Torsemide

What it is

Torsemide is a synthetic loop diuretic. Its label describes it as a diuretic of the pyridine-sulfonylurea class with molecular weight 348.43, supplied as a white to off-white crystalline powder pressed into tablets. It is taken by mouth and is almost completely absorbed, with a bioavailability of about 80 percent. It is metabolised mainly by the liver enzyme CYP2C9.

What the research says

The evidence is strongest for what torsemide does to the kidney and weakest for whether choosing it over the older loop diuretic furosemide changes how long people live. It makes the kidney pass more sodium, chloride and water, which relieves the fluid overload of heart failure, kidney disease and liver cirrhosis, and it lowers blood pressure. The largest randomised trial ever run on this question, TRANSFORM-HF in 2,859 people hospitalised with heart failure, found 26.1 percent died in the torsemide group and 26.2 percent in the furosemide group. Meta-analyses of smaller trials do suggest fewer heart-failure hospitalisations with torsemide, but they agree with the big trial that mortality is no different, and a mechanistic substudy found no pharmacokinetic advantage at all.

Evidence grade: Well established.

How it works

Drug class: Loop diuretic of the pyridine-sulfonylurea class

Torsemide works from inside the tubule of the loop of Henle in the kidney, blocking the transporter that reabsorbs sodium, potassium and chloride. The result is that more sodium, chloride and water leave the body in the urine, without the drug itself changing how fast the kidney filters blood. Lower body fluid volume means less congestion in the lungs and legs, and lower blood pressure. (Source 1)

What it is used for

• Torsemide reliably increases sodium and water excretion and is approved for the oedema of heart failure. It has not been shown to reduce death compared with furosemide; the largest trial found 26.1 percent versus 26.2 percent mortality. Evidence: established. (Source 2)
• Approved on the basis of its diuretic effect. The label notes that in renal failure a smaller fraction of each dose reaches the site of action, and we found no outcome trial in this population. Evidence: limited. (Source 2)
• Approved, with the label directing lower starting doses given together with an aldosterone antagonist or potassium-sparing diuretic. The supporting trials are small and from the 1990s. Evidence: limited. (Source 2)
• Torsemide lowers blood pressure, and the label reasons from the wider antihypertensive literature that lowering blood pressure reduces cardiovascular events. The label itself states plainly that no controlled trial has shown risk reduction with torsemide. Evidence: limited. (Source 2)
• A 118-woman randomised placebo-controlled trial of 20 mg daily for five days found 44 percent versus 58 percent persistent postpartum hypertension, a difference that did not reach statistical significance (relative risk 0.76, 95% CI 0.5-1.1). Evidence: not-supported. (Source 3)
• This is a prescribing choice rather than a separate licensed indication. TRANSFORM-HF tested it directly and found no mortality difference, and the TRANSFORM-Mechanism substudy found no pharmacokinetic or pharmacodynamic advantage. Evidence: not-supported. (Source 4)

Interactions

• NSAIDs such as ibuprofen and naproxen (label): Taking an NSAID with torsemide has been linked to acute kidney failure, and NSAIDs blunt both the diuretic and the blood-pressure-lowering effect. (Source 5)
• High-dose aspirin and other salicylates (label): Torsemide and salicylates compete for the same secretion route in the kidney tubule, so people on high salicylate doses may develop salicylate toxicity. (Source 5)
• Amiodarone, fluconazole, miconazole (CYP2C9 inhibitors) (label): These slow torsemide's breakdown and raise its blood level, which can increase both diuresis and side effects. (Source 5)
• Warfarin and phenytoin (case reports): The case report says only that torsemide shares metabolic pathways with warfarin, which could allow a clinically important interaction, and that the evidence for it is limited and inconsistent. In this single case the INR rose and bleeding risk increased after torsemide was started, and the INR settled once torsemide was replaced by furosemide. No enzyme mechanism was measured or named. (Source 6)
• Lithium (label): Like other diuretics, torsemide reduces how fast the kidney clears lithium, creating a high risk of lithium toxicity. (Source 7)
• Cholestyramine (a bile-acid-binding powder) (theoretical): In animals, taking cholestyramine at the same time reduced how much oral torsemide was absorbed; the label says to separate the doses by at least an hour before, or four to six hours after, cholestyramine. The combination has not been studied in humans. (Source 7)
• Aminoglycoside antibiotics such as gentamicin (case reports): Loop diuretics add to the hearing toxicity of these antibiotics; this has been reported with torsemide and gentamicin specifically. (Source 7)
• Corticosteroids and ACTH (label): Used together with torsemide these may increase the risk of low blood potassium. (Source 7)
• Potassium, magnesium and calcium (the minerals people most often supplement) (label): Torsemide can lower blood potassium, sodium, magnesium and calcium. Anyone taking these as supplements, or being monitored for low levels, is dealing with the same chemistry the drug acts on; the label directs periodic electrolyte monitoring rather than routine supplementation. (Source 8)
• Thiamine (vitamin B1) (case reports): Thiamine deficiency has been reported during post-approval use of torsemide. These are spontaneous reports, so a rate cannot be calculated and cause is not established. (Source 9)
• Food (pharmacokinetic study): Taking torsemide with food delays the peak blood level by about 30 minutes but does not change total absorption or diuretic effect, so it can be taken with or without food. (Source 10)

Stopping it

• Torsemide is not a drug of dependence and no withdrawal syndrome is described for it. What stopping risks is the return of fluid overload. The ReBIC-1 randomised double-blind trial withdrew low-dose furosemide in 188 stable heart-failure outpatients with no fluid retention and found no increase in breathlessness and no significant increase in restarting the diuretic (75.3% versus 83.7% free of reuse; odds ratio 1.69, 95% CI 0.82-3.49). (Source 11)
• Those authors concluded that stopping the diuretic may deserve consideration specifically in stable outpatients with no signs of fluid retention who are on optimal medical therapy. The trial tested furosemide, not torsemide, and ran for 90 days. (Source 12)
• The label's own stopping instruction sits in its patient counselling section and is narrower and symptom-driven: patients are told that if fainting occurs, torsemide should be discontinued until the prescriber has been consulted. (Source 13)

What goes wrong

In that substudy the higher torsemide doses clinicians actually used produced more sodium loss but also more neurohormonal activation and kidney dysfunction, with no gain in fluid status. (Source 14)

• Blood level study, Moderate certainty.
• Size: 88 participants.
• Who: Patients with heart failure.
• How long: 30 days.
• Result: Significant increases in renin, aldosterone and norepinephrine (P < 0.05 for all) with the higher torsemide doses; plasma volume (P = 0.52) and body weight (P = 0.89) did not improve versus furosemide.
• Funding: not stated in the record fetched.

Higher diuretic doses in the torsemide group resulted in mild perturbations in kidney function and significant increases in renin, aldosterone, and norepinephrine (P < 0.05 for all).

In placebo-controlled trials, excessive urination was the adverse effect clearly attributable to torsemide, and it was dose related. (Source 15)

• Official position, Certainty not rated.
• Size: 564 torsemide and 274 placebo recipients in US-based trials (about 4,000 subjects in all pre-approval studies)
• Who: Patients with hypertension, heart failure, renal or hepatic disease.
• How long: 1 to 52 days, median 41 days.
• Result: Excessive urination 6.7% torsemide versus 2.2% placebo; in hypertension studies 1% placebo, 4% on 5 mg daily, 15% on 10 mg daily.
• Funding: US regulatory document.

In United States placebo-controlled trials excessive urination occurred in 6.7% of patients compared with 2.2% of patients receiving placebo.

Torsemide can cause clinically important electrolyte disturbances and raise blood glucose. (Source 8)

• Official position, Certainty not rated.
• Size: Not quantified in this section.
• Who: Patients taking torsemide.
• How long: Ongoing therapy.
• Result: Hypokalaemia, hyponatraemia, hypomagnesaemia, hypocalcaemia and hypochloraemic alkalosis described qualitatively; in hypertension trials at 5-10 mg daily the mean fall in serum potassium was about 0.1 mEq/L and 1.5% on torsemide versus 3% on placebo had a potassium below 3.5 mEq/L at any time.
• Funding: US regulatory document.

Torsemide can cause potentially symptomatic hypokalemia, hyponatremia, hypomagnesemia, hypocalcemia, and hypochloremic alkalosis.

Excessive diuresis with torsemide can cause dehydration, low blood pressure and worsening kidney function, including acute kidney failure. (Source 8)

• Official position, Certainty not rated.
• Size: Not quantified.
• Who: Patients taking torsemide, especially if salt-depleted or on renin-angiotensin-aldosterone blockers.
• How long: Ongoing therapy.
• Result: No rates given in this section; the label directs periodic monitoring of volume status and renal function.
• Funding: US regulatory document.

Excessive diuresis may cause potentially symptomatic dehydration, blood volume reduction and hypotension and worsening renal function, including acute renal failure particularly in salt-depleted patients or those taking renin-angiotensin aldosterone inhibitors.

Tinnitus and hearing loss have been seen with torsemide, as with other loop diuretics. (Source 8)

• Official position, Certainty not rated.
• Size: Not quantified.
• Who: Patients on loop diuretics, higher risk with high doses, severe renal impairment and low blood protein.
• How long: Ongoing therapy.
• Result: Described as usually reversible; no rate given.
• Funding: US regulatory document.

Tinnitus and hearing loss (usually reversible) have been observed with loop diuretics, including torsemide.

Severe skin reactions including Stevens-Johnson syndrome and toxic epidermal necrolysis have been reported after approval, alongside pancreatitis, blood count abnormalities and thiamine deficiency. (Source 9)

• Official position, Very low certainty.
• Size: Spontaneous reports from a population of uncertain size.
• Who: Patients using torsemide after approval.
• How long: Post-approval surveillance, no fixed duration.
• Result: No rates can be calculated; the label itself says frequency cannot be estimated reliably and causality cannot be established.
• Funding: US regulatory document.

Skin/hypersensitivity: Stevens-Johnson syndrome, toxic epidermal necrolysis, photosensitivity reaction, pruritus

A 2026 case report describes a rising INR and bleeding risk when torsemide was started in a man on warfarin, which settled after torsemide was swapped for furosemide. (Source 6)

• Case report, Very low certainty.
• Size: 1 patient.
• Who: 66-year-old man with type 2 diabetes, hypertension, chronic kidney disease and a prior aortic valve replacement on chronic warfarin.
• How long: Single hospital admission.
• Result: Progressive INR elevation needing repeated fresh frozen plasma; Naranjo scale assessed as a possible interaction.
• Funding: not stated in the record fetched.

Limit of this finding: This is one patient, and the authors' own rating on the Naranjo scale is only 'possible', not 'probable' or 'definite'. They also say the evidence on a torsemide-warfarin interaction 'remains limited and inconsistent'. A single possible case does not establish that torsemide raises INR, and the paper names no enzyme mechanism.

During hospitalization, warfarin (3 mg once daily) and Torsemide (20 mg once daily) were initiated concurrently. Subsequently, the patient developed a progressive elevation in international normalized ratio (INR), necessitating repeated fresh frozen plasma transfusions due to increased bleeding risk.

What the evidence supports

A randomised double-blind withdrawal trial found that stopping a low-dose loop diuretic in stable heart failure outpatients did not increase breathlessness or the need to restart it. (Source 11)

• Randomized trial, Moderate certainty.
• Size: 188 patients (95 withdrawal, 93 maintenance) at 11 heart-failure clinics in Brazil.
• Who: Stable outpatients with mild heart failure, no fluid retention, on optimal therapy; 25% female, mean age 59, mean LVEF 32%.
• How long: 90 days.
• Result: Free of furosemide reuse: 70 patients (75.3%) in the withdrawal group and 77 patients (83.7%) in the maintenance group, as the paper reports them; odds ratio for additional furosemide use with withdrawal 1.69 (95% CI 0.82-3.49; P = 0.16). Dyspnoea co-primary endpoint, the area under the curve of a visual-analogue dyspnoea score at baseline, Day 15, Day 45 and Day 90: median AUC 1875 (IQR 383-3360) with withdrawal versus 1541 (IQR 474-3124) with maintenance; P = 0.94.
• Funding: not stated in the record fetched; PubMed indexes it as Research Support, Non-U.S. Gov't.

Limit of this finding: The paper's counts and its own percentages do not match. It randomised 95 patients to withdrawal and 93 to maintenance, but 70 patients is 75.3% of 93, not of 95, and 77 patients is 83.7% of 92, not of 93 - so the percentages appear to be calculated on smaller analysed groups than the randomised ones. Take the counts and percentages as the paper prints them; do not build a fraction out of them.

For the second co-primary endpoint, 70 patients (75.3%) in the withdrawal group and 77 patients (83.7%) in the maintenance group were free of furosemide reuse during follow-up (odds ratio for additional furosemide use with withdrawal 1.69, 95% confidence interval 0.82-3.49; P = 0.16).

What the evidence does not support

The largest randomised trial of torsemide versus furosemide after a heart-failure hospitalisation found no difference in death. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 2,859 participants randomised at 60 US hospitals (1,431 torsemide, 1,428 furosemide)
• Who: Adults hospitalised with heart failure regardless of ejection fraction.
• How long: Median follow-up 17.4 months; mortality followed to 30 months.
• Result: Death 373/1431 (26.1%) with torsemide versus 374/1428 (26.2%) with furosemide; hazard ratio 1.02 (95% CI 0.89-1.18). Absolute difference 0.1 percentage points, so no number needed to treat can be calculated.
• Funding: not stated in the record fetched; PubMed indexes it as Research Support, N.I.H., Extramural.

Death occurred in 373 of 1431 patients (26.1%) in the torsemide group and 374 of 1428 patients (26.2%) in the furosemide group (hazard ratio, 1.02 [95% CI, 0.89-1.18]).

The same trial found no difference in the combined outcome of death or any hospitalisation over 12 months. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 2,859 participants.
• Who: Adults hospitalised with heart failure.
• How long: 12 months.
• Result: Death or all-cause hospitalisation 677/1431 (47.3%) torsemide versus 704/1428 (49.3%) furosemide; hazard ratio 0.92 (95% CI 0.83-1.02). Absolute difference 2.0 percentage points, confidence interval crossing no effect.
• Funding: not stated in the record fetched; PubMed indexes it as Research Support, N.I.H., Extramural.

Over 12 months following randomization, all-cause mortality or all-cause hospitalization occurred in 677 patients (47.3%) in the torsemide group and 704 patients (49.3%) in the furosemide group (hazard ratio, 0.92 [95% CI, 0.83-1.02]).

The same meta-analysis found no significant difference in all-cause mortality or NYHA class, and reported all-cause hospitalisation as not significantly different even though its own p value was below its stated threshold. (Source 16)

• Meta-analysis, Low certainty.
• Size: 4,115 patients across 12 randomised controlled trials.
• Who: Patients with heart failure.
• How long: Varied by trial.
• Result: All-cause mortality RR 0.98 (95% CI 0.87-1.10; p=0.73; I2=0%); all-cause hospitalisation RR 0.93 (95% CI 0.86-1.00; p=0.04; I2=0%); NYHA class improvement RR 1.25 (95% CI 0.92-1.68; p=0.15; I2=0%); NYHA class change MD -0.04 (95% CI -0.24 to 0.16; p=0.70; I2=15%)
• Funding: not stated in the record fetched.

Limit of this finding: The paper contradicts itself on one of these outcomes. Its methods set significance at the 0.05 level, yet it calls all-cause hospitalisation 'no significant difference' while reporting p=0.04, which is below that threshold, with a confidence interval (0.86 to 1.00) whose upper limit sits exactly on no effect. Read all-cause hospitalisation as borderline and unresolved rather than as either a clear reduction or a clear null. The mortality and NYHA results are not affected.

There was no significant difference in all-cause hospitalizations (RR 0.93; 95% CI, 0.86-1.00; p=0.04; I2=0%), all-cause mortality (RR 0.98; 95% CI, 0.87-1.10; p=0.73; I2=0%), NYHA class improvement (RR 1.25; 95% CI, 0.92-1.68; p=0.15; I2=0%), or NYHA class change (MD -0.04; 95% CI, -0.24 to 0.16; p=0.70; I2=15%) between groups.

A mechanistic substudy of TRANSFORM-HF found torsemide delivered LESS drug to its site of action in the kidney than furosemide, contradicting the usual argument for preferring it. (Source 14)

• Blood level study, Moderate certainty.
• Size: 88 participants.
• Who: Patients with heart failure randomised to oral furosemide or torsemide.
• How long: Assessments at baseline and 30 days.
• Result: Kidney bioavailability median 17.1% (IQR 12.3%-23.5%) with torsemide versus 24.8% (16.6%-34.1%) with furosemide, P < 0.001; furosemide had the longer duration of kidney drug delivery and natriuresis (P <= 0.004 for both)
• Funding: not stated in the record fetched.

Kidney bioavailability, or the proportion of dose delivered to the tubular site of action, was significantly less with torsemide compared with furosemide (median, 17.1% [interquartile range, 12.3%-23.5%] versus 24.8% [16.6%-34.1%], P < 0.001).

Torsemide's FDA label states that no controlled trial has shown that torsemide reduces cardiovascular risk, even though it is approved for hypertension. (Source 2)

• Official position, Certainty not rated.
• Size: Not applicable.
• Who: US prescribing population.
• How long: FDA-approved label, SPL version effective 8 May 2024.
• Result: No outcome figures given; the label reasons by class from other antihypertensive drug trials.
• Funding: US regulatory document.

There are no controlled trials demonstrating risk reduction with torsemide tablets.

An off-label randomised trial of torsemide to prevent persistent high blood pressure after preeclampsia was null. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 118 women randomised (59 torsemide, 59 placebo)
• Who: Women with preeclampsia at a single tertiary centre.
• How long: 20 mg/day for 5 days, with follow-up to 6 weeks postpartum.
• Result: Persistent postpartum hypertension 44% torsemide versus 58% placebo; relative risk 0.76 (95% CI 0.5-1.1). Absolute difference 14 percentage points but the confidence interval includes no effect.
• Funding: not stated in the record fetched; PubMed indexes it as Research Support, Non-U.S. Gov't.

The rate of persistent postpartum hypertension was 44% in the torsemide and 58% in the placebo group (relative risk 0.76, 95% CI 0.5-1.1).

Where the evidence is mixed

Over a median 17.4 months, 113 of the 2,859 participants withdrew consent before the trial ended, part of the loss to follow-up the authors say limits how firmly the null result can be read. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 2,859 participants.
• Who: Adults hospitalised with heart failure.
• How long: Median 17.4 months.
• Result: 113 patients withdrew consent before completion: 53 (3.7%) in the torsemide group and 60 (4.2%) in the furosemide group. No figure for crossover or nonadherence is given in the abstract.
• Funding: not stated in the record fetched; PubMed indexes it as Research Support, N.I.H., Extramural.

Over a median follow-up of 17.4 months, a total of 113 patients (53 [3.7%] in the torsemide group and 60 [4.2%] in the furosemide group) withdrew consent from the trial prior to completion.

The trial's own authors flag that crossover between the two drugs, nonadherence and loss to follow-up limit how firmly the null result can be read. (Source 17)

• Randomized trial, Moderate certainty.
• Size: 2,859 participants.
• Who: Adults hospitalised with heart failure.
• How long: Median 17.4 months.
• Result: No numerical estimate of crossover or nonadherence is reported in the abstract; the authors state the limitation in words only.
• Funding: not stated in the record fetched; PubMed indexes it as Research Support, N.I.H., Extramural.

However, interpretation of these findings is limited by loss to follow-up and participant crossover and nonadherence.

A 2024 meta-analysis of 12 randomised trials found fewer heart-failure hospitalisations with torsemide but no difference in death. (Source 16)

• Meta-analysis, Low certainty.
• Size: 4,115 patients across 12 randomised controlled trials.
• Who: Patients with heart failure.
• How long: Varied by trial; not pooled.
• Result: Hospitalisation for heart failure RR 0.60 (95% CI 0.43-0.83; p=0.002; I2=0%); cardiovascular hospitalisation RR 0.72 (95% CI 0.60-0.88; p=0.0009; I2=0%); LVEF MD 4.51% (95% CI 2.94 to 6.07; p<0.0001; I2=0%)
• Funding: not stated in the record fetched.

Torsemide significantly reduced HHF (RR 0.60; 95% CI, 0.43-0.83; p=0.002; I2=0%), hospitalization for cardiovascular causes (RR 0.72; 95% CI, 0.60-0.88; p=0.0009; I2=0%), and improved LVEF (MD 4.51%; 95% CI, 2.94 to 6.07; p<0.0001; I2=0%) compared with furosemide.

A second independent 2023 meta-analysis reached the same split verdict: fewer hospitalisations, identical mortality. (Source 18)

• Meta-analysis, Low certainty.
• Size: 4,127 patients across 10 randomised controlled trials (2,088 furosemide, 2,039 torsemide); 56% men, mean age 68.
• Who: Patients with heart failure.
• How long: Varied by trial.
• Result: All-cause mortality RR 1.02 (95% CI 0.91 to 1.15, p = 0.70) for furosemide versus torsemide; cardiovascular hospitalisation RR 1.36 (95% CI 1.13 to 1.65, p = 0.001); heart-failure hospitalisation RR 1.65 (95% CI 1.21 to 2.24, p = 0.001); all-cause hospitalisation RR 1.06 (95% CI 1.01 to 1.11, p = 0.02)
• Funding: not stated in the record fetched.

patients treated with furosemide compared with torsemide had higher risks of cardiovascular hospitalizations (RR 1.36, 95% CI 1.13 to 1.65, p = 0.001), HF-related hospitalizations (RR 1.65, 95% CI 1.21 to 2.24, p = 0.001), and all-cause hospitalizations (RR 1.06, 95% CI 1.01 to 1.11, p = 0.02).

Overall withdrawal from treatment because of side effects was no more common with torsemide than with placebo in the registration trials. (Source 15)

• Official position, Certainty not rated.
• Size: 564 torsemide and 274 placebo recipients.
• Who: Patients in US-based pre-approval trials.
• How long: 1 to 52 days, median 41 days.
• Result: Discontinuation for adverse reactions 3.5% torsemide versus 4.4% placebo.
• Funding: US regulatory document.

Discontinuation of therapy due to adverse reactions occurred in 3.5% of United States patients treated with torsemide and in 4.4% of patients treated with placebo.

Where the research disagrees

Whether torsemide is a better loop diuretic than furosemide in heart failure

• Mentz and the TRANSFORM-HF investigators (2023), Open-label pragmatic randomised trial in 2,859 patients: Among patients discharged after hospitalization for heart failure, torsemide compared with furosemide did not result in a significant difference in all-cause mortality over 12 months. (Source 17)
• Singh and colleagues (2023), meta-analysis of 10 randomised trials, Random-effects meta-analysis of 10 RCTs, 4,127 patients: patients treated with furosemide compared with torsemide had higher risks of cardiovascular hospitalizations (RR 1.36, 95% CI 1.13 to 1.65, p = 0.001) (Source 18)
• Rao and the TRANSFORM-Mechanism investigators (2025), Mechanistic pharmacokinetic/pharmacodynamic substudy in 88 patients: We observed no meaningful pharmacokinetic/pharmacodynamic advantages for torsemide versus furosemide. (Source 14)

Whether torsemide's approval for hypertension implies it reduces cardiovascular events

• The FDA-approved label (SPL effective 8 May 2024), Regulatory position reasoning by class from other antihypertensive trials: There are no controlled trials demonstrating risk reduction with torsemide tablets. (Source 2)

How much

• Reference intake: Dosing is set by the prescriber, not by a reference intake. The FDA label's starting doses (SPL effective 8 May 2024) are 10 mg or 20 mg once daily for oedema of heart failure, 20 mg once daily for oedema of chronic renal failure, 5 mg or 10 mg once daily for oedema of hepatic cirrhosis given with an aldosterone antagonist or potassium-sparing diuretic, and 5 mg once daily for hypertension. (Source 19)
• Upper limit: The label does not set a maximum dose but states that doses above 200 mg have not been adequately studied, and above 40 mg in hepatic cirrhosis. For hypertension the label stops at 10 mg once daily before adding another drug. (Source 19)
• Studied: In the US placebo-controlled trials behind approval, daily doses ranged from 1.25 mg to 20 mg, with most patients receiving 5 mg to 10 mg, for 1 to 52 days (median 41 days). (Source 15)
• Studied: The off-label postpartum hypertension trial gave oral torsemide 20 mg per day for 5 days starting within 24 hours of delivery. (Source 3)
• Studied: In the TRANSFORM-Mechanism substudy clinicians on average used roughly a 2:1 furosemide-to-torsemide dose conversion, while about 4:1 produced equal sodium loss. (Source 14)

A common belief, and what the research shows

The belief: Torsemide is simply the better, more reliable loop diuretic, so switching from furosemide should help you live longer.

What the research shows: The reliability claim rests on pharmacology, not outcomes, and when it was tested directly it did not hold. TRANSFORM-HF randomised 2,859 people and reported that 'Death occurred in 373 of 1431 patients (26.1%) in the torsemide group and 374 of 1428 patients (26.2%) in the furosemide group (hazard ratio, 1.02 [95% CI, 0.89-1.18]).' The mechanistic substudy went further and found the opposite of the expected pharmacokinetic advantage: 'Kidney bioavailability, or the proportion of dose delivered to the tubular site of action, was significantly less with torsemide compared with furosemide'. Meta-analyses of smaller trials do find fewer heart-failure hospitalisations with torsemide, but every one of them agrees with the big trial that mortality is unchanged.

Questions and answers

What is it?

Torsemide is a prescription loop diuretic, a water tablet. Its label describes it as a diuretic of the pyridine-sulfonylurea class, a white to off-white crystalline powder with a molecular weight of 348.43, pressed into tablets. It is sold as a generic and under the brand names Demadex and Soaanz. (Source 20)

What does it do in the body?

It acts inside the loop of Henle in the kidney, blocking the transporter that pulls sodium, potassium and chloride back into the body. More sodium, chloride and water leave in the urine. Importantly, it does not change how fast the kidney filters blood or the body's acid-base balance. Clinical pharmacology studies confirmed this site of action in humans. (Source 1)

Is it good or bad for you?

It depends entirely on whether you are carrying excess fluid. In heart failure, kidney disease or liver cirrhosis with oedema it relieves congestion, and in the registration trials people stopped it for side effects no more often than placebo (3.5% versus 4.4%). It is harmful when diuresis goes too far: the label warns of dehydration, low blood pressure and worsening kidney function including acute kidney failure, particularly in people who are salt-depleted or taking renin-angiotensin-aldosterone blockers. (Source 8)

How do you get more of it?

It is a prescription-only medicine, so the only route is a prescriber increasing the dose. The label's approach is to start at 10 mg or 20 mg once daily for heart-failure oedema and roughly double the dose until the wanted diuretic response is reached; doses above 200 mg have not been adequately studied. Nothing in food or supplements raises torsemide levels, and taking it with food only delays the peak by about 30 minutes. (Source 19)

If it is harmful, what reduces it?

Torsemide is a short-acting tablet rather than something that accumulates, so the question is usually about stopping it rather than clearing it. The label's immediate instruction is that if fainting occurs torsemide should be discontinued until the prescriber has been consulted. ReBIC-1 then showed that in stable patients with no fluid retention a loop diuretic can be withdrawn without a rise in breathlessness and without most people needing to restart it. Taking cholestyramine at the same time reduced absorption of an oral dose in animals, which is a documented way of blunting it, not a recommendation. (Source 12)

Why might someone be low in it or missing it?

The question for a drug is why it might stop working rather than why you are low in it. The label gives several documented reasons: CYP2C9 inducers such as rifampin increase torsemide clearance and lower its blood level; NSAIDs partly block its sodium-losing effect, shown with indomethacin under dietary sodium restriction but not at normal sodium intake; probenecid and other organic anion drugs reduce its secretion into the tubule where it acts; and cholestyramine taken at the same time reduced absorption of an oral dose in animals. (Source 5)

Which whole foods contain it or feed it?

No whole food contains torsemide; it is a synthetic sulfonylurea-class molecule made in a factory, supplied as a crystalline powder in tablets with crospovidone, lactose monohydrate, magnesium stearate, microcrystalline cellulose and povidone. The food question that does matter is sodium intake, because the drug's job is to remove sodium, and whether food is in the stomach, which delays the peak level by about 30 minutes without changing the overall effect. (Source 20)

What happens if you do not have it?

If someone with fluid overload is not treated, the congestion stays. But ReBIC-1 tested the opposite case directly: in 188 stable heart-failure outpatients with no fluid retention who were on optimal therapy, withdrawing a low-dose loop diuretic did not increase breathlessness and 75.3 percent never needed to restart it, versus 83.7 percent in the group who kept taking it. So being without a loop diuretic is not automatically harmful; it depends on whether fluid is accumulating. (Source 11)

How can you test for it?

There is no routine blood test for torsemide itself, and therapeutic drug monitoring is not used. What is monitored is the drug's effect: the label directs monitoring serum electrolytes and blood glucose periodically, and monitoring volume status and renal function periodically. Those tests are reliable and widely available, but they measure consequences rather than adherence or drug level. The label also notes that asymptomatic high uric acid can occur and gout may rarely be precipitated, so a raised urate on a routine panel is an expected finding rather than a surprise. (Source 8)

References

1. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 12.1 Mechanism of Action (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
2. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 1 INDICATIONS AND USAGE (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
3. Obstetrics and gynecology. Torsemide for Prevention of Persistent Postpartum Hypertension in Women With Preeclampsia: A Randomized Controlled Trial.. 2018. PMID 30303905, DOI 10.1097/AOG.0000000000002941. Read the source
4. JAMA. Effect of Torsemide vs Furosemide After Discharge on All-Cause Mortality in Patients Hospitalized With Heart Failure: The TRANSFORM-HF Randomized Clinical Trial. (Results). 2023. PMID 36648467, DOI 10.1001/jama.2022.23924. Read the source
5. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Sections 7.1-7.2 Drug Interactions (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
6. Indian journal of pharmacology. Torsemide and warfarin: A cautionary case of altered international normalized ratio and bleeding risk.. 2026. PMID 42060815, DOI 10.4103/ijp.ijp_1300_25. Read the source
7. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Sections 7.3-7.9 Drug Interactions (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
8. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 5 WARNINGS AND PRECAUTIONS (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
9. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 6.2 Postmarketing Experience (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
10. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 12.3 Pharmacokinetics (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
11. European heart journal. Short-term diuretic withdrawal in stable outpatients with mild heart failure and no fluid retention receiving optimal therapy: a double-blind, multicentre, randomized trial. (Methods and results). 2019. PMID 31424503, DOI 10.1093/eurheartj/ehz554. Read the source
12. European heart journal. Short-term diuretic withdrawal in stable outpatients with mild heart failure and no fluid retention receiving optimal therapy: a double-blind, multicentre, randomized trial. (Conclusions). 2019. PMID 31424503, DOI 10.1093/eurheartj/ehz554. Read the source
13. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 17 PATIENT COUNSELING INFORMATION (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
14. Journal of the American Society of Nephrology : JASN. Mechanistic Differences between Torsemide and Furosemide.. 2025. PMID 39196651, DOI 10.1681/ASN.0000000000000481. Read the source
15. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 6.1 Clinical Trials Experience (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
16. Arquivos brasileiros de cardiologia. Torsemide versus Furosemide in the Treatment of Heart Failure: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.. 2024. PMID 39046046, DOI 10.36660/abc.20230825. Read the source
17. JAMA. Effect of Torsemide vs Furosemide After Discharge on All-Cause Mortality in Patients Hospitalized With Heart Failure: The TRANSFORM-HF Randomized Clinical Trial. (Conclusions and relevance). 2023. PMID 36648467, DOI 10.1001/jama.2022.23924. Read the source
18. The American journal of cardiology. Effect of Furosemide Versus Torsemide on Hospitalizations and Mortality in Patients With Heart Failure: A Meta-Analysis of Randomized Controlled Trials.. 2023. PMID 37677884, DOI 10.1016/j.amjcard.2023.08.079. Read the source
19. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 2 DOSAGE AND ADMINISTRATION (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source
20. DailyMed, U.S. National Library of Medicine. Torsemide tablets, USP - FDA prescribing information, Section 11 DESCRIPTION (Aurobindo Pharma Limited, SPL effective 2024-05-08). 2024. Read the source