Abstract
BACKGROUND AND AIMS: Acquired Immune Deficiency Syndrome (AIDS), caused by Human Immunodeficiency Virus (HIV), leads to severe immunodeficiency, making patients susceptible to opportunistic infections and malignancies. Lymphadenopathy is a common symptom in AIDS patients, reflecting immune system responses but also indicating potential disease progression. Distinguishing between benign and malignant lymphadenopathy is crucial for appropriate treatment. This study aimed to develop a diagnostic method for differentiating benign and malignant lymph nodes in HIV-infected patients using clinical and ultrasound data. METHODS: The study was conducted at Hangzhou Xixi Hospital from March 2016 to March 2024, including 149 HIV patients with confirmed lymphadenopathy. Ultrasound examinations were performed to assess lymph node characteristics, and biopsies were conducted for pathological confirmation. Statistical analysis involved the least absolute shrinkage and selection operator (LASSO) regression to identify significant predictors and construct a nomogram for predicting lymph node malignancy. RESULTS: The malignant lymph nodes had larger short and long diameters, and differences in shape, echogenicity, and hilum compared to benign lymph nodes. Lymphocyte count and T cell subsets were higher in malignant lymph nodes. The LASSO regression model identified short diameter, lymphocyte ratio, CD3(+) T cell count, and CD4(+) T cell ratio as significant predictors. The nomogram constructed based on these features demonstrated good predictive accuracy (AUC = 0.904). CONCLUSIONS: In conclusion, our study developed a diagnostic nomogram based on clinical and ultrasound data to differentiate benign and malignant lymph nodes in HIV patients. This tool had diagnostic accuracy and offers practical guidance for clinical management of HIV patients with lymphadenopathy.