We examined the prevalence of left ventricular structural and functional abnormalities in previously untreated subjects by performing echocardiography in 89 normal volunteers, 57 patients with established hypertension, and 38 patients with mild or borderline hypertension. We measured left ventricular mass, wall thickness, internal diameter, and wall thickness/radius ratio. Because of intergroup differences hi body size, we used covariance analysis to index these variables to a common value of 1.8 m
2. No adjustment was needed for the wall thickness/radius ratio. Functional variables determined were fractional shortening and transmitral early/late flow velocity ratio (the latter was standardized by analysis of covariance to age 40 years). The prevalence of left ventricular mass index values more than 2 SD above the mean of the normal group was 30% in the patients with established hypertension and 12-15% hi the patients with mild hypertension. Corresponding figures for wall thickness index were 65% and 32% and for the wall thickness/radius ratio 60% and 40%. The prevalence of abnormality in the transmitral flow velocity was 28% in the patients with established hypertension and 12% in the patients with mild hypertension. A multivariate discriminant function that used combined anatomic and functional variables provided the most reliable classification; it was correct in 82% of normal subjects, 65% of patients with established hypertension, and 61% of patients with mild hypertension. The majority of patients with hypertension have cardiac structural or functional abnormalities, or both. {Hypertension 1989; 13:151-162) I n human primary hypertension the reported prevalence of left ventricular hypertrophy (LVH) ranges between 20 and 80%. >-7 The majority of investigators suggest that the prevalence is closer to the lower end of this range. However, even the upper end of the range is below the almost universal prevalence of LVH in animals with genetic or experimental hypertension.
-9 Hypertension is usually more severe in the experimental models, which could account for the greater incidence of LVH. In addition, most experiments are performed in inbred strains of animals, where the normal range of variation of left ventricular size may be smaller than in a mixed population 89 where genetic or environmental factors are not as well controlled. Many of the previous human series have included patients previously treated with antihyper-