EMPA-REG: Does 'Super Fuel' Explain Surprise Effect?

Eleuterio Ferrannini, MD, PhD


August 02, 2016

This feature requires the newest version of Flash. You can download it here.

Early this morning, I participated in a session dedicated to understanding the mechanisms, or the potential mechanisms, by which the EMPA-REG OUTCOME study[1] got its results. These results were unexpected in a positive way, even to the investigators. The main result was the large reduction in cardiovascular (CV) and total mortality, and in hospital admissions for heart failure, which are big and increasing problems in patients with type 2 diabetes.

Because the treatment benefit with empagliflozin was seen relatively early after study randomization and continued at a nearly constant rate for the 3-year study duration, the question in everybody's mind has been: What could possibly account for the large reduction in CV mortality and hospitalization for heart failure in these patients with type 2 diabetes and a heavy load of CV risk?

Previous CV outcome trials with antidiabetic drugs have been mostly neutral. The size and the time of the effect have led most people to argue that results could not possibly be due to a reduction in atherosclerosis or even a slowing down of atherogenesis; curves between the placebo and the treatment arm begin to diverge in a matter of a few weeks after randomization, which is too short to be due to atherosclerotic CV disease.

This has aroused a lot of attention and thinking, and this session was devoted to new ideas: What else could potentially explain this result? Some people argued that sodium-glucose cotransporter-2 (SGLT2) inhibitors cause a reduction in blood pressure—not as large as with an antihypertensive, but consistent across members of the class—which is particularly marked in the first few weeks after randomization. This would clearly reduce the afterload to the heart and be beneficial for reducing cardiac work. Other people argued that the increase in hematocrit due to hemoconcentration would also reduce the preload to the heart and thereby reduce the burden on a heart which is at the brink of breaking down.

The hypothesis presented today is not in contrast with these two possibilities, which clearly must have made some contribution to the trial benefit, but is complementary and more basic. The hypothesis hinges upon metabolic changes that may have occurred as a consequence of the glucosuria induced by the SGLT2 inhibition, which can be beneficial simultaneously to the heart and to the kidney. As is going to be presented at this meeting, progression of renal disease, progression of the microalbuminuria, and doubling of serum creatinine were also positively affected in the trial, and to almost the same extent as the CV outcomes.[2]

Why was the effect size the same in two completely different organs and the CV benefit still demonstrable even in people with impaired renal function in the data from EMPA-REG? Some people hypothesize that one way in which the function of the heart and the kidney could be improved in the EMPA-REG population is through a change in substrate composition; by that, one means the mixture of substrates that is offered to the kidney and to the heart to be utilized for energy production.

A study is not just good for what it shows, but also for the questions that it raises.

The discussion and the argument have been about the possibility that the moderate rise in ketonemia that is observed in patients once they go on an SGLT2 inhibitor may be responsible for these metabolic changes by making available beta-hydroxybutyrate—a "super fuel"—because it can be taken up freely by both the kidney and the heart, and it is not insulin dependent. It can be utilized as a substrate to produce energy in a very efficient way; by efficient, one means using less oxygen for the same amount of adenosine triphosphate (ATP) produced, or conversely, using the same amount of oxygen to produce more ATP, to then be invested in contractility.

The session really described some of the cellular mechanisms whereby this shift in substrate delivery to the organs—from the liver producing ketone bodies to the heart and the kidney utilizing beta-hydroxybutyrate—could have made a contribution to the observed outcomes.

An important part of this discussion is that a study is not just good for what it shows, but also for the questions that it raises in terms of new biology or physiology that could be learned from the results. We are going to see in the near future a flourishing of studies aimed at defining the circumstances, the extent, and the patient population in which these new mechanisms may be operative.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: