Preclinical diabetic cardiomyopathy: relation of left ventricular diastolic dysfunction to cardiac autonomic neuropathy in men with uncomplicated well-controlled type 2 diabetes.

Diabetic cardiomyopathy is an ill-defined entity. This study was designed to explore the possible association between left ventricular diastolic dysfunction (LVDD) and cardiac autonomic neuropathy (CAN) independently from metabolic control. Three groups of 10 age-matched men each with well-controlled type 2 diabetes were studied: (1) subjects with normal diastolic function, (2) subjects with LVDD characterized by impaired LV relaxation, and (3) subjects with a more severe form of LVDD characterized by a pseudonormalized pattern of LV filling. No subject had evidence of clinical diabetic complications, coronary artery disease (CAD), hypertension, congestive heart failure, or thyroid or overt renal disease, and all had a negative maximal exercise test. LVDD was evaluated by Doppler echocardiographic and CAN was evaluated using spectral analysis of heart rate variability (HRV; time and frequency domains) from 24-hour Holter recordings. Findings showed that the high frequency power (HF: 0.15 to 0.4 Hz) tends to decrease with worsening diastolic function; 5.0 +/- 0.2 ms(2) (mean +/- SE) in group 1, 4.2 +/- 0.3 ms(2) in group 2, and 3.9 +/- 0.4 ms(2) (P =.03) in group 3, respectively, whereas the low frequency power (LF: 0.04 to 0.15 Hz) was similar between groups. In the time domain, the mean squared differences of the successive RR intervals (rMSDD) also showed the same pattern, ie, 31.0 +/- 2.8 ms, 23.8 +/- 1.6 ms, and 21.5 +/- 2.9 ms in groups 1, 2, and 3, respectively (P =.03). The E/A ratio correlated significantly with indices of parasympathetic modulation (HF; r = 0.448, P =.013; rMSDD: r = 0.457, P =.011; pNN50: r = 0.425, P =.019). LVDD and CAN are associated in patients with otherwise uncomplicated well-controlled type 2 diabetes. The parameters defining these 2 abnormalities may serve to better define diabetic cardiomyopathy as a distinct entity and could eventually become useful prognostic indicators as it has been shown in nondiabetic populations.