Background— Prominent features of myocardial remodeling in heart failure with preserved ejection fraction (HFPEF) are high cardiomyocyte resting tension (F _passive ) and cardiomyocyte hypertrophy. In experimental models, both reacted favorably to raised protein kinase G (PKG) activity. The present study assessed myocardial PKG activity, its downstream effects on cardiomyocyte F _passive and cardiomyocyte diameter, and its upstream control by cyclic guanosine monophosphate (cGMP), nitrosative/oxidative stress, and brain natriuretic peptide (BNP). To discern altered control of myocardial remodeling by PKG, HFPEF was compared with aortic stenosis and HF with reduced EF (HFREF). Methods and Results— Patients with HFPEF (n=36), AS (n=67), and HFREF (n=43) were free of coronary artery disease. More HFPEF patients were obese ( P <0.05) or had diabetes mellitus ( P <0.05). Left ventricular myocardial biopsies were procured transvascularly in HFPEF and HFREF and perioperatively in aortic stenosis. F _passive was measured in cardiomyocytes before and after PKG administration. Myocardial homogenates were used for assessment of PKG activity, cGMP concentration, proBNP-108 expression, and nitrotyrosine expression, a measure of nitrosative/oxidative stress. Additional quantitative immunohistochemical analysis was performed for PKG activity and nitrotyrosine expression. Lower PKG activity in HFPEF than in aortic stenosis ( P <0.01) or HFREF ( P <0.001) was associated with higher cardiomyocyte F _passive ( P <0.001) and related to lower cGMP concentration ( P <0.001) and higher nitrosative/oxidative stress ( P <0.05). Higher F _passive in HFPEF was corrected by in vitro PKG administration. Conclusions— Low myocardial PKG activity in HFPEF was associated with raised cardiomyocyte F _passive and was related to increased myocardial nitrosative/oxidative stress. The latter was probably induced by the high prevalence in HFPEF of metabolic comorbidities. Correction of myocardial PKG activity could be a target for specific HFPEF treatment.