The hypothesis tested was that myocardial venous O2 saturation (SvO2) heterogeneity, a measure of microregional O2 supply/consumption balance, would increase under hypoxic and CO-hypoxia conditions. Since we are able to determine both O2 supply and the O2 supply/consumption ratio, we could also determine whether regional myocardial O2 consumption was heterogeneous. Twenty open-chest anesthetized dogs were studied under control and four hypoxic conditions, hypoxic hypoxia induced by ventilation with either an 8% O2 (SaO2 = 56%) or a 6% O2 (SaO2 = 40%) gas mixture for 20 min, or CO hypoxia induced by ventilation with a 1% CO gas mixture for either 7 min (SaO2 = 67%) or 20 min (SaO2 = 40%). Regional myocardial blood flow was measured using radioactive microspheres in 40 pieces (approximately 0.5 g) of the left ventricular free wall. Arterial and venous O2 saturations were determined with a four-wavelength microspectrophotometric method. A total of 28 veins (20-100 microns) were examined to determine SvO2 for each condition within each animal. The coefficient of variation (CV = SD/mean x 100), an index of heterogeneity, was calculated for both flow and SvO2 under each condition. Flow increased with increasing severity of hypoxia but its heterogeneity did not change with hypoxic or CO hypoxia. However, SvO2 heterogeneity significantly increased with increasing severity of hypoxia. A linear regression of SvO2 CV and mean SvO2 showed a significant correlation (CV = -0.84 (mean SvO2) + 51.1, R = 0.59). All possible myocardial O2 consumptions were calculated by multiplying all of the flows and O2 extractions. In 53 subepicardial and subendocardial measurements, only 10% of the flow and O2 supply/consumption heterogeneity observations could be explained by uniform O2 consumption if our acceptance criterion was 0.06-0.1 ml O2/min/100 g, and 50% could be explained with an acceptance criterion of 0.3-0.4 ml O2/min/100 g. Therefore, there must be some regional myocardial O2 consumption heterogeneity. The increase in venous O2 saturation heterogeneity during hypoxia may be due to increased variation in regional myocardial O2 consumption or variation in the control of O2 supply/consumption coupling.