The clinical and immunological features of CMV infection in rheumatic patients: a nested case-control study

Cytomegalovirus (CMV) is a common opportunistic pathogen that poses significant health risks to immunosuppressed individuals, including those with rheumatic diseases (RD). CMV infection in RD patients can exacerbate disease activity and complicate clinical outcomes, leading to increased morbidity and potentially poorer prognosis. Understanding the risk factors associated with CMV infection in RD patients is crucial for improving diagnosis, management, and treatment strategies.

This study is aimed at analyzing the potential associated risk factors for CMV infection in RD, constructing a prediction model to predict the risk of CMV infection in RD patients, and exploring the impact of CMV DNA load on prognosis and virus conversion of patients.

In this retrospective nested case-control study, 660 patients with RD who were hospitalized at Tongji Hospital from January 2017 to October 2023 were reviewed, of whom 146 were diagnosed with CMV infection during follow-up. Based on a 1:2 propensity score matching with sex, age, and disease type, 292 RD patients without CMV infection were included as controls. In addition, the CMV infection group was further divided into high and low DNA load groups based on the median CMV DNA load (1600 copies/mL). Multivariate logistic regression analysis was conducted to identify independent risk factors for CMV infection in RD patients. A nomogram prediction model was constructed and validated based on multivariate logistic regression results.

Symptoms such as fever and diarrhea were more common in patients with CMV infection. The proportion of peripheral blood lymphocyte (LYM) count, immunoglobulin (Ig) A, IgG, IgM, CD3⁺CD4⁺ T cells, CD3⁺CD8⁺ T cells, CD3^-CD19⁺ B cells, CD3⁺CD19^- T cells, and helper T (Th)/suppressor T (Ts) ratio below the lower limit of normal (LLN) was significantly higher in the CMV infection group compared to the non-CMV infection group. Multivariate logistic regression analysis revealed that IgG < LLN, Th/Ts ratio < LLN, high-dose glucocorticoid (GC) treatment in the past month, and use of cyclophosphamide (CTX) were risk factors for CMV infection in RD patients. The nomogram model based on the aforementioned identified risk factors showed a good prediction performance of CMV infection, with an area under the curve (AUC) of 0.898 (95% CI 0.890 ~ 0.979). Additionally, the PGA score significantly decreased after treatment compared to the period of CMV infection. Finally, high CMV DNA load was tightly correlated with down-regulated immunological features and poorer prognosis.

This study revealed that low IgG levels, decreased Th/Ts ratio, high-dose GC treatment, and use of CTX have important effects on CMV infection. Prediction of CMV infection based on these factors may help guide the early intervention in the potential CMV infection, and high CMV DNA load may indicate a poorer prognosis in RD patients. Key Points • RD patients with CMV infection are more immunosuppressive and exhibit severer clinical features. • A novel nomogram can help identify the potential CMV infection in RD patients. • High CMV DNA load is tightly correlated with down-regulated immunological features and poorer prognosis.