Mitral flow velocity pattern in patients with left ventricular (LV) diastolic dysfunction usually includes decreased peak early diastolic filling velocity (E), slowed deceleration of the early diastolic filling wave and increased peak filling velocity at atrial contraction (A). However, the abnormal mitral flow velocity pattern can be normalized in the presence of concomitant congestive heart failure. In such cases E can be equal to or even higher than normal, its deceleration is normal or faster than normal value, and A can be normal or lower than normal value. Clinical observations in patients with severe heart failure showed that the mitral flow velocity pattern changes with vasodilating therapy, reflecting the changes in the left atrial (LA) to LV pressure difference rather than those in the absolute LA pressure or LV pressure alone. This was validated in the canine study in which levels of LV dysfunction were made by the injection of microspheres into the left coronary artery to study the interrelation among the mitral flow velocity pattern and LA and LV pressures. In this experiment, the changes in the mitral flow velocity pattern could not be explained by the changes in LA or LV pressure alone but was better explained by the changes in the LA to LV pressure difference. Not only LA-LV crossover pressure but also LA compliance seem to be important as determinants of LA pressure level in diastole. In addition to LV relaxation rate, incompleteness of relaxation, elastic recoil and LV passive elastic properties, extracardiac constraint is also considered to be an important determinant of the level of the LV diastolic pressure and hence of the mitral flow velocity pattern at least in the presence of congestive heart failure. Thus, mitral flow velocity pattern is determined by the interaction of LA and LV pressures, both of which are affected by chamber properties as well as loading conditions.