Pulsatility and VWF for Mechanical Circulatory Support

Study Questions:

What is the relation between pulsatility and von Willebrand factor (VWF) under continuous-flow mechanical circulatory support (CF-MCS)?


The study investigators used one in vitro endothelial-free mock circulatory loop model and two experimental swine models with CF-MCS to investigate the relationship between pulsatility and VWF multimerization. Using these devices, they investigated in a dose-effect model (model 2) three levels of pulsatility in three groups of swine. In a cross-over model (model 3), they studied the effects of sequential changes of pulsatility on VWF. They used two percutaneous micro-axial catheter-mounted high shear rotary pumps adapted from Impella-CP (MCS-A) and Impella-5.0 (MCS-B), and using a dedicated cannula, adapted to pig anatomy constraints (Abiomed Europe-GmbH, Aachen, Germany). The two pumps were designed to induce very similar shear with a tip velocity of 9.5 and 10.0 m × s-1 for MCS-A and -B, respectively. During all of the experiments, the pumps were used at a constant and maximal speed with a maximum flow of 3.2 L/min (MCS-A) and 4.5 L/min (MCS-B). They reported the evolution of VWF multimerization in a patient undergoing serial CF-MCS and/or pulsatile-MCS; the results obtained in this patient are part of the “first clinical use of a bioprosthetic total artificial heart” CARMAT study. To compare the dynamic of high molecular weight (HMW) multimers between the groups, they fitted a repeated measures two-way analysis of variance (ANOVA) model with time and groups as factors.


In the in vitro model, a rapid time-dependent loss of HMW multimers (>60% after 5 minutes and >90% after 30 minutes) was observed in both MCS-A and -B experiments. The loss of HMW multimers coincided with a loss of collagen-binding activity. Loss of HMW multimers and collagen-binding activity followed a similar time course in MCS-A and -B (p = 0.578 and p = 0.771 for HMW multimers and VWF:CB/VWF:Ag, respectively).

In the swine model (model 2 or dose-effect of pulsatility on VWF parameters) when combining all data points, a strong correlation was observed between arterial pulsatility and HMW multimer ratio (r = 0.73; p < 0.01). Of note, no significant increase AMP in LDH was observed during the time of the experiment (30 minutes), with no difference between the two devices (MCS-A or -B) in any of the locations (left ventricle or aorta). The swine model (model 3 or consequences of sequential changes of pulsatility on VWF parameters) showed that pulse pressure (PP) can be significantly and easily manipulated by relocating the same device (MCS-B) from one position to another (left ventricle vs. aorta and vice versa). The first low-pulsatility phase (phase 1:MCS in LV-1) was associated with a rapid and significant loss of HMW multimers (0.74 [IQR: 0.58-0.88], 0.51 [IQR: 0.38-0.60], and 0.39 [IQR: 0.22-0.48] at 5, 15, and 30 minutes, respectively vs. 0.90 [IQR: 0.84-1.10] at baseline; ANOVA p < 0.0001). After stopping the pump for 60 minutes (phase 2: native heart only-1) and restoring a low shear and a normal pulsatility, a complete recovery of HMW-multimer ratio (at 1.0 [IQR: 0.89-1.26]) was obtained. During the normal PP phase (phase 3: MCS in aorta), a decrease of HMW multimer ratio was also observed (0.87 [IQR: 0.69-1.04], 0.76 [IQR: 0.73-0.88], and 0.74 [IQR: 0.62-0.82] at 5, 15, and 30 minutes, respectively; ANOVA p < 0.0001).

In the patient with cardiogenic shock the study investigators observed that the PP during the first phase of CF-MCS was very low (<10 mm Hg). In the second phase, after PF-MCS implantation occurred, there was a high increase of PP (51 mm Hg [IQR: 42.5-54 mm Hg]). Then, during the last phase of dual support (CF-MCS + PF-MCS), a significant drop of PP (25 mm Hg [IQR: 20-30 mm Hg] vs. 51 mm Hg [IQR: 42.5-54 mm Hg], p < 0.01) occurred. Under CF-MCS (high shear and low pulsatility), a marked decrease of both VWF:Act/VWF:Ag and HMW-multimer ratio (0.6 and 0.3, respectively) were observed. Three hours after implantation, the PF-MCS support provided a rapid restoration of functional VWF with a complete normalization of both VWF:Act/VWF:Ag and HMW multimers and a parallel increase in VWFpp/VWF:Ag ratio from 1.0 to 2.2. During the 7 days of PF-MCS (low shear and normal pulsatility), VWF:Act/VWF:Ag and HMW multimers remained into the normal range (VWF:Act/VWF:Ag ratio = 1.11 [IQR: 1.04-1.22], HMW-multimers ratio = 0.87 [IQR: 0.78-0.98]). The hemodynamic and respiratory failure during PF-MCS required the addition of CF-MCS to PF-MCS (phase 3: high shear and low pulsatility). Upon initiation of PF-MCS, the investigators noted a recurrence of a functional VWF defect that remained stable over time (VWF:Act/VWF:Ag ratio = 0.51 [IQR: 0.47-0.55], HMW-multimers ratio = 0.10 [IQR: 0.07-0.10]). Under single PF-MCS support, a continuous increase inVWF:Ag occurred from 295 IU/dl just after implantation to 581 IU/dl after completing 7 days of PF-MCS. After implantation of additional CF to PF-MCS, VWF:Ag dropped to 369 IU/dl within the first day of dual support. Then, a progressive and milder rise of VWF:Ag was observed from 326 IU/dl in the second day to 491 IU/dl after completing 11 days of dual support. A progressive rise of VWFpp was noted under PF-MCS from 385 to 660 IU/dl between day 6 and day 12. After the addition of CF-MCS to PF-MCS, a progressive decrease of VWFpp occurred, from 660 IU/dl at day 13 to 363 IU/dl at day 23.


The study investigators concluded that the VWF defect reflects the balance between degradation induced by the shear stress and the endothelial release of new VWF triggered by the pulsatility. And this modulation of VWF levels could explain the relationship between pulsatility and bleeding observed in CF-MCS recipients. They opined that reservation of pulsatility may be a new target to improve clinical outcomes of patients.


The findings from the swine model and case report in a patient with cardiogenic shock support the hypothesis (N Engl J Med 2013;368:579-80) that pulsatile flow during MCS preserves VWF from excessive degradation. More research is needed to confirm the benefits of pulsatility in long-term MCS patients. Until then, it is reasonable to strive for as much pulsatile flow as possible to reduce the risk of bleeding in MCS patients.

Clinical Topics: Anticoagulation Management, Cardiac Surgery, Dyslipidemia, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Cardiac Surgery and Heart Failure, Lipid Metabolism, Acute Heart Failure, Heart Failure and Cardiac Biomarkers

Keywords: Anticoagulants, Blood Pressure, Catheters, Cardiac Surgical Procedures, Extracorporeal Membrane Oxygenation, Heart, Artificial, Heart Failure, Hemorrhage, Pulsatile Flow, Respiratory Insufficiency, RNA-Binding Proteins, Shock, Cardiogenic, von Willebrand Factor

< Back to Listings