Association of Glasgow coma scale and endotracheal intubation in predicting mortality among patients admitted to the intensive care unit

Abstract

Background: We assessed predictors of mortality in the intensive care unit (ICU) and investigated if Glasgow coma scale (GCS) is associated with mortality in patients undergoing endotracheal intubation (EI).

Methods: From February 2020, we performed a 1-year study on 2,055 adult patients admitted to the ICU of two teaching hospitals. The outcome was mortality during ICU stay and the predictors were patients' demographic, clinical, and laboratory features.

Results: EI was associated with a decreased risk for mortality compared with similar patients (adjusted odds ratio [AOR], 0.32; P=0.030). This shows that EI had been performed correctly with proper indications. Increasing age (AOR, 1.04; P<0.001) or blood pressure (AOR, 1.01; P<0.001), respiratory problems (AOR, 3.24; P<0.001), nosocomial infection (AOR, 1.64; P=0.014), diabetes (AOR, 5.69; P<0.001), history of myocardial infarction (AOR, 2.52; P<0.001), chronic obstructive pulmonary disease (AOR, 3.93; P<0.001), immunosuppression (AOR, 3.15; P<0.001), and the use of anesthetics/sedatives/hypnotics for reasons other than EI (AOR, 4.60; P<0.001) were directly; and GCS (AOR, 0.84; P<0.001) was inversely related to mortality. In patients with trauma surgeries (AOR, 0.62; P=0.014) or other surgical categories (AOR, 0.61; P=0.024) undergoing EI, GCS had an inverse relation with mortality (accuracy=82.6%, area under the receiver operator characteristic curve=0.81).

Conclusions: A variety of features affected the risk for mortality in patients admitted to the ICU. Considering GCS score for EI had the potential of affecting prognosis in subgroups of patients such as those with trauma surgeries or other surgical categories.

Keywords: Glasgow coma scale; endotracheal intubation; intensive care unit; logistic models; mortality; risk factors.

Figure 1.

Bar chart of Glasgow coma scale (GCS) categories in patients with or without undergoing endotracheal intubation. Intubated patients had a GCS of 10 or less.

Figure 2.

Variable importance for predicting mortality rate. The white boxplots illustrate the confirmed selected features. PaO₂: partial pressure of oxygen; PaCO₂: partial pressure of carbon dioxide; EI: endotracheal intubation; FiO₂: fraction of inspired oxygen; SBP: systolic blood pressure; GCS: Glasgow coma scale; COPD: chronic obstructive pulmonary disease; MI: myocardial infarction.

Figure 3.

The area under the receiver operator characteristic (ROC) curve (0.81). The black line represents the model prediction and the diagonal gray line denotes the ROC curve of a random classifier.