Cardiovascular complications during or after transcatheter aortic valve replacement (TAVR) are associated with extremely high mortality, but extracorporeal membrane oxygenation (ECMO) can be used as procedural rescue option to improve outcomes when patients experience respiratory or cardiac arrest. From 2012 to 2014, 230 patients underwent TAVR and 10 patients (4.3%) required emergent venous-arterial ECMO support. Mean age was 83 years, median Society of Thoracic Surgeons (STS) score was 15, and mean aortic gradient was 45 mm Hg. Median left ventricular ejection fraction was 35%. Access for most ECMOs was femoral; two patients required central arterial and femoral venous access. Extracorporeal membrane oxygenation was initiated in response to hemodynamic collapse due to perforation of left ventricle (n = 2), aortic root rupture (n = 1), moderate-to-severe aortic insufficiency (n = 1), left main impingement (n = 1), valve embolization (n = 1), severe hypotension and cardiac arrest after prolonged rapid pacing sequence (n = 1), ventricular fibrillation (n = 2), and ventricular tachycardia (n = 1). Median time of ECMO support was 87 minutes. There were three hospital deaths. Post-TAVR mean aortic gradient was 8 mm Hg and median hospital stay was 19 days. Additional procedures included valve-in-valve placement (n = 1), percutaneous coronary intervention (n = 1), surgical LV repair (n = 2), surgical valve replacement (n = 1), aortic root rupture repair, and coronary bypass grafting (n = 1). Extracorporeal membrane oxygenation is rescue therapy for hemodynamic instable patients who develop TAVR-related cardiac complications.

Abnormalities in markers of liver injury after venoarterial extracorporeal membrane oxygenation (VA-ECMO) initiation are of unclear distribution and clinical significance. This study included all consecutive adult patients from a single institution who underwent VA-ECMO cannulation between May 2012 and September 2016 and had liver function panels drawn during their admission (n = 223). Data points include: age, sex, body mass index, diagnosis, duration of ECMO cannulation, duration of hospitalization, pre-ECMO cardiac arrest, central nervous system (CNS) injury, the presence of chronic kidney disease or acute renal failure, renal replacement therapy utilization, lactate levels, duration of pre-ECMO intubation, admission and peak bilirubin/aspartate aminotransferase (AST)/alanine aminotransferase (ALT)/alkaline phosphatase (ALP) levels, and time to peak bilirubin/AST/ALT/ALP in relation to cannulation. Multivariate Poisson regression analyses were performed to determine associations with mortality. In-hospital mortality was 66%. Serum bilirubin elevation appeared to significantly correlate continuously with mortality. Other markers of liver injury were not significant in final multivariate models. As a univariate factor, no patient survived with a total serum bilirubin greater than 30 mg/dl, and specificity for 90% mortality was crossed at 11 mg/dl. Mortality was also significantly associated with the presence of CNS injury and elevation of lactic acid levels. Postcannulation liver injury is significantly associated with increased mortality and total serum bilirubin appears to be a biomarker of considerable clinical significance.

No abstract