医学
四分位间距
急诊科
置信区间
优势比
插管
凝视
屏幕时间
气管插管
急诊医学
麻醉
外科
物理疗法
内科学
计算机科学
人工智能
精神科
体力活动
作者
Preston Dean,Katherine Edmunds,Ashish S. Shah,Mary Frey,Yin Zhang,Stephanie Boyd,Benjamin T. Kerrey
标识
DOI:10.1016/j.annemergmed.2021.11.019
摘要
Our study objectives were to describe patterns of video laryngoscope screen visualization during tracheal intubation in a pediatric emergency department (ED) and to determine their associations with procedural performance.We conducted a prospective, observational, video-based study of pediatric ED patients undergoing tracheal intubation with a standard geometry video laryngoscope (Storz C-MAC; Karl Storz, Tuttlingen, Germany). Our primary exposure was video screen visualization patterns, measured by the percentage of each attempt spent viewing the screen and the number of times the proceduralist changed their gaze between the patient and screen (gaze switches). Our primary outcome was first-pass success. We compared measures of screen visualization between successful and unsuccessful first attempts using a generalized linear mixed model.From December 2019 to October 2021, we collected data on 153 patients. The first-pass success rate was 79.1%. Proceduralists viewed the video screen during 80.4% of attempts; the median percentage of each attempt spent viewing the video screen was 42.1% (interquartile range 8.7% to 65.5%). The median number of gaze switches per attempt was 3 (interquartile range 1 to 6, maximum 22). The percentage of each attempt spent viewing the video screen was not associated with success (adjusted odds ratio 1.00, 95% confidence interval 0.93 to 1.08); additional gaze switches were associated with a lower likelihood of success (adjusted odds ratio 0.80, 95% confidence interval 0.71 to 0.90).We found wide variation in how proceduralists viewed the video laryngoscope screen during intubations in a pediatric ED. We illustrate the application of 2 objective screen visualization measures to quantify and understand how clinicians actually use video laryngoscopy.
科研通智能强力驱动
Strongly Powered by AbleSci AI