In vitro and in vivo testing showed a safe and effective function of the ReinHeart-TAH system. This has the potential to become an alternative to transplantation. However, before a first-in-man implant, chronic animal trials still have to be completed.

The hybrid MCL that we developed allows variable and continuous in vitro evaluation of mechanical circulatory support devices in TAH configuration and particularly their control algorithms in response to various cardiovascular conditions. The system has been built in a modular configuration to allow testing of different types of devices and thus provides a valuable test platform prior to animal experiments.

Hemocompatibility of blood pumps is a crucial parameter that has to be ensured prior to in vivo testing. In contrast to rotary blood pumps, a standard for testing a pulsatile total artificial heart (TAH) has not yet been established. Therefore, a new mock circulation loop was designed to investigate hemolysis in the left ventricle of the ReinHeart TAH. Its main features are a high hemocompatibility, physiological conditions, a low priming volume, and the conduction of blood through a closed tubing system. The mock circulation loop consists of a noninvasive pressure chamber, an aortic compliance chamber, and an atrium directly connected to the ventricle. As a control pump, the clinically approved Medos-HIA ventricular assist device (VAD) was used. The pumps were operated at 120 beats per minute with an aortic pressure of 120 to 80 mm Hg and a mean atrial pressure of 10 mm Hg, generating an output flow of about 5 L/min. Heparinized porcine blood was used. A series of six identical tests were performed. A test method was established that is comparable to ASTM F 1841, which is standard practice for the assessment of hemolysis in continuous-flow blood pumps. The average normalized index of hemolysis (NIH) values of the VAD and the ReinHeart TAH were 0.018 g/100 L and 0.03 g/100 L, respectively. The standard deviation of the NIH was 0.0033 for the VAD and 0.0034 for the TAH. Furthermore, a single test with a BPX-80 Bio-Pump was performed to verify that the hemolysis induced by the mock circulation loop was negligible. The performed tests showed a good reproducibility and statistical significance. The mock circulation loop and test protocol developed in this study are valid methods to investigate the hemolysis induced by a pulsatile blood pump.

No abstract

BackgroundGood anatomical compatibility is an important aspect in the development of cardiovascular implants. This work analyzes the interaction of the pump unit of an electrically driven pulsatile Total Artificial Heart (TAH) and the mediastinum. For an adequate compliance, both overall dimensions and alignment of inlets and outlets must be matched.MethodsCross-sectional medical image data of 27 individuals, including male and female patients suffering from end stage heart failure, was segmented and reconstructed to three dimensional (3D) surface models. Dimensions and orientations of relevant structures were identified and analyzed. The TAH surface model was virtually placed in orthotopic position and aligned with atrioventricular valves and big vessels. Additionally seven conventional cadaver studies were performed to validate different pump chamber designs based on virtual findings. Thereby 3D-coordinates were captured and introduced to the virtual environment to allow quantitative comparison between different individuals.ResultsSpatial parameters varied more in male patients with higher values if heart failure persists. Good correlation of the virtual analysis both to literature data and conventional cadaver studies could be shown. The full data of the 27 individuals as well as the summarized values found in literature are enclosed in the appendix. By superimposing the TAH-volume model to the anatomy, various misalignments were found and the TAH-design was adjusted.ConclusionsVirtual fitting allows implant design adjustments in realistic anatomy which has not been influenced by thoracotomy. Higher numbers of relevant individuals can be reasonably investigated in the virtual environment and quantitatively correlated. Using this approach, conventional cadaver studies can be significantly reduced but not obviated, due to the unavailable haptic feedback and immobility of potentially compressed structures.