Morgenstern, J. The DEVICE trial: Not as convincing as you think?, First10EM, October 9, 2023. Available at:
Is there anything emergency doctors love to debate more than airway? We have fancy new tools versus old school badassery. I sort of want to fit into both camps. I want to be the doc who has always succeeded with direct laryngoscopy, while also being young enough to adapt to new technology as it comes along. Both are a different kind of cool. But you know what is actually the coolest? Science.
The DEVICE trial: Prekker ME, Driver BE, Trent SA, et al. Video versus Direct Laryngoscopy for Tracheal Intubation of Critically Ill Adults. N Engl J Med. 2023 Aug 3;389(5):418-429. doi: 10.1056/NEJMoa2301601. Epub 2023 Jun 16. PMID: 37326325 NCT05239195
The DirEct versus VIdeo LaryngosCopE (DEVICE) trial was a pragmatic, multicentre, unblinded RCT from 17 sites, including 7 emergency departments and 10 ICUs in 11 medical centers across the United States.
Critically ill adults undergoing orotracheal intubation in the ED and ICU.
Exclusions: Patients under 18 years old, pregnant patients, prisoners, urgency of intubation precludes enrollment, “operator has determined that the use of a video laryngoscope or use of a direct laryngoscope is required or contraindicated for the optimal care of the patient.”
Intubation with a video laryngoscope. (The protocol did not specify the brand of VL or the geometry of the blade.)
Intubation with a direct laryngoscope. (The protocol did not specify the geometry of the blade)
All other aspects of intubations, such as preoxygenation, drugs, and use of a bougie or stylet, was at the discretion of the intubator.
The primary outcome is successful intubation on the first attempt. (Defined as “placement of an endotracheal tube in the trachea with a single insertion of a laryngoscope blade into the mouth and either a single insertion of an endotracheal tube into the mouth or a single insertion of a bougie into the mouth followed by a single insertion of an endotracheal tube into the mouth.”
The trial was powered for 2000 patients, but was stopped early based on a predetermined endpoint after enrolling 1000 patients. It seems like it took them a while to run the interim analysis, because before actually stopping the trial, they enrolled another 420, for a total study population of 1420.
Median age was 55. Most (70%) of intubations were in an emergency department. About 45% were for altered mental status and 35% for respiratory distress. 92% of the intubations were performed by an emergency resident or critical care fellow. Clinicians had a median of 50 prior intubations, and about 70% of those prior intubations had been with a video laryngoscope.
For the primary outcome of first pass success, VL was better than DL (85% vs 71%, absolute risk difference 14.3% 95% CI, 9.9 to 18.7; p<0.001). The benefit from VL was strongly associated with the total number of prior intubations, with an ARR of 26% among clinicians with less than 25 prior intubations and an ARR of only 6% among clinicians with more than 100 prior intubations.
Severe complications were identical between the two groups (21% vs 21%). Desaturation to less than 80% occurred in 10% of both groups. There were 3 peri-intubation cardiac arrests in both groups.
Time to intubation was 38 seconds with VL and 46 seconds with DL (-8 seconds, 95% CI -12 to -4).
Esophageal intubations occurred in 1% of both groups. There was one cricothyrotomy in the trial, in the DL group.
Although this is the largest and best trial comparing video and direct laryngoscopy, there are a number of issues that impact our interpretation of the results.
Unblinded trials that use subjective outcomes are always problematic. This is especially true in the VL vs DL debate, because a good view might encourage you to prolong your first attempt, resulting in a ‘first pass success’, even if a quick repositioning of the DL might result in a quicker overall intubation, but on the ‘second pass’.
Although a pragmatic trial makes some sense, the lack of standardization of things like preoxygenation makes it hard to know whether these results apply to my practice. Given the first pass success rates of 70-85%, there is reason to believe that these results will not apply to many experienced airway practitioners. (That being said, the reason for the low numbers might also be because of a strict research definition of first pass.) Generalizability is also impacted by the fact that this was primarily trainees (with reasonable experience) intubating a narrow range of sicker patients.
The trial really doesn’t compare apples to apples. Not only are you comparing a video screen to no video screen, but the devices come in a large variety of shapes, which adds significant heterogeneity to these results.
We have to remember that first pass success is a surrogate outcome, and although there is a strong correlation in prior research between first pass success and adverse events, the stricter definition of first pass in research might limit that correlation. (If you take the laryngoscope out, gently bag the patient, and easily intubate on a second attempt with no desaturation, that is not a bad outcome, but it would count against you in this trial.) Surrogate outcomes are really important in small trials, but if you are planning a 2000 patient trial, it is probably time to start looking at real patient oriented outcomes. Patient oriented outcomes, such as severe adverse events, rate of desaturation, and cardiac arrest, were all identical in this study, which probably means that the surrogate outcome is not all that important. In other words, for outcomes that really matter to patients, this trial seems to demonstrate equivalence between the two tools. (Of course, it would be difficult to show a difference in these more important outcomes, because skilled airway operators can prevent most adverse events with meticulous preoxygenation, and adequate plans to adjust course in the face of poor glottic visualization.)
I always have trouble interpreting studies that allow the practicing clinician to exclude patients when they think one option is better than the other. There is a clear bias when that occurs, which could dramatically skew the results. The mistake could occur in either direction. Sometimes the patients excluded did not require the ‘necessary’ treatment, and the exclusion just gives us an excuse to continue using a treatment despite negative studies. On the other hand, excluding the patients most likely to benefit will bias the trial towards the null hypothesis if there is a true benefit. In this trial, 76 patients were excluded because VL was ‘required’ as compared to only 6 for DL, a clear (but predictable) bias.
Although it is only a secondary analysis, I believe the results that show a correlation between training and DL success. The learning curve is quicker for video laryngoscopy, but once you have mastered airway anatomy, the differences between the devices becomes a lot smaller. This data tells me that it makes a lot of sense to train people using VL, but introduce DL once a critical number of intubations has been reached.
I don’t really think there is a debate between video and direct laryngoscopy. The only reason there was a debate in the past is because we were conflating two different issues. When I started, the only video laryngoscopes available were hyper-angulated. Therefore, there was a significant tradeoff between the better view and impaired tube delivery. There was also a debate about whether training would adequately cross-over between the two devices. Now that standard geometry video laryngoscopes are widely available, the debate is settled. We can have the better visualization offered by video, with no impairment of tube delivery, and perfect training. (The training is way better than before, because the standard blade can be used exactly like a direct blade, but your teacher can see exactly what is happening and provide real time feedback.) Even the old concerns about whether you would be adequately trained in the case of an equipment failure are completely moot, because the standard geometry VL can be used as a DL. Aside from cost questions for hospitals that don’t already have VLs, I think this debate is over.
I don’t think this trial is as definitive as some people have made it out to be. It is a large RCT with a positive primary outcome suggesting benefit from video laryngoscopes, but the trial really shows equivalence on the more important outcomes, like significant adverse events. I don’t think you can use this trial to make a strong argument for VL, but I also I don’t think you need science to tell you that VL is a better tool.
Prekker ME, Driver BE, Trent SA, Resnick-Ault D, Seitz K, Russell DW, Gandotra S, Gaillard JP, Gibbs KW, Latimer A, Whitson MR, Ghamande S, Vonderhaar DJ, Walco JP, Hansen SJ, Douglas IS, Barnes CR, Krishnamoorthy V, Bastman JJ, Lloyd BD, Robison SW, Palakshappa JA, Mitchell S, Page DB, White HD, Espinera A, Hughes C, Joffe AM, Herbert JT, Schauer SG, Long BJ, Imhoff B, Wang L, Rhoads JP, Womack KN, Janz D, Self WH, Rice TW, Ginde AA, Casey JD, Semler MW; DEVICE investigators and the Pragmatic Critical Care Research Group. DirEct versus VIdeo LaryngosCopE (DEVICE): protocol and statistical analysis plan for a randomised clinical trial in critically ill adults undergoing emergency tracheal intubation. BMJ Open. 2023 Jan 13;13(1):e068978. doi: 10.1136/bmjopen-2022-068978. PMID: 36639210