Diabetic cardiomyopathy can be a major complication of diabetes. This complication is not only present in established diabetes but also in prediabetes. Increased diastolic stiffness develops early in experimental diabetic animals which is associated with interstitial accumulation of connective tissue.1 ,2

The Otsuka Long-Evans Tokushima Fatty (DM) strain of rat was established as an animal model of congenital diabetes, manifesting stable clinical and pathological features that resemble human non-insulin dependent diabetes mellitus (NIDDM).3 In our previous study we found alteration in left ventricular diastolic function and accumulation of myocardial collagen in prediabetic rats.4

Troglitazone, an insulin sensitising thiazolidinedione, improves metabolic deterioration in diabetic animal models, obese subjects, and patients with NIDDM.5 The drug has an acute effect on cultured cardiac myocytes, on isolated rat hearts, and prevents high glucose induced insulin resistance in cultured rat fibroblasts.

Our study aimed to examine the long term effects of troglitazone on myocardial collagen content using histopathological methods in prediabetic rats.

Male DM rats and age matched male non-DM rats were randomly divided into two groups, respectively: (1) treated DM (n = 10); (2) untreated DM (n =10); (3) treated non-DM (n = 10); (4) untreated non- DM (n = 10). Rat chow (Oriental Yeast Co, Tokyo, Japan) with or without 0.2% troglitazone (Sankyo, Japan) was given from the age of 5 weeks up to 15 weeks. Before and after treatment, blood sampling for fasting blood glucose, fasting plasma total cholesterol (cholesterol oxydase-peroxydase method), and fasting plasma triglyceride (glycerokinase-glycerol-3-phosphooxydase method) were obtained after an overnight fast. An oral glucose tolerance test (OGTT), plasma insulin concentration (enzyme linked immunosorbent assay method, Levis insulin kit, Japan), plasma alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were measured at 15 weeks of age.

Histomorphometrical and chemical analysis of collagen were performed as in our previous study4 after treatment in all four groups.

Intergroup comparisons were performed with an unpairedt test that was corrected by using a Bonferoni factor for multiple comparisons whereby significance was defined as p < 0.05/k, where k = number of comparisons. Group data are expressed as mean (SD).

There were no abnormal values for ALT or AST in any of the four groups. There was no significant difference in fasting blood glucose between treated and untreated groups. Troglitazone treatment decreased fasting plasma triglyceride in diabetic rats (untreated 0.64 (0.17) mmol/lv treated 0.26 (0.18) mmol/l, p < 0.001). At 15 weeks of age there were significant differences in two hour blood glucose on OGTT (mmol/l) and plasma insulin concentration (ng/ml) between untreated DM (10.4 (1.5) mmol/l, 8.1 (1.5) ng/ml) and untreated-non DM (6.1 (0.4) mmol/l, 3.8 (2.1) ng/ml, p < 0.001, respectively). Troglitazone treatment improved glucose tolerance and insulin sensitivity of diabetic rats (two hour blood glucose on OGTT 7.7 (1.1) mmol/lv untreated DM, p < 0.005; plasma insulin concentration 3.2 (0.4) ng/ml v untreated DM, p < 0.001).

In both the treated and untreated groups there was no dilation or hypertrophy observed on heart excisions. Troglitazone treat-ment decreased left ventricular collagen content/dry weight in diabetic rats (1.8 (0.1) mg/g vuntreated 2.2 (0.3) mg/g), although no significant difference was observed. No effect on non-DM rats was observed (treated 1.8 (0.4) mg/g v untreated 1.8 (0.5) mg/g). In histomorphometry, troglitazone treatment decreased left ventricular collagen area percentage in diabetic rats (3.3 (0.9)%v untreated 8.1 (1.4)%, p < 0.0001) but had no effect on non-DM rats (treated 4.1 (2)%v untreated 5.1 (1.6)%) (fig 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Histomorphometric analysis at 15 weeks of age in diabetic and non-diabetic rats after treatment with troglitazone. The collagen fibres appear blue in Azan-mallory stain and black in threshold image. In the threshold image, significant differences in left ventricular collagen area percentage in treated DM versus untreated DM, and untreated non-DM versus untreated DM (p < 0.0001) were observed.

Troglitazone improves hyperinsulinaemia in diabetic humans and other animals. Hyperinsulinaemia is associated with lipid abnormalities and is an independent risk factor for ischaemic heart disease that may lead to cardiac dysfunction. Fasting plasma total cholesterol in our untreated DM rats was not significantly different from those in untreated non-DM rats, and no atherosclerotic lesion was histopathologically observed in any group. It is clear that at this stage diabetic rats are free from premature atherosclerosis (as shown by histological and serological findings) and hypertension, although these factors may be relevant in those adult rats with NIDDM and cardiac dysfunction. Administration of troglitazone decreased values of collagen content/dry weight and collagen area percentage of left ventricle in DM rats in the present study. Hepatic dysfunction may influence collagen metabolism in multiple organs, including the heart. However, in the present study, troglitazone treatment decreased interstitial fibrosis and collagen content without hepatic dysfunction. The improvement in cardiac fibrosis in this study is not the result of side effects of troglitazone and might have played a role in improvement of cardiac function.

Gene expression of the transforming growth factor or enhanced gene expression of collagen might participate in the onset of cardiac fibrosis by stimulating extracellular matrix synthesis.4We speculate that the effect of troglitazone in the present study may have been related to improvement in the metabolic state and the regulation of gene expression of collagen in prediabetic rats. To our knowledge, this is the first study to investigate the effect of troglitazone on left ventricular collagen content in a spontaneous NIDDM rat model. Additional work is required for clarification of the mechanism of troglitazone's effect on collagen content.