Minimally invasive left ventricular assist device implantation: optimizing device design for this approach.

Introduction: The global heart failure (HF) burden is expected to increase due to aging populations, increasing number of end-stage HF patients and adverse lifestyle changes. Mechanical circulatory support (MCS) devices such as left ventricular assist devices (LVADs) have become a promising treatment option for short-term and long-term circulatory support of end-stage HF patients.Areas covered: Recent developments in MCS technology have been focused on miniaturization leading to the development of minimally invasive surgical procedures for LVAD implantation. This helps overcome possible postoperative complications such as major incisions and poor outcomes due to infections, right heart failure, and bleeding. This article discusses clinical and technological developments in the field of minimally invasive procedures for LVAD implantation.Expert opinion: Most patients might benefit from minimally invasive LVAD implantation performed through a limited left lateral thoracotomy associated with an upper hemisternotomy or a right anterior thoracotomy. The thoracotomy approach can also be considered in case of pump exchange or pump explant. The success of these techniques is mainly based on the optimization of LVAD pump design, inflow cannula insertion, and outflow graft as well as driveline exit sites. The future direction of the LVAD field is likely to include less-invasive approaches and smartificial technologies.

Photographic LVAD Driveline Wound Infection Recognition Using Deep Learning.

The steady increase in the number of patients equipped with mechanical heart support implants, such as left ventricular assist devices (LVAD), along with virtually ubiquitous 24/7 internet connectivity coverage is motive to investigate and develop remote patient monitoring. In this study we explore machine learning approaches to infection severity recognition on driveline exit site images. We apply a U-net convolutional neural network (CNN) for driveline tube segmentation, resulting in a Dice score coefficient of 0.95. A classification CNN is trained to predict the membership of one out of three infection classes in photographs. The resulting accuracy of 67% in total is close to the measured expert level performance, which indicates that also for human experts there may not be enough information present in the photographs for accurate assessment. We suggest the inclusion of thermographic image data in order to better resolve mild and severe infections.

In vitro study for the evaluation of transluminal aspiration as a novel treatment option for thrombosis in the HeartWare HVAD.

INTRODUCTION:: Pump thrombosis of left ventricular assist devices remains a devastating complication with high morbidity and mortality. Despite the improvements made, the matter affects many patients and the treatment options are limited to thrombolysis and surgical replacement. An alternative approach using the aspiration Indigo catheter was tested. METHODS:: An Indigo thrombectomy catheter was used within an in vitro model to assess the direct aspiration of prefabricated clots from three different positions within the HeartWare HVAD (inlet, outlet, and housing). The experiments were conducted with a straight and an angled catheter. The aspiration pressure was constant. The flow, power consumption, and pressure head of the left ventricular assist devices were measured at pre-defined measuring points. RESULTS:: The device was more effective (success rate 71%) at inlet and outlet of the left ventricular assist device. In addition, the duration of aspiration and the aspiration volume were shorter in comparison to the aspiration in the housing (inlet M = 19.75 s, outlet M = 60.50 s, and housing M = 38.75 s). Moreover, the aspiration volume was associated with the aspiration duration and the weight of thrombi but not with their volume. Noteworthy, the angled catheter showed an improved performance compared to the straight one (67%-33%). The recorded parameters showed no major changes during the use of the catheter. After application of the Indigo catheter, flow and pressure head of the pump could be restored. CONCLUSIONS:: The aspiration system showed promising results under specific conditions for the treatment of pump thrombosis in an in vitro model. However, further examination, including in vivo experiments, will justify its effectiveness.

An acoustic method for systematic ventricular assist device thrombus evaluation with a novel artificial thrombus model.

BACKGROUND: Pump thrombosis (PT) is still one of the major adverse events in patients supported with left ventricular assist devices. Nowadays, thrombus detection relies on clinical parameters like reoccurring heart failure symptoms, on changes in pump power consumption, and on laboratory parameters such as increased LDH and hemolysis. Once detected PT is most often persistent and refractory to medical therapy. We therefore designed a novel, non-invasive acoustic method for early pump thrombus detection in an in vitro artificial thrombus model. METHODS: The study was performed in vitro using a mock circulation loop, artificial blood (water-glycerin) and artificial thrombus material (silicon) allowing for repeatable and defined testing. Tested ventricular assist device (VAD) type was HVAD (Medtronic). Three different thrombus locations were evaluated: on the tilted pad of the rotor, in the primary flow path, and in the secondary flow path beneath the rotor. After evaluating baseline parameters (no thrombus, n=20 for each pump), the influence of thrombi of seven different masses (no thrombus, 0.5-5.0 mg) on pump power consumption and acoustic emission of four HVAD devices was investigated via a microphone system (Sennheiser) and subsequent frequency spectrum analysis (n=12). The acoustic analysis algorithm included the number of frequency peaks recorded. RESULTS: Measurements with thrombi on the tilted pad showed an increased number of frequency peaks with all thrombus sizes compared to baseline measurements without any thrombus (baseline: 32.7±7.4; 0.5 mg: 45.3±10.4 up to 5 mg: 80.4±5.5). Power consumption was relevantly elevated in 5mg thrombus measurement only (6.3±1.29 W compared to 4.9±0.14 W at baseline). Measurements with thrombi in the primary and secondary showed no relevant alteration in power consumption and frequency peak count. CONCLUSIONS: We present an acoustic method that detects pump thrombi located on the tilted pad of the HVAD rotor requiring ten times less mass compared to thrombi detected by power consumption alterations used in current detection algorithms. Assuming that pump thrombi are growing over several days, the presented method may detect PT much earlier thereby increasing efficacy of medical therapy and helping to avoid pump exchange.