Abstract:

Objective

To analyze the risk factors and develop a nomogram-based predictive model for early acute kidney injury （AKI） in patients with severe burns.

Methods

A retrospective analysis was conducted on 337 patients with severe burns （≥30% TBSA） admitted to Senior Department of Burns and Plastic Surgery in the Fourth Medical Center of PLA General Hospital between January 2015 and December 2023.The dataset was randomly split into a training set （70%） and a validation set （30%） using a fixed random seed （1222） to ensure reproducibility.Least absolute shrinkage and selection operator （LASSO） regression and multivariate logistic regression analyses were used to identify predictive variables for constructing the early AKI risk nomogram.The model's discriminative ability， calibration， and clinical utility were evaluated using the receiver operating characteristic （ROC） curve， calibration curve， and decision curve analysis （DCA）.

Results

Greater body weight， larger burn area of above third-degree， presence of shock on admission， prolonged time from injury to hospital admission， and higher blood glucose and white blood cell counts within 48 hours of admission were independent risk factors for early AKI in patients with severe burns.The nomogram model， based on these six variables， achieved AUC values of 0.828 （95%CI： 0.770-0.886） in the training set and 0.826 （95%CI： 0.743-0.909） in the validation set.The calibration curve analysis yielded P-values of 0.787 and 0.125， indicating good agreement between predicted and observed outcomes.DCA demonstrated that the nomogram model provided a high net clinical benefit.

Conclusion

The nomogram prediction model score based on body weight， burn area of above third-degree， whether shock occurred at admission， time from injury to hospital admission， and blood glucose and white blood cell counts within 48 hours of admission can be used to predict early AKI in severe burn patients.

Key words: Burns, Acute kidney injury, Risk factors, Prediction model