|| Checking for direct PDF access through Ovid
Urgent tracheal intubation is performed frequently in intensive care units and incurs higher risk than when intubation is performed under more controlled circumstances. Video laryngoscopy may improve the chances of successful tracheal intubation on the first attempt; however, existing comparative data on outcomes are limited.To compare first-attempt success and complication rates during intubation when using video laryngoscopy compared with traditional direct laryngoscopy in a tertiary academic medical intensive care unit.We prospectively collected and analyzed data from a continuous quality improvement database of all intubations in one medical intensive care unit between January 1, 2012, and December 31, 2014. Propensity matching and multivariable logistic regression were used to reduce the risk of bias and control for confounding.A total of 809 intubations took place over the study period. Of these, 673 (83.2%) were performed using video laryngoscopy and 136 (16.8%) using direct laryngoscopy. First-attempt success with video laryngoscopy was 80.4% (95% confidence interval [CI], 77.2-83.3%) compared with 65.4% (95% CI, 56.8-73.4%) for intubations performed with direct laryngoscopy (P < 0.001). In a propensity-matched analysis, the odds ratio for first-attempt success with video laryngoscopy versus direct laryngoscopy was 2.81 (95% CI, 2.27-3.59). The rate of arterial oxygen desaturation events during the first intubation attempt was significantly lower for video laryngoscopy than for direct laryngoscopy (18.3% vs. 25.9%; P = 0.04). The rate of esophageal intubation during any attempt was also significantly lower for video laryngoscopy (2.1% vs. 6.6%; P = 0.008).Video laryngoscopy was associated with significantly improved odds of first-attempt success at tracheal intubation by nonanesthesiologists in a medical intensive care unit. Esophageal intubation and oxygen desaturation occurred less frequently with the use of video laryngoscopy. Randomized clinical trials are needed to confirm these findings.