A nomogram risk prediction model for HIV-infected individuals with HBV infection was constructed based on clinical characteristics and T cell subsets

WU Congxia; XU Jingjing; WEN Xiaoping; ZHU Xiaohong; HUANG Zuoyu; CAO Li; WANG Juan; ZHAI Xiangjun; ZOU Meiyin

doi:10.3969/j.issn.1007-8134.2026.01.002

Abstract：Objective To analyze the factors associated with co-infection with hepatitis B virus(HBV)in human immunodeficiency virus(HIV)-infected patients,and to develop and validate a nomogram prediction model based on clinical characteristics and T-cell subsets.Methods A retrospective study was conducted on 150 HIV-infected patients admitted to Nantong Third People's Hospital between January 2017 and June 2022.Based on their HBV co-infection status,the patients were divided into co-infected group and non-co-infected group.Clinical data and laboratory indicators of patients were collected and compared between the 2 groups.Variables that showed significant differences in univariate analysis were screened for multicollinearity,and those without collinearity were included in multivariate logistic regression analysis to identify predictors factors for HBV co-infection.Using R software,a nomogram prediction model was constructed based on statistically significant variables from the multivariate logistic analysis,and internal validation was performed.Results Among the 150 HIV-infected individuals,22 cases(14.67%)were co-infected with HBV and 128 cases(85.33%)were not co-infected.There were statistically significant differences between the 2 groups in terms of age,family history of HBV,history of hepatitis B vaccination,CD4+T lymphocytes,and CD4+/CD8+levels(P<0.05),and there was no collinearity problem among these variables(VIF≤10,tolerance≥0.1).Multivariate Logistic regression analysis showed that Age(OR=3.846,P=0.029),family history of HBV(OR=46.750,P=0.001),and no history of hepatitis B vaccination(OR=3.278,P=0.035)were risk factors for concurrent HBV infection(P<0.01).CD4+T lymphocytes(OR=0.942,P=0.001)and CD4+/CD8+(OR=0.004,P=0.001)were protective factors(P<0.01).A nomogram prediction model was constructed based on the above 5 predictors.Internal validation showed that the area under the receiver operating characteristic curve was 0.955(95%CI:0.913-0.998),the calibration curve fitted well(P=0.353),Cox-Snell R2=0.689,Nagelkerke R2=0.39,suggesting that the model had good discrimination and calibration,and no overfitting occurred.Decision curve analysis shows that this nomogram has a relatively high clinical net benefit within a large threshold range.Conclusion Age,family history of HBV,history of hepatitis B vaccination,CD4+T lymphocytes,and CD4+/CD8+are all independent influencing factors affecting HBV infection in HIV-infected individuals.The nomogram prediction model constructed based on the above factors has good predictive efficacy.It can provide quantitative tools for the early stratified screening of the risk of HBV infection in HIV-infected individuals.