Severity score for predicting pneumonia in inhalation injury patients

Abstract Inhalation injuries contribute significantly to morbidity and mortality in both children and adults with burns. Pneumonia is a major compromising factor in these patients. The purpose of this article was to evaluate the characteristics, impact factors, incidence, morbidity, and mortality of pneumonia in inhalation injuries. Furthermore, a severity score has been formulated to help predict the probability of developing pneumonia following inhalation injuries. A retrospective study was performed of 214 patients, treated for inhalation injuries from 1999 to 2009 at the Burn Center in Chang Gung Memorial Hospital, Linkou, Taiwan. Patients’ characteristics, length of hospitalization, total burn surface area, initial PaO 2 :FiO 2 ratio, number of intubated days, bronchoscope grade, initial carboxyhemoglobin level (COHb) and mortality rate were recorded. A Student's t -test was used for comparison of inhalation injury patients with and without pneumonia and was also used for comparing a TBSA of >20% to those with a TBSA of ≤20% in patients with inhalation injury and pneumonia. Logistic regression analyses were utilized to create a severity score related to pneumonia. 129 patients with inhalation injury were included in the analysis. Overall, 38% (49/129) patients developed pneumonia. Pneumonia associated with inhalation injury occurred more often in patients with a TBSA>20% ( P P 2 :FiO 2 ratio (PaO 2 /FiO 2 ) was significantly lower in patients with pneumonia ( P P  20%. Logistic regression analysis was performed to find out the impact factors of pneumonia in inhalation injury patients and to set a severity score. Patients age >60 years, TBSA >20%, bronchoscope grade is 3 or 4, initial PaO 2 /FiO 2  ≦ 300 and initial COHb level>10% showed a significant difference ( P This study had identified the significant risk factors for potential development of pneumonia in a group of inhalation injury patients. The impact of these risk factors should be validated in further prospective trials to improve outcome or at least reduce the incidence of the surrogate diagnosis of pneumonia.