The causes of reduced levels of omega-3 and omega-6 highly unsaturated fatty acids ("HUFA deficiency") in heart failure remain unresolved. HUFA profiles were examined in the serum of 331 patients with failing versus nonfailing heart disease. Arachidonic acid was positively correlated (P < 0.001) with eicosapentaenoic acid (EPA) (r = 0.40) and docosahexaenoic acid (DHA) (r = 0.53) and negatively with palmitic (r = 0.42), palmitoleic (r = 0.38), and oleic acid (r = 0.48). Delta-5 desaturase activity was reduced (P < 0.01) in heart failure patients with low ejection fraction, dilatation, increased wall stress, and reduced heart rate variability (SDNN). In these patients, the reduced (P < 0.01) HUFA and increased palmitic (P < 0.01) and oleic acid (P = 0.05) arose from separate influences involving reduced cardiac contractility (arachidonic acid and palmitic acid predicted by ejection fraction) and chamber dilatation (DHA and oleic acid predicted by end-diastolic diameter). A low DHA (0.2%-0.9% versus 1.4%-3.1%) was associated (P < 0.025) with atrial dilatation (44 ± 8 mm versus 40 ± 8 mm). Equidirectional but less pronounced effects on HUFA were induced by sympathetic activation and (or) insulin resistance (fat and sugar fed to deoxycorticosterone acetate (DOCA)-salt rats) but not by compensated cardiac overload alone (DOCA-salt or aortic constriction), or reduced fatty acid oxidation (CPT-1 inhibition). Based on administration of omega-3 HUFA (OMACOR), dilatation is identified as a target for 1-2 g omega-3 HUFA·day(-1). Interventions for reduced arachidonic acid remain to be explored.