Diabetes mellitus (DM) has been increasing in both developed and developing communities as a result of epidemic obesity and increasingly sedentary lifestyles.1) The relationship between DM and cardiovascular diseases2) and the fact that cardiovascular diseases are the major reason for morbidity and mortality in diabetics have been known for a long period of time.3) Heart failure is a common and serious complication of diabetes. The incidence of cardiomyopathy-induced congestive heart failure (CHF) has been found to be increased in diabetic patients even in the absence of coronary atherosclerosis.4-6) Diabetic cardiomyopathy can be described as abnormalities in myocardial function or structure in the absence of hypertension (HT), alongside coronary artery disease (CAD) and serious valvular heart disease. The development of diabetic cardiomyopathy has been considered to be a multifactorial condition. Increased stiffness due to the accumulation of connective tissue and insoluble collagen on left ventricle (LV) wall,7) autonomic dysfunction,8) endothelial dysfunction, impairment in various ligand sensitivity9) and anomalies in the proteins which arrange ion flow, particularly intracellular calcium, are some of the factors that play a role in development of disease.10)11) In early stages of diabetic cardiomyopathy, the systolic and diastolic functions of LV were found to be preserved on conventional echocardiography (CE). Diastolic dysfunction can be sensitively detected at the early stage by tissue Doppler echocardiography (TDE).12)13) Diastolic dysfunction, systolic dysfunction and myocardial structural changes are three main clinicopathologic features of diabetic cardiomyopathy.

The basis of noninvasive evaluations of diastolic dysfunction is constituted by Doppler studies of transmitral flow, mitral flow velocities, deceleration times, isovolumic relaxation time (IVRT) and flow patterns. As the degree of diastolic dysfunction deteriorates, the "early diastolic LV filling (E wave)" decreases and a delayed relaxation pattern appears. However, by increasing left atrial pressure, the E wave can be normalized and turns into a mitral flow wave, in a so-called 'pseudonormal pattern' which cannot be distinguished from that of a normal wave. The advantages of this technique are that it achieves perfect temporal resolution and mono-directional correlation with progressive cardiac anomalies. In one study, the mitral early peak diastolic velocities (E) and late peak diastolic velocities (A) of 27 type 1 and 25 type 2 diabetic patients without HT and CAD were examined and the E/A ratio was found to be more markedly decreased in type 2 diabetics.14) In another study, it was reported that ventricular filling, particularly early peak filling velocity, was more significantly impaired in type 2 diabetics compared to type 1 diabetics.15) Several studies have demonstrated early impaired diastolic function, despite systolic function parameters remaining normal, in diabetics. This finding may be related to using more sensitive techniques in the detection of diastolic dysfunction and the absence of sufficiently adequate techniques to evaluate systolic dysfunction.

Recently, a diabetes-specific cardiomyopathy has been demonstrated in diabetic patients; however the pathophysiology and diagnostic criteria of this condition have not yet been elucidated. This condition has been described in diabetics without CAD, HT or valvular heart disease, and in the early stage of the disease systolic functions are preserved and diastolic dysfunction develops. Although diabetic cardiomyopathy develops in patients with overt DM, the stage of this condition has not yet been clarified in the prediabetic period. The aim of the present study was to evaluate whether tissue Doppler imaging can detect a pre-clinical impairment of diastolic function in prediabetic patients with preserved systolic functions in CE, so as to detect possible early cardiac dysfunction.

The main features of the subjects studied are summarized in Table 1. Baseline demographic and clinical characteristic were comparable across the four groups. The CE and TDE parameters of all groups are demonstrated in Table 2 and 3, respectively. The four groups were similar in terms of CE parameters. In TDE evaluation. The early peak diastolic velocity of medial mitral annulus (Ea) of impaired fasting glucose (IFG) patients were similar to controls (11.65±0.66 vs. 9.72±1.58, p=0.084). However, statistically significant difference between controls and impaired glucose tolerance (IGT) patients (11.65±0.66 vs. 9.06±1.07, p≤0.001) and between controls and CGT patients (11.65±0.66 vs. 9.74±1.09, p=0.008) were found. However, there was no difference between groups in terms of measurements from the lateral mitral annulus (SlatA, ElatA, AlatA) and the lateral myocardium (SlatM, ElatM, AlatM).

The present study has two significant findings. Firstly, it was shown that the process of cardiomyopathy initiates in IGT and CGT conditions before overt diabetes develops. In measurements by TDE, we detected that the systolic functions of prediabetics were preserved whereas diastolic functions were impaired. This condition was found to be prominent in IGT and CGT patients. In a study performed with TDE, García-Fernández et al.18) demonstrated that LV diastolic dysfunction might be regional.18) Although in line with this view we did detect diastolic dysfunction findings in the medial of mitral annulus, diastolic functions were within normal ranges in the lateral annulus and lateral myocardium in the present study. Secondly, we can propose that insulin resistance plays a pivotal role in the effects of diabetes on the myocardium. There is evidence that the cardiac muscle in type 2 diabetics is resistant to insulin, although underlying CAD was not excluded.19) In an animal study by Mizushige et al.20) it was found that diastolic function in prediabetic rats might be associated with insulin resistance.

The prevalence of CHF and diastolic dysfunction has been increasing in diabetic patients.21) LV diastolic dysfunction may present as the first stage of CHF and is associated with high mortality/morbidity.22) In various studies it has been demonstrated that diabetic cardiomyopathy contributed to the incidence of increased heart failure in diabetics. Almost all these studies were conducted on overt diabetics with normal systolic/diastolic function. However, this condition has not been investigated in prediabetics such as those with IFG and IGT. Importantly, CE and normal cardiac functions have not been examined by TDE. Recently, noninvasive techniques that evaluate diastolic function have improved. However, we do not have enough data about early stage of diabetic cardiomyopathy and the pathophysiology of this disease remains unclear.

If physiologic adaptation to metabolic changes and degenerative changes, which comprise two main aspects of diabetic cardiomyopathy, are treated in the early stages of diabetes, clinicians may be able to intervene in the progress of the disease. Studies on antihyperglysemic treatment advocate the view that myocardial function and structural differences are correlated with glysemic control. Pogátsa et al.23) investigated the effects of antihyperglysemic treatment on diabetic dogs with serious hyperglicemia and it was found that LV passive elastic module (the marker of stiffness) and LV end-diastolic pressure were higher and cardiac output was lower in non-treated dogs. Similarly, in another study it was indicated that diabetic cardiomyopathy-induced changes in diabetic rats may be reversible with insulin treatment.24) Contrary to these findings, in a study conducted on diabetic dogs Regan et al.25) suggested that diabetic cardiomyopathy was not reversible with antihyperglycemic treatment.

The current study has some potential limitations that should be noted. First, the study sample size was small. Second, we did not measure the insulin sensitivity of subjects since insulin resistance and hyperinsulinemia are already present in patients with IGT, IFG and CGI. Third, one of the important supplementary methods for assessment of diastolic function is pulmonary venous flow velocity evaluation, which was not reported in this study. Although this study has several limitations, the results nevertheless provide important information about cardiac functions in prediabetic patients.

In conclusion, there as yet exists no effective treatment for diabetic cardiomyopathy, which is one of the leading causes of morbidity and mortality in patients with overt diabetes. Therefore it is important to prevent the occurrence or progression of the disease. Studies have clearly demonstrated that prediabetes is a reversible condition. In the present study, we obtained data indicating that the effect of prediabetes may be reversible. This finding needs be supported by standardized methods and studies conducted on a wide number of patient populations.