Background: Several studies of simultaneous pulmonary capillary wedge pressure (PCWP) and direct left atrial pressure (LAP) in cardiac surgical patients have shown good correlation between these measurements. However, discrepancy between PCWP and LAP have been reported after cardiopulmonary bypass (CPB). This study examined the relationship between PCWP and LAP after CPB in surgical patients undergoing mitral valve replacement and undergoing myocardial revascularization. Methods: Thirty-three patients undergoing myocardial revascularization (group 1) and fifty-two patients undergoing mitral valve replacement (group 2) were examined. Group 2 was divided into two subgroups by preoperative pulmonary hypertension (mean pulmonary artery pressure more than 30mmHg), without pulmonary hypertension (group 2a) and with pulmonary hypertension (group 2b). Simultaneous measurements of pulmonary artery diastolic pressure (PADP), PCWP and LAP were obtained after cardiopulmonary bypass and after sternal closure. Results: The correlation between PCWP and LAP showed statistically significant in all groups. Good correlation was seen in group I (r=0.72, p<0.01) and in group 2a (r=0.65, p<0.01). However, the correlation was not good in group 2b (r=0.53, p<0.01). There was no significant difference between PCWP and LAP in group I and 2a. However, discrepancy between PCWP and LAP was showed in group 2b. There was statistical difference between PADP and LAP in all groups. The correlation coefficient between PADP and LAP was significant in all groups but the correlation was not good. Conclusions: The PCWP was shown to be a reliable index of LAP in patients without pulmonary hypertension following CPB. However, the PCWP measurement was not accurate in assessing LAP in patients who had pulmonary hypertension. Therefore, use of LAP may lead to accurate ventricular function curve and treatment. (Korean J Anesthesiol 1997; 33: 833∼838)