Portal de Periódicos da CAPES

RESPONSE: LETTER TO THE EDITOR

2008; Wiley; Volume: 31; Issue: 2 Linguagem: Inglês

10.1111/j.1540-8159.2007.00983.x

1540-8159

Giovanni Targher, Alessandro Costa, Lorenzo Bertolini, Débora Mariano Cian, Giulio Molon,

Cardiac Imaging and Diagnostics

To the Editor: We appreciate the comments by Dr. Madias regarding our recently published article assessing the prevalence of abnormal microvolt T-wave alternans (TWA) in type 2 diabetic patients and its relationships with the degree of glycemic control.1 We obviously agree with the observations by Dr. Madias that the results of our cross-sectional study suggesting a possible cause-effect relationship between poor glycemic control and abnormal microvolt TWA will be further strengthened when the follow-up data on the whole cohort of diabetic patients will be available. By study design, our type 2 diabetic patients were clinically free of diagnosed ischemic heart disease and systolic dysfunction, and were selected for matching for age, gender, and blood pressure values to control subjects (who also included some subjects with raised blood pressure values). Very recently, we have completed the collection of echocardiographic data in the whole cohort of type 2 diabetic patients (by an experienced cardiologist [D.C.], who was blind to subjects' details), whereas echocardiographic examinations in the control cohort are still ongoing. Importantly, interventricular septal wall thickness (11.8 ± 0.7 vs 11.6 ± 1.0 mm; P = 0.28) and the early to atrial diastolic filling (E/A) ratio of the transmitral pulsed Doppler signal (0.71 ± 0.12 vs 0.74 ± 0.12; P = 0.23)—the most widely used parameters for estimating left ventricular hypertrophy (LVH) and diastolic dysfunction—were essentially similar between diabetic patients with normal and abnormal TWA. Almost identical results were found between those with lower and those with higher hemoglobin A1c (HbA1c) values. Moreover, all of these patients had normal left ventricular ejection fraction (LVEF >60%), and did not have any signs of prior myocardial infarction at echocardiographic evaluations. Thus, we believe that these new echocardiographic data suggest that any specific impact of LVH and left ventricular diastolic dysfunction on the value of microvolt TWA in type 2 diabetic patients is likely to be minimal. According to the Dr. Madias's suggestions, we have also calculated the actual microvolt TWA values in all study participants and not only in type 2 diabetic patients with microvolt TWA positivity, as previously reported in our article.1 The microvolt T-wave amplitude tended to be higher in type 2 diabetic patients than in nondiabetic controls (2.6 ± 3.1 vs 1.7 ± 1.5 μV; P = 0.18). In addition, we found a significant positive association between HbA1c and numerical TWA values in the diabetic cohort (r-value = 0.40; P = 0.0034). Finally, HbA1c remained an independent positive predictor of the magnitude of microvolt TWA (= 0.0041) even when this latter variable was included as a continuous measure in a multivariate linear regression analysis. As regards to the possible use of receiver-operator-curves (ROC) statistics, we found that the widely used cut-point of >1.9 μV for classifying TWA positivity discriminates quite effectively type 2 diabetic patients with higher HbA1c (i.e., 7.5% or more) from those with lower HbA1c (sensitivity 0.51, specificity 0.89).