Western Journal of Emergency Medicine: Integrating Emergency Care with Population Health

Introduction: During endotracheal intubation, equipment setup typically includes the patient monitor fixed at the head of the bed behind the back of the operator. Inability to directly visualize the patient monitor may result in delayed recognition of desaturation. Our primary aim in this study was to measure the association between patient monitor position and the time to recognition of a desaturation event during endotracheal intubation.

Methods: We performed a randomized crossover trial of emergency medicine residents across two Accreditation Council for Graduate Medical Education-accredited programs. Subjects were asked to perform direct and video-assisted laryngoscopy once on a difficult airway trainer in a simulation. The sequence of laryngoscopy modality (direct vs video-assisted) and monitor position (head vs left vs right of bed) were randomized prior to each attempt. The simulated patient’s peripheral capillary oxygen saturation (SpO2 ) was programmed to begin at 100% and decrease at a rate of 1% per second 10 seconds after the start of the subject’s attempt. The primary outcome measure was time to operator recognition of hypoxia defined as the observed period during which the simulated SpO2 was < 90%. Secondary outcomes were operator failure to visualize a desaturation event. We rendered Kaplan-Meier curves illustrating the time to visualization of hypoxia and performed a Cox regression adjusting for laryngoscopy modality and total number of previous intubations performed. Using multivariable linear regression, we modeled the association between time to recognition of hypoxia in seconds and patient monitor position with similar adjustments. To assess differences in the number of observed failure events between study arms, we used chi-squared or Fisher exact tests. The threshold for statistical significance was a two-sided P < 0.05.

Results: We observed 68 attempts by 34 subjects. Twenty-two (32.5%), 22 (32.5%), and 24 (35%) intubations were performed with the monitor positioned at the head, left, and right, respectively. The median times to recognition of hypoxia were 37 seconds [sec] (95% CI, 25-86 sec) for the head, 32 sec (95% CI, 18-50 sec) for the left, and 23 sec (95% CI, 19-34 sec) for the right groups, respectively. Cox regression demonstrated hazard ratios of 2.7 and 3.2 for the left and right groups when compared to the head group, and these findings were statistically significant (P = .04 and P = 0.03, respectively). We found no statistically significant associations between time to recognition of hypoxia and laryngoscopy modality or total number of previous intubations (P = .82 and .21, respectively). Failure rates across head, left, and right groups were similar, at 36%, 59%, and 50%, respectively (P = .30).

Conclusion: Positioning of the patient monitor at the head of the bed results in delayed visualization of desaturation events during simulated direct and video-assisted laryngoscopy across different levels of experience. Ideally, the patient monitor should be positioned on the side of the bed and within the operator’s direct line of sight. Further research is warranted to assess how equipment setup may impact procedural performance, operator ergonomics, and patient safety.