Multivariate associates of QT interval parameters in diabetic patients with arterial hypertension: importance of left ventricular mass and geometric patterns.

The aim of the study was to assess the determinants of increased QT interval parameters in diabetic patients with arterial hypertension and, in particular, the strength of their relationships to echocardiographically derived left ventricular mass (LVM) and geometric patterns. In a cross-sectional study with 289 hypertensive type 2 diabetic outpatients, maximal QT and QTc (heart rate-corrected) intervals, and QT, QTc, and number-of-leads-adjusted QT interval dispersions were manually measured from standard baseline 12-lead ECGs. Electrocardiographic criteria for left ventricular hypertrophy (LVH) were either Sokolow-Lyon or Cornell sex-specific voltages. LVM and geometric patterns were determined by 2D echocardiography. Statistical analyses involved bivariate tests (Mann-Whitney, chi2, Spearman's correlation coefficients, ANOVA and receiver-operating-characteristic (ROC) curve analyses) and multivariate tests (multiple linear and logistic regressions). QT dispersion measurements showed significant correlations with echocardiographic LVM (r=0.26-0.27). ROC curves demonstrated a poor isolated predictive performance of all QT parameters for detection of LVH (areas under curve: 0.58-0.59), comparable to that of electrocardiographic voltage criteria. Only patients with concentric hypertrophy had significantly increased QT dispersion (QTd) when compared to those with normal geometries (64.24+/-21.09 vs 53.20+/-15.35, P<0.05). In multivariate analyses, both electrocardiographic and echocardiographic LVH were independent predictors of increased QTd, as well as only QTd and gender were determinants of LVM. In conclusion, increased QT interval dispersion is associated with LVM and concentric hypertrophy geometric pattern in diabetic hypertensive patients, although in isolation neither QTd nor any QT parameter presents enough predictive performance to be recommended as screening procedures for detection of LVH.