Annali italiani di chirurgia · 2026
Factors Associated With Intraoperative Acquired Pressure Injury in Total Knee Arthroplasty Patients: Development of Predictive Models.
СтатьяВалидационное исследованиеНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Zhou Jie, Yang Xuezhi, Xie Xiaoxiu, Zhou Hongli, Chen Xin, Liu Juanjuan
Аннотация
AIM: To construct and validate a risk prediction model for intraoperatively acquired pressure injury (IAPI) in total knee arthroplasty (TKA), thereby improving the accuracy of early diagnosis and intervention.
METHODS: This retrospective study included 546 patients who underwent elective total knee arthroplasty at Chengdu 363 Hospital Affiliated to Southwest Medical University and Chengfei Hospital. According to predefined inclusion and exclusion criteria, 278 cases from Chengdu 363 Hospital Affiliated to Southwest Medical University between January 2022 and December 2023 were used as the training set, while 118 cases from 2024 served as the internal validation set; 150 cases from Chengfei Hospital in 2024 were used as the external validation set. Feature variables were screened using multivariable logistic regression and Lasso regression analyses. Sensitivity, specificity, accuracy, F1-score (F1), and area under the curve (AUC) were used to evaluate discriminative performance. External validation was performed using AUC to evaluate generalizability. The optimal model was further interpreted by the Shapley additive explanation (SHAP) method to identify key risk factors.
RESULTS: Among the four machine learning algorithms tested, the gradient boosting decision tree (GBDT) model demonstrated the best discriminative performance (AUC 0.867, sensitivity 0.725, specificity 0.836, accuracy 0.788, and F1 value 0.747). The five most influential variables associated with IAPI risk were body mass index (BMI), Braden score, age, American Society of Anesthesiologists (ASA) classification, and surgical duration.
CONCLUSIONS: The GBDT-based prediction model, combined with the SHAP interpretation, effectively identifies risk factors for intraoperative IAPI in TKA. This model provides strong support for early clinical intervention and contributes to improving the outcomes of IAPI care.
METHODS: This retrospective study included 546 patients who underwent elective total knee arthroplasty at Chengdu 363 Hospital Affiliated to Southwest Medical University and Chengfei Hospital. According to predefined inclusion and exclusion criteria, 278 cases from Chengdu 363 Hospital Affiliated to Southwest Medical University between January 2022 and December 2023 were used as the training set, while 118 cases from 2024 served as the internal validation set; 150 cases from Chengfei Hospital in 2024 were used as the external validation set. Feature variables were screened using multivariable logistic regression and Lasso regression analyses. Sensitivity, specificity, accuracy, F1-score (F1), and area under the curve (AUC) were used to evaluate discriminative performance. External validation was performed using AUC to evaluate generalizability. The optimal model was further interpreted by the Shapley additive explanation (SHAP) method to identify key risk factors.
RESULTS: Among the four machine learning algorithms tested, the gradient boosting decision tree (GBDT) model demonstrated the best discriminative performance (AUC 0.867, sensitivity 0.725, specificity 0.836, accuracy 0.788, and F1 value 0.747). The five most influential variables associated with IAPI risk were body mass index (BMI), Braden score, age, American Society of Anesthesiologists (ASA) classification, and surgical duration.
CONCLUSIONS: The GBDT-based prediction model, combined with the SHAP interpretation, effectively identifies risk factors for intraoperative IAPI in TKA. This model provides strong support for early clinical intervention and contributes to improving the outcomes of IAPI care.