SaO₂-SvO₂ difference for risk stratification of patients with sepsis and septic shock

Background

Assessment and monitoring of hemodynamics is a cornerstone in critically ill patients as hemodynamic alteration may become life-threatening in a few minutes [1,2]. This study aimed to determine whether the SaO₂-SvO₂ difference could be used as risk stratification for patients with sepsis and septic shock.

Materials and methods

A retrospective cohort study conducted in the ICU of Hospital Santa Luzia, Brasilia, DF, Brazil, during 6 months. An arterial blood sample was collected at admission. Patients with sepsis were divided in two groups: survivors group (SG) and nonsurvivors group (NSG). The accuracy of SaO₂-SvO₂ difference to predict ICU mortality was measured with the area under the receiver operating characteristic curve.

Results

A total of 131 patients with sepsis were enrolled, 11.5% (n = 15) with septic shock. Age was 66 ± 21 years, SAPS3: 37 ± 17, APACHE II: 14 ± 8, PaO₂/FiO₂: 342 ± 142 and SaO₂/FiO₂: 347 ± 109. ICU mortality was 18% (n = 23). The main sites of infections were respiratory (56.5%, n = 74), urinary (19%, n = 25) and cutaneous (7.6%, n = 10). Nonsurvivor patients had higher SaO₂-SvO₂ difference (26 ± 9 vs. 19 ± 9, P = 0.03). In the group of patients with septic shock, SaO₂-SvO₂ difference was also higher in nonsurvivors (29 ± 3 vs. 10 ± 2, P = 0.02). All patients with septic shock who died had SaO₂-SvO₂ difference greater than 25%. The SaO₂-SvO₂ difference area under ROC curve was 0.714 (95% CI 0.534 to 0.894).

Conclusions

A higher SaO₂-SvO₂ difference is associated with mortality in patients with sepsis, especially in patients with septic shock.