We evaluated the usefulness of a combination of QRS voltage and the pattern of ST-T abnormality in the electrocardiographic diagnosis of left ventricular hypertrophy (LVH) in 100 middle-aged men: 32 normals, 59 with hypertension (HT), and 9 with hypertrophic cardiomyopathy (HCM) with evidence of LVH (RV5 or RV6 > or = 2.6 mV, SV1 + RV5 or SV1 + RV6 > or = 3.5 mV). They The subjects were classified into three groups based on ST-T pattern: normal (group N), early strain (group ES), and strain (group S). Echocardiographic evidence of LVH was present in 52.0% (52/100) of the subjects: 72.8% (43/59) of the patients with HT, all 9 patients with HCM, and none of the 32 normals. Echocardiographic evidence of LVH was present in 31.3% (20/64) of group N, 73.3% (11/15) of group ES, and all 21 subjects of group S. In patients with HT, the incidence of echocardiographic LVH was higher in group S (100%) than in both group ES (78.6%) and group N (60.6%). QRS voltage (RV5, RV6, RV5 + SV1, and RV6 + SV1) was significantly correlated with interventricular spetal thickness (IVST), IVST+LVPWT/2, and LV mass, as determined by echocardiography, in patients with LVH (IVST or left ventricular posterior wall thickness (LVPWT) of > or = 12 mm) (r = 0.55 to r = 0.75, p < 0.05), but not in patients without LVH (IVST and LVPWT < 12 mm). There were significant correlations between QRS voltage indices (RV5, RV6, RV5 + SV1, and RV6 + SV1) and IVST, IVST+LVPWT/2, and LV mass in group S (r = 0.68 to r = 0.86, p < 0.05), but not in group N. Values for IVST and LV mass were significantly greater in group S than in groups ES or N. The combination of QRS criteria and ST-T findings reflected the echocardiographic assessment of LVH, especially in patients with HT.(ABSTRACT TRUNCATED AT 250 WORDS)