A prospective randomized trial comparing oxygen delivery versus transcutaneous pressure of oxygen values as...

Transcutaneous pressure of oxygen (PtcO2) correlates with arterial pressure of oxygen (PaO2) in nonshock states, but in shock states, PtcO2 approximates cardiac output with no response to increasing fraction of inspired oxygen (FiO2) and PaO2. An incremental change of more than 21 mmHg in PtcO2 in response to an FiO2 of 1.0 (identified as the oxygen challenge test [OCT]) implies adequate tissue perfusion, and lack of response has been associated with mortality. Patients with severe sepsis and septic shock requiring pulmonary artery catheters were randomized to two groups: the oxygen delivery (DO2) group was treated to a DO2 and mixed venous oxygen saturation goals, and the PtcO2 group was treated to achieve an OCT value of 40 mmHg or more. The DO2 (n = 30) and PtcO2 (n = 39) groups were similar in baseline characteristics. Mortality rate was 12 (40%) of 39 for the DO2 group and 5 (13%) of 39 for the PtcO2 group (P = 0.02). Logistic regression analysis of the statistically significant variables between survivors and nonsurvivors demonstrated that inability to reach the PtcO2 goal at 24 h after resuscitation (T24) and a positive cardiac history are associated with mortality (P < 0.001). The area under the receiver operating curve was 0.824 for the OCT at T24. The best OCT value was 25 mmHg at T24 with positive and negative predictive values of 87% and 90%, respectively. Treating patients with severe sepsis/septic shock to an OCT value of 25 mmHg or more may provide a specific end point of resuscitation that may be associated with better survival than resuscitating to the central hemodynamic parameters of DO2 and mixed venous oxygen saturation.