Abstract--QT interval parameters have been associated with left ventricular hypertrophy (LVH) in hypertensive patients. The aim of this study is to assess this relationship in resistant hypertension and, in particular, to evaluate whether any QT interval parameter could provide additive information for LVH beyond that obtained from the best electrocardiographic voltage criterion. In a cross-sectional study, 471 resistant hypertensives were submitted to standard 12-lead ECGs, 24-hour ambulatory blood pressure monitoring, and 2D echocardiographic examinations. QT interval durations and QRS voltages were measured, and maximum rate-corrected QT interval duration (QTc_max) and dispersion (QTd), and Sokolow’s and Cornell’s voltage product were calculated. Statistical analyses involved bivariate tests and multivariate logistic regression, with LVH as the dependent variable. A total of 383 patients (81%) had echocardiographic LVH. In bivariate comparisons, both QT interval parameters showed a predictive performance for LVH similar to Cornell’s product, the best ECG voltage criterion. In multivariate analysis, QT parameters and Cornell’s product were independently associated with LVH, after adjustment for other LVH determinants. QTc interval >440 ms^1/2 and dispersion >60 ms were associated with a 2-fold (95% confidence interval [CI], 1.1 to 3.8) greater chance of having LVH, whereas Cornell’s product >240 mV·ms implied a 2.6-fold (95% CI, 1.2 to 6.1) increased chance of LVH. The combination of prolonged QT interval and increased Cornell’s product was associated with a 5.3- to 9.3-fold higher chance of having LVH. Hence, although in isolation, no QT interval parameter performs better for LVH detection than simpler Cornell’s product, it provides additive information and can be used in combination with voltage criteria to refine LVH risk stratification in resistant hypertension.