Changes of central venous oxygen saturation define fluid responsiveness in patients with septic shock: A prospective observational study

Abstract

Purpose: To evaluate whether the changes of central venous oxygen saturation (Scvo₂) after fluid challenge can define fluid responsiveness in patients with septic shock.

Methods: In this prospective observational study, septic shock patients with invasive cardiac output monitoring requiring fluid challenge were included. Cardiac index (CI) and Scvo₂ were measured before and after fluid challenges. The changes of CI (ΔCI) and the changes of Scvo₂ (ΔScvo₂) were calculated and analyzed using Pearson correlation. Receiver operating characteristics curve (ROC) analysis was used to classify fluid responders and nonresponders. Area under ROC was calculated.

Results: Forty patients were included and 18 patients (45%) were fluid responders. In the responders, CI increased from 3.4±1.1 to 4.4±1.0 L min^-1 m^-2 and Scvo₂ from 69.6%±9.8% to 77.1%±8.9% (both P<.001) after fluid challenge. In the nonresponders, neither CI nor Scvo₂ changed (4.1±1.3 vs 4.1±1.3 L min^-1 m^-2, 71.0%±13.8% vs 70.6%±14.1%, both P>.05). The correlation between ΔScvo₂ and ΔCI was significant (r=0.702, P<.001). The area under ROC of ΔScvo₂ to define fluid responsiveness was 0.88 (95% confidence interval [95% CI], 0.76-0.99). A ΔScvo₂ threshold value of 5.0% discriminated responders from nonresponders with sensitivity of 0.78 (95% CI, 0.52-0.93) and specificity of 0.95 (95% CI, 0.75-1.00).

Conclusions: The changes of Scvo₂ correlate with the changes of CI, and the changes of Scvo₂ define fluid responsiveness in patients with septic shock.

Trial registration: ClinicalTrials.gov NCT01941472.

Keywords: Central venous oxygen saturation; Fluid responsiveness; Septic shock.