Management of the difficult-to-liberate adult patient in the intensive care unit

INTRODUCTION — 

Many patients in intensive care units (ICUs) are difficult to wean from mechanical ventilation, resulting in delayed extubation and liberation from the device. The management of patients who are difficult to liberate from mechanical ventilation in the ICU is reviewed here. Details regarding readiness testing, methods of weaning, and the management of patients who require prolonged mechanical ventilation in long-term care facilities are provided separately. (See "Management and prognosis of patients requiring prolonged mechanical ventilation in long-term acute care hospitals (LTACH)".)

DEFINITION AND INCIDENCE — 

Weaning from mechanical ventilation is also referred to as "liberation" from mechanical ventilation and can be classified as simple, difficult, or prolonged [1].

●Simple-to-wean – Patients in this classification have successfully passed their first spontaneous breathing trial (SBT). Approximately one-half to two-thirds of patients in ICUs undergo simple weaning, many of which will proceed with extubation. Details regarding what constitutes an SBT and extubation management are discussed separately [2,3]. (See "Initial weaning strategy in mechanically ventilated adults" and "Extubation management in the adult intensive care unit".)

●Difficult-to-wean – Patients in this classification fail their first SBT and then require up to three SBTs or seven days to pass an SBT [1]. The incidence ranges from 26 to 39 percent [2,3]. This population mostly includes those who require mechanical ventilation in the first few weeks of acute ICU admission. This population is discussed in this topic review.

●Prolonged weaning – Patients in this classification have failed at least three SBTs or require more than seven days to pass an SBT. The incidence ranges from 6 to 14 percent [2,3]. Patients who require more than seven days to achieve liberation from mechanical ventilation are at increased risk for death [3] and are also more likely to require reintubation compared with those who are simple to wean [4]. While in the ICU, many of these patients are managed similarly to patients who are difficult to wean, and many will require tracheostomy. (See "Tracheostomy: Rationale, indications, and contraindications" and "Management and prognosis of patients requiring prolonged mechanical ventilation in long-term acute care hospitals (LTACH)".)

Recognizing that some patients are liberated from invasive mechanical ventilation without a well-defined SBT, a modification of the above classification system has been proposed [5]. In this alternative classification, a "separation attempt" was defined as an SBT (with or without extubation) or extubation (planned or unplanned) without a preceding SBT. "Successful separation" was defined as extubation within seven days or discharge from the ICU without invasive mechanical ventilation. Among 2907 patients, 24 percent never had attempted weaning or separation while 57 percent had weaning of less than 24 hours (short weaning); 95 percent of these patients achieved successful separation. Ten percent had weaning duration from two to six days (difficult weaning), with 84 percent achieving successful separation. In 9 percent, weaning lasted more than one week (prolonged weaning) and 62 percent eventually achieved successful separation. Importantly, mortality rates rose from one group to the next (short 5 percent, difficult 15 percent, prolonged 30 percent) [5,6].

IDENTIFY AND CORRECT THE CAUSES LIMITING SUCCESSFUL LIBERATION — 

Repeated unsuccessful attempts at liberation from mechanical ventilation usually signify incomplete resolution of the illness that precipitated mechanical ventilation and/or the development of one or more new problems that prevent liberation. The clinician should identify and treat these issues before resuming further efforts at liberation from mechanical ventilation.

●Identify and treat the cause(s) – Numerous factors contribute to failure to wean (table 1) by causing an imbalance between respiratory muscle strength and the work of breathing. Respiratory and cardiac issues are common while psychological, ventilator, and nutrition-related issues are less common. Most of the etiologies are apparent on routine clinical examination, laboratory testing, arterial blood gas analysis, electrocardiography, and chest radiography, as well as an assessment of patient medications and the ventilator circuit. Additional investigations may be indicated depending upon the suspicion for select etiologies. (See 'Respiratory or ventilatory causes' below and 'Cardiac causes' below and 'Psychological causes' below and 'Ventilator circuit issues' below and 'Nutritional issues' below.)

When feasible, identified etiologies should be treated to improve the probability of successful liberation [7,8].

●Resume weaning – Once it is felt that the likely cause of continued ventilator dependency has been corrected, readiness testing (table 2) should be performed to determine whether weaning can be resumed. (See "Weaning from mechanical ventilation: Readiness testing".)

A predictive index has been proposed (I-TRACH score), but is not in clinical use [9].

Respiratory or ventilatory causes — Respiratory/ventilatory causes of failure to wean, their detection and treatment are listed in the table (table 1).

Clinical manifestations are often nonspecific and include tachycardia, tachypnea, respiratory distress, use of accessory muscles, or oxygen desaturation during a breathing trial.

Routine testing including chest, cardiac, and neurologic examination (including an assessment for delirium), complete blood count and chemistries (including calcium, magnesium, and phosphate), chest radiography, arterial blood gas analysis, and nutrition assessment will narrow the differential considerably. A review of recently administered or current medications to identify potential contributors to respiratory depression, somnolence, or neuromuscular weakness, can also be helpful. Additional testing is individualized and targeted at specific suspected etiologies (eg, computed tomography [CT] of the chest and/or abdomen, CT pulmonary angiography, thoracic ultrasound [lung, pleura, diaphragm, cardiac], nerve conduction studies, bronchoscopy [to look for airway pathology]).

Frequently overlooked causes include:

●Auto-positive end-expiratory pressure (PEEP) – The detection and treatment of auto-PEEP are discussed separately. (See "Positive end-expiratory pressure (PEEP)", section on 'Intrinsic (auto) PEEP'.)

●Overventilation – One common mistake contributing to difficulty in ventilator liberation in patients with chronic hypercapnia is to set minute ventilation with the goal of normalizing the arterial carbon dioxide tension (PaCO₂). This causes the pH to rise and prompts renal excretion of bicarbonate over three to five days until the pH normalizes. When the patient resumes spontaneous ventilation, such as that during an SBT, an acute respiratory acidosis will result because there is insufficient bicarbonate for buffering (thereby constituting a failed SBT). Thus, patients with chronic hypercapnia should be mechanically ventilated in between SBTs with a minute ventilation that maintains the patient's usual PaCO₂.

●Imposed work of breathing by endotracheal tubes – Small-sized endotracheal tubes (ETTs) or ETTs narrowed by secretions can significantly increase the work of breathing and produce a pattern of rapid shallow breathing observed in patients with weaning failure (akin to breathing through a straw) [10]. Use of a pressure support SBT may overcome the imposed work of breathing by the ETT and facilitate successful weaning [11].

Moderate to large pleural effusions are common in ventilated patients at the time of SBTs and in one prospective observational study, were found to be an independent predictor of weaning failure [12]. However, while drainage may improve oxygenation and lung volumes [13] and improve diaphragmatic contractility and respiratory system compliance [14], it has not been conclusively shown to reduce mechanical ventilation days. Thus, we typically avoid large volume thoracentesis in this setting unless indicated for another reason.

Cardiac causes — Weaning-induced myocardial ischemia or cardiac dysfunction (called "weaning-induced pulmonary edema") may contribute to difficult weaning in 20 to 60 percent of patients and when identified, should be treated accordingly (eg, optimize beta-blockade, diuresis) (table 1) [15-17]. Because of the increased work involved in spontaneous breathing, myocardial demand increases, thereby resulting in ischemia in susceptible individuals and pulmonary edema in those with underlying cardiac dysfunction. Ischemia and/or pulmonary edema, in turn, further increase the work of breathing, creating a vicious cycle.

Because PEEP is a treatment for heart failure, weaning-induced cardiac dysfunction may be hard to detect when SBT strategies that include PEEP, rather than a T-tube or non-PEEP strategies, are used [18]. Thus, cardiac dysfunction should be suspected in patients who pass an SBT that included PEEP but who fail extubation and require reintubation for subsequent development of pulmonary edema.

Manifestations include tachycardia, tachypnea, oxygen desaturation, hypertension or even hypotension during a breathing trial. Patients rarely complain of chest pain, and ST depressions are only occasionally appreciated on electrocardiography (EKG). Risk factors identified in one retrospective study included chronic obstructive pulmonary disease (COPD), previous cardiomyopathy, and obesity [15].

When ischemia is suspected, most experts use continuous multilead EKG during the SBT or record an EKG pre- and post-trial. Rarely, patients require cardiac catheterization for diagnosis. If ischemia is identified, it should be treated. The treatment of myocardial ischemia and cardiac dysfunction are described separately. (See "Overview of the acute management of non-ST-elevation acute coronary syndromes".)

When weaning-induced pulmonary edema is suspected, transthoracic echocardiography is used in conjunction with clinical examination and brain natriuretic peptide (BNP) to justify optimizing cardiac medications and altering fluid management. (See "Treatment and prognosis of heart failure with preserved ejection fraction" and "Overview of the management of heart failure with reduced ejection fraction in adults".)

●BNP levels – The possibility that weaning is limited by cardiac dysfunction is suggested by an elevated BNP or N-terminal pro-BNP prior to the weaning trial, an elevated N-terminal pro-BNP at the end of the trial, or a >20 percent increase in BNP during the SBT [8,19-21]. Interventions based upon BNP levels alone are not typically routine. However, one randomized trial reported that BNP-guided fluid management, particularly in patients with left ventricular systolic dysfunction, reduced time to extubation (59 versus 42 hours) [22].

●Echocardiography – The identification of systolic or diastolic dysfunction by transthoracic echocardiography may favor cardiac dysfunction as cause of failure to wean. By finding patients at risk for weaning-induced pulmonary edema [21], transthoracic echocardiography identifies those likely to benefit from treatment for heart failure. As an example, in a single-center observational study of 59 patients with COPD or heart failure, echocardiography-guided treatment resulted in negative fluid balance, decreased left ventricular filling pressures and weaning success in 12 patients with weaning-induced pulmonary edema [23]. (See "Transthoracic echocardiography: Normal cardiac anatomy and tomographic views".)

While not routine, some echocardiographic parameters and stress echocardiography have been described as potentially useful in identifying weaning-induced cardiac dysfunction. In a meta-analysis of 11 studies, transthoracic echocardiographic parameters indicating decreased left ventricular diastolic function and increased left ventricular filling pressure were associated with weaning failure [24].

●Other reported methods that require further validation include:

•Passive leg raising (PLR) – One study reported that patients who fail to increase cardiac output by >10 percent during PLR (negative PLR test) are significantly more likely to fail an SBT [25]. In another trial, a positive PLR was associated with a successful weaning trial while a negative one was associated with failure [15]. (See "Novel tools for hemodynamic monitoring in critically ill patients with shock", section on 'Passive leg raising or fluid bolus challenge'.)

Psychological causes — Psychological issues (eg, depression, anxiety, delirium) can be significant impediments to successful weaning. When suspected as contributors to difficult weaning, medications should be adjusted and the psychological issue treated, if feasible (eg, antipsychotic, anxiolytic, analgesic).

Depression, anxiety, and delirium are common in patients on mechanical ventilation. Depressive disorders are present in approximately 40 percent of patients undergoing weaning from prolonged mechanical ventilation (eg, patients in long-term acute care) and their presence adversely affects weaning success [26]. In another single center study, delirium was associated with two-fold higher odds for patients to be categorized as having difficult weaning versus simple weaning [27].

Such disorders may be hard to assess and may simply manifest as agitation, grimacing, disorientation, tachycardia, and hypertension during the trial.

Interventions that may minimize these factors include explaining the weaning plan to the patient, family, and other caretakers; arranging for a trusted caretaker to provide reassurance and explanation during weaning trials; biofeedback (if feasible or available) that displays the breathing pattern [28]; ensuring adequate sleep; and environmental stimulation during the trial, such as television, radio, or books. Efforts to optimize medications for treating delirium and/or agitation (eg, switching anxiolytic to dexmedetomidine [29]) and pain should also be performed. Oversedation should be avoided. (See "Sedative-analgesia in ventilated adults: Management strategies, agent selection, monitoring, and withdrawal" and "Sedative-analgesia in ventilated adults: Medication properties, dose regimens, and adverse effects".)

Ventilator circuit issues — Ventilator circuit-related problems can increase the work of breathing and contribute to SBT failure. Potential problems include equipment dead space, poor circuit compliance, low gas compression volume, exhalation valve function, and increased resistance caused by the ETT, inspiratory circuit, or expiratory circuit [30]. Consulting respiratory therapy and examining the pressure and flow tracings on the ventilator screen can help identify and eliminate potential ventilator-related problems. As examples:

●An elevated peak inspiratory airway pressure (Ppeak) with a sizeable difference between the Ppeak and plateau pressure (Pplat; eg, >10 cm H₂O) may suggest an obstructed ETT or abnormally increased resistance in ventilator circuit tubing (assuming that the patient is on volume-cycled ventilation). Examining the ETT and circuit for obstructions or secretions may reveal an obvious source that needs to be treated accordingly (eg, replacing ETT or tubing, frequent suctioning, removing the obstruction). Notably, a substantial difference between Ppeak and Pplat can also indicate issues with increased airways resistance within the tracheobronchial tree unrelated to the ventilator circuit.

●An increase in both Ppeak and Pplat may suggest a malfunction of the exhalation valve, which should be checked if this waveform abnormality is found. However, such changes may also result from issues not related to the ventilator circuit (ie, processes that decrease respiratory system compliance such as pulmonary edema, pneumothorax, pleural effusion).

●Ventilator asynchrony may be evident on waveforms. This manifests as minimal or no inspiratory flow, despite patient inspiratory effort due to continued neural inspiration (contraction of inspiratory muscles) after the ventilator has cycled to the expiratory phase (ie, premature opening of the exhalation valve). This may lead to a transient but prominent decrease in peak expiratory flow, followed by increase and then a slow decrease again as neural expiration ends [31]. Such inspiratory effort increases the work of breathing and may injure the inspiratory muscles as they contract eccentrically (eg, the muscle contracts while lengthening). If suspected, the cause of ventilator asynchrony should be sought and treated (eg, auto-PEEP). (See "Acute respiratory distress syndrome: Ventilator management strategies for adults", section on 'Treat dyssynchrony'.)

Nutritional issues — Protein catabolism induced by critical illness leads to decreased respiratory muscle mass, strength, and endurance, which may lead to difficulty weaning from mechanical ventilation [32]. Sarcopenia (determined by measuring total psoas muscle area by CT) is an independent risk factor for difficulty weaning in surgical patients [33]. Although it is rare for malnutrition to be the sole cause, it can contribute to difficulty weaning such that this population should be evaluated by a nutrition expert and their nutrition support optimized. This approach is based upon studies that report a possible increase in the number of ventilator-free days when adequate nutrition is provided, the details of which are discussed separately. (See "Nutrition support in critically ill adult patients: Initial evaluation and prescription".)

Overfeeding with excessive carbohydrates can impair weaning success [34], presumably by leading to excess carbon dioxide production and an increased ventilatory load on the respiratory muscles. While in the past overfeeding was common, it is now a rare phenomenon.

TRIALS TO LIBERATE PATIENTS FROM MECHANICAL VENTILATION — 

Once the potential causes have been identified and treated, weaning trials may resume.

General strategies

Posture — Patients should be placed in the posture that they prefer during a spontaneous breathing trial. In patients with chronic obstructive pulmonary disease, the optimal posture varies. Some patients have less dyspnea when they are lying supine, whereas others prefer to lean forward. Patients with diaphragmatic paralysis generally prefer and perform better in an upright position because their vital capacity decreases when they are horizontal. In contrast, patients with intercostal muscle weakness (eg, due to a low cervical cord lesion) may prefer being supine because their lung volumes tend to increase when they move from an upright to a supine position.

Airway management — Airway secretions should be suctioned before every weaning trial.

Theoretically, bronchodilators may facilitate weaning in patients with airway obstruction by reducing airway resistance and the work of breathing [35,36]. In a randomized study, nebulized budesonide reduced weaning duration in chronic obstructive pulmonary disease (COPD) patients who required mechanical ventilation with more than three SBTs or more than seven days from the first SBT [37]. We use short-acting inhaled beta-adrenergic and/or anticholinergic agents prior to a weaning trial in patients with airflow obstruction unrelated to the artificial airway. We use nebulized corticosteroids in difficult-to-wean COPD patients. (See "Invasive mechanical ventilation in acute respiratory failure complicating chronic obstructive pulmonary disease", section on 'Bronchodilators'.)

Trials longer than 30 minutes — In patients who are difficult-to-wean, spontaneous breathing trials (SBTs) are the same as those in patients who are weaning for the first time with one major difference being that trials are longer than the typical 30 minutes in duration (often up to two hours long). Data to support this strategy are discussed separately. (See "Initial weaning strategy in mechanically ventilated adults".)

Importantly, a weaning trial should be terminated early if the patient is failing, since respiratory muscle fatigue may develop and further decrease the chances of successful weaning. Rest is the only treatment for such fatigue, and recovery can take several days.

Method of trial — In our clinical practice, we favor once daily SBTs to wean patients who have failed an initial weaning attempt. We return the patients to a supportive mode of mechanical ventilation after a failed SBT, look for and correct reversible causes of weaning intolerance, and reassess their readiness to wean the next day. We typically use SBT with pressure support (5 to 8 cm H₂O) and positive end-expiratory pressure (PEEP; 5 cm H₂O). The one exception to this strategy is in patients with weaning failure due to cardiac dysfunction where some experts prefer to resume SBTs using a T-tube or pressure support without PEEP. This preference ensures that weaning-induced heart failure is not concealed by the use of PEEP (which is a therapy for heart failure). We do not use intermittent mandatory ventilation weaning. (See "Initial weaning strategy in mechanically ventilated adults", section on 'Choosing a weaning method'.)

A strategy of extubation with immediate application of noninvasive ventilation (NIV) may facilitate earlier extubation in select patients who might otherwise be difficult to wean. While this approach is not necessary as a routine, it is beneficial in patients who fail the initial SBT and are at risk of failing extubation, including patients with COPD or chronic hypercapnic respiratory failure. This approach should not be used in patients with difficult airways, excessive secretions, or an impaired mental status.

The practice of early extubation to NIV is supported by randomized trials and meta-analyses [38-41]. In a meta-analysis of 16 trials (1156 patients), hospital mortality was lower in the NIV weaning group compared with control (14 versus 20 percent; odds ratio [OR] 0.58, 95% CI 0.29-0.89) [40]. However, the difference was largely accounted for by the large effect size in trials involving patients with COPD (10 versus 20 percent; OR 0.43, 95% CI 0.13-0.81); whereas mortality rates were similar in both groups in trials involving mixed ICU populations (18 versus 20 percent; OR 0.88, 95% CI 0.25-1.48). NIV weaning also reduced mechanical ventilation duration and ICU length of stay, with little to no effect on hospital length of stay. The effect on these outcomes was also largely accounted for by the effect in trials involving patients with COPD, with smaller effect sizes or no effect in trials involving mixed ICU populations. NIV weaning reduced rates of ventilator-associated pneumonia (VAP) both in trials involving patients with COPD and trials in mixed ICU populations (overall VAP rates 10 versus 38 percent; OR 0.22, 95% CI 0.13-0.33). However, VAP definitions varied between trials and many trials had unusually high rates of VAP in the control arm.

One randomized trial included 364 patients with respiratory failure predominantly due to pneumonia or postsurgical complications who were deemed ready to wean but who had failed a SBT [39]. Patients were either extubated without passing an SBT and treated with NIV or were extubated only when an SBT was passed. Although the median time to liberation from ventilation (invasive or noninvasive) was no different (4.3 versus 4.5 days), the NIV group, as expected, received fewer days of invasive mechanical ventilation (1 versus 4 days) and had a lower ICU length of stay (10.8 versus 12.2 days). However, despite fewer invasive and total ventilation days, early extubation to NIV had no impact on mortality or rates of reintubation or tracheostomy. Notably, most of the included patients had hypoxemic nonhypercapnic respiratory failure and only 4 percent had an admission diagnosis of COPD or asthma exacerbation. Similarly, in a small randomized study of hypoxemic patients, early extubation to NIV reduced the duration of mechanical ventilation and length of hospital stay without reducing mortality [42].

Mechanical ventilation between trials — One hour of rest on ventilatory support after a successful SBT (but prior to extubation) may reduce the risk of extubation failure in critically ill patients [43]. A subsequent randomized controlled trial did not confirm these findings but did show, in a prespecified subgroup analysis, a reduced risk of extubation failure among patients who had been ventilated for >72 hours [44]. Similarly, appropriate ventilatory settings must be used to allow respiratory muscles to rest after a failed weaning trial (between SBTs), where the risk for respiratory muscle fatigue is substantial. In contrast, inappropriate settings can create more work than is required for spontaneous breathing, leading to respiratory muscle fatigue and impairing successful weaning.

●Mode, tidal volume, respiratory rate, and/or pressure level – Any mode may be used as long as the settings (ie, tidal volume, respiratory rate, and/or pressure level) are such that the patient is comfortable and performing minimal work. Reasonable settings are those that result in a respiratory rate between 12 and 20 breaths per minute, a tidal volume between 6 and 8 mL/kg, and a minute volume between 6 and 12 L/minute. For those on pressure-controlled ventilation a reasonable goal is a plateau pressure <30 cm H₂O (the lower the better). For those on pressure support ventilation (PSV), a pressure of 7 to 15 cm H₂O usually achieves these goals. Most patients have a PEEP between 5 and 8 cm H₂O. Importantly, patients who have chronic hypercapnia should receive a minute ventilation that targets their baseline arterial carbon dioxide tension. (See "Overview of initiating invasive mechanical ventilation in adults in the intensive care unit", section on 'Settings' and 'Respiratory or ventilatory causes' above and "Weaning from mechanical ventilation: Readiness testing", section on 'Clinical criteria'.)

●Trigger method, trigger sensitivity, and inspiratory flow – A reasonable approach is to use flow triggering in patients receiving intermittent mandatory ventilation and in patients receiving PSV who have increased inspiratory effort during triggering. Either pressure or flow triggering is acceptable in other patients. A trigger sensitivity of -1 to -2 cm H₂O during pressure triggering or -1 to -2 L/minute during flow triggering is appropriate. An initial inspiratory flow rate of 60 L/minute reasonable for most patients. The flow rate can be increased to as needed if a patient appears to be struggling. (See "Overview of initiating invasive mechanical ventilation in adults in the intensive care unit", section on 'Flow rate and pattern' and "Overview of initiating invasive mechanical ventilation in adults in the intensive care unit", section on 'Trigger sensitivity'.)

IMPROVING RESPIRATORY MUSCLE STRENGTH — 

Respiratory muscle weakness is common among mechanically ventilated patients; it may be present at the time of intubation or result from ICU-acquired paresis or ventilator-induced respiratory muscle weakness. As an example, lower diaphragm thickness, measured by ultrasonography within 36 hours of intubation, was associated with prolonged weaning [45]. One study found diaphragm dysfunction, defined by vertical excursion of <10 mm or paradoxical motion during inspiration, in 29 percent of patients ready for weaning. Another study found that 27 percent of patients demonstrated less than a 30 percent increase in diaphragm thickening during inspiration, an indication of dysfunction [46]. Another study reported that 41 percent of patients developed signs of diaphragm atrophy (>10 percent decrease in thickness) by day 4 of mechanical ventilation [47].

Respiratory muscle strength is typically evaluated by clinical examination at the bedside by asking the patient to take a maximal inspiratory effort; weak effort or low lung volumes suggest respiratory muscle weakness. While not routine, more objective bedside measures such as a low negative inspiratory force (eg, <60 cm H₂O) and poor diaphragmatic excursion by ultrasound can be used to support clinical findings. Respiratory muscle strength can also be inferred by measurement of handgrip strength using handheld dynamometry [48]. (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation" and "Tests of respiratory muscle strength".)

Physical therapy is the mainstay of treatment. Inspiratory muscle strength training (IMST) is of unclear benefit and not routinely used. Importantly, methods that improve muscle strength, including physical therapy, typically take days to weeks for improvement to occur. (See 'Investigational strategies (inspiratory respiratory muscle training)' below.)

Physical therapy — Early mobilization/physical therapy in mechanically ventilated patients is supported by the American Thoracic Society and American College of Chest physicians as a preventive measure to promote liberation from mechanical ventilation [38,49]. These data are discussed separately. (See "Post-intensive care syndrome (PICS): Treatment and prognosis", section on 'Early ambulation/physical therapy'.)

Investigational strategies (inspiratory respiratory muscle training) — Inspiratory muscles can be trained to improve strength and endurance so that fatigue occurs less easily [50,51]. However, whether IMST consistently results in earlier liberation from mechanical ventilation remains unproven. Consequently, it is not routinely used.

IMST is performed by adding a resistance device to the inspiratory limb of the ventilator circuit. A systematic review of small randomized trials and observational studies reported that compared with sham or no training, inspiratory muscle training improved maximal inspiratory pressures (increase by -7 cm H₂O), rapid shallow breathing index (+15 breaths/minute/L) (see "Weaning from mechanical ventilation: Readiness testing", section on 'Rapid shallow breathing index') and weaning success (risk ratio 1.34) [50]. Duration of mechanical ventilation improved only in the subgroup with known weaning difficulty. However, there was significant heterogeneity among trials suggesting that larger trials are needed to confirm or exclude this benefit. In a subsequent trial of tracheostomized patients, IMST resulted in improved muscle strength, 60-day survival (70 versus 49 percent), and weaning success (75 versus 45 percent) compared with traditional T-piece weaning [51]. However, the trial was small and had several limitations. A randomized trial of 104 tracheostomized patients found no improvement in weaning success or duration when comparing electronically assisted IMST with spontaneous breathing with T-piece [52].

REFRACTORY PATIENTS — 

Some patients remain difficult-to-wean even though their acute illness and factors contributing to failure to wean have been resolved. In such patients, a tracheostomy is frequently placed and the patient transferred to a long-term acute care facility/hospital (LTAC) where further weaning efforts can be undertaken (ie, prolonged weaning). (See 'Definition and incidence' above.)

Tracheostomy — A tracheostomy can be advantageous in those in whom weaning is expected to be prolonged. The advantages, disadvantages and timing of tracheostomy are discussed in detail separately. (See "Tracheostomy: Rationale, indications, and contraindications", section on 'Rationale for tracheostomy versus endotracheal tube' and "Tracheostomy: Rationale, indications, and contraindications", section on 'Optimal timing in mechanically ventilated patients'.)

Ongoing long-term acute care — The number of patients in specialized weaning units increased threefold between 1997 and 2006, reflecting the trend to manage such patients outside of the ICU [53]. Observational data suggests delaying ventilated patients' discharge to an LTAC may decrease the likelihood of successful weaning [54]. However, one single-institution LTAC ventilator weaning trial reported that as many as one-third of this population could have been successfully liberated from mechanical ventilation prior to transfer to an LTAC [55].

Several factors are considered when determining whether a patient is appropriate for transfer to a specialized weaning center. First, the acute illness should be resolved. Second, the patient should be on stable ventilator settings with fraction of inspired oxygen requirements below 0.5. Third, patients should have a stable airway and route to receive nutrition, which usually consists of a tracheostomy and enteral feeding tube, respectively [56].

LTAC facilities provide a site where weaning and rehabilitation are the primary focus of care. They foster involvement of the patient in decision-making and provide an environment that integrates the family and other caregivers into a supportive health care team. This team includes intensivists, internists, nurses, and respiratory therapists who identify weaning goals and coordinate the weaning process. In addition, there is daily patient-centered rehabilitation that includes physical exercise for regaining muscle strength and practice performing activities of daily living. Skilled therapists are also available to focus on issues such as speech and communication, nutrition support, physical and recreational activities, and counseling. Strategies for weaning patients who are in an LTAC facility for prolonged mechanical ventilation and their outcomes are described in detail separately. (See "Discussing goals of care" and "Management and prognosis of patients requiring prolonged mechanical ventilation in long-term acute care hospitals (LTACH)".)

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Weaning from mechanical ventilation".)

SUMMARY AND RECOMMENDATIONS

●Definition – Patients are considered difficult to liberate from mechanical ventilation if they fail their first spontaneous breathing trial (SBT) and then require up to three SBTs or seven days to pass an SBT. Up to 40 percent of patients mechanically ventilated for an acute illness in the intensive care unit (ICU) are difficult to liberate. (See 'Definition and incidence' above.)

●Correct the cause – Repeat unsuccessful attempts at liberation usually signify incomplete resolution of the illness that precipitated mechanical ventilation and/or the development of one or more new problems that prevent liberation. The clinician should identify and treat these issues (eg, respiratory, cardiac, psychological, circuit, nutritional) (table 1) before resuming further weaning trials. (See 'Identify and correct the causes limiting successful liberation' above.)

●Resume weaning trials – Once the potential causes have been identified and treated, SBTs may resume. SBTs are similar to those undergoing first-time weaning, except special attention should be paid to comfortable posture, airway management, and appropriate ventilation in between trials; in addition, trials are typically longer (up to two hours) and pressure support ventilation without positive end-expiratory pressure (PEEP) or T-piece trials may be preferred in those with cardiac dysfunction. (See 'Trials to liberate patients from mechanical ventilation' above.)

●Respiratory muscle strength training – Early mobilization/physical therapy may be used to maintain strength and potentially promote liberation from mechanical ventilation. However, specific therapeutic strategies for difficult-to-liberate patients have not been studied. Inspiratory muscle training is of unclear benefit and therefore, not typically used. (See 'Investigational strategies (inspiratory respiratory muscle training)' above and 'Improving respiratory muscle strength' above.)

●Refractory patients – Some patients remain difficult to liberate even though their acute illness and factors contributing to failure have been resolved. In such patients, a tracheostomy is frequently placed and the patient transferred to a long-term acute care facility/hospital where further liberation efforts can be undertaken (ie, prolonged weaning). (See 'Ongoing long-term acute care' above and 'Refractory patients' above.)

ACKNOWLEDGMENT — 

The UpToDate editorial staff acknowledges Martin F Joyce-Brady, MD, who contributed to earlier versions of this topic review.