Predictors of reintubation in trauma intensive care unit : Qatar experience

Abstract EN

Objectives: To determine the causes, predictors and outcomes of reintubation.

Methods: Retrospective analysis of data collected from the trauma data base registry was conducted to identify the extubation failure cases in Trauma ICU at Hamad General Hospital, the only Level I trauma center in Qatar between January 2009 and December 2010.

Demographics, mechanism of Injury, complications, injury severity score (ISS), Glasgow Coma Scale (GCS), ICU-length of stay (LOS), and mortality were analyzed among trauma patients who need reintubation within 48 hrs after extubation (group 1) compared to successfully extubated patients (group 2).

Result: A total of 954 patients were admitted to the trauma ICU, of which 343 were intubated orotracheally.

The mean age of patients was 32±12 years with male predominance (95%).

Motor vehicle crash (41%), pedestrian injury (20%) and falls (18%) were the most common mechanisms of injury.

Reintubation (group 1) was required in 24 patients (7%).

Patients in group 1 had higher rate of head injury mainly SAH (88%), pneumonia (79%) and pulmonary contusion (58%).

The mean ICU-LOS was higher in the reintubated patients (p=0.010) in comparison to group 2.

Forty-six percent of reintubated patients required tracheostomy.

The mean age, ISS, GCS and tube size was comparable among the two groups.

Furthermore, reintubation was not associated with higher mortality rate (p=0.910).

However, Ventilator-associated pneumonia (VAP) (odd ratio=3.61 [95% CI 1.25-10.44]; p=0.020) and ventilator days (odd ratio=1.09 [95% CI 1.024-1.153]; p=0.006) were independent predictors of reintubation by multivariate analysis.

Conclusion: Re-intubation is associated with increased ICU-LOS and need for tracheostomy.

VAP and prolonged intubation are independent predictors of re-intubation.

Our finding addresses the value of prevention and early treatment of infection in intubated patients.

This study may represent an audit of local practice as well.