Therapeutic and Prognostic Significance of Arachidonic Acid in Heart Failure

Background: Accurate prediction of death is an unmet need in patients with acute decompensated heart failure (HF). Arachidonic acid (AA) metabolites play an important role in the multiple pathophysiological processes. We aimed to develop an AA score to accurately predict mortality in patients with acute decompensated HF and explore the causal relationship between the AA predictors and HF. Methods: The serum AA metabolites was measured in patients with acute decompensated HF (discovery cohort n=419; validation cohort n=386) by mass spectroscopy. We assessed the prognostic importance of AA metabolites for 1-year death using Cox regression and machine learning approaches. A machine learning-based AA score for predicting 1-year death was created and validated. We explored the mechanisms using transcriptome and functional experiments in a mouse model of early ischemic cardiomyopathy. Results: Among the 27 AA metabolites, elevated 14,15-DHET/14,15-EET ratio was the strongest predictor of 1-year death (hazard ratio, 2.10, P=3.1×10^-6). Machine learning-based AA score using a combination of the 14,15-DHET/14,15-EET ratio, 14,15-DHET, PGD2, and 9-HETE performed best (area under the curve [AUC]: 0.85). The machine learning-based AA score provided incremental information to predict mortality beyond BNP (B-type natriuretic peptide; ΔAUC: 0.19), clinical score (ΔAUC: 0.09), and preexisting Acute Decompensated Heart Failure National Registry, Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure, and Get With The Guidelines Heart Failure scores (ΔAUC: 0.17, 0.17, 0.15, respectively). In the validation cohort, the AA score accurately predicted mortality (AUC:0.81). False-negative and false-positive findings, as classified by the BNP threshold, were correctly reclassified by the AA score (46.2% of false-negative and 84.5% of false-positive). In a murine model, the expression and enzymatic activity of sEH (soluble epoxide hydrolase) increased after myocardial infarction. Genetic deletion of sEH improved HF and the blockade of 14,15-EET abolished this cardioprotection. We mechanistically revealed the beneficial effect of 14,15-EET by impairing the activation of monocytes/macrophages. Conclusions: Our studies propose that the AA score predicts death in patients with acute decompensated HF and inhibiting sEH serves as a therapeutic target for treating HF. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT04108182.