Risk of ESRD in Prior Living Kidney Donors

Jennifer L. Wainright; Amanda M. Robinson; Amber R. Wilk; David K. Klassen; Wida S. Cherikh; Darren E. Stewart


American Journal of Transplantation. 2018;18(5):1129-1139. 

In This Article

Abstract and Introduction


We studied End–Stage Renal Disease (ESRD) in living kidney donors (LKDs) who donated in the United States between 1994 and 2016 (n = 123 526), using Organ Procurement and Transplantation Network and Centers for Medicare and Medicaid Services data. Two hundred eighteen LKDs developed ESRD, with a median of 11.1 years between donation and ESRD. Absolute 20–year risk was low but not uniform, with risk associated with race, age, and sex and increasing exponentially over time. LKDs had increased risk of ESRD if they were male (adjusted hazard ratio [aHR]: 1.75, 95% confidence interval [95%CI]: 1.33–2.31), had higher BMI (aHR: 1.34 per 5 kg/m2, 95%CI: 1.10–1.64) or lower estimated GFR (aHR: 0.89 per 10 mL/min, 95% CI: 0.80–0.99), were first–degree relatives of the recipient (parent: [aHR: 2.01, 95% CI: 1.26–3.21]; full sibling [aHR: 1.87, 95%CI: 1.23–2.84]; identical twin [aHR: 19.79, 95%CI: 7.65–51.24]), or lived in lower socioeconomic status neighborhoods at donation (aHR: 0.87 per $10k increase; 95%CI: 0.77–0.99). We found a significant interaction between donation age and race, with higher risk at older ages for white LKDs (aHR: 1.26 per decade, 95%CI: 1.04–1.54), but higher risk at younger ages for black LKDs (aHR: 0.75 per decade, 95%CI: 0.57–0.99). These findings further inform risk assessment of potential LKDs.


In recent years, awareness has increased about long–term risks for living kidney donors (LKDs). The procedure, once believed to confer minimal surgical and long–term risks, needs additional study to assess longer–term risk for donors.[1–5] With increasing numbers of LKDs needing a kidney transplant after donating—currently 45–50 LKDs are added to the Organ Procurement and Transplantation Network (OPTN) kidney waiting list each year[6]—this issue has become critically important.

In addition to the small perioperative risks of donation[7–10] and mixed evidence for longer–term mortality,[3,7,11] there is new awareness of increased risk of nonrenal morbidity for LKDs.[2,5] Understandably, however, the largest focus has been on postdonation kidney function, with recent studies finding an increase in risk of End–Stage Renal Disease (ESRD) among LKDs relative to comparably healthy nondonors.[3,4] The existing body of evidence, while finding slightly different results in different cohorts, has identified predonation risk factors associated with higher rates of ESRD in LKDs: black or African American race, male sex, biological relatedness to recipient, and older age, blood pressure, obesity, and eGFR (estimated glomerular filtration rate) at donation.[1,3,4,12–17]

Despite important contributions of previous research, questions remain about long–term risks of donation and how to best counsel potential LKDs. Numerous studies[4,12–19] have found a relationship between race and risk of ESRD among LKDs, but the mechanisms responsible remain unclear. Despite evidence regarding the role of APOL1 as a risk factor in people of African descent,[20–25] disparities in access to health care among racial minorities in the United States[26] raise the question of whether differences are explained primarily by genetic factors[27,28] or also by other factors, such as socioeconomic status (SES). Garg and colleagues found that lower income was associated with higher risk of death or first major cardiovascular event for a cohort of mostly white LKDs,[11] and Lentine and colleagues investigated the role of SES on donor outcomes and found no significant associations, but cohort limitations likely limited the effect of SES.[16,17]

An issue intertwined with race in previous research is age at donation. Several studies have found associations between older age at donation and greater risk of ESRD for at least some donors.[1,4,14,15] Another important question, however, is the safety of allowing donation by younger donors, who have higher predonation lifetime ESRD risk.[29–32] Steiner documents the greater lifetime risk of ESRD for young African Americans, and because young donors are evaluated for donation before signs of renal disease would be expected to manifest, we would expect younger age at donation to be a significant risk factor for ESRD among black donors.[14,18,29,30,32,33] A recent study found older age at donation predicted greater risk of ESRD for nonblack donors, but did not find younger age at donation to be a risk factor for black donors.[14]

As with other factors, our understanding of the effect of donor relatedness to the recipient has evolved. In the early years of living donation, LKDs consisted almost exclusively of donors who were first–degree relatives of their recipients,[34] making study of this issue difficult. Some previous research[4] failed to identify a relatedness effect, but classified donors into 2 broad groups: blood relatives and unrelated donors. Newer research[14] found being a first–degree relative of the recipient was a significant predictor of ESRD among LKDs. This is an important finding, but additional questions remain about the relative risks among various types of relatives (ie, is donation riskier for parents or siblings of a recipient?), and no study to date has investigated potential effects of donation by the identical twin of a recipient.

Our study assesses LKDs' 20–year risk of ESRD and investigates other remaining questions about risk factors for ESRD among LKDs, such as the precise nature of the relationship between race and age at donation in terms of risk of ESRD among LKDs. We hypothesize there is greater 20–year risk associated with donation at a younger age for some donors, and this age effect varies by race. This study also investigates whether SES predicts ESRD in LKDs, independent of race. We hypothesize that SES predicts ESRD risk independent of the well–documented risk associated with black donors. Finally, we examine relatedness to the recipient on a more granular level (ie, parent, child, sibling, and identical twin) than explored in past research, and we hypothesize that more closely related donors have a higher risk of ESRD than do less closely related or nonrelated donors.