Prediction of all-cause mortality after liver transplantation using left ventricular systolic and diastolic function assessment

• Published in: PloS one Vol. 14; no. 1; p. e0209100
• Main Authors: Moon, Young-Jin, Kim, Jung-Won, Bang, Yun-Sic, Lim, Young Su, Ki, Yumin, Sang, Bo-Hyun
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 25-01-2019; Public Library of Science (PLoS)
• Subjects: Adolescent; Adult; Anesthesiology; Biology and Life Sciences; Cardiomyopathy; Causes of; Confidence intervals; Diastole - physiology; Doppler effect; Echocardiography; Electronic health records; Heart failure; Heart Failure - mortality; Heart Failure - surgery; Hospitals; Humans; Inflow; Laboratories; Left ventricular function; Liver; Liver Transplantation; Liver transplants; Medical records; Medicine; Medicine and Health Sciences; Middle Aged; Mortality; Multivariate Analysis; Pain; Patient outcomes; Patients; Physiological aspects; Proportional Hazards Models; Prospective Studies; Research and Analysis Methods; Retrospective Studies; South Korea; Stroke; Stroke volume; Studies; Survival; Systole - physiology; Transplantation; Transplants & implants; Velocity; Ventricle; Ventricular Function, Left - physiology; Young Adult; South Korea
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0209100

Although pretransplant cardiac dysfunction is considered a major predictor of poor outcomes after liver transplantation (LT), the ability of left ventricular (LV) systolic/diastolic function (LVSF/LVDF), together or individually, to predict mortality after LT is poorly characterized. We retrospectively evaluated pretransplant clinical and Doppler echocardiographic data of 839 consecutive LT recipients from 2009 to 2012 aged 18-60 years. The primary endpoint was all-cause mortality at 4 years. The overall survival rate was 91.2%. In multivariate Cox analysis, reduced LV ejection fraction (LVEF, P = 0.014) and decreased transmitral E/A ratio(P = 0.022) remained significant prognosticators. In LVSF analysis, patients with LVEF≤60% (quartile [Q]1) had higher mortality than those with LVEF>60% (hazard ratio = 1.90, 95% confidence interval = 1.15-3.15, P = 0.012). In LVDF analysis, patients with an E/A ratio<0.9(Q1) had a 2.19-fold higher risk of death (95% confidence interval = 1.11-4.32, P = 0.024) than those with an E/A ratio>1.4(Q4). In combined LVDF and LVSF analysis, patients with an E/A ratio<0.9 and LVEF≤60% had poorer survival outcomes than patients with an E/A ratio≥0.9 and LVEF>60% (79.5% versus 93.3%, P = 0.001). Patients with an early mitral inflow velocity/annular velocity (E/e' ratio)>11.5(Q4) and LV stroke volume index (LVSVI)<33mL/m2(Q1) showed worse survival than those with an E/e' ratio≤11.5 and LVSVI ≥33mL/m2(78.4% versus 92.2%, P = 0.003). A combination of LVSF and LVDF is a better predictor of survival than LVSF or LVDF alone.