An attempt was made to estimate noninvasively cardiac index (CI) and pulmonary capillary wedge pressure (PCWP) in patients with acute myocardial infarction (AMI) using the theory of quantification, a kind of multivariate analysis. Eighty seven patients with AMI in whom hemodynamic measurements had been made were studied. Thirteen parameters (age, sex, height, body weight, systolic, diastolic and mean BP, rales, S3 or S4, pretibial edema, urine protein, leukocyte count, S-LDH) were selected for the estimation of CI. Six parameters (HR, rates, pretibial edema, S-GOT, S-LDH, ESR) were chosen for the estimation of PCWP. The estimated CI was significantly correlated with the data obtained by thermodilution with a correlation coefficient of 0.749 (p<0.01). The estimated PCWP was also significantly correlated with the data obtained by Swan-Ganz catheter with a correlation coefficient of 0.762 (p<0.01). Among 13 parameters, age, sex, rates, pretibial edema and S-LDH contributed greatly to the estimation of CI. Rales and S-GOT influenced largely for the estimation of PCWP.When the formulae were applied for subsequent 50 patients with AMI (external samples), the correlation coefficient between estimated and measured CI was 0.453 (p<0.01). The correlation coefficient between estimated and measured PCWP was 0.640 (p<0.01). The difference exceeding 11/min/M' between estimated and measured CI was observed in 8 cases and the difference exceeding 10 mmHg between estimated and measured PCWP was observed in 6 cases. Deviations of estimated values from actual measurements appeared to occur under the following situations : complete A-V block, immediately after ventricular fibrillation, exceeding 72 h after the attack.The results indicate that the formulae are useful clinically to estimate CI and PCWP in patients with AMI.