Predicting Mortality in Kidney Transplant Recipients

Accurately predicting kidney recipient risk of death has a crucial interest because of the organ shortage, the need to optimize allograft allocation by identifying high-risk patients who may not benefit from a transplant and improve the clinical decision-making after transplant to ensure that each patient survives as long as possible.

However, according to a literature review the investigators performed, studies attempting to develop a kidney recipient death prediction model suffer from many shortcomings, including the lack of key risk factors, use of biased registry data, small sample size, lack of external validation in different countries and subpopulations, and short follow-up.

The present study thus aimed to address these limitations and develop a robust, generalizable kidney recipient death prediction model.

The number of individuals suffering from end-stage chronic renal disease (ESRD) worldwide has increased over time, exceeding seven million of patients in 2020. For individuals with ESRD, kidney transplantation is the best treatment in terms of patient survival, quality of life and from a cost-effective standpoint, as compared with dialysis, even in comorbid or elderly populations.

Although the number of kidney transplantations performed each year has increased as well, it follows a lower pace than the increase of individuals on the waiting-list, resulting in an organ shortage. There is therefore a need to optimize allograft allocation by identifying the high-risk patients who may not benefit from a transplant and improve the clinical decision-making after transplant to ensure that each patient survives as long as possible.

In this context, a kidney recipient death prediction model may improve transplant clinical practice, allowing for the ability to evaluate the individual risk of post transplant mortality, already before undergoing transplantation, thereby guiding decision making. However, developing such a model is a very difficult task, as death after kidney transplantation depends on many parameters, such as donor age, history or cause of death, imaging parameters, patients' past medical history (e.g. diabetes, dialysis duration, hypertension), patients' biological parameters, as well as the function of the allograft, which depends on patients' immunological factors, or allograft related parameters such as HLA mismatches or cold ischemia time.

The goal of the present study was therefore to identify the determinants of death after kidney transplantation, and to develop and validate a prediction model that would help optimize allograft allocation and post-transplant patient management, using a large, international, highly phenotyped cohort of kidney recipients with extensive data collection and long-term follow-up.