Peak rapid filling rate (PRFR) is often used clinically as an index of left ventricular relaxation, i.e., of early diastolic function. This study tests the hypothesis that early filling rate is a function of the atrioventricular pressure difference and hence is influenced by the left atrial pressure as well as by the rate of left ventricular relaxation. As indexes, we chose the left atrial pressure at the atrioventricular pressure crossover (PCO), and the time constant (T) of an assumed exponential decline in left ventricular pressure. We accurately determined the magnitude and timing of filling parameters in conscious dogs by direct measurement of phasic mitral flow (electromagnetically) and high-fidelity chamber pressures. To obtain a diverse hemodynamic data base, loading conditions were changed by infusions of volume and angiotensin LI. The latter was administered to produce a change in left ventricular pressure of less than 35% (A-1) or a change in peak left ventricular pressure of greater than 35% (A-2). PRFR increased with volume loading, was unchanged with A-1, and was decreased with A-2; T and PCO increased in all three groups (p < .005 for all changes). PRFR correlated strongly with the diastolic atrioventricular pressure difference at the time of PRFR (r = .899, p < .001) and weakly with both T (r = .369, p < .01) and PCO (r = .601, p < .001). The correlation improved significantly when T and PCO were both included in the multivariate regression (r = .797, p < .0001). PRFR is thus determined by both the left atrial pressure and the left ventricular relaxation rate and should be used with caution as an index of left ventricular diastolic function.Circulation 74, No. 1, 187-196, 1986. RELAXATION ABNORMALITIES are one of the earliest manifestations of cardiac dysfunction and frequently precede systolic dysfunction in many disease states.'' Early filling function has been evaluated in a variety of diseases, e.g., coronary artery disease, hypertrophic cardiomyopathy, hypertensive heart disease, aortic valve disease, and congestive cardiomyop-

We studied left ventricular relaxation in the filling and transiently nonfilling working hearts of seven open-chest pentobarbital-anesthetized dogs by totally occluding the mitral annulus during one systole. In the completely isovolumic nonfilling cycle, the ventricle relaxes to a lower pressure minimum (usually negative) than in the normal filling cycle. By clamping the ventricle at end systole, we determined the pressure asymptote (Poo) under dynamic conditions. With this information, we evaluated the validity of a monoexponential characterization of relaxation. P = (P0 - Poo) exp(-t/T) + Poo (T, time constant, P0, pressure at t = 0). Plots of In(P-Poo) versus t are nonlinear and concave to the origin, thereby revealing that late relaxation is more rapid than predicted by a monoexponential relation. Nevertheless, the monoexponential T remains a useful index of relaxation and correlates well with other temporal indexes (isovolumic relaxation time and relaxation half-time). When T is calculated from a filling cycle by assuming a zero pressure asymptote, i.e., the conventional way, there is no significant difference with the true value based on the nonfilling cycle.

SUMMARY The dynamics of acute mitral regurgitation were studied in six open-chest dogs in whom a portion of the anterior leaflet was excised. Phasic mitral and aortic flows were measured electromagnetically and left ventricular filling volume, regurgitant volume (RV) and forward stroke volume (SV) were calculated. The systolic pressure gradient (SPG) between the left ventricle (LV) and left atrium (LA) was obtained from highfidelity pressure transducers. The effective mitral regurgitant orifice area (MRA) was calculated from the hydraulic equation of Gorlin.Volume infusion resulted in significant increases in both left atrial and left ventricular pressures; thus, the SPG was unchanged and the increase in RV was due primarily to the increase in MRA. Angiotensin infused to raise arterial pressure resulted in greater increments in left ventricular than left atrial pressure, so that SPG rose significantly. The increase in RV was due to increases in both MRA and SPG. Norepinephrine infusion increased systolic left ventricular pressure and SPG, while left ventricular end-diastolic pressure and left atrial pressure diminished. Despite a significant increase in SPG, RV did not increase, due to a substantial decrease in MRA. Thus, angiotensin and volume infusion induced a substantial increase in regurgitation due to the increase in MRA, while augmentation of contractility after norepinephrine infusion resulted in a decrease in regurgitation through reduction of MRA. These findings support the clinical view that maintaining a small LV with sustained myocardial contractility will reduce mitral regurgitation. Alternatively, left ventricular dilatation can enhance mitral regurgitation by increasing the effective regurgitant orifice independent of SPG.NORMAL MITRAL VALVE FUNCTION depends on the mechanical integrity of the mitral annulus, valve leaflets, chordae tendineae, papillary muscles and the contraction of the free left ventricular (LV) wall. The factors that determine the regurgitant flow in mitral insufficiency are the systolic pressure gradient (SPG) between the left ventricle and left atrium, the size of the regurgitant orifice or the area of the deficit in mitral closure, and the duration of the regurgitation or the length of ventricular systole.'s 2 It has been traditionally accepted that the regurgitant orifice is fixed under different circulatory states,' except in the clinical syndrome of papillary muscle dysfunction in which the properties of the supporting structures may change with time.3 However, recent studies of experimental acute mitral insufficiency in the dog have suggested that the size of the regurgitant orifice is not fixed.' The present study was designed to define the effects of alterations in ventricular volume, pressure loading and myocardial contractility on the regurgitation of experimental acute mitral insufficiency, and to determine if the size of the regurgitant orifice is altered by these interventions. Materials and MethodsSix mongrel dogs weighing 20-30 kg were anesthetized with sodium ...

To examine the mechanisms of mitral valve motion in mid diastole and at closure, we simultaneously measured mitral flow (electromagnetic), valve motion (echo), and atrioventricular pressures (micromanometer). Peak valve excursion (E point) occurs early 46 +/- 7 ms) after opening and always precedes peak flow; therefore, mid-diastolic closing motion (EF slope) is not due to flow deceleration or vortex formation. Large variations in peak flow are accompanied by small variations in valve excursion (coefficient of variation 41 vs. 12%, respectively). We conclude that the valve overshoots its equilibrium position and that the chordae produce tension on the valve during diastole. This approach is supported by data from papillary muscle rupture, prolonged P-R interval, and mathematical modeling. We offer a valve-closure theory unifying chordal tension, flow deceleration, and vortices, with chordal tension as a necessary condition for the proper function of the other two. Nevertheless, prolonged periods of diastasis and ventricular premature contractions indicate that competent valve closure may occur in the absence of vortices and flow deceleration.

Midwall sarcomere lengths near maximal have been reported at left ventricular end-diastolic pressures at the upper limits of normal, but an ascending limb of cardiac function can exist at much higher filling pressures. Accordingly, an analysis was made of sarcomere length distributions across the left ventricular wall over a range of end-diastolic pressures between 2 and 20 mm Hg. In 11 dogs, significant ascending limbs of ventricular function were documented at filling pressures between 2 and 12 mm Hg and between 12 and 20 mm Hg. Nine hearts were arrested and rapidly fixed in diastole with intracoronary glutaraldehyde perfusion, and tissues from five sites equally distributed across the left ventricular free wall were examined by electron microscopy. At filling pressures at the lower end of the normal range (2, 6, and 12 mm Hg), an uneven distribution of sarcomere lengths across the wall was noted: mean sarcomere lengths were shortest at the subendocardial layer, longest at a site between the subendocardial and the midwall layers, and then progressively shorter toward the epicardium. All differences were highly significant. At an end-diastolic pressure of 20 mm Hg, the difference in sarcomere lengths between layers was small; sarcomere lengths decreased only slightly from endocardium to epicardium. Despite maximal sarcomere lengths just inside the midwall in hearts fixed at 12 mm Hg (2.253/A), sarcomere lengths in other layers-the subendocardial and the subepicardial regionswere increased substantially from 2.129 and 2.183^,, respectively, at 12 mm Hg to 2.283 and 2.247/x at 20 mm Hg (P < 0.001 and P < 0.002, respectively). Therefore, sarcomeres from all layers of the wall increase in length up to a filling pressure of 12 mm Hg; however, as filling pressure is increased further, only short sarcomeres from the subendocardial and the subepicardial layers are recruited. The latter mechanism forms the ultrastructural basis for the ascending limb of normal ventricular function at elevated left ventricular filling pressures. • According to Starling's law of the heart, ventricular performance is a function of initial muscle length: an increase in ventricular volume or filling pressure produces an increase in actively developed pressure or stroke volume (1, 2). In isolated cardiac muscle, over the physiological range of the length-tension curve (3, 4), sarcomere length tends to ,be a direct function of muscle length, providing an ultrastructural explanation for the Frank-Starling mechanism (5, 6).From the Cardiovascular Division, Department of Medicine, University of California San Diego, School of Medicine, La Jolla, California 92037.This work was supported by U. S. Public Health Service Grant HL-12373 from the National Heart and Lung Institute. Dr. Covell is the recipient of Career Development Award HE-21132 from the National Heart and Lung Institute.This investigation was presented in part before the American Federation for Clinical Research, Atlantic City, New Jersey, April 30, 1972. Received July 14, 1972...