Oxygen transport patterns in patients with sepsis syndrome or septic shock

Objective To investigate the relationship between oxygen transport patterns and outcome in patients with sepsis syndrome or septic shock managed according to two different treatment regimens. Design Retrospective study of a subgroup of patients with sepsis syndrome or septic shock taken from a randomized, prospective, controlled trial. Setting General intensive care units in a teaching and a district general hospital. Patients Seventy-eight patients classified according to predetermined criteria as having sepsis syndrome or septic shock were drawn retrospectively from a larger study group of 109 consecutive patients considered to be at risk for developing multiple organ failure. Interventions All patients received volume expansion to an optimal pulmonary artery occlusion pressure. If the therapeutic goals (cardiac index of >4.5 L/min/m sup 2, oxygen delivery [DO2] of >600 mL/min/m2, and oxygen consumption [VO2] of >170 mL/min/m2) were not achieved with fluids alone, patients were randomized to either a control group or a treatment group. In the treatment group, dobutamine (5 to 200 micro g/kg/min) was administered to increase cardiac index and DO2 until all three goals were simultaneously achieved. In the control group, dobutamine was administered only if the cardiac index was <2.8 L/min/m2. In both groups, norepinephrine was infused to maintain mean arterial pressure at 80 mm Hg. Measurements and Main Results Hemodynamic, oxygen transport, and lactate measurements were made at the time of admission to the study, at the time of optimal volume administration, at 1, 2, 4, 8, 12, 16, 20, and 24 hrs, then every 6 hrs for the next 24 hrs, and at least every 8 hrs thereafter. The time at which all therapeutic goals were first achieved simultaneously or the time of maximal DO2 was identified and termed "tmax." Survivors from both the control and treatment groups significantly (p < .001) increased cardiac index and DO2 in response to maximal resuscitation, and despite an associated decrease in oxygen extraction (p < .01), there was a significant (p < .01) increase in VO2. In nonsurvivors from both groups, despite significant increases in cardiac index (p < .05) and DO2 (p < .01) at tmax, oxygen extraction decreased (p < .01) and VO2 remained unchanged. DO2 and VO2 were significantly lower (p < .05) at tmax in nonsurvivors than in survivors from both groups. Persistently high lactate concentrations were characteristic of nonsurvivors. Conclusions Survivors of sepsis syndrome or septic shock are characterized by an ability to increase both DO2 and VO2. In contrast, nonsurvivors typically have reduced cardiac reserve, they fail to increase VO2 following resuscitation, and when delivery is enhanced with aggressive inotropic support, oxygen extraction falls. These patterns of response were similar in both treatment and control groups, although the magnitude of the changes was exaggerated in the treatment group. These observations may help to explain the findings by some investigators that treatment aimed at achieving survivor values of cardiac index, DO2, and VO2 fails to improve outcome when instituted following admission to intensive care. (Crit Care Med 1997; 25:926-936)