Spontaneous Breathing Trial
Spontaneous Breathing Trial
Spontaneous Breathing Trial
OVERVIEW
Spontaneous breathing trials (SBT) are used to identify patients who are likely to fail
liberation from mechanical ventilation
SBT is “the defacto litmus test for determining readiness to breathe without a
ventilator”
Ideally, during an SBT we want to observe the patient under conditions of respiratory
load that would simulate those following extubation
METHOD
SBT involves the following steps:
It be conducted while the patient is still connected to the ventilator circuit, or the
patient can be removed from the circuit to an independent source of oxygen (T-piece)
When using the ventilator a PS of 5 – 7 cmH2O and 1-5 cmH20 PEEP (so called
‘minimal ventilator settings’) will overcome increased work of breathing through the
circuit (i.e. ETT)
If still on the ventilator the patient should have ‘minimal ventilator settings”
Initial trial should last 30 – 120 minutes
If it is not clear that the patient has passed at 120 minutes the SBT should be
considered a failure
In general, the shorter the intubation time the shorter the SBT required
80% of patients who tolerate this time can be permanently removed from the ventilator
Martin Tobin has argued that adding either 5 cm H2O as “physiologic” PEEP or
pressure support of 7 cm H2O to overcome the resistance in an endotracheal tube
(or both, as is usually done) may actually reduce the “spontaneously” breathing
patient’s workload by >40%
It has been shown experimentally that the work of breathing through an endotracheal
tube, compared to the work of breathing following extubation, is almost identical due
to upper airway edema resulting from an ETT being in place for several days
Tobin argues for wider use of true T-piece spontaneous breathing trials, especially in
those at high risk of failed extubation and when the consequences of failed
extubation may be catastrophic
An alternative is to have the ventilator set on “flow-by,” with pressure support and
PEEP set at zero
There is no strong evidence in favour of any of these approaches