Comparing velour versus silicone interfaces at the driveline exit site of HeartMate II devices: infection rates, histopathology, and ultrastructural aspects.

BACKGROUND: Driveline exit site (DLES) infection is a major complication of ventricular assist devices (VADs). Differences in the sheath material interfacing with exit site tissue appear to affect healing time and infection risk more than site hygiene, but the mechanistic basis for this is not clear. METHODS: Health record data from Utah Artificial Heart Program patients with HeartMate II (HMII) devices implanted from 2008 to 2012 were retrospectively reviewed, with particular attention to interface type, incorporation (healing) time, and infections. Tissue samples from the DLES were collected at the time of VAD removal in a small subset. These samples were examined by routine histology and environmental scanning electron microscopy (ESEM). RESULTS: Among 57 patients with sufficient data, 15 had velour interfaces and 42 had silicone. Indications for and duration of support were similar between the groups. The silicone group had shorter incorporation time (45 ±22 vs. 56 ±34 days, P=.17) and fewer DLES infections (20% vs. 1.7%, P=.026, for patient infections and 0.0340 vs. 0.166, P=.16, for infections per patient-year). Tissues from five patients, three with velour, were examined. Velour interfaces demonstrated more hyperkeratosis, hypergranulosis, and dermal inflammation. By ESEM, the silicone driveline tracts appeared relatively smooth and flat, whereas the velour interface samples were irregular with deep fissures and globular material adhering to the surface. CONCLUSIONS: Using the silicone portion of the HMII driveline at the DLES was associated with fewer infections and a trend toward faster healing in this small retrospective series. Whether the intriguing microscopic differences directly account for this needs further study on a larger scale.

Allograft rejection in patients supported with continuous-flow left ventricular assist devices.

BACKGROUND: Both pulsatile-flow and continuous-flow left ventricular assist devices (LVADs) successfully provide patients a bridge to transplantation. Some data suggest that continuous-flow pumps increase the risk of allograft rejection, contributing to posttransplantation morbidity and mortality. We sought to analyze the relationship between LVAD flow characteristics and subsequent allograft rejection in bridge to transplant (BTT) patients. METHODS: Patients with LVADs from the UTAH Transplant Affiliated Hospitals were retrospectively analyzed. Rejection was determined pathologically according to the International Society for Heart and Lung Transplantation revised cardiac allograft rejection scale. Multimodal statistical analyses were applied. RESULTS: Of 1,076 patients who underwent transplantation over a 26-year period, 151 had LVADs. Of these, 111 (77 pulsatile flow, 34 continuous flow) patients had pathologic data available. There was no difference in overall rejection (grades 1R to 3R) between the pulsatile-flow LVAD and continuous-flow LVAD groups (2.00 ± 1.43 versus 1.50 ± 1.16 episodes/year; p = 0.076.) Patients with pulsatile-flow LVADs had more clinically relevant (grades 2R to 3R) rejection than did patients with continuous-flow LVADs (0.49 ± 0.72 versus 0.12 ± 0.33 episodes/year; p < 0.001). There was no survival difference at 1 year (p = 0.920) or 4 years (p = 0.721) after transplantation. CONCLUSIONS: Patients with continuous-flow LVADs have similar overall rejection rates and a reduced rate of clinically relevant rejection compared with patients with pulsatile-flow LVADs during the first year after transplantation. Although there is theoretical concern that nonphysiologic, nonpulsatile flow could alter the neurohormonal profile of patients in heart failure, we are encouraged that the type of LVAD circulation does not influence posttransplantation allograft survival.

Differential impact on post-transplant outcomes between pulsatile- and continuous-flow left ventricular assist devices.

BACKGROUND: The HeartMate II (HMII) left ventricular assist device (LVAD) has proven reliable and durable and has become the preferred choice for bridge to transplant therapy (BTT) when compared with the pulsatile HeartMate XVE (XVE). In this study, we compared the post-transplant (PTx) outcomes between XVE and HMII using a large national data registry. METHODS: The Organ Procurement and Transplantation Network (OPTN)/United Network for Organ Sharing (UNOS) Thoracic Registry database was queried for all patients implanted with either an XVE or an HMII as BTT during 2004-2009. Statistical analysis between XVE and HMII were performed using Kaplan-Meier survival analysis and Cox regression analyses. RESULTS: A total of 673 patients were implanted with the XVE and 484 with HMII. When adjusted for age, gender, ethnicity, intra-aortic balloon pump, ventilator, inotropes, dialysis, body mass index, creatinine, bilirubin, transfusion, pulmonary capillary wedge, and pulmonary arterial pressures, the HMII had similar one- and three-yr survival (hazard ratio = 0.95, CI = 0.64, 1.42) and rejection-free survival PTx compared to XVE. The XVE group had more early incidences of allograft rejection (AR) and hospitalization for infection (HI). CONCLUSIONS: Compared to XVE, patients with HMII have similar one- and three-yr survival after heart transplantation with less risk of early graft rejection and significant infection. With a strong shift toward use of continuous-flow LVADs, PTx outcomes are expected to continue to improve.

A novel non-invasive method to assess aortic valve opening in HeartMate II left ventricular assist device patients using a modified Karhunen-Loève transformation.

BACKGROUND: Thrombus formation on or near the aortic valve has been reported in HeartMate II (Thoratec, Pleasanton, CA) left ventricular assist device (LVAD) patients whose aortic valves do not open. With an akinetic valve, thrombogenesis is more likely. Thrombus formation may lead to neurologic events, placing the patient at greater risk. Aortic valve stenosis and/or regurgitation have also been observed with akinetic aortic valves. Assessing aortic valve opening is crucial when optimizing rotations per minute (rpm) to minimize embolic risk and aortic valve stenosis but presently relies solely on echocardiography, intermittent decreases in rpms to force aortic valve opening, and monitoring of pulse pressure. We hypothesized the electrical current waveforms of the HeartMate II would reveal whether the aortic valve was opening due to pressure changes in the left ventricle to allow for continuous monitoring and control of aortic valve opening ratios. METHODS: Electrical HeartMate II current waveforms of patients from 2008 to 2009 that were recorded at the time of echocardiograph procedures were analyzed using a modified Karhunen-Loève transformation with a training set of electrical waveforms from 8,860 HeartMate II electrical current recordings from 2001 to 2009. RESULTS: The study included 6 patients. The electrical current magnitude of the projection of the electrical current waveforms onto the training set's eigenvectors was statistically significantly greater in 4 of the 6 patients when the aortic valve was closed, confirmed by echocardiography. The 2 patients who did not have a large increase in the magnitude had mild aortic valve regurgitation. CONCLUSION: Electrical current analysis for rotary non-pulsatile pumps is a means to develop a physiologic feedback algorithm for an auto-mode, which currently does not exist. Constant regulation and optimization of rotary non-pulsatile LVADs would minimize patients' risk for neurologic events and aortic valve stenosis.