As a means of assessing ventricular performance, we analyzed the time-varying ratio of instantaneous pressure, P(t), to instantaneous volume, V(t), in the canine left ventricle. Intraventricular volume was measured by plethysmography, while the right heart was totally bypassed. The cardiac nerves were sectioned, and an epinephrine infusion was used to alter the contractile state. The instantaneous pressure-volume ratio was defined aswhere V d is an experimentally determined correction factor. We found that (1) all the E(t) curves thus defined were similar in their basic shape and attained their peak near the end of the ejection phase regardless of the mechanical load, the contractile state, or the heart rate, (2) under a constant heart rate and contractile state extensive changes in preload, afterload, or both did not alter the peak value of E ( t ) , Emax, or the time to Emax from the onset of systole, Tmax, and (3) these parameters of E(t) markedly changed with epinephrine infusion or increases in heart rate. At an epinephrine infusion rate of 2 fig/kg min" 1 , Emax increased to 12.2 ± 4.5 (SD) mm Hg/ml (N = 9) from its control value of 6.6 ± 1.2 mm Hg/ml before the infusion. Simultaneously, Tmax shortened from 191 ± 29 msec to 157 ± 26 msec. Increases in the paced heart rate proportionally shortened Tmax (45% per 100-beats/min change in heart rate) without any effect on Emax. We concluded that E ( t ) , represented by Emax and Tmax, explicitly reflects the ventricular contractility.

We developed a framework of analysis to predict the stroke volume (SV) resulting from the complex mechanical interaction between the ventricle and its arterial system. In this analysis, we characterized both the left ventricle and the arterial system by their end systolic pressure (Ps)-SV relationships and predicted SV from the intersection of the two relationship lines. The final output of the analysis was a formula that gives the SV for a given preload as a function of the ventricular properties (Ees, V0, and ejection time) and the arterial impedance properties (modeled in terms of a 3-element Windkessel). To test the validity of this framework for analyzing the ventriculoarterial interaction, we first determined the ventricular properties under a specific set of control arterial impedance conditions. With the ventricular properties thus obtained, we used the analytical formula to predict SVs under various combinations of noncontrol arterial impedance conditions and four preloads. The predicted SVs were compared with those measured while actually imposing the identical set of arterial impedance conditions and preload in eight isolated canine ventricles. The predicted SV was highly correlated (P less than 0.0001) with the measured one in all ventricles. The average correlation coefficient was 0.985 +/- 0.004 (SE), the slope 1.00 +/- 0.04, and the gamma-axis intercept 1.0 +/- 0.2 ml, indicating the accuracy of the prediction. We conclude that the representations of ventricle and arterial system by their Ps-SV relationships are useful in understanding how these two systems determine SV when they are coupled and interact.

We have previously shown in the normally ejecting canine left ventricle that E ( t ), the time-varying ratio of instantaneous pressure, P ( t ), to instantaneous volume, V ( t ), is little affected by end-diastolic volume or aortic pressure. The present study on an excised, supported canine heart preparation indicates that the thesis on E ( t ) is also valid for either totally isovolumic or auxobaric beats. Intraventricular volume was measured more accurately than it was in the previous study by a new volumetric system. Regression analysis of the data showed that the instantaneous pressure-volume relationship could be approximated by the equation P ( t ) = E ( t ).[ V ( t ) - V d ], where V d is an empirical constant, over a wide range of intraventricular volume. Similar E ( t ) curves were obtained from both isovolumic and auxobaric beats for a given contractile state. When the contractile state of the preparation was enhanced by a constant-rate infusion (0.2 µg/min) of norepinephrine or isoproterenol into the coronary artery, the peak magnitude of E ( t ) increased 63% from 3.6 mm Hg/ml and the time to peak E ( t ) shortened 10% from 175 msec. We conclude that the present investigation substantiates our earlier study which established a link between E ( t ) and the contractile state of the heart.

In a previous analysis of ventricular arterial interaction (Sunagawa et al., 1983), we represented the left ventricle as an elastic chamber which periodically increases its volume elastance to a value equal to the slope of the linear end-systolic pressure-volume relationship. Similarly, the arterial load property was represented by an effective elastance which is the slope of the arterial end-systolic pressure-stroke volume relationship. Since the maximal transfer of potential energy from one elastic chamber to another occurs when they have equal elastance, we hypothesized that the left ventricle would do maximal external work if the ventricular elastance and the effective arterial elastance were equal. We tested this hypothesis in 10 isolated canine left ventricles, ejecting into a simulated arterial impedance, by extensively altering arterial resistance and finding the optimal resistance that maximized left ventricular stroke work under various combinations of end-diastolic volume, contractility, heart rate, and arterial compliance. Each of these parameters was set at one of three levels while others were at control. The optimal resistance varied only slightly with arterial compliance, whereas it varied widely with contractility and heart rate. We thus determined that the ratio of the optimal effective arterial elastance to the given ventricular elastance remained nearly unity. This result supports the hypothesis that the left ventricle does maximal external work to the arterial load when the ventricular and arterial elastances are equalized.