Satoshi Miyashita, Raquel Lamarche, Peter Oro, Roshan B Patel, Karthik V, Rana Abualsaud, Yuichiro Yano, Toshio Naito, W H Wilson Tang, Daniel Burkhoff, Adriano R Tonelli, Matthew T Siuba
During a median follow-up of 1.66 years, 18 patients died and 21 underwent liver transplantation. Mean CI was 5.5 ± 1.2 L/min/m2, mean MELD-Na score was 24 ± 9, and echocardiography demonstrated preserved left ventricular ejection fraction (65 ± 11%) with disproportionate left atrial enlargement relative to left ventricular volume, reflected by a mean LA/LV volume ratio of 0.84 ± 0.31. In univariable Cox regression analyses, higher LA/LV volume ratio, CI, and end-systolic elastance (Ees) were associated with increased mortality when modeled per 1-standard deviation increase, with hazard ratios of 2.13 (95% confidence interval, 1.20-3.79; p = 0.010), 1.64 (95% confidence interval, 1.12-2.42; p = 0.012), and 1.93 (95% confidence interval, 1.09-3.44; p = 0.024), respectively. In MELD-Na-adjusted Cox models, LA/LV volume ratio, CI, and Ees remained associated with mortality, with adjusted hazard ratios of 2.02 (95% confidence interval, 1.12-3.65; p = 0.019), 1.66 (95% confidence interval, 1.08-2.56; p = 0.021), and 1.92 (95% confidence interval, 1.09-3.39; p = 0.024), respectively. Similar directionality was observed in Firth penalized Cox models and Fine-Gray competing-risk models treating liver transplantation as a competing event.
INTRODUCTION: Cirrhosis is a recognized cause of high-output heart failure (HOHF), yet the prognostic relevance of paired invasive hemodynamic and echocardiographic remodeling profiles in this population remains incompletely defined.
METHODS: We retrospectively analyzed 41 adults with cirrhosis-associated HOHF who underwent right heart catheterization between 2015 and 2023. HOHF was defined as thermodilution cardiac index (CI) ≥ 4.0 L/min/m2 with elevated filling pressures. The primary outcome was all-cause mortality.
RESULTS: During a median follow-up of 1.66 years, 18 patients died and 21 underwent liver transplantation. Mean CI was 5.5 ± 1.2 L/min/m2, mean MELD-Na score was 24 ± 9, and echocardiography demonstrated preserved left ventricular ejection fraction (65 ± 11%) with disproportionate left atrial enlargement relative to left ventricular volume, reflected by a mean LA/LV volume ratio of 0.84 ± 0.31. In univariable Cox regression analyses, higher LA/LV volume ratio, CI, and end-systolic elastance (Ees) were associated with increased mortality when modeled per 1-standard deviation increase, with hazard ratios of 2.13 (95% confidence interval, 1.20-3.79; p = 0.010), 1.64 (95% confidence interval, 1.12-2.42; p = 0.012), and 1.93 (95% confidence interval, 1.09-3.44; p = 0.024), respectively. In MELD-Na-adjusted Cox models, LA/LV volume ratio, CI, and Ees remained associated with mortality, with adjusted hazard ratios of 2.02 (95% confidence interval, 1.12-3.65; p = 0.019), 1.66 (95% confidence interval, 1.08-2.56; p = 0.021), and 1.92 (95% confidence interval, 1.09-3.39; p = 0.024), respectively. Similar directionality was observed in Firth penalized Cox models and Fine-Gray competing-risk models treating liver transplantation as a competing event.
DISCUSSION: In patients with cirrhosis-associated HOHF, disproportionate left atrial remodeling, elevated CI, and higher Ees were associated with mortality. These findings support further evaluation of integrated invasive hemodynamic and quantitative echocardiographic assessment for characterizing risk-related phenotypes in cirrhosis-associated HOHF.