Anticipate, optimize, recognize failure, rescue, and escalate. A working mental model for the ICU airway, built around the DAS 2025 algorithm and the Vortex approach.
You pushed the meds. You inserted the blade. You cannot pass the tube. SpO2 is falling. What happens next depends on which of three outcomes you are in, not on trying the same thing harder.
Plan A works
Video laryngoscopy with a stylet, tube in, done. This happens roughly 85% of the time with VL.
Plan A fails but you can oxygenate
You have time. Stop. Think. Regroup.
Nothing works and you cannot oxygenate
Cannot intubate, cannot oxygenate (CICO). The emergency inside the emergency.
The Difficult Airway Society (DAS) 2025 guidelines keep a linear structure across four plans:
| Plan | Approach |
|---|---|
| A | Tracheal intubation |
| B | Supraglottic airway (SGA) ventilation |
| C | Face mask ventilation |
| D | Emergency front of neck airway (eFONA) |
Maximize first attempt success. Do not just manage failure after it happens.
The best time to identify a difficult airway is before you push the meds. The MACOCHA score (0 to 12) was built specifically for the ICU, not the OR.
| Factor | Points |
|---|---|
| Mallampati III or IV | 5 |
| Obstructive sleep apnea | 2 |
| C-spine limitation | 1 |
| Mouth opening < 3 cm | 1 |
| Coma, GCS < 8 | 1 |
| Hypoxemia, SpO2 < 80% | 1 |
| Non-anesthesiologist operator | 1 |
A score of 3 or higher predicts a difficult airway. Verbalize it. Prepare for it.
If MACOCHA is high and CICO risk is high, consider awake intubation.
Make the first attempt your best attempt. DAS 2025, Gottlieb (BMJ 2026), and Freund (Intensive Care Medicine 2026) all agree that video laryngoscopy is first line for most intubations, not just the anticipated difficult ones.
Optimize everything before the blade goes in:
Each failed attempt roughly doubles the complication rate. First pass success is the goal. Everything after is rescue.
A 2026 Australia and New Zealand survey of 211 ICU clinicians (Toolis et al.) found a gap between equipment availability and consistent execution.
We have the equipment. We have the evidence. We have the guidelines. What is missing is consistent implementation.
First attempt failed. SpO2 is holding. Do not immediately reinsert the blade. Ask three questions.
Can I oxygenate?
BVM or SGA. If yes, you have time.
Why did it fail?
Poor view? Could not pass the tube? Secretions?
What will I change?
Different blade, bougie, or a different operator.
Maximum of 3 total intubation attempts. Each attempt must be different from the last. After 3, declare failed intubation and move to Plan B.
Do not repeat the same technique hoping for a different result. That is perseveration. That is how people die.
Failed intubation but you can oxygenate: place an SGA. It is your bridge. It buys time, restores oxygenation, and gives you options.
Second generation SGAs (i-gel, LMA Supreme) are preferred for their better seal and gastric drainage port.
SGA has been Plan B, ahead of face mask, since the DAS 2015 guideline, reflecting its reliability as a rescue device. The 2025 update keeps that order.
The SGA is not a failure. It is a lifeline. Use it.
If the SGA fails, optimize face mask ventilation.
Can you oxygenate via face mask?
Yes: you have time. Regroup, call for help, and consider intubation through the SGA with a bronchoscope, an awake approach, or elective FONA.
No: you are in a CICO emergency. Declare it. Say the words out loud: "This is a CICO emergency."
This is the fork in the road. One path has time. The other does not. Know which one you are on.
Cannot intubate, cannot oxygenate. Emergency front of neck airway. Three things happen simultaneously.
Declare CICO out loud
Everyone in the room needs to hear it.
Confirm or establish full neuromuscular blockade
Give rocuronium if not already paralyzed. A tense patient makes FONA harder.
One final attempt at any untried technique
Only if it has not been tried yet.
Scalpel to neck. Immediately. No more delays.
The recommended technique per DAS 2025, CAFG, and ASA. DAS 2025 standardized a vertical skin incision.
Stabilize the larynx with the non-dominant hand.
Vertical skin incision, roughly 4 cm, over the cricothyroid membrane.
Horizontal stab through the cricothyroid membrane.
Turn the scalpel 90 degrees, sharp edge caudal, maintain traction.
Slide the bougie through the incision, feeling for tracheal ring clicks.
Railroad a 6.0 cuffed ETT over the bougie.
Inflate the cuff, confirm with EtCO2, secure.
Why scalpel-bougie-tube: higher success under stress, a definitive airway immediately, and fewer failure points than a cannula technique.
Traditional algorithms are flowcharts. In a crisis, it is hard to read a flowchart. The Vortex is a visual cognitive aid built for real time use.
Each lifeline gets a best effort. If it fails, move to the next.
All three lifelines exhausted. CICO → eFONA.
The Vortex does not tell you how to intubate. It tells you when to stop trying and move on. That is the part we are bad at.
The most common human factors in airway disasters are behavioral, not technical.
Endlich et al. (2026) applied a Safety-II framework to DAS 2025. Traditional airway management focuses on preventing failure (Safety-I). DAS 2025 legitimizes escalation as an adaptive response, not a failure.
Escalation is not failure. It is adaptation under pressure.
"This is a low, moderate, or high risk airway.
Plan A is video laryngoscopy with a stylet. Bougie is at the bedside.
Plan B is the i-gel, sized by ideal body weight.
Plan C is face mask ventilation.
Plan D is a surgical airway. Cric kit is on the cart.
We have all of our equipment and adjuncts at hand.
Any questions before we start?"
This takes 30 seconds. It prevents chaos.
Emergency cricothyrotomy is needed in under 1% of difficult airways. It is rare. But when it is needed and not performed, or performed too late, people die.
The NAP4 audit found that failure to perform timely cricothyrotomy was a recurring theme in airway related ICU deaths.
The biggest barrier is not skill. It is psychological.
The algorithm gives you permission to cut. Training and simulation give you the skill. If your institution has a Difficult Airway Response Team, utilize it early. But you have to make the decision.
Equipment for Plan D: #20 scalpel, bougie, 6.0 cuffed ETT, 10 mL syringe. Know where it is.
The difficult airway algorithm is not about memorizing a flowchart. It is about building a mental model so that when Plan A fails, and it will eventually, you do not freeze.
Anticipate (MACOCHA) → Optimize (first pass success) → Recognize failure (3-attempt rule) → Rescue (SGA → BVM) → Escalate (eFONA)
Escalation is not failure. It is adaptation under pressure. The best airway experts are not the ones who never fail. They are the ones who recognize failure early and act decisively.
Primary sources: DEVICE trial, NEJM 2023 (Prekker et al.) · Difficult Airway Society (DAS) 2025 guidelines · ID-ACCT study, UK ICU difficult airway incidence · CAFG 3-attempt rule · Vortex approach (Chrimes, 2016) · NAP4 audit, Royal College of Anaesthetists · Toolis et al. 2026, ANZ ICU airway practice survey · Endlich et al. 2026, Safety-II framework applied to airway management.