Does Video Laryngoscopy or Direct Laryngoscopy affect first pass success rates for intubation among Attending and Non-Attending Emergency Physician in the Emergency Department?


 Background: To our knowledge, there has been no study comparing intubation characteristics between attending and non-attending Emergency Physicians in South-East Asia. We aim to identify whether the use of Direct Laryngoscopy (DL) compared to Video Laryngoscopy (VL) affects first pass success rates between Attending Emergency Physicians (AEP) and Non-Attending Emergency Physicians (NAEP). Materials and Methods: Retrospective analysis of data from 2009 to 2016 in an existing airway registry managed by and academic Emergency Department in Singapore. Primary outcome is first pass success intubation rate. The secondary outcome was first pass success rate for difficult intubations. Difficult intubations were defined as LEMON score of more than 1 or more than 1 attempt at intubation. Results: There were 2909 intubation carried out by emergency physicians in the Emergency Department from 2009 to 2016. AEP conducted 1748 intubations while NAEP conducted 1161 intubations. The first pass success rates for AEP was 84.2% while that for NAEP was 67.4%. 86.2% of intubations by AEP were done with a direct laryngoscope. 89.0% of the intubations by NAEP were done with a direct laryngoscope. 18.9% of intubations by the AEP were difficult compared to 35.2% by the NAEP (p<0.01 95% C.I 13.0%-19.6%). First pass success rate with VL was lower than DL for all intubations (OR 0.66, 95% C.I 0.51-0.84). In the subgroup of difficult intubations, VL did not improve first pass success rate among AEP (OR 0.77, 95% C.I 0.38-1.58) but it did for NAEP (OR 2.46, 95% C.I 0.94-6.45). Conclusion: Our study showed that VL has a poorer first pass success rate for all intubations in general. However, specifically for difficult intubations, VL is associated with improved first pass success rates among NAEP.


Introduction
First pass success in emergency intubations has been associated with fewer adverse events 1 than if more than a single attempt is needed at intubation. Level of experience and choice of Video Laryngoscopy (VL) or Direct Laryngoscopy (DL) have been reported to be features associated with first pass success in intubations 2,3,4 . In South-East Asia, there have been no studies examining the relationship between these two factors and how they relate to first pass success in intubation. We aim to identify whether the use of VL compared to DL affects first pass success rates for regular or difficult endotracheal intubations when performed by Attending Emergency Physicians (AEP) and Non-Attending Emergency Physicians (NAEP).

Materials And Methods
This was a single-center retrospective observational study. Data was obtained from the Airway Registry of an academic Emergency Department (ED) at the Singapore General Hospital (SGH), Singapore, between 2009 and 2016. All intubations conducted in the ED during this period were documented on a standardized hardcopy form. The entries were then transcribed into an electronic Airway Registry. Approval from the Institution Review Board was obtained for data collection into the Airway Registry. The primary outcome was rate of successful first pass intubation. A single pass was defined as an attempt to pass the endotracheal tube (ETT) through the vocal cords. A pass will usually be followed by airway maneuvers e.g. bagging or suctioning.
The LEMON (Look externally, Evaluate 3-3-2, Mallampati, Obstruction, Neck Mobility) criteria is a commonly used tool to assess a difficult airway. The presence of any of these features in the criteria will be given a score of 1. A difficult intubation is defined as a LEMON score of more than 1 or the number of attempts required for a successful intubation being more than 1.
Secondary outcomes were the first pass intubation success rate for difficult intubations. Categorical variables were analysed using Chi-Squared Test for differences between groups. The t-test was used to identify differences between groups for continuous variables. Statistical significance was considered if p<0.05. Odds Ratios (OR) and 95% confidence intervals (95% CI) were presented where applicable.

All Intubations
AEPs had a higher overall first pass success rate than NAEPs (84.2% versus 67.4%, p<0.01). The mean number of attempts was 1.24 and 1.46 for AEPs and NAEPs, respectively.

Video Laryngoscopy
The GlideScope had the highest first pass success rate among all VL (OR 2.85, 95% C.I 0.39-20.7), followed by C-MAC (OR 1.79 95% C.I 0.24-13.14) and lastly McGrath. For each variant of VL used, there was no statistically significant difference in the first past success rates between AEP and NAEP (Table 3). This finding was also consistent regardless of whether it was a difficult intubation.

Discussion
The observed finding of VL having lower overall first pass success rate compared to DL (p = 0.0008, OR 0.66 95% C.I 0.51-0.84) is dissimilar to existing literature. Existing p u b l i c a t i o n s 6,7,8,9,10 report no statistically significant difference in first pass success rates between VL and DL. The greater familiarity with DL likely accounted for the higher rates of DL use in our study and the lower rates of first pass success with VL.
For difficult intubations, VL was associated with marginally better overall first pass success rates (p = 0.3081, OR 1.34, 95% C.I 0.76-2.39). With VL, the GlideScope had the highest first pass success rate compared to the other two VL. The presence of a video camera at the distal end of the VL allows for visualization of the vocal cords without alignment of the various intubation axes. Specifically, the rigid stylet for the GlideScope Video Laryngoscope allows for smooth insertion 11,12 of the ETT along the hyperacute angle of the laryngoscope blade. These features likely helped improve first pass success rates in difficult intubations 7,11,12,13 , especially where small mouth opening, and limited neck mobility posed a challenge. The C-Mac allows the user to perform intubation as per DL or via its attached video camera with an external monitor. The McGrath has a monitor attached to the laryngoscopy blade, allowing the user to maintain line of sight of the patient and hand positions while intubating. Despite the various traits of each device, our study did not show statistically significant improved first pass intubation rates when these devices were used by AEPs than by supervised NAEPs (Table 3).
Moreover, these results are further augmented in the subgroup of difficult intubations among the NAEP 7,11,12,13 . An improved glottic view and ease in anatomy identification are reported mechanisms resulting in success with VL among NAEP for difficult intubations. In addition, the ability to visualize the NAEP's field of view with VL also allows the supervising AEP to provide timely and appropriate advice during the intubation. Therefore, video laryngoscope-assisted intubations are useful for training relative novices in endotracheal intubations skills, especially during difficult intubations by these groups of healthcare workers. 14,15 Our study clearly demonstrated that persons performing endotracheal intubations performed better with devices they were familiar with, such as with direct laryngoscopy. Therefore, trained AEPs did better, as would be expected, than NAEPs with DL and also used DL more effectively than VL. However, when encountering difficult airways the large differences in first-pass successful intubation rates between DL and VL were greatly reduced. These attest to the need for training with use of VL and to the use of VL as a secondary adjunct for gaining airway access during difficult intubations.
First pass success during Emergency Intubations are associated with fewer adverse effects and better patient outcomes 14,15 . Therein lies the importance to balance operator level of training and device choice so that an optimal first pass success rate is achieved. If novice operators are not provided with the opportunity to conduct Emergency Intubations, this may eventually affect their first pass success rates as future airway managers.

Limitations
This was a single-center study with data derived from an Airway Registry. Cases were entered into the Airway Registry by manually transcribing into the electronic database from a hard copy form. This allows for errors in transcription. In addition, since the forms were completed by the operators after completion of the resuscitation which included the intubation, recall bias may contribute to inaccurately recorded data. This may thus affect the generalizability of the results. Of note, our results specifically for the difficult intubation subgroup are consistent with that of existing literature.
Our definition of a difficult intubation did not use a single standardized definition such as the Intubation Difficulty Score (IDS) 16 . In addition to the LEMON score, we included all first pass intubation failures. The greater unfamiliarity of NAEPs with endotracheal intubation likely contributed to the higher proportion of difficult intubations in this group and the larger overall number of such intubations. The LEMON score, by itself, has been shown to correlate well with difficulty of intubation 17,18 .

Conclusions
Our study shows that VL has a poorer first pass success rate for all intubations in general.
However, specifically for difficult intubations, VL is associated with improved first pass success rates especially when used by supervised NAEPs. Between AEPs and NAEPs, while DL use was easier with trained AEPs, whether for standard or difficult airways, there was no statistically significant difference in first pass intubation rates with VL use.

Ethics Approval and Consent to Participate
Approval from the Singhealth Institution Review Board was obtained for data collection into the Airway Registry used for this study.

Availability of Data and Materials
The dataset analysed in this current study may be made available on reasonable request.
Please contact WPW.

Competing Interests
There are no competing interests to declare.

Not applicable
Authors' Contributions WPW, NDZ, EW developed the study and supervised its data collection. SGC assisted with data collection and entry. WPW analysed the data. WPW drafted the manuscript. NDZ and EW provided advice the manuscript and contributed as corresponding authors. WPW take responsibility for the paper.