Preservation of von Willebrand Factor Activity With the ModulHeart Device

Figures

• 

  Figure 1. ModulHeart Device (A) The ModulHeart device in its assembled configuration with the 3 pumps anchored in parallel within a self-expandable anchor (expanded). (B) The ModulHeart device implanted transaxillary and positioned in the proximal abdominal aorta. The arrows indicate the direction of pump outflow.

• 

  Figure 2. Study Design (A) Model 1: von Willebrand factor (VWF) activity over pump support time (60 minutes) was evaluated in an endothelium-free mock circulatory loop. The ModulHeart device was compared to a no pump control, and 2 Impella devices (CP and 5.0). (B) Model 2: VWF activity was assessed in an in vivo swine model during multiple 60-minute cycles with the ModulHeart device at 14,000 rpm. (C) Model 3: VWF was evaluated in 4 patients undergoing high-risk percutaneous coronary intervention.

• 

  Figure 3. Custom Mock Circulatory Loop Set-Up Four identical mock circulatory loops were built to simulate intra-aortic placement of circulatory support devices. Loops were fitted with the ModulHeart device, Impella CP, Impella 5.0, and 1 loop received no device (negative control). All devices were run simultaneously at matching flow rates. Shown in the loop diagram are (A) clamp-on flow meter, (B) 9.5-mm tubing, (C1 and C2) flow restriction clamps, (D) device being tested, (E) 22-mm tubing, (F) loop reservoir, (G) sampling port, and (H) temperature sensor.

• 

  Figure 4. VWF Parameters Measured in an Endothelium-Free Continuous-Flow Circulatory Loop (A) There was no significant difference between von Willebrand factor (VWF) activity with the ModulHeart device compared with the no pump control (P = 0.86; repeated measures 2-way analysis of variance with time and groups as factors). VWF activity was 1.05 ± 0.14 vs baseline levels after 60 minutes of ModulHeart operation at 14,000 RPM. There was a 27% loss in VWF activity with the Impella CP and 19% loss with the Impella 5.0 at 60 minutes. Level of change in VWF activity over time between the devices did not reach statistical difference. (B) VWF antigen levels remained stable over time in each experimental setup (repeated measures 1-way analysis of variance P = 0.49, P = 0.46, P = 0.17, and P = 0.51 for negative control, ModulHeart, Impella 5.0, and Impella CP, respectively). Results are presented as mean ± SD. CBA:Ag = collagen binding activity:antigen.

• 

  Figure 5. Change in VWF Parameters in Swine Under ModulHeart Support (A) There was no change in VWF activity after 60 minutes of ModulHeart support at 14,000 RPM in 3 healthy swine (repeated measures 1-way analysis of variance P = 0.13). (B) VWF antigen levels were preserved over time (repeated measures 1-way ANOVA P = 0.40). Results are presented as mean ± SD. Abbreviations as in Figure 4.

• 

  Figure 6. VWF in Patients Undergoing High-Risk PCI With ModulHeart Support (A) Postprocedure von Willebrand factor (VWF) activity remained unchanged compared with preprocedure levels (Wilcoxon signed-rank test P = 0.88). (B) There was no significant variation in VWF antigen levels postprocedure (paired t-test P = 0.078). Results are presented as mean ± SD. PCI = percutaneous coronary intervention.