Weaning from Mechanical Ventilation — SBT, Protocols & Prolonged Weaning

Overview & definition

Weaning (liberation) from mechanical ventilation is the process of discontinuing ventilatory support. It has three steps: a daily readiness screen, a spontaneous breathing trial (SBT), and extubation if the trial is tolerated. Around 70 per cent of patients wean easily after a single SBT; a minority wean with difficulty and need the underlying cause addressed.[1][1]

Weaning classification — simple, difficult, prolonged

The WIND (Weaning according to a New Definition) classification is the modern framework used to stratify weaning. Patients fall into one of three groups according to the number of failed SBTs and the time from the first SBT to successful extubation:[1]

Why the 70/20/10 split matters for the exam

Simple (70 per cent), difficult (20 per cent), prolonged (10 per cent) — this is the single most testable number in weaning. The exam wants the candidate to recognise that the vast majority of patients wean easily with a protocolised daily screen and SBT, and that the difficult and prolonged groups are where the mark is won or lost. The candidate is expected to recite the causes of difficult weaning (load-versus-capacity imbalance) and to discuss a structured approach: address the cause, repeat SBTs, avoid SIMV weaning, consider tracheostomy for prolonged weaning, and recognise weaning-induced cardiac failure.[1][1]

Step 1 — the daily readiness screen

Before each SBT, screen for readiness:[1]

• The underlying cause of respiratory failure is resolving.
• Adequate oxygenation — a PaO2/FiO2 over about 150-200, PEEP under about 5-8 cmH2O, FiO2 under about 0.4-0.5.
• Haemodynamically stable — no or low-dose vasopressors, no active myocardial ischaemia.
• Awake and cooperative — able to follow commands, initiate inspiration, and cough.
• No new serious problem (no ongoing deep sedation, no severe acidosis).

A daily screen, combined with a daily sedation interruption (a spontaneous awakening trial), is the foundation of efficient weaning.[1]

The daily readiness screen is a filter — its job is to safely and quickly identify patients who can attempt an SBT without missing anyone capable of breathing. The screen has positive criteria (must be present) and negative criteria (must be absent). A protocolised screen, run by nurses or respiratory therapists, is faster than waiting for a doctor and reduces ventilator days.[5]

[1]

The cuff leak, cough, and gag — airway protection before the trial

A readiness screen that ignores the airway is incomplete. Before extubating you must confirm three things: (i) cuff leak is adequate (no laryngeal oedema — see the extubation section below); (ii) cough peak flow is at least 60 L/min (the patient can clear secretions — a strong cough on suctioning is the bedside surrogate); and (iii) gag and swallow are intact so the patient can protect the airway from aspiration. A weak cough (peak flow under 60 L/min, Smina Chest 2003) and copious secretions predict extubation failure independent of the SBT result.[11]

The screen + SBT takes a few minutes to set up and an hour to run; it should be a daily routine, not a special event. A patient who passes is extubated; a patient who fails goes back on support, the cause is addressed, and the trial is repeated the next day. [1]

Step 2 — the spontaneous breathing trial (SBT)

The SBT tests whether the patient can breathe with minimal support:[1]

• Method — a T-piece (no ventilatory support), or low pressure support (5-7 cmH2O) with PEEP 5, for 30-120 minutes.
• Tolerance (pass) — respiratory rate under 35, SpO2 over 90 per cent, heart rate under 140 (and not changing by more than 20 per cent), no respiratory distress (no accessory-muscle use, no diaphoresis, no agitation), stable blood pressure, and a stable gas exchange.
• Failure — a rising respiratory rate, falling SpO2, distress (accessory muscles, diaphoresis, agitation), haemodynamic instability, or arrhythmia — resume ventilatory support and address the cause.

A patient who tolerates the SBT proceeds to extubation (after confirming airway protection — see the extubation topic).[1]

The ideal SBT loads the respiratory muscles with the work they will face after extubation (an endotracheal tube adds resistive work, so the post-extubation work is similar). Three techniques are accepted and roughly equivalent in trial populations; the T-piece most closely mimics unassisted breathing, while PSV 5-7 cmH2O + PEEP 5 partly compensates for the resistance of the endotracheal tube.[3][4]

SBT duration — 30-120 minutes

The standard duration is 30 to 120 minutes. Subirà (JAMA 2019) randomised patients to a 30-minute vs a 120-minute SBT and found the shorter trial was non-inferior for successful extubation — there is no benefit in prolonging the trial in patients who are passing. The corollary: a patient who is going to fail usually does so within the first 20-30 minutes (rising rate, falling saturation, distress). A trial of 120 minutes is reserved for the difficult or chronically ventilated patient where longer observation is informative.[12]

Step 3 — weaning protocols and the ABC trial

Protocolised weaning (a nurse- or respiratory-therapist-driven daily screen and SBT) reduces the duration of mechanical ventilation compared with physician-directed weaning, because it removes the delay of waiting for a doctor to assess the patient.[1]

The ABC trial (Awakening and Breathing Controlled, Girard, Lancet 2008) paired a daily sedation interruption (a spontaneous awakening trial) with a daily SBT. The combined protocol reduced ventilator-free days and improved outcomes (with no excess self-extubation), establishing the paired SAT plus SBT as standard.[1]

The evidence for protocolised weaning is strong. Ely (NEJM 1996) randomised patients to a nurse/therapist-driven protocol versus physician-directed weaning and found the protocol group was extubated a median of 1.5-2 days earlier, with no increase in reintubation. The mechanism is simple: doctors round once or twice a day; nurses are at the bedside continuously, so a protocolised screen + SBT removes the decision-to-extubate delay.[5]

The ABC trial — pairing sedation interruption with the SBT

The Awakening and Breathing Controlled trial (Girard, Kress, Fuchs; Lancet 2008) was a multicentre randomised trial pairing a spontaneous awakening trial (SAT, a daily sedation interruption) with a spontaneous breathing trial (SBT) versus usual care with sedation directed by the clinician. The combined protocol increased ventilator-free days, ICU days, and 28-day survival, with no excess self-extubation or reintubation. The principle: the patient must be awake enough to breathe before you can test whether they can breathe. This pairing is now the global standard.[1]

Daily sedation interruption — the Kress evidence

Kress (NEJM 2000) showed that a daily interruption of sedative infusions (until the patient was awake) halved the duration of mechanical ventilation and the length of ICU stay, with no excess psychological harm or agitated delirium. It is the single most effective, cheapest weaning intervention. Pair it with the SBT (the ABC trial) for the maximum effect.[6]

The rapid shallow breathing index (RSBI)

The RSBI = respiratory rate divided by tidal volume in litres, measured during a brief unsupported breath. An RSBI under 105 predicts SBT success; over 105 predicts failure (shallow, rapid breathing — the pattern of fatigue). It is a guide, not absolute, and is used alongside the clinical SBT.[1]

How the RSBI was derived and how to measure it

Yang & Tobin (NEJM 1991) prospectively measured a battery of physiological indices in ventilated patients being tested for weaning, and asked which one best predicted success or failure. The f/VT ratio (respiratory frequency divided by tidal volume in litres) outperformed every other index — a threshold of 105 breaths/min/L gave a sensitivity and specificity of ~0.80 for SBT success. The physiology: as the diaphragm fatigues, the patient adopts a rapid, shallow breathing pattern (high rate, small tidal volume) — a textbook fatigue pattern. To measure it, place the patient on a T-piece or CPAP 5 (no pressure support) for one minute, then read the rate and the exhaled Vt (in litres) from the ventilator.[2]

Pitfalls of the RSBI

The RSBI is a guide, not absolute. It over-estimates failure (a falsely high value) when the endotracheal tube is narrow or partly blocked (high resistance inflates the rate), and during the first minute of a trial before the patient settles. It under-estimates failure when the cuff is leaking (Vt measured by the ventilator is spuriously low, lowering the RSBI). Use it alongside the clinical SBT — a patient with an RSBI of 100 who is in distress has failed; a patient with an RSBI of 120 who is comfortable and passing the trial can still be extubated.[2]

Other predictors of weaning outcome

The RSBI is the headline index, but a modern difficult-weaning assessment adds:

• Maximal inspiratory pressure (MIP or NIF) — under -20 to -30 cmH2O suggests adequate reserve; over -20 (i.e. weaker) predicts failure.
• Cough peak flow — under 60 L/min predicts extubation failure (Smina, Chest 2003).[11]
• Diaphragmatic thickening fraction on ultrasound — under 30-36 per cent on M-mode predicts failure; ultrasound is increasingly used at the bedside.
• Lung ultrasound score — a high B-line count predicts weaning-induced pulmonary oedema.
• Cardiac biomarkers (BNP/NT-proBNP) — a rise during the SBT predicts weaning-induced cardiac failure.

Failed SBT — explicit criteria and what to do

The failed SBT is defined by any one of the following during the trial:[1]

When the SBT fails, resume the previous ventilatory support immediately and address the cause. Document the mode of failure — respiratory (rate, saturation), cardiac (rate, BP, echo), or neurological (agitation, sedation). Each failure should teach you something about the dominant barrier, and the next day's plan should target it. Repeat the SBT daily (or every other day in the prolonged weaner); do not abandon daily trials because the patient has failed a few times.[1]

Difficult and prolonged weaning

Weaning difficulty is categorised:[1]

• Simple — passes the first SBT and is extubated.
• Difficult — up to three failed SBTs, or up to seven days from the first SBT.
• Prolonged — more than three failed SBTs, or more than seven days.

The fundamental problem in difficult weaning is a respiratory load that exceeds the capacity of the respiratory muscles. The causes:[1]

• Reduced capacity (respiratory-muscle weakness) — critical-illness polyneuromyopathy, prolonged neuromuscular blockade, malnutrition, and electrolyte disturbance (hypophosphataemia, hypokalaemia, hypomagnesaemia).
• Increased load — obstructive disease (COPD with high resistive load and dynamic hyperinflation), pneumonia and copious secretions, pulmonary oedema, obesity and chest-wall restriction.
• Reduced drive — over-sedation, central nervous system depression.
• Cardiac — weaning-induced cardiac failure: resuming spontaneous breathing increases venous return (preload) and left-ventricular afterload (the loss of positive intrathoracic pressure), unmasking occult LV dysfunction and causing pulmonary oedema (suspect a patient who fails the SBT with a rising respiratory rate and crackles; diagnose with an echo during a trial).
• Metabolic and psychological — acidosis, delirium, anxiety.

The framework to keep in your head is the load-versus-capacity balance: the respiratory muscles pump air against a load (resistance, compliance, secretions). When capacity falls (weakness) or load rises (obstruction, oedema), the pump fails — and the patient weans slowly or not at all. A separate axis is drive: even with adequate capacity and load, over-sedation abolishes the central drive to breathe. Diagnose the dominant axis at each failed SBT.[1]

Management of difficult weaning

Address the cause:[1][1]

• Muscle weakness — nutrition, correct electrolytes (phosphate, potassium, magnesium), avoid prolonged paralysis, early mobility.
• Increased load — bronchodilate obstruction, treat infection and clear secretions, diurese pulmonary oedema.
• Reduced drive — reduce sedation (daily SAT).
• Cardiac failure — diurese, afterload reduction, treat ischaemia; an echo during a failed SBT confirms weaning-induced pulmonary oedema.
• Weaning strategy — repeat daily SBTs (the preferred approach); gradual pressure-support weaning as an alternative. Avoid SIMV weaning, which prolongs weaning. For prolonged weaning, a tracheostomy reduces dead space and eases secretion clearance and weaning (timing does not change mortality, per TracMan).
• Treat delirium, anxiety, and optimise sleep.[1]

Extubation — cuff leak, post-extubation stridor, reintubation

Passing the SBT is necessary but not sufficient for extubation. The airway must be safe: there must be no upper-airway obstruction (laryngeal oedema), and the patient must protect the airway (cough, gag, swallow). Two bedside tests answer these questions: the cuff leak test and the cough on suction. [1]

The cuff leak test — predict laryngeal oedema

The cuff leak is the difference between the inspired and the expired tidal volume when the cuff is deflated. A small cuff leak (under 110 mL, or under 25 per cent of the delivered tidal volume) suggests that the tube fills the trachea snugly because the surrounding laryngeal and tracheal tissues are oedematous — predicting post-extubation stridor. Risk factors for a low cuff leak include a traumatic or difficult intubation, prolonged intubation (over 7 days), large-bore tube, female sex, and active cuff infection. A patient at high risk of post-extubation stridor should receive prophylactic steroids (methylprednisolone 20-40 mg every 4-6 hours for 4 doses, starting 12-24 hours before extubation), which reduce the rate of reintubation for stridor.[15]

[1]

Post-extubation stridor

Post-extubation stridor is inspiratory stridor within 24-48 hours of extubation caused by laryngeal oedema (or, less commonly, bilateral vocal-cord paralysis). It occurs in 2-15 per cent of extubations and carries a higher reintubation and mortality rate. Risk factors (Jaber, CCM 2003): female sex, prolonged intubation (over 36-48 hours), traumatic or repeated intubation, large tube, high cuff pressure, and aggressive suctioning.[15]

Management:

• Racemic adrenaline nebulised (1:1000 adrenaline 5 mL) — topical vasoconstrictor, fast onset.
• Dexamethasone IV (systemic anti-inflammatory; 4-8 mg) — onset over hours, treats laryngeal oedema.
• Heliox (helium-oxygen mixture) — lower density reduces turbulent-flow work of breathing across the obstruction; a temporising bridge.
• Non-invasive ventilation — CPAP/PSV stents the upper airway open; useful if the patient is tiring but not in failure.
• Reintubate early — a worsening stridor with rising respiratory rate is a reintubation indication; do not wait for a respiratory arrest. Use a smaller tube this time. [1]

Reintubation — 10-20 per cent of extubations

Reintubation rates run 10-20 per cent overall, and higher in medical and neurology ICUs. Reintubation within 48-72 hours is an extubation failure; the cause is usually one of: upper-airway obstruction (stridor), inadequate cough or secretion clearance, encephalopathy with loss of airway protection, or residual cardiac/respiratory failure. A failed extubation carries a markedly higher mortality (some studies show mortality 2-5 times higher than successful extubation), partly because reintubation is itself hazardous (aspiration, haemodynamic instability, desaturation) and partly because failure marks a sicker patient.[8][9]

The cuff leak, the cough, and the swallow — the airway-protection bundle

Before pulling the tube, complete the airway bundle: (1) cuff leak over 110 mL (or a positive leak after steroids); (2) strong cough on suctioning (peak cough flow over 60 L/min); (3) gag reflex intact and the patient able to swallow secretions. A patient who passes the SBT but fails the airway bundle (weak cough, absent gag, copious secretions) will likely fail extubation — leave the tube in, address the cause (suction, treat delirium, mobilise), and reassess daily.[11]

[1]

Tracheostomy for prolonged weaning

For the prolonged weaner (over three failed SBTs, over seven days), a tracheostomy reduces dead space, lowers the work of breathing, eases secretion clearance, and is more comfortable than an endotracheal tube (the patient can eat, speak, mobilise). The TracMan trial (NEJM 2013) compared early (day 1-4) versus late (day 10+) tracheostomy and found no difference in mortality or ventilator-free days — so the timing is driven by patient comfort and expected duration of ventilation rather than a hard deadline. A percutaneous dilatational tracheostomy at the bedside (Ciaglia technique) is the standard approach in ICU. Decannulation follows once the patient tolerates a speaking valve and a capped tracheostomy on room air for 24-48 hours.[7]

The evidence base — landmark trials

[1]

High-yield points for the CICM/FFICM/EDIC exam

[1] [1] [1]

Red flags

References

1. [1]Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial Lancet, 2008.PMID 18191684
2. [2]Yang KL, Tobin MJ. In vitro release of growth hormone-releasing factor (GRF) from the hypothalamus: somatostatin inhibits GRF release Regul Pept, 1991.PMID 1675472
3. [3]Esteban A, Frutos F, Tobin MJ, et al. Marijuana, the AIDS wasting syndrome, and the U.S. government N Engl J Med, 1995.PMID 7637743
4. [4]Brochard L, Rauss A, Benito S, et al. Ozone dose and effect in humans and rats. A comparison using oxygen-18 labeling and bronchoalveolar lavage Am J Respir Crit Care Med, 1994.PMID 8087337
5. [5]Ely EW, Baker AM, Dunagan DP, et al. Persistent hepatitis C viremia predicts late relapse after sustained response to interferon-alpha in chronic hepatitis C. TriVeneto Viral Hepatitis Group Ann Intern Med, 1996.PMID 8633819
6. [6]Kress JP, Pohlman AS, O'Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation N Engl J Med, 2000.PMID 10816184
7. [7]Young D, Harrison DA, Cuthbertson BH, Rowan K; TracMan Collaborators. Start codon targeted (SCoT) polymorphism in toxic and non-toxic accessions of Jatropha curcas L. and development of a codominant SCAR marker Plant Sci, 2013.PMID 23602106
8. [8]Ferrer M, Esquinas A, Arancibia F, et al. Complement receptor 1 gene polymorphisms are associated with idiopathic pulmonary fibrosis Am J Respir Crit Care Med, 2003.PMID 12773320
9. [9]Esteban A, Frutos-Vivar F, Ferguson ND, et al. Racial variations in causes of vision loss in nursing homes: The Salisbury Eye Evaluation in Nursing Home Groups (SEEING) Study Arch Ophthalmol, 2004.PMID 15249367
10. [10]Schweickert WD, Pohlman MC, Pohlman AS, et al. Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial Lancet, 2009.PMID 19446324
11. [11]Smina M, Salam A, Khamiees M, Gada P, Amoateng-Adjepong Y, Manthous CA. Left ventricular hypertrophy is a common echocardiographic abnormality in severe obstructive sleep apnea and reverses with nasal continuous positive airway pressure Chest, 2003.PMID 12907548
12. [12]Subirà C, Hernández-Tejedor A, Delgado M, et al. Association of Changes in Air Quality With Incident Asthma in Children in California, 1993-2014 JAMA, 2019.PMID 31112259
13. [13]Dres M, Schmidt M, Ferre A, Demoule A. [Ethical principles in psychiatric action] Psychiatr Prax, 2014.PMID 24983582
14. [14]Ferrer M, Sellares J, Valencia M, et al. Estimation of Potential Availability of Essential Oil in Some Brands of Herbal Teas and Herbal Dietary Supplements PLoS One, 2015.PMID 26110869
15. [15]Jaber S, Chanques G, Matecki S, et al. A germin-like protein of wheat leaf apoplast inhibits serine proteases J Exp Bot, 2003.PMID 12709479