[Factors affecting BK polyomavirus infection after kidney transplantation in post-school children and a predictive infection model]

Abstract

Objective: To analyze high-risk factors affecting BK polyomavirus (BKPyV) infection and to construct a prediction model for BKPyV infection in children after renal transplantation. Methods: The clinical data of 332 children who received allogeneic kidney transplantation in the First Affiliated Hospital of Zhengzhou University from January 2014 to March 2022 were retrospectively collected. According to the BKPyV load level, the dynamic change process of lymphocytes at different time points were analyzed. The factors that have potential influence on BKPyV infection were screened by Cox regression analysis, and the receiver operating characteristic curve (ROC) was used to evaluate the sensitivity and specificity of the predictive model of infection. Results: Among the 332 children, there were 215 males and 117 females; the age of transplantation was (12.2±3.9) years old; 37 cases were preschool (1-5 years old), and 295 cases were post-school age (6-18 years old). BKPyV load in 224 urine samples and 30 blood samples of children were detected. There were 9 cases of BKPyV-associated viruria and 3 cases of BKPyV associated viremia in pre-school children, 76 cases BKPyV associated viruria and 14 cases of BKPyV associated viremia in post-school children. Multivariate Cox regression analysis showed that higher body mass index (BMI) (HR=1.105, 95%CI: 1.020-1.197), antithyroglobulin (ATG) application (HR=2.196, 95%CI: 1.335-3.613), and higher tacrolimus concentration (HR=2.484, 95%CI: 1.298-4.753), higher natural killer (NK) lymphocyte count (HR=1.193, 95%CI: 1.009-1.411), higher CD14⁺⁺CD16^-cell count (HR=1.096, 95%CI: 1.024-1.173) were independent risk factors for BKPyV associated viruria in post-school children. Delayed graft function (DGF) (HR=4.993, 95%CI: 1.555-16.038), Acute rejection (AR) (HR=6.021, 95%CI: 1.930-18.787), higher CD14⁺⁺CD16^-cell count (HR=1.227, 95%CI: 1.081-1.392) were independent risk factors for BKPyV associated viremia in post-school children. The results of ROC curve analysis showed that combined BMI, immune induction drugs, tacrolimus concentration, NK cell count, and CD14⁺⁺CD16^-cell count predicted the occurrence of BKPyV associated viruria in post-school children after kidney transplantation at 0.5, 1, 2, and 5 years with area under curve (AUC) of 0.712 (95%CI: 0.626-0.798), 0.708 (95%CI: 0.612-0.804), 0.754 (95%CI: 0.668-0.840) and 0.767 (95%CI: 0.685-0.849). The sensitivity and specificity of the model were 64.9%, 61.4%, 61.6%, 55.8% and 70.9%, 72.4%, 76.0%, 84.0%, respectively. Combined with DGF, AR, and CD14⁺⁺CD16^-cell counts predicted the occurrence of BKPyV-associated viremia at 0.5, 1, 2, and 5 years after renal transplantation in post-school children with AUC of 0.791 (95%CI: 0.631-0.951), 0.744 (95%CI: 0.547-0.936), 0.786 (95%CI: 0.629-0.946) and 0.812 (95%CI: 0.672-0.948). The sensitivity and specificity of the model were 76.1%, 67.1%, 75.0%, 77.9% and 88.9%, 89.0%, 89.9%, 88.0%, respectively. Conclusions: The postoperative CD14⁺⁺CD16^-cell level can be used as an independent predictor of BKPyV infection in post-school children after renal transplantation. Combined BMI, immune induction drugs, tacrolimus concentration, NK cell count, CD14⁺⁺CD16^-cell count and combined DGF, AR, CD14⁺⁺CD16^-cell count show good fitting effect in predicting the occurrence of BKPyV-associated viruria and viremia after transplantation in post-school children respectively.