Video beats direct laryngoscopy for intubation--even for experts
The question now is, how to teach and maintain DL skills in the video era?
Video laryngoscopy was introduced in 2001 as a technique for orotracheal intubation. Although an increasing share of intubations have been performed using video laryngoscopy, and it’s recommended for difficult airways, uncertainty has remained as to whether video was superior than conventional direct laryngoscopy for average risk intubations.
In the relatively few studies comparing video to direct laryngoscopy, intubations were usually performed by inexperienced trainees, and sample sizes were small. The real world success rates by operators experienced with intubation remained unknown.
That uncertainty has been largely dispelled by large randomized trials in 2023 and 2024 confirming that video laryngoscopy is a superior method for obtaining first pass intubation success for patients with average risk airways. The results seem to hold for both major video laryngoscopes (GlideScope and McGrath).
Cleveland Clinic: DL vs. GlideScopes in the OR (JAMA 2024)
In a cluster-crossover design, Cleveland Clinic’s two sets of 11 operating rooms randomly switched between direct laryngoscopy and video with a hyperangulated blade (GlideScope) for a week at a time over 18 months in 2021-2022, while intubating patients for elective or emergent cardiac, thoracic, or vascular surgical procedures. Intubations for over 8,400 surgical procedures (85% elective, 15% emergent) in 7,736 patients were studied.
Patients randomized to direct laryngoscopy were four times more likely to require multiple intubation attempts (7.6% vs. 1.7%).
Direct laryngoscopy patients were thirteen times more likely to experience intubation failure (4.0% vs 0.27%), compared to patients randomized to video laryngoscopy. This outcome measure was less robust than intubation attempts, because the “failure” definition included simply switching devices, which the intubator was permitted to do at any time for any reason.
What makes this study highly convincing for video laryngoscopy’s superiority is the skill level of the intubators. Eighty-five percent of intubations were performed by CRNAs, anesthesiology residents, fellows, or attendings; only 15% were performed by student CRNAs or medical students.
It's an important point, because prior trials suggesting video laryngoscopy’s superiority largely studied emergency medicine and critical care trainees, who may only need to complete a few dozen intubations throughout their entire training program to be considered competent. Although most EM and CCM trainees intubate more than that, fewer than 30 intubations per year for EM residents, and fewer than 100 total for critical care fellows over a fellowship are common benchmarks.
Anesthesiology residents literally stop counting, generally performing well over a thousand intubations in a four-year residency (compared to pulmonary-critical care fellows’ mere nine months total in the ICU on a traditional research/clinical split), including extensive experience on difficult airways and advanced airway management.
For their part, CRNAs also log hundreds of intubations during training, then they do it every workday, practicing without physician supervision in more than half the states in the U.S.
In other words, this is a group of predominantly highly skilled intubators, and still they:
Required multiple DL attempts once every 14 patients (vs. needing multiple attempts once every 59 intubations with video), and
Abandoned DL to switch to video in one of every 25 patients (vs. failing every 371 patients with video).
Although the results were not broken down granularly by experience level, there was no evidence of heterogeneity of effects based on training or role (i.e., it wasn’t the student CRNAs skewing the results).
This is highly convincing evidence that that video is better overall for first-pass intubation success.
EMMA Trial: DL vs. McGrath MAC in the OR (Anaesthesia 2023)
In a multicenter trial (four centers in Germany, one in Canada in 2015-2019), 2000 patients underwent intubation for elective surgery by a mix of trainees (66%) and attendings (33%), randomized to either direct laryngoscopy or McGrath MAC video laryngoscopy for the first attempt(s). The McGrath MAC has a blade designed to emulate the traditional direct laryngoscopy Macintosh blade (i.e., it is not hyperangulated like the GlideScope).
Among 2092 intubations:
94% of patients randomized to video laryngoscopy were intubated on the first attempt, vs. 82% with direct laryngoscopy.
6% of airways were considered difficult with DL, but only 1% with video.
In about 2% of intubations with the McGrath, screen flickering or camera fogging made the intubation difficult.
Trainees’ success was more improved with video, relative to attendings’—but attendings also had higher first pass success rates with video than DL.
ED and CCM fellows intubating unstable patients (DEVICE trial, NEJM 2023)
Using a variety of video laryngoscopy devices (vs. direct laryngoscopy) when intubating 1417 unstable patients, ED and CCM fellows had an 85% first pass success rate with video, compared to 71% with direct laryngoscopy.
We reviewed the DEVICE trial in 2023, which you can read about here:
The Goalposts Are Moving
An increasing share of intubations have been performed with video since 2001. Each of those intubations represents one less rep for someone to practice direct laryngoscopy. It seems inevitable that on the whole, direct laryngoscopy skills would decline, at least marginally, making video look better (and be better) by comparison. The EMMA trial findings (in which video helped the less-experienced more) bear that out.
At one brand name program 15 years into the video era, pulmonary/critical care fellows successfully intubated on their first attempt with direct laryngoscopy only 40% of the time. I doubt this is an outlier.
Should DL still be taught at all? Of course! Batteries can die, the power can go out, and screens can fog over or stop working. (There is at least one program that has abandoned DL training during critical care fellowship, though.)
Is it ethical to intubate with DL, knowing a patient is being subjected to an increased risk of multiple attempts, which (especially on crashing patients in the ED and ICU) are associated with bad things happening? This gets into complex terrain.
In the interests of all patients (not to mention our own careers and skill sets), trainees need to learn, and experienced clinicians still need to practice and maintain their DL skills.
One imperfect option is training and retraining using the McGrath MAC, without relying on the video screen. Its Macintosh-like shape simulates DL. It would be a small technical feat for the manufacturer to create a Bluetooth-enabled version that broadcasts video to a nearby iPhone (the instructor’s), while keeping the laryngoscope screen (the intubator’s view) dark. (Or just cover up the screen with your hand.) If safety becomes an issue, display the screen.
Conclusion
There’s very little quality data regarding safety of intubation over time, but it seems indisputable that video laryngoscopy has reduced the rate of difficult intubations and the uncommon but devastating catastrophes that can result.
Recent large randomized trials confirm the secular trend toward video laryngoscopy is appropriate and justified. Video improves first pass intubation success, compared to direct laryngoscopy.
The challenge for educators, trainees, and experienced clinicians alike will be to maintain proficiency with the “inferior,” but essential skill of direct laryngoscopy as we go forward in the video era.
4% failure rate is really high. This makes this single center trial suspicious. In addition, video laryngoscopes are effective when the defect is an anteriorly displaced larynx. I suspect some of this difference is crossover from the DL cohort to the VL cohort when an anteriorly displaced larynx is discovered. It is true that anesthesiologists at all training levels abandon now DL in favor of VL sooner than they would have 10 or 15 years ago. This means they get little experience with other (e.g. straight Miller) blades and maneuvers which increased DL success in the past. This is how all technical advances work; the skillset that developed around the old methods atrophies. This is a good thing, generally. The problem we have observed mostly among ED and critical care practitioners, is that when the defect is not an anteriorly displaced larynx, VL can be disadvantageous. Obstructing masses, airway hemorrhage or a very distorted anatomy from perhaps radiation or trauma is often harder to identify and intubate with a VL than a DL. At our place, this has led to some really poor outcomes. Rare events to be sure, but morbid.
There is an easy solution to this - Standard geometry VL(SGVL). While CMAC is the OG SGVL brand, Glidescope also makes them. They make training airway skills so much less stressful as a staff - you just turn the screen so you can see it but the trainee cannot during the attempt. You can give them active feedback knowing exactly what they're seeing and also knowing if they goose it before you inflate the gut.
As a gas-crit I will say that (anecdotally) SGVL is also less time consuming with less futzing around than with Hyperangulated VL (HAVL) . Unless someone has a (literal) fixed neck I will reach for a SGVL 100/100 times.