Postoperative organ failures commonly occur after major abdominal surgery and substantially increase the utilization of resources and the costs of care, and are favored by tissue hypoxia. A therapeutic strategy designed to detect and reverse tissue hypoxia, as diagnosed by an increase of oxygen extraction over a pre-defined threshold could decrease the incidence of organ failures.

The development of postoperative organ failures severely affects the prognosis of surgical patients, and substantially increases the utilization of resources and the costs of care. The presence of tissue hypoxia is particularly common and likely results from an impairment of the adaptive mechanisms of myocardial function, leading to an inappropriately low cardiac output and from the effects of inflammatory mediators. Importantly, tissue hypoxia can be detected early. Indeed, hypoxic tissues will compensate the hypoperfusion by increasing oxygen extraction (O2ER) over a threshold value, a change that will also be reflected by a decreased venous oxygen saturation, and if uncompensated by lactic acidosis. Hence, the use of oxygen extraction calculated from arterial and central venous oxygen saturation as a therapeutic goal is appropriate to monitor goal-directed hemodynamic strategies, as it reflects the balance between oxygen delivery and consumption. In place of O2ER or its surrogate, mixed venous oxygen saturation, the use of oxygen-derived values measured in central blood drawn from the routinely placed central line, in place of pulmonary artery catheter, may represent a valuable alternative. Indeed, recent evidence suggests that a multifaceted goal-directed strategy, including fluid challenge, blood transfusion and inotropes titrated to keep central venous oxygen saturation higher than a predetermined threshold of 70% was associated with decreased mortality and rate of organ failures when applied from the early phase of septic shock or severe sepsis. The aim of the present multicentre prospective randomized study was to compare the outcomes of patients randomized to a conventional management or to a therapeutic strategy guided by O2ER estimate (O2ERe) calculated from the arterial oxygen saturation and the central venous saturation ScvO2. Specifically, we hypothesized that the use of a goal-directed protocol aimed at maintaining the O2ERe below a previously defined "critical" (able to discriminate survivors from non-survivors) value of 26.7% during surgical interventions in high-risk patients will reduce the rate of postoperative organ failures, hospital length of stay and mortality, as compared with the standard management based on the monitoring of mean arterial pressure, central venous pressure and urine output.

High Risk surgical patients scheduled for elective abdominal extensive surgery were eligible. After enrollment, the patients were randomized to one of the two groups of treatment (A or B) by a phone system on a 24-hour-a-day, 7-day-a-week basis. Randomization was based on a permuted-block algorithm, allowing stratification for each center. The exclusion criteria from the study were : age below 16 years, preexistent neurologic, or malignant haematologic diseases.

In preparation for surgery, the patients were equipped with central and peripheral venous and arterial catheters. Standard monitoring included continuous recording of electrocardiography; body temperature, heart rate, pulse oxymetry and arterial blood pressure. Central venous pressure, ScvO2 , arterial blood gases, lactate concentration, body temperature, and urine output were recorded hourly. Hemoglobin concentration was measured when deemed necessary by the anesthesiologist. For the purpose of the study, blood gases measured on arterial and central venous samples, arterial lactate and O2ERe (SaO2-ScvO2/SaO2 [SaO2 : arterial oxygen saturation, ScvO2 : central venous saturation]) were recorded after induction of anesthesia (T0), hourly after cutaneous incision (T1a-f), throughout surgery, during the first 6 hours of the postoperative period (T2a-f) and on postoperative day 1.

In both groups, the patients were managed to achieve predefined standard goals, i.e. a meen arterial pressure (MAP) >80 mmHg and an urine output > 0.5ml/kg/h. The patients of the "protocol group" (Group A) were managed to keep O2ER below 26.7%, following algorithms detailed in figure 1. In brief, a fluid challenge (colloids 250-1000 ml infused over 30 min to restore central venous pressure to at least 10 mmHg), dobutamine (incremental doses of 3 µg/kg/min up to 15 µg/kg/min) and/or packed red cells (in cases of hemoglobin below 10 g/dl or intraoperative hemorrhage > 1000 ml) could be administered. There was no specific requirement regarding the type of anesthesia in any of the groups.

Inclusion Criteria:

• high risk surgical patients requiring major abdomina surgery

Exclusion Criteria:

• age below 16 years, preexistent neurologic, or malignant haematologic diseases