The discontinuation of ventilatory support is best achieved not by gradually reducing the level of ventilatory support (weaning) but by systematically identifying and eliminating the precipitants of respiratory failure. Once this goal has been achieved, the ventilator is no longer necessary. However, if precipitants are still present or recovery is incomplete, reducing needed ventilatory support is more likely to delay recovery. It is now clear that daily spontaneous breathing trials on a T-piece reduce the duration of mechanical ventilation compared with gradual reduction of the respiratory rate using synchronized intermittent mandatory ventilation and, in some studies, compared with pressure support trials.

Once the patient is no longer in shock, has an adequate arterial saturation on a fractional inspired O₂ (Fio₂) ≤ 0.5 with a positive end-expiratory pressure (PEEP) ≤ 7.5 cm H₂O and does not have an obviously unsustainable respiratory load (eg, minute ventilation > 20 L/min), a daily spontaneous breathing trial is done using a T-piece or continuous positive airway pressure (CPAP) of 5 cm H₂O. Patients capable of sustaining spontaneous breathing generally breathe slowly and deeply, instead of rapidly and shallowly. This observation has been formalized as the rapid shallow breathing (RSB) index, determined by dividing the patient's unassisted respiratory rate (in breaths/min) by the tidal volume (in L). A value < 105 suggests that spontaneous breathing is likely to be successful, although a single isolated measurement is not perfectly predictive of success. Recently, the decision of whether to extubate a patient after a spontaneous breathing trial has shifted away from the use of the RSB index and has relied more on clinical assessment during the course of the trial, supplemented by measuring ABGs. Patients who fare well during a brief 1- to 2-h spontaneous breathing trial and who have favorable ABGs are good candidates for extubation. The decision to extubate is a separate one from the decision to stop ventilatory support and requires evaluation of the patient's mentation and airway protective reflexes, as well as the patency of the airway.

Although sedatives and opioids are essential for ensuring comfort, rest, and synchrony with the ventilator, their use may prolong mechanical ventilation. Such drugs may accumulate and cause protracted sedation, frustrating attempts to do spontaneous breathing trials even when the cause of respiratory failure has been corrected. The level of sedation should be continually assessed, and progressive sedative withdrawal should be begun as soon as possible. Formal protocols can be used, or simple daily interruption can be carried out. The infusion is stopped until the patient is either awake and following commands or needs re-sedation for agitation, breathing asynchronously with the ventilator, or other physiologic derangements. If sedation is still needed, it is restarted at half the previous dose and titrated as necessary. Several studies have shown that the mean duration of mechanical ventilation is reduced in institutions that use either daily "sedation vacations" or other sedation protocols, as well as daily spontaneous breathing trials.

Last full review/revision August 2007 by Brian K. Gehlbach, MD; Jesse B. Hall, MD

Content last modified February 2012