Aerobic ATP synthesis via oxidative phosphorylation causes a proportional production of respiratory water. Thus the amount of respiratory water produced at a given time should be a reliable measure of the current ATP demand of the mammalian myocardium. Respiratory water from isolated rabbit hearts was labeled by using the stable oxygen isotope 18O. The hearts were perfused according to the method of Langendorff (O. Langendorff. Pfluegers Arch. 61: 291-332, 1895) with 18O2-equilibrated Krebs-Henseleit solution. Control hearts were exclusively perfused with carbogen-equilibrated Krebs-Henseleit solution. Myocardial tissue was then lyophilized; the extracted water and samples from the coronary venous effluent were converted to CO2 by using the guanidine hydrochloride technique. The delta 18O values within the CO2 samples were determined by mass spectrometry and related to the standard mean ocean water (SMOW) scale. Compared with control hearts, the 18O-labeled hearts exhibited a significant increase of delta 18O values from tissue water (-47.50 +/- 0.64 vs. -40.35 +/- 2.05% SMOW; P < 0.05). The values were also significantly increased in the coronary venous effluent after a perfusion time of only 50 s (-47.50 +/- 0.64 vs. -43.66 +/- 0.91% SMOW; P < 0.05). Thus this first adaptation of the guanidine hydrochloride technique on microliter samples of myocardial tissue water and coronary venous effluent demonstrates that this method can be used to evaluate both respiratory activity and the kinetics of cardiac metabolic processes.