The drowned airway refers to massive regurgitation of blood or vomitus that prevents either intubation or mask-ventilation. This is a rare and uniquely perilous situation. Massive vomiting simultaneously impairs many airway techniques (e.g. direct laryngoscopy, indirect laryngoscopy, mask ventilation, fiberoptic intubation).
The literature describes various techniques for the drowned airway. However, there doesn’t appear to be any airway algorithm which integrates these into a coherent strategy. A collection of airway tricks without any plan is a formula for disaster. Generic airway algorithms work poorly, because they include steps which are unsafe (e.g. mask ventilation) and exclude techniques which may be helpful (e.g. esophageal ETT diversion). This post will attempt to construct an airway algorithm specific to the drowned airway (1).
Getting started: Prevention & preparation
- Active gastrointestinal disease (e.g. upper GI bleed, obstruction, ileus)
- Gastroparesis, diabetes
- Ultrasonography showing gastric distension (best predictor, for more information see gastricultrasound.org).
Drainage should be considered if a significant volume of gastric fluid is suspected or seen on ultrasonography (Koenig 2011).
- NG tube suction. This is often helpful, although not necessarily 100% effective (the tube can become occluded by debris). Use of a larger NG tube might deter blockage. For patients who are too agitated to permit NG tube placement, a dissociative dose of ketamine may be used to facilitate this procedure (similar to delayed-sequence intubation).
- Encourage vomiting. Some patients will suppress the urge to vomit. With encouragement and perhaps some gentle epigastric pressure, these patients can often vomit and empty their stomachs immediately prior to intubation. This strategy requires that the patient must be awake and able to protect their airway while vomiting. Some patients with hematemesis are vomiting periodically already – the best time to intubate is immediately after an episode of vomiting.
- Prokinetic medications: Evidence does support the ability of prokinetics to empty gastric contents (especially 250 mg IV erythromycin). However logistically it is usually impossible to order erythromycin, receive it from pharmacy, push it over five minutes, allow it to work for 15-30 minutes, and then intubate the patient (Czarnetzki 2015).
Preparation for aspiration
If the stomach cannot be drained or the risk remains unclear, procedural preparations include:
- Intubate in steep reverse Trendelenberg: Head elevation may prevent passive regurgitation.
- Avoid mask ventilation: This may promote gastric insufflation and regurgitation.
- Intubation by most experienced operator: The best shot at intubation is immediately after paralysis: there is often a short period of time when the airway is clear, before a tsunami of vomit pours in. A nimble intubator can capitalize on this narrow window of opportunity.
- Prepare two large-bore suction catheters: Ideally, large-bore suction catheters should be used routinely for all intubations. If your shop isn’t using commercial large-bore suction devices, you can MacGyver one easily from suction tubing as shown here.
- Personal protective equipment: These intubations can involve spraying of vomit or blood (e.g. during cricothyrotomy).
Designing the algorithm: Foundational concepts
There are roughly four common adverse outcomes from a drowned airway:
- Anoxic brain injury
- Aspiration causing ARDS
Avoiding immediate death or anoxic brain injury is obviously the first priority. The second priority is avoiding a large-volume aspiration leading to ARDS (which carries a 30% mortality). Avoidance of cricothyrotomy is the lowest priority, since it causes the lowest morbidity.
#2: Lack of re-oxygenation options
Conventional airway algorithms are based on alternation between intubation attempts and re-oxygenation (usually with a bag-valve mask). Re-oxygenation may be dangerous or impossible in a drowned airway:
- Mask ventilation or LMA may propel vomitus deeper into the lungs, causing aspiration and ARDS.
- If aspiration has already occurred, the trachea may become obstructed with thick vomitus, making it impossible to ventilate with either a mask or LMA.
Lack of a re-oxygenation option means that the airway must be secured immediately, prior to critical desaturation (within the “safe apnea time”). For a healthy patient who is perfectly preoxygenated, this may provide 5-10 minutes. Unfortunately, for a critically ill patient with morbid obesity or respiratory failure, the safe apnea time will be much shorter. The drowned airway algorithm must therefore be achievable within a few minutes. This creates logistic constraints:
- Techniques involved must be simple and fast.
- Materials and personnel required must be immediately available.
Any good airway algorithm should be simple:
- In a crisis, nobody will remember a complicated algorithm.
- The existence of numerous possible options causes problems (e.g. half-hearted attempt at one strategy, then another strategy, then a third strategy, then a fourth strategy).
Ideally the algorithm should be simple enough for everyone on the team to easily understand (e.g. physicians, respiratory therapists, and nurses). This puts everyone on the same page. If the physician becomes fixated on a specific task (e.g. repeated laryngoscopy attempts), other staff should prompt them to proceed down the algorithm.
#4: Definitive airway control
Drowning will generally cause ongoing soilage of the upper airway. A definitive airway (cuffed plastic in the trachea) is needed for airway protection and toileting. Alternative strategies for oxygenation are less desirable (e.g. trans-tracheal jet ventilation, laryngeal mask ventilation).
Algorithm for the drowned airway
If massive regurgitation occurs unexpectedly, the following steps are helpful:
- Inform the airway team that the drowned airway algorithm will be followed.
- Prepare for cricothyrotomy (e.g. recruit any additional staff that may be nearby to assist with this).
Ideally the team should recognize the risk of drowning and prepare for this in advance. Regardless, it is useful for the entire team to be aware of the situation and possible outcomes (e.g. cricothyrotomy, esophageal diversion) at the earliest timepoint possible.
Step 1: Attempt clearance of the airway with a large-bore suction device.
Ideally, every intubation should be performed with a commercial large-bore suction catheter (figure below). These devices are only trivially more expensive than Yankauer catheters, yet vastly preferable. Routine use of large-bore suction devices puts the operator in a continuous state of readiness.
The first step in dealing with regurgitation is suction. If there is an ongoing stream of blood or vomitus from the esophagus, this may be dealt with by leaving the suction catheter “parked” with its tip in the esophagus, while the endotracheal tube is inserted. The suction catheter may be parked to the left side of the laryngoscope, so that it doesn’t interfere with endotracheal tube insertion (James DuCanto has described this as the “Salad Park Maneuver”).
Step 2: Attempt Seldinger intubation of the trachea with large-bore suction catheter
In some cases, it will be impossible to maintain a view of the glottis (e.g. the airway keeps filling up with vomitus). However, it may be possible to intermittently visualize the glottis. In this situation, the following technique may be used:
- Insert the large-bore suction device through the vocal cords into the trachea while you have an adequate view (2).
- Pass a bougie or airway-exchange catheter through the suction catheter into the trachea.
- Withdraw the suction catheter, leaving the bougie or airway-exchange catheter in the trachea.
- Railroad an ETT into the trachea.
Step 3: Esophageal diversion with an ETT
- Blindly insert an ETT and inflate the cuff.
- Look at what comes out of the ETT (Ideally the ETT should be attached to suction, but it doesn’t have to be. If you don’t have suction, put the tip of the ETT into a basin and allow fluid to drain out).
- Most of the time, the ETT will end up in the esophagus. Vomitus or blood will start regurgitating up through the ETT.
- Rarely, the ETT might end up in the trachea. If this happens, you’re done!
- The esophageal ETT should divert the flow of regurgitated material. With the esophageal ETT in place, attempt laryngoscopy again. It may now be possible to suction the oropharynx and visualize the glottis. Insert a second ETT into the trachea if possible.
- If the esophageal ETT is successful at diverting the vomit and allowing clearance of the airway, this may allow for ventilation with a LMA or mask (even if you aren’t able to intubate the trachea immediately).
Endgame: Scalpel-Finger-Bougie cricothyrotomy
Cricothyrotomy is your eject button. Most patients with massive regurgitation will have normal neck anatomy (3). Thus, surgical cricothyrotomy is usually a simple and fast procedure to gain definitive airway control.
This is neither sterile nor subtle. Antiseptic is foolish, because the surgical field will inevitably lose sterility once vomitus starts bubbling out of the cricothyroid membrane (4). Start with a generous vertical incision, then cut across the membrane, insert a finger, place a bougie, and railroad in the ETT. If the incisions are adequately long and deep, the procedure can be accomplished in about a minute.
- Premature cricothyrotomy exposes the patient to the risk of a minor surgical procedure unnecessarily.
- Delayed cricothyrotomy will result in anoxic brain injury.
The most common error is delaying cricothyrotomy until it’s too late. If you wait until cricothyrotomy is obviously needed (e.g. bradycardia, severe hypoxemia), the procedure will be finished too late to salvage the patient. One of the primary goals of this algorithm is to develop a structured approach to facilitate performance of cricothyrotomy early enough to succeed. Proposed indications for cricothyrotomy are thus:
- Inability to secure the airway despite large-bore suction or esophageal ETT diversion.
- Desaturation below 88%. Once the saturation falls below 88%, the patient is on the steep descent of their desaturation curve.
Standard practice is to start with placement of a small ETT (e.g. #6 ETT) or cuffed tracheostomy tube. However, if the trachea is already impacted with vomitus, a small ETT can become obstructed by debris. It might be necessary to place a larger ETT (e.g. #7 ETT). Make sure not to insert the ETT too deep (starting at the cricothyroid membrane, you only want to insert the ETT a few centimeters).
After the dust has settled, the trans-cricoid ETT may be managed as follows:
- Place an orogastric tube to decompress the stomach.
- Suction the upper airway.
- Under direct vision, intubate the vocal cords via the mouth. As you are inserting the new ETT through the vocal cords, simultaneously remove the old endotracheal tube through the cricoid membrane (5). Secure the trans-oral ETT in the usual fashion.
- A surgeon will often be able to close the neck skin incision at the bedside.
- Consider antibiotic prophylaxis to prevent neck infection (6).
Other possible algorithms?
There are many ways to skin the airway cat. The above algorithm seems reasonable for my context, but it certainly isn’t the only possible approach. Some promising techniques which I left out are:
- Digital intubation: This is a nice technique for dealing with the soiled airway, because your fingers can still feel the epiglottis. However, inter-operator performance is variable depending on hand geometry and experience. If you are adept at this technique then it might be useful for you. This is only useful if you are paralyzing with rocuronium; with succinylcholine the paralysis may wear off and you could get bitten.
- Combitube or King Airway: These devices occlude the esophagus, in a similar manner compared to the esophageal ETT diversion technique. I haven’t included them because they aren’t widely available within hospitals that I’ve worked at. However, if you have these devices and know how to use them, they could be a nice salvage device (especially for EMS folks who don't have lots of support staff).
- Laryngeal Mask Airway: The use of LMA in the drowned airway is controversial, depending on several factors. A properly seated LMA can provide some protection of the airway. However, an incorrectly placed LMA can propel vomitus into the trachea. An advanced-generation LMA could be a reasonable option if it contains a built-in suction channel and the operator is adept at correct placement.
What is your algorithm for the drowned airway? Brew some coffee, take a moment, and create an algorithm that makes sense for your practice.
- The drowned airway is a rare and uniquely perilous situation were copious regurgitation prevents either intubation or mask ventilation. These patients are at very high risk of aspiration, ARDS, anoxic brain damage, or cardiac arrest.
- Massive regurgitation may be prevented in some cases by using gastric ultrasonography to detect large volumes of gastric fluid. NG tube drainage before intubation may mitigate risk.
- Front-line techniques for management of the drowned airway are large-bore catheter suctioning, Seldinger intubation using a large-bore suction catheter, or esophageal diversion with an endotracheal tube.
- There should be a low threshold to perform surgical cricothyrotomy if other techniques fail or if the patient begins desaturating. Cricothyrotomy is highly effective, but in order to work it must be initiated early (before the patient develops massive aspiration or anoxic brain injury).
- An airway algorithm is proposed for management of the drowned airway.
- Large-bore suction for intubation: strategies & devices(PulmCrit)
- Having a vomit SALAD with Dr. Jim Ducanto (EMCrit)
- Cricothyrotomy – Cut to Air: Emergency Surgical Airway(EMCrit)
- Intentional esophageal intubation(ScanCrit)
- Very little evidence is available regarding this rare situation, so it's impossible to claim that this is the best algorithm possible. I look forward to feedback from airway experts.
- This technique is impossible with a standard suction catheter (e.g. Yankhaur catheter), which isn’t large enough to allow passage of a bougie or airway exchange catheter.
- Patients with anatomically challenging airways may be more likely to have difficult cricothyrotomy, because similar factors will often lead to both difficult laryngoscopy and difficult cricothyrotomy (e.g. morbid obesity, neck radiation, neck/tracheal mass). In contrast, the risk factors for massive regurgitation (e.g. gastrointestinal pathology) will generally not correlate with an increased risk of difficult cricothyrotomy. Thus, on average these patients are fairly easy to cric.
- Antisepsis could theoretically reduce the risk of infection, but it’s not worth delaying the procedure. You’re racing the clock here to prevent anoxic brain injury, so delaying the procedure in an attempt to be sterile just isn’t worth it.
- Once the crisis has passed and the patient is saturating well, you now have plenty of time to perform suctioning and laryngoscopy. It generally isn’t difficult to intubate the patient under these more forgiving circumstances.
- There is no evidence regarding this, nor is there ever likely to be. Many of these patients will be sick enough that they’re already receiving antibiotics anyway.