Acquired Von Willebrand syndrome in patients on long-term support with HeartMate II

Abstract

INTRODUCTION

Acquired Von Willebrand syndrome (AVWS) has been reported to contribute to bleeding complications in patients with left ventricular assist devices (VAD) including HeartMate II (HMII) [1–8]. Acquired Von Willebrand syndrome is caused by increased shear stress resulting from the contact of blood with foreign surfaces in the device and the cannulas that connect the device with the heart and the great vessels. The shear stress leads to unfolding of the high-molecular-weight multimers of the Von Willebrand factor (VWF) and finally to cleavage of the VWF by the protease ADAMTS-13 [9]. AVWS is characterized by the loss of the high-molecular-weight (HMW) multimers of VWF. Consequently, the collagen-binding capacity of the VWF (VWF:CB) and the binding to the platelet glycoprotein Ib/IX (GPIb/IX) receptor complex (measured as VWF:activity, VWF:A) are decreased. We reported previously that the onset of AVWS after VAD implantation occurs early and can be detected in most patients within 1 day after VAD implantation [10–12]. Although fast reversal of AVWS after explanation of the VAD has been shown [5, 13], data regarding the presence of AVWS in patients on long-term VAD support are rare. The aim of this prospective study was to collect data regarding VWF parameters and bleeding symptoms of patients with VAD support (HMII) over a long-term period.

METHODS

Patients

Long-term follow-up was scheduled for 3, 6 and 12 months after left ventricular assist device implantation and every year thereafter. Intermediate analyses were done occasionally. Overall, 278 data sets of 74 patients (14 women, 60 men, age 54.4 ± 14.3, median 58.5, range 18.3–75.2 years) were analysed at time points ranging from 3 to 80 months (mean 11.2 ± 12.1 months, median 6.3 months). The data were obtained within the scope of our institutional monitoring programme on haemostatic changes during support by mechanical VADs that had been approved by the ethics committee of the University of Freiburg. All patients provided their informed consent before any data were used for research purposes.

Our standard anticoagulation protocol for patients with VADs consists of phenprocoumon with a target international normalized ratio (INR) of 2.5–3.0 and acetylsalicylic acid (ASS) 100 mg daily In case of recurrent bleeding episodes ASS is discontinued. The patients received unfractionated heparin during the first days after surgery. We always measured not only activated partial thromboplastin time (aPTT), but also the heparin level (anti-FXa). When the patients had stabilized after surgery and the chest tube had been removed, the patients were switched to cumarine (phenprocoumon) and ASS. The cumarine therapy was monitored by Quick and INR.

Surgical procedures: VADs were implanted according to the usual surgical technique as previously described [4, 14].

Laboratory analyses

Sodium citrate plasma was obtained by centrifugation at 1500 × g for 15 min at 20 °C and analysed within 4 h after blood was drawn or stored at −80 °C. VWF antigen (VWF:Ag; Siemens Healthcare Diagnostics, Tarrytown, NY, USA) and ristocetin cofactor activity (VWF:RCo; Siemens Healthcare Diagnostics) were measured using the Behring Coagulation System according to standard procedures. Collagen type I (Nycomed Pharma, Unterschleissheim, Germany) was immobilized on a microtitre plate, and the collagen binding capacity (VWF:CB) in plasma was determined photometrically using the ELISA technique. We calculated the ratios of VWF:RCo and VWF:CB, respectively, to VWF:Ag (VWF:RCo/VWF:Ag, normal: >0.65; VWF:CB/VWF:Ag, normal: >0.7). These values reflect the biological activity of the available VWF with regard to binding to platelets and to collagen.

Von Willebrand factor multimers were separated on SDS-agarose low-resolution gels (1.0% agarose) and blotted on a polyvinylidene difluoride membrane to assess the HMW multimers. Von Willebrand factor was detected using the appropriate primary and secondary antibodies (DAKO, Hamburg, Germany) and stained with 3,3'-diaminobenzidine/cobalt chloride (Bio-Rad, Munich, Germany). Standard human plasma (Siemens Healthcare Diagnostics) was used for the control. An AVWS was diagnosed if HMW multimers were missing and at least one functional VWF parameter was below the normal range [15–18].

Statistics

The program IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp., Armonk, NY, USA) was used for all calculations. Variables are expressed as mean ± standard deviation, median and range as indicated in the text. The relation between bleeding events and the ratios VWF:RCo/VWF:Ag and VWF:CB/VWF:Ag was determined by binary logistic regression analysis using the mean ratios of the individual patients.

RESULTS

We determined the VWF:RCo, the VWF:CB and the VWF:Ag levels, and we calculated the ratios VWF:RCo/VWF:Ag and VWF:CB/VWF:Ag, respectively. Acquired Von Willebrand syndrome was diagnosed if a loss of VWF HMW multimers was detected and the ratios of either VWF:RCo/VWF:Ag or VWF:CB/VWF:Ag or both were reduced.

We found the ratio of VWF:RCo/VWF:Ag to be reduced in a majority of the samples, whereas VWF:CB/VWF:Ag was reduced in almost all samples (Fig. 1). Multimeric analyses of VWF were performed for 181 blood samples. We detected a loss of HMW multimers (Fig. 2) in 170 samples from 73 patients, confirming the diagnosis of AVWS. Only one of the 74 patients displayed normal patterns in both of his two samples.

Figure 1:

Ratios of (A) ristocetin-cofactor activity (VWF:RCo) and of (B) collagen-binding capacity (VWF:CB), respectively, to VWF antigen (VWF:Ag). VWF:RCo: ristocetin cofactor activity; VWF:CB: collagen-binding capacity; VWF:Ag: antigen.

Open in new tabDownload slide

Figure 2:

Presence of HMW multimers of VWF in the control and HMW multimers missing (arrow) in the plasma of a patient with an HMII. HMW: high molecular weight; VWF: Von Willebrand factor; HMII: HeartMate II.

Open in new tabDownload slide

Thirteen patients had bleeding complications. Statistical analysis showed no relationship between the ratios of the VWF parameters and the occurrence of bleeding (VWF:RCo/VWF:Ag, P = 0.618, VWF:CB/VWF:Ag, P = 0.456).

DISCUSSION

In this study we collected VWF parameters in patients with a left ventricular HMII for a long-term period (up to 80 months). We found that AVWS persisted longer than the first 30 days postoperatively. Impaired binding of VWF to platelets was found in a majority and to collagen in almost all patients as reflected by the ratios of VWF:RCo/VWF:Ag and VWF:CB/VWF:Ag, respectively. Further, HMW multimers of VWF were missing in most samples. The data indicated that AVWS seems to be a typical phenomenon in almost all patients on long-term support with HMII. There were no time-dependent changes in the VWF parameters. The long-term persistence of AVWS should be considered when these patients develop bleeding during the long period after HMII implantation.

The ratio of VWF:CB/VWF:Ag corresponded even more closely with the results of the VWF multimeric analyses (showing loss of VWF HMW multimers) than the VWF:RCo/VWF:Ag-ratio; i.e. the ratio of VWF:RCo/VWF:Ag was less sensitive for AVWS. This high specificity of collagen binding capacity was probably due to the use of a non-commercial in-house ELISA with collagen type I. Collagen type I has been shown to have a higher affinity for HMW multimers of VWF compared with collagen type III and is therefore more discriminative for AVWS [19]. In contrast, VWF:RCo/VWF:Ag has been shown to vary to a larger extent than VWF:CB/VWF:Ag with respect to AVWS [20, 21].

Uriel et al. [7] reported that 44% of patients with HMII had major bleeding events during a follow-up period of 112 ± 183 days, with 50% of them experiencing an event after 2 months. In a previous study, we found bleeding complications in almost two-thirds of patients on HMII support for longer than 2 months [22].

The bleeding tendency of patients with VAD depends on the challenge, i.e. surgery or trauma. Patients with HMII often bleed after surgery. Therapy is possible in a few patients with VADs who experience bleeding. Tranexamic acid or desmopressin can be used if the bleeding is not severe. When the bleeding is severe, VWF containing FVIII or even activated FVII (aFVII) can be applied. This therapy may help for only for a short time; however, it can stop the bleeding and improve wound healing. Therefore, if the diagnosis (AVWS) is known early, treatment (i.e. with tranexamic acid) can be started early, and further bleeding can be prevented. In addition, as mentioned above, the therapy can be modified according to the injury and the bleeding symptoms.

To our knowledge, patients have not been observed for a long period, up to 6.5 years, as in our study. The results support the assumption that AVWS is a persistent problem in patients with the HMII, which should be considered when these patients develop symptoms such as epistaxis or gastrointestinal bleeding in the long course after HMII implantation.

ACKNOWLEDGEMENTS

We are thankful for the support of our perfusionist team, and we appreciate the technical support provided by Ulrike Heizmann and Simone Rosenfelder.

Funding

This work was supported by the German Research Foundation, Bonn, Germany [ZI 486/7-1].

Conflict of interest: none declared.

REFERENCES

Author notes