Mechanical analysis of ventricular interaction

A unified theoretical model of the right and left ventricles and the surrounding pericardium is proposed to examine the mechanisms underlying direct ventricular interaction. The biventricular model consists of three elastic cylindrical membrane segments, representing the left ventricular and right ventricular free walls and the septum, joined at the interventricular sulci. This configuration highlights forces at the sulci, and is useful in analyzing the mechanical interplay of forces at these junctions. The model shows that both diastolic and systolic interaction are sensitive to the material properties of the heart wall as well as to cardiac dimensions. More specifically, diastolic interaction, similar to that measured experimentally, appears to be possible only if the material nonlinearities of the three walls are different. Septal shift, which is related to the overall shape of the biventricular cross-section, is shown to depend on the balance of wall forces at the sulci as well as the transseptal pressure gradient. Further, the model shows that, at end-systole, a significant right ventricular pressure is generated even when the right free wall is non functional. Right ventricular pressure can approach normal levels if left ventricular pressure is abnormally elevated, or if right ventricular stiffness is greatly increased as with chronic myocardial infarction. This supports speculations that right ventricular pumping is normally assisted by the left ventricle. Three models of the pericardium are examined to investigate how the pericardium augments interaction. The most realistic is one where there is sufficient lubrication in the pericardial space to allow equalization of tension in the pericardium at end-diastole. Model results show that pericardial contact pressure is not uniform over the left and right ventricles. Furthermore, contact pressure is lowest--perhaps even negative--in the area overlying the interventricular grooves. Standard indices for measuring the magnitude of interaction are desirable. It is shown here that the magnitude of interaction is dependent on the filling state (a particular V$\sb{r}$, V$\sb{l}$) of the ventricles, and that at any filling state twenty-four indices are necessary for a complete description. However, only three independent indices need be determined; these can then be used to compute the remaining twenty-one