Background: Living-donor kidney transplantation yields the best outcomes for patients with kidney failure. As donor safety is required and literature limited, there is a need to evaluate the impact of pre-donation characteristics on long-term kidney function in living donors. This study aims to 1) identify characteristics influencing longitudinal living-donor kidney function; 2) construct a prediction model to forecast kidney function in prospective donors; and 3) incorporate the model into a web application. The model will be validated against the Dutch donor population and will compare living-donor kidney function trajectories with matched controls from the general population. Here, we present the results of the first aim.
Method: All living donors who donated a kidney between 1981 and 2019 in our center were included. The dataset was compiled from our center’s database, the Dutch Organ Transplantation Registry (NOTR), and the electronic hospital information system. The evolution of post-donation kidney function was investigated using linear mixed-effects models, correcting for pre-donation characteristics as detailed in the results. Kidney function was defined as the estimated glomerular filtration rate (eGFR) using the 2009 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.
Results: A total of 2,212 individuals underwent donor nephrectomy. Among them, 2,181 donors had at least one post-donation eGFR measurement, contributing to 19,109 observations between March 31, 1982 and July 23, 2023. Figure 1 shows the time evolution of post-donation eGFR, considering different donor ages (A), pre-donation eGFR values (B) and donation periods (C). Post-donation eGFR increased in the first decade and slowly decreased thereafter within a difference of 4 mL/min/1.73 m2 over 20 years. The overall and time effects of pre-donation eGFR (B) and donation period (C) were both significant (p < 0.001). The overall effect of donor age (A) was significant as well (p < 0.001). However, its interaction with time was not significant (p = 0.28). The effects of BMI, kinship with the recipient, smoking at donation, and ethnicity were statistically significant (p < 0.01). However, differences in eGFR trajectories were less than 4 mL/min/1.73 m2 and are deemed as not clinically relevant. The effect of ethnicity is more difficult to interpret and needs further analysis. Characteristics such as cardiovascular disease, antihypertensive medication, malignancy, sex, and urological conditions did not have a significant effect (p > 0.05).

Conclusion: In our living-donor population, kidney function decreased less than 4 ml/min/1.73 m2 over 20 years after donation. Apart from expected influencing characteristics, such as donor age and pre-donation eGFR, other donor variables did not show a clinically relevant effect. These results offer encouragement for developing a prediction model to enhance individualized understanding of long-term kidney function following donation.