Previous reports cite optimization of O2 delivery (DO2) to 660 mL/min/m2, O2 consumption (VO2) to 170 mL/min/m2, and cardiac index (CI) of 4.5 L/min as predicting survival. We prospectively evaluated 76 consecutive patients with multiple trauma admitted directly to the ICU from the operating room or emergency department. Patients had serum lactate levels and oxygen transport measured on ICU admission and at 8, 16, 24, 36, and 48 hours. Patients were analyzed with respect to survival (S) versus nonsurvival (NS), lactate clearance to normal (< or = 2 mmol/L) by 24 and 48 hours, hemodynamic optimization as defined above, as well as Injury Severity Score (ISS), ICU stay (LOS), and admission blood pressure. All patients achieved non-flow-dependent VO2. There was no difference in CI, DO2, VO2, or ISS when S was compared with NS. All 27 patients whose lactate level normalized in 24 hours survived. If lactate levels cleared to normal between 24 and 48 hours, the survival rate was 75%. Only 3 of the 22 patients who did not clear their lactate level to normal by 48 hours survived. Ten of the 25 nonsurvivors (40%) achieved the above arbitrary optimization criteria. Fifteen of the survivors never achieved any of these criteria. Optimization alone does not predict survival. However, the time needed to normalize serum lactate levels is an important prognostic factor for survival in severely injured patients.