Abstract 13702: Serum Albumin Predicts Mortality in Patients With Pulmonary Arterial Hypertension

Abstract

Introduction: Low serum albumin predicts adverse outcomes in heart failure, reflecting poor nutritional status, liver dysfunction and systemic inflammation. The predictive value of albumin for mortality is not well-described in pulmonary arterial hypertension (PAH). Hypothesis: Low serum albumin will significantly predict increased mortality in patients with PAH. Methods: We constructed a database of 273 WHO group 1 PAH patients followed at our center from 03/2001 through 08/2008, including echocardiography, invasive hemodynamics, laboratory data and mortality outcomes. Serum albumin levels were available in 163 patients. The association between albumin and mortality was assessed using Cox regression and Kaplan-Meier analysis. Results: Patients were followed for a mean of 3.6 ± 2.6 years. On univariate analysis, serum albumin significantly predicted mortality with HR = 0.746 (p = 0.01, 95% CI 0.598-0.932) per 1 g/dl increase. On multivariate analysis, serum albumin significantly predicted mortality with HR = 0.491 (p = 0.009, 95% CI 0.288-0.835) per 1 g/dl increase when adjusted for age, gender, NYHA functional class, mean right atrial pressure, Fick cardiac output, connective tissue disease, right ventricular function, right atrial systolic area and serum creatinine. Patients with serum albumin below the 25th percentile (<=3.3 g/dl) had significantly higher mortality by Kaplan-Meier curve analysis (p = 0.01 by log-rank test) despite similar clinical characteristics, when compared with patients with higher serum albumin. Conclusion: Higher serum albumin is an independent predictor of decreased mortality in PAH patients when adjusted for clinical, echocardiographic and hemodynamic variables. Patients with low serum albumin values had significantly higher mortality despite similar characteristics.