The influence of diabetes on novel electrocardiographic indexes of arrhythmic risk in patients with stable coronary artery disease Panagiotis Korantzopoulos a, ⁎, Konstantinos P. Letsas b , Zacharias Christogiannis a , Kallirroi Kalantzi a , Haralampos J. Milionis c , Ilias Massis a , Christos Pappas a , John A. Goudevenos a a Department of Cardiology, University of Ioannina School of Medicine, 45110 Ioannina, Greece b Second Department of Cardiology, Evangelismos General Hospital of Athens, Athens, Greece c Department of Internal Medicine, University of Ioannina School of Medicine, Ioannina, Greece article info Article history: Received 5 September 2010 Accepted 23 October 2010 Available online 20 November 2010 Keywords: Coronary artery disease Diabetes Repolarization T peak-to-end Diabetes is a worldwide epidemic and is considered a coronary artery disease (CAD) equivalent [1]. Moreover, epidemiological evidence suggests that diabetes is a strong predictor of sudden cardiac death (SCD) [2,3]. On the other hand, CAD is the condition most commonly associated with SCD [2] whereas diabetes is a significant predictor of SCD in CAD [4]. A well-known pathogenetic factor for malignant ventricular arrhythmias is the increased dispersion of repolarization which reflects the heterogeneity rather than the total duration of repolarization [5]. The T peak-to-end (Tpe) interval and the Tpe/QT ratio represent novel electrocardiographic indexes of arrhythmic risk that possibly corre- spond to the spatial dispersion of ventricular repolarization [5–7]. In this pilot observational study, we sought to investigate the impact of diabetes on the aforementioned electrocardiographic (ECG) indexes in the setting of stable CAD. Consecutive patients with stable CAD who were seen in the outpatient clinic were screened. Exclusion criteria were recent acute coronary syndrome within the past 6 months, recent percutaneous coronary intervention or cardiac surgery, any physical disability, congestive heart failure with NYHA class > II, presence of atrial fibrillation, presence of orthostatic hypotension, presence of non- sustained ventricular tachycardia on Holter monitoring, presence of bundle brunch block, QRS duration > 120 ms, presence of second or third degree AV block, previous implantation of a pacemaker or a defibrillator, administration of antiarrhythmic drugs, administration of drugs that prolong the QT interval, thyroid dysfunction, renal failure, and electrolyte disturbances. All participants were able to perform the normal daily activities and their functional capacity was satisfactory. A total of 94 patients were finally included in the analysis (mean age: 63 ± 8.4 years, 85 men). Demographic, clinical as well as electrocardiographic indexes of repolarization were carefully recorded. Specifically, The QT and the QTpeak intervals were measured manually on ECG recordings at a paper speed of 50 mm/s. The Tpe interval was calculated as QT–QTpeak. The QT interval was measured in as many of the 12 leads as possible while Tpe interval was assessed ⁎ Corresponding author. Tel.: + 30 26510 99347; fax: + 30 26510 07017. E-mail addresses: p.korantzopoulos@yahoo.gr, pan-kor@mailbox.gr (P. Korantzopoulos). Table 1 Demographic and clinical characteristics of the patients. No diabetes (n = 79) Diabetes (n = 15) P value Age (years) 62.2 ± 8.1 65 ± 7.2 0.35 Sex (male) 71 (89.9%) 14 (93.3%) 0.67 CHF 10 (12.7%) 3 (20%) 0.45 LVEF (%) 43.8 ± 7.3 41.5 ± 8.3 0.60 Hypertension 28 (35.4%) 7 (46.7%) 0.56 Previous MI 44 (55.7%) 8 (53.3%) 0.86 Drug therapy B-blockers 63 (79.7%) 13 (86.7%) 0.73 Diltiazem 6 (7.6%) 1 (6%) 0.69 ACEI/ARBs 43 (54.4%) 10 (66.7%) 0.41 Statins 73 (92.4%) 13 (86.7%) 0.37 Aspirin 68 (86.1%) 13 (86.7%) 0.95 Clopidogrel 50 (63.3%) 11 (73.3%) 0.56 Table 2 Electrocardiographic parameters. Values presented as median [25th–75th percentile]. No diabetes (n = 79) Diabetes (n = 15) P value HR (bpm) 66 [58–75] 73 [56–87] 0.45 QTc (ms) 390 [371–421] 385 [357–396] 0.22 Tpe (ms) 80 [64–86] 84 [62–88] 0.09 Tpe/QT 0.20 [0.16–0.22] 0.24 [0.18–0.27] 0.03 in the precordial leads [6]. The Tpe and the Tpe/QT ratio were calculated using the corresponding values from each lead. The measurements were obtained in three consecutive complexes of each lead, and the resulting average value was finally accepted. In order to avoid diurnal variations, we obtained the ECG recordings during the same time interval (from 9 a.m. to 11 a.m.). QT interval corrected for heart rate (QTc) was calculated using the Bazett's formula (QTc=QT/RR -2 ) [8]. The Tpe and QTc reported values were the maximum obtained values. All measurements were performed by one experienced investigator unaware of the clinical characteristics of the study participants. To identify intraobserver variability, the ECG tracings of 10 randomly selected patients were reexamined 10 days after the initial evaluation. Intraobserver variation was less than 5%. Continuous variables are expressed as mean ± SD, or as median [25th–75th percentile]. Categorical variables are presented as abso- lute numbers and frequencies. Comparisons of continuous variables performed using the unpaired t-test or the non-parametric Mann– Whitney U test, as appropriate. Comparisons between categorical variables were performed by the chi-square test. A two-tailed p value <0.05 was considered significant. All analyses were performed using the SPSS software (version 13.0; SPSS Inc., Chicago, Illinois). The demographic and clinical characteristics of non-diabetic (n = 79) and diabetic patients (n = 15) are presented in Table 1 and were comparable between the 2 groups. The ECG indexes are presented in Table 2. The QTc interval was similar between the 2 groups. However, diabetics had longer Tpe interval (statistical trend; p = 0.09) and significantly increased Tpe/QT ratio (p = 0.03). 267 Letters to the Editor