Central venous oxygen saturation as predictor of a successful spontaneous breathing trial outcome in mechanically ventilated patients

Abstract

Rationale: Weaning from mechanical ventilation is a key element in the care of the critically ill patient and spontaneous breathing trial (SBT) is a crucial step in this procedure. This study aimed to evaluate the diagnostic accuracy of central venous oxygen saturation (ScvO2), a marker of tissue oxygenation, as a potential predictor of successful SBT.

Methods: This is a nested case-control study. A consecutive population of patients who were mechanically ventilated for ≥48h and fulfilled the initial criteria of readiness to wean, according to guidelines¹, constituted the study population. All patients attempted a SBT. Patients were categorized to cases (SBT success) and controls (SBT failure), based on a combination of specific criteria¹ which indicated whether SBT was successful or not. Binary logistic regression analysis was utilized to indicate independent predictors of SBT success and ROC curve was used to demonstrate diagnostic accuracy of indices.

Results: 77 patients (64.8±16.3 years; 62.3% male) constituted the study population. SBT was successful in 63.6% of them. SvO2 predicted SBT success (OR: 1.122; 95% CI:1.005-1.257) independently of age, hemoglobin concentration, arterial oxygen saturation (SaO2), arterial oxygen partial pressure (PaO2) and respiratory rate to tidal volume ratio. Diagnostic accuracy for ScvO2 alone (ROC area=0.685) was superior of that of either SaO2 (0.625) or PaO2 (0.619) alone to predict SBT success.

Conclusion: ScvO2 seems to be a useful parameter which could further facilitate the accurate categorization among those patients who pass or fail the SBT.

¹ Boles J.M., et al. Eur Respir J 2007; 29: 1033–1056.