Sensitivity and specificity of criteria of left ventricular hypertrophy for the prediction of cardiovascular events in the Bordeaux cohort of hypertensive patients

Left ventricular hypertrophy (LVH) is associated with an increased risk of cardiovascular complications independently of other known risk factors. Sin ce 1983, we have followed up a cohort of initially untreated hypertensive p atients with echocardiographic measurements of left ventricular mass (LVM). We analyzed the data on 474 patients with more than five years follow-up t o assess the prognostic value of LVM and the sensitivity and specificity of the different ways to index for LVM through ROC curves. Forty patients wer e lost to follow-up. The mean follow-up period was 89 +/- 31 months. A card iovascular complication was recorded in 40 individuals. There was a strong link between increased LVM and the occurrence of complications. Indexation by body surface area or height to the power 2.7 give the [GRAPHIC] greatest area under the curve (AUC) to discriminate between patient with or without cardiovascular events. To get the same sensitivity in females and in males for the prediction of future events, lower cut-off must be used in females.