MODELO PREDITIVO PARA O DESENVOLVIMENTO DE CHOQUE SÉPTICO E HIPOVOLÊMICO EM PACIENTES DE UNIDADE DE TERAPIA INTENSIVA

Abstract

Shock affects most patients admitted to the intensive care unit (ICU) and is one of the main causes of death in these places. One of the challenges for reducing mortality is the early identification of shock for immediate initiation of treatment. Therefore, the objective of this study was to create and validate a model for predicting septic or hypovolemic shock from easily obtainable variables collected at admission of patients admitted to an intensive care unit. For this purpose, a predictive modeling study was developed with data from a concurrent cohort carried out in a hospital in the interior of northeastern Brazil. Patients aged 18 years or older without vasoactive drug use on the day of admission and who were hospitalized between November 2020 and July 2021 were included. Decision Tree, Random Forest, AdaBoost, Gradient Boosting and XGBoost classification algorithms were tested for the construction of the model and the evaluation metrics used were recall, precision and area under the Receiver Operating Characteristic curve (AUC-ROC). 720 patients were used to create and validate the model. The models showed high predictive capacity with an AUC-ROC of 0.979; 0.983; 0.979; 0.997 and 1.00 for the Decision Tree, Random Forest, AdaBoost, Gradient Boosting and XGBoost algorithms respectively. In this study, the predictive model created and validated showed a high ability to predict septic and hypovolemic shock from the moment of admission of patients to the ICU.