Fulminant Myocarditis Supported with Extracorporeal Membrane Oxygenation: An Individual Patient Data Meta-Analysis on 30-day Mortality and Impact of Prognostic Factors.

Purpose: Fulminant myocarditis (FM) is a rare disease that may progress rapidly to refractory cardiogenic shock requiring temporary mechanical circulatory support. Venoarterial extracorporeal membrane oxygenation (V-A ECMO) has been used to provide biventricular support in these patients, however, studies assessing its benefit have demonstrated conflicting results, likely due to small sample size. We conducted an individual patient data (IPD) meta-analysis to evaluate 30-day mortality and associated prognostic factors after V-A ECMO in patients with FM. Methods: A systematic search of 7 relevant databases was performed for studies including adult (> 18 years old) FM patients who required V-A ECMO and reported on mortality after the initiation of V-A ECMO. Authors of each identified study were invited to participate in the IPD meta-analysis. The primary outcome was 30-day mortality after V-A ECMO initiation in patients with FM supported with V-A ECMO. A mixed effect multivariable model with a random intercept to account for clustering of patients was used to examine independent prognostic factors associated with 30-day mortality. Results: We combined IPD from 14 centers including 307 patients, mean age was 45+17 years, and 54% were females. FM was verified by biopsy in 52%, most commonly being lymphocytic (65%). The most common cannulation site for V-A ECMO was peripheral (76%), 32% had a cardiac arrest prior to V-A ECMO. The proportion of patients on mechanical ventilation was 87%, and left ventricular (LV) venting was used in 50%, most commonly intra-aortic balloon pump (56%). Thirty-day mortality was 37% (95%CI 32-43). Patients who died were signifcantly older, had higher levels of cardiac enzymes, a higher frequency of cardiac arrest prior to V-A ECMO and were treated with central V-A ECMO. By multivariable analsysis, independent factors associated with 30-day mortality after V-A ECMO were older age (OR 1.03, 1.01-1.04), cardiac arrest (OR 2.80, 1.63-4.90), and central V-A ECMO cannulation (OR 2.06, 1.06-4.01), whereas mechanical ventilation and LV-venting were not. Conclusion: In patients with FM requiring V-A ECMO, age, cardiac arrest prior to V-A ECMO, and central cannuation were identified as independent risk factors of 30-day mortality.