As mentioned above, inhibition of the carbonic anhydrase enzyme reduces the concentration of H+ and HCO3-, and leads to the excretion of Na+ and water. Acetazolamide is a diuretic capable of inhibiting carbonic anhydrase, and its use for decongestion was tested in a study called ADVOR.⁵

The ADVOR study was carried out to evaluate whether acetazolamide associated with a loop diuretic increases the chance of effective decongestion in patients hospitalized for HF. The multicenter study randomized 519 patients with acute decompensated HF to receive intravenous acetazolamide 500 mg/day or placebo, in addition to the loop diuretic. The primary outcome was an improvement in congestion assessed by a composite score of the presence and intensity of peripheral edema, pleural effusion, and ascites after 72 hours, which was achieved in 42.2% of the patients in the acetazolamide group and 30.5% of patients in the placebo group (p<0.001). The acetazolamide group had a lower rate of residual congestion and a shorter length of stay, without significantly increasing the incidence of adverse events and with no impact on outcomes of death or re-hospitalization.⁵ It is worth noting that patients using iSGLT-2 were excluded. and to date, there are no studies that contemplate the combined blockage of the proximal convoluted tubule (acetazolamide + iSGLT-2). With the introduction of these as disease-modifying drugs in HF, more studies should be carried out to evaluate the effect of the association with other diuretics, such as acetazolamide.

Sodium glucose linked transporter type 2 (SGLT-2) is responsible for the reabsorption of sodium and glucose in the proximal tubule of the nephron. Thus, its inhibition can lead to glycosuria and increased diuresis and natriuresis. However, the potential for iSGLT-2 to induce these effects is related to the level of serum glycemia, which is why patients tend to present a low risk of hypoglycemia. iSGLT-2s also reduce sodium reabsorption by inhibiting the sodium-hydrogen cotransporter (NHE3) in the proximal convoluted tubule.¹¹

In addition to these mechanisms, iSGLT-2 has also demonstrated robust benefits in the treatment of heart failure with reduced ejection fraction (HFrEF), in terms of reducing hospitalizations and mortality, in important studies such as DAPA-HF (dapaglifozin) and EMPEROR-reduced (empaglifozin). In the setting of refractory congestion, initial studies demonstrated a limited power of these drugs in promoting diuresis and natriuresis;¹¹ however, more recently new and larger studies have been carried out to evaluate their potential to improve diuretic efficacy. The main studies were: EMPULSE, EMPAG-HF, EMPA RESPONSE-AHF, DAPA RESIST, and DICTATE-AHF.

The EMPULSE study randomized 530 patients into a placebo group versus empaglifozin 10 mg/day in patients with acute HF, regardless of ejection fraction, 24 hours after admission. The primary outcome was a clinical benefit defined by the set of time to death, number, and time to occurrence of a HF event or 5-point variation in the Kansas Quality of Life Questionnaire (KCCQ) at 90 days. The outcome occurred in 53.9% of the empaglifozin group and 39.7% of the placebo group (p= 0.0054). Regarding the treatment of congestion, a more significant weight loss was observed in the empaglifozin group (-1.5 kg) when compared to placebo (p=0.014).¹⁶

The EMPAG-HF study randomized 60 patients with acute decompensated HF within the first 12 hours of hospitalization, who received 25 mg/day of empaglifozin versus the placebo, in addition to standard loop diuretic therapies. The primary outcome was cumulative diuresis over a five-day period. The addition of empaglifozin to the standard therapy resulted in a 25% increase in cumulative urine output over a five-day period (group difference 2.2 L; p=0.003), with no difference in weight change. The daily and cumulative dose of loop diuretics was lower in the empaglifozin group. iSGLT2 increased diuretic efficiency without altering markers of kidney function and injury, reduced the N-terminus of the B-type natriuretic peptide prohormone (NT-proBNP), and proved to be a safe and effective strategy to aid in conventional decongestive treatment.¹⁷

Also with empaglifozin, the EMPA RESPONSE AHA study randomized 80 patients with acute HF to receive either 10 mg/day of empaglifozin or the placebo for 30 days. The primary outcome consisted of changes in dyspnea assessed by the visual analogue scale, diuresis, variation in NT-proBNP, or length of hospital stay, and showed no difference between the groups. Despite this, in the empaglifozin group, there was an increase in urinary output and a decrease in the combined outcome of worsening HF, re-hospitalization for HF, or death within 60 days, with no increase in adverse effects in relation to placebo.¹⁸

Dapaglifozin was tested in a multicenter study called DAPA RESIST involving 61 patients and aimed to compare the decongestion effect promoted by dapaglifozin versus metolazone in patients with acute decompensated HF with diuretic resistance. Patients must have insufficient decongestion (defined by loss <1 kg or a water balance <1 liter after 24 hours, despite treatment with furosemide ≥160 mg/day or equivalent) associated with brain natriuretic peptide (BNP) ≥100 pg/ml or NT- proBNP ≥400 pg/ml, persistence of congestion (peripheral edema, ascites, increased central venous pressure, pulmonary congestion in chest radiography or ultrasound), and the prospect of hospitalization for more than 3 days. Subjects were randomized into two groups: 10 mg/day of dapaglifozin and 5 or 10 mg/day of metolazone for up to 3 consecutive days. No difference in the primary outcome (weight change) was found between the groups. The dapaglifozin group showed lower diuretic efficacy (characterized by the need for higher doses of furosemide) and a lower incidence of hyponatremia and hypokalemia.¹⁹

Recently, the DICTATE-AHF study randomized 240 patients hospitalized for acute decompensated HF with signs of hypervolemia in the first 24 hours of hospitalization to dapaglifozin 10 mg/day or placebo in combination with a loop diuretic. The primary outcome was diuretic efficacy (defined as weight change over cumulative dose of loop diuretic), showing no difference between groups. Despite this, early initiation of dapaglifozin demonstrated an increase in natriuresis and diuresis and a reduction in the necessary amount of loop diuretics and length of hospital stay in relation to placebo.²⁰

The loop of Henle is responsible for the reabsorption of approximately 25% of the sodium. The proximal convoluted tubule reabsorbs around 70% of the filtrate; however, the agents that act on these sites (acetazolamide and iSGLT-2) have failed to demonstrate any major capacity to promote sodium and water loss when used alone, given that a large part of this filtrate is reabsorbed at a later moment in the loop of Henle.⁹

Loop diuretics act by inhibiting the NA-K-Cl pump in the ascending segment of the loop of Henle, resulting in a greater excretion of sodium, chlorine, and potassium. These are transported by proteins, such as albumin, and secreted into the proximal convoluted tubule.⁹

In 2011, one of the most important studies with loop diuretics was carried out, DOSE, which tested the effectiveness of using loop diuretics in 308 patients with decompensated HFrEF in relation to dose (high x low) and infusion regimen (bolus x infusion continuous) in a 1:1:1:1 randomization. A low dose was defined as an intravenous dose equivalent to the patient's oral dose and a high dose as an intravenous dose equivalent to 2.5 times the previous oral dose. The primary outcomes were clinical improvement on a dyspnea scale and an increase in creatinine after 72 hours of treatment. Comparing the “bolus” group with the “continuous infusion” group, no statistical significance was found when considering the efficacy and safety outcomes. The comparison of the “high dose” versus “low dose” groups suggested that, although no difference was demonstrated between the groups, the use of high doses of furosemide was able to promote a greater relief of symptoms, increased diuresis, and a higher loss of weight, at the expense of a greater increase in creatinine when compared to the low dose group.²¹

Torsemide differs from other loop diuretics, such as furosemide, in that it has a longer half-life and duration of action with less variability in oral bioavailability, suggesting a greater diuretic effect. To test the hypothesis of superiority of torsemide in relation to furosemide, the TORIC study was published in 2002, an open cohort, with 1,377 patients, classified by the New York Heart Association (NYHA) II and III, who received 10 mg/day of torsemide orally versus 40 mg/day of furosemide or other oral diuretics, for 12 months, in addition to standard HF therapy. This study showed that torsemide is safe and, although it was not designed to evaluate outcomes, such as mortality, it suggested a lower incidence in patients treated with torsemide. Patients showed functional improvement (NYHA) and a lower incidence of hypokalemia.²²

In 2022, TRANSFORM-HF was published, with 2,859 patients, designed to determine whether torsemide would result in a reduction in mortality compared to furosemide in patients hospitalized for HF. Although previous studies and meta-analysis suggest advantages of using torsemide over furosemide, this study demonstrated no benefit in relation to mortality or hospitalization due to HF.²³ However, some questions should be considered, such as the non-adherence of participants, crossover between groups, discontinuation of the use of diuretics during follow-up, and the use of new therapies, such as INRA (neprilysin and angiotensin receptor inhibitors) and iSGLT2, as they potentially reduce diuretic needs.

Thiazide diuretics inhibit sodium re-uptake in the distal convoluted tubule by blocking the NaCl co-transporter. In healthy individuals, this segment is responsible for only 5% of sodium reabsorption. However, in individuals with a chronic use of loop diuretics, compensatory mechanisms can lead to hypertrophy of the tubular cells of the distal nephron, generating an increase in sodium reabsorption and a secondary decline in natriuresis.⁹ Thiazides bind to proteins, requiring an adequate renal flow so that they can be secreted into the tubules and their effect may be reduced in the presence of severe renal dysfunction. These can cause significant hypokalemia, given that, for every sodium ion lost, 2-3 potassium ions are excreted. Metolazone, chlorthalidone, and hydrochlorothiazide represent this class of medications. Chlorthalidone has slower gastrointestinal absorption, with a long half-life (24 to 72 hours) compared to hydrochlorothiazide (6 to 12 hours). Metolazone is more potent than hydrochlorothiazide and maintains its action even when there is a severe decrease in the glomerular filtration rate.¹³

In 2023, the CLOROTIC trial was published with the aim of evaluating the addition of hydrochlorothiazide to furosemide as a strategy for treating congestion. The study randomized 230 patients, in functional class (FC) III and IV, with a full spectrum of ejection fraction and high levels of natriuretic peptides. The primary outcome was weight change and improvement in patient-reported dyspnea within 72 hours. The thiazide group showed greater weight loss (-1.5 kg, p = 0.002) and increased diuresis and natriuresis when compared to the placebo, with no difference in patient perception regarding improvement in dyspnea, hospitalization rate, or mortality. In the safety outcome, there was a higher rate of increase in creatinine in the thiazide group (46.5% x 17.2%; p<0.001), but no difference was found in hypokalemia and hyponatremia. In a post-hoc analysis using potassium levels ≤3.5 and ≤3.0 mmoL/L, hypokalemia was more frequent with thiazides.²³ The similar prevalence of hypotension between groups is important, as this is a frequent concern in the management of patients with acute HF. CLOROTIC also showed no differences in the length of stay.²⁴

Spironolactone, eplerenone, and finerenone are mineralocorticoid receptor antagonists (MRAs). They work by blocking the action of aldosterone, a hormone that stimulates sodium reabsorption and potassium excretion in the distal nephron. For many years, it was believed that medications that inhibit the RAAS, such as angiotensin-converting enzyme inhibitors (ACEIs), would be sufficient to suppress the production of aldosterone. Therefore, it would not be necessary to combine these medications with an MRA, which would increase the risk of hyperkalemia. However, recent data have shown that aldosterone inhibition by ACE inhibitors is only transient. The first study then emerged to test the effect of spironolactone, an MRA, on morbidity and mortality in patients with severe HF. RALES (1999) randomized 822 patients, NYHA III-IV, left ventricular ejection fraction (LVEF) < 35%, to 25 mg/day of spironolactone versus the placebo, with a primary outcome of death from all causes. This study was interrupted early due to evidence of superiority of spironolactone.²⁵ Therefore, for the first time, a diuretic reduced the risk of morbidity and mortality in patients with severe HF, becoming a class I recommendation as a disease-modifying agent, most likely due to cardioprotective and non-diuretic mechanisms.

However, with a focus on treating congestion, the ATHENA-HF study was conducted to test the incremental efficacy of high doses of MRA, in addition to standard loop diuretic therapy, in the treatment of acute HF. Therapy with a dose of 100 mg of spironolactone daily proved to be no better than a dose of 25 mg in reducing NT-proBNP, increasing diuresis after 96 hours, reducing weight, and improving symptoms or congestion scores. High doses of MRA were safe, not triggering hyperkalemia or a worsening of renal function.²⁶ In addition to the benefit in clinical outcomes, MRAs are useful to compensate for hypokalemia, resulting from the use of loop diuretics and thiazides, maintaining an electrolyte balance.

Vasopressin V2 receptor antagonists (aquaretics) act by blocking these receptors and inhibiting water absorption by the distal nephron. They are effective in treating hyponatremia and can also be used to reduce volume overload in patients with HF. Nevertheless, the EVEREST study showed no decrease in morbidity and mortality in acute HF due to tolvaptan.²⁷ This class of drugs is not routinely used to treat congestion in patients with HF.

Based on the hypothesis that ultrafiltration (UF) provides benefits, such as greater control over decongestion and reduced neuroendocrine activation, studies have evaluated the real benefit and safety of this technique in patients with HF. The two largest studies, UNLOAD and CARRESS HF, present conflicting results.

The UNLOAD trial in 2006 evaluated 200 patients with acute HF with signs of congestion. Patients were randomized to receive UF or intravenous diuretics. The primary outcome assessed weight loss and dyspnea 48 hours after randomization. The trial showed that UF was safe and effective in reducing weight and dyspnea after 90 days, and the UF group presented lower rates of readmission and a worsening of HF.³¹

The CARRESS HF study randomized 2,033 patients, comparing UF with a fixed removal of 200 ml/h to goal-based pharmacological therapy for decongestion in patients with congested HF and a worsening renal function. The trial failed to demonstrate the superiority of UF over the use of diuretics and was associated with a higher rate of adverse events.³² Ultrafiltration is not without risks and can present complications, including infection, hemorrhage, electrolyte disturbances, anemia, and thrombocytopenia.

Currently, ultrafiltration is used as a rescue therapy to relieve congestion in refractory cases or with a significant worsening of renal function during decongestion, when dialysis is necessary. The PURE HF study should bring more answers regarding UF, as it serves to evaluate whether or not peripheral veno-venous ultrafiltration complementary to low-dose diuretics is effective in reducing outcomes, such as mortality and hospitalization for HF, within 90 days after randomization when compared to isolated intravenous diuretics.