Quantification of incomplete left ventricular relaxation: relationship to the time constant for isovolumic pressure fall

We initially found that the fall in left ventricular pressure during isovolumic relaxation is exponential and therefore characterized by a time constant (T). To the extent that isovolumic pressure fall reflects myocardial events during relaxation, T indexes the time course of relaxation of the left ventricle. In isolated canine hearts sudden beat to beat loading changes from isovolumic to ejection shorten T. In contrast, under stable hemodynamic conditions T is independent of load across a broad, physiologically significant range. For example:(1) in nine isovolumic hearts with end-diastolic pressure from 7±0.5 to 22±0.5 mm Hg and peak left ventricular pressure (LVP) from 95 ± 11 to 155 ± 12 mm Hg, T did not change significantly; (2) in seven ejecting hearts from a stroke volume of 10 to 20 ml and from a peak LVP of 85 to 192 mm Hg, T was not altered. T changed significantly with factors thought to affect the active cardiac relaxing system such as heart rate, catecholamines and recovery from ischaemia. To quantify the extent of incomplete left ventricular relaxation between beats and to identify the time at which relaxation no longer influences the diastolic ventricular pressure-dimension relationships, we measured pressure and endocardial dimension by sonomicrometry in ten working dog hearts. We obtained the fully relaxed pressure-dimension relationship for individual hearts from pressure-dimension values following prolonged diastoles. From the linear relationship between pressure and dimension we predicted the left ventricular dimension which would be present at any observed diastolic pressure if relaxation were complete. At intervals of 0.25 T or less we compared this predicted fully relaxed dimension with the actual dimension. The predicted and actual dimension were not significantly different at end-diastole in any beat where the next beat began > 3.5 T after maximal negative dP/dt. This indicates completion of relaxation prior to end-diastole in all such beats. An end-diastolic difference in predicted v. actual dimension (when the beat began < 3.5 T after maximal negative dP/dt) identified incomplete relaxation. This difference between the predicted fully relaxed dimension and the actual dimension quantified the extent of incomplete relaxation. Earlier in diastole this difference quantified the effect of relaxation on ventricular dimension and thus diastolic filling. Under widely varying hemodynamic conditions, except with severe failure, the time of the end of the effect of relaxation on diastolic pressure-dimension was 2.85 ± 0.33 (s.d.) T after maximal negative dP/dt. Under severe hemodynamic stress this time occurred earlier in diastole. This early end of the influence of relaxation on diastolic pressure-dimension relationships could be predicted from P at maximal negative dP/dl, T, and the fully relaxed diastolic pressure-dimension relationship. Thus, T is a load-independent index of left ventricular pressure fall under stable hemodynamic conditions. Use of T aids in the identification of changes in the relaxation time course, quantification of incomplete relaxation and the time course and extent of the effect of relaxation on pressure-dimension relationships during the diastolic filling period. The extent of incomplete relaxation and/or the effect of relaxation on diastolic dimensions throughout diastole can be quantified by comparing the actual dimension to the fully relaxed dimension at any pressure during diastole.