Comparing the effect of pre-anesthesia clonidine and tranexamic acid on intraoperative bleeding volume in rhinoplasty: a machine learning approach.

Effectively managing intraoperative blood loss during surgery is crucial, as it significantly impacts patient outcomes and the surgeon's reputation. This importance is amplified in rhinoplasty, where minimizing bleeding is vital for optimal results. Consequently, this research investigates how various preoperative medications, such as clonidine and tranexamic acid (TXA), affect blood loss during rhinoplasty, with an emphasis on enhancing surgical efficiency and patient safety. This retrospective cross-sectional study examined data from 120 patients who underwent rhinoplasty from 2019 to 2022. The data were preprocessed and analyzed using various regression models, including Linear regression, random forest (RF), support vector regression (SVR), Extreme Gradient Boosting (XGBoost), Gradient Boosting, Ridge, and least absolute shrinkage and selection operator (LASSO), to forecast blood loss associated with the use of clonidine and TXA. The findings were then compared to determine the most effective medication and the best predictive model. The results revealed that the Linear and Ridge regression algorithms outperformed all other models based on three evaluation metrics: mean absolute error (MAE), mean square error (MSE), and R-squared. The estimated blood loss during rhinoplasty was 112 mL with clonidine, 132 mL with TXA, and 157 mL without any medication. This study highlights the significance of preoperative medications in controlling blood loss during rhinoplasty. The findings indicate that regression models can accurately predict blood loss, with clonidine demonstrating a significant impact on reducing hemorrhage. This reduction enhances the surgeon's visibility and may contribute to a shorter duration of the surgical procedure.