There is virtually no definitive evidence delineating the mechanism(s) of diabetic cardiomyopathy in patients. This applies to both proximate mechanisms (e.g., defects in excitation-contraction coupling) and ''upstream'' mechanisms (e.g., |
possible alterations in gene expression that result in changes in the proteins responsible for excitation-contraction coupling). Even in experimental animal preparations, it is at present impossible to understand fully all of the potentially causative factors activated as a result of the presence of diabetes. Despite the paucity of hard evidence, there are a number of putative candidates as well as approaches that can be taken to at least ask the right questions about mecha-
& u nism(s) and begin to make correlations in patients. This section will address some of the proposed mechanisms. What follows, however, is an incomplete list.
Microvascular disease related to diabetes has often been considered to be a mechanism underlying diabetic cardiomyopathy. Myocardial abnormalities in diabetic patients without clinically manifest heart disease are correlated with mi-crovasculopathy in other organs. As judged from endomyocardial biopsy studies, there is no question that myocardial microvessels are affected in patients with diabetes even in the absence of confounding features such as hypertension or large-vessel coronary disease. Thus, microvessels (small arteries, arterioles, capillaries) display ''diabetic'' histological abnormalities such as thickening of the medial layer and basement membrane and capillary microaneurysms. As in other organs, myocardial capillaries are ''leaky'' in animal models with diabetes. In vivo coronary flow velocity studies in patients reveal evidence of endothelial dysfunction. What is unclear is how these microvascular abnormalities could actually cause cardiomyopathy. Animal models are of little use in clarifying this issue because they do not manifest microvascular disease comparable to that of human patients.
In the absence of angiographically detectable coronary artery disease and/ or hypertension, there is little to suggest ischemic dysfunction at the microvascu-lar level (e.g., areas of microinfarction with replacement fibrosis or dysfunction consistent with stunned and/or hibernating myocardium). As an alternative to microvascular ischemia, local elaboration by the endothelium of various growth factors and cytokines that influence myocardial signaling pathways and can also impair contractile function is a possible mechanism. High levels of circulating insulin and PAI-1, present in many diabetic patients, could be involved in causing and/or reflecting abnormal signaling.
The heart has usually been considered to be an organ in which the metabolic substrate is primarily fatty acids. However, normal myocardium does take up glucose to a limited extent at rest, and glucose uptake and use of the glycolytic pathway are substantially increased with stress. Moreover, certain key ion pumps, for example, the myocardial sarcoplasmic reticulum calcium ATPase (SERCA-2), may be particularly dependent on glycolytic energy supply. Human myocardium is insulin sensitive. Thus, the metabolic consequences of impaired glucose entry should be considered as a possible mechanism contributing to diabetic car-diomyopathy. In animal preparations, impaired glucose entry with associated increased fatty acid oxidation results in accumulation of toxic intermediates, especially long-chain acyl carnitines, that can result in free-radical-mediated damage to both the sarcolemma and intracellular membranes. This in turn could result in impaired ion pumping and exchange mechanisms. The latter are well described in the myocardium in experimental diabetes and in nonmyocardial tissue in diabetic patients, and include abnormalities of calcium pumping and exchange as well as the sarcolemmal Na-K ATPase. Although there is no information with a
& u regard to free-radical-mediated damage in human myocardium, in view of the results in experimental diabetes this could contribute to the depressed FFR discussed earlier.
Hyperglycemia per se may cause myocardial damage by glycation of myo-cardial matrix proteins with formation of advanced glycosylation end-products (AGEs) and associated free-radical-mediated damage that, in the case of collagen, causes increased cross-linking. This may be the mechanism underlying the common histological finding of increased PAS positive material in myocardium from diabetic patients and is an attractive possibility as an explanation for echocardio-graphic Doppler findings consistent with decreased compliance. However, very limited immunohistopathological studies directed specifically at detection of AGEs in myocardium of autopsied patients with diabetes have not shown extensive deposition. In very small studies examining the histological appearance of myocardium obtained by endomyocardial biopsy in diabetic patients with evidence of cardiomyopathy who do not have coronary artery disease or hypertension, only very modest increases in connective tissue have been reported, and most of it has been perivascular. Increased cross-linking, however, may be at least as important functionally as changes in collagen content or type. When diabetes is combined with hypertension, however, the result seems to be much more pronounced increases in connective tissue. It is also possible for cytosolic proteins, including contractile proteins, to be glycated. Although abnormalities of myofilament calcium sensitivity have been reported in experimental diabetes, it is unknown whether this is due to glycation. This question has not been studied in patients.
Three other putative abnormalities that could cause a cardiomyopathy in patients with diabetes merit consideration. First, a defect in pyruvate kinase activity has been observed in some diabetic patients. If present in human myocardium, it could contribute to abnormal carbohydrate metabolism. Second, mitochondrial damage in diabetic myocardium could limit energy reserves derived from oxida-tive metabolism during stress. However, there is at present no objective evidence to support this. In our studies on the FFR, diastolic tension does not rise at high stimulation frequencies, as would be expected if energy reserves were exceeded because of increased demand. Third, protein kinase C (PKC) has been found to be activated in some forms of experimental diabetes. PKC activation is thought to occur in dilated cardiomyopathy and is known to depress crossbridge cycling and may also have deleterious effects on excitation-contraction coupling. More- |
over, transgenic mice with selective overexpression of a specific PKC isoform £
develop a dilated cardiomyopathy. However, it is unknown whether PKC is activated in human diabetes.
From the above, it is apparent that there are a number of different manifesta- Ja tions as well as potential mechanisms underlying diabetic cardiomyopathy. It is not known which manifestations and mechanisms are actually operative in pa-
& u tients. Indeed, it may well be that they can be multiple in a given patient and perhaps vary with age, gender, ethnicity, and duration and severity of diabetes. As noted above, there have not been obvious differences in the clinical manifestations of diabetic cardiomyopathy with respect to the type of diabetes or insulin use, although such differences may emerge.
Was this article helpful?