BACKGROUND: Infection, which is the most serious complication in burns has been little studied. There the aim of the present was to study this complication and its risk factors in patients admitted into a Burn Intensive Care Unit. METHODS:Patients admitted from 1988-1992 were prospectively studied. The characteristics of the patients were systematically collected on admission (i.e. age, body area burned [BAB], previous disease) as were those of the patient's evolution in the Burn Unit (i.e. infection, manipulation, and therapy). Prior to analysis the patients were randomly divided into two groups: a "study" group with 455 patients and a "validation" group of 163 patients. A uni and bivariate descriptive study was performed in addition to a multivariate study by unconditional multiple logistic regression obtaining one predictive and one explicative equation. The former was thereafter validated in a sample of 163 patients. RESULTS: No significant differences were observed between the two groups which were made up of severely ill patients (20% BAG 2/3 with dermodeep flame burns, mean age of 43 years) with a similar rate of infection 7.2% and 8.6%, respectively. The ethiologic agent was Pseudomonas aeruginosa in both groups. On applying logistic regression a predictive equation of burn infection and the main risk factors of infection were obtained: BAB (between 15-30%, odds ratio of 4, being 15 if BAB > 30%), duration of central catheterization prior to infection (1-15 days, odds ratio of 6, being 31 if it is longer than 15 days) and preinfection stay (between 15-30 days, odds ratio of 3, being 10 if the stay is longer than 30 days). The validation was obtained on applying the predictive equation to the patients in both groups to calculate the number of infected patients. In both cases the prediction was correct. CONCLUSIONS: The main risk factors of burn infection are: body area burned, central catheterization and preinfection hospital stay. The equation which adequately predicts the probability of infection based on the characteristics of the burned patients is a good tool for reducing the frequency of infections in burned patients.
RCT Entities:
BACKGROUND:Infection, which is the most serious complication in burns has been little studied. There the aim of the present was to study this complication and its risk factors in patients admitted into a Burn Intensive Care Unit. METHODS:Patients admitted from 1988-1992 were prospectively studied. The characteristics of the patients were systematically collected on admission (i.e. age, body area burned [BAB], previous disease) as were those of the patient's evolution in the Burn Unit (i.e. infection, manipulation, and therapy). Prior to analysis the patients were randomly divided into two groups: a "study" group with 455 patients and a "validation" group of 163 patients. A uni and bivariate descriptive study was performed in addition to a multivariate study by unconditional multiple logistic regression obtaining one predictive and one explicative equation. The former was thereafter validated in a sample of 163 patients. RESULTS: No significant differences were observed between the two groups which were made up of severely ill patients (20% BAG 2/3 with dermodeep flame burns, mean age of 43 years) with a similar rate of infection 7.2% and 8.6%, respectively. The ethiologic agent was Pseudomonas aeruginosa in both groups. On applying logistic regression a predictive equation of burn infection and the main risk factors of infection were obtained: BAB (between 15-30%, odds ratio of 4, being 15 if BAB > 30%), duration of central catheterization prior to infection (1-15 days, odds ratio of 6, being 31 if it is longer than 15 days) and preinfection stay (between 15-30 days, odds ratio of 3, being 10 if the stay is longer than 30 days). The validation was obtained on applying the predictive equation to the patients in both groups to calculate the number of infectedpatients. In both cases the prediction was correct. CONCLUSIONS: The main risk factors of burn infection are: body area burned, central catheterization and preinfection hospital stay. The equation which adequately predicts the probability of infection based on the characteristics of the burned patients is a good tool for reducing the frequency of infections in burned patients.