Construction and evaluation of risk prediction model for acute skin failure occurring in patients with septic shock in intensive care unit
Objective To construct a risk prediction model for acute skin failure occurring in patients with septic shock in ICU.Methods The clinical data from 491 patients with septic shock in ICU of a hospital from January 2020 to January 2022 was selected.The in-fluencing factors of acute skin failure in patients with septic shock in ICU were screened out by univariate and multivariate analysis.The early warning model was constructed based on influencing factors and the predictive effect of the model was verified.Results Acute skin failure oc-curred in 71(14.46%)of 491 patients with septic shock in ICU.Body surface area(OR=38.030),national early warning score(OR=2.457),sequential organ failure assessment score(OR=4.018),arterial lactate value(OR=7.404),peripheral perfusion index(OR=0.003),and capillary refill time(OR=36.864)were all influencing factors of acute skin failure occurring in patients with septic shock(P＜0.01 or P＜0.05).The prediction model was built based on above influencing factors and the nomogram was drawn.The area under ROC was 0.964,and risk prediction value corresponding to the maximum Youden index was 82.71%.The predicted critical score was≥233 points,while Hos-mer-Leme showed that the goodness of fit test was ?2=0.586,P=0.996.Actual curve of calibration curve almost coincided with the ideal curve in model.Conclusion In the study,based on the risk factors of acute skin failure in patients with septic shock in ICU,the nomogram prediction model has good consistency and differentiation,and the predicted risk value score is≥233 points,providing reference for risk warning and prevention of acute skin failure in patients with septic shock in ICU.
万方数据
维普
