AimsAbnormally low right ventricular ejection fraction (RVEF) is a predictor of poor outcomes in chronic heart failure (HF) patients with low left ventricular ejection fraction (LVEF). However, little is known about the relationship between LVEF and RVEF in these patients.Methods and resultsOf the 2707 Beta-blocker Evaluation of Survival Trial (BEST) participants with ambulatory chronic HF, New York Heart Association class IIIIV symptoms, and LVEF ≤35, 2008 patients had gated-equilibrium radionuclide angiographic data on baseline LVEF and RVEF. Patients were categorized into quartiles by LVEF <29 (n= 507), 2328 (n= 513), 1722 (n= 538), and <17 (n= 450). Logistic regression models were used to determine the association of LVEF quartiles (reference, <29) with abnormally low RVEF (<20). The prevalence of RVEF <20 for patients with LVEF quartiles <29, 2328, 1722, and <17 were 3, 6, 15, and 32, respectively. Unadjusted odds ratios [95 confidence intervals (CIs)] for RVEF <20 (vs. <20) associated with LVEF quartiles 2328, 1722, and <17 (reference, <29) were 2.18 (1.144.17; P 0.018), 6.32 (3.5411.30; P< 0.001), and 16.67 (9.4629.39; P< 0.001), respectively. Respective multivariable-adjusted odds ratios (95 CIs) were 1.82 (0.943.54; P 0.076), 4.55 (2.488.35; P< 0.001), and 10.53 (5.7019.44; P< 0.001), respectively. Heart failure symptoms and signs had unadjusted associations with low RVEF, but lacked intrinsic associations.ConclusionIn patients with advanced systolic HF, LVEF has a strong dose-dependent relationship with RVEF which is independent of other characteristics, and low LVEF is useful as a surrogate marker of abnormally low RVEF in these patients. © 2010 The Author.