Approach to the patient suddenly impossible to ventilate

The approach to a patient who suffers sudden respiratory failure (or complains of difficulty breathing) while being ventilated via tracheostomy is composed of a series of steps, which are so standardised that they could almost be made into an algorithm or protocol pathway. In a manner similar to a primary survey or the ALS algoritm, one assesses the patient in an orderly fashion, moving from mst lifethreatening issues to least. The first step in most situations is to take the patient off the ventilator and manually ventilate themn, thereby both excluding a major list of differentials and at the same time providing 100% pure oxygen. The pathway for a tracheostomy patient typically leads to the exclusion of a dislodged or blocked tracheostomy, and ends up with the patient being reintubated orally (the dislodged tracheostomy scenario is discussed in the Airway Management section).

The college loves these questions. They have been repeated many times. The patient described in these scenarios is usually tracheostomised, and usually complains that they can't get enough air. Occasionally, patient-centered medicine is left out of it, and the scenario alerts you to the dterioration by way of desaturation and high airway pressures.

• Question 7 from the first paper of 2021 (suddenly impossible to ventilate)
• Question 2 from the second paper of 2019 (a low pressure alarm)
• Question 23 from the second paper of 2016 ("can't get enough air")
• Question 12 from the first paper of 2015 (suddenly impossible to ventilate)
• Question 4 from the second paper of 2012 ("can't get enough air")
• Question 26 from the second paper of 2009 (sudden onset of surgical emphysema)
• Question 27 from the second paper of 2006(suddenly impossible to ventilate)

The most sensible-looking reference for this comes from Jairo Santanilla's "The Crashing Ventilated Patient". The structured approach suggested below has been based on this reference.

Troubleshooting the intubated patient suddenly impossible to ventilate

Immediate management:

• Increase the FiO2 to 100%
• Disconnect from the ventilator, and manually bag-ventilate them.
• Simultaneously assess and manage threats to life in a systematic manner.
• If the lung compliance is good, the patient's ventilator or its tubing is the problem, and you can keep bagging the patient until the ventilator is changed.
• if the bag ventilation is difficult, one must conclude that the patient or the tube are the problem.

If the bag ventilation is easy and the patient improves with it:

• Machine factors are to blame.
• Check the circuit:
  • Check for condensation in the ventilator tubing
  • Change HME 
  • Change the expiratory filter
  • If there is nothing obviously wrong with the tubing, the ventilator may be malfunctioning. Change the ventilator while manually bagging the patient.

If the bag ventilation is difficult and the patient is still unwell:

• Patient factors are to blame.
• Either the airway or the rest of the respiratory system is somehow compromised.
• Address the airway first:
  • In the intubated patient:
    • Is the ETT blocked?
    • Pass a suction catheter down and suction the patient
    • Ensure the patient is not biting the tube.
    • Has the ETT migrated? Is there a cuff herneation?
    • Auscultate both lungs; ensure equal air entry
    • Listen for cuff leak
    • Ensure satisfactory cuff pressure
  • In the tracheostomy patient:
    • check tracheostomy diameter (too narrow?)
    • check inner cannula (encrusted with inspissated secretions?)
    • check tracheostomy patency (blocked with secretions?)
    • Check tracheostomy position (dislodged during last turn?)
    • Check for subcutaneous emphysema
    • Suction the patient, loking for fresh blood and clots (unrecognised pulmonary haemorrhage?)
• Let's say the airway is fine. The rest of the respiratory system must be somehow compromised.  The possibilities include:
  • Bronchial occlusion, eg. by sputum plug or clot
  • Bronchospasm
  • Pulmonary embolism
  • Pulmonary oedema
  • Pleural pathology, eg. pneumothorax, haemothorax or pleural effusion
  • Abdominal pathology, eg. massive distension
• These possibilities need to be investigated systematically:
  • Auscultation of the chest will immediately identify lateralising pathology, and may reveal pulmonary oedema
  • A bedside chest ultrasound will immediately confirm or exclude pneumothorax, haemothorax or large pleural effusion.
  • A bedside TTE should immediately exclude severe LV failure and massive PE.
  • ECG will exclude MI
  • ABG will identify metabolic acidosis
  • CXR to confirm/exclude large bronchus obstruction
  • Bronchoscopy to relieve this mechanical obstruction
• If there is no problem with the respiratory system, but the patient is still "impossible to ventilate", consider the following extrapulmonary possibilities:
  • Patient-ventilator dyssynchrony
  • Pain of respiration (eg. in context of rib fractures or thoracotomy)
  • Increased ventilatory demand:
    • Severe agitation
    • Seizures
    • Fever and rigors
    • Metabolic acidosis

The low pressure alarm

Instead of painfully recapitulating the entire section on ventilator sensors and alarms from the Part One section on mechanical ventilation, one may summarise this issues as follows:

• The low pressure alarm on most ventilators is sounded when the circuit pressure drops below a certain pressure value 
• The alarm is usually set (manually) to about 2cm H₂O below the PEEP value.
• The sensor for this is usually in the afferent (pre-patient) limb of the circuit, and it is involved in a feedback loop with the solenoid flow-limitng valve which contols flow into the circuit
• The ventilator is usually able to compensate for a depressurised circuit by increasing the flow rate.

So, the low pressure alarm has sounded. This can mean several things:

• There is no problem:
  • The pressure sensor is malfunctioning
  • The pressure sensor is working fine, but somebody has either set the alarm threshold too high, or has decreased the PEEP without also adjusting the alarm.
• There is a real probem:
  • The problem is with the circuit
  • The problem is with the patient
  • The ventilator has failed in some fundamental way and is not producing flow (eg. in a turbine-driven ventilator, the turbine has failed)
  • The flow is being "stolen" by the suction set

It's impossible to determine which one it is without doing a bit of troubleshooting. If there is a real problem, it could be:

• Patient problem:
  • Leak could be around the ETT
    • The tube has migrated out past the cords, and now and there is a massive leak
    • The cuff has deflated, and now there is a cuff leak
    • The ETT itself is damaged, and now there is a leak out of the defect (this is rare, as it takes quite a lot of effort to damage those things).
  • Leak could be via the patient:
    • Large volume air leak via the ICC
• Circuit problem:
  • Disconnection of the circuit from the ventilator
  • Disconnection of the patient from the circuit
• Failure of the ventilator

So, the best way to approach this:

1) Ensure the patient is safe:

• Disconnect the patient from the ventilator and the inline suction set, and bag them manually, noting the pressure generated thereby
• Note whether there is an audible leak; if yes, then the endotracheal tube is to blame
• Assure that the patient's oxygenation is adequate
• Assure that the end-tidal CO₂ is still generating a waveform (i.e. the tube is not displaced into the oesophagus)

3) Systematically troubleshoot the circuit

• Start with the patient
  • Listen to the chest: is there air entry?
  • ETT:
    • Make sure it is at the same depth as was previously documented
    • Check the cuff pressure with the manometer
    • Check for cuff leak
    • Perform laryngoscopy if there is any doubt
  • Check the chest drain: is there now a vigorous air leak?
    May need a chest Xray
  • Check the ventilator: while bagging the patient manually, see if the problem is reproduced with the same circuit and a "test lung".
  • Check the closed suction set: did the trigger valve become disabled by some accident (i.e. is the suctioin constantly "stealing" airway pressure?)

The low pressure alarm, though not as sexy a topic as the high pressure alarm, is explored in the excellent article by Raphael (1999), which appears to be the only usueful thing written on the subject. Though the article is specifically designed for troubleshooting circle circuits, Raphael's algorithm is reproduced below to stimulate some thought.