ObjectiveTo characterize the obese heart failure with preserved ejection fraction (HFpEF) phenotype in a multicenter cohort. Patients and MethodsThis was a secondary analysis of the randomized clinical trial RELAX (Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure with Preserved Ejection Fraction) performed between October 1, 2008, and February 1, 2012. Patients with HFpEF were classified by body mass index (BMI) as obese (BMI≥35 kg/m2) and nonobese (BMI<30 kg/m2) for comparison. ResultsObese patients with HFpEF (n=81) were younger (median age, 64 [interquartile range (IQR), 67-79] years vs 73 [IQR, 56-70] years; P<.001) but had greater peripheral edema (31% [25] vs 9% [6]; P<.001), more orthopnea (76% [56] vs 53% [35]; P=.005), worse New York Heart Association class (P=.006), and more impaired quality of life (P<.001) as compared with nonobese patients with HFpEF (n=70). Despite more severe signs and symptoms, obese patients with HFpEF had lower N-terminal pro B-type natriuretic peptide level (median, 481 [IQR, 176-1183] pg/mL vs 825 [IQR, 380-1679] pg/mL [to convert to pmol/L, multiply by 0.118]; P=.007) and lower left atrial volume index (median, 38 [IQR, 31-47] mL/m2 vs 54 [IQR, 41-63] mL/m2; P<.001). Serum C-reactive protein (median, 5.0 [IQR, 2.4-9.9] mg/dL vs 2.7 [IQR, 1.6-5.4] mg/dL [to convert to mg/L, multiply by 10−3]; P<.001) and uric acid (median, 7.8 [IQR, 6.1-8.7] mg/dL vs 6.8 [IQR, 5.5-8.3] mg/dL; P=.03) levels were higher in obese HFpEF, indicating greater systemic inflammation, than in nonobese HFpEF. Peak oxygen consumption was impaired in obese HFpEF (median, 11.1 [IQR, 9.6-14.4] mL/kg per minute vs 13.1 [IQR, 11.3-14.7] mL/kg per minute; P=.008), as was submaximal exercise capacity (6-minute walk distance, 272 [IQR, 200-332] m vs 355 [IQR, 290-415] m; P<.0001). ConclusionObese HFpEF is associated with decreased quality of life, worse symptoms of heart failure, greater systemic inflammation, worse exercise capacity, and higher metabolic cost of exertion as compared with nonobese HFpEF. Further study is required to understand the pathophysiology and potential distinct treatments for patients with the obese phenotype of HFpEF. Trial Registrationclinicaltrials.gov Identifier: NCT00763867