Recently, it was shown that proportional relationships exist between systolic, diastolic and mean pulmonary artery pressure (P(sys), P(dia) and P(mean)) and that they are maintained under various conditions in both health and disease. An arterial-ventricular interaction model was used to study the contribution of model parameters to the ratios P(sys)/P(mean), and P(dia)/P(mean). The heart was modeled by a time-varying elastance function, and the arterial system by a three-element windkessel model consisting of peripheral resistance, R(p), arterial compliance C(a), and pulmonary artery characteristic impedance Z(0). Baseline model parameters were estimated in control subjects and compared to values estimated in patients with pulmonary hypertension. Results indicate that experimentally derived ratios P(sys)/P(mean) and P(dia)/P(mean) could be accurately reproduced using our model (1.59 and 0.61 vs. 1.55 and 0.64, respectively). Sensitivity analysis showed that the (empirical) constancy of P(sys)/P(mean) and P(dia)/P(mean) was primarily based on the inverse hyperbolic relation between total vascular resistance (R(T); calculated as R(p) + Z(0)) and C(a), (i.e. constant R(T)C(a) product). Of the cardiac parameters, only heart rate affected the pressure ratios, but the contribution was small. Therefore, we conclude that proportional relations between systolic, diastolic and mean pulmonary artery pressure result from the constancy of R(T)C(a) thus from pulmonary arterial properties, with only little influence of heart rate.