Incidence of low central venous oxygen saturation during unplanned admissions in a multidisciplinary intensive care unit: an observational study

Abstract

Introduction: It has been shown that early central venous oxygen saturation (ScvO2)-guided optimization of hemodynamics can improve outcome in septic patients. The early ScvO2 profile of other patient groups is unknown. The aim of this study was to characterize unplanned admissions in a multidisciplinary intensive care unit (ICU) with respect to ScvO2 and outcome.

Methods: Ninety-eight consecutive unplanned admissions to a multidisciplinary ICU (median age 63 [range 19 to 83] years, median Simplified Acute Physiology Score [SAPS II] 43 [range 11 to 92]) with a clinical indication for a central venous catheter were included in the study. ScvO2 was assessed at ICU arrival and six hours later but was not used to guide treatment. Length of stay in ICU (LOSICU) and in hospital (LOShospital) and 28-day mortality were recorded.

Results: ScvO2 was 70% +/- 12% (mean +/- standard deviation) at admission and 71% +/- 10% six hours later (p = 0.484). Overall 28-day mortality was 18%, LOSICU was 3 (1 to 28) days, and LOShospital was 19 (1 to 28) days. Patients with an ScvO2 of less than 60% at admission had higher mortality than patients with an ScvO2 of more than 60% (29% versus 17%, p < 0.05). Changes in ScvO2 during the first six hours were not predictive of LOSICU, LOShospital, or mortality.

Conclusion: Low ScvO2 in unplanned admissions and high SAPS II are associated with increased mortality. Standard ICU treatment increased ScvO2 in patients with a low admission ScvO2, but the increase was not associated with LOSICU or LOShospital.

Figure 1

Receiver operator characteristic (ROC) analysis of central venous oxygen saturation (ScvO₂) and Simplified Acute Physiology Score (SAPS II). Outcome parameter for ROC curves of (a) ScvO₂ and (b) SAPS II was 28-day mortality. Area under the curve (A) values were 0.53 for ScvO₂ and 0.89 for SAPS II.

Figure 2

Central venous oxygen saturation (ScvO₂) trends in the different patient groups. Trends are divided into an intensive care unit admission ScvO₂ of less than (a) and greater than (b) 60%. ScvO₂ is displayed as mean ± standard deviation. *Student's paired t test, p < 0.05, versus admission ScvO₂. CNS, central nervous system; n, absolute number of patients per subgroup.

Figure 3

Patient outcome in all patients and in the different patient groups. The numbers in the bars indicate the absolute number of patients per subgroup. Eight patients were lost to follow-up and had to be excluded from outcome analysis. No statistical difference was found for patient location at 28 days in the whole collective and in the different patient groups. CNS, central nervous system.