OBJECTIVES: Data about inhospital outcomes in bleeding complications during extracorporeal life support (ECLS) have been poorly investigated. DESIGN: Retrospective observational study. SETTING: Patients reported in Extracorporeal Life Support Organization Registry.… Click to show full abstract
OBJECTIVES: Data about inhospital outcomes in bleeding complications during extracorporeal life support (ECLS) have been poorly investigated. DESIGN: Retrospective observational study. SETTING: Patients reported in Extracorporeal Life Support Organization Registry. PATIENTS: Data of 53.644 adult patients (greater than or equal to 18 yr old) mean age 51.4 ± 15.9 years, 33.859 (64.5%) male supported with single ECLS run between 01.01.2000 and 31.03.2020, and 19.748 cannulated for venovenous (V-V) ECLS and 30.696 for venoarterial (V-A) ECLS. INTERVENTIONS: Trends in bleeding complications, bleeding risk factors, and mortality. MEASUREMENT AND MAIN RESULTS: Bleeding complications were reported in 14.786 patients (27.6%), more often in V-A ECLS compared with V-V (30.0% vs 21.9%; p < 0.001). Hospital survival in those who developed bleeding complications was lower in both V-V ECLS (49.6% vs 66.6%; p < 0.001) and V-A ECLS (33.9 vs 44.9%; p < 0.001). Steady decrease in bleeding complications in V-V and V-A ECLS was observed over the past 20 years (coef., –1.124; p < 0.001 and –1.661; p < 0.001). No change in mortality rates was reported over time in V-V or V-A ECLS (coef., –0.147; p = 0.442 and coef., –0.195; p = 0.139). Multivariate regression revealed advanced age, ecls duration, surgical cannulation, renal replacement therapy, prone positioning as independent bleeding predictors in v-v ecls and female gender, ecls duration, pre-ecls arrest or bridge to transplant, therapeutic hypothermia, and surgical cannulation in v-a ecls. CONCLUSIONS: A steady decrease in bleeding over the last 20 years, mostly attributable to surgical and cannula-site–related bleeding has been found in this large cohort of patients receiving ECLS support. However, there is not enough data to attribute the decreasing trends in bleeding to technological refinements alone. Especially reduction in cannulation site bleeding is also due to changes in timing, patient selection, and ultrasound guided percutaneous cannulation. Other types of bleeding, such as CNS, have remained stable, and overall bleeding remains associated with a persistent increase in mortality.
               
Click one of the above tabs to view related content.