Cox Regression Model Analysis of Infection in Renal Transplants After Operation.

Abstract

The objective of this study was to explore the factors that affect infections after renal transplant, establishing the Cox model to forecast infection for patients of renal transplant. Data were collected from patients who had renal transplantation in Nanking Jinlin Hospital from January 2011 to April 2015 (n= 305 transplants). There were 296 individual data that could be used after deleting the people who were lacking some data, changing the main immunosuppressants during the first year, losing follow-up, and data writing that was not fully 1 year after the operation; 296 individuals were divided by 3:7. The 206 data of patients (7/10 of the total individuals) were used to analyze and build a model, and the rest of the data were used to verify the model, analyzing the 206 data with Cox regression, discovering the factors that affect the infection after renal transplant independently, building the model, and verification. Cox regression showed that there are three independent factors that affect infections after renal transplant: X3, the donor type (relative risk [RR]= 1.929, P= .037); X9, dialysis time (RR= 1.017, P= .032); and X13, human leukocyte antigen (HLA) match (RR= 0.257, P= .013). The model is: PI= 0.657X3+ 0.017X9- 1.359X13. All PI for the 206 individuals were calculated and then divided into three groups: the low-risk group, the median-risk group, and the high-risk group. The model was verified by calculating the PI for all 90 people. The log-rank test showed that the survival rates among these groups were significantly different (P< .001). Donor type, dialysis time, and HLA match are all factors that affect infection after renal transplant. Donor type and dialysis time were the dangerous factors for infection, but HLA match was the protecting factor. The model depends on these three factors and could forecast infection after renal transplant.