What to Do If a Patient Self-Extubated

When a patient pulls out their own endotracheal tube, the first priority is stabilizing the airway and assessing whether the patient can breathe adequately on their own. Roughly 60% of patients who self-extubate do not need to be reintubated, which means the event, while alarming, does not automatically require emergency re-securing of the airway. The immediate clinical response, the decision about reintubation, and the longer-term prevention strategies all rest on a body of evidence that has grown considerably over the past two decades.

How Common Self-Extubation Actually Is

Self-extubation is not rare. Across large ICU databases, it accounts for the majority of unplanned extubation events. A prospective multicenter study of French ICUs found a pooled unplanned-extubation rate of about 1.0 per 100 ventilator days, with self-extubation making up 88% of those events.¹ A large French network analysis covering over 12,000 mechanically ventilated patients reported a similar rate of 0.81 self-extubation events per 100 ventilator days, with about 5.6% of all ventilated patients experiencing the event.² A case-control study at a single center documented a slightly higher rate of 1.3 per 100 ventilator days and noted that the risk was greatest during the first two days of mechanical ventilation, especially in agitated patients.³

These numbers mean most busy ICUs will see self-extubation regularly. Treating it as a rare aberration rather than a predictable event is one of the reasons prevention programs often fall short.

The First Minutes After the Tube Comes Out

The immediate response follows the same logic as any acute airway emergency, but calibrated to the fact that most self-extubated patients are not in cardiac arrest. The clinical team should move through a rapid sequence of assessments and interventions:

• Call for help: Alert the physician and respiratory therapist immediately. Even if the patient looks stable at first glance, you want intubation equipment and personnel at the bedside in case things deteriorate.
• Position and assess: Sit the patient upright if possible. Look at respiratory rate, chest movement, oxygen saturation, and level of consciousness. Listen for stridor, which suggests laryngeal swelling or injury.
• Provide supplemental oxygen: Apply a high-flow nasal cannula or non-rebreather mask right away while you assess. The goal is to keep saturation above 90% while the team decides on next steps.
• Suction the airway: The patient may have secretions pooled above the cuff site. Oropharyngeal suctioning can prevent aspiration and help you see whether the patient is managing their own airway.
• Check the tube: Look at the removed tube. If the cuff is still inflated and intact, the tube came out cleanly. If the cuff is torn or a piece is missing, there may be a foreign body in the airway, which changes the urgency dramatically.
• Secure IV access and prepare for reintubation: Have medications for rapid-sequence intubation drawn up. Even if the patient appears to be tolerating extubation, respiratory failure can develop over the next several hours.

The mistake to avoid in these first minutes is reflexively reintubating every patient. That said, the opposite mistake, assuming everything is fine because saturations look decent initially, is equally dangerous. The safest approach is to prepare for reintubation while watching closely for signs that the patient can sustain independent breathing.

Who Needs Reintubation and Who Does Not

The reintubation rate after self-extubation sits in the range of roughly 30–45% in most studies. The large French network analysis found that about 39% of patients required reintubation within 48 hours.² An earlier study reported 44% reintubation in a cohort of 100 unplanned extubations.⁴ Older data from a 12-month review found a 31% reintubation rate overall, dropping to just 11% among patients whose self-extubation appeared deliberate rather than accidental.⁵

That last finding is worth pausing on. Some patients pull out their tubes because they are genuinely ready to be liberated from the ventilator. The tube is uncomfortable, they are alert, and from a physiological standpoint they were close to meeting extubation criteria already. These patients tend to do well. The patients who self-extubate while confused, deeply sedated, or in the middle of high-support ventilation are the ones who run into trouble.

Several clinical factors have been identified as predictors of needing reintubation. A landmark study found that four or more of the following factors together predicted reintubation with about 92% accuracy: the patient was on a controlled ventilator mode with a set rate above 6 breaths per minute; their most recent arterial pH was 7.45 or higher; their oxygen-to-FiO2 ratio was below 250; their peak heart rate in the preceding 24 hours exceeded 120; they had three or more coexisting medical conditions; their mental status was not fully alert; and their original reason for intubation was something other than a planned surgical procedure.⁶ A separate Korean study confirmed that medical (as opposed to surgical) patients, those on sedatives, those with copious secretions, and those with a low oxygen ratio were at elevated risk of needing reintubation.⁷

In practical terms, the bedside clinician should quickly assess: Was this patient already being weaned? Are they alert? Can they protect their airway, cough, and handle secretions? Is their oxygenation holding? If the answers trend positive, close observation with supplemental oxygen is reasonable. If the patient was on high ventilator settings, is not fully conscious, or is deteriorating, early reintubation avoids a more chaotic emergency intubation later.

Why Reintubation Matters for Outcomes

Patients who require reintubation after self-extubation have substantially worse outcomes than those who tolerate it. In one study, reintubated patients had longer ICU and hospital stays, longer total durations of mechanical ventilation, higher rates of ICU-acquired infections, and higher mortality compared with patients who did not need reintubation.⁴ This is not entirely because reintubation itself is harmful, though it does carry risks. It largely reflects the fact that patients who cannot sustain their own breathing were sicker to begin with. Still, the act of reintubation can introduce new complications: aspiration during the intubation attempt, airway trauma, and a fresh period of mechanical ventilation with all its attendant risks.

Bridging With Non-Invasive Support

Not every patient who struggles after self-extubation needs to go straight back on the ventilator. Non-invasive ventilation and high-flow nasal cannula can bridge the gap for patients who are borderline. A randomized trial in patients at very high risk for extubation failure found that non-invasive ventilation with active humidification led to a reintubation rate of about 23%, compared with roughly 39% for high-flow nasal cannula alone.⁸ That study was designed for planned extubations, but the principle applies: for a patient who is borderline after self-extubation, non-invasive support buys time and may prevent the need for reinstrumentation of the airway.

The caveat is that non-invasive support should not be used to delay a clearly necessary reintubation. If the patient is obtunded, cannot protect their airway from aspiration, or has significant upper-airway swelling, non-invasive support is not appropriate and will only waste precious time.

Complications Beyond the Airway

The most immediate risk after self-extubation is airway compromise, but complications can extend further. Pulling out a tube with an inflated cuff can damage the larynx. A retrospective study of self-extubation injuries found that hoarseness was the most commonly reported symptom, present in about 62% of cases, and posterior laryngeal edema was the most common physical finding, seen in roughly 58%.⁹ Vocal cord injuries, arytenoid dislocation, and tracheal mucosal tears are less common but can occur, particularly in patients who are agitated and yank the tube forcefully.

Beyond the throat, the loss of positive-pressure ventilation can trigger rapid desaturation, aspiration of pooled secretions, and hemodynamic instability. In the perioperative period especially, cardiovascular collapse and cardiac arrest are recognized risks when tube displacement goes unnoticed.¹⁰

Screening for Swallowing Problems

One often-overlooked step after any extubation, planned or not, is assessing the patient’s ability to swallow safely. Prolonged intubation weakens the muscles of the throat and disrupts the coordinated swallowing reflex, leaving patients vulnerable to aspiration. A retrospective study found that nurse-performed bedside screening for swallowing difficulty after extubation was associated with a large reduction in post-extubation pneumonia and an increase in the likelihood of resuming oral feeding by ICU discharge.¹¹ This screening is particularly important when a patient self-extubates, because the event is uncontrolled and the larynx may already be swollen or injured. Keep the patient NPO (nothing by mouth) until you can assess swallowing, or until a formal speech-language pathology evaluation can be arranged if the patient was intubated for more than a couple of days.

Agitation and Delirium as the Root Cause

Understanding why a patient self-extubated is essential for preventing it from happening again if they are reintubated. Agitation and delirium are overwhelmingly the leading drivers. A case-control study found that the lowest levels of sedation on the Ramsay scale (meaning the most awake and agitated patients) were associated with a 25- to 30-fold increase in the odds of unplanned extubation.¹² A separate retrospective cohort found that agitation alone carried roughly a 9-fold risk of self-extubation.¹³

This puts clinicians in a bind. Modern ICU practice has moved toward lighter sedation and daily spontaneous-awakening trials because deep sedation carries its own risks, including longer ventilation times, more delirium, and higher mortality. A well-known trial pairing daily awakening with spontaneous breathing trials found that the intervention group had more self-extubations, but critically, the number of patients who actually required reintubation afterward was similar between the groups, and total reintubation rates were equivalent.¹⁴ In other words, lighter sedation led to more tube removals, but many of those patients turned out to be ready for extubation anyway. The concern among bedside nurses about patient safety during sedation interruptions is real and well-documented, but the data suggest that the strategy’s net benefits outweigh the increased self-extubation events.¹⁵

The Physical Restraints Debate

Wrist restraints are widely used in ICUs with the explicit goal of preventing tube removal. The evidence that they actually work is surprisingly thin. An integrative review found that while physical restraints are the first choice for preventing treatment interference and most clinicians believe in their value, their effectiveness in preventing self-extubation is questionable.¹⁶ A more recent narrative review echoed this, noting that high-quality evidence, including the R2D2-ICU trial, has challenged the assumption that reducing restraints leads to more tube removals, while confirming that restraints are associated with a range of negative physical and psychological outcomes.¹⁷

A qualitative study exploring why restraints persist despite weak evidence uncovered a compelling pattern of reasoning. Nurses tended to interpret events in a way that always reinforced restraint use: if restraints were applied and the patient did not self-extubate, the restraints got the credit; if the patient self-extubated despite restraints, nurses blamed the technique rather than questioning whether restraints work; and if restraints were removed and the patient pulled the tube, nurses blamed themselves for removing them.¹⁸ This kind of confirmation bias is understandable given the high stakes, but it means clinical teams should not assume restraints are a reliable safeguard. They may be part of a prevention strategy, but they are not a substitute for adequate sedation management, frequent reassessment, and proper tube securement.

How the Tube Is Secured Makes a Difference

One of the more actionable prevention measures is ensuring the endotracheal tube is well-secured in the first place. A comparison of a commercial tube holder versus adhesive tape found that tape allowed the tube to slide an average of 3.4 cm under standardized tension, whereas the holder limited movement to just 0.3 cm. Nearly all patients (97%) experienced clinically significant tube movement with tape, compared with only 3% with the holder.¹⁹ A randomized controlled trial confirmed this in clinical practice: tube dislodgement occurred in about 4% of patients with a commercial fastener versus roughly 10% of those secured with tape.²⁰ A meta-analysis pooling several studies also showed a trend toward fewer dislodgements with commercial holders, though the result did not quite reach statistical significance.²¹

In practical terms, if your unit still uses tape, consider whether a switch to a commercial holder is feasible. The difference in tube movement under tension is substantial, and the clinical trial data lean in favor of the holders, especially for patients who are agitated or who tend to sweat or drool enough to loosen adhesive.

Staffing and Surveillance

Self-extubation is more likely on understaffed shifts. When fewer nurses are present, there is less chance that someone notices a patient working free of their restraints or becoming restless before the tube comes out.²² This is not the kind of finding that any individual clinician can fix at the bedside, but it is an important piece of the puzzle for unit leaders and hospital administrators. Adequate nurse-to-patient ratios, particularly during night shifts and weekends when self-extubation tends to cluster, are a structural prevention strategy that no bundle of bedside interventions can replace.

Pediatric and Neonatal Self-Extubation

Children, and especially neonates, present a distinct challenge. The trachea of a newborn is short enough that even small amounts of tube movement can result in dislodgement. Neonatal ICUs also tend to use lighter sedation than pediatric or cardiac ICUs, and the combination of a short airway and a relatively awake patient makes neonates especially vulnerable to unplanned extubation events.²³ Prevention bundles that include standardized tube securement, daily assessment of tube position, and team huddles focused on high-risk patients have been shown to reduce unplanned extubation rates in pediatric settings. The JAMA Pediatrics study evaluating one such bundle found meaningful reductions, especially when the bundle achieved high compliance.

The reintubation decision in neonates is also more urgent because these patients have smaller oxygen reserves and can desaturate within seconds rather than minutes. In a pediatric self-extubation event, the threshold for reintubation is typically lower than in adults.

Traumatic Brain Injury and Self-Extubation

Patients with head injuries present a specific scenario. Altered consciousness from the injury itself interacts with sedation in unpredictable ways, and agitation during the recovery phase is common. A study of over 300 head-injury patients on mechanical ventilation found that 12% experienced self-extubation, and about 31% of those required reintubation.²⁴ The reintubation rate is comparable to the general ICU population, but the stakes are higher: hypoxia in the setting of a brain injury can cause secondary neurological damage that compounds the original insult. For this patient population, airway management after self-extubation demands especially rapid assessment and a low threshold for re-securing the airway.

Building a Prevention Culture

Individual interventions, whether better tube holders, sedation protocols, or staffing adjustments, all help, but the strongest results come from bundled quality-improvement programs that combine multiple strategies and embed them in the daily workflow. An overview of systematic reviews found that root-cause analysis programs, where each self-extubation event triggers a structured review of what went wrong and what could be changed, reduced unplanned extubation rates over time.²⁵ These programs work because they treat self-extubation as a systems failure rather than a patient behavior problem. When the question shifts from “why did this patient pull out their tube” to “what in our workflow allowed this to happen,” the solutions become structural and sustainable. Documenting each event, identifying patterns (time of day, staffing level, sedation depth, tube securement method), and testing countermeasures in a plan-do-study-act cycle is how units achieve lasting reductions rather than temporary vigilance spikes after a bad event.