Predictors of mechanical ventilation after burn injury

Introduction

It is important to predict the need for mechanical ventilation (MV) in burn patients in order to improve their management and also to optimize resource ultilization. However, there are scant data on the predictors of MV in burn patients. The purpose of this study was to determine the risk factors for MV in a cohort of our burn patients.

Methods

A retrospective chart review was completed for patients admitted to Baskent University Burn Institute, from January 2003 through December 2008. Patients were divided into two groups based on whether they required MV during their first 3 days of hospitalization or not. After comparing the two groups for the recorded variables, a binary logistic regression model was developed using statistical and clinical significance to identify the risk factors for MV.

Results

Out of 160 patients, 33 patients (21%) required MV during their first 3 days of hospitalization. Patients who required MV were similar to those who did not in terms of demographic features. Patients who required MV were significantly different from those who did not regarding the mean burn percentage (56% ± 23% vs 19% ± 15%, P < 0.001), mechanism of burn (flame, 73% vs 37%, P = 0.001), inhalational injury (27% vs 3%, P < 0.001), and burn affecting head and neck (72% vs 37%, P < 0.001). Binary logistic regression revealed that inhalational injury (OR, 8.4; 95% CI, 1.1 to 66.1; P = 0.042) and burn percentage (OR, 1.1; 95% CI, 1.1 to 1.2; P < 0.001) were independent risk factors for MV.

Conclusions

Our results demonstrate that inhalational injury and burn percentage are predictors of MV in our series of burn patients. Particularly, patients with inhalational injury were eight times more likely to require MV during their first 3 days of hospitalization in our series.