Multicenter evaluation of left ventricular assist device implantation with or without ECMO bridge in cardiogenic shock.

BACKGROUND: The efficacy of extracorporeal membrane oxygenation (ECMO) as a bridge to left ventricular assist device (LVAD) remains unclear, and recipients of the more contemporary HeartMate 3 (HM3) LVAD are not well represented in previous studies. We therefore undertook a multicenter, retrospective study of this population. METHODS AND RESULTS: INTERMACS 1 LVAD recipients from five U.S. centers were included. In-hospital and one-year outcomes were recorded. The primary outcome was the overall mortality hazard comparing ECMO versus non-ECMO patients by propensity-weighted survival analysis. Secondary outcomes included survival by LVAD type, as well as postoperative and one-year outcomes. One hundred and twenty-seven patients were included; 24 received ECMO as a bridge to LVAD. Mortality was higher in patients bridged with ECMO in the primary analysis (HR 3.22 [95%CI 1.06-9.77], p = 0.039). Right ventricular assist device was more common in the ECMO group (ECMO: 54.2% vs non-ECMO: 11.7%, p < 0.001). Ischemic stroke was higher at one year in the ECMO group (ECMO: 25.0% vs non-ECMO: 4.9%, p = 0.006). Among the study cohort, one-year mortality was lower in HM3 than in HeartMate II (HMII) or HeartWare HVAD (10.5% vs 46.9% vs 31.6%, respectively; p < 0.001) recipients. Pump thrombosis at one year was lower in HM3 than in HMII or HVAD (1.8% vs 16.1% vs 16.2%, respectively; p = 0.026) recipients. CONCLUSIONS: Higher mortality was observed with ECMO as a bridge to LVAD, likely due to higher acuity illness, yet acceptable one-year survival was seen compared with historical rates. The receipt of the HM3 was associated with improved survival compared with older generation devices.

Survival and Outcomes After Cardiac Arrest With VA-ECMO Rescue Therapy.

BACKGROUND: Extracorporeal membrane oxygenation (ECMO) use in patients with cardiac arrest is increasing. Utilization remains variable between centers using ECMO as a rescue therapy or early protocolized extracorporeal cardiopulmonary resuscitation. METHODS: Single-center, retrospective evaluation of cardiac arrest with cardiopulmonary resuscitation and rescue ECMO support from 2011 through 2019. Study objectives included survival, non-neurologic, and neurologic outcomes; validation of the SAVE and modified SAVE (mSAVE) scores for survival and favorable neurologic outcome; and predictive factor identification in cardiac arrest with ECMO rescue therapy. RESULTS: Eighty-nine patients were included. In-hospital survival was 38.2% and median CPC score was 2. Survivors had lower BMI (27.9 ±â€Š4.2 kg/m2 vs. 32.3 ±â€Š7.5 kg/m2, P = 0.003), less obesity (BMI ≥ 30 kg/m2) (26.5% vs. 49.1%, P = 0.035), shorter CPR duration (35.5 ±â€Š31.7 m vs. 58.0 ±â€Š49.5 m, P = 0.019), more tracheostomy (38.2% vs. 7.3%, P < 0.001), and less renal replacement therapy (RRT) (17.6% vs. 38.2%, P = 0.031). Patients with a favorable neurologic outcome had lower body weight (86.2 ±â€Š17.9 kg vs. 98.1 ±â€Š19.4 kg, P = 0.010), lower BMI (28.1 ±â€Š4.5 kg/m2 vs. 33.9 ±â€Š7.9 kg/m2, P < 0.001), and less obesity (29.7% vs. 56.3%, P = 0.026). mSAVE score predicted in-hospital survival (OR 1.11; 95%CI 1.03-1.19; P = 0.004) and favorable neurologic outcome (OR 1.11; 1.03-1.20; P = 0.009). Multivariate analysis for in-hospital survival included mSAVE, BMI, CPR-time, tracheostomy, and RRT (c-statistic: 0.864). Favorable neurologic outcome included mSAVE and BMI (c-statistic: 0.805). CONCLUSIONS: mSAVE, BMI, RRT, and tracheostomy are predictors of in-hospital survival and mSAVE and BMI are predictors of favorable neurologic outcome in cardiac arrest with ECMO rescue therapy.

Left Ventricular Assist Device Multialarm Emergency: A High-Fidelity Simulation Case for Emergency Medicine Residents.

Introduction: As left ventricular assist devices (LVADs) become more prevalent in the treatment of patients with end-stage heart failure, emergency physicians must become experts in the management and resuscitation of patients with LVADs. As with other high-acuity, low-occurrence scenarios, managing the unstable LVAD patient makes for an ideal topic for simulation-based resident education. Methods: By incorporating a high-fidelity HeartMate 3 LVAD task trainer, our program developed and executed a novel LVAD simulation activity for our emergency medicine resident physicians. In the scenario, a 65-year-old male with recent LVAD placement arrived at a community hospital with undifferentiated hypotension. Various device alarms activated during the scenario and required intervention. Ultimately, the patient was found to be in septic/hypovolemic shock and only survived with appropriate resuscitation. We implemented a postscenario survey to assess the effectiveness of the simulation activity and administered it to 27 residents. Results: Content and delivery of our simulation were found to be effective; all survey questions regarding content and delivery obtained a mean score of 4.5 or greater on a 5-point Likert scale. Residents reported an overall high level of confidence in achieving most of the skill-based learning objectives (most scores > 4.1). The two objectives with the lowest confidence ratings were troubleshooting an LVAD and its various alarms (3.8) and demonstrating the ability to assess an LVAD patient (3.9). Discussion: Our LVAD simulation activity was successful and also revealed several potential areas for future research and simulation improvement.