Skip to main content

Factors associated with mortality in severe burn patients

Objective

To estimate mortality and the factors associated with mortality in severe burn patients.

Methods

We included burn patients older than 18 years admitted to the burns ICU from Hospital Universitario de Getafe from May 1992 to November 2005. The following variables were collected: age, gender, mechanism of burn, total and deep burn surface, inhalation injury and need of mechanical ventilation. To estimate the factors associated with mortality we split the population into two randomized cohorts: model cohort (including 80%) and validation cohort (20%). We performed a backward stepwise logistic regression entering the variables with P < 0.05 in the univariate analysis. The Hosmer–Lemeshow goodness of fit was used to evaluate the calibration. Discrimination was tested by measuring the area under the receiver-operating characteristic (aROC) curve.

Results

In the period of study 1014 patients were admitted. The mean age was 46 years (SD 20), 25% was female. The mean deep burn surface was 14% (SD 19%). Inhalation injury was diagnosed in 33%. Mechanical ventilation was required in 44%. Median length of stay in the unit was 9 days (interquartile range: 2, 24). Overall mortality was 15%. Factors associated with mortality are presented in Table 1. The obtained model had a satisfactory calibration (Hosmer-Lemeshow goodness of fit χ2 = 10.49; P = 0.10) and an excellent discrimination (aROC: 0.94; 95% CI: 0.90–0.98; P < 0.001).

Table 1

Conclusion

In our cohort of burn patients we observed a mortality of 15%. Factors associated with mortality were: age, deep burn surface, inhalation injury and mechanical ventilation.

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Peñuelas, O., Cerdá, E., Bustos, A. et al. Factors associated with mortality in severe burn patients. Crit Care 10, P425 (2006). https://doi.org/10.1186/cc4772

Download citation

Keywords

  • Public Health
  • Logistic Regression
  • Univariate Analysis
  • Mechanical Ventilation
  • Emergency Medicine