Background: Echocardiogram has provided an excellent technique to differentiate the spectrum of geometric remodelling as well as to diagnose left ventricular hypertrophy (LVH). The purpose of our study was to determine the relationship of geometric left ventricular remodelling patterns to electrocardiographic (ECG) findings in hypertensive patients. Methods: We arbitrarily divided patients into 4 groups according to the LV mass index (LVMI) and relative wall thickness, group 1 (normal): normal LVMI and normal relative wall thickness, group 2 (concentric remodeling): increased relative wall thickness and normal LVMI, group 3 (concentric hypertrophy): increased relative wall thickness and increased LVMI, group 4 (eccentric hypertrophy): normal relative wall thickness and increased LVMI. We studied total 100 patients (mean age 60 짹 11, female 48) with 25 patients in each group. Patients were categorized according to the presence or absence of ECG LVH by 3 major ECG criteria. Echocardiographic LVH was defined by LV mass index (LVMI) > 104 g/m2 in women and 116 g/m2 in men. Results: The prevalence of ECG LVH by Sokolow-Lyon voltage criteria, Cornell voltage criteria, and Romhilt-Estes point score were 12%, 0% and 0% in group 1, 20%, 12%, and 12% in group 2, 52%, 44%, 44% in group 3, and 60%, 56%, and 44% in group 4, respectively. The QRS voltages by Sokolow-Lyon criteria were 27.0 짹 6.9, 30.0 짹 12.8, 33.7 짹 11.7, and 40.6 짹 15.7 mV in group 1, 2, 3, and 4, respectively (p=0.001). The QRS voltages by Cornell criteria were 14.5 짹 4.8, 17.4 짹 5.6, 21.5 짹 8.6, and 23.9 짹 9.9 mV in group 1, 2, 3, and 4, respectively, p<0.001. The Romhilt-Estes scores were 0.6 짹 1.2, 1.2 짹 2.4, 3.8 짹 2.7, and 4.6 짹 3.0 in group 1, 2, 3, and 4, respectively, p<0.001. Conclusion: Our study indicates that the ECG findings of LVH were significantly different in 4 groups categorized by geometric remodeling in hypertensive patients.