What is the clinical evidence from this study?

Study design: Observational. Population: Post-induction hypotension (n=5406). Intervention: Logistic regression model vs. Random forest, XGBoost, and neural network models. Primary outcome: Prediction of post-induction hypotension (AUROC) (AUROC 0.74, 95% CI 0.71-0.77).

February 22, 2025Open Access

Machine learning-based prediction of post-induction hypotension: identifying risk factors and enhancing anesthesia management

Resultado clave

A logistic regression model predicted post-induction hypotension with an AUROC of 0.74, demonstrating superior calibration and clinical utility compared to more complex machine learning algorithms.

Diseño del estudio

Tipo

Observational (n=5,406)

Multicéntrico

PICO estructurado

Can machine learning models using routine electronic medical record data accurately predict post-induction hypotension in non-cardiac surgical patients?

Población

5,406 patients undergoing non-cardiac surgeries whose routine clinical data were analyzed to develop and validate machine learning models for predicting post-induction hypotension.

Exposición

Machine learning predictive models (logistic regression, random forest, XGBoost, and neural network) using preoperative and baseline clinical features.

Comparador

Comparison among the different machine learning models.

Resultado

Post-induction hypotension (PIH), defined as a mean arterial pressure (MAP) less than 55 mm Hg between the induction of anesthesia and the start of surgery.safety

A logistic regression model utilizing routine electronic medical record data can effectively predict post-induction hypotension, providing a practical tool for personalized preoperative risk assessment.

Resultado numérico

Estimación del efecto: AUROC 0.74 (95% CI 0.71-0.77)

Tasa de eventos absoluta: 0.74% vs 0.71%

Limitaciones

Dataset from a single institutional database may introduce biases and limit generalizability
Reliance on noninvasive blood pressure measurements rather than invasive measurements
Lack of detailed information on vasoactive drug use
Did not include additional intraoperative features such as monitoring waveform data, intubation-related data, and ventilator parameters

Resumen

BACKGROUND: Post-induction hypotension (PIH) increases surgical complications including myocardial injury, acute kidney injury, delirium, stroke, prolonged hospitalization, and endangerment of the patient's life. Machine learning is an effective tool to analyze large amounts of data and identify perioperative complication factors. This study aims to identify risk factors for PIH and develop predictive models to support anesthesia management. METHODS: A dataset of 5406 patients was analyzed using machine learning methods. Logistic regression, random forest, XGBoost, and neural network models were compared. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC), calibration curves, and decision curve analysis (DCA). RESULTS: The logistic regression model achieved an AUROC of 0.74 (95% CI: 0.71-0.77), outperforming the random forest (AUROC: 0.71), XGBoost (AUROC: 0.72), and neural network (AUROC: 0.72) models. In terms of calibration, logistic regression demonstrated superior performance, as reflected by Brier Scores and calibration curves, followed by XGBoost, random forest, and neural network. Decision curve analysis indicated that the logistic regression model provided the greatest clinical utility among all models. Baseline blood pressure, age, sex, type of surgery, platelet count, and certain anesthesia-inducing drugs were identified as important features. CONCLUSIONS: This study provides a valuable tool for personalized preoperative risk assessment and customized anesthesia management, allowing for early intervention and improved patient outcomes. Integration of machine learning models into electronic medical record systems can facilitate real-time risk assessment and prediction.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo