Ex vivo performance of muscle powered cardiac assist device: potential for right ventricular support

The present study addressed the filling sensitivity and power output of the TOYOBO VAD (TVAD) and Utah soft artificial ventricle (USAV) to evaluate these pumps as a muscle powered cardiac assist device (MCAD). Two pumps were assembled with three different types of driving chambers placed underneath the latissimus dorsi (LD) muscle: the soft spindle (SS) type, elastic spindle (ES) type and bellows (B) type. The USAV required a filling time of greater than 1000 msec at any preload with any driving chamber, although the TVAD allowed this only with the ES-type chamber assembly. In an ex vivo mock study, the ES- and SS-type chamber assembly demonstrated a maximum stroke volume (SV) of 7.7 mL (39% stroke) by the USAV and 12 mL (17% stroke) by the TVAD at an afterload below 20 mmHg. On the other hand, the B-type chamber increased SV according to afterload in both pumps, resulting in a maximum SV of 25 mL (36% stroke) by the TVAD and 7.6 mL (38% stroke) by the USAV at a 30 mmHg afterload. The maximum stroke work (SW) achieved was 1.3 x 10(6) erg by the TVAD and 0.5 x 10(6) erg by the USAV, which exceeds that of the canine right ventricle. In conclusion, the ES-type chamber provided best pump filling and the B-type chamber best pump ejection, but back pressure at the chamber diaphragm determined performance efficiency. An active filling mechanism for the driving chamber will be necessary to offset the low preload requirements of a VAD, and provide the maximum power output necessary for right ventricular support.