Type 2 Diabetes Mellitus, Interaction Between Left Ventricle and Large Arteries

A direct relationship between brachial pulse pressure (PP) and N-terminal prohormone of brain natriuretic peptide (NT-proBNP) has been demonstrated. NT-proBNP is a marker of cardiac overload and sensitive marker of heart failure. Brachial PP is an indirect marker of arterial stiffness and central pressure, whether arterial markers, mild elevation of BNP, or both directly contribute to excessive rate of events is unknown (question mark). Whether other mechanisms such as endothelial dysfunction or small artery disease contributes is unknown. Type 2 diabetes mellitus (T2D) is a growing pandemic worldwide, affecting up to 10% of the adult population. This pandemic is particularly growing in low-middle-income countries.1 T2D is a strong risk factor for incident cardiovascular (CV) events, in addition to neural degeneration, kidney failure, blindness, and dementia. T2D is closely associated with common CV risk factors such as sedentary lifestyle, obesity, metabolic disorders, and hypertension and is associated with most CV events, but particularly stoke, peripheral artery disease, coronary disease, and heart failure. Both the microvasculature and large vessels are affected in what is now considered as a classical cross-talk—microvascular remodeling and dysfunction affecting large arteries by increasing the mechanical load, both sites sharing functional dysfunction at endothelial level, smooth muscle cells, and extracellular matrix. Furthermore, neurological, ocular, and renal complications of T2D are mainly of vascular origin.2 For all these combined reasons, measuring the consequences of cumulated risk factors on target organ damage may carry a particular value for predicting those at high risk of developing CV events. In this context, Dillinger et al. tested the association between a crude marker of vascular dysfunction, pulse pressure (PP), and incident cases of heart failure, as defined by brain natriuretic peptide (BNP) levels above 50 pg/mL, in a hospital consultation of 3,367 T2D patients, free of prevalent CV disease, followed 2.5 years.3 The main finding was that high PP was independently associated with the risk of elevated BNP, after adjustment of main confounders, with doubling of the risk for PP above the median. This leads to a progression in the C-statistics, and improvement of reclassification that was small but significant. The authors further showed in a subpopulation having central blood pressure measurements and carotid-femoral pulse wave velocity was associated with BNP levels. They conclude that PP could be useful to detect which patients will develop heart failure in future. The article has several merits and strengths, among them addressing an important problem, and providing answers with a reasonably sized sample population. The article also raises some questions as it associates 2 surrogate markers—on one side heart failure and the other arterial stiffness. Let us start with BNP as surrogate for diagnosis of heart failure. Atrial natriuretic peptides were discovered in the 80s by de Bolt4 finding that homogenates of right atrium-induced natriuresis. The later identification of chromaphin granules in the right atrium, and by showing that these granules secreted atrial natriuretic peptides (further identified in the brain as BNP, also secreted by the ventricular cardiomyocytes, and preferred as biomarkers because of slower kinetics of degradation).5 Cardiac natriuretic peptides have vasodilatory, natriuretic properties, dependent on cyclic guanosine monophosphate (cGMP). Cardiac natriuretic peptides are important targets for treatments (BNP analogues, neutral endopeptidase inhibitors). Physiologists rapidly showed that increased hemodynamic load, as represented by preload (volemia), but also afterload led to changes in phenotype of cardiomyocytes, and a higher proportion of them exhibiting secretory phenotypes.5 Afterload for atria is represented by ventricular stiffness, itself related to left ventricular concentric hypertrophy, remodeling and associated changes in extracellular matrix (fibrosis). Afterload of the left and right ventricles are represented by aortic/pulmonary impedance, respectively, which themselves directly depend on vascular resistance and vascular stiffness. BNP is highly sensitive to hemodynamic factors, and BNP has been shown to be highly correlated with left ventricular hypertrophy indices in a seminal paper from the Framingham group.6 BNP is also secreted in response to important physiological cardiovascular stress such as strenuous exercise,7 reaching concentrations well above the threshold retained here. BNP level is highly predictive of reduced ejection fraction (<40%), however with modest incremental value above other risk factors,8 whereas the odd ratios for a more clinical definition of heart failure is very strong.8 This might be explained by the fact that almost half of the cases of heart failure have preserved ejection fraction (HF-PEF), in which, by definition, ejection fraction is not reduced.9 As a matter of fact, patients with T2D exhibit HF-PEF more often than HF-REF 10. So it appears to us that defining heart failure on the sole basis of BNP may be hazardous. On the other hand, the authors used brachial PP as a proxy for aortic stiffness. This is important since PP is easy to obtain and could be implemented in clinical care with minimal effort. PP has long been recognized as an important contributor to CV risk,11 notably heart failure. PP is related to cardiac output and total arterial compliance (TAC) by the following formula TAC = stroke volume/PP, so any decrease in stroke volume without changes in aortic stiffness will result in a decrease in PP. Conversely, if stroke volume is constant, PP is directly proportional to arterial stiffness. This is likely to be the case in heart failure based on the Starling’s Law whereby dilatation helps to compensate for the reduction in cardiac output by maintaining stroke volume (i.e., decreasing ejection fraction). We have shown that arterial stiffness is strongly related with concentric remodeling (i.e., pressure adaptation of the left ventricle),12 which itself is strongly associated with elevated BNP.13 Brachial PP at consultation is not necessarily representative of the true load applied to the left ventricle. It is affected by white coat hypertension and masked hypertension, both frequently observed conditions. It is not representative of 24-hour BP load either because of the high incidence of nondipper14 and obstructive sleep apnea syndrome in T2D patients.15 Brachial PP is not even a direct measurement of the true hemodynamic load to the left ventricle because PP should be measured centrally.16 In addition to the above, most patients with T2D included in the present study were hypertensives, and therefore treated with several antihypertensive drugs, including RAS blockers and calcium channel blockers, both shown to differentially influence central and brachial BP (including PP),17 but also left ventricular diastolic function.18 Although the authors adjusted for antihypertensive treatments, however treatments represent a bias factor rather than an explanatory factor. Similar considerations could be also be applied to antidiabetic drugs, several of them positively affecting large and small arteries.17 The recent identification of SGLT2 inhibitors as potent treatments for heart failure, together with natriuretic and microvascular actions is particularly interesting in this context with many ongoing trials.19 Although mild elevation in BNP cannot be considered as equivalent to heart failure, and PP cannot be considered as a true hemodynamic load the left ventricle is exposed to, the present article does underline the strong relationship that exists between cardiac overload indexes and hemodynamic load. Given that in a small subpopulation of 117 individuals, a relationship between central PP, arterial stiffness, and BNP was observed, this reinforces the notion that arterial stiffness (and maybe microvascular dysfunction) may be a hidden contributor to heart failure. This article underlines the fact that T2D is truly a CV disease and that identification of patients at high risk must include CV markers. Taking PP into account may help to achieve such a goal, and this article pleads for performing more direct assessments of vascular function, both at the level of the large and small arteries, together with heart vessel coupling, to better risk stratify patients and enable personalized preventative strategies to improve CV disease outcomes. The authors declared no conflict of interest.