The objective of this study was to determine the electrocardiographic left ventricular hypertrophy (LVH) criterion that best correlated with vascular structure and function parameters in hypertensive patients. A cross-sectional study involving 347 hypertensive patients was performed. The mean age of the subjects was 54.9±11.8 years, and 61% were male. Electrocardiography was used to detect LVH based on the evaluation of 10 criteria, and we defined the voltage-duration product (VDP) complex criterion. The vascular structure was evaluated according to carotid intima-media thickness (C-IMT), and vascular function was evaluated according to pulse wave velocity (PWV), the ambulatory arterial stiffness index (AASI), the home arterial stiffness index, and the peripheral (PAIx) and central (CAIx) augmentation indices. LVH according to at least some electrocardiographic criteria was recorded in 29.10% of the patients (34.10% of females; 25.90% of males). The vascular structure and function parameters showed higher values in the hypertensive patients with LVH. The criterion most closely correlated with C-IMT was Lewis-VDP (r=0.257); with PWV and AASI, the criterion was the Framingham-adjusted Cornell voltage (r=0.228 and r=0.195, respectively); and with CAIx and PAIx, the criterion was Novacode (r=0.226 and r=0.277, respectively). In the multivariate analysis, the association of the vascular structure and function parameters, the VDP complex (multiple linear regression) and the presence of LVH (logistic regression) disappeared after adjusting for age, sex and antihypertensive drugs. The relationship between the electrocardiographic criteria used to detect LVH in hypertensive patients and the vascular structure and function parameters were fundamentally conditioned by age and antihypertensive drug treatment.