Home    中文  
 
  • Search
  • lucene Search
  • Citation
  • Fig/Tab
  • Adv Search
Just Accepted  |  Current Issue  |  Archive  |  Featured Articles  |  Most Read  |  Most Download  |  Most Cited

Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition) ›› 2025, Vol. 19 ›› Issue (06): 335-344. doi: 10.3877/cma.j.issn.1674-1358.2025.06.003

• Research Article • Previous Articles    

Risk prediction model for incomplete immune reconstitution in patients with human immunodeficiency virus infection/acquired immune deficiency syndrome after 4 years of anti-retroviral therapy

Yaqiong Zhang1, Yulin Zhang1, Ke Ding1, Xiaoxia Zhang1, Wenjia Hu1, Tielong Chen2, Shihui Song1, Yong Xiong1,()   

  1. 1 Department of Infectious Diseases, Zhongnan Hospital of Wuhan University, AIDS Research Center of Wuhan University, Wuhan 430071, China
    2 Phase I Clinical Research Laboratory, Zhongnan Hospital of Wuhan University, Wuhan 430071, China
  • Received:2025-04-30 Online:2025-12-15 Published:2026-02-12
  • Contact: Yong Xiong

Abstract:

Objective

To analyze the risk factors for incomplete immune reconstitution in patients with human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/AIDS), and to construct and validate a visual nomogram prediction model.

Methods

The clinical data of 302 patients with HIV/AIDS who were treated for the first time at Zhongnan Hospital of Wuhan University from January 1st, 2016 to November 30th, 2020 before the initiation of antiretroviral therapy (ART) were analyzed, retrospectively. The patients were divided into immunological response group (IRs group) (202 cases) and incomplete immune reconstitution group (INRs group) (100 cases) based on the CD4+ T lymphocyte count for four consecutive years after the initiation of ART. All participants were randomly assigned to the training set (212 cases) and the validation set (90 cases) with a ratio of 7∶3. Random forest, Lasso regression and multivariate Logistic regression analysis were used to screen the predictive factors and construct a nomogram prediction model. The performance of the model was evaluated by comparing the receiver operating characteristic (ROC) curves, calibration curves and clinical decision curves of the training set and the validation set.

Results

Among the 212 cases in the training set, 72 cases had incomplete immune reconstitution, with an incidence rate of 33.96%; among the 90 cases in the validation set, 28 cases had incomplete immune reconstitution, with an incidence rate of 31.11%. In the training set, the baseline CD4+ T lymphocyte count, white blood cell (WBC), platelet (PLT), and hemoglobin (Hb) levels of INRs group were significantly lower than those of IRs group (all P<0.05); the proportion of patients with co-infection of hepatitis B virus (HBV), age at ART initiation, baseline HIV load, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels of INRs group were significantly higher than those of the IRs group (all P<0.05); the route of infection and WHO stage between the two groups were with significant differences (both P<0.05). However, there were no statistically significant differences in gender, hepatitis C virus (HCV) antibody, ART regimen, body mass index (BMI), ART initiation delay, serum creatinine (SCr), triglyceride (TG) and cholesterol (TC) between the two groups (all P>0.05). The immune reconstitution incomplete predictors selected by random forest were baseline CD4+ T lymphocyte count, ART initiation age and WBC; the immune reconstitution incomplete predictors selected by Lasso regression were baseline CD4+ T lymphocyte count, ART initiation age and HBV co-infection. The multivariate Logistic regression analysis of the above factors (baseline CD4+ T lymphocyte count, ART initiation age, WBC and HBV co-infection) showed that a higher baseline CD4+ T lymphocyte count (OR=0.986, 95%CI: 0.982-0.990, P<0.001) was a protective factor for incomplete immune reconstitution in patients with HIV/AIDS , while a later ART initiation age (OR=1.035, 95%CI: 1.009-1.061, P=0.008) and HBV co-infection (OR=8.326, 95%CI: 1.836-37.765, P=0.006) were all independent risk factors for incomplete immune reconstitution in HIV/AIDS patients. Based on the above three predictors (baseline CD4+ T lymphocyte count, ART initiation age and HBV co-infection), a risk prediction model for incomplete immune reconstitution was constructed. The area under the ROC curve (AUC) of the training set and validation set were 0.882 and 0.825, respectively. The mean absolute errors of the calibration curves of the training set and validation set were 0.038 and 0.030, respectively; and the Brier scores were 0.135 and 0.154, respectively. The clinical decision curve (DCA) analysis indicated that the prediction model had clinical practical value in both the training set and validation set within the threshold probability range of 10% to 70%.

Conclusions

The nomogram model constructed based on baseline CD4+ T lymphocyte count, age at ART initiation and HBV co-infection has high performance in predicting immune reconstitution failure, and is helpful for clinical risk assessment and clinical decision-making.

Key words: Human immunodeficiency virus, Acquired immune deficiency syndrome, Antiretroviral therapy, Incomplete immune reconstitution, Prediction model

京ICP 备07035254号-20
Copyright © Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), All Rights Reserved.
Tel: 010-85322058 E-mail: editordt@163.com
Powered by Beijing Magtech Co. Ltd