We evaluated whether a goal-directed protocol, without measurement of central venous oxygen saturation, would improve survival in medical intensive care unit (ICU) patients with septic shock. This is a prospective, controlled study in a 24-bed medical ICU at a tertiary care hospital. From a total of 241 consecutive patients with septic shock, 224 were randomly assigned to receive therapy with or without a written protocol using central venous pressure, mean arterial pressure, and urine output as therapeutic goals. Baseline characteristics were similar between groups. Implementation of goal-directed therapy caused a more rapid reversal of persistent shock (47 +/- 22.8 vs. 65.4 +/- 32.1 h, P = 0.006) and decreases of ICU (50% vs. 67.2%, P = 0.009) and in-hospital (53.7% vs. 71.6%, P = 0.006) mortality rates compared with non-goal-directed therapy. Patients receiving goal-directed therapy also had less risk for developing central nervous system or renal failure than patients without. Patients with goal-directed therapy received more fluid during the period of persistent shock (136.2 +/- 119 vs. 88.6 +/- 57.7 mL h, P = 0.034) and less delay in vasopressor administration (78 +/- 22.2 vs. 104.4 +/- 29 min, P = 0.001) than patients with non-goal therapy. Implementation of a goal-directed protocol improves survival and clinical outcomes in ICU patients with septic shock. These benefits may arise from adequate fluid resuscitation, earlier vasopressor administration, rapid shock reversal, and protection of major organ function. With central venous oxygen saturation measurement to detect tissue perfusion, the clinical outcomes may be further improved.