Effect of volume loading, pressure loading, and inotropic stimulation on left ventricular torsion in humans.

Abstract

BACKGROUND The transmural distribution of fiber angles and the extent of shortening among obliquely oriented fibers are likely to be major determinants of the twisting motion that accompanies left ventricular (LV) ejection. As such, measurements of torsion may provide useful information about LV contractile function, but other factors, such as ventricular loading conditions, may also regulate this motion. METHODS AND RESULTS Torsion angles (theta i) of midventricular and apical regions were measured relative to a reference minor axis near the base in seven human cardiac allografts from biplane radiographic images of metallic midwall markers. Pressure loading with methoxamine (5-10 muk/kg/min) increased LV end-systolic pressure by 41 +/- 14 mm Hg (p less than 0.0001). Volume loading with normal saline raised LV end-diastolic pressure from 9.9 +/- 5.2 to 19.6 +/- 4.9 mm Hg (p less than 0.0001). These alterations in LV loading conditions were associated with no change in theta i (difference not significant) for any marker site. Inotropic stimulation with dobutamine (5 micrograms/kg/min) increased values of theta i by as much as twofold (p less than 0.05); this response varied considerably depending on marker location, with the middle and apical inferior wall and the apical lateral wall being the most sensitive. When the marker site associated with the largest torsion angle (theta max) was considered in each patient, dobutamine increased theta max in all cases (25.2 +/- 10.5 degrees versus 15.8 +/- 7.7 degrees, p less than 0.001), whereas pressure and volume loading had negligible effects. This 59% increase in theta max was greater than that of conventional load-dependent indexes of LV systolic performance such as stroke volume (16%), ejection fraction (22%), and maximum rate of LV pressure rise (52%). CONCLUSIONS This component of LV motion is relatively insensitive to alterations in preload and afterload, while changes in contractile state influence LV torsion in a regionally heterogeneous manner. Quantification of LV torsion may, therefore, provide a sensitive and relatively load-independent measure of contractile performance that may prove to be useful in the serial assessment of LV function.