The time course and the magnitude of the effect of glyceryl trinitrate (GTN) on central venous (pulmonary artery diastolic pressure-PAPd) and peripheral arterial (a/b-ratio of the finger pulse wave) haemodynamics were compared in a randomized double-blind cross-over study in 12 patients suffering from congestive heart failure (NYHA II-III) with elevated PAPd at rest (> or = 15 mmHg). The data were obtained in a bioavailability study of two sprays of glyceryl trinitrate, which differed in their galenical characteristics and in the dose of GTN (0.4 mg vs. 0.8 mg). Following sublingual administration of each spray, PAPd, a/b-ratio and the plasma concentrations of GTN and its metabolites were measured up to 30 min. The relative bioavailability of GTN of the test preparation was estimated to be 157%, 161% and 147%, when calculated from the plasma concentration-time data or the integrated effect of GTN on a/b-ratio or PAPd, respectively. The mean time courses of the decrease in PAPd and the increase in the a/b-ratio of the finger pulse curve were mirror images. Thus, there was a strong correlation between the mean values of PAPd and a/b-ratio following the administration of glyceryl trinitrate. Since the slope of the relationship differed considerably between the patients, the magnitude of effect of GTN on PAPd in the individual patient could not be predicted from the changes in a/b-ratio.