A multicenter evaluation of external outflow graft obstruction with a fully magnetically levitated left ventricular assist device.

BACKGROUND: The HeartMate 3 (HM 3; Abbott) left ventricular assist device (LVAD) has improved hemocompatibility-related adverse outcomes. In sporadic cases, external compression of the outflow graft causing obstruction (eOGO) can result from substance accumulation between the outflow graft and its bend relief. We sought to evaluate the prevalence, course, and clinical implications of eOGO in an international study. METHODS: A multicenter retrospective analysis of HM 3 LVADs implanted between November 2014 and April 2021 (n = 2108) was conducted across 17 cardiac centers in 8 countries. We defined eOGO as obstruction >25% in the cross-sectional area in imaging (percutaneous angiography, computed tomography, or intravascular ultrasound). The prevalence and annual incidence were calculated. Serious adverse events and outcomes (death, transplantation, or device exchange) were analyzed for eOGO cases. RESULTS: Of 2108 patients, 62 were diagnosed with eOGO at a median LVAD support duration of 953 (interquartile range, 600-1267) days. The prevalence of eOGO was 3.0% and the incidence at 1, 2, 3, 4, and 5 years of support was 0.6%, 2.8%, 4.0%, 5.2%, and 9.1%, respectively. Of 62 patients, 9 were observed, 27 underwent surgical revision, 15 underwent percutaneous stent implantation, 8 received a heart transplant, and 2 died before intervention. One patient underwent surgical revision and later stent implantation. The mortality with therapeutic intervention was 9/53 (17.0%). CONCLUSIONS: Although uncommon, HM 3 LVAD-supported patients might develop eOGO with an increasing incidence after 1 year of support. Although engineering efforts to reduce this complication are under way, clinicians must maintain a focus on early detection and remain vigilant.

Association of thrombophilia prospective detection with hemocompatibility related outcomes in left ventricular assist device patients.

INTRODUCTION: Inherited thrombophilias represent a concerning risk factor due to a proclivity to an aberrant clot formation. However, in patients with left ventricular assist device (LVAD), their impact on bleeding and thrombotic complications remains still poorly understood. The aim of the present study was to evaluate the effect of thrombophilic mutation directed anticoagulation therapy on adverse clinical outcomes in LVAD patients. MATERIALS AND METHODS: About 138 consecutive patients indicated for LVAD implant (HeartMate II, Abbott, Plymouth, USA) were prospectively screened for three major thrombophilic mutations: factor II (prothrombin), factor V Leiden, and homozygous methylenetetrahydrofolate reductase (MTHFR). Subsequently, discordant individualized anticoagulation targets of INR 2.5-3.0 in thrombophilia positive and INR 1.8-2.2 in negative patients were established; notably without anti-platelet agents given the center standard of care. RESULTS: Mean age was 50 ± 12.7 years, 83% male. Mean duration of support was 464.5 days (SD 482.9; SEM 41.1) and median of 310 days (IQR 162; 546). Full thrombophilia positive cohort analysis has not revealed any significant impact on event free survival. In contrast, detailed analysis of specific thrombophilias subsets has revealed Factor II prothrombin mutation as a significant predisposition for the pump thrombosis risk (SHR 10.48; p = 0.001) despite more aggressive prespecified anticoagulation target. Moreover, the incidence of bleeding events in prothrombin group was also significantly increased (SHR 6.0; p = 0.03). CONCLUSIONS: Our observations suggest that specific thrombophilias in LVAD patients may pose different intensity predisposition for thrombotic complications. Factor II (prothrombin) positive mutation was identified as significant risk factor associated with the pump thrombosis.

Increased pulsatility index is associated with adverse outcomes in left ventricular assist device recipients.

AIMS: Recipients of left ventricular assist devices (LVAD) are exposed to increased risk of adverse clinical events. One of the potential contributing factors is non-pulsatile flow generated by LVAD. We evaluated the association of flow patterns in carotid arteries and of increased arterial stiffness with death and cerebrovascular events in LVAD recipients. METHODS AND RESULTS: We analysed data from 83 patients [mean age 54 ± 15 years; 12 women; HeartMate II (HMII), n = 34; HeartMate 3 (HM3), n = 49]. Pulsatile and resistive indexes, atherosclerotic changes in carotid arteries (measured by duplex ultrasound), and arterial stiffness [measured by Endo-PAT 2000 as the augmentation index standardized for heart rate (AI@75)] were evaluated 3 and 6 months after LVAD implantation. Sixteen patients died during follow-up (27.3 months; interquartile range 15.7-44.3). After adjusting for the main variables examined, the pulsatility index measured at 3 months was positively associated with increased hazard ratios (HR) for death and cerebrovascular events [HR 9.8, 95% confidence interval (CI) 1.62-59.42], with HR increasing after adding AI@75 to the model (HR 18.8, 95% CI 2.44-145.50). In HM3 recipients, HR was significantly lower than in HMII recipients (HR 0.31, 95% CI 0.11-0.91), but the significance disappeared after adding AI@75 to the model (HR 0.33, 95% CI 0.09-1.18). CONCLUSIONS: The risk of death and cerebrovascular events in LVAD recipients is associated with increased pulsatility index in carotid arteries and potentiated by increased arterial stiffness. The same risk is attenuated by HM3 LVAD implantation, but this effect is weakened by increased arterial stiffness.

The effect of long-term left ventricular assist device support on flow-sensitive plasma microRNA levels.

BACKGROUND: Implantation of current generation left ventricular assist devices (LVADs) in the treatment of end-stage heart failure (HF), not only improves HF symptoms and end-organ perfusion, but also leads to cellular and molecular responses, presumably in response to the continuous flow generated by these devices. MicroRNAs (miRNAs) are important post-transcriptional regulators of gene expression in multiple biological processes, including the pathogenesis of HF. In our study, we examined the influence of long-term LVAD support on changes in flow-sensitive miRNAs in plasma. MATERIALS AND METHODS: Blood samples from patients with end-stage heart failure (N = 33; age = 55.7 ± 11.6 years) were collected before LVAD implantation and 3, 6, 9, and 12 months after implantation. Plasma levels of the flow-sensitive miRNAs; miR-10a, miR-10b, miR-146a, miR-146b, miR-663a, miR-663b, miR-21, miR-155, and miR-126 were measured using quantitative PCR. RESULTS: Increasing quantities of miR-126 (P < 0.03) and miR-146a (P < 0.02) was observed at each follow-up visit after LVAD implantation. A positive association between miR-155 and Belcaro score (P < 0.04) and an inverse correlation between miR-126 and endothelial function, measured as the reactive hyperemia index (P < 0.05), was observed. CONCLUSIONS: Our observations suggest that after LVAD implantation, low pulsatile flow up-regulates plasma levels of circulating flow-sensitive miRNAs, contributing to endothelial dysfunction and vascular remodeling.

An Integrative Study of Aortic mRNA/miRNA Longitudinal Changes in Long-Term LVAD Support.

Studying the long-term impact of continuous-flow left ventricular assist device (CF-LVAD) offers an opportunity for a complex understanding of the pathophysiology of vascular changes in aortic tissue in response to a nonphysiological blood flow pattern. Our study aimed to analyze aortic mRNA/miRNA expression changes in response to long-term LVAD support. Paired aortic samples obtained at the time of LVAD implantation and at the time of heart transplantation were examined for mRNA/miRNA profiling. The number of differentially expressed genes (Pcorr < 0.05) shared between samples before and after LVAD support was 277. The whole miRNome profile revealed 69 differentially expressed miRNAs (Pcorr < 0.05). Gene ontology (GO) analysis identified that LVAD predominantly influenced genes involved in the extracellular matrix and collagen fibril organization. Integrated mRNA/miRNA analysis revealed that potential targets of miRNAs dysregulated in explanted samples are mainly involved in GO biological process terms related to dendritic spine organization, neuron projection organization, and cell junction assembly and organization. We found differentially expressed genes participating in vascular tissue engineering as a consequence of LVAD duration. Changes in aortic miRNA levels demonstrated an effect on molecular processes involved in angiogenesis.

The Effect of Artificial Pulsatility on the Peripheral Vasculature in Patients With Continuous-Flow Ventricular Assist Devices.

BACKGROUND: Implantation of left-ventricular assist systems (LVASs) has become the standard of care for advanced heart failure (HF). The absence of pulsatility in previous devices contributes to vascular and endothelial dysfunction related to atherosclerotic or vascular complications. We hypothesized that the artificial pulsatility provided by the HeartMate 3 (HM3) (Abbott, Chicago, IL) LVAS would exert a favourable effect on the vasculature. METHODS: In 32 patients implanted with HM3 (5 female patients, mean age 55 ± 13.6 years), the reactive hyperemia index (RHI) and peripheral augmentation index (AI), markers of endothelial function and arterial stiffness, were measured with an EndoPAT2000 before and in the third and sixth month after implantation. RHI and AI data from 30 HeartMate II (HM II) (Abbott) recipients in the third and sixth month after implantation, from 15 patients with advanced HF without LVASs and from 13 healthy volunteers were also analyzed. RESULTS: In HM3 recipients, the mean RHI significantly decreased at 3 and 6 months after implantation. The RHI was substantially lower at baseline than that of healthy or the HF reference group. Increasing AI values, indicating worsening arterial stiffness, were also observed. Similar trends were observed in HM II recipients between the third and sixth months but with higher absolute values of RHI and AI. CONCLUSIONS: We detected impaired vascular function in HM3 patients and provided additional evidence on the negative effect of low pulsatility on vascular function after LVAS implantation. The results suggest that the artificial pulsatility of the HM3 does not avert the progression of endothelial dysfunction.

Evaluation of low-intensity anti-coagulation with a fully magnetically levitated centrifugal-flow circulatory pump-the MAGENTUM 1 study.

BACKGROUND: The HeartMate 3 left ventricular assist system is engineered to avoid pump thrombosis, yet bleeding complications persist. We investigated the safety of low-intensity anti-coagulation in patients with the HeartMate 3. METHODS: The Minimal AnticoaGulation EvaluatioNTo aUgment heMocompatibility (MAGENTUM 1) pilot study is a prospective, single-arm study of low-intensity warfarin anti-coagulation in patients implanted with the HeartMate 3 pump. After standard warfarin anti-coagulation (international normalized ratio [INR] 2.0 to 3.0) and aspirin for 6 weeks post-implant, patients were transitioned to a lower INR target range of 1.5 to 1.9. The primary end-point was a composite of survival free of pump thrombosis, disabling stroke (modified Rankin score [MRS] >3), or major bleeding (excluding peri-operative bleeding) with at least 6-month post-implant follow-up. Time in therapeutic range (TTR) was measured to assess anti-coagulation target efficacy using the Rosendaal method. A safety algorithm to monitor for signs of pump thrombosis was developed and implemented. RESULTS: We enrolled 15 patients (mean age 57.3 ± 13.3 years), 13 men with advanced heart failure (67% with INTERMACS Profiles 2 or 3), irrespective of therapeutic goal of bridge-to-transplant or destination therapy. The primary end-point was met in 14 of 15 (93 ± 6%) patients; 1 patient developed recurrent gastrointestinal bleeding. The TTR during the reduced anti-coagulation phase (6 weeks to 6 months) was 75.3 ± 8.6%. No thrombotic events occurred. CONCLUSIONS: This pilot study suggests low-intensity anti-coagulation targeting an INR between 1.5 and 1.9 is achievable and safe with the HeartMate 3 cardiac pump in the short-term phase, 6-months post-implant. A large-scale trial is now warranted.