How to manage the transition from diabetes to heart failure and arrhythmias

Epidemiology

In the Framingham Heart Study, diabetes mellitus type 2 (T2DM) increases the risk of heart failure (HF) up to 2-fold in men and 5-fold in women (1), independently with age, hypertension, hypercholesterolaemia, or coronary artery disease (CAD). More recent studies show that diabetic patient has 30% higher risk to develop HF (2) and ~33% increase of risk of hospitalisation for HF (3). HF patients with diabetes are at increased risk for mortality and re-hospitalisation after hospitalisation for HF (4). Conversely, HF is an independent risk for developing type 2 diabetes. (5) The association of diabetes and HF is of dire prognosis (6) (7).

Atrial fibrillation (AF) incidence increases 1.4- to 1.6-fold in diabetic patients (8). Although some conflicting data exist, possibly reflecting a shared risk factor profile of diabetes mellitus with systemic inflammation, autonomic dysfunction, obesity, CAD, and HF, T2DM seems to be independently associated with increased risk of AF (9). In a recent observational study of AF patients, diabetes was associated with worse AF symptoms, lower quality of life, and increased risk of death and hospitalisations (10). A higher occurrence of bradyarythmias has been too recently reported, resulting in an increased demand for pacemaker treatment.[11]

Diabetes is also a risk factor for ischaemic stroke in patients with AF, and has been therefore included in the stroke risk scoring systems which are used to guide anticoagulation therapy in AF patients.

Pathophysiology

1. T2DM and HF. Ventricle dysfunction in patients with diabetes, independently of hypertension or atherosclerotic coronary disease, defines the Diabetic Cardiomyopathy (12), in which toxic lipid accumulation in the heart may represent the essential hallmark (13). Cardiac hypertrophy (CH) (14), microvascular disease (15), endothelial dysfunction (16) are the major actors of ventricular remodeling, leading to systolo-diastolic dysfunction. However, because diabetes is often associated with obesity, hypertension and dyslipidemia, one must be aware of the complexity of an interrelated pathophysiology that may provoke HF by multiple biochemical, anatomical and functional alterations in cardiomyocytes (17) Beyond the basic concept of diabetic myocardiopathy, the role of the triade (Diabetic cardiomyopathy, Left ventricular (LV) hypertrophy and CAD), recently extended to the “cardiotoxic tetrad” (Fluid overload contributing in increasing pressure overload in a stiffened ventricle) (18) has been pointed out (19).
2. T2DM and AF. T2DM promotes profibrillatory structural remodeling (changes in atrial tissue properties, size, and cellular ultrastructure) mediated by oxidative stress, advanced glycosylation end products, and connective tissue modifications. Associated electromechanical, and autonomic nervous system changes finally contribute to the onset of FA. (9) (20)
3. Potential underlying mechanisms. Diabetes mellitus is associated with increased myocardial Fat Acid (FA) utilization, decreased glucose utilization (glycolysis and glucose oxidation), increased myocardial oxygen consumption and decreased cardiac efficiency. Multiple hypothesis may account for myocardial dysfunction and arrhythmogenic disturbances in T2DM patients; they are resumed in Table 1. More extensive data may be found in two recent reviews (17)( 21).

Table 1: Potential Mechanisms Underlying Diabetic Cardiomyopathy

(adapted from (16) and (20)

How to detect HF in patients with T2DM

HF represents one of the most common (14.7%) initial presentations of CVD in T2DM patients. This suggests the need for earlier screening. (22) However, no present data allow a systematic screening for features of HF in asymptomatic people with T2DM. Were such a strategy to be adopted, careful selection of patients would be critical. Patients should be asked if they have symptoms of HF and if they can climb more than one flight of stairs without having uncomfortable dyspnoea. However, early myocardial dysfunction may not be identified by basic clinical evaluation. A two-dimensional and pulsed Doppler echocardiography might be useful to visualize the cardiac and structural changes that underlie HF: abnormal LV systolic function, diastolic dysfunction, and changes in LV geometry. (19) (21)

A recent analysis has shown three patterns in this population:

1. a group with systolic and diastolic function that is less impaired than the remainder, associated with a favorable prognosis;
2. patients (mostly women) with obesity, hypertension and diastolic dysfunction;
3. a cluster (mostly men) with LV hypertrophy and systolic dysfunction.

The latter two groups have a less favorable prognosis. (23) Plasma BNP has been shown to have a sensitivity of 92% and a specificity of 72% in detecting HF with reduced Ejection Fraction (HFrEF). Its diagnostic value for HF with preserved EF (HFpEF) has been evaluated in a recent systematic review [ Remmelzwaal S. et al, BMC in press] Thus, a natriuretic peptide–based screening, based on low cutoff levels, may be effective for detecting ventricular dysfunction. Use of other biomarkers have also been reported (circulating microribonucleic acids or glucose metabolites) but no consensus has yet been reached on the clinical role of such strategies in T2DM patients. (21)

How to prevent HF and AF in patients with T2DM

The incidence of cardiovascular outcomes in patients with diabetes and HF is related to the level of hyperglycaemia as indicated by the haemoglobin A1c (HbA1c). (24) However, there is little evidence to indicate that improved glycemic control itself improves HF outcomes in this population. (17)(25) In a meta-analysis of randomized controlled studies, intensive glucose lowering compared with less stringent control had no impact on HF events (26) The assumption has been made that the excess risk associated with type 2 diabetes may depend, not only on the hyperglycemia level, but on the associated risk-factors. A recent cohort study (27) evaluated cardiovascular outcomes in T2DM patients as compared with controls, according to five risk-factor variables (HbA1c and LDL cholesterol level, arterial blood pressure, albuminuria and smoking). Concerning the hospitalizations for HF, patients younger than 55 years with all five risk-factor variables outside the target ranges had the highest excess risk; the presence of atrial fibrillation, a high body-mass index, a glycated hemoglobin level and renal function outside the target were the strongest predictors. Interestingly, normalization of glycemia did not restore risk of cardiovascular disease (CVD) to the control population without diabetes at baseline, suggesting that the control of the whole spectrum of the risks (particularly obesity and renal function) is crucial inT2DM patients with HF, especially among the younger ones. Among antidiabetic drugs, considering the reductions in hospitalisation for HF observed in the recent trials of Sodium-glucose co-transporter-2 inhibitors (SGLT2i), these drugs may be of a great value in the primary prevention of HF in T2DM patients. ( 28)

Concerning AF, the mechanisms that may underpin its relation with T2DM remain speculative. (29) In a prospective study, patients with DM and incident AF had an increased risk for morbidity and mortality, but an intensive glycaemic control did not affect the rate of new-onset. (30) No specific strategy may be actually established concerning the prevention of AF in T2DM patients.

How to treat T2DM patients with HF

In patients with HF, interventions that reduce morbidity and mortality confer similar benefit in the presence or absence of diabetes. (31)

Renin Aldosterone Angiotensin System (RAAS) inhibitors

The activation of the RAAS, is a crucial anomaly occurring both in T2DM and in HF. Angiotensin Converting Enzyme (ACE) inhibitors reduce the risk for new onset HF in patients with established CVD or diabetes mellitus, and represent, with Ang II (angiotensin II) type 1 receptor blockade, a first line therapy for HF prevention in patients with diabetes mellitus.

Mineralocorticoid/aldosterone receptor antagonists

Aldosterone receptor antagonists, such Aldactone and Eplerenone may be of interest in the diabetic myocardiopathy, via a potential effect on inflammation and insulin resistance. (16)

Angiotensin receptor neprilysin inhibitor

Recently, Valsartan in combination with the neprilysin inhibitor, sacubitril, has shown impressive results on HFrEF management in general; its specific value in HfrEF subjects with T2DM has to be evaluated. (32)

Beta-blockade

Clinical guidelines support the use of β-blockade in individuals with T2DM and HF, with no regards about the potential increase in risk of hypoglycemia. Meta-analyses confirm the benefits of β-blockade in this population, although its magnitude is attenuated, possibly because of associated autonomic dysfunction. (17)

Glucose lowering drugs

(17) (24) (33) Unfortunately, most of the safety trials on new glucose lowering agents did not consider HF as a primary endpoint; only recently major trials do include, as a secondary endpoint, hospitalization for HF.

As shown in Table 2:

• Thiazolidinediones (TZDs; e.g. rosiglitazone and pioglitazone) increase insulin sensitivity, which may lead to increased sodium and water retention. In clinical trials, they have been associated with an increased risk for HF. These drugs are contraindicated in patients with HF.
• An increased risk for HF hospitalization was shown for Saxagliptin and a non-significant trend for Alogliptin. Such an effect was not found for Sitagliptin that remains the only dipeptidyl peptidase 4 inhibitor (DPP4-i) drug to be used in people with T2DM and HF.
• Glucagon like peptide 1 receptor agonists (GLP1-RA) reduced atherosclerotic major cardiovascular events (MACE), in T2DM patients with established atherosclerotic cardiovascular disease (ASCVD) but demonstrated a neutral effect on the risk of hospitalization for HF.
• Sodium-glucose co-transporter-2 inhibitors (SGLT2i) represent the first class of therapeutic glucose-lowering agents that have definitively reduced the risk of HF in T2DM patients (34,35,36). Overall, they significantly reduced the risk for the composite of cardiovascular death or hospitalization for HF by 23%, and hospitalization for HF by 31%, regardless of baseline atherosclerotic risk category or history of heart failure (28). An important reduction in progression of renal disease was consistently observed. Moreover, recent data show that Dapagliflozin (37) and Empagliflozin (38 ) reduce HF outcomes regardless the presence or absence of T2DM, with a 30% reduction of HF hospitalizations for both of them and a reduction of cardiovascular (18%) and all-cause mortality (17%) for Dapagliflozin. SGLT2i drugs where generally well tolerated and safe, except for the risk of genital infections. An increased risk of fractures and lower limb amputations was reported with canagliflozin, but not with empagliflozin or dapagliflozin.
  Finally, all these data represent an exciting challenge regarding the future management of T2DM and HF, and justify the recent (39) and impending modifications of our guidelines.

Table 2: Effects of antihyperglycaemic drugs on Major Adverse Cardiac Events (MACE) and HF hospitalisations

(Adapted from D. H. Fitchett (23))

How to treat T2DM patients with AF

AF treatment in T2DM patients must rely on the current guidelines (40) The frequent association with other risks factors (renal failure, obesity etc.) must be taken into account. A longer duration of diabetes appears to confer a higher risk of thrombo-embolism, but without greater risk of bleeding events. Intensive glycaemic control does not affect the rate of new-onset AF (30), although metformin seems to be associated with a decreased long-term risk of AF. In T2DM patients, AF ablation may be affected by lower efficacy and a higher incidence of long-term follow-up complications, resulting in a limited impact on long-term survival. (41)

Conclusions

Heart failure is one of the major challenges in the treatment of T2DM patients. A codified screening of this population should result in an earlier specific prescription. Even more importantly, the identification of new personalised strategies should clarify the use of the new antidiabetic drugs in order not only to control the glycemic level, but also to reduce the progression, and possibly to avoid the onset of HF. The recent results observed with the SGTL2i class are particularly encouraging in this viewpoint.