Recent clinical trials and observational studies show that, on average, 30% to 40% of the patients with heart failure have diabetes mellitus. The proportion is similar for patients with chronic heart failure, patients with acute heart failure, and patients with either HFREF or HFPEF; however, it may increase to more than 50% in some geographic areas, such as the Middle East.

Conversely, heart failure events are very common in patients with diabetes mellitus. In particular, the prevalence of heart failure hospitalizations has been largely underestimated; for example, in the control arm of some clinical trials that included people with type 2 diabetes mellitus, heart failure events are as common or even more frequent than coronary events. Data from large international registries suggest that the presence of diabetes is a strong predictor of future heart failure hospitalizations. A recent meta-analysis pooling more than 70 000 people with diabetes mellitus shows that each one point increase in HbA is associated with a 20% increase in the risk of heart failure hospitalizations.

The coexistence of heart failure and diabetes mellitus is also associated with a significant increase in the risk of rehospitalizations for heart failure and in-hospital or 1-year mortality. In particular, the level of glycemia at referral is a strong predictor of in-hospital mortality. Therefore, these data confirm that the association of heart failure with diabetes mellitus is deadly.

There is no fundamental difference in the management of HFREF patients with or without diabetes mellitus. All classes of medications that showed benefit in HFREF patients are equally effective in patients with diabetes mellitus: this holds true for angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, B-adrenergic blockers, mineralocorticoid receptor antagonists, ivabradine, and sacubitril/valsartan. The evidence is derived from subgroup analyses performed on the large outcome trials that were conducted with these classes of heart failure medications. The magnitude of the effect on clinical outcomes is similar to that observed in people without diabetes; however, since people with diabetes have a higher cardiovascular risk than those without diabetes, the absolute benefit is usually greater in people with diabetes. 

The safety is also excellent, except for the increase in the risk of hyperkalemia or renal function deterioration observed with renin-angiotensin-aldosterone system blockers. This observation derives from subanalyses of the CHARM trial with candesartan, the EPHESUS trial with eplerenone, and observational studies. As a result, regular monitoring of renal function and kalemia is recommended after the initiation or during the uptitration of renin-angiotensin-aldosterone system inhibitors in patients with both diabetes mellitus and heart failure.

GLP-1 analogs 

This new drug class has been extensively studied in patients with diabetes mellitus. The long-acting GLP-1 analog liraglutide was tested in LEADER, whereas the very long-acting semaglutide was evaluated in SUSTAIN.. Both drugs significantly reduced the primary outcome in patients with diabetes and a high cardiovascular risk; however, there was a nonsignificant risk reduction in heart failure hospitalizations. These results were also observed in the recently presented EXSCEL trial that used the long-acting exenatide, where there was no increase (but also no improvement) in the risk of cardiovascular events, specifically in heart failure hospitalizations. However, two small studies conducted in HFREF patients with or without diabetes mellitus provided a negative signal on outcomes. One potential explanation is that GLP-1 analogs increase heart rate; this effect might be harmful in patients with reduced cardiac function as demonstrated in heart failure. In addition, only a minority of patients enrolled in the large outcome trials had a his- tory of heart failure. Therefore, uncertainty remains on the cardiovascular safety of GLP-1 analogs in patients with a reduced ejection fraction.

SGLT inhibitors

A cotransporter of sodium and glucose exists in the proximal tubule (type 1 and principally type 2) of the kidneys (type 2) and in the gut (type 1). In normal conditions, this cotransporter reabsorbs the totality of the glucose that is filtered in the glomeruli together with sodium. Blocking this cotransporter with inhibitors causes osmotic glycosuria that leads to better glycemic control in diabetes mellitus and better sodium excretion. Empagliflozin and canagliflozin have been tested in large outcome trials, EMPA-REG OUTCOME and CANVAS, respectively, in large populations of diabetic patients with a high cardiovascular risk.10.11 Both trials showed a significant reduction in the primary end point, which is mainly driven by a reduction in mortality and a spectacular reduction in heart failure hospitalizations. The beneficial effect on heart failure events remains unclear; however, possible hypotheses for SGLT inhibitors include a diuretic effect without neurohormonal stimulation, a reduction in blood pressure, an improvement in kidney function, and a direct myocardial metabolic effect. The benefit observed on heart failure

The cardiovascular safety of antidiabetic drugs raised concern following a controversy with rosiglitazone, a thiazolidinedione that was developed in the early 2000s. A meta-analysis suggested an increased risk of myocardial infarction in patients treated with rosiglitazone, a concern that was not confirmed by subsequent analyses. Nevertheless, this prompted the Food and Drug Administration, followed by the European Medicines Agency, to establish rules regarding development programs of new antidiabetic drugs in order to detect potential harm in terms of cardiovascular events. 

There is a lack of robust data on the cardiovascular safety of sulfonylureas, metformin, and insulin because these drugs were widely used for patients with diabetes mellitus before the onset of evidenced-based medicine. Therefore, most of the available data are derived from observational studies. 

Although heart failure is a classic contraindication for the use of metformin due to the risk of lactic acidosis, there is no signal for potential harm when using this drug in patients with diabetes and heart failure. On the contrary, observational data suggest that metformin is associated with improved outcomes, such as mortality, compared with the control therapy of sulfonylurea. Data regarding sulfonylureas are somewhat difficult to interpret, but some large observational studies suggest that the risk of heart failure is significantly increased compared with metformin, particularly with first-generation sulfonylureas.

Although insulin is associated with water and sodium retention and sympathetic nervous system activation during hypoglycemia, there are no data suggesting harm in terms of cardiovascular outcomes in patients with diabetes. One randomized controlled trial, ORIGIN, was conducted with insulin glargine, and it showed no increase in heart failure hospitalizations in the insulin arm compared with the control arm.

Thiazolidinediones.

Pioglitazone and rosiglitazone, two thiazolidinediones, were developed in large outcome trials in combination with traditional antidiabetic drugs. Both drugs are associated with a substantial increase in the risk of heart failure since these PPAR-y agonists upregulate a channel in the kidneys, that leads to sodium retention. As a result, these drugs should not be used in patients with diabetes who have overt heart failure or significantly impaired cardiac function.

DPP4 inhibitors

DPP4 inhibitors are a widely used new class of glucose-lowering agents. They were tested in large outcome trials and were shown to be neutral toward myocardial ischemic events. However, saxagliptin, which was tested in the SAVOR-TIMI 53 trial, was surprisingly associated with a 27% increased risk of heart failure hospitalizations compared with the control arm., Older patients and patients with a history of heart failure or high baseline plasma BNP levels were more likely to develop heart failure. Another DPP4 inhibitor, sitagliptin, was tested in a large outcome trial, TECOS, however, there was no negative signal on heart failure events with this compound. There is no plausible biological explanation for why saxagliptin was associated with an increased risk of heart failure; although DPP4 has multiple substrates, including vasoactive peptides, it also degrades BNP so that inhibition of DPP4 results in elevated levels of this peptide, which has beneficial effects on the cardiovascular system. The play of chance cannot be excluded to explain why saxagliptin was associated with an increased risk of heart failure events in SAVOR- TIMI 53.

Nevertheless, when considering the introduction of a DPP4 inhibitor in a patient with diabetes and heart failure, preference should be given to sitagliptin due to its neutral effect on heart failure hospitalizations.

GLP-1 analogs

This new drug class has been extensively studied in patients with diabetes mellitus. The long-acting GLP-1 analog liraglutide was tested in LEADER, whereas the very long-acting semaglutide was evaluated in SUSTAIN.. Both drugs significantly reduced the primary outcome in patients with diabetes and a high cardiovascular risk; however, there was a nonsignificant risk reduction in heart failure hospitalizations. These results were also observed in the recently presented EXSCEL trial that used the long-acting exenatide, where there was no increase (but also no improvement) in the risk of cardiovascular events, specifically in heart failure hospitalizations. However, two small studies conducted in HFREF patients with or without diabetes mellitus provided a negative signal on outcomes. One potential explanation is that GLP-1 analogs increase heart rate; this effect might be harmful in patients with reduced cardiac function as demonstrated in heart failure. In addition, only a minority of patients enrolled in the large outcome trials had a his- tory of heart failure. Therefore, uncertainty remains on the cardiovascular safety of GLP-1 analogs in patients with a reduced ejection fraction. 

SGLT inhibitors

A cotransporter of sodium and glucose exists in the proximal tubule (type 1 and principally type 2) of the kidneys (type 2) and in the gut (type 1). In normal conditions, this cotransporter reabsorbs the totality of the glucose that is filtered in the glomeruli together with sodium. Blocking this cotransporter with inhibitors causes an osmotic glycosuria that leads to better glycemic control in diabetes mellitus and better sodium excretion. Empagliflozin and canagliflozin have been tested in large outcome trials, EMPA-REG OUTCOME and CANVAS, respectively, in large populations of diabetic patients with a high cardiovascular risk.10.11 Both trials showed a significant reduction in the primary end point, which is mainly driven by a reduction in mortality and a spectacular reduction in heart failure hospitalizations. The beneficial effect on heart failure events remains unclear; however, possible hypotheses for SGLT inhibitors include a diuretic effect without neurohormonal stimulation, a reduction in blood pressure, an improvement in kidney function, and a direct myocardial metabolic effect. The benefit observed on heart failure events prompted companies developing SGLT inhibitors to consider conducting new clinical trials on heart failure patients with or without diabetes mellitus and with or without a reduced ejection fraction. While these trials are conducted, SGLT
inhibitors appear to be a treatment of choice for patients with diabetes mellitus and heart failure. The safety issues are mainly urine infections and an increase in the risk of amputations that is possibly related to hemoconcentration.

The combination of diabetes mellitus and heart failure is common and it is associated with poor outcomes. The management of HFREF is similar in patients with or without diabetes mellitus, although there is an increased risk of hyperkalemia when using renin-angiotensin-aldosterone system inhibitors. Thiazolidinediones should not be used in patients with heart failure. There is uncertainty regarding the safety of saxagliptin in patients with heart failure, whereas another DPP4 inhibitor, sitagliptin, is neutral. GLP-1 analogs have a beneficial effect on cardiovascular events, but there is some concern on their use in patients with diabetes and a reduced ejection fraction. Finally, SGLT inhibitors have a strong beneficial effect on heart failure events, although the mechanism of this effect remains unclear.