If one turns back the wheel of time to contemplate on the relationship of diabetes and heart failure, the Framingham Study comes to mind, that, early on, identified an increased incidence of heart failure in patients with diabetes [1]. The reasons for this association were not quite clear. The topic therefore deserves more scrutiny, and the work by Halon et al. [2] presented in the previous issue of Cardiology is of great relevance.
The authors studied a community-recruited cohort of type 2 diabetics without any clinical history of cardiovascular disease. At baseline, a 64-slice CT coronary angiography was performed, and the patients were followed for a mean period of 8 years. Only 3% (i.e., 23 of 735 patients) developed heart failure during the observation period. This finding suggests that heart failure is not very common in a low-risk population of type-2 diabetes patients. Furthermore, it was reassuring that the authors succeeded in identifying several clinical risk factors that can predict the development of heart failure. These included age, HbA1c, microvascular disease, elevated systolic blood pressure, and left atrium enlargement (expressed as left atrium/right atrium volume ratio). While these factors had independent prognostic value for the development of heart failure, the existence and extent of coronary heart disease as measured by CT coronary angiography had no influence on the incidence of heart failure 8 years later.
The first assumption one might draw from these observations is that coronary heart disease is apparently not the most common cause of heart failure in patients with diabetes. Can heart failure, then, rather be attributed to diabetic cardiomyopathy, i.e., heart failure without significant coronary heart disease or hypertension? The clinical course of cardiac dysfunction in diabetes often evolves from subclinical cardiac abnormalities such as left ventricular fibrosis and diastolic dysfunction, to severe diastolic heart failure with preserved ejection fraction, and eventually to systolic dysfunction accompanied by heart failure with a reduced ejection fraction [3]. Several mechanisms for the particular development of diabetic cardiomyopathy have been proposed [4]. One of them is “stiffening” of the heart likely due to the occurrence of advanced glycation end products (AGE), including the glycation of collagen in the heart [5,6]. Another hypothesis proposes gliptins as target molecules for protection [7].
In this context, it would have been of considerable interest had Halon et al. [2] measured biomarkers for heart failure, especially high-sensitivity troponins and natriuretic peptides, such as BNP or NT-pro-BNP, as these have been shown to be important prognostic markers for subsequent heart failure [8].
The diagnosis of coronary heart disease in diabetes may be difficult as the classic symptoms are often absent in diabetics, especially in women [9]. Interestingly, another publication by the same study group refers to a previous study in the same patient cohort where maximal treadmill exercise was studied in diabetic patients [10]. It was shown that the maximal exercise level was not an independent predictor of coronary heart disease outcome events. Indeed, the treadmill exercise test has been shown to be unreliable in type 1 diabetics in whom silent coronary artery disease is common [11].
It is obvious from the paper by Halon et al. [2] that the population studied was a mixture of low- and high-risk patients, also correctly pointed out by the authors. Going forward, it is now mandatory to define a high-risk population. In fact, this is straightforward, based on the clinical parameters that the paper identifies for predicting heart failure, namely, age, HbA1c, systolic blood pressure, and microalbuminuria (nowadays termed “moderately increased albuminuria” in revised terminology). But, how does one make sense of the fact that HbA1c is an independent predictor for heart failure? In a recent article from the Diabetes Control and Complications Trial (DCCT), a study on type 1 diabetic patients, the researchers show that the unequivocal effect of high blood glucose that persists for many years on cardiovascular disease is increasingly mediated by traditional risk factors such as hypertension, low-density lipoprotein cholesterol, and triglycerides. The authors conclude that it is important to control blood glucose from the start of the disease and to treat other risk factors aggressively as well [12]. On a side-note, but pertinent to this issue, physicians have to be alert to the cardiovascular safety of antidiabetic drugs [13,14,15,16].
The relationship between micro- and macrovascular disease manifestations in the heart in diabetics is yet another fascinating topic. Heart failure with preserved ejection fraction is often encountered in diabetic patients, shifting its pathophysiologic emphasis from left ventricular afterload to a systemic inflammatory state. This, in turn, conceivably induces oxidative stress in the coronary microvascular endothelium, at least as one of several mechanisms [17].
Are there therapeutic lessons to draw from this study? Patients with type 2 diabetes and microalbuminuria have a 4-times increased risk for heart failure. In the STENO-2 study, researchers randomized type 2 diabetics with microalbuminuria to 2 groups: the first underwent multifactorial intervention (statins, antihypertensive treatment, and intensified lowering of blood glucose), and the other remained on standard care. The results showed that multifactorial interventions led, notably, to a 70% reduction in heart failure in these patients, with microalbuminuria, based on a 20-year follow-up [18].
Taken together, it will be mandatory in the future for all physicians, be it general practitioners, internists, endocrinologists, cardiologists, or nephrologists, involved in the care of diabetic patients to collaborate and share clinical data such as the HbA1c and microalbuminuria status. This will enable the treating physicians to better assess and treat patients with diabetes in a comprehensive manner.
