论文部分内容阅读
Introduction and aims: Predicting survival among victims of major burns trauma remains challenging, but can be useful to plan surgical management or palliation.The aims of this study were to retrospectively analyze burns patients to develop a mathematical model of predicting mortality based on admission characteristics.Materials and Methods: Data on all the burn patients presenting to Institute of Burn Research, Southwest Hospital, Third Military Medical University from January of 1999to December of 2008 were extracted from the departmental registry.Univariate associations with mortality were identified and independent associations were derived from multivariate logistic regression analysis.Using variables independently and significantly associated with mortality, a mathematical model to predict mortality was developed using the support vector machine (SVM) model.The predicting ability of this model was evaluated and verified.Results: There were 6220 cases included in this study with an overall mortality at hospital discharge of 1.8%.Variables at presentation independently associated with mortality were gender, age, total burn area, full thickness burn area, combination of inhalation injury, shock, period before admission, involvement of the head, face, neck and perineum, multiple injury, and injury mechanism(explosion or electricity).The sensitivity and specificity of logistic model were 99.75% and 85.84% respectively, with an area under the receiver operating curve of 0.989 (95% CI: 0.979-1.000; p<0.01).The model correctly classified 99.50% of cases.The subsequently developed SVM model correctly classified nearly 100%of test cases, which could not only predict adult group but also pediatric group, with pretty high robustness (92%-100%).Conclusions: It is possible to accurately predict mortality post burns injury using variables recorded at hospital presentation.These models need to be validated both internally and externally.In high volume burns centers, predictive models may be used clinically for forward planning of surgery or effective palliation.