Impaired myocardial bioenergetics is a hallmark of many cardiac diseases. There is a need of a simple and reproducible method of assessment of mitochondrial function from small human myocardial tissue samples. In this study we adopted high-resolution respirometry to homogenates of fresh human cardiac muscle and compare it with isolated mitochondria. We used atria resected during cardiac surgery (n = 18) and atria and left ventricles from brain-dead organ donors (n = 12). The protocol we developed consisting of two-step homogenization and exposure of 2.5% homogenate in a respirometer to sequential addition of 2.5 mM malate, 15 mM glutamate, 2.5 mM ADP, 10 μM cytochrome c, 10 mM succinate, 2.5 μM oligomycin, 1.5 μM FCCP, 3.5 μM rotenone, 4 μM antimycin and 1 mM KCN or 100 mM Sodium Azide. We found a linear dependency of oxygen consumption on oxygen concentration. This technique requires < 20 mg of myocardium and the preparation of the sample takes <20 min. Mitochondria in the homogenate, as compared to subsarcolemmal and interfibrillar isolated mitochondria, have comparable or better preserved integrity of outer mitochondrial membrane (increase of respiration after addition of cytochrome c is up to 11.7±1.8% vs. 15.7±3.1%, p˂0.05 and 11.7±3.5%, p = 0.99, resp.) and better efficiency of oxidative phosphorylation (Respiratory Control Ratio = 3.65±0.5 vs. 3.04±0.27, p˂0.01 and 2.65±0.17, p˂0.0001, resp.). Results are reproducible with coefficient of variation between two duplicate measurements ≤8% for all indices. We found that whereas atrial myocardium contains less mitochondria than the ventricle, atrial bioenergetic profiles are comparable to left ventricle. In conclusion, high resolution respirometry has been adapted to homogenates of human cardiac muscle and shown to be reliable and reproducible.