Analysis of influencing factors and construction of prediction model for infection in patients with open skin injuries
HUANG Jingkai
NONG Dahuan
HOU Honglei
ZHANG Hongbo
Abstract：Objective:To explore the influencing factors of infection after bleeding from open skin injuries and con-struct a column chart prediction model.Methods:A total of 375 patients with open skin injury and bleeding who were treated in hospital from February 2021 to March 2023 were enrolled in the modeling group.According to whether infec-tion occurred,they were divided into the occurrence group(n=45)and the non-occurrence group(n=330).Addition-ally,160 patients with the same type of injury admitted to Beihai Skin Disease Prevention Hospital from April 2023 to April 2024 were included in the validation group,which was similarly divided into the occurrence group(n=20)and the non-occurrence group(n=140).The general data of the modeling group were analyzed retrospectively,and the inde-pendent influencing factors of infection were screened by univariate and multivariate Logistic regression analysis.Based on the screening results,a nomogram prediction model was constructed,and the model was verified by calibration curve and ROC curve.Results:Univariate analysis showed that there was no significant difference in sex,weight index,smok-ing and drinking,injury site,dressing type and wound number between the two groups(P＞0.05).There were signifi-cant differences in age,underlying diseases,use of antibiotics,injury area,wound depth and time from injury to debride-ment(P＜0.05).ROC curve analysis showed that the area of injury,the depth of wound and the time from injury to de-bridement had certain predictive value for the infection after bleeding of patients with open skin injury,and the AUC of the modeling group was 0.817,0.737 and 0.880,respectively(P＜0.05).Logistic regression analysis showed that age,injury area,wound depth and time from injury to debridement were independent influencing factors of infection after bleeding of open skin injury(P＜0.05).The consistency index(C-index)of modeling group is 0.942(0.908～0.975),DCA results show that when the risk threshold is greater than 0.22,the model can provide standardized net in-come.Conclusion:Age,injury area,wound depth and time from injury to debridement are the independent influencing factors of infection after bleeding of open skin injury.Constructing nomogram prediction model can effectively predict the risk of infection,help to make accurate clinical decisions and take targeted preventive measures to reduce the inci-dence of infection.
Keywords：open skin injurybleedinginfectedcolumn chart
Publication Date：2025-10-10
Online Publishing Date：2025-10-29(First online date of this platform, not the publication date of the document)
Pages：6( 748-753 )