Send to

Choose Destination
Neurocrit Care. 2018 Feb;28(1):93-96. doi: 10.1007/s12028-017-0398-3.

Extubating the Neurocritical Care Patient: A Spontaneous Breathing Trial Algorithmic Approach.

Author information

Department of Neurology, University of Missouri, 5 Hospital Drive CE 540, Columbia, MO, 65211, USA.
Department of Neurology, University of Missouri, 5 Hospital Drive CE 540, Columbia, MO, 65211, USA.



Delaying extubation in neurologically impaired patients otherwise ready for extubation is a source for significant morbidity, mortality, and costs. There is no consensus to suggest one spontaneous breathing trial (SBT) over another in predicting extubation success. We studied an algorithm using zero pressure support and zero positive end-expiratory pressure (ZEEP) SBT followed by 5-cm H2O pressure support and 5-cm H2O positive end-expiratory pressure (i.e., 5/5) SBT in those who failed ZEEP SBT.


This is a retrospective analysis of intubated patients in a neurosciences intensive care unit. All eligible patients were initially challenged with ZEEP SBT. If failed, a 5/5 SBT was immediately performed. If passed either the ZEEP SBT or the subsequent 5/5 SBT, patients were liberated from mechanical ventilation.


In total, 108 adult patients were included. The majority of patients were successfully liberated from mechanical ventilation using ZEEP SBT alone (82.4%; p = 0.0007). Fifteen (13.8%) patients failed ZEEP SBT but immediately passed 5/5 SBT (p = 0.0005). One patient (0.93%) required reintubation. We found high sensitivity of this extubation algorithm (100; 95% CI 95.94-100%) but poor specificity (6.67; 95% CI 0.17-31.95%).


This study showed that the majority of patients could be successfully liberated from mechanical ventilation after a ZEEP SBT. In those who failed, a 5/5 SBT increased the successful liberation from mechanical ventilation.


Extubation; Mechanical ventilation; Neurocritical care; Spontaneous breathing trial (SBT); ZEEP


Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center