Skeletal muscle ventricles in continuity with the bloodstream.

BACKGROUND AND AIM OF STUDY

The prevalence of end-stage congestive heart failure and limitation of clinical alternative treatments present the need for creative new solutions. Formation of a ventricle from skeletal muscle (SMV) has shown promise in the animal laboratory. Two modes of the SMV for cardiac assistance, the counterpulsation (CP-SMV) and the ventricular assist (VA-SMV), using the latissimus dorsi muscle were applied in a canine model. Ability to augment arterial pressure was assessed. The effect of stimulation delay on the degree of augmentation was also evaluated.

METHODS AND RESULTS

Thirty-five SMVs were connected in continuity with the bloodstream in the two modes: (1) CP-SMV (aorta-to-aorta) (n = 12); and (2) VA-SMV (left ventricular [LV] apex-to-aorta) (n = 23). In the CP-SMV mode, designed to simulate the intra-aortic balloon pump, the SMV was simply interposed into the path of the descending aorta (DAo) without prosthetic valves in either the inflow or the outflow conduit. In order to obligate blood flow through the SMV, the DAo was ligated between the two grafts. In the VA-SMV mode, the connection was made with valved conduits from the LV apex (inflow) to the ascending aorta (outflow) (n = 11) or to the DAo (n = 12). The ascending aorta (AAo) was also ligated proximal to the outflow conduit for the same reason of obligating blood flow through the SMV. The SMV was timed to contract in diastole in both the CP-SMV mode and the VA-SMV mode. In the VA-SMV mode, the average systolic pressure without stimulation was 101.6 +/- 2.2 mmHg and with stimulation 118.21 +/- 4.78 mmHg (mean augmentation, 14.5 +/- 2.6 mmHg) (p < 0.01). In the CP-SMV mode, the average systolic pressure without stimulation was 97 +/- 32 mmHg and with stimulation, 122 +/- 26 mmHg (mean augmentation, 25 +/- 8.6 mmHg) (p < 0.001). We also extended earlier work on timing of stimulation of isolated SMV by evaluating the effect of stimulation delay on the degree of augmentation in continuity with the bloodstream with the SMV in the VA-SMV configuration. Delays of 50 msec to 225 msec were evaluated. SMV stimulation was via the thoracodorsal nerve at an amplitude of 1.5 V and a frequency of 25 Hz. The greatest augmentation occurred at a stimulation delay of 150 msec (p < 0.001).

CONCLUSION

Both counterpulsation and assist configurations produced effective diastolic augmentation. Although diastolic augmentation occurred with all timing delays, the optimal delay was 150 msec. Complications in the survival animals include AAo problems, SMV rupture, respiratory insufficiency, intraoperative instability, and thrombosis (which occurred in 51% [18/35] of the animals). This high frequency of thrombosis in the canine model suggests the use of a less thrombogenic SMV lining, more aggressive or prolonged anticoagulation, or an alternative animal model.