Donation by older donors#

The demographic shift toward older kidney donors necessitates improved characterization of risks to enhance consent processes and outcomes.

Introduction#

The number of older adults (≥50 years) undergoing live donor nephrectomy has quadrupled over the last two decades, a trend expected to persist given demographic shifts, updates in kidney transplant allocation, and changes to Kidney Disease: Improving Global Outcomes (KDIGO) guidelines. Despite this increase, critical knowledge gaps persist in understanding how older donors cope with the physical stress of nephrectomy, particularly the role of age-related kidney changes (glomerular senescence) and the body’s ability to adapt following a 50% nephron mass loss. Approximately 80% of older donors experience a post-donation GFR <60 ml/min/1.73m², meeting the definition of chronic kidney disease (CKD), placing them at increased risk of frailty and hospitalization as they progress toward CKD. However, no studies have identified clinical signatures of resiliency in this unique population, even as nephrectomy directly impacts their renal system and overall health.

This thesis focuses on addressing these gaps by prioritizing the patient’s central question, “Can I safely donate my kidney?” over population-level metrics. The scientific aims include: (1) quantifying donation-attributable risks of perioperative death, (2) long-term risks of ESRD and (3) mortality; (4) describing hospitalization prevalence in older donors compared to non-donors; and (5) implementing a risk calculator for perioperative death, long-term ESRD and mortality risks, and sentinel hospitalization in older donors.

This thesis will frame its work in the context of various areas of inquiry, including kidney disease, aging research, and data science. The results of all four aims are delivered in an innovative tool—an app—to enhance provider-patient dialogue about kidney donation risks (aim 5). But the architecture of the app offers so much more that will be discussed in the appendix.

This thesis addresses the critical need to improve understanding of perioperative and long-term risks for older living kidney donors. By identifying resiliency markers and individualizing risk assessment, the study aims to empower older adults and providers with evidence-based tools to make informed decisions in a growing and vulnerable donor population.

Perioperative Mortality#

The 90-day risk of death following living kidney donation has been estimated at 3 per 10,000. ² Notably, some of these deaths were due to homicide, an outcome unlikely to reflect a perioperative risk directly attributable to nephrectomy. ³

To contextualize these risks, we introduce a control population to quantify the baseline 90-day mortality risk faced by healthy eligible donors who do not proceed with donation. Using data from the Organ Procurement and Transplantation Network (OPTN) and the National Health and Nutrition Examination Survey (NHANES), we estimate the 90-day risk of death for both donors and their healthy nondonor counterparts through registry linkages to the National Death Index. Cumulative incidence estimates derived from Cox regression are presented in an accompanying app (see App Features below).

For white female nondonors aged 40 years or younger who self-reported “Very Good or Excellent” health at the time of survey, the 90-day risk of death was estimated at 1.3 per 10,000 (95% CI: 0.5-3.5 per 10,000). Among comparable donors, the 90-day risk of death was 2.2 per 10,000 (95% CI: 1.0-4.7 per 10,000). Among nondonors aged over 50 years, the 90-day risk of death was no higher than that observed in nondonors with similar demographic and health characteristics.

The nephrectomy-attributable risk for a typical young donor is approximately 0.9 per 10,000, substantially lower than the risk currently cited during the informed consent process. For older donors, the nephrectomy-attributable risk may be even lower, though it remains insufficiently quantified due to the limited availability of data.

Long-term Risk of ESRD#

Donation-attributable risk of end-stage renal disease (ESRD) has been estimated at 26.9 per 10,000 at 15 years. Subgroup estimates for white, black, and Hispanic donors have also been reported. ⁴ However, personalized risk estimates that take into account age and other associated risk factors remain unknown. ⁵

We compare the risk of ESRD in kidney donors (using OPTN data) with that of a cohort of nondonors (using NHANES data). See Perioperative Mortality above for more information. Cumulative incidence estimates derived from Cox regression are presented in an accompanying app (see App Features below).

The maximum follow-up period was 15 years, with a median follow-up of 7.6 years for kidney donors (interquartile range [IQR], 3.9–11.5) and 20 years for matched healthy nondonors (IQR, 18–35). The estimated risk of ESRD at 6.5 years for the typical 40 year old white female with SBP 120 mmHg, DBP 80mmHg, BMI 24 kg/$m^2$, eGFR 95 ml/min/1.73$m^2$, urine albumin/creatinine ratio of 4mg/g, with a college level education, “Very Good” self-rated health status, no history of diabetes, hypertension, and smoking was significantly higher in kidney donors (5.6 per 10,000; 95% CI, 1.5–20.1) compared to matched healthy nondonors (1.3 per 10,000; 95% CI, 0.8–2.8). Donation by older donors such as the one seen above: 84 year old white male, SBP 140 mmHg, BMI 24 $kg/m^2$, eGFR 80ml/min/1.73$m^2$, no history of diabetes or smoking, but has well-controlled hypertension. He is a college graduate. We estimate his risk if his 6.5 year risk at (34.6 per 10,000; 95% CI, 2.2–552.0) compared to the counterfactual scenario in which he had no donated (9.2 per 10,000; 95% CI, 5.2–16.1).

Our personalized donation-attributable risk sets a standard for informed consent among the diverse range of those who seek to donate a kidney. But uncertainty of our estimates for such a rare donor phenotype highlights the challenges that remain.

Long-term Risk of Death#

Coming Soon!

Hospitalization Risks#

Hospitalization may be a sentinel event signaling risk for adverse outcomes, including end-stage renal disease (ESRD) and mortality, among living kidney donors (LKDs). However, only two years of follow-up are mandated for LKDs, limiting long-term risk characterization. To address this, we used a multicenter retrospective cohort study of LKDs with over 40 years of follow-up to identify factors associated with patient-reported all-cause hospitalization.

Patient factors were captured from self-report and supplemented by linkage to the SRTR database. Among an cohort of 3,084 LKDs, 2,251 (73%) donors who donated between May 1968 and December 2019 responded to the survey.

Overall, 1575 (70%) reported any hospitalization a median (interquartile range) of 13 (10-20+) years post-nephrectomy, with surgery/procedure as the most common cause (57%). The cumulative incidence of hospitalization was 47.2% (95%CI: 45.5-48.9) at 20 years post-donation for a 40 year old white female donor with BMI of 25 $kg/m^2$, SBP of 120 mmHg, DBP of 80 mmHg, eGFR of 95 ml/min/1.73$m^2$ (see App Features below). In a parsimonious model, age at donation (aHR 1.01–1.13–1.27, p=0.04), female sex (aHR 1.07–1.39–1.80, p=0.01), and post-donation diabetes/hypertension (aHR 1.22–1.52–1.88, p<0.01) were associated with hospitalization (see appendix).

While self-reported, the frequency of hospitalization among LKDs beyond two years post-nephrectomy suggests longer follow-up may be warranted for this population. Furthermore, surveillance and follow-up should be emphasized in populations at higher risk of developing diabetes and hypertension after nephrectomy. The absence of a control population with ascertainment of hospitalization remains a critical limitation of this study.

App Features#

Explore the interactive visualizations below, designed to personalize risk profiles for older kidney donors. The app integrates data-driven insights from all four prior aims.