Library › Pharmacology › SGLT2 inhibitors

SGLT2 inhibitors in heart failure, kidney disease, and diabetes

Sodium-glucose cotransporter 2 inhibitors from renal glucose physiology and mechanism through the heart failure, kidney, and diabetes outcome trials, labeled indications and doses, adverse effects, perioperative and sick-day management, and special populations.

Renal glucose handling and the SGLT2 transporter

The glomeruli filter about 180 g of glucose per day at a normal glomerular filtration rate (GFR) and plasma glucose. In health virtually all of it is reabsorbed in the proximal tubule, and urine is free of glucose. Two sodium-coupled transporters perform the reabsorption. SGLT2, encoded by SLC5A2, is a high-capacity, low-affinity transporter in the S1 and S2 segments of the proximal convoluted tubule. It couples one sodium ion to one glucose molecule and reabsorbs about 90% of the filtered glucose. Sodium-glucose cotransporter 1 (SGLT1), encoded by SLC5A1, is a low-capacity, high-affinity transporter in the S3 segment. It couples two sodium ions to each glucose molecule and reabsorbs the remainder. Glucose leaves the tubular cell across the basolateral membrane through the facilitative glucose transporter 2 (GLUT2). The sodium gradient that drives both cotransporters is maintained by the basolateral sodium-potassium ATPase, so glucose reabsorption consumes a substantial share of proximal tubular oxygen.

Reabsorption is saturable. The maximal tubular transport capacity for glucose is about 375 mg/min, and glucose appears in the urine when plasma glucose exceeds a renal threshold of about 180 mg/dL. In type 2 diabetes, SGLT2 expression and the transport maximum rise, which raises the threshold and returns more glucose to the circulation at the time it is least needed. SGLT1 is also the principal glucose and galactose transporter of the small intestine, where it governs postprandial glucose absorption, and it is expressed at lower levels in the heart and skeletal muscle. SGLT2 is expressed almost exclusively in the kidney. Whether cardiomyocytes express SGLT2 protein is uncertain, and most studies find little or none.

Two findings guided development of the drug class. Phlorizin, a glucoside isolated from apple tree bark in the nineteenth century, inhibits both SGLT1 and SGLT2 and produces glycosuria; its poor oral bioavailability and intestinal SGLT1 inhibition made it unusable as a drug. Familial renal glucosuria, caused by loss-of-function variants in SLC5A2, produces persistent glycosuria with normal plasma glucose and a benign course, which indicated that lifelong SGLT2 inhibition is tolerated. The C-glucoside SGLT2 inhibitors were then designed for oral absorption, a once-daily half-life, and high selectivity for SGLT2 over SGLT1.

SGLT2 inhibition lowers the renal glucose threshold so that glycosuria occurs at normal plasma glucose. A typical dose causes excretion of about 60 to 80 g of glucose per day in a patient with normal kidney function, a loss of roughly 240 to 320 kcal. Only about half of the filtered glucose is excreted, because increased delivery to the S3 segment is reabsorbed by SGLT1. Glycosuria is self-limiting: as plasma glucose falls, the filtered load falls, and the drug causes no hypoglycemia on its own.

Glucose lowering and its dependence on kidney function

Glycosuria is the product of the filtered glucose load and the fraction left unreabsorbed. The filtered load equals GFR multiplied by plasma glucose, so the glucose-lowering effect declines in proportion to kidney function. With preserved eGFR, SGLT2 inhibitors lower glycated hemoglobin (HbA1c) by about 0.5 to 1.0 percentage points relative to placebo, more when baseline HbA1c is high. At an eGFR of 30 to 45 mL/min/1.73 m², the reduction is small, and below 30 it is negligible. For this reason, the US labels set glycemic thresholds higher than the thresholds for cardiorenal indications. Dapagliflozin and ertugliflozin are not recommended for glycemic control below an eGFR of 45, and empagliflozin, canagliflozin, and bexagliflozin are not recommended for glycemic control below 30. A patient with type 2 diabetes and an eGFR of 30 who takes an SGLT2 inhibitor for kidney or heart protection usually needs another drug for glucose control.

The natriuretic, hemodynamic, and kidney-protective effects are preserved at low eGFR. In DAPA-CKD, EMPA-KIDNEY, and the heart failure trials, relative risk reductions for cardiovascular and kidney outcomes were similar above and below an eGFR of 30 to 45. The dissociation between glycemic and cardiorenal effects is the strongest clinical evidence that the outcome benefit is not mediated by glucose lowering. In the trials enrolling patients with type 2 diabetes, the HbA1c difference between groups was about 0.2 to 0.6 percentage points, too small to explain an early fall in cardiovascular death or heart failure hospitalization.

Glycosuria interferes with two glycemic tests. Urine glucose is positive throughout treatment and cannot be used for monitoring, and the serum 1,5-anhydroglucitol assay is unreliable because its renal reabsorption competes with glucose.

The individual agents: selectivity, pharmacokinetics, and labeling

Six SGLT2 inhibitors are marketed in the United States. They share the same mechanism and class labeling for adverse effects, but their outcome evidence, labeled indications, and kidney thresholds differ. Selectivity for SGLT2 over SGLT1 in vitro ranges from about 2,500-fold for empagliflozin to about 20-fold for sotagliflozin. Canagliflozin is the only selective agent with clinically relevant intestinal SGLT1 inhibition at its 300-mg dose, and sotagliflozin is a deliberate dual inhibitor. All are cleared mainly by glucuronidation, so none requires dose reduction for kidney function on pharmacokinetic grounds; the eGFR thresholds in the labels reflect efficacy and trial enrollment.

Half-lives and metabolic pathways from the current US prescribing information. Selectivity ratios are from in vitro inhibitory concentrations and vary between assays. UGT, uridine 5′-diphospho-glucuronosyltransferase.

Empagliflozin and dapagliflozin carry the broadest evidence and labeling, are the agents used for heart failure and nondiabetic CKD, and are given as 10 mg once daily for all cardiorenal indications. Canagliflozin has cardiovascular and kidney outcome evidence in type 2 diabetes but was not tested in heart failure or nondiabetic CKD. Ertugliflozin was noninferior for cardiovascular safety and did not reduce the primary or key secondary composite outcomes in its trial, and bexagliflozin has no outcome trial with a labeled cardiorenal indication. Sotagliflozin is labeled only for cardiovascular risk reduction and has no glycemic indication.

Natriuresis, plasma volume, and hemodynamic effects

Each glucose molecule that SGLT2 fails to reabsorb leaves behind a sodium ion, and the glucose that reaches the distal nephron acts as an osmotic diuretic. SGLT2 inhibition also reduces proximal sodium reabsorption through the sodium-hydrogen exchanger 3 (NHE3), which is functionally coupled to SGLT2 in the brush border. The combined result is a natriuresis and osmotic diuresis that are largest in the first days of therapy. Urine volume rises by a few hundred milliliters per day at first, and sodium balance is restored within weeks as distal reabsorption increases. In a 12-week mechanistic study in type 2 diabetes, plasma volume fell by a median of 7.3% with dapagliflozin, and hematocrit rose.

The volume effect differs from that of a loop diuretic in three ways. First, it is self-limited and modest, so hypovolemia is uncommon in a patient who is not already receiving intensive diuretic therapy. Second, the osmotic diuresis removes electrolyte-free water as well as sodium, and modeling studies suggest a preferential reduction of interstitial over intravascular volume, which may relieve congestion with less arterial underfilling. Third, it is not accompanied by the reflex sympathetic activation that follows loop diuretic use: plasma renin activity rises modestly and heart rate does not increase despite the fall in blood pressure. Because SGLT2 inhibition acts upstream of the loop of Henle, the drug increases distal sodium delivery and potentiates loop and thiazide diuretics, which is the basis for its use in diuretic resistance.

Systolic blood pressure falls by about 4 to 6 mm Hg and diastolic pressure by about 1 to 2 mm Hg in patients with type 2 diabetes and hypertension. The fall is smaller in heart failure trials, where patients enter with lower pressure and many are taking several vasoactive drugs. Arterial stiffness and pulse wave velocity improve modestly. These changes reduce preload and afterload, but their magnitude does not account for the early reduction in heart failure events, and loop diuretics, which cause larger volume losses, do not reduce mortality.

Tubuloglomerular feedback and kidney protection

In diabetes and in other states of nephron loss, increased proximal reabsorption of glucose and sodium reduces sodium chloride delivery to the macula densa. Tubuloglomerular feedback interprets the low distal delivery as volume depletion and dilates the afferent arteriole. The result is glomerular hyperfiltration, a raised intraglomerular pressure that damages the remaining nephrons over time. SGLT2 inhibition restores sodium chloride delivery to the macula densa, which releases adenosine and constricts the afferent arteriole. Intraglomerular pressure and single-nephron GFR fall. Renin-angiotensin system inhibitors act at the efferent arteriole by the opposite route, and the two classes lower glomerular pressure additively.

This mechanism explains the three kidney effects seen in the trials. Whole-kidney eGFR falls acutely by about 3 to 5 mL/min/1.73 m² within 2 to 4 weeks and returns toward baseline when the drug is stopped, which marks a hemodynamic and reversible change. Albuminuria falls by about 30% to 40%. After the initial dip, the slope of eGFR decline is flattened for as long as the drug is taken, so that the treated curve crosses the placebo curve within 1 to 2 years and kidney function is better preserved thereafter.

Additional effects contribute. Lower proximal tubular workload reduces oxygen consumption and may relieve cortical hypoxia. Kidney weight and tubular injury biomarkers fall in experimental models. Acute kidney injury is also less frequent with SGLT2 inhibitors than with placebo despite the early dip: in a meta-analysis of 13 placebo-controlled trials with more than 90,000 participants, it fell by 23% (relative risk 0.77).

Metabolic, myocardial, and erythropoietic mechanisms

The heart failure benefit appears within weeks, persists without diabetes, and extends across the ejection fraction range. No single mechanism explains it, and the label indications do not depend on any one of them. Hypotheses supported mainly by animal or cellular data are identified as such.

Ketone production and myocardial fuel. Glycosuria causes a continuous calorie loss, insulin falls, and the glucagon-to-insulin ratio rises. Lipolysis and hepatic ketogenesis increase, and β-hydroxybutyrate rises modestly, typically into the range of a few tenths of a millimole per liter. The failing heart oxidizes less fatty acid and glucose and readily takes up ketones in proportion to their arterial concentration. The "thrifty substrate" hypothesis proposes that ketone oxidation improves the work produced per unit of oxygen consumed. The hypothesis is plausible and consistent with human catheterization data on ketone extraction, but improved energetics has not been shown to mediate the clinical benefit. The same metabolic shift, carried to an extreme, causes euglycemic ketoacidosis.

Erythropoiesis and oxygen delivery. Hematocrit rises by about 2 to 4 percentage points within weeks. Part of the rise is hemoconcentration, but erythropoietin and reticulocyte counts also rise, and red cell mass increases. The leading explanation is relief of tubular hypoxia in the renal cortex, which allows peritubular fibroblast-like cells to resume erythropoietin production through hypoxia-inducible factor 2α. Iron indices change in the same direction: hepcidin and ferritin fall as iron is mobilized for erythropoiesis. In mediation analyses of the outcome trials, the change in hematocrit was the variable most strongly associated with the cardiovascular benefit, but such analyses do not establish causation.

Sympathetic activity. Heart rate does not rise despite natriuresis and lower blood pressure, and experimental studies show reduced renal and systemic sympathetic activity. Afferent renal nerve signaling from a less stressed kidney is the proposed link.

Myocardial ion handling and the sodium-hydrogen exchanger 1 (NHE1). Early cellular studies reported that empagliflozin and dapagliflozin inhibit cardiac NHE1, lowering cytosolic sodium and calcium and improving relaxation. Later work using direct measurements of NHE1 activity in isolated cardiomyocytes found no inhibition by empagliflozin at clinically relevant concentrations, and no fall in intracellular sodium. The NHE1 hypothesis is therefore contested and is not an established mechanism. Effects on late sodium current, calcium/calmodulin-dependent protein kinase II, and mitochondrial function have been reported in experimental models.

Remodeling, inflammation, and fat depots. In EMPA-TROPISM, a randomized cardiac magnetic resonance study in nondiabetic heart failure with reduced ejection fraction (HFrEF), empagliflozin reduced left ventricular volumes and mass over 6 months. Epicardial and visceral fat decrease. Markers of inflammation and fibrosis fall, and experimental studies implicate activation of adenosine monophosphate-activated protein kinase and sirtuin signaling, increased autophagy, and suppression of the NLRP3 inflammasome. These pathways mimic a fasting state and are an active area of investigation.

Kidney-heart interaction. Preservation of kidney function, lower risk of acute kidney injury, lower potassium, and fewer interruptions of renin-angiotensin and mineralocorticoid therapy contribute to the long-term benefit, particularly in patients with combined heart and kidney disease.

Effects on blood pressure, weight, uric acid, lipids, potassium, and hematocrit

Most laboratory and physiologic changes after initiation are expected and require no intervention, but each can be mistaken for an adverse event or disease progression.

Blood pressure, eGFR, and albuminuria from pooled diabetes trial data; LDL and hematocrit from the JARDIANCE prescribing information. Magnitudes are approximate and smaller in normoglycemic patients.

Serious hyperkalemia (central laboratory potassium of 6.0 mmol/L or more) was 16% less frequent with SGLT2 inhibitors than with placebo in an individual participant meta-analysis of six trials in type 2 diabetes with almost 50,000 participants (hazard ratio 0.84; 95% confidence interval [CI] 0.76 to 0.93), without an increase in hypokalemia. The reduction was consistent in patients taking renin-angiotensin inhibitors, diuretics, or mineralocorticoid receptor antagonists. In the heart failure trials, patients assigned to an SGLT2 inhibitor were less likely to stop a mineralocorticoid receptor antagonist.

Heart failure with reduced ejection fraction: DAPA-HF and EMPEROR-Reduced

Two trials established SGLT2 inhibitors as the fourth foundational drug class in HFrEF. Both enrolled patients with New York Heart Association (NYHA) class II to IV symptoms, left ventricular ejection fraction (LVEF) of 40% or less, and elevated natriuretic peptides who were receiving contemporary therapy. Both included patients with and without type 2 diabetes in roughly equal proportions.

DAPA-HF. In 4,744 patients with an eGFR of 30 mL/min/1.73 m² or more, dapagliflozin 10 mg daily reduced the composite of worsening heart failure (hospitalization or an urgent visit requiring intravenous therapy) or cardiovascular death over a median of 18.2 months (16.3% versus 21.2%; hazard ratio 0.74; 95% CI 0.65 to 0.85). Cardiovascular death (hazard ratio 0.82; 95% CI 0.69 to 0.98) and death from any cause (hazard ratio 0.83; 95% CI 0.71 to 0.97) were both reduced. The number needed to treat to prevent one primary event over 18 months was about 21. The effect was consistent in patients with and without diabetes, and the curves separated within the first month. Symptoms measured by the Kansas City Cardiomyopathy Questionnaire improved. Volume depletion, kidney adverse events, and major hypoglycemia were uncommon and similar to placebo.

EMPEROR-Reduced. The trial enrolled 3,730 patients with an eGFR of 20 or more and selected a population with more severe disease: most had an LVEF of 30% or less, and natriuretic peptide entry thresholds rose as LVEF increased. Empagliflozin 10 mg daily reduced cardiovascular death or heart failure hospitalization over a median of 16 months (19.4% versus 24.7%; hazard ratio 0.75; 95% CI 0.65 to 0.86). Total heart failure hospitalizations fell by 30% (hazard ratio 0.70). Cardiovascular death was not significantly reduced (hazard ratio 0.92). The annual decline in eGFR was slower with empagliflozin (−0.55 versus −2.28 mL/min/1.73 m² per year), and a composite of serious kidney outcomes was reduced. Benefit was consistent with and without diabetes.

A meta-analysis of the two trials (8,474 patients) found a 13% reduction in death from any cause (hazard ratio 0.87; 95% CI 0.77 to 0.98) and a 14% reduction in cardiovascular death (hazard ratio 0.86; 95% CI 0.76 to 0.98), which resolved the apparent difference in mortality between them. Benefit was consistent with and without sacubitril/valsartan, across the range of baseline eGFR, and in older patients. The 2022 American Heart Association/American College of Cardiology/Heart Failure Society of America (AHA/ACC/HFSA) guideline recommends an SGLT2 inhibitor in symptomatic chronic HFrEF to reduce heart failure hospitalization and cardiovascular mortality, irrespective of diabetes (Class 1, A). Sequencing with angiotensin receptor-neprilysin inhibitors, beta blockers, and mineralocorticoid receptor antagonists is covered in heart failure with reduced ejection fraction.

Heart failure with LVEF above 40%: EMPEROR-Preserved and DELIVER

Before these trials, no drug had clearly reduced a primary composite outcome in heart failure with preserved ejection fraction (HFpEF). Both trials enrolled patients with LVEF above 40%, NYHA class II to IV, structural heart disease, and elevated natriuretic peptides, so they included heart failure with mildly reduced ejection fraction (HFmrEF, LVEF 41% to 49%) and HFpEF.

EMPEROR-Preserved. In 5,988 patients with an eGFR of 20 or more, empagliflozin 10 mg daily reduced cardiovascular death or heart failure hospitalization over a median of 26.2 months (13.8% versus 17.1%; hazard ratio 0.79; 95% CI 0.69 to 0.90). The reduction came from fewer heart failure hospitalizations (total hospitalizations, hazard ratio 0.73), and cardiovascular death was not significantly reduced. Results were consistent with and without diabetes. In a pooled analysis of the two EMPEROR trials across the ejection fraction spectrum, the effect on heart failure hospitalization was attenuated at LVEF of 65% or more.

DELIVER. In 6,263 patients with an eGFR of 25 or more, dapagliflozin 10 mg daily reduced worsening heart failure or cardiovascular death over a median of 2.3 years (16.4% versus 19.5%; hazard ratio 0.82; 95% CI 0.73 to 0.92). DELIVER enrolled patients during or shortly after hospitalization and patients whose ejection fraction had improved from 40% or less (heart failure with improved ejection fraction), and benefit was consistent in both groups and above and below an LVEF of 60%. Symptom scores improved.

In a pooled analysis of the two trials (12,251 patients), SGLT2 inhibitors reduced cardiovascular death or first heart failure hospitalization (hazard ratio 0.80; 95% CI 0.73 to 0.87), with a borderline reduction in cardiovascular death (hazard ratio 0.88; 95% CI 0.77 to 1.00) and a reduction in first hospitalization (hazard ratio 0.74). The 2022 AHA/ACC/HFSA guideline rates SGLT2 inhibitors as reasonable in HFmrEF and HFpEF (Class 2a, B-R), a recommendation written before DELIVER. The 2026 ACC expert consensus decision pathway on HFpEF recommends an SGLT2 inhibitor in all patients with HFpEF unless contraindicated. The 2023 European Society of Cardiology (ESC) focused update recommends empagliflozin or dapagliflozin in HFmrEF and HFpEF with a Class I recommendation. The diagnosis of HFpEF and its other therapies are covered in heart failure with preserved ejection fraction.

Pooled heart failure evidence across ejection fraction and diabetes status

A meta-analysis of the five large heart failure trials (DAPA-HF, EMPEROR-Reduced, EMPEROR-Preserved, DELIVER, and SOLOIST-WHF; 21,947 participants) quantified the class effect. SGLT2 inhibitors reduced the composite of cardiovascular death or heart failure hospitalization (hazard ratio 0.77; 95% CI 0.72 to 0.82), cardiovascular death (0.87; 0.79 to 0.95), first heart failure hospitalization (0.72; 0.67 to 0.78), and death from any cause (0.92; 0.86 to 0.99). Effects on the primary composite were consistent across 14 subgroups, including ejection fraction, diabetes, age, sex, eGFR, and care setting.

Three conclusions follow. First, SGLT2 inhibitors are the only drug class with consistent outcome benefit across the full range of ejection fraction, which makes them the first drug started when the ejection fraction is unknown or changing. Second, the benefit does not depend on diabetes, HbA1c, or glucose lowering. Third, the hospitalization benefit is large and early in every phenotype, while the mortality benefit is clearest in HFrEF. The absolute benefit is greatest in patients at highest risk: those recently hospitalized, with high natriuretic peptides, or with CKD.

The heart failure benefit also extends to specific cardiomyopathies that were represented in the trials, including ischemic, nonischemic, and hypertensive disease. Observational data in transthyretin cardiac amyloidosis associate SGLT2 inhibitor use with fewer heart failure events and better diuretic response, without randomized confirmation; that disease is covered in cardiac amyloidosis. Patients with hypertrophic cardiomyopathy, infiltrative disease, constrictive physiology, and complex congenital heart disease were excluded or few in number.

Starting SGLT2 inhibitors during hospitalization for heart failure

Drugs not started during a heart failure hospitalization are often never started, and the weeks after discharge carry the highest risk of readmission and death. Four randomized trials and a meta-analysis inform in-hospital initiation.

EMPULSE. In 530 patients hospitalized for acute de novo or decompensated chronic heart failure, regardless of ejection fraction or diabetes, empagliflozin 10 mg was started after stabilization at a median of 3 days after admission. At 90 days, empagliflozin produced clinical benefit on a hierarchical composite of death, heart failure events, and change in symptom score (win ratio 1.36; 95% CI 1.09 to 1.68). Serious adverse events were less frequent with empagliflozin (32.3% versus 43.6%). The early eGFR dip was small (about 2 mL/min/1.73 m² at day 15), and investigator-reported acute renal failure was less frequent with empagliflozin (7.7% versus 12.1%).

SOLOIST-WHF. In 1,222 patients with type 2 diabetes hospitalized for worsening heart failure, sotagliflozin was started before discharge in about half and within a median of 2 days after discharge in the rest. Over a median of 9 months, total cardiovascular deaths, heart failure hospitalizations, and urgent heart failure visits were reduced (51.0 versus 76.3 events per 100 patient-years; hazard ratio 0.67; 95% CI 0.52 to 0.85). The trial ended early for loss of funding.

DICTATE-AHF. In 240 patients with hypervolemic acute heart failure randomized within 24 hours of presentation to dapagliflozin 10 mg or structured usual care, dapagliflozin did not significantly improve diuretic efficiency, defined as cumulative weight change per cumulative loop diuretic dose (odds ratio 0.65; 95% CI 0.41 to 1.02). Patients receiving dapagliflozin needed less intravenous loop diuretic (median 560 versus 800 mg furosemide equivalent) and fewer diuretic up-titrations, had greater 24-hour natriuresis and urine output, completed intravenous diuresis sooner, and were discharged earlier, without an increase in kidney, glycemic, or cardiovascular safety events.

DAPA ACT HF-TIMI 68. In 2,401 patients hospitalized for heart failure (71.5% with LVEF of 40% or less, 44.7% with newly diagnosed heart failure), in-hospital dapagliflozin did not significantly reduce cardiovascular death or worsening heart failure through 2 months (10.9% versus 12.7%; hazard ratio 0.86; 95% CI 0.68 to 1.08). Death from any cause was 3.0% versus 4.5% (hazard ratio 0.66; 95% CI 0.43 to 1.00). Symptomatic hypotension (3.6% versus 2.2%) and worsening kidney function (5.9% versus 4.7%) were slightly more frequent with dapagliflozin. The trial had fewer events than planned and limited power.

In the prespecified meta-analysis of DAPA ACT HF-TIMI 68, EMPULSE, and the in-hospital cohort of SOLOIST-WHF (3,527 patients hospitalized for heart failure), in-hospital initiation of an SGLT2 inhibitor reduced the early risk of cardiovascular death or worsening heart failure (hazard ratio 0.71; 95% CI 0.54 to 0.93). The risk of death from any cause was also lower (hazard ratio 0.57; 95% CI 0.41 to 0.80). The 2022 AHA/ACC/HFSA guideline recommends that guideline-directed therapy be started during hospitalization after clinical stability is achieved in patients with HFrEF (Class 1). The 2024 ACC expert consensus decision pathway for patients hospitalized with heart failure states that SGLT2 inhibitors can be started at any point during the admission regardless of ejection fraction. Clinical stability for this purpose means no requirement for intravenous inotropes or vasodilators, systolic blood pressure above about 100 mm Hg in most trial protocols, and no symptomatic hypotension. An ongoing intravenous loop diuretic is not a barrier. The sotagliflozin label allows it to start as soon as the patient is hemodynamically stable, including during hospitalization. Diuresis, vasodilators, and inotropes in the acute phase are covered in acute decompensated heart failure.

Patients who already take an SGLT2 inhibitor continue it during hospitalization for heart failure unless they have hypotension with hypoperfusion, ketoacidosis, poor oral intake with a nothing-by-mouth order for a procedure, or another acute illness that raises ketoacidosis risk. Critical illness, sepsis, and shock are reasons to hold the drug and restart it once the patient is eating and stable.

Kidney outcome trials: CREDENCE, DAPA-CKD, and EMPA-KIDNEY

Kidney outcomes were secondary end points in the early cardiovascular trials, which consistently showed slower eGFR decline and less albuminuria. Three dedicated kidney trials followed, each stopped early at a planned interim analysis.

CREDENCE. The trial enrolled 4,401 patients with type 2 diabetes, an eGFR of 30 to 90 mL/min/1.73 m², and a urine albumin-to-creatinine ratio (UACR) above 300 to 5,000 mg/g, all receiving renin-angiotensin blockade. Canagliflozin 100 mg daily reduced a composite of end-stage kidney disease, doubling of serum creatinine, or kidney or cardiovascular death over a median of 2.62 years (43.2 versus 61.2 events per 1,000 patient-years; hazard ratio 0.70; 95% CI 0.59 to 0.82). The kidney-specific composite was reduced by 34% (hazard ratio 0.66). The trial was stopped early at a planned interim analysis. It was the first positive kidney outcome trial in diabetic kidney disease since the angiotensin receptor blocker trials of 2001.

DAPA-CKD. The trial enrolled 4,304 patients with an eGFR of 25 to 75 and a UACR of 200 to 5,000 mg/g, about one-third without diabetes. Dapagliflozin 10 mg daily reduced a composite of a sustained eGFR decline of at least 50%, end-stage kidney disease, or kidney or cardiovascular death over a median of 2.4 years (9.2% versus 14.5%; hazard ratio 0.61; 95% CI 0.51 to 0.72). Death from any cause was reduced (hazard ratio 0.69). The trial was stopped early for efficacy, and the effect was consistent with and without diabetes and across causes of CKD, including IgA nephropathy and chronic glomerulonephritis. Patients with polycystic kidney disease, lupus nephritis, antineutrophil cytoplasmic antibody-associated vasculitis, or recent immunosuppressive therapy were excluded.

EMPA-KIDNEY. The trial enrolled 6,609 patients with an eGFR of 20 to less than 45, or an eGFR of 45 to less than 90 with a UACR of 200 mg/g or more. More than half had no diabetes, and many had little albuminuria. Empagliflozin 10 mg daily reduced progression of kidney disease or cardiovascular death over a median of 2.0 years (13.1% versus 16.9%; hazard ratio 0.72; 95% CI 0.64 to 0.82). Hospitalization from any cause was reduced. The relative effect on kidney progression was larger with higher albuminuria, and in patients with a UACR below 30 mg/g the effect on the primary outcome was uncertain, although the eGFR slope was slowed.

A collaborative meta-analysis of 13 large placebo-controlled trials (90,409 participants, 15,605 without diabetes) found that SGLT2 inhibitors reduced kidney disease progression by 37% (relative risk 0.63; 95% CI 0.58 to 0.69), acute kidney injury by 23%, and cardiovascular death or heart failure hospitalization by 23%. Relative effects were similar with and without diabetes and across baseline eGFR. The absolute benefits outweighed the serious hazards of ketoacidosis and amputation.

Dedicated kidney and cardiorenal trials. eGFR in mL/min/1.73 m². CV, cardiovascular; ESKD, end-stage kidney disease; HF, heart failure; UACR, urine albumin-to-creatinine ratio.

Kidney Disease: Improving Global Outcomes (KDIGO) 2024 recommends an SGLT2 inhibitor for adults with type 2 diabetes, CKD, and an eGFR of 20 or more (1A). It also recommends an SGLT2 inhibitor for adults with CKD and an eGFR of 20 or more who have a UACR of 200 mg/g or more, or heart failure irrespective of albuminuria (1A). It suggests one for adults with an eGFR of 20 to 45 and a UACR below 200 mg/g (2B). Practice points state that the initial eGFR dip is not a reason to stop and that continuing the drug below an eGFR of 20 is reasonable unless it is not tolerated or kidney replacement therapy begins. They also state that the drug is withheld during prolonged fasting, surgery, or critical illness.

Cardiovascular outcome trials in type 2 diabetes

After the 2008 US Food and Drug Administration (FDA) guidance requiring cardiovascular safety trials for new glucose-lowering drugs, four SGLT2 inhibitors were tested against placebo in patients with type 2 diabetes and atherosclerotic disease or multiple risk factors. All met noninferiority for major adverse cardiovascular events (MACE: cardiovascular death, myocardial infarction, or stroke). Their consistent finding was a reduction in heart failure hospitalization, which prompted the dedicated heart failure trials.

EMPA-REG OUTCOME. In 7,020 patients with type 2 diabetes and established cardiovascular disease, empagliflozin 10 or 25 mg daily reduced MACE over a median of 3.1 years (10.5% versus 12.1%; hazard ratio 0.86; 95.02% CI 0.74 to 0.99). The effect came entirely from cardiovascular death (3.7% versus 5.9%; a 38% relative reduction); myocardial infarction and stroke were not reduced. Heart failure hospitalization fell by 35% (2.7% versus 4.1%), and death from any cause by 32% (5.7% versus 8.3%). The curves separated within months, too early for an antiatherosclerotic effect. Genital infections were more frequent with empagliflozin.

CANVAS Program. In 10,142 patients with type 2 diabetes and high cardiovascular risk (65.6% with established cardiovascular disease), canagliflozin reduced MACE (26.9 versus 31.5 per 1,000 patient-years; hazard ratio 0.86; 95% CI 0.75 to 0.97). Albuminuria progression and a composite kidney outcome were reduced. Lower-limb amputation, mostly of the toe or metatarsal, was doubled (6.3 versus 3.4 per 1,000 patient-years; hazard ratio 1.97; 95% CI 1.41 to 2.75), and fractures were increased.

DECLARE-TIMI 58. In 17,160 patients, 10,186 of them without established atherosclerotic disease, dapagliflozin 10 mg daily over a median of 4.2 years was noninferior for MACE and did not reduce it (8.8% versus 9.4%; hazard ratio 0.93; 95% CI 0.84 to 1.03). It reduced the coprimary composite of cardiovascular death or heart failure hospitalization (4.9% versus 5.8%; hazard ratio 0.83; 95% CI 0.73 to 0.95), driven by fewer heart failure hospitalizations, and reduced a kidney composite. Diabetic ketoacidosis was more frequent (0.3% versus 0.1%).

VERTIS CV. In 8,246 patients with type 2 diabetes and atherosclerotic disease followed for a mean of 3.5 years, ertugliflozin was noninferior for MACE (11.9% in both groups; hazard ratio 0.97; 95.6% CI 0.85 to 1.11) and did not significantly reduce cardiovascular death or heart failure hospitalization (hazard ratio 0.88; 95.8% CI 0.75 to 1.03). First heart failure hospitalization, an exploratory outcome, was reduced (hazard ratio 0.70).

Cardiovascular outcome trials in type 2 diabetes. MACE, major adverse cardiovascular events; CV, cardiovascular; HF, heart failure; HR, hazard ratio.

Across these trials, the reduction in MACE was confined to patients with established atherosclerotic disease, and the reduction in heart failure hospitalization (about 30%) and kidney outcomes was present with or without atherosclerotic disease and with or without a history of heart failure. SGLT2 inhibitors therefore prevent heart failure in diabetes more reliably than they prevent myocardial infarction or stroke. Glucagon-like peptide-1 (GLP-1) receptor agonists have the opposite profile: they reduce atherosclerotic events and stroke but have no effect on heart failure with reduced ejection fraction. The 2022 AHA/ACC/HFSA guideline recommends SGLT2 inhibitors in patients with type 2 diabetes and either established cardiovascular disease or high cardiovascular risk to prevent heart failure hospitalization (Class 1, A).

Sotagliflozin and dual SGLT1/SGLT2 inhibition

Sotagliflozin inhibits SGLT2 in the kidney and SGLT1 in the intestine. Intestinal SGLT1 inhibition delays glucose absorption, blunts postprandial glucose, and increases release of glucagon-like peptide-1 and peptide YY from distal gut cells exposed to unabsorbed glucose. At the doses used, renal SGLT1 inhibition is modest, so glycosuria resembles that of a selective SGLT2 inhibitor. Diarrhea is more frequent than with selective agents because unabsorbed carbohydrate reaches the colon.

Two trials, both stopped early when the sponsor withdrew funding, form the evidence. SOLOIST-WHF enrolled patients with type 2 diabetes recently hospitalized for heart failure across the ejection fraction range. SCORED enrolled 10,584 patients with type 2 diabetes, an eGFR of 25 to 60, and cardiovascular risk factors, with or without albuminuria. In SCORED, sotagliflozin reduced the total number of cardiovascular deaths, heart failure hospitalizations, and urgent heart failure visits (5.6 versus 7.5 events per 100 patient-years; hazard ratio 0.74; 95% CI 0.63 to 0.88). Diarrhea (8.5% versus 6.0%), genital mycotic infections (2.4% versus 0.9%), volume depletion, and diabetic ketoacidosis were more frequent with sotagliflozin.

The FDA approved sotagliflozin (INPEFA) to reduce cardiovascular death, heart failure hospitalization, and urgent heart failure visits in adults with heart failure, or with type 2 diabetes, CKD, and other cardiovascular risk factors. Its label covers heart failure across the ejection fraction range and does not require diabetes, although both trials enrolled only patients with type 2 diabetes. It is not indicated for glycemic control. The dose is 200 mg once daily, taken no more than 1 hour before the first meal, increased after at least 2 weeks to 400 mg as tolerated. Initiation is not studied below an eGFR of 25. Sotagliflozin increases digoxin exposure, and digoxin concentrations are monitored.

In type 1 diabetes, sotagliflozin 400 mg added to insulin improved HbA1c, weight, and insulin dose in the inTandem program, but diabetic ketoacidosis occurred in 3.0% versus 0.6% with placebo in inTandem3. The FDA declined approval for type 1 diabetes in 2019 and again after a 2024 advisory committee review. In hypertrophic cardiomyopathy, the SONATA-HCM trial tests sotagliflozin against placebo in obstructive and nonobstructive disease, with change in the Kansas City Cardiomyopathy Questionnaire clinical summary score at 26 weeks as the primary end point. Current management of that disease is covered in hypertrophic cardiomyopathy.

SGLT2 inhibitors after myocardial infarction: DAPA-MI and EMPACT-MI

Heart failure after myocardial infarction is common and prognostically important, and the rapid benefit of SGLT2 inhibitors in heart failure suggested a role in prevention. Two trials tested them early after infarction.

EMPACT-MI. The trial randomized 6,522 patients hospitalized for acute myocardial infarction with new LVEF below 45% or congestion, plus at least one enrichment risk factor, to empagliflozin 10 mg or placebo within 14 days of admission. Over a median of 17.9 months, first heart failure hospitalization or death from any cause was not significantly reduced (8.2% versus 9.1%; 5.9 versus 6.6 events per 100 patient-years; hazard ratio 0.90; 95% CI 0.76 to 1.06). First heart failure hospitalization (hazard ratio 0.77; 95% CI 0.60 to 0.98) and total heart failure hospitalizations (rate ratio 0.67) were lower in analyses without formal hypothesis testing. Death from any cause was similar (5.2% versus 5.5%).

DAPA-MI. In 4,017 patients with acute myocardial infarction and impaired regional or global systolic function, without diabetes or chronic heart failure, dapagliflozin 10 mg was compared with placebo. Low event accrual led to a change of the primary outcome to a hierarchical composite, which favored dapagliflozin (win ratio 1.34; 95% CI 1.20 to 1.50) because of cardiometabolic components such as weight and new diabetes. Cardiovascular death or heart failure hospitalization did not differ (2.5% versus 2.6%; hazard ratio 0.95).

Neither trial supports routine SGLT2 inhibitor therapy after myocardial infarction solely because an infarction has occurred. The drugs are given after infarction when the patient has an established indication: heart failure at any ejection fraction, type 2 diabetes with cardiovascular disease, or CKD. Post-infarction management is covered in ST-elevation myocardial infarction and non-ST-elevation acute coronary syndrome.

Atrial fibrillation, ventricular arrhythmia, and sudden death

SGLT2 inhibitors are not antiarrhythmic drugs, but several analyses suggest fewer arrhythmic events. In DECLARE-TIMI 58, dapagliflozin reduced reported atrial fibrillation or flutter events by 19% (7.8 versus 9.6 events per 1,000 patient-years), irrespective of prior atrial fibrillation or heart failure. The outcome was identified from adverse event reports and was not a prespecified efficacy end point. Meta-analyses of the outcome trials report similar relative reductions, with low absolute rates.

In a post hoc analysis of DAPA-HF, dapagliflozin reduced the composite of serious ventricular arrhythmia, resuscitated cardiac arrest, or sudden death (5.9% versus 7.4%; hazard ratio 0.79; 95% CI 0.63 to 0.99). Adjudicated sudden death made up most of these events, and 41% of cardiovascular deaths in the trial were sudden. Pooled analyses of the heart failure trials show a smaller and less certain effect on sudden death. These findings do not change indications for implantable cardioverter-defibrillators, which are selected after guideline-directed medical therapy has been optimized, including an SGLT2 inhibitor. Rhythm management is covered in atrial fibrillation and ventricular tachycardia and sudden cardiac death.

Guideline recommendations for SGLT2 inhibitors

US heart failure, kidney, diabetes, and perioperative guidance agree on the core indications and differ mainly in strength of recommendation and in kidney thresholds.

Class of recommendation and level of evidence as published. ACC pathway documents are expert consensus decision pathways and do not assign classes. UACR, urine albumin-to-creatinine ratio.

European guidance differs in one main respect: the 2023 ESC focused update on heart failure gives empagliflozin or dapagliflozin a Class I recommendation in HFmrEF and HFpEF, stronger than the Class 2a in the 2022 US guideline. In HFrEF, the 2022 AHA/ACC/HFSA guideline attached a value statement to SGLT2 inhibitors and rated them as intermediate economic value at their prices at the time.

FDA-labeled indications, doses, and kidney thresholds

The thresholds are the eGFR values below which the label does not recommend initiation or, where marked as trial enrollment, below which initiation was not studied. Continuation below them is permitted for some agents and indications, because the trials did not stop the drug when eGFR fell.

From the current US prescribing information for FARXIGA, JARDIANCE, INVOKANA, STEGLATRO, BRENZAVVY, and INPEFA. The dapagliflozin and empagliflozin labels do not recommend them for CKD in polycystic kidney disease or in patients requiring, or recently treated with, immunosuppressive therapy for kidney disease (for empagliflozin, intravenous immunosuppression or more than 45 mg of prednisone or equivalent). None is approved for type 1 diabetes.

The JARDIANCE and FARXIGA labels no longer list dialysis as a contraindication. They state that the trials did not enroll patients with an eGFR below 20 (empagliflozin) or 25 (dapagliflozin) or on dialysis. They also state that enrolled patients were not required to discontinue treatment when eGFR fell below these levels (below 30 in DAPA-HF) or dialysis began. Doses for cardiorenal indications are fixed; there is no target dose to titrate toward with dapagliflozin or empagliflozin, and the 25-mg empagliflozin dose and 300-mg canagliflozin dose add only glucose lowering.

SGLT2 inhibitor indication, dose, and surgical hold calculator

The calculator applies the US labeling, the trial enrollment thresholds, KDIGO 2024, and the FDA perioperative labeling to one patient. It returns whether the selected agent carries the selected indication, the labeled dose, the eGFR threshold for initiation, and, when a surgery date is entered, the date of the last dose before surgery. Volume status, blood pressure, ketoacidosis risk factors, and drug interactions are outside its inputs.

Thresholds from the US prescribing information and the trial enrollment criteria (empagliflozin 20 for heart failure and CKD and 30 for cardiovascular death in type 2 diabetes, dapagliflozin and sotagliflozin 25, and canagliflozin 30 for cardiorenal use; eGFR in mL/min/1.73 m²). The last-dose date counts the labeled hold (3 days; 4 days for ertugliflozin) plus the day of surgery, consistent with the 3- to 4-day hold in the 2024 AHA/ACC perioperative guideline. Procedures with prolonged fasting, including bowel preparation, are treated as surgery.

Initiation, monitoring, and the expected eGFR dip

Before initiation, eGFR, potassium, and volume status are assessed, and volume depletion is corrected. In a patient with type 2 diabetes, the insulin or sulfonylurea dose is reviewed. The patient is told about genital hygiene, the symptoms of genital and urinary infection, and the sick-day rules for holding the drug. Dapagliflozin and empagliflozin are started at 10 mg once daily, taken in the morning with or without food; no titration is required.

Kidney function and electrolytes are checked about 1 to 4 weeks after initiation in patients with reduced eGFR, those taking loop diuretics, or those at risk of volume depletion, and then as clinically indicated. In the DAPA-HF analysis of the early dip, eGFR fell by a mean of 4.2 mL/min/1.73 m² at 2 weeks with dapagliflozin and 1.1 with placebo. A decline of more than 10% occurred in 38% of patients receiving dapagliflozin, and larger declines were uncommon. Patients with an early dip derived the same clinical benefit as those without one, and the dip did not predict worse kidney outcomes. Similar observations were made in EMPA-REG OUTCOME, CREDENCE, and the kidney trials.

A fall in eGFR of up to about 30% after initiation is accepted without stopping the drug. A larger fall, a continuing decline after the first month, oliguria, or hyperkalemia leads to evaluation for volume depletion, hypotension, urinary obstruction, nephrotoxic drugs, renal artery stenosis, or intercurrent illness. The diuretic dose is reduced when the patient is volume depleted. Stopping the SGLT2 inhibitor for a stable early dip forfeits its long-term kidney and cardiac benefit; in the trials, eGFR in the treated group crossed above the placebo curve within 1 to 2 years. Monitoring during combined titration of the four heart failure drug classes is covered in heart failure with reduced ejection fraction.

Long-term follow-up includes eGFR and UACR at intervals set by CKD stage, HbA1c in diabetes, inspection of the feet in patients with diabetes and peripheral artery disease, and review of genital or urinary symptoms. Hematocrit may rise above the reference range; in a euvolemic patient this is expected and does not require phlebotomy, but polycythemia from another cause is not attributed to the drug without evaluation.

Volume depletion, hypotension, and diuretic adjustment

Volume depletion is the most frequent reason for intolerance. Risk factors are older age, low baseline blood pressure, an eGFR below 60 mL/min/1.73 m², intensive loop diuretic therapy, and poor oral intake. In the heart failure trials, adverse events of volume depletion were only slightly more frequent with the SGLT2 inhibitor than with placebo, and symptomatic hypotension was uncommon.

In a euvolemic patient taking a stable loop diuretic dose, the SGLT2 inhibitor is started without changing the diuretic, and the dose is reduced later if weight falls below the dry weight or orthostatic symptoms appear. In a patient with low blood pressure, borderline volume status, or recent intensive diuresis, the loop diuretic dose is reduced by about one-quarter to one-half at initiation, with daily weights and early follow-up. In the heart failure trials, the need for loop diuretic intensification fell over time in the SGLT2 inhibitor groups. Thiazide and thiazide-like diuretics add distal natriuresis, and a thiazide combined with an SGLT2 inhibitor and a loop diuretic is used for diuretic resistance with close electrolyte monitoring.

Systolic pressure of 90 to 100 mm Hg is not a contraindication in heart failure if perfusion is adequate. The SGLT2 inhibitor has less effect on blood pressure than an angiotensin receptor-neprilysin inhibitor and is usually retained when blood pressure limits titration. Nonessential vasodilators and alpha-blockers are reduced first.

Euglycemic diabetic ketoacidosis

Diabetic ketoacidosis during SGLT2 inhibitor therapy often occurs with plasma glucose below 250 mg/dL, and sometimes near normal. Glycosuria removes glucose while insulin secretion or dosing falls and glucagon rises, so ketogenesis proceeds without the hyperglycemia that normally signals ketoacidosis. Volume depletion and a fall in renal ketone clearance amplify the acidosis. The FDA issued a safety communication in 2015, and all SGLT2 inhibitor labels carry the warning. Case series that followed described delayed recognition because glucose was not elevated.

Incidence. In type 2 diabetes, ketoacidosis is uncommon. In DECLARE-TIMI 58 it occurred in 0.3% with dapagliflozin and 0.1% with placebo over 4.2 years, and meta-analyses of large trials estimate roughly double the risk at an absolute rate well below 1 per 100 patient-years. It is very rare in patients without diabetes; in the heart failure and kidney trials, cases occurred almost exclusively in patients with diabetes. In type 1 diabetes the risk is several-fold higher.

Precipitants. Most cases follow a recognizable trigger: surgery or a procedure with fasting, an acute medical illness, vomiting or diarrhea, low-carbohydrate or ketogenic diets, prolonged fasting, heavy alcohol intake, a sharp reduction or omission of insulin, pancreatic insufficiency or a history of pancreatitis or pancreatic surgery, latent autoimmune diabetes of adults, and pregnancy. Insulin-deficient type 2 diabetes with long duration and low body weight carries higher risk.

Diagnosis. Nausea, vomiting, abdominal pain, malaise, dyspnea, and tachypnea in a patient taking an SGLT2 inhibitor lead to measurement of serum or capillary β-hydroxybutyrate, venous pH, bicarbonate, and anion gap, irrespective of glucose. A β-hydroxybutyrate of 3.0 mmol/L or more with a pH below 7.3 or bicarbonate below 18 mEq/L meets criteria for ketoacidosis. Urine ketone strips detect acetoacetate and underestimate β-hydroxybutyrate, so a blood measurement is preferred. Glucose may be normal, and the anion gap may be blunted when hyperchloremia from saline therapy develops.

Treatment. The SGLT2 inhibitor is stopped. Ketoacidosis is treated with intravenous fluid, insulin, and potassium replacement as in other forms of ketoacidosis, with one adjustment: because glucose is low, dextrose is given from the start alongside insulin so that insulin can continue until ketosis resolves. Glycosuria and ketonuria persist for several days after the last dose, longer in patients with reduced kidney function, and recurrence of ketosis is watched for during this period. Rechallenge after an episode with a clear precipitant is an individual decision; after an unprovoked episode the class is usually not restarted.

Genital mycotic infections, urinary tract infection, and Fournier gangrene

Genital mycotic infections are the most frequent adverse effect. Vulvovaginal candidiasis in women and balanitis or balanoposthitis in men occur in roughly 3% to 6% of treated patients with diabetes, several times the placebo rate. Risk is higher in women than in men, in uncircumcised men, and in patients with a history of such infections. In EMPA-REG OUTCOME, genital infections were reported in 6.4% with empagliflozin and 1.8% with placebo. Infections usually appear in the first months, respond to topical or single-dose oral antifungal therapy, and rarely lead to discontinuation. Genital hygiene after voiding reduces recurrence. The rate is lower in patients without diabetes, whose urine contains less glucose.

Urinary tract infection. Pooled trial data show at most a small increase in urinary tract infection, and the large outcome trials showed no consistent excess of serious urinary infection. The labels warn of urosepsis and pyelonephritis after postmarketing reports, and patients with recurrent complicated urinary infection, urinary retention, or indwelling catheters are monitored more closely. A urinary tract infection is treated in the usual way, and the drug is held during severe infection or sepsis.

Fournier gangrene. Necrotizing fasciitis of the perineum is rare and life-threatening. The FDA added a warning in 2018. A review of FDA adverse event reports identified 55 cases in patients receiving SGLT2 inhibitors from March 2013 through January 2019, all requiring surgical debridement, with 3 deaths. Pain, tenderness, erythema, or swelling of the genital or perineal area with fever or malaise leads to immediate surgical evaluation, broad-spectrum antibiotics, and discontinuation of the drug. A causal relation has not been established, because the event is too rare for the trials to resolve it.

Hypoglycemia with insulin and sulfonylureas

SGLT2 inhibitors do not cause hypoglycemia on their own, because glycosuria falls as plasma glucose falls and insulin secretion is unaffected. Hypoglycemia occurs when they are added to insulin or an insulin secretagogue. In patients with type 2 diabetes whose HbA1c is near target, the labels advise a lower dose of insulin or an insulin secretagogue at initiation: the sulfonylurea is reduced or stopped, insulin is reduced modestly, and glucose is monitored over the first weeks. The reduction is smaller when eGFR is low, because glucose lowering is small. Insulin is never stopped in a patient who requires it, because insulin deficiency is the main precondition for ketoacidosis.

Amputation, fracture, and other safety signals

Lower-limb amputation. The CANVAS Program found about twice the rate of amputation with canagliflozin, mostly of the toe or metatarsal, and the FDA added a boxed warning in 2017. CREDENCE, with closer foot surveillance, found no significant excess, and the boxed warning was removed in 2020, although the labeling still lists the risk. Amputation rates in the empagliflozin outcome trials were 5.0 versus 4.3 per 1,000 patient-years (hazard ratio 1.05), and VERTIS CV and the bexagliflozin trial reported numerically higher rates with the active drug. The class labels advise monitoring for lower-limb ulcers and infection. Patients with active foot ulceration, critical limb ischemia, or prior amputation receive foot care and vascular assessment, and the drug is held during active foot infection or ischemia. Volume contraction and reduced limb perfusion are the proposed mechanisms.

Fracture. Canagliflozin increased fractures in CANVAS, mainly early and in older patients, and the label carries a fracture warning. Fracture rates were not increased in CREDENCE or in the trials of other agents. Mechanisms proposed include falls from volume depletion and effects on phosphate, parathyroid hormone, and fibroblast growth factor 23.

Other effects. Acute kidney injury is less frequent than with placebo, although postmarketing reports describe it in volume-depleted patients. Hypersensitivity reactions, including angioedema, occur rarely. Bladder cancer imbalances in the dapagliflozin glycemic program were not confirmed in DECLARE-TIMI 58. Polyuria and nocturia are noticed by some patients in the first weeks.

Adverse effects of SGLT2 inhibitors from the outcome trials and US labeling.

Holding SGLT2 inhibitors before surgery, procedures, and during acute illness

Perioperative fasting, surgical stress hormones, reduced insulin doses, and volume shifts create the conditions for euglycemic ketoacidosis, and postoperative cases have been reported up to several days after surgery. In 2020 the FDA required the labels to advise withholding canagliflozin, dapagliflozin, and empagliflozin for at least 3 days and ertugliflozin for at least 4 days before scheduled surgery. Bexagliflozin and sotagliflozin carry 3-day holds. The 2024 AHA/ACC perioperative guideline recommends withholding SGLT2 inhibitors for 3 to 4 days before noncardiac surgery, when feasible, to reduce the risk of perioperative metabolic acidosis (Class 1, C-LD). It also states that continuing other guideline-directed heart failure therapy through surgery is reasonable (Class 2a, C-LD). The SGLT2 inhibitor is restarted once the patient is clinically stable and has resumed a normal diet.

For surgery on a Thursday, the last dose of dapagliflozin or empagliflozin is taken on Sunday, and for ertugliflozin on Saturday. The hold applies to procedures with prolonged fasting or bowel preparation, including colonoscopy, and to cardiac surgery. Short procedures that require only an overnight fast fall outside the labeled hold; institutional protocols commonly omit the dose on the morning of the procedure and resume it with the first meal. The drug is restarted when the patient is eating and drinking normally, is hemodynamically stable, and has no ketosis or acidosis, usually 24 to 72 hours after surgery.

For urgent or emergency surgery in a patient who has taken the drug within 3 days, surgery is not delayed. Blood β-hydroxybutyrate and acid-base status are measured before and after surgery, dextrose-containing fluid and insulin are given as needed, and ketones are rechecked until the patient is eating. A 2026 multidisciplinary consensus from the Society for Perioperative Assessment and Quality Improvement recommends individualizing the hold by diabetes status, surgical complexity, indication, and duration of fasting.

Sick-day rules. During acute illness with vomiting, diarrhea, fever, poor oral intake, or dehydration, the SGLT2 inhibitor is withheld and restarted 24 to 48 hours after the patient is eating and drinking normally. The same rule applies to metformin, diuretics, renin-angiotensin inhibitors, and other drugs that raise the risk of acute kidney injury or lactic acidosis during volume depletion. Patients with diabetes who feel unwell test blood ketones if a meter is available, continue basal insulin, and seek care when ketones are elevated. KDIGO 2024 and the ADA Standards of Care endorse temporary withholding during prolonged fasting, surgery, and critical illness.

From the US prescribing information, the 2024 AHA/ACC perioperative guideline, KDIGO 2024, and the ADA Standards of Care.

SGLT2 inhibitors in type 1 diabetes

No SGLT2 inhibitor is approved in the United States for type 1 diabetes. Every label states that the drugs are not recommended to improve glycemic control in type 1 diabetes and that they increase the risk of ketoacidosis in these patients. In trials of sotagliflozin, dapagliflozin, and empagliflozin added to insulin, HbA1c fell by about 0.4 percentage points and weight and insulin requirements fell, but ketoacidosis was several times more frequent than with placebo. Dapagliflozin and sotagliflozin were briefly authorized in Europe for selected adults with type 1 diabetes and a body mass index of 27 kg/m² or more, and both authorizations were later withdrawn.

When a patient with type 1 diabetes has HFrEF or CKD, the cardiorenal benefit of an SGLT2 inhibitor is plausible but untested, because type 1 diabetes was excluded from the outcome trials. Use is an individualized specialist decision. An international consensus on managing ketoacidosis risk in type 1 diabetes advises use only in adults with stable insulin requirements, a body mass index of 27 kg/m² or more, no recent ketoacidosis, reliable insulin dosing, and access to blood ketone testing. Insulin is not reduced by more than about 10% to 20%. Ketones are checked when unwell, and the drug is stopped during illness, fasting, alcohol excess, or reduced insulin delivery, for example pump failure.

Dialysis, kidney transplantation, and polycystic kidney disease

Dialysis. Patients on dialysis were not enrolled in any outcome trial, and an SGLT2 inhibitor produces little glycosuria or natriuresis in a patient with minimal residual function. Initiation is not recommended. In DAPA-CKD, DELIVER, EMPA-KIDNEY, and the EMPEROR trials, patients who reached dialysis during the trial could continue the drug, and KDIGO 2024 advises continuation until kidney replacement therapy begins. Continuation after dialysis starts is individualized, and the RENAL LIFECYCLE trial randomizes patients with advanced CKD, patients on dialysis, and kidney transplant recipients to dapagliflozin or placebo.

Kidney transplantation. Transplant recipients were excluded from the outcome trials. Observational studies in recipients with diabetes after transplantation report glucose lowering and weight loss without excess rejection, with a higher frequency of urinary infection in some series. The drug is not started early after transplantation, while immunosuppression is intense and urinary infections are common. In a stable recipient with diabetes, heart failure, or proteinuria, an SGLT2 inhibitor is sometimes used after discussion with the transplant team, with attention to calcineurin inhibitor exposure and volume status.

Polycystic kidney disease and immunosuppressed glomerular disease. Polycystic kidney disease was excluded from DAPA-CKD and EMPA-KIDNEY, and the labels do not recommend dapagliflozin or empagliflozin for CKD in that disease. Experimental data raise concern that increased vasopressin and glycosuria could promote cyst growth. Patients receiving, or recently treated with, immunosuppressive therapy for glomerular disease were also excluded, and the labels state that the drugs are not expected to be effective in these groups.

SGLT2 inhibitors in older adults and frailty

Relative benefit was consistent across age groups in the heart failure, kidney, and diabetes trials, and because older patients have higher event rates, the absolute benefit is larger. In DECLARE-TIMI 58, the reduction in cardiovascular death or heart failure hospitalization was consistent in patients younger than 65, 65 to 74, and 75 years or older, and acute kidney injury was reduced in all age groups. Pooled analyses of DAPA-HF and DELIVER and of the EMPEROR trials found consistent effects up to and beyond 80 years.

Older patients have more volume depletion, orthostatic hypotension, and falls, especially with loop diuretics, low body weight, poor fluid intake, or cognitive impairment. Genitourinary infection and incontinence from polyuria are more troublesome in frail patients. The drug is started with a review of diuretic doses and fluid intake, and the sick-day rules are explained to the patient and caregiver. Frailty, limited life expectancy, and the goal of avoiding hospitalization favor the drug in heart failure, because the hospitalization benefit appears within weeks. In older patients with diabetes and low HbA1c, insulin and sulfonylurea doses are reduced to avoid hypoglycemia.

SGLT2 inhibitors in pregnancy and lactation

The labels do not recommend SGLT2 inhibitors during the second and third trimesters, based on animal studies showing renal pelvic and tubular dilatation at exposures corresponding to late human renal development. Human data are limited. In women of reproductive potential with heart failure or CKD, contraception is discussed, and the drug is stopped when pregnancy is planned or recognized. The labels do not recommend breastfeeding during treatment, because human kidney maturation continues during the first 2 years of life. After delivery, heart failure therapy uses agents with lactation data; the SGLT2 inhibitor is resumed after breastfeeding ends.

Combining SGLT2 inhibitors with diuretics, renin-angiotensin inhibitors, and mineralocorticoid receptor antagonists

The SGLT2 inhibitor is one of the four foundational drugs in HFrEF, with an angiotensin receptor-neprilysin inhibitor (or an angiotensin-converting enzyme [ACE] inhibitor or angiotensin receptor blocker), an evidence-based beta blocker, and a mineralocorticoid receptor antagonist. The 2024 ACC decision pathway recommends starting all four in the order the clinical situation allows and completing optimization within about 3 months. The SGLT2 inhibitor is well suited to early initiation: it needs no titration, lowers blood pressure little, does not raise potassium, and offsets some adverse effects of the other drugs. It lowers the risk of hyperkalemia from mineralocorticoid antagonists, attenuates the long-term eGFR decline after renin-angiotensin inhibition, and adds natriuresis that allows the loop diuretic dose to fall. Sequencing by blood pressure, heart rate, potassium, and kidney function is covered in heart failure with reduced ejection fraction.

In CKD, the SGLT2 inhibitor is added to a maximally tolerated ACE inhibitor or angiotensin receptor blocker; almost all patients in CREDENCE and DAPA-CKD, and most in EMPA-KIDNEY, received renin-angiotensin blockade. The two classes reduce intraglomerular pressure by complementary actions on the afferent and efferent arterioles. Finerenone, a nonsteroidal mineralocorticoid receptor antagonist, reduces kidney and cardiovascular events in type 2 diabetes with albuminuric CKD and in HFmrEF and HFpEF (FINEARTS-HF). In those trials, the effect of finerenone was consistent with and without SGLT2 inhibitor use, and the SGLT2 inhibitor lowered the risk of hyperkalemia. The two are used together, with potassium monitored.

Loop diuretics remain the treatment of congestion. In acute heart failure, an SGLT2 inhibitor increases natriuresis and reduces the loop diuretic dose needed for decongestion, as in DICTATE-AHF. Acetazolamide and thiazide-like diuretics act at different nephron segments, and sequential blockade with an SGLT2 inhibitor raises the risk of hypokalemia, hyponatremia, and volume depletion, so electrolytes and creatinine are checked frequently during combined use.

Combining SGLT2 inhibitors with GLP-1 receptor agonists

The two classes have complementary outcome profiles. SGLT2 inhibitors reduce heart failure hospitalization, cardiovascular death in heart failure, and kidney disease progression with or without diabetes. GLP-1 receptor agonists reduce atherosclerotic events and stroke, produce larger weight loss and HbA1c reduction, and, with semaglutide, slow kidney disease progression in type 2 diabetes. Semaglutide and tirzepatide improve symptoms and heart failure events in obesity-related HFpEF but have no benefit in HFrEF. No randomized trial has compared the combination with either drug alone. Subgroup analyses of GLP-1 receptor agonist trials show consistent effects in participants already taking an SGLT2 inhibitor, and observational data associate the combination with fewer cardiovascular events than either class alone.

The ADA Standards of Care state that combined therapy with an SGLT2 inhibitor and a GLP-1 receptor agonist, each with demonstrated cardiovascular benefit, may be considered in type 2 diabetes with established atherosclerotic disease or multiple risk factors to reduce cardiovascular and kidney events further (grade B). The principal interaction is volume depletion. Three contributors add up: natriuresis from the SGLT2 inhibitor; nausea, vomiting, and reduced intake during GLP-1 receptor agonist titration; and loop diuretics. Together they increase the risk of hypotension, acute kidney injury, and ketoacidosis during intercurrent illness. Both drugs are withheld during vomiting or poor intake. Before procedures under sedation, GLP-1 receptor agonists raise aspiration risk and SGLT2 inhibitors raise ketoacidosis risk, so the two holds are planned together. Incretin therapy is covered in GLP-1 receptor agonists and tirzepatide.

Drug interactions and laboratory interference

SGLT2 inhibitors have few pharmacokinetic interactions. Pharmacodynamic interactions with diuretics and glucose-lowering drugs are described in the sections above. General principles are covered in cardiovascular drug interactions and dosing.

• UGT inducers (rifampin, phenytoin, phenobarbital, ritonavir) reduce exposure to canagliflozin, sotagliflozin, and bexagliflozin. During inducer therapy, the canagliflozin label raises the dose from 100 mg to 200 mg, with a maximum of 300 mg when eGFR is 60 or more.
• Digoxin exposure rises modestly with canagliflozin and with sotagliflozin 400 mg; digoxin concentrations are monitored. Dapagliflozin and empagliflozin do not alter digoxin pharmacokinetics.
• Lithium concentrations may fall when an SGLT2 inhibitor is started; lithium concentrations are checked more often during initiation and dose changes.
• Insulin and sulfonylureas raise hypoglycemia risk, and insulin reduction raises ketoacidosis risk; doses are reduced in moderation.
• Loop and thiazide diuretics add to volume depletion.
• Urine glucose and 1,5-anhydroglucitol tests are invalid during treatment.

Cost, access, and underuse of SGLT2 inhibitors

SGLT2 inhibitors remain underused in eligible patients. Registry studies in heart failure, CKD, and diabetes have found that a minority of eligible patients receive them, with lower use in women, older patients, Black patients, patients seen without cardiology or nephrology care, and patients with high out-of-pocket costs. Clinician factors include concern about the eGFR dip, misattribution of the drug to diabetes care only, and uncertainty about which specialty prescribes it.

Out-of-pocket cost is the main patient barrier in the United States. Generic dapagliflozin is available. Empagliflozin and dapagliflozin were among the first drugs with prices negotiated by Medicare under the Inflation Reduction Act, and the Medicare Part D annual out-of-pocket cap limits patient costs for beneficiaries. Manufacturer assistance programs, 90-day prescriptions, and formulary preference for one agent improve access. Because the outcome evidence is consistent between dapagliflozin and empagliflozin for heart failure and CKD, the less expensive of the two on a given plan is an appropriate choice. When neither is affordable in HFrEF, the other three foundational classes remain indicated.

Systems that increase use include in-hospital initiation before discharge, electronic health record prompts for patients with heart failure or albuminuric CKD, pharmacist-led titration clinics, and shared protocols between cardiology, nephrology, endocrinology, and primary care that assign responsibility for starting the drug.

Sources of error in SGLT2 inhibitor prescribing

• Stopping the drug for the expected early eGFR dip: a stable fall of up to about 30% in the first weeks is hemodynamic, and discontinuation forfeits long-term kidney and cardiac protection.
• Withholding the drug because the patient has no diabetes or because HbA1c is at target: the heart failure and kidney benefits do not depend on diabetes or glucose lowering.
• Withholding the drug in heart failure or CKD because eGFR is below the glycemic threshold: the glycemic thresholds (45 or 30) do not apply to the cardiorenal indications (20 for empagliflozin, 25 for dapagliflozin).
• Prescribing an agent without the indication: canagliflozin, ertugliflozin, and bexagliflozin have no heart failure indication, and canagliflozin is labeled for kidney protection only in diabetic nephropathy with albuminuria.
• Titrating empagliflozin to 25 mg or canagliflozin to 300 mg for heart or kidney benefit: the higher doses add only glucose lowering.
• Dismissing nausea, vomiting, dyspnea, or malaise because glucose is normal: euglycemic ketoacidosis is diagnosed by blood ketones and acid-base status, not glucose.
• Continuing the drug up to the day of surgery, or restarting it before the patient eats: the labels require a hold of at least 3 days (4 for ertugliflozin) and restart only with normal oral intake and clinical stability.
• Cutting insulin sharply when the drug is started in a patient with insulin-deficient diabetes: the resulting insulin deficiency is the principal precondition for ketoacidosis.
• Starting the drug in a hypovolemic patient or adding it to intensive diuresis without reducing the loop diuretic in a patient with low blood pressure: the added natriuresis produces volume depletion and hypotension.
• Starting the drug after myocardial infarction solely because of the infarction: neither DAPA-MI nor EMPACT-MI reduced the composite of death and heart failure hospitalization (the death component was death from any cause in EMPACT-MI and cardiovascular death in DAPA-MI). The fewer heart failure hospitalizations in EMPACT-MI were a secondary finding after a neutral primary result, and the drug is given for an established indication.
• Using urine glucose or 1,5-anhydroglucitol to monitor glycemic control in a treated patient: both are unreliable during treatment.
• Starting the drug in polycystic kidney disease, in patients on dialysis, or in type 1 diabetes outside a specialist protocol: these groups were excluded from the outcome trials.

Prognosis and expected benefit by indication

Absolute benefit depends on baseline risk. In HFrEF, treating about 21 patients for 18 months prevents one cardiovascular death or worsening heart failure event (DAPA-HF). In a cross-trial analysis, the combination of an angiotensin receptor-neprilysin inhibitor, a beta blocker, a mineralocorticoid receptor antagonist, and an SGLT2 inhibitor was estimated to add 2.7 years (at age 80) to 8.3 years (at age 55) free from cardiovascular death or first heart failure hospitalization. The estimated gain in survival was 1.4 to 6.3 years. Both estimates are relative to an ACE inhibitor or angiotensin receptor blocker plus a beta blocker. In HFmrEF and HFpEF, the main gain is fewer heart failure hospitalizations and better symptoms, with a smaller effect on cardiovascular death. In CKD, the absolute benefit is largest in patients with heavy albuminuria and rapid eGFR decline, in whom a few years of treatment can delay kidney failure substantially. In type 2 diabetes without heart failure or CKD, the absolute reduction in heart failure hospitalization is small but persistent. Across all groups, the benefit accrues within weeks for heart failure and over years for kidney function, so the drug is started early and continued long term unless it causes a specific adverse effect.

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 2022;145:e1033, 2022;146:e185, and 2023;147:e674). doi:10.1161/CIR.0000000000001063
2. Maddox TM, Januzzi JL Jr, Allen LA, et al. 2024 ACC expert consensus decision pathway for treatment of heart failure with reduced ejection fraction: a report of the American College of Cardiology Solution Set Oversight Committee. J Am Coll Cardiol. 2024;83:1444–1488. doi:10.1016/j.jacc.2023.12.024
3. Kittleson MM, Panjrath GS, Bates K, et al. Management of heart failure with preserved ejection fraction: 2026 ACC expert consensus decision pathway. J Am Coll Cardiol. 2026;88:1817–1846. doi:10.1016/j.jacc.2026.06.018
4. McDonagh TA, Metra M, Adamo M, et al. 2023 focused update of the 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2023;44:3627–3639. doi:10.1093/eurheartj/ehad195
5. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int. 2024;105(4S):S117–S314. doi:10.1016/j.kint.2023.10.018
6. American Diabetes Association Professional Practice Committee for Diabetes. 9. Pharmacologic approaches to glycemic treatment: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(Suppl 1):S183–S215. doi:10.2337/dc26-S009
7. American Diabetes Association Professional Practice Committee for Diabetes. 10. Cardiovascular disease and risk management: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(Suppl 1):S216–S245. doi:10.2337/dc26-S010
8. Thompson A, Fleischmann KE, Smilowitz NR, et al. 2024 AHA/ACC/ACS/ASNC/HRS/SCA/SCCT/SCMR/SVM guideline for perioperative cardiovascular management for noncardiac surgery: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2024;150:e351–e442 (correction 2024;150:e466). doi:10.1161/CIR.0000000000001285
9. Oprea AD, Mohamed B, Hepner DL, et al. Perioperative management of patients taking sodium-glucose cotransporter 2 inhibitors: Society for Perioperative Assessment and Quality Improvement (SPAQI) multidisciplinary consensus statement. Br J Anaesth. 2026;136:1776–1799. doi:10.1016/j.bja.2026.02.031
10. McMurray JJV, Solomon SD, Inzucchi SE, et al. Dapagliflozin in patients with heart failure and reduced ejection fraction (DAPA-HF). N Engl J Med. 2019;381:1995–2008. doi:10.1056/NEJMoa1911303
11. Packer M, Anker SD, Butler J, et al. Cardiovascular and renal outcomes with empagliflozin in heart failure (EMPEROR-Reduced). N Engl J Med. 2020;383:1413–1424. doi:10.1056/NEJMoa2022190
12. Zannad F, Ferreira JP, Pocock SJ, et al. SGLT2 inhibitors in patients with heart failure with reduced ejection fraction: a meta-analysis of the EMPEROR-Reduced and DAPA-HF trials. Lancet. 2020;396:819–829. doi:10.1016/S0140-6736(20)31824-9
13. Anker SD, Butler J, Filippatos G, et al. Empagliflozin in heart failure with a preserved ejection fraction (EMPEROR-Preserved). N Engl J Med. 2021;385:1451–1461. doi:10.1056/NEJMoa2107038
14. Solomon SD, McMurray JJV, Claggett B, et al. Dapagliflozin in heart failure with mildly reduced or preserved ejection fraction (DELIVER). N Engl J Med. 2022;387:1089–1098. doi:10.1056/NEJMoa2206286
15. Vaduganathan M, Docherty KF, Claggett BL, et al. SGLT-2 inhibitors in patients with heart failure: a comprehensive meta-analysis of five randomised controlled trials. Lancet. 2022;400:757–767. doi:10.1016/S0140-6736(22)01429-5
16. Voors AA, Angermann CE, Teerlink JR, et al. The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial (EMPULSE). Nat Med. 2022;28:568–574. doi:10.1038/s41591-021-01659-1
17. Bhatt DL, Szarek M, Steg PG, et al. Sotagliflozin in patients with diabetes and recent worsening heart failure (SOLOIST-WHF). N Engl J Med. 2021;384:117–128. doi:10.1056/NEJMoa2030183
18. Cox ZL, Collins SP, Hernandez GA, et al. Efficacy and safety of dapagliflozin in patients with acute heart failure (DICTATE-AHF). J Am Coll Cardiol. 2024;83:1295–1306. doi:10.1016/j.jacc.2024.02.009
19. Berg DD, Patel SM, Haller PM, et al. Dapagliflozin in patients hospitalized for heart failure: primary results of the DAPA ACT HF-TIMI 68 randomized clinical trial and meta-analysis of sodium-glucose cotransporter-2 inhibitors in patients hospitalized for heart failure. Circulation. 2025;152:1411–1422. doi:10.1161/CIRCULATIONAHA.125.076575
20. Perkovic V, Jardine MJ, Neal B, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy (CREDENCE). N Engl J Med. 2019;380:2295–2306. doi:10.1056/NEJMoa1811744
21. Heerspink HJL, Stefánsson BV, Correa-Rotter R, et al. Dapagliflozin in patients with chronic kidney disease (DAPA-CKD). N Engl J Med. 2020;383:1436–1446. doi:10.1056/NEJMoa2024816
22. The EMPA-KIDNEY Collaborative Group; Herrington WG, Staplin N, Wanner C, et al. Empagliflozin in patients with chronic kidney disease. N Engl J Med. 2023;388:117–127. doi:10.1056/NEJMoa2204233
23. Nuffield Department of Population Health Renal Studies Group; SGLT2 inhibitor Meta-Analysis Cardio-Renal Trialists' Consortium. Impact of diabetes on the effects of sodium glucose co-transporter-2 inhibitors on kidney outcomes: collaborative meta-analysis of large placebo-controlled trials. Lancet. 2022;400:1788–1801. doi:10.1016/S0140-6736(22)02074-8
24. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes (EMPA-REG OUTCOME). N Engl J Med. 2015;373:2117–2128. doi:10.1056/NEJMoa1504720
25. Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes (CANVAS Program). N Engl J Med. 2017;377:644–657. doi:10.1056/NEJMoa1611925
26. Wiviott SD, Raz I, Bonaca MP, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes (DECLARE-TIMI 58). N Engl J Med. 2019;380:347–357. doi:10.1056/NEJMoa1812389
27. Cannon CP, Pratley R, Dagogo-Jack S, et al. Cardiovascular outcomes with ertugliflozin in type 2 diabetes (VERTIS CV). N Engl J Med. 2020;383:1425–1435. doi:10.1056/NEJMoa2004967
28. Bhatt DL, Szarek M, Pitt B, et al. Sotagliflozin in patients with diabetes and chronic kidney disease (SCORED). N Engl J Med. 2021;384:129–139. doi:10.1056/NEJMoa2030186
29. Garg SK, Henry RR, Banks P, et al. Effects of sotagliflozin added to insulin in patients with type 1 diabetes (inTandem3). N Engl J Med. 2017;377:2337–2348. doi:10.1056/NEJMoa1708337
30. Butler J, Jones WS, Udell JA, et al. Empagliflozin after acute myocardial infarction (EMPACT-MI). N Engl J Med. 2024;390:1455–1466. doi:10.1056/NEJMoa2314051
31. James S, Erlinge D, Storey RF, et al. Dapagliflozin in myocardial infarction without diabetes or heart failure (DAPA-MI). NEJM Evid. 2024;3:EVIDoa2300286. doi:10.1056/EVIDoa2300286
32. Heerspink HJL, Perkins BA, Fitchett DH, Husain M, Cherney DZI. Sodium glucose cotransporter 2 inhibitors in the treatment of diabetes mellitus: cardiovascular and kidney effects, potential mechanisms, and clinical applications. Circulation. 2016;134:752–772. doi:10.1161/CIRCULATIONAHA.116.021887
33. Ferrannini E, Mark M, Mayoux E. CV protection in the EMPA-REG OUTCOME trial: a "thrifty substrate" hypothesis. Diabetes Care. 2016;39:1108–1114. doi:10.2337/dc16-0330
34. Chung YJ, Park KC, Tokar S, et al. Off-target effects of sodium-glucose co-transporter 2 blockers: empagliflozin does not inhibit Na+/H+ exchanger-1 or lower [Na+]i in the heart. Cardiovasc Res. 2021;117:2794–2806. doi:10.1093/cvr/cvaa323
35. Adamson C, Docherty KF, Heerspink HJL, et al. Initial decline (dip) in estimated glomerular filtration rate after initiation of dapagliflozin in patients with heart failure and reduced ejection fraction: insights from DAPA-HF. Circulation. 2022;146:438–449. doi:10.1161/CIRCULATIONAHA.121.058910
36. Neuen BL, Oshima M, Agarwal R, et al. Sodium-glucose cotransporter 2 inhibitors and risk of hyperkalemia in people with type 2 diabetes: a meta-analysis of individual participant data from randomized, controlled trials. Circulation. 2022;145:1460–1470. doi:10.1161/CIRCULATIONAHA.121.057736
37. Curtain JP, Docherty KF, Jhund PS, et al. Effect of dapagliflozin on ventricular arrhythmias, resuscitated cardiac arrest, or sudden death in DAPA-HF. Eur Heart J. 2021;42:3727–3738. doi:10.1093/eurheartj/ehab560
38. Zelniker TA, Bonaca MP, Furtado RHM, et al. Effect of dapagliflozin on atrial fibrillation in patients with type 2 diabetes mellitus: insights from the DECLARE-TIMI 58 trial. Circulation. 2020;141:1227–1234. doi:10.1161/CIRCULATIONAHA.119.044183
39. Peters AL, Buschur EO, Buse JB, et al. Euglycemic diabetic ketoacidosis: a potential complication of treatment with sodium-glucose cotransporter 2 inhibition. Diabetes Care. 2015;38:1687–1693. doi:10.2337/dc15-0843
40. Danne T, Garg S, Peters AL, et al. International consensus on risk management of diabetic ketoacidosis in patients with type 1 diabetes treated with sodium-glucose cotransporter (SGLT) inhibitors. Diabetes Care. 2019;42:1147–1154. doi:10.2337/dc18-2316
41. Bersoff-Matcha SJ, Chamberlain C, Cao C, Kortepeter C, Chong WH. Fournier gangrene associated with sodium-glucose cotransporter-2 inhibitors: a review of spontaneous postmarketing cases. Ann Intern Med. 2019;170:764–769. doi:10.7326/M19-0085
42. FARXIGA (dapagliflozin) tablets: US prescribing information. AstraZeneca Pharmaceuticals; 2026. DailyMed
43. JARDIANCE (empagliflozin) tablets: US prescribing information. Boehringer Ingelheim Pharmaceuticals; revised January 2026. DailyMed
44. INVOKANA (canagliflozin) tablets: US prescribing information. Janssen Pharmaceuticals; revised June 2026. DailyMed
45. INPEFA (sotagliflozin) tablets: US prescribing information. Lexicon Pharmaceuticals; 2026. DailyMed
46. Lambers Heerspink HJ, de Zeeuw D, Wie L, Leslie B, List J. Dapagliflozin a glucose-regulating drug with diuretic properties in subjects with type 2 diabetes. Diabetes Obes Metab. 2013;15:853–862. doi:10.1111/dom.12127
47. Santos-Gallego CG, Vargas-Delgado AP, Requena-Ibanez JA, et al. Randomized trial of empagliflozin in nondiabetic patients with heart failure and reduced ejection fraction (EMPA-TROPISM). J Am Coll Cardiol. 2021;77:243–255. doi:10.1016/j.jacc.2020.11.008
48. 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: a report of the American College of Cardiology Solution Set Oversight Committee. J Am Coll Cardiol. 2024;84:1241–1267. doi:10.1016/j.jacc.2024.06.002
49. Butler J, Packer M, Filippatos G, et al. Effect of empagliflozin in patients with heart failure across the spectrum of left ventricular ejection fraction. Eur Heart J. 2022;43:416–424. doi:10.1093/eurheartj/ehab798
50. Vaduganathan M, Claggett BL, Jhund PS, et al. Estimating lifetime benefits of comprehensive disease-modifying pharmacological therapies in patients with heart failure with reduced ejection fraction: a comparative analysis of three randomised controlled trials. Lancet. 2020;396:121–128. doi:10.1016/S0140-6736(20)30748-0