基于Lasso-列线图构建脑部手术患者医疗器械相关压力性损伤风险预测模型OA

Objective To investigate the risk factors for medical device-related pressure injury(MDRPI)in patients undergoing brain surgery and to establish a risk prediction model.Methods Using convenience sampling,a total of 482 patients who underwent brain surgery in the Department of Neurosurgery,the First Affiliated Hospital of Xinjiang Medical University from June 2023 to April 2024 were selected as the study subjects and divided into an MDRPI group(107 cases)and a non-MDRPI group(375 cases)according to the occurrence of MDRPI.General information,clinical data,initial laboratory data upon admission,device usage data,and surgery-related data were collected and compared between the two groups.Lasso regression was used to screen for predictors.Subsequently,multivariate Logistic regression analysis was performed to examine the association between each predictor and the occurrence of MDRPI in patients undergoing brain surgery,and a Nomogram risk prediction model was constructed.The receiver operating characteristic(ROC)curve,calibration curve,decision curve,and Hosmer-Lemeshow(H-L)test were used to validate the model's discrimination,calibration,clinical benefit,and goodness-of-fit.Results The incidence of MDRPI among the 482 study subjects was 22.2%(107/482).Statistically significant differences were observed between the MDRPI group and the non-MDRPI group in Acute Physiology and Chronic Health EvaluationⅡscore,presence of edema,use of endotracheal tube,duration of device use,duration of anesthesia,intraoperative blood loss,albumin,hemoglobin,oxygenation index,C-reactive protein,interleukin-6,and blood lactate(P<0.05).After including all independent variables in the Lasso regression analysis,five core predictors were identified.Further multivariate Logistic regression analysis showed that the use of endotracheal tube(OR=2.836),duration of device use(OR=1.438),duration of anesthesia(OR=1.026),intraoperative blood loss(OR=1.015),and blood lactate(OR=2.771)were risk factors for MDRPI in patients undergoing brain surgery(P<0.05).The area under the curve(AUC)of the constructed nomogram prediction model for MDRPI in patients undergoing brain surgery was 0.899(95%CI=0.869-0.934),with a sensitivity of 0.850 and a specificity of 0.816,indicating good model discrimination.The calibration curve showed a high degree of agreement between the predicted probabilities and the actual incidence,indicating ideal calibration.The H-L test suggested a good model fit(χ2=3.803,P>0.05).The decision curve showed that the model had a high clinical net benefit.Conclusion The use of endotracheal tube,duration of device use,duration of anesthesia,intraoperative blood loss,and blood lactate are risk factors for MDRPI in patients undergoing brain surgery.The constructed nomogram risk prediction model demonstrates good predictive performance.