Bipyridine derivatives have recently been introduced as a new class of inodilator drugs in the intravenous therapy of heart failure. A member of this class is milrinone, which improves the inotropic state and reduces ventricular afterload, leading to improved hemodynamics. Because systolic and diastolic function are intimately related, it can be expected that the diastolic muscle properties are influenced by changes in systolic function and therefore by milrinone therapy. Since end-diastolic pressure may shift as a result of a change in ventricular volume alone, a complete left ventricular diastolic pressure volume (LVDPV) relationship must always be measured before one can make firm conclusions regarding changes in diastolic function. After a LVDPV relationship is obtained, one should identify the variables that can modify this relationship without directly affecting the intrinsic diastolic muscle properties. These variables can be divided into static effects (coronary vascular bed volume, right ventricular pressure, and pericardial pressure) and dynamic effects (viscoelasticity and myocardial active relaxation). Increments in coronary perfusion pressure of perfusion flow (vascular bed volume) are known to stiffen the cardiac wall (turgor effect). Changes in right ventricular pressure or pericardial pressure are other factors affecting the LVDPV relationship by changing transmural pressure. This effect is more pronounced when the ventricle is already stiff, such as in patients with myocardial hypertrophy. Dynamic effects become important when a LVDPV relationship is measured during isolated cardiac cycles; they include viscoelasticity and abnormal myocardial relaxatlon. Clinical assessment of diastolic cardiac performance assumes a model in which the heart is considered an elastic body (while in fact it is viscoelastic). Therefore any increment in filling rate ( dv dt) will decrease the distensibility and result in a higher end-diastolic pressure. Abnormal relaxation is consistently present in patients with heart failure. It increases early diastolic myocardial stiffness caused by longer muscle activation (passive), and thus the LVDPV relationship will shift upward. In view of these considerations, how does one interpret the effects of milrinone on diastolic performance? The improved contractility increases cardiac metabolism and coronary blood flow. As a result, the cardiac wall stiffens by the erectile effect. An increase in atrial contractility must also be considered. This will increase the filling rate and will stiffen the wall further because of a viscoelastic effect in late diastole. The relaxation rate is affected by changes in afterload (vasodilation) and level of contractility. Both of these effects tend to increase the relaxation rate and distensibility of the left ventricle measured from the LVDPV relationship. This change in relaxation rate is independent of the direct changes in muscle properties. The erectile and viscoelastic effects counteract active relaxation, reflecting a direct milrinone effect. These effects probably explain why in our study there was an increase in distensibility in some patients and a decrease in distensibility in others after milrinone therapy. In view of the problems in controlling the variables in humans affecting the LVDPV relationship, the present results make it difficult to draw firm conclusions regarding the effect of milrinone on cardiac performance.