Introduction

Acute respiratory failure remains one of the three leading causes of admission to the intensive care unit (ICU). Self-extubation is an adverse outcome requiring reintubation in 50% of cases.

Objective

To assess for determinants (risk factors) of self-extubation and mortality in the ICU by using a generalized estimation equation model (GEE).

Methods

The data was collected from a retrospective cohort study from 2017-2020 including all patients admitted to the ICU with mechanical ventilation. Univariate and bivariate analyses were performed. Then, a GEE model was conducted to predict the risk of self-extubation and mortality.

Results

A total of 857 subjects were included, with a mean age of 60.5 +/- 17 years-old. Most of the subjects were males (55.2%). An 8.99-fold risk (95%CI 3.83-21.1, p<0.0i) of self-extubation was identified in patients with agitation. Exposure to infusion of neuromuscular blockade was also found to increase the risk of self-extubation 3.37 times (95%CI 1.31-8.68, p=0.01). No associations were identified between immobilization and self-extubation (OR 1.38 95%CI 0.76-2.51, p=0.29). Finally, light sedation according to the Richmond Sedation Scale (RASS) between 0 to -2 rather than moderate (RASS-3) reduces the risk of mortality (OR 0.57, 95%CI 0.38-0.83, p<0.01).

Conclusions

The main factors resulting in self-extubation were: agitation, delirium, and infusion of neuromuscular blocking agents. An association was found between light sedation and a lower risk of mortality. No association was found between the use of physical restraint and the desired outcome.

        · abstract in Spanish     · text in English | Spanish     · English ( pdf ) | Spanish ( pdf )