The relationship between cardiac and vascular abnormalities was studied in 68 patients with established septic shock. At time of hemodynamic evaluation, after initial resuscitation, there was no significant difference in arterial pressure, pulmonary artery pressure, cardiac filling pressures, and cardiac index between the 38 survivors of shock and the 30 patients who died of shock, but the left ventricular stroke work index and the right ventricular (RV) stroke work index were higher in survivors than in those who died (mean +/- SD: 25.0 +/- 9.1 vs 20.1 +/- 9.4 gm/m2 [p less than 0.05] and 6.6 +/- 3.6 vs 4.8 +/- 2.8 gm/m2 [p less than 0.05], respectively). Survivors had also higher thermodilution RV ejection fraction and lower RV end-diastolic volumes than had those who died (43.9% +/- 16.3% vs 31.1% +/- 13.7% [p less than 0.01] and 82 +/- 30 vs 99 +/- 31 ml/m2 [p less than 0.05], respectively). Calculated systemic vascular resistance was similar in the two groups, but vasopressors had been required in 22 (58%) of 38 survivors and 25 (83%) of 30 patients who died (p less than 0.01). Moreover, when the patients were separated into two groups according to their cardiac output, higher or lower than 3 L/min/m2, in both subgroups patients who died had lower blood pressure than had survivors. Blood lactate levels were significantly lower in survivors than in nonsurvivors (5.1 +/- 2.1 vs 8.1 +/- 4.7 mEq/L, p less than 0.01). Final data obtained before recovery of shock or death indicated that the survivors had higher arterial pressure, lower pulmonary artery pressure and right atrial pressure, higher stroke volume, and higher RV ejection fraction than had the patients who died. No survivors but all patients who died had been treated with vasopressors. These data therefore indicate that death as a result of septic shock is characterized by both myocardial depression and altered vascular tone and both are probably interrelated.