The Heidelberg Kidney Donor Study (HeiKiD) is a prospective cohort study established to evaluate long-term outcomes after LKD and to enhance shared decision-making and informed consent.

The study was approved by the ethics committee of the University Hospital Heidelberg (S104-2011), and all participants provided written informed consent. Data collected prior to 2011 were obtained retrospectively from medical records. Since 2011, donors have been followed prospectively within a structured cohort study with predefined follow-up intervals.

Data handling complied with the European General Data Protection Regulation.

Eligible donors were aged ≥18 years, had completed the pre-donation evaluation, and had at least 5 years of follow-up after donation. The analysis included 632 caucasian donors who donated between 01/1991 and 06/2020 for recipients across 48 dialysis centers in southern and central Germany. Demographic data were obtained via structured questionnaires. Clinical and laboratory data were collected before donation, 1 year post-donation, and annually thereafter. In accordance with German living donor evaluation standards, all potential donors underwent comprehensive cardiovascular assessment before donation, including echocardiography and exercise ECG; if coronary heart disease was known or non-invasive testing was abnormal, coronary angiography was performed. Donors with clinically significant or uncontrolled cardiovascular disease were generally excluded, and only mild, stable, or previously treated cardiovascular conditions were accepted after interdisciplinary evaluation.

Primary outcomes were:

≥50% decline in estimated glomerular filtration rate (eGFR) from baseline to last follow-up

The threshold corresponds to CKD stage 3b and was selected because this level of kidney function has been consistently associated with a substantially increased risk of cardiovascular events, CKD progression, and mortality in large population-based studies [19, 20].

Secondary endpoints included the incidence of eGFR <30 mL/min/1.73 m², the need for renal replacement therapy, composite major adverse cardiovascular events (MACE), individual components of MACE (myocardial infarction, stroke, cardiovascular death, peripheral artery disease events (revascularization, amputation)), and death.

Beyond binary renal endpoints, analyses focused on early post-donation change, achieved post-adaptation kidney function, and long-term eGFR trajectories.

Analyses were stratified by donor age (<40 years, 40–60 years, and >60 years) to explore age effects. In addition, the impact of baseline comorbidity on longterm outcome was analyzed. A dedicated subanalysis was performed in donors aged ≥70 years.

Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range, IQR) and compared using the Student’s t-test or Mann–Whitney U test. Categorical variables were presented as counts and percentages and compared using the chi-square or Fisher’s exact test. Bias was minimized through consecutive enrollment and consistent data collection. Analyses were performed using complete case data without imputation.

Linear mixed-effects models were used to evaluate the association of age at donation with eGFR levels and longitudinal eGFR trajectories.

To separate the expected early post-nephrectomy adaptation from subsequent long-term changes, early eGFR change was quantified from baseline to 12 months after donation. Early change was defined as ΔeGFR0–12 (eGFR1y − eGFRbaseline) and, in sensitivity analyses, as percent change (%ΔeGFR). Determinants of early eGFR change were assessed using multivariable linear regression with prespecified predictors: baseline eGFR (CKD-EPI), age at donation (continuous), preexisting hypertension, BMI, sex, and donation era (grouped into 1991–2000, 2001–2010, 2011–2020). To facilitate comparison of effect sizes across predictors, standardized regression coefficients were calculated for continuous variables.

To assess determinants of achieved kidney function at long-term follow-up, multivariable linear regression models were fitted with eGFR at last follow-up as the dependent variable and adjusted for baseline eGFR, early eGFR change (ΔeGFR0–12), follow-up duration, age at donation, hypertension, BMI, and sex.

A total of 632 living kidney donors with a minimum follow-up of 5 years were included. Of these, 93 were aged <40 years (14.7%), 424 were aged 40–60 years (67.1%), and 115 were aged >60 years (16.6%), Table 1. The mean age at donation was 50.6 ± 10.6 years (range 19–77). Mean donor age increased from 43.9 years in 1990–1994 to 53.4 years in 2015–2020.

Overall, 62.5% were female, ranging from 53.9% among donors >60 years of age to 71.0% in the youngest group. Nearly two-thirds were related to their recipients (392/632; 62.0%), most frequently parent-to-child, especially mother-to-son (16.9%), Table 1.

The median follow-up time after donation was 12 years (IQR 9–16).

Cardiometabolic profiles showed clear age-related trends. Mean BMI was comparable across all groups, but overweight and obesity were more common in older donors. Smoking was more prevalent in younger individuals, while hypertension and the use of antihypertensives increased with age. Accordingly, office and 24-h blood pressure measurements were higher in older donors. Dyslipidemia and thyroid disorders also increased with age (Table 1).

Hematuria was observed in 9.6% of donors and occurred more frequently in female than male donors (12.8% vs. 4.2%). In cases of unexplained hematuria, a pre-donation kidney biopsy was performed in 14 donors. Histopathological findings were generally mild and major glomerulopathies were excluded. Thin basement membrane alterations were reported in 3 donors.

A history of malignancy was documented in 14 donors (2.2%), including breast cancer (n = 2), cervical cancer (n = 3), colon cancer (n = 1), papillary thyroid carcinoma (n = 1), renal cell carcinoma (n = 1), seminoma (n = 1), appendiceal carcinoid tumor (n = 1), pleural tumor (n = 1), and basal cell carcinoma (n = 3). Psychiatric disorders were reported in 16 donors (2.5%), predominantly depressive disorders (n = 12), followed by bipolar disorder (n = 2), panic attacks (n = 1), and adjustment disorder/burnout (n = 1). The prevalence of these comorbidities at baseline and at long-term follow-up is shown in Figure 1, both overall and stratified by age group (<40, 40–60, >60 years).

Baseline-eGFR decreased progressively with age and fell by 0.58 mL/min/1.73 m² per year of age (β = −0.58; 95% CI −0.67 to −0.48; p = 3.26 × 10⁻²), which corresponds to −5.76 mL/min/1.73 m² per decade.

During the study, CKDepi eGFR decreased from 95.8 ± 14.3 to 66.5 ± 16.6 mL/min/1.73 m² (Δ-30.6%, p < 0.001, Table 2). This decline was observed in all age groups but was most pronounced in donors >60 years (−35.1%, p < 0.001). Transitions between eGFR categories are shown in Figure 2. Within the first year after donation, eGFR decreased by an average of 26.0% (Δ–22.4% < 40 years; 24.9% 40–60 years; −31.8% > 60 years) and remained stable with minimal long-term change (+0.4%, −1.2%, +3.2%, respectively), Figure 3.

Across longitudinal follow-up, older age at donation was associated with lower eGFR values (−0.65 mL/min/1.73 m² per year of age, p < 0.001), whereas the rate of eGFR decline did not differ by age group (interaction p = 0.54). Donors with hypertension demonstrated the largest early decline and slightly lower long-term eGFR values (Figure 4).

A ≥50% eGFR decline at long-term was detected in 6.8% of donors, with the prevalence increasing sharply with age (4.8% < 40 years, 5.3% 40–60 years, 14.4% > 60 years, p = 0.0053), Table 3. Even among younger donors, the incidence ranged from 3.0% in donors with a favorable profile to 14.3% in donors with both hypertension and obesity. In older donors, the risk was 12.5% in donors without risk factors, 22.4% in donors with one risk factor, and 45.5% in hypertensive obese donors (Figure 5).

eGFR <45 mL/min/1.73 m² was identified in 8.0% of donors, with the prevalence increasing sharply with age (1.2% < 40 years, 5.3% 40–60 years, 20.4% > 60 years). The incidence of an eGFR <45 mL/min/1.73 m² rose further with the presence of risk factors such as hypertension and obesity. In older donors, the risk was 12.5% in donors without risk factors, 22.4% in donors with one risk factor, and 45.5% in hypertensive obese donors (Figure 5).

In the univariate analysis (Table 4), age at donation strongly predicted a ≥50% decline in eGFR: each additional year increased the probability (≈odds) of a substantial GFR loss by about 4.7% (OR 1.047, p = 0.006). Donors >60 years had a more than three times higher risk compared to donors <40 years (OR 3.34, p = 0.038).

Among preexisting comorbidities, hypertension and thyroid disease proved to be significant predictors, while hyperlipidemia, cardiovascular disease, and prediabetes were not predictive. The use of antihypertensive, lipid-lowering, and thyroid medications were also linked to a higher risk.

Each additional month since donation, increased the risk. Older age at follow-up also remained a strong predictor. At long-term follow-up, hypertension (both preexisting and new-onset) was associated with GFR loss, and the presence of malignancy also increased the risk.

Older age at donation increased the long-term risk of an eGFR <45 mL/min/1.73 m² by approximately 10.6% per years (OR 1.11, p < 0.001), Table 5. Donors older than 60 years had a substantially higher risk compared with those younger than 40 years (OR 6.62, p < 0.001). Pre-existing hypertension was associated with an approximately twofold increased risk of an eGFR <45 mL/min/1.73 m² (OR 2.07, p = 0.002. A similar increase was observed for hyperlipidemia (OR 2.40, p = 0.036). Thyroid disease was also associated with a significantly higher risk of reduced eGFR (OR 2.30, p = 0.020).

A lower baseline GFR as measured by MDRD, CKDepi, and Cockcroft-Gault formulas predicted a higher risk (all p < 0.001).

At long-term follow-up, older age continued to increase the risk by approximately 10.5% per year. Hypertension, hyperlipidemia, and malignant diseases were also significantly associated with increased risk.

In the multivariate analysis, age at donation, hypertension and baseline eGFR (OR 0.93), remained independent predictors (Figure 6).

In multivariable linear regression analyses with absolute early eGFR change (ΔeGFR 0–12 months) as the dependent variable, baseline eGFR showed the strongest association with early decline (standardized β = −0.29, p < 0.001). Age at donation was independently associated with early decline but with a smaller effect size (standardized β = −0.16, p = 0.002). Hypertension showed a modest additional association (standardized β = −0.10, p = 0.023), whereas BMI and donation era were not independently associated with early eGFR change. When early eGFR change was expressed as a percentage of baseline function, age at donation and baseline eGFR showed comparable but smaller effect sizes (standardized β = −0.16 and −0.12, respectively), while BMI and donation era remained non-significant.

Severe renal impairment was rare, an eGFR <30 mL/min/1.73 m² was observed in 1.2% and <15 mL/min/1.73 m² in 0.3% of donors. Two patients required renal replacement therapy. Preservation of renal function (≤20% decline) was noted in 20.0% overall (32.4% in donors <40 years, 22.0% in donors 40–60 years, and 15.4% in donors >60 years).

MACE occurred in 4.3% (n = 27) of donors (Table 2), increasing with age (mean age at donation 55 years) and longer follow-up.

Twenty-one deaths occurred predominantly among older donors (mean age 73 ± 8.7 years at death), with malignancies accounting for 52.4%. Median time from donation to death was 14 years (IQR 10–19). Overall long-term survival was high across all age groups (Table 3).

Glycemic control remained stable, although HbA1c levels increased slightly within the normal range from 5.45% ± 0.40% to 5.56% ± 0.58% (p < 0.001), mainly in donors aged 40–60 years.

Total cholesterol levels remained largely stable across the entire cohort, except for a small but significant increase in older donors (p = 0.0017), Table 2. Hemoglobin and hematocrit showed minimal changes, apart from a mild decline in hemoglobin among donors >60 years (p = 0.0092), Table 2.

An exploratory risk stratification was performed to describe patterns of long-term renal outcomes according to baseline donor characteristics. Using baseline variables that were independently associated with an eGFR <45 mL/min/1.73 m², including age (β = 0.062 per year), hypertension (β = 1.113), and baseline eGFR (β = −0.069 per mL/min/1.73 m²), a composite risk score was constructed to summarize their combined effects.

A two-dimensional risk map illustrated the joint association of age, hypertension, and baseline eGFR with the prevalence of eGFR <45 mL/min/1.73 m², showing a progressive increase with advancing age and lower baseline kidney function, particularly among hypertensive donors (Figure 7). When grouped into tertiles, the observed prevalence of eGFR <45 mL/min/1.73 m² increased from 1.0% in the lowest to 20.0% in the highest risk category (Figure 8A). Across age strata, predicted and observed prevalences were closely aligned (<40 years: 1.4% vs. 1.2%; 40–60 years: 5.8% vs. 5.3%; >60 years: 21.9% vs. 24.0%; Figure 7B).

Model discrimination was quantified for descriptive purposes and showed an AUC of 0.84 (95% CI 0.78–0.89; Figure 9).

Cardiovascular disease at baseline was uncommon overall but more frequent in older donors. Among donors >60 years, 19 of 115 (16.5%) had cardiovascular disease at baseline. Baseline kidney function was comparable between donors with and without cardiovascular disease. Long-term renal outcomes were also similar: eGFR <45 mL/min/1.73 m² occurred in 21.1% versus 21.9%, and a ≥50% decline in eGFR in 23.5% versus 13.8%, respectively. Major adverse cardiovascular events occurred in 10.5% versus 6.3%.

Among donors aged ≥70 years (mean 72, range 70–78 years), 55% were male. Most donated to their children (45%) or partners (35%). The mean follow-up period was 10.1 ± 2.5 years (range 7.1–14.7). Baseline eGFR (CKDepi) was 84.7 ± 9.0 mL/min/1.73 m², and 75% had at least one comorbidity, most commonly hypertension (65%), hyperlipidemia (30%), and cardiovascular disease (30%), Table 6.

After donation, the mean eGFR decreased by 37.7% and subsequently remained stable (year 1: 54.5 ± 20.0 mL/min/1.73 m² at year one, last follow-up: 53.6 ± 14.7 mL/min/1.73 m²; mean age 82 years). A ≥50% decline in eGFR occurred in 18.8%, and 25% had eGFR <45 mL/min/1.73 m², but none dropped below 30 mL/min/1.73 m² or required renal replacement therapy.

MACE occurred in 15%, while diabetes, malignancy, and psychiatric disorders were rare. One donor died from prostate cancer during long-term follow-up.