Skip to main content

Table 2 Multivariate analysis of variables associated with intubation and mortality in the overall population and in the propensity score-matched cohort

From: Is immunosuppression status a risk factor for noninvasive ventilation failure in patients with acute hypoxemic respiratory failure? A post hoc matched analysis

Variables

Adjusted odds ratio (95% confidence interval)

p value

In the overall population

Factors associated with intubation a

Expired tidal volume after 1 h under noninvasive ventilation, per 100 mL increase

1.27 (1.03–1.58)

0.03

PaO2/FiO2 after 1 h under noninvasive ventilation, per mmHg drop

1.006 (1.002–1.010)

0.003

Factors associated with ICU mortality b

Immunodepression

2.64 (1.24–5.67)

0.01

Expired tidal volume after 1 h of noninvasive ventilation, per 100 mL increase

1.37 (1.09–1.74)

0.008

In the propensity score-matched cohort

Factors associated with intubation c

Heart rate, per beat per min drop

1.02 (1.001–1.01)

0.048

PaO2/FiO2 after 1 h of noninvasive ventilation, per mmHg drop

1.007 (1.002–1.011)

0.007

Factors associated with ICU mortality d

Immunosuppression

2.56 (1.54–6.57)

0.04

Heart rate, per beat per min drop

1.03 (1.01–1.05)

0.01

  1. aVariables included in the backward stepwise logistic regression model were age, immunosuppression, Glasgow score of oxygen, bilateral lung infiltrates, expired tidal volume after 1 h under noninvasive ventilation, PaO2/FiO2 after 1 h under noninvasive ventilation
  2. bVariables included in the backward stepwise logistic regression model were age, immunosuppression, bilateral lung infiltrates, expired tidal volume after 1 h of noninvasive ventilation, PaO2/FiO2 after 1 h of noninvasive ventilation
  3. cVariables included in the backward stepwise logistic regression model were age, systolic blood pressure under oxygen, heart rate under oxygen, and PaO2/FiO2 after 1 h under noninvasive ventilation. Immunosuppression was forced into the model
  4. dVariables included in the backward stepwise logistic regression model were age, immunosuppression, heart rate under oxygen, tidal volume under noninvasive ventilation, PaO2/FiO2 after 1 h of noninvasive ventilation