THE EFFECTS OF LEFT-HEART ASSIST ON RIGHT-VENTRICULAR MUSCLE MECHANICS AND VENTRICULAR COUPLING IN THE INJURED HEART

So that we could better characterize the effects of left heart assist on right ventricular myocardial muscle mechanics and ventricular mecha nical coupling in the injured heart, nine dogs underwent 30 minutes of global cardiac ischemia supported by cardiopulmonary bypass followed by randomly varied levels of left heart assist at 0, 1.0, and 2.0 L/mi n (0, 37 +/- 4, and 74 +/- 7 ml/kg per minute). A centrifugal pump wit h left ventricle-to-aorta bypass was used with the intent to cause lef t ventricular volume unloading but without complete left ventricular p ressure unloading. Right ventricular regional free wall and septal-fre e wall dimensions were measured by a sonomicrometer and right ventricu lar pressure by a micromanometer. Pressure and dimension data were acq uired over a range of preloads produced by transient vena caval occlus ion and at steady state at an initial control point and after ischemia at each level of left heart assist. Right ventricular regional early diastolic function was assessed by percent segmental relaxation during the first third of diastole, end-diastolic compliance by the end-dias tolic pressure-dimension relationship, systolic contractile performanc e by the slope (M(w)) and dimension axis intercept (L(w)) of the linea r preload recruitable stroke work relationship, and right ventricular isovolumic relaxation by the pressure decay time constant. Ischemia re duced M(w) of both the free wall (38.3 +/- 16.1 to 16.4 +/- 4.2 erg cm (-3) 10(3), p < 0.01) and septal free wall (30.2 +/- 12.7 to 13.4 +/- 4.9 erg.cm(-3).10(3), p < 0.01) and shifted L(w) rightward (1.3 +/- 0. 3 to 1.4 +/- 0.3 mm, p < 0.01, and 2.8 +/- 0.8 to 3.0 +/- 0.9 mm, p < 0.01), which confirmed myocardial ischemic injury. There were no effec ts of left heart assist on free wall or septal-free wall systolic cont ractile performance assessed by M(w) and L(w) or on early diastolic re laxation assessed by percent segmental relaxation during the first thi rd of diastole in either right ventricular region (all p = not signifi cant). There were also no observed characteristic alterations of free wall or septal-free wall end-diastolic pressure-dimension relationship s with left heart assist. The pressure decay time constant decreased w ith increasing levels of left heart assist (51 +/- 14, 49 +/- 16, and 43 +/- 11 msec, p < 0.05), which indicated an improvement in right ven tricular isovolumic relaxation attributable to left heart assist. Thes e data demonstrate that mechanical ventricular interactive effects dur ing left heart assist are beneficial, but limited to isovolumic relaxa tion in the injured heart. The likely optimal method of left heart ass ist for postcardiotomy support should sufficiently augment cardiac out put and arterial pressure but maintain left ventricular systolic press ure generation to preserve beneficial ventricular mechanical coupling.