Regression of left ventricular hypertrophy (LVH) in the setting of a well-planned intervention study has been associated with longer survival in hemodialysis patients. Whether changes in left ventricular mass (LVM) in clinical practice predict survival and cardiovascular events in these patients is still unknown. In a prospective study in 161 hemodialysis patients we tested the prognostic value of changes in LVM on survival and incident cardiovascular events. Echocardiography was performed twice, 18 +/- 2 SD months apart. Changes in LVM occurring between the first and the second echocardiographic study were then used to predict mortality and cardiovascular events during the ensuing 29 +/- 13 months. The prognostic value of LVM changes was tested in a multivariate Cox's model with LVM index (LVMI) [expressed as LVM/height(2.71)], included as a covariate to control for regression to the mean. The rate of increase of LVMI was significantly (P= 0.029) higher in patients with incident cardiovascular events than in those without such events. Accordingly, cardiovascular event-free survival in patients with changes in LVMI below the 25th percentile was significantly (P= 0.004) higher than in those with changes above the 75th percentile. In a multiple Cox regression analysis, including age, diabetes, smoking, homocysteine, 1 g/m(2.7)/month increase in LVMI was associated with a 62% increase in the incident risk of fatal and nonfatal cardiovascular events [hazard ratio 1.62 (95% CI 1.13-2.33), P= 0.009]. Changes in LVMI have an independent prognostic value for cardiovascular events and provide scientific support to the use of repeated echocardiographic studies for monitoring cardiovascular risk in dialysis patients.