Sodium Glucose Cotransporter 2 Inhibitors: Searching for Mechanisms in the Wake of Large, Positive Cardiovascular Outcomes Trials

Mark C. Petrie, MBChB


Circulation. 2019;140(21):1703-1705. 

At the 2019 European Society of Cardiology congress, sodium glucose cotransporter 2 (SGLT2) inhibitors were firmly planted at the feet of cardiologists. The Dapa-HF trial (Study to Evaluate the Effect of Dapagliflozin on the Incidence of Worsening Heart Failure or Cardiovascular Death in Patients With Chronic Heart Failure) reported a clear-cut clinical benefit when these drugs are used to treat patients with established heart failure and reduced ejection fraction without (as well as with) diabetes mellitus.[1] These diabetes drugs should not now be left solely in the hands of diabetologists for use in people with diabetes mellitus. Their benefits in preventing heart failure had been previously firmly established in 4 large, well-conducted cardiovascular outcomes trials,[2] but this cutting of the diabetic umbilical cord should help with their acceptance and adoption by cardiologists.[3,4]

In this issue of Circulation, Verma and colleagues[5] have investigated the mechanism of action of the cardiac effects of empagliflozin. A rigorously designed, placebo-controlled, randomized mechanistic trial was conducted, EMPA-HEART (Effects of Empagliflozin on Cardiac Structure in Patients With Type 2 Diabetes). Such trials require sweat and devotion from the trial team and large donations of goodwill from patients. Empagliflozin reduced the primary end point of indexed left ventricular mass (measured by cardiac magnetic resonance imaging) in an EMPA-REG OUTCOME-like cohort (BI 10773 [Empagliflozin] Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients). This reduction in left ventricular mass may have been simply attributable to the sizeable reduction in blood pressure (7 mm Hg systolic and 3 mm Hg diastolic) or attributable to many other, as yet unidentified, mechanisms. These are novel and enlightening data, but this is very much a first step in understanding how these drugs work. In trials like EMPA-HEART, there are limits to the amount of data that can be gathered. Patients need to be protected from an excessive burden of trial-related procedures. Multiple imaging techniques, frequent large-volume phlebotomy, and many return visits are unreasonable to request from patients who have volunteered their time and patience. As a result, EMPA-HEART does not shed light on cardiac metabolism, renal, diuretic, vascular, or other mechanisms.

EMPA-HEART was designed to answer the pertinent question when the trial was conceived: the mechanism of benefit in the first SGLT2 cardiovascular outcome trial (EMPA-REG-OUTCOME[6]). When we consider mechanisms that result in the benefits observed in renal disease (CREDENCE[7] [Evaluation of the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Participants With Diabetic Nephropathy]) or heart failure with reduced ejection fraction (Dapa-HF), EMPA-HEART cannot help. Patients with estimated glomerular filtration rates <60 mL·min−1·1.73 m−2 were excluded and only 2 patients had heart failure.

Many research groups are searching for the mechanism of action, but it seems unlikely that there is one holy-grail answer. Multiple editorials and opinion pieces with colorful diagrams have postulated scores of elegant (and sometimes confusing) theories, but little firm evidence exists from studies in humans with cardiovascular disease.[8] EMPA-HEART is one of the first of many SGLT2 inhibitor mechanistic trials (in humans) to report. Many more trials will be required to piece together what are likely to be different mechanisms of action in the diverse patient populations in which benefits have been shown. There are at least 4 broad populations that have been investigated: those with and without established atherosclerotic cardiovascular disease, those with established heart failure, and those with renal impairment.

In the EMPA-REG-OUTCOME–, CANVAS (Canagliflozin Cardiovascular Assessment Study)–, and DECLARE (Multicenter Trial to Evaluate the Effect of Dapagliflozin on the Incidence of Cardiovascular Events)–like populations,[2] like the one recruited in EMPA-HEART, there are potential groups that might benefit from well-established effects of these drugs: patients with obesity (weight loss), those with hypertension (blood pressure reduction), patients with recognized or unrecognized heart failure (diuretic effects), and patients with renal disease (renal effects). Of course, there may be straightforward or complex reasons why each of these simple mechanisms might improve cardiovascular outcomes. For example, obesity has a myriad of effects on the cardiovascular system, so weight loss might result in benefit by many mechanisms. Similarly, blood pressure reduction might result in improved cardiac function by simple mechanisms (reduced left ventricular mass or improved cardiac index) or by a host of secondary mechanisms. Many of the more elaborate theories that have nothing to do with obesity, blood pressure, or renal disease might hold true, but these must be proven and not remain as speculation.

Perhaps the most controversial mechanism of action is the diuretic effect of SGLT2 inhibitors. Patients with cardiovascular disease frequently have unrecognized congestion. Clinicians often do not increase diuretics unless patients display clear evidence of edema or pulmonary rales. A reduction in venous pressure after SGLT2 inhibition as a result of diuresis might be a major mediator of renal benefits. A consequent increase in the transrenal blood pressure gradient might result in improved renal perfusion. At present, we know little of the diuretic and hemodynamic effects of these agents, and these were outside the scope of EMPA-HEART.

It is fashionable to state that glucose lowering is in no way responsible for the mechanism of SGLT2 inhibitor clinical efficacy. Two potential glucose-related mechanisms are possible, or even likely. The first is that quality of life might be improved with tighter glycemic control (fewer hyperglycemic symptoms). A second possible mechanism is a SGLT2 inhibitor–mediated improvement in glucose-related cardiac or cardiovascular abnormalities (diabetic cardiomyopathy[5]). Of course, glucose lowering cannot be responsible for beneficial effects in patients without diabetes mellitus. The benefits seen in patients without diabetes mellitus in the Dapa-HF trial[1] will spark a swathe of trials looking at potential mechanisms of action in those without diabetes mellitus.

Do SGLT2 inhibitors have more of a role in cardiology than has already been established? It is very likely that patients with post–myocardial infarction heart failure (or severe left ventricular dysfunction) are likely to benefit. Most of the drug classes that are used as evidence-based, guideline-directed heart failure treatments have proven to be beneficial after large myocardial infarctions: angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, β-blockers, and mineralocorticoid receptor antagonists. If a trial of SGLT2 inhibitors were conducted after large myocardial infarctions, this would be very likely to lead to these drugs joining this list of coronary care unit stalwarts.

SGLT2 inhibitors have now been proven to be very useful as cardiology drugs. Can we adopt any more of the diabetes drugs into the cardiology armamentarium? Glucagon-like protein 1 receptor agonists (GLP1-RAs) may have a small beneficial effect in preventing heart failure in patients with risk factors or established cardiovascular disease.[9] Details of the phenotype of the heart failure that is prevented are currently lacking. There is so far no evidence that GLP1-RAs can be used as treatments for heart failure. Two small trials of GLP1-RAs in heart failure with reduced ejection fraction were discouraging but not terminal.[10,11] It seems possible, or maybe probable, that GLP1-RAs could benefit patients with heart failure with preserved ejection fraction. The GLP1-RAs cause substantial weight reduction, blood pressure reduction, and a reduction in inflammation and atherosclerosis. These are all prevalent in heart failure with preserved ejection fraction.

What will become of the other drugs formally known as drugs for diabetes mellitus? Drugs used in patients with diabetes mellitus should not just be glucose lowering, but also have benefits in terms of reducing cardiovascular events. Metformin is under scrutiny because the evidence of cardiovascular benefit appears weak when juxtaposed with evidence from modern cardiovascular outcomes trials.[12,13] The gradual relegation that is taking place in guidelines is warranted.[14] The only nonneutral cardiovascular result in trials of dipeptidyl peptidase-4 inhibitors has been the negative finding of increased heart failure hospitalizations caused by saxagliptin.[15] Although the safety of dipeptidyl peptidase-4 inhibitors has been trumpeted, is it acceptable to recommend drugs with no cardiovascular benefit, even if they may offer some microvascular benefits over the longer term?

These are exciting times for SGLT2 inhibitors. Many large cardiovascular outcome trials are ongoing in heart failure and renal disease in patients with and without diabetes mellitus, so the coming few years will see a wealth of new evidence. The host of mechanistic trials that are progressing may steadily fill in the mechanistic jigsaw puzzle, although these insights will continue to lag behind the clinical benefit that has blazed the way.