Chapter 16: Ventilator Liberation

The ultimate goal of mechanical ventilation is ventilator discontinuation. Most patients can be liberated from the ventilator when the physiologic reason for ventilatory support is reversed. In others, this may be a more prolonged process and associated with chronic critical illness. Because of their underlying disease process, some patients may become chronically ventilator-dependent (eg, those with neuromuscular disease). This chapter addresses issues defining readiness for ventilator discontinuation, assessments that predict readiness for ventilator liberation, approaches to liberation from ventilator support, use of protocols, automated weaning, and assessment for extubation. The content of this chapter is written to be consistent with evidence-based clinical practice guidelines (Table 16-1).

There are four commonsense factors to be assessed to determine readiness for ventilator discontinuation: (1) reversal of the indication for ventilator support, (2) adequate gas exchange, (3) ability to initiate a breath, and (4) hemodynamic stability.

Reversal of Indication for Ventilator Support

The most important indicator of readiness for discontinuation of ventilatory support is some reversal of the indication for ventilatory support. In addition to respiratory failure, this includes consideration of fever, nutrition, and electrolyte balance. Renal, liver, or gastrointestinal dysfunction may adversely impact the ability to liberate the patient from the ventilator and, thus, attention should be given to correcting these abnormalities.

Gas Exchange

There should be acceptable gas exchange before initiation of the ventilator discontinuation process. From an oxygenation perspective, this typically means a Pao₂ more than 60 mm Hg with an Fio₂ less than or equal to 0.50 and positive end-expiratory pressure (PEEP) less than or equal to 8 cm H₂O. A high ventilation requirement for a pH more than 7.25 decreases the likelihood of successful liberation. A dead space to tidal volume ratio (V_D/V_T) less than 60% and a minute ventilation less than 12 L/min is associated with a greater potential for ventilator liberation.

Ability to Initiate a Breath

To be liberated from mechanical ventilation, the patient must be able to initiate a breath. This means that there are no central ventilatory drive issues. The principal reason for ventilatory drive suppression is excessive sedation. For this reason, a spontaneous awakening trial (SAT) in which sedation is stopped is an important step in the ventilator liberation process. A daily SAT safety screen should be performed, which consists of the following: not receiving a sedative infusion except for active seizures or alcohol withdrawal, not receiving escalating sedative doses due to ongoing agitation, not receiving neuromuscular blockers, no evidence of active myocardial ischemia in the past 24 hours, and no evidence of increased intracranial pressure. Patients passing the screen should receive an SAT, during which time all sedatives and analgesics used for sedation are stopped for up to 4 hours, but analgesics needed for active pain are continued. Patients are deemed to have passed the SAT if they open their eyes to verbal stimuli. Patients are deemed to have failed the SAT if they develop sustained anxiety, agitation, or pain; respiratory rate more than 35 breaths/min for 5 minutes or longer; Spo₂ less than 88% for 5 minutes or longer; acute cardiac dysrhythmia; or two or more signs of respiratory distress including tachycardia, bradycardia, use of accessory muscles, abdominal paradox, diaphoresis, or marked dyspnea. For a failed SAT, sedatives are started at half the previous dose and then titrated to achieve patient's comfort.

The choice of sedative agent may also affect the ventilator liberation process. Benzodiazepine administration is associated with the development of delirium, and delirium has been associated with a longer period of ventilator dependence. On the other hand, when compared with a benzodiazepine, use of dexmedetomidine resulted in less time on the ventilator and less delirium.

Hemodynamic Stability

Cardiovascular function should be optimized before initiation of the ventilator discontinuation process. Arrhythmias, fluid overload, and myocardial contractility should be managed appropriately. The patient should be hemodynamically stable with minimal cardiovascular support, no cardiac ischemia, and no unstable arrhythmia.

A number of so-called weaning parameters have been introduced, the intent of which is to identify extubation readiness. Most predictors of weaning outcome focus on the ability to achieve or sustain a specific ventilatory parameter. Unfortunately, no predictive parameter is 100% accurate in identifying individuals who will successfully be liberated from the ventilator. In fact, the best predictor of successful ventilator liberation is the patient's response to an SBT. High-level evidence supporting the use of weaning parameters is lacking.

The most commonly used predictors of weaning success are listed in Table 16-2. They are grouped into indices that evaluate ventilatory drive, ventilatory muscle capability, and ventilatory performance. The most accurate predictor of weaning success is the rapid-shallow breathing index (RSBI). This index is determined by dividing the respiratory rate by the tidal volume in liters, determined 1 minute after removing the patient from the ventilator. If the RSBI is less than or equal to 105, the probability of successful weaning is high and if it is more than 105, the probability of failure is high. However, more recent reports have shown that the RSBI may not be as highly predictive of ventilator liberation as originally reported.

The spontaneous breathing trial (SBT) is the best way to determine readiness for ventilator liberation. Patients who tolerate an SBT for 30 to 120 minutes should be considered liberated from ventilatory support and candidates for extubation. The traditional way of performing an SBT is with a T-piece connected to the endotracheal tube, providing humidified oxygen. Many clinicians use the ventilator for SBTs, which has the advantage of maintaining a precise Fio₂ and patient monitoring during the SBT. Moreover, if the SBT fails, ventilatory support can be quickly reestablished. If the SBT is to simulate a T-piece trial, the pressure support and PEEP should both be set at 0.

It is the practice of some clinicians to perform the SBT using low levels of pressure support (5-10 cm H₂O), PEEP (5 cm H₂O), or tube compensation. Most patients do equally well on T-piece trials, pressure support and PEEP set at 0, or low levels of pressure support and PEEP. The intent of using a low level of pressure support or tube compensation is to overcome the resistive load of the endotracheal tube. However, the resistive load of the endotracheal tube is usually not excessive unless the tube size is small. Patients with small endotracheal tubes or nasal intubation may benefit from the application of low levels of pressure support (5-10 cm H₂O). Otherwise, breathing through the endotracheal tube with no support from the ventilator closely simulates the resistance through the upper airway after extubation.

Use of low levels of PEEP during the SBT is discouraged. In patients with chronic obstructive pulmonary disease (COPD), the application of 5 cm H₂O PEEP during the SBT may counterbalance auto-PEEP. In patients with poor cardiac function, a low level of PEEP during the SBT may be sufficient to keep the patient out of failure. In patients with COPD or poor cardiac failure, the use of PEEP during the SBT might predict success, only to have the patient develop respiratory failure soon after extubation.

Patients successfully completing an SBT of 30 to 120 minutes are considered for extubation (Table 16-3). If the patient does not tolerate the SBT, the ventilator is set to provide a comfortable level of support. Once a comfortable level of support is provided, evidence is lacking for reducing the level of support before the next SBT. Before the next SBT, usually on the following day, an attempt should be made to identify and correct all potential causes of the failed SBT.

Approaches to a Failed Spontaneous Breathing Trial

A common reason for a failed SBT is an imbalance between the capacity of the respiratory muscles (weakness) and the load that is placed on them (Table 16-4). Causes of respiratory muscles weakness include critical illness weakness, electrolyte imbalance, malnutrition, and primary neuromuscular disease. An excessive respiratory muscle load can be the result of high airways resistance, as in the patient with COPD, or a low compliance, as in a patient with pneumonia or pulmonary edema. Auto-PEEP and a high minute ventilation requirement also increase the load on the respiratory muscles.

The maximal inspiratory pressure (Pi_max) is used to measure respiratory muscle strength. To ensure the best results, measurement of Pi_max should be performed at residual volume. To achieve this, a one-way valve as illustrated in Figure 16-1 is used. This allows exhalation but not inspiration. Thus the lung volume at which Pi_max is measured decreases with each breathing attempt. The Pi_max measurement should be performed for about 20 seconds provided no arrhythmias or desaturation occurs.

There is increasing evidence supporting the role of early mobility of critically ill mechanically ventilated patients. This improves skeletal muscle conditioning, including the respiratory muscles. Early mobility has also been associated with a lower incidence of delirium, which might also result in earlier liberation from mechanical ventilation.

Iatrogenic causes also contribute to a failed SBT. Examples included a partially obstructed endotracheal tube, dead space in the ventilator circuit, or increased airways resistance while breathing aerosolized water from the T-piece.

A gradual reduction in synchronized intermittent mandatory ventilation rate has been used for weaning. Similarly, a gradual reduction in the level of pressure support ventilation has been used for weaning. The intent in both cases is to gradually transfer the work-of-breathing from the ventilator to the respiratory muscles of the patient. However, evidence is lacking that gradual reductions of support facilitate liberation more quickly than daily SAT and SBT. Ventilator modes such as SmartCare are available that provide automated weaning. Although these automated modes may be equivalent to clinician-directed weaning, additional data supporting fewer ventilator days are needed before they can be recommended.

A successful approach to liberating patients from ventilatory support is the use of protocols implemented by respiratory therapists and nurses. These protocols result in shorter weaning times and shorter lengths of mechanical ventilation than traditional physician-directed weaning. The primary reason for the success of protocols is that they are developed by multidisciplinary teams and are implemented by respiratory therapists and nurses empowered to make clinical decisions. Although protocols can be developed for any approach to weaning, the approach that has been most commonly used is based on SBTs. Elements of a ventilator discontinuation protocol are shown in Figure 16-2.

Once the patient has successfully completed an SBT of 30 to 120 minutes, consideration should be given for extubation. Even if the patient has successfully completed an SBT, many clinicians tend to be too conservative, wanting to avoid reintubation. Although reintubation is associated with morbidity and mortality risk, prolonged intubation in a patient who can be extubated is also associated with poorer outcomes. A reasonable reintubation rate in general medical or surgical units is about 10% to 20%.

Inability to perform four simple tasks (open eyes, follow with eyes, grasp hand, and stick out tongue), to generate a cough peak flow less than or equal to 60 L/min, and with secretions more than or equal to 2.5 mL/h have been reported to increase the risk of reintubation (Table 16-5). The potential for upper airway obstruction following extubation should always be considered. Deflating the cuff and assessing for air leak around the tube and through the upper airway when positive pressure is applied is sometimes performed to assess this. If no air movement around the tube is identified, this is suggestive of upper airway swelling and potential for airway obstruction after removal of the tube. However, use of the cuff leak test is associated with false-positive and false-negative results. Its use is controversial and should not be used routinely, but rather only in patients with a high probability of upper airway swelling. If upper airway swelling is suspected, a short course of steroid therapy is indicated before extubation, and personnel trained to reintubate should be at the bedside at the time of extubation.

In patients at risk for extubation failure, preventative use of noninvasive ventilation (NIV) is indicated. However, the use of NIV to rescue a failed extubation should be limited to patients with hypercapnic respiratory failure.

The patient with chronic critical illness (CCI) has survived an acute critical illness or injury but has not yet recovered to the point of liberation from life-sustaining therapies. Such patients are weak, deconditioned, often delirious or comatose, and receiving prolonged mechanical ventilation (PMV). Other forms of organ support such as vasopressors, inotropes, and renal replacement therapy may be required, and the patient may be receiving one of several courses of broad-spectrum antibiotics for ongoing or recurrent infections. A tracheostomy is often in place as well as a feeding tube. The decision about placing a tracheostomy is a point of demarcation often used to identify CCI, as is the need for PMV. A common definition of CCI is PMV more than or equal to 21 consecutive days for more than or equal to 6 h/d. When medically stable, patients who have failed ventilator discontinuation attempts in the ICU should be transferred to long-term acute-care facilities who have demonstrated success and safety in accomplishing ventilator discontinuation. Weaning strategy in the prolonged mechanically ventilated patient should be slow-paced, and should include gradually lengthening SBTs.

Patients with irreversible neuromuscular diseases who require long-term mechanical ventilation are a unique group. Unlike patients with CCI, they have not suffered from acute illnesses or injuries that involve systemic inflammation and multiorgan failure. Therefore their outcomes and resource needs are different than the typical patient with CCI. Important discussions regarding their care usually revolve around safe and effective provision of home mechanical ventilation rather than recovery from multiorgan failure. The relationships between CCI, PMV, and long-term mechanical ventilation are represented schematically in Figure 16-3. Unless there is evidence for clearly irreversible disease (eg, high spinal cord injury, advanced amyotrophic lateral sclerosis), a patient requiring prolonged mechanical ventilatory support for respiratory failure should not be considered permanently ventilator-dependent until 3 months of weaning attempts have failed.