H. Hayes

Management Options to Treat Gastrointestinal Bleeding in Patients Supported on Rotary Left Ventricular Assist Devices: A Single‐Center Experience

Gastrointestinal (GI) bleeding in ventricular assist devices (VADs) has been reported with rotary devices. The pathophysiological mechanisms and treatments are in evolution. We performed a retrospective review of GI bleeding episodes for all VADs implanted at our institution. Five male patients experienced GI bleeding-age 63.6 ± 3.64 years. VAD type VentrAssist n = 1, Jarvik 2000 n = 2, and HeartWare n = 2. All patients were anticoagulated as per protocol with antiplatelet agents (aspirin and/or clopidogrel bisulfate [Plavix] and warfarin (therapeutic international normalized ratio 2.0-3.5). There was no prior history of gastric bleeding in this group. Ten episodes of bleeding requiring blood transfusion occurred in five patients. Some patients had multiple episodes (1 × 5, 1 × 2, 3 × 1). The events occurred at varying times post-VAD implantation (days 14, 21, 26, 107, 152, 189, 476, 582, 669, and 839). Octreotide (a long-acting somatostatin analogue that reduces splanchnic arterial and portal blood flow) was administered subcutaneously or intravenously. Three patients received infusions of adrenaline at 1 µg/min to enhance pulsatility. Anticoagulation was interrupted during bleeding episodes but successfully introduced post bleeding event. GI bleeding is a significant complication of VAD therapy. In this article, we discuss diagnosis and management options.

Factors associated with anti-human leukocyte antigen antibodies in patients supported with continuous-flow devices and effect on probability of transplant and post-transplant outcomes

BACKGROUNDnOne major disadvantage of ventricular assist device (VAD) therapy is the development of human-leukocyte antigen (HLA) antibodies. We aimed to identify factors associated with HLA antibodies during continuous flow (CF)-VAD support and assess the effect on transplant probability and outcomes.nnnMETHODSnWe included 143 consecutive heart failure patients who received a CF-VAD as a bridge-to-transplant at 3 institutions. Factors associated with post-VAD peak panel reactive antibodies (PRA) among several measurements were identified using multivariable linear regression. A parametric survival model was used to assess transplant waiting time and probability, risk of rejection, and a composite outcome of rejection, graft failure, and death.nnnRESULTSnThirty-six patients (25%) were female; mean age was 47 ± 13 years. Eighty-one patients (57%) had a pre-VAD PRA of 0%, and 16 were highly sensitized (PRA > 80%). Age, female sex, and pre-VAD PRA were independently associated with post-VAD PRA. A 10-year increase in age was associated with a 5% decrease in post-VAD PRA (p = 0.03). Post-VAD PRA was 19% higher in women vs men (p < 0.01). A 10%-increase in pre-VAD PRA was associated with a 4.7% higher post-VAD PRA (p < 0.01). During a mean follow-up of 12 ± 11 months, 90 patients underwent cardiac transplantation. A 20% increase in post-VAD PRA was associated with 13% lower probability of transplant (hazard ratio, 0.87; 95% confidence interval, 0.76-0.99). A high PRA was not associated with adverse post-transplant outcomes.nnnCONCLUSIONSnYounger age, female sex, and pre-VAD PRA were independent predictors of elevated PRA post-VAD. Higher PRA was significantly associated with lower transplant probability but not increased rejection, graft failure, or death after transplant.

Clinical Outcomes of Patients Treated With Pulmonary Vasodilators Early and in High Dose After Left Ventricular Assist Device Implantation.

Right ventricular failure (RVF) is common after left ventricular assist device (LVAD) implantation and a major determinant of adverse outcomes. Optimal perioperative right ventricular (RV) management is not well defined. We evaluated the use of pulmonary vasodilator therapy during LVAD implantation. We performed a retrospective analysis of continuous-flow LVAD implants and pulmonary vasodilator use at our institution between September 2004 and June 2013. Preoperative RVF risk was assessed using recognized variables. Sixty-five patients (80% men, 50u2009±u200914 years) were included: 52% HeartWare ventricular assist device (HVAD), 11% HeartMate II (HMII), 17% VentrAssist, 20% Jarvik. Predicted RVF risk was comparable with contemporary LVAD populations: 8% ventilated, 14% mechanical support, 86% inotropes, 25% BUN >39u2009mg/dL, 23% bilirubin ≥2u2009mg/dL, 31% RV : LV (left ventricular) diameter ≥0.75, 27% RA : PCWP (right atrium : pulmonary capillary wedge pressure) >0.63, 36% RV stroke work index <6u2009gm-m/m(2)/beat. The majority (91%) received pulmonary vasodilators early and in high dose: 72% nitric oxide, 77% sildenafil (max 200u2009±u200979u2009mg/day), 66% iloprost (max 126u2009±u200937u2009μg/day). Median hospital stay was 26 (21) days. No patient required RV mechanical support. Of six (9%) patients meeting RVF criteria based on prolonged need for inotropes, four were transplanted, one is alive with an LVAD at 3 years, and one died on day 35 of intracranial hemorrhage. Two-year survival was 77% (92% for HMII/HVAD): transplanted 54%, alive with LVAD 21%, recovery/explanted 2%. A low incidence of RVF and excellent outcomes were observed for patients treated early during LVAD implantation with combination, high-dose pulmonary vasodilators. The results warrant further investigation in a randomized controlled study.