Abstract
AIMS: Heart failure with preserved ejection fraction (HFpEF) poses significant diagnostic, prognostic and therapeutic challenges, with high morbidity and mortality rates. Currently, there are limited predictors of outcomes in HFpEF patients. Circulating oxidized polyunsaturated fatty acyl lipids, or oxylipins, are known to initiate and resolve inflammation in cardiovascular diseases. However, their ability to predict mortality in HFpEF has not been established. We hypothesize that a panel of oxylipins can predict and stratify mortality risk in HFpEF patients. METHODS AND RESULTS: Venous and arterial blood samples were collected during right heart catheterization from 90 HFpEF patients at a single institution. Patients were followed for 5 years to determine morbidity and mortality rates. We measured 143 arterial and 143 venous oxylipins in all study participants using liquid chromatography-mass spectrometry. Volcano plots were used to visualize differences in oxylipins between survived and deceased groups. Receiver operator characteristic (ROC) curves were used to determine optimal biomarker cut-points, and the relationship between the most significant oxylipins and mortality was assessed with Kaplan-Meier (KM) curves. HFpEF patients with 5-year mortality had increased age, decreased body mass index, decreased diastolic blood pressure and worse renal function at baseline. They also had more severe pulmonary hypertension (PH) and right heart dysfunction. Volcano plot analysis revealed that arterial oxylipin 15-keto prostaglandin F2a (PGF2a) was significantly associated with 5-year mortality. ROC curve analysis identified an optimal cut-point for 15-keto PGF2a, and participants with elevated arterial 15-keto PGF2a had significantly increased 5-year mortality on KM curves. Multivariable adjusted analysis identified 15-keto PGF2a as a significant predictor of 5-year mortality (OR 1.82; CI 1.03, 3.5). CONCLUSIONS: In this cohort of patients with HFpEF, arterial 15-keto PGF2a, a stable metabolite of PGF2a, significantly predicted 5-year mortality.