the first. The dose is doubled.
• Once-daily furosemide in a patient with high diuretic requirements: natriuresis lasts 6 to 8 hours, and postdiuretic sodium retention during the rest of the day can cancel it.
• Equating oral and intravenous furosemide milligram for milligram as maintenance doses: oral bioavailability averages about 50%, so 80 mg intravenously corresponds to about 160 mg orally, and discharge on an oral dose equal to the intravenous dose often undertreats.
• Declaring resistance when the problem is absorption: a patient who responds briskly to intravenous furosemide after failing oral therapy needs a better-absorbed agent or a parenteral route, not a second diuretic.
• Missing excess sodium intake or NSAIDs: a high urine sodium with weight gain points to intake; over-the-counter NSAIDs blunt every diuretic.
• Adding metolazone daily without an end date: its long action and synergy with loop diuretics can cause profound hypokalemia, hyponatremia, and volume depletion within days.
• Correcting hypokalemia without magnesium and chloride: potassium does not correct while magnesium is low, and alkalosis persists while chloride is depleted.
• Stopping diuresis for a small creatinine rise while congestion persists: residual congestion at discharge predicts readmission and death more strongly than the creatinine change.
• Assuming a sulfonamide antibiotic allergy forbids loop or thiazide diuretics: cross-reactivity is not supported by cohort data, and ethacrynic acid is needed only after a severe reaction to a sulfonamide diuretic.
• Expecting torsemide to improve survival: TRANSFORM-HF showed no mortality difference; the agent is chosen for absorption, duration, and cost.
• Leaving the loop dose unchanged when an SGLT2 inhibitor or ARNI is started in a euvolemic older patient: additive natriuresis can cause hypotension and kidney dysfunction that lead to withdrawal of the disease-modifying drug.
• Using once-daily furosemide as an antihypertensive: its short action leaves most of the day uncovered; chlorthalidone or indapamide, or torsemide when a loop diuretic is required, is used.

References

1. Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2022;145:e895–e1032. Corrections in Circulation. 2022;145:e1033, 2022;146:e185, and 2023;147:e674. doi:10.1161/CIR.0000000000001063
2. Hollenberg SM, Stevenson LW, Ahmad T, et al. 2024 ACC expert consensus decision pathway on clinical assessment, management, and trajectory of patients hospitalized with heart failure focused update. J Am Coll Cardiol. 2024;84:1241–1267. doi:10.1016/j.jacc.2024.06.002
3. Jones DW, Ferdinand KC, Taler SJ, et al. 2025 AHA/ACC/AANP/AAPA/ABC/ACCP/ACPM/AGS/AMA/ASPC/NMA/PCNA/SGIM guideline for the prevention, detection, evaluation and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2025;152:e114–e218. Corrections in Circulation. 2025;152:e403, 2026;153:e1121, and 2026;153:e1359. doi:10.1161/CIR.0000000000001356
4. Mullens W, Damman K, Harjola VP, et al. The use of diuretics in heart failure with congestion: a position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail. 2019;21:137–155. doi:10.1002/ejhf.1369
5. McDonagh TA, Metra M, Adamo M, et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021;42:3599–3726. doi:10.1093/eurheartj/ehab368
6. Felker GM, Ellison DH, Mullens W, Cox ZL, Testani JM. Diuretic therapy for patients with heart failure: JACC state-of-the-art review. J Am Coll Cardiol. 2020;75:1178–1195. doi:10.1016/j.jacc.2019.12.059
7. Ellison DH, Felker GM. Diuretic treatment in heart failure. N Engl J Med. 2017;377:1964–1975. doi:10.1056/NEJMra1703100
8. Brater DC. Diuretic therapy. N Engl J Med. 1998;339:387–395. doi:10.1056/NEJM199808063390607
9. Oh SW, Han SY. Loop diuretics in clinical practice. Electrolyte Blood Press. 2015;13:17–21. doi:10.5049/EBP.2015.13.1.17
10. Ellison DH, Velázquez H, Wright FS. Adaptation of the distal convoluted tubule of the rat: structural and functional effects of dietary salt intake and chronic diuretic infusion. J Clin Invest. 1989;83:113–126. doi:10.1172/JCI113847
11. Felker GM, Lee KL, Bull DA, et al. Diuretic strategies in patients with acute decompensated heart failure. N Engl J Med. 2011;364:797–805. doi:10.1056/NEJMoa1005419
12. Mentz RJ, Anstrom KJ, Eisenstein EL, et al. Effect of torsemide vs furosemide after discharge on all-cause mortality in patients hospitalized with heart failure: the TRANSFORM-HF randomized clinical trial. JAMA. 2023;329:214–223. doi:10.1001/jama.2022.23924
13. Keshvani N, Grewal A, Rizvi SK, et al. Diuretic resistance risk and the efficacy of torsemide vs furosemide: a post-hoc analysis of the TRANSFORM-HF trial. Eur J Heart Fail. 2026. Published online ahead of print. doi:10.1093/ejhf/xuag207
14. Cosín J, Díez J; TORIC investigators. Torasemide in chronic heart failure: results of the TORIC study. Eur J Heart Fail. 2002;4:507–513. doi:10.1016/S1388-9842(02)00122-8
15. Murray MD, Deer MM, Ferguson JA, et al. Open-label randomized trial of torsemide compared with furosemide therapy for patients with heart failure. Am J Med. 2001;111:513–520. doi:10.1016/S0002-9343(01)00903-2
16. López B, Querejeta R, González A, Sánchez E, Larman M, Díez J. Effects of loop diuretics on myocardial fibrosis and collagen type I turnover in chronic heart failure. J Am Coll Cardiol. 2004;43:2028–2035. doi:10.1016/j.jacc.2003.12.052
17. Testani JM, Hanberg JS, Cheng S, et al. Rapid and highly accurate prediction of poor loop diuretic natriuretic response in patients with heart failure. Circ Heart Fail. 2016;9:e002370. doi:10.1161/CIRCHEARTFAILURE.115.002370
18. ter Maaten JM, Beldhuis IE, van der Meer P, et al. Natriuresis-guided diuretic therapy in acute heart failure: a pragmatic randomized trial. Nat Med. 2023;29:2625–2632. doi:10.1038/s41591-023-02532-z
19. Dauw J, Charaya K, Lelonek M, et al. Protocolized natriuresis-guided decongestion improves diuretic response: the multicenter ENACT-HF study. Circ Heart Fail. 2024;17:e011105. doi:10.1161/CIRCHEARTFAILURE.123.011105
20. Segar MW, Khan MS, Patel KV, et al. A phenomapping tool and clinical score to identify low diuretic efficiency in acute decompensated heart failure. JACC Heart Fail. 2024;12:508–520. doi:10.1016/j.jchf.2023.09.029
21. Hanberg JS, Rao V, ter Maaten JM, et al. Hypochloremia and diuretic resistance in heart failure: mechanistic insights. Circ Heart Fail. 2016;9:e003180. doi:10.1161/CIRCHEARTFAILURE.116.003180
22. Ezekowitz JA, Colin-Ramirez E, Ross H, et al. Reduction of dietary sodium to less than 100 mmol in heart failure (SODIUM-HF): an international, open-label, randomised, controlled trial. Lancet. 2022;399:1391–1400. doi:10.1016/S0140-6736(22)00369-5
23. Cox ZL, Hung R, Lenihan DJ, Testani JM. Diuretic strategies for loop diuretic resistance in acute heart failure: the 3T trial. JACC Heart Fail. 2020;8:157–168. doi:10.1016/j.jchf.2019.09.012
24. Trullàs JC, Morales-Rull JL, Casado J, et al. Combining loop with thiazide diuretics for decompensated heart failure: the CLOROTIC trial. Eur Heart J. 2023;44:411–421. doi:10.1093/eurheartj/ehac689
25. Mullens W, Dauw J, Martens P, et al. Acetazolamide in acute decompensated heart failure with volume overload. N Engl J Med. 2022;387:1185–1195. doi:10.1056/NEJMoa2203094
26. Voors AA, Angermann CE, Teerlink JR, et al. The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial. Nat Med. 2022;28:568–574. doi:10.1038/s41591-021-01659-1
27. Cox ZL, Collins SP, Hernandez GA, et al. Efficacy and safety of dapagliflozin in patients with acute heart failure. J Am Coll Cardiol. 2024;83:1295–1306. doi:10.1016/j.jacc.2024.02.009
28. Yeoh SE, Osmanska J, Petrie MC, et al. Dapagliflozin vs. metolazone in heart failure resistant to loop diuretics. Eur Heart J. 2023;44:2966–2977. doi:10.1093/eurheartj/ehad341
29. Butler J, Anstrom KJ, Felker GM, et al. Efficacy and safety of spironolactone in acute heart failure: the ATHENA-HF randomized clinical trial. JAMA Cardiol. 2017;2:950–958. doi:10.1001/jamacardio.2017.2198
30. Bart BA, Goldsmith SR, Lee KL, et al. Ultrafiltration in decompensated heart failure with cardiorenal syndrome. N Engl J Med. 2012;367:2296–2304. doi:10.1056/NEJMoa1210357
31. Konstam MA, Gheorghiade M, Burnett JC Jr, et al. Effects of oral tolvaptan in patients hospitalized for worsening heart failure: the EVEREST Outcome Trial. JAMA. 2007;297:1319–1331. doi:10.1001/jama.297.12.1319
32. Griffin M, Soufer A, Goljo E, et al. Real world use of hypertonic saline in refractory acute decompensated heart failure: a U.S. center's experience. JACC Heart Fail. 2020;8:199–208. doi:10.1016/j.jchf.2019.10.012
33. Lee TH, Kuo G, Chang CH, et al. Diuretic effect of co-administration of furosemide and albumin in comparison to furosemide therapy alone: an updated systematic review and meta-analysis. PLoS One. 2021;16:e0260312. doi:10.1371/journal.pone.0260312
34. Sica DA, Muntendam P, Myers RL, et al. Subcutaneous furosemide in heart failure: pharmacokinetic characteristics of a newly buffered solution. JACC Basic Transl Sci. 2018;3:25–34. doi:10.1016/j.jacbts.2017.10.001
35. Rohde LE, Rover MM, Figueiredo Neto JA, et al. Short-term diuretic withdrawal in stable outpatients with mild heart failure and no fluid retention receiving optimal therapy: a double-blind, multicentre, randomized trial. Eur Heart J. 2019;40:3605–3612. doi:10.1093/eurheartj/ehz554
36. Ishani A, Cushman WC, Leatherman SM, et al. Chlorthalidone vs. hydrochlorothiazide for hypertension–cardiovascular events. N Engl J Med. 2022;387:2401–2410. doi:10.1056/NEJMoa2212270
37. Agarwal R, Sinha AD, Cramer AE, et al. Chlorthalidone for hypertension in advanced chronic kidney disease. N Engl J Med. 2021;385:2507–2519. doi:10.1056/NEJMoa2110730
38. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288:2981–2997. doi:10.1001/jama.288.23.2981
39. Beckett NS, Peters R, Fletcher AE, et al. Treatment of hypertension in patients 80 years of age or older. N Engl J Med. 2008;358:1887–1898. doi:10.1056/NEJMoa0801369
40. Lee CJ, Ihm SH, Shin DH, et al. Spironolactone vs amiloride for resistant hypertension: a randomized clinical trial. JAMA. 2025;333:2073–2082. doi:10.1001/jama.2025.5129
41. Williams B, MacDonald TM, Morant SV, et al. Endocrine and haemodynamic changes in resistant hypertension, and blood pressure responses to spironolactone or amiloride: the PATHWAY-2 mechanisms substudies. Lancet Diabetes Endocrinol. 2018;6:464–475. doi:10.1016/S2213-8587(18)30071-8
42. Pedersen SA, Gaist D, Schmidt SAJ, Hölmich LR, Friis S, Pottegård A. Hydrochlorothiazide use and risk of nonmelanoma skin cancer: a nationwide case-control study from Denmark. J Am Acad Dermatol. 2018;78:673–681.e9. doi:10.1016/j.jaad.2017.11.042
43. Strom BL, Schinnar R, Apter AJ, et al. Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med. 2003;349:1628–1635. doi:10.1056/NEJMoa022963
44. Ono R, Chimura M, Docherty KF, et al. Optimal serum potassium concentrations in heart failure: an individual patient data meta-analysis. Eur Heart J. 2026;47:4480–4499. doi:10.1093/eurheartj/ehag341
45. Biggins SW, Angeli P, Garcia-Tsao G, et al. Diagnosis, evaluation, and management of ascites, spontaneous bacterial peritonitis and hepatorenal syndrome: 2021 practice guidance by the American Association for the Study of Liver Diseases. Hepatology. 2021;74:1014–1048. doi:10.1002/hep.31884
46. Licata G, Di Pasquale P, Parrinello G, et al. Effects of high-dose furosemide and small-volume hypertonic saline solution infusion in comparison with a high dose of furosemide as bolus in refractory congestive heart failure: long-term effects. Am Heart J. 2003;145:459–466. doi:10.1067/mhj.2003.166
47. Paterna S, Fasullo S, Parrinello G, et al. Short-term effects of hypertonic saline solution in acute heart failure and long-term effects of a moderate sodium restriction in patients with compensated heart failure with New York Heart Association class III (Class C) (SMAC-HF Study). Am J Med Sci. 2011;342:27–37. doi:10.1097/MAJ.0b013e31820f10ad