A comprehensive comparison of the in vitro hemocompatibility of extracorporeal centrifugal blood pumps

Abstract

Purpose: Blood damage has been associated with patients under temporary continuous-flow mechanical circulatory support. To evaluate the side effects caused by transit blood pumping, in vitro hemocompatibility testing for blood damage in pumps is considered a necessary reference before clinical trials. Methods: The hemocompatibility of five extracorporeal centrifugal blood pumps was investigated comprehensively, including four commercial pumps (the Abbott CentriMag, the Terumo Capiox, the Medos DP3, and the Medtronic BPX-80) and a pump in development (the magAssist MoyoAssist^®). In vitro, hemolysis was tested with heparinized porcine blood at nominal operating conditions (5 L/min, 160 mmHg) and extreme operating conditions (1 L/min, 290 mmHg) using a circulation flow loop. Hematology analyses concerning the blood cell counts and the degradation of high-molecular-weight von Willebrand factor (VWF) during 6-h circulation were also evaluated. Results: Comparing the in vitro hemocompatibility of blood pumps at different operations, the blood damage was significantly more severe at extreme operating conditions than that at nominal operating conditions. The performance of the five blood pumps was arranged in different orders at these two operating conditions. The results also demonstrated superior hemocompatibility of CentriMag and MoyoAssist^® at two operating conditions, with overall low blood damage at hemolysis level, blood cell counts, and degradation of high-molecular-weight VWF. It suggested that magnetic bearings have an advantage in hemocompatibility compared to the mechanical bearing of blood pumps. Conclusion: Involving multiple operating conditions of blood pumps in in vitro hemocompatibility evaluation will be helpful for clinical application. In addition, the magnetically levitated centrifugal blood pump MoyoAssist^® shows great potential in the future as it demonstrated good in vitro hemocompatibility.

Keywords: ECMO; blood damage; centrifugal pump; hemolysis; von Willebrand factor.

FIGURE 1

Schematic diagram and picture of the blood circulation loop. The loop setup allows for control of temperature, flow rate, and pressure head.

FIGURE 2

Averages of (A) NIH and (B) MIH for all the centrifugal blood pumps. The sample sizes were MoyoAssist^® (n = 3; n = 3), CentriMag (n = 3; n = 3), DP3 (n = 3; n = 1), Capiox (n = 3; n = 2), and BPX-80 (n = 2; n = 1), representing nominal operating conditions and extreme operating conditions. Porcine blood subjected to circulation loops for 6 h. Hemolysis was measured and calculated in accordance with ASTM. A value of p < 0.05 (**) was considered to be statistically significant.

FIGURE 3

Comparison of temporal changes among control and centrifugal blood pumps on (A) white blood cells at nominal operating conditions, (B) platelets at nominal operating conditions, (C) white blood cells at extreme operating conditions, and (D) platelets at extreme operating conditions. The sample sizes were the same as those for the hemolysis test. Porcine blood was subjected to a circulation loop for 6 h and measured by automatic hematology analysis every 2 h. Results expressed as mean ± SD, % relative to each time zero samples. A value of p < 0.05 (**) was considered to be statistically significant.

FIGURE 4

Representative immunoblot gel image of VWF multimers for all the centrifugal blood pumps at (A) nominal operating conditions and (B) extreme operating conditions. The sample sizes were the same as those for the hemolysis test. Porcine blood was subjected to circulation loops for 6 h and then measured by immunoblotting.

FIGURE 5

Degradation of high-molecular weight VWF multimers for all the centrifugal blood pumps at (A) nominal operating conditions and (B) extreme operating conditions. Results expressed as mean ± SD, % relative to each time zero sample. A value of p < 0.05 (**) was considered to be statistically significant.