Palak Shah

Adverse events in contemporary continuous-flow left ventricular assist devices: A multi-institutional comparison shows significant differences

OBJECTIVES We review differences in the incidence and timing of adverse events in patients implanted with continuous-flow left ventricular assist devices (LVADs), using the Mechanical Circulatory Support Research Network registry. METHODS From May 2004 to September 2014, a total of 734 patients (591 men; median age: 59 years) underwent primary continuous-flow LVAD implantation at our institutions. Patients implanted with the HeartMate II (HMII) (560 [76%] patients), compared with the HeartWare ventricular assist device (HVAD; 174 [24%]) were more often receiving destination therapy (47% vs 20%; P < .01), had a lower preoperative creatinine level (1.2 vs 1.3; P = .01), and had less median preoperative right ventricular dysfunction (mild vs moderate; P < .01). Ischemic etiology, prior sternotomy, and median INTERMACS profile were similar. RESULTS Overall mortality was 54 of 734 (7.4%); 41 of 560 (7.3%) in the HMII group, and 13 of 174 (7.5%) in the HVAD group (P = .95). Follow-up was available in 100% of early survivors for a median of 1 year (max: 10 years) and a total of 1120 patient-years of support (HMII: 940 patient-years [median: 1.1 years, max: 5.3 years] and HVAD: 180 patient-years [median: 0.6 year, max: 10.4 years]). On multivariable analysis, GI bleeding (P = .63), any infection (P = .32), driveline infection (P = .10), and pump thrombus (P = .64) were similar between devices while HeartWare HVAD was associated with higher risk of stroke (HR: 1.8, [1.25, 2.5], P = .003). CONCLUSIONS In this pooled analysis, a trend was found for higher incidence of percutaneous driveline infections in patients treated with the HMII; a higher incidence of stroke and time-related cumulative risk of any infection and stroke was found in patients treated with the HVAD, which was independently associated with higher stroke risk.

Treatment of device thrombus in the HeartWare HVAD: Success and outcomes depend significantly on the initial treatment strategy

BACKGROUND Pump thrombosis is a major adverse event in patients supported with a left ventricular assist device (LVAD). Treatment approaches include device exchange, lytic therapy, or augmentation of anticoagulation or antiplatelet therapy. The optimal strategy in the HeartWare HVAD Ventricular Assist System (HeartWare, Framingham, MA) is uncertain, and because few large studies have examined differing treatment outcomes, we have reviewed findings from the Mechanical Circulatory Support Research Network registry. METHODS Between March 2009 and August 2014, 175 patients (133 male) underwent implantation of the HeartWare HVAD at institutions that comprise the Mechanical Circulatory Support Research Network. Median age at implant was 59 years (range, 18-76 years). Follow-up was available in all patients for a median of 6 months (maximum, 61 months) and for a total of 163 patient-years of support. There were 36 pump thromboses (using Interagency Registry for Mechanically Assisted Circulatory Support criteria) in 21 patients for a total event rate of 0.22 events/patient-year of support; 13 patients had 1 event, 4 had 2, 2 had 3, 1 had 4, and 1 had 5. The median time to the first thrombosis was 6.4 months, and to each subsequent thrombosis was 4, 3, 2, and 2 months, respectively. Primary treatment success was defined as the patient remaining alive and within the first 30 days of the initial treatment be free from stroke, recurrence of pump thrombosis, device exchange, or urgent transplantation (United Network of Organ Sharing Status 1A). Medical treatment was defined as tissue plasminogen activator, heparin plus glycoprotein IIb/IIIa inhibitor, or heparin alone, not followed by surgical treatment within 72 hours. RESULTS Initial medical treatment was used in 29 episodes (tissue plasminogen activator in 24, heparin alone in 4, and heparin plus glycoprotein IIb/IIIa in 1) and surgical (device exchange) in 7. Medical treatment was successful in 14 of 29 episodes (48%). Complications of medical treatment included hemorrhagic stroke in 6 patients (21%), need for urgent device exchange/transplant in 6 (21%), and death in 3 (10%). Surgical treatment was successful in all 7 patients (100%). No significant early complications or early deaths occurred after device exchange. CONCLUSIONS In this large multicenter analysis, we observed that medical therapy, as the initial treatment strategy for HeartWare HVAD thrombosis, is associated with low success (48%) and a significant risk of hemorrhagic stroke (21%) and death (10%). However, initial treatment with device exchange was uniformly successful and not associated with significant early morbidity or death. Although the optimal treatment approach for HeartWare HVAD pump thrombosis remains undecided, these data do not support the routine use of medical therapy as an initial treatment strategy.

Uncorrected pre-operative mitral valve regurgitation is not associated with adverse outcomes after continuous-flow left ventricular assist device implantation

BACKGROUND Mitral valve regurgitation (MR) is prevalent in patients with heart failure. Because very few data exist examining the influence of significant pre-operative MR on outcomes after left ventricular assist device (LVAD) implantation, we evaluate our experience. METHODS Between October 1996 and August 2013, 756 patients underwent primary LVAD implantation at our institutions. Of these, 508 patients received a continuous-flow LVAD and represent the contemporary cohort for this analysis. Devices implanted included the HeartMate II in 410 patients (81%) and HeartWare HVAD in 98 patients (19%). Based on availability of pre-operative echocardiography, 491 patients were divided into 2 study groups according to degree of pre-operative MR; 189 patients (39%) had moderate to severe or greater MR (MR group), and 302 (61%) had less than moderate to severe MR (less MR group). Median age at operation (60 years in MR group vs 58 years in less MR group, p = 0.19), male sex (78% in MR group vs 81% in less MR group, p = 0.42), and ischemic etiology (46% in MR group vs 51% in less MR group, p = 0.35) were similar between groups. RESULTS There were 40 early deaths (7.9%), and follow-up was available in all 468 early survivors for 641 patient-years of support. Patients in the MR group had higher late survival (2 years, 75%; 4 years, 65%) compared with patients in the less MR group (2 years, 66%; 4 years, 48%; p < 0.04). Cox proportional hazards model confirmed the independent interaction between MR and late survival (hazard ratio 0.62, p = 0.04). CONCLUSIONS There was improved survival in patients with severe pre-operative MR after continuous-flow LVAD implantation in our cohort. These findings may lend insight into the possible lack of value of addressing significant MR at the time of LVAD implantation.

INTERMACS profiles and modifiers: Heterogeneity of patient classification and the impact of modifiers on predicting patient outcome

BACKGROUND Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) patient profiles and modifiers are descriptors of patient illness severity before durable ventricular assist device implantation. It is unknown how individual U.S. institutions and practitioners assign profiles and if modifiers improve on risk discrimination. METHODS Respondents (n = 212) to a web-based survey answered questions about the INTERMACS profile assignment process in their institution. For 5 hypothetical clinical scenarios, respondents assigned the best profile. The INTERMACS registry (2009-2014) was queried, and hazard ratio (HR) (95% confidence interval [CI]) for mortality between profiles as well as based on the presence of temporary circulatory support (TCS), frequent flyer (FF), or arrhythmia modifiers was calculated. RESULTS Respondents included 131 (62%) cardiologists, 30 (14%) surgeons, and 51 (24%) physician extenders/coordinators. Institutional INTERMACS profile assignment was variable (63% assigned by cardiologists/surgeons; 10% by research coordinators; 27% by physician extenders). Profile assignments in hypothetical patient scenarios were heterogeneous, especially for contiguous profiles. The 1-year survivals for Profiles 1, 2, and 3 were 77 ± 1.2%, 80 ± 0.7%, and 84 ± 0.7% (p < 0.001). Although Profile 1 patients had worse adjusted survival than Profile 3 patients (p = 0.001), survival for Profile 1 patients vs Profile 2 patients was similar (adjusted HR = 1.01 [95% CI = 0.88-1.12]). The TCS (adjusted HR = 1.1 [95% CI = 0.94-1.2]) and arrhythmia (adjusted HR = 1.1 [95% CI = 0.97-1.2]) modifiers were not predictive of mortality, but the FF modifier was (HR = 1.3 [95% CI = 1.02-1.63]). CONCLUSIONS Substantial heterogeneity exists in the process and assignment of INTERMACS profiles. This heterogeneity could affect mortality estimates used for risk stratification. Only the FF modifier appears to improve risk discrimination beyond that of known risk factors. Adding objective descriptors may reduce profile heterogeneity.

Clinical Outcomes of Advanced Heart Failure Patients with Cardiogenic Shock Treated with Temporary Circulatory Support Before Durable LVAD Implant.

Temporary circulatory support (TCS) is used to improve hemodynamics in patients with cardiogenic shock as a bridge to durable ventricular assist device (dVAD). Data from dVAD recipients with or without TCS (extracorporeal membranous oxygenation [ECMO], n = 14; TandemHeart [TH], n = 26) were evaluated. Clinical characteristics and hemodynamics were analyzed for patients before and after TCS and compared with profile 1 (n = 29) or profile 2–3 (n = 269) patients without TCS before dVAD. Extracorporeal membranous oxygenation patients had the highest use of preoperative mechanical ventilation, vasopressors, and the highest HeartMate II risk score before dVAD (p < 0.01). Patients receiving TCS before dVAD implant had hemodynamics comparable with patients in Profiles 2–3 and superior to that of Profile 1 patients without TCS. Operative survival after dVAD was lower in patients receiving ECMO (57%) compared with Profile 1 patients receiving TH (88%), Profile 1 without TCS (82%), or Profile 2–3 patients (97%; all p < 0.01). Despite improved clinical stability with TCS, patients bridged to a dVAD have event-free survival that parallels patients in profile 1 without TCS. Our data suggest that patients requiring TCS before dVAD implant should be still classified Interagency Registry for Mechanically Assisted Circulatory Support profile 1.

A multi-institutional outcome analysis of patients undergoing left ventricular assist device implantation stratified by sex and race

BACKGROUND Earlier studies have demonstrated disparities in patients undergoing left ventricular assist device (LVAD) implantation when stratified according to sex and race. Because very few data exist from large investigations, we reviewed data from the registry of the Mechanical Circulatory Support Research Network. METHODS Between May 2004 and September 2014, 734 patients underwent primary LVAD implantation at our institutions. Median age at implant was 57 (range 18 to 82) years and there were 577 males (80%). Race included Caucasian (C) in 586 patients (82%), African-American (AA) in 112 (16%), and other (O) in 21 (3%). Between sexes, significant pre-operative differences most commonly included median age at implant (males 60 years, females 57 years), ischemic etiology (53% vs 35%) and mean INTERMACS profile (2.9 vs 2.5). Between races, significant pre-operative differences most commonly included median age at implant (C = 61 vs AA = 51 vs O = 51), New York Heart Association functional class (85% vs 100% vs 92%) and ischemic etiology (55% vs 24% vs 40%). RESULTS There were no significant differences in survival at 1, 3 or 5 years by sex or race. Similarly, there were no differences in time-related freedom from stroke, drive-line infection, gastrointestinal bleeding or pump thrombus by sex or race. After controlling for differences, neither sex nor race was associated with survival (p = 0.09 and p = 0.18, respectively), stroke (p = 0.28 and p = 0.21), drive-line infection (p = 0.9 and p = 0.92), gastrointestinal bleed (p = 0.48 and p = 0.45) or pump thrombus (p = 0.99 and p = 0.8). CONCLUSIONS In this large, multi-institutional analysis, although some pre-operative clinical characteristics varied, they did not translate into any significant differences in late survival or complications while on LVAD support.

Bleeding and thrombosis associated with ventricular assist device therapy

Over the past decade, continuous-flow rotary pumps have dramatically improved survival for patients with advanced systolic heart failure. Bleeding and thrombosis, however, continue to be the Achilles heel of left ventricular assist device (LVAD) therapy. There is a dynamic and complex interaction between the patient and pump. The net effect of a variety of hematologic derangements, such as hemolysis, high-molecular-weight von Willebrand degradation, platelet activation and diminished pulsatility, is poorly understood. A combination of these factors mediates the common adverse events of gastrointestinal bleeding, device thrombosis and stroke. In this review we incorporate information from translational investigations in LVAD patients to understand how continuous-flow pumps activate the coagulation system and platelets predisposing to thrombosis, while, in parallel, degrade high-molecular-weight von Willebrand factor and trigger abnormal angiogenesis predisposing to bleeding. Finally, we propose novel strategies to develop a personalized approach to anti-thrombotic monitoring and titration of anti-coagulants to minimize the bleeding and thrombotic event rates of future LVAD recipients.

MicroRNAs in Heart Failure, Cardiac Transplantation, and Myocardial Recovery: Biomarkers with Therapeutic Potential

Purpose of ReviewHeart failure is increasing in prevalence with a lack of recently developed therapies that produce major beneficial effects on its associated mortality. MicroRNAs are small non-coding RNA molecules that regulate gene expression, are differentially regulated in heart failure, and are found in the circulation serving as a biomarker of heart failure.Recent FindingsData suggests that microRNAs may be used to detect allograft rejection in cardiac transplantation and may predict the degree of myocardial recovery in patients with a left ventricular assist device or treated with beta-blocker therapy. Given their role in regulating cellular function, microRNAs are an intriguing target for oligonucleotide therapeutics, designed to mimic or antagonize (antagomir) their biological effects.SummaryWe review the current state of microRNAs as biomarkers of heart failure and associated conditions, the mechanisms by which microRNAs control cellular function, and how specific microRNAs may be targeted with novel therapeutics designed to treat heart failure.

Left ventricular assist device outcomes based on flow configuration and pre-operative left ventricular dimension: An Interagency Registry for Mechanically Assisted Circulatory Support Analysis

BACKGROUND Axial configuration (AC) and centrifugal configuration (CC) left ventricular assist devices (LVAD) have different flow characteristics, and whether the interaction between device flow configuration and the pre-operative left ventricular internal diastolic diameter (LVIDD) mediates adverse events after LVAD implantation is unknown. METHODS We queried 9,424 continuous-flow LVAD recipients who received LVADs from April 2008 to June of 2015 in the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS). The pre-operative LVIDD * flow configuration interaction term was tested in multivariable models to determine its relationship to adverse events. RESULTS The pre-operative LVIDD * flow configuration interaction was a significant predictor of device thrombosis. As the LVIDD increased, the risk of AC device thrombosis increased compared with CC devices (p = 0.0099). At 7.0 cm, the hazard ratio (HR) for AC device thrombosis compared with the CC device was 1.61 (95% confidence interval [CI], 1.17-2.22; p = 0.004) and continued to rise as the LVIDD increased. The LVIDD * flow configuration interaction did not predict stroke, gastrointestinal bleeding, or patient survival. In multivariable models, the hazard of stroke was higher with the CC device regardless of the LVIDD (HR, 1.96; 95% CI, 1.64-2.36; p < 0.0001). Adjusted analyses showed LVAD implantation into a larger left ventricle was associated with a lower risk of death (HR, 0.90; 95% CI, 0.85-0.95; p = 0.0004) per centimeter of LVIDD. CONCLUSIONS Our study suggests that the pre-operative LVIDD, flow configuration, and interaction terms should be considered individually when choosing the appropriate LVAD to mitigate the rates of device thrombosis, stroke, and death.

Percutaneous Driveline Fracture Following Implantation of the HeartMate II Left Ventricular Assist Device: How Durable is Driveline Repair?

Durability of the percutaneous driveline is critical for the optimal long-term support of patients after left ventricular assist device (LVAD) implantation. There are no data specifically examining the durability of major repair for driveline fracture or its effect on patient outcomes. Between May 2004 and August 2014, 560 patients underwent implantation with the HeartMate II LVAD at our respective institutions. Median age at LVAD implantation was 59 years (range 18–82 years) and 465 patients (83%) were male. The indication for LVAD implantation was bridge to transplant (BTT) in 296 patients (53%). Follow-up was available in all early survivors (n = 519) for a total of 940 patient-years of support (median 1.1 years, max. 10.4 years). Percutaneous driveline fracture was identified in 17 patients (3.2%), eight of whom were implanted as destination therapy. Median time from LVAD implantation to driveline fracture was 1.3 years (range, from 4 months to 3.8 years). Two of these patients underwent device exchange as primary treatment, while 15/17 (88%) underwent repair. Three of these 15 patients required a driveline reintervention, including device exchange (n = 1), rerepair (n = 1), and ungrounded cables (n = 1). Median time of support after driveline repair was 10 months (range from 3 months to 5.4 years). There were no late deaths after driveline repair during the follow-up period with 14/15 patients (93%) active on support and one having undergone transplant. The incidence of percutaneous driveline fracture after HeartMate II LVAD implantation is low (3.9%). The majority of driveline repairs for driveline fracture are durable with reintervention required in a minority of patients. There was no adverse effect of driveline fracture requiring repair on late outcome noted in this series with all patients either active on support or transplanted at last follow-up.