Previous studies reporting pathologic oxygen supply dependency calculated VO2 as CO x C(a-v)O2. We investigated whether pathologic oxygen supply dependency exists in septic and postoperative patients if VO2 and DO2 are assessed independently. In septic patients, VO2 was 164 +/- 31 and DO2 was 633 +/- 209 ml/min/m2. The slope (b) of the VO2-DO2 regression line VO2 = b x DO2 + a ranged from -0.10 to 0.08 (mean, 0.02 +/- 0.01, p less than 0.05) and was statistically significant in two patients (b = 0.05 and b = 0.08, p less than 0.05). In postoperative patients VO2 was 136 +/- 19 and DO2 was 481 +/- 160 ml/min/m2; b ranged from -0.07 to 0.09 (mean, 0.04 +/- 0.01, p less than 0.001) and was statistically significant in one patient (b = 0.09, p less than 0.01). The lack of a close relationship between independently measured VO2 and DO2 may indicate that septic and postoperative patients in stable hemodynamic condition have no pathologic oxygen supply dependency. Analysis of the VO2-DO2 relationship may not be useful to guide therapy or predict outcome.