The use of synchronized biventricular epicardial compression achieved by inflation of a pneumatically driven bladder surrounding the heart has been proposed as a means of assisting the failing heart.1,2 Unlike other types of assist devices, such compression devices do not contact the blood and thus may be associated with fewer complications related to thromboembolic events. The physiologic principles by which direct cardiac compression (DCC) enhances ventricular pumping strength have been demonstrated in isolated canine hearts.3-6 The end-systolic pressurevolume relationship (ESPVR) is shifted upward (indicating increased effective contractile strength) by an amount proportional to the inflation pressure of the compression bladder.
In both the isolated ejecting heart and the heart in situ, application of biventricular DCC also results in decreased right ventricular (RV) and left ventricular (LV) preload volumes and pressures.6 On the basis of measurements showing that the end-diastolic pressure volume relationship (EDPVR) is minimally altered by the device,7 it has been demonstrated that the reduction in preload is mainly a consequence of the increased pumping capacity of the supported heart. Although reductions of preload are clinically beneficial, they decrease the net flow and pressuregenerating capacity of the heart (Starling mechanism) even in the presence of DCC.6,7 For example, in a canine model of severe acute ischemic heart failure (defined as a reduction in cardiac output to ~30% of normal), DCC doubled cardiac output, but net flow during support amounted to only about 60% of normal. This increase in cardiac output was associated with a decrease in LV end-diastolic pressure (EDP) from a mean value of 22 to 12 mm Hg and an increase in arterial pressure from 55 to 100 mm Hg.7 However, according to physiologic principles, cardiac output can be normalized if preload and afterload are appropriately manipulated.