A comparison of pulsatile and non-pulsatile flow in ventricular assist devices: effects on bacteraemia, end-organ function and vascular reactivity

Abstract

SUMMARY BACKGROUND In the setting of a progressive shortage of donors for heart transplantation, there has been a movement in recent years towards a new generation of ventricular assist devices (VAD’s). These VAD’s are based on continuous rather than pulsatile flow and are smaller and thus easier to implant. They offer greater durability with the prospect of implantation as a permanent (destination) device. Pulsatile VADs implanted at the Alfred Hospital (Melbourne, Australia) include the Thoratec and the Novacor. Since the end of 2006, the pulsatile VADs have become obsolete and are no longer implanted. The centrifugal Ventrassist VAD has been used at the Alfred hospital since 2003. Although widely adopted, there remains a lack of knowledge of what effect non-pulsatile devices have on bacteraemia, end- organ function and vascular structure and reactivity. The first two studies were retrospective studies (Chapters 3 & 4) and the remainder of the chapters were prospective studies (Chapters 5, 6 & 7). The aims of this research were: 1. To determine the effect of device type (pulsatile vs non-pulsatile) on the incidence of bacteraemia 2. To compare the incidence of impaired end organ function including gastrointestinal complications, neurological symptoms and renal impairment in non-pulsatile vs pulsatile VADs 3. To determine the effect of VAD implantation on vascular function as assessed by flow mediated dilation (FMD) and forearm blood flow (FBF) measurements 4. To determine the effect of VAD implantation and of pulsatile and non-pulsatile flow on the endothelium as assessed by blood borne endothelial markers 5. To determine the effect of cardiac transplantation on vascular function and endothelial markers. 6. To determine the effect of the following on intima-medial thickness: heart failure, VAD implantation whether pulsatile or non-pulsatile and transplantation. Bacteraemia and VAD infections Infection is a significant source of morbidity and mortality after VAD implantation. Since the introduction of the considerably smaller non-pulsatile flow VADs, which require much smaller pump pockets than the pulsatile VADs, we assume that they would have less exposure to infection. With much smaller drive-line exit sites, the non-pulsatile devices certainly have an potential advantage of lower risk of infection and being the newer device on the market, medical staff looking after patients with a VAD in situ, have a much better understanding of detecting and treating infections than before. In our study we examined the incidence of bacteraemia and infectious complications in patients receiving pulsatile and non-pulsatile VADs. At the end of the study, we concluded that device type including differences in pulsatility made no difference to the incidence of bacteraemia. GI complications and End-organ function We observed more gastrointestinal (GI) problems with the CF than with the pulsatile devices. Motility problems and bowel infarctions were likely to be due to gut ischemia contributed by non-pulsatile flow. In the current era an overall reduction in GI problems has been clearly demonstrated with the newer CF devices. These improvements are likely to be due to more careful patient selection, better preoperative optimization, improved operative techniques, and enhanced postoperative management. Since these improvements have been introduced, the incidence of ischaemic-type GI complications fallen. However, GI bleeding remains a problem. VASCULAR FUNCTION AND STRUCTURE Function and structure of the vascular system was evaluated utilising the following approaches: • Measurement of endothelium-dependent regulation of vascular tone at focal sites of the circulation (Flow Mediated Vasodilation [FMD]) • Measurement of vascular reactivity using Venous Occlusion Plethysmography / Forearm Blood Flow [FBF] • Measurement of morphologic characteristics of the vascular wall using Intima- media thickness (IMT) • Determination of soluble endothelial markers (e.g. Von Willebrand factor, plasminogen activator and adhesion molecules) Flow Mediated Dilation Arterial flow-mediated dilation (FMD), assessed by high-resolution ultrasonography, reflects endothelium-dependent vasodilator function. Endothelial function is known to be impaired in patients with heart failure (HF), but the vascular effects of a non-pulsatile ventricular assist device (VAD) implantation, now used as either a bridge to transplantation or as destination therapy, remain unclear. Using flow-mediated vasodilation (FMD), this study aimed to provide greater insight into VAD-induced changes in vascular function. In our study we did not see significant changes either VAD implantation, whether pulsatile or non-pulsatile, or transplantation, but observed a deterioration over time. Intima Medial Thickness The carotid intima medial thickness was increased in heart failure patients, but no effects of VAD implantation or pulsatility were observed. IMT measurements did however decrease over time after VAD insertion and also after heart transplantation. Forearm Blood Flow In this study, patients with heart failure had decreased vascular reactivity compared to the normal subjects. Patients receiving a VAD showed an increase in vascular function, however, patients supported with pulsatile VADs had superior vascular reactivity compared to patients supported with non-pulsatile flow VADs, indicating that the pulsatile blood flow characteristics of pulsatile VADs offer significant benefits in vascular reactivity over the continuous blood flow characteristics of non- pulsatile flow VADs. There was no change observed over time or after heart transplantation. Endothelial Markers After analysing the endothelial markers before and after VAD insertion, between pulsatile and non-pulsatile flow, it was clear that there were no real differences between the two groups. The markers for both groups were elevated within the first few days after device implantation but declined later as would be expected when a foreign body is introduced. As was shown with the FMD and FBF, the body adapts to these devices over time, whether pulsatile or non-pulsatile. Summary A marked increase in GI complications was found in the patients who received a non-pulsatile device. Pulsatile devices were found to improve vascular function as evidenced by FBF. Therefore it is likely that having a pulse should reduce GI and other complications, particularly in the early post-operative phase when patients are at their most vulnerable. The lack of pulsatility may be the one factor that has been overlooked all these years and future research should focus on methods to generate pulsatility in these newer continuous flow devices.