Early left ventricular filling: an approach to its multifactorial nature using a combined hemodynamic-Doppler technique.

It has recently been shown that early left ventricular filling is a multifactorially determined phenomenon, the characteristics of which are highly dependent on relative changes in any of its determinants (left ventricular end-systolic volume, the constant of isovolumic left ventricular pressure decay, left atrial pressure at the onset of mitral valve flow, and left ventricular and left atrial compliance). Thus changes in the pattern of filling do not necessarily reflect changes in diastolic properties; they might instead simply reflect changes in loading conditions. To define a clinically implemented approach where the contribution of each of the covariates of early filling to the filling process and their modification by load manipulation could be assessed, nine patients with ischemic heart disease underwent simultaneous assessment of micromanometer left ventricular pressure and two-dimensional echo-guided Doppler mitral flow velocity before and after administration of nitroglycerin (0.2 mg intravenously). Nitroglycerin induced a significant reduction in the early-filling E wave (from 41 +/- 5 cm/sec to 32 +/- 7 cm/sec; p less than 0.002), whereas the late-filling A wave did not change (from 51 +/- 12 cm/sec to 55 +/- 9 cm/sec; p = 0.15), so that the E/A ratio decreased 27 +/- 16% (p = 0.016). End-systolic volume, the constant of isovolumic left ventricular pressure decay, and left atrioventricular pressure crossover at the onset of mitral flow decreased (from 49 +/- 37 to 43 +/- 38 ml [p = 0.016], from 52 +/- 14 to 47 +/- 13 msec [p = 0.016], and from 19 +/- 10 to 12 +/- 7 mm Hg [p = 0.08], respectively), whereas left atrial compliance (defined as stroke volume/atrioventricular pressure crossover) and left ventricular compliance (computed as change in volume/change in pressure at early and late diastole) did not change (p = 0.15 and p = 0.38, respectively); the diastolic pressure-volume relationship, however, was displaced slightly leftward and markedly downward, suggesting relief of pericardial constraint. A multilinear regression analysis, performed with pooled data at baseline and during infusion of nitroglycerin in each patient, identified left atrioventricular pressure crossover at the onset of mitral flow as the only significant predictor (p less than 0.02) of peak E wave velocity in the circumstances considered. Thus the interaction among covariates of early left ventricular filling and the relationship between filling and diastolic left ventricular and left atrial properties can be addressed with relative ease by means of this clinically implemented approach, in an effort to sort out the contribution of each cofactor to such a complex event.