The aim of this study was to clarify the mechanisms responsible for the increase in early filling rate observed during oral nisoldipine therapy in patients with ischaemic left ventricular (LV) dysfunction. For that purpose, the global and regional LV function was analysed before and after 2 months of double-blind monotherapy with nisoldipine (10 mg twice daily) or a placebo, in 17 patients with a previous anterior myocardial infarction. The baseline LV ejection fraction ranged from 34-51% and no patient had heart failure. Compared to the placebo, nisoldipine significantly lowered LV systolic pressure and end-diastolic pressure (-3 mmHg vs +6 with the placebo; P less than 0.01) and the LV pressure at the time of mitral opening (-2.0 +/- 3.4 mmHg vs +3.5 +/- 3.0; P less than 0.01). Despite this reduction in driving pressure, the global LV early peak filling rate improved with nisoldipine only and this improvement was related to a selective increase in expansion rate of the anterior areas, from 1010 +/- 360 to 1339 +/- 496 mm2.s-1 (P less than 0.001). The time to regional peak filling rate (-8%; P less than 0.01), the asynchrony of diastolic wall motion and the regional ejection fraction (33 +/- 10 to 38 +/- 12%; P less than 0.001) also improved in the anterior areas with nisoldipine but not with the placebo. In contrast, in the inferior, control zones, the regional ejection fraction and filling rate remained unchanged, both when compared to baseline and to the placebo.(ABSTRACT TRUNCATED AT 250 WORDS)