Abstract
Background:
The Centers for Disease Control and Prevention surveillance algorithm identifies complications associated with invasive mechanical ventilation as ventilator-associated events (VAE). There is little known about reason for intubation and VAE risk. We aim to evaluate the association between reason for intubation and development of VAE. We hypothesize that subjects with primary pulmonary reasons for intubation have increased risk of VAE compared with subjects whose reason for intubation is nonpulmonary in origin.
Methods:
This is a single-center retrospective study of medical intensive care unit (MICU) subjects mechanically ventilated ≥4 days and admitted between January and December 2019, at an urban university hospital. Subjects were categorized as primary pulmonary versus nonpulmonary based on reason for intubation. For the primary predictor of VAE, we performed multivariate logistic regression adjusting for Mortality Probability Model (MPM0-III) variables. We accounted for the time-varying risk of VAE using a Cox proportional hazard model. Secondary outcomes included hospital and ICU mortality and stay.
Results:
After exclusions, there were 250 subjects in our sample 98 (39.2%) in the pulmonary and 152 (60.8%) in the nonpulmonary groups for comparison. Subjects in the nonpulmonary group had higher severity-of-illness as measured by the MPM0-III (P = .002). There was no difference in crude VAE rates between groups. In the fully adjusted model, there was no higher incidence of VAE among subjects intubated for primary pulmonary reasons (OR: 0.78, 95% CI: 0.34–1.78, P = .55). Time to VAE was not impacted by reason for intubation. There was no difference in mortality and stay.
Conclusions:
Our results do not suggest an association between reason for intubation and risk for VAE. This reinforces the validity of the VAE surveillance algorithm by shifting focus to modifiable therapeutic choices during a course of invasive mechanical ventilation to reduce VAE risk.
Get full access to this article
View all access options for this article.
