ABSTRACT
Background
Human donor kidneys release (pro)renin, erythropoietin (EPO), active vitamin D, and urodilatin during normothermic machine perfusion (NMP). However, whether the endocrine function of donor kidneys is associated with post‐transplant kidney function is unclear.
Methods
We studied 28 donor kidneys, including seven from donation after brain death (DBD) donors and 21 from donation after circulatory death (DCD) donors. Prior to transplantation, we measured levels of (pro)renin, EPO, 1,25(OH)2D in the perfusate, and urodilatin in urine during NMP. Hormone release rates were compared between kidneys with and without delayed graft function (DGF), and correlations were assessed between hormone release rates and donor characteristics and transplant outcome, including DGF duration, serum creatinine levels at 1‐week post‐transplant, and estimated glomerular filtration rate at 1‐month post‐transplant.
Results
DBD kidneys secreted significantly less EPO and more active vitamin D than DCD kidneys. Kidneys with DGF exhibited significantly higher release rates of active vitamin D and lower release rates of urodilatin compared to those without DGF. In addition, EPO release rate was positively correlated with serum creatinine levels at 1‐week post‐transplant. Finally, urodilatin release rates were negatively correlated with DGF duration and positively correlated with urine output.
Conclusions
Urodilatin release in urine and EPO and active vitamin D release in perfusate during NMP may serve as potential biomarkers for predicting early post‐transplant outcomes.
Trial Registration: ClinicalTrials.gov identifier: NCT04882254
Keywords: donor kidney, early post‐transplant outcome, endocrine function, hormone release, normothermic machine perfusion
Abbreviations
- 1,25(OH)2D
1,25‐dihydroxyvitamin D / active vitamin D
- 25(OH)D
25‐hydroxyvitamin D
- CIT
cold ischemia time
- DBD
donation after brain death
- DCD
donation after circulatory death
- DGF
delayed graft function
- ECD‐DBD
expanded criteria donation after brain death
- eGFR
estimated glomerular filtration rate
- EPO
erythropoietin
- HMP
hypothermic machine perfusion
- IQR
interquartile range
- IRMA
immunoradiometric assay
- NMP
normothermic machine perfusion
- PNF
primary non‐function
- RAS
renin‐angiotensin system
- WIT‐1
the initial period of warm ischemia time experienced by DCD kidneys
1. Introduction
Ex situ normothermic machine perfusion (NMP) is a transplant organ preservation technique in which warmed (37°C), oxygenated perfusate, enriched with cellular or acellular oxygen carriers, is circulated through the donor organ. Studies are ongoing to explore whether NMP can improve the outcome of kidney transplantation [1, 2, 3]. Furthermore, NMP mimics physiological conditions and therefore supports metabolic activity and kidney function, which allows for viability and functional assessment (e.g., urine production, hormone secretion) of donor kidneys prior to transplantation [1, 4‐6].
The growing shortage of donor kidneys for transplantation has led to increased interest in the use of suboptimal kidneys from expanded criteria donation after brain death (ECD‐DBD) donors and donation after circulatory death (DCD) donors [7, 8]. However, the quality of these kidneys and their post‐transplant outcome vary, resulting in unpredictable post‐transplant function. Thorough quality assessment tools are required to accurately assess whether suboptimal donor kidneys are suitable for transplantation; however, no such tools exist today. Biomarkers such as lactate dehydrogenase, kidney injury molecule‐1, and neutrophil gelatinase‐associated lipocalin have shown promise in assessing donor kidney quality during NMP [9]. However, these biomarkers serve as indicators of kidney injury and do not per se predict kidney function. A recent trial found that changes in the glutathione S‐transferase (GST)‐Pi concentrations in the perfusate during NMP were strongly correlated with graft function 12 months post‐transplant [10]. Research on the development of assays for functional assessment of kidneys during NMP has demonstrated that porcine kidneys can be classified into functional, limited functional, and non‐functional categories based on inulin clearance during NMP [11]. Additionally, urine output during NMP, a key parameter in Hosgood and Nicholson's quality assessment score, has been shown to be significantly higher in transplanted kidneys compared to those that were declined for transplantation [12, 13].
In addition to urine production, hormone secretion is also a critical component of the kidney's function. The kidneys produce (pro)renin, erythropoietin (EPO), 1,25‐dihydroxy vitamin D [1,25(OH)2D], and urodilatin, amongst other hormones [14]. Renin and its precursor prorenin are synthesized and secreted by juxtaglomerular cells [15]. Renin is the rate‐limiting enzyme of the renin‐angiotensin system (RAS), which regulates blood pressure, fluid and electrolyte balance [16]. Studies have shown elevated renin expression in individuals with kidney disease [17]. Additionally, changes in renin production and activity are associated with impaired kidney function in chronic kidney disease [18, 19, 20]. EPO is primarily synthesized by EPO‐producing cells in the kidneys and is induced by hypoxia. It is crucial for the production of red blood cells [21]. A decline in kidney function is often accompanied by decreased production of EPO [22, 23], and the loss of EPO‐producing cells is associated with renal fibrosis [24]. Active vitamin D [1,25(OH)2D] is derived from 25‐hydroxy vitamin D [25(OH)D] after conversion by proximal tubular cells in the kidney. 1,25(OH)2D plays a vital role in maintaining calcium homeostasis and regulating bone formation [25], and its deficiency is associated with kidney function loss [26, 27]. Studies have shown correlations between plasma levels of EPO and 25(OH)D in kidney transplant recipients and post‐transplant graft function [28, 29]. Urodilatin is a natriuretic peptide synthesized by distal tubular cells and released into the pro‐urine to regulate sodium and water balance [30]. The administration of urodilatin has demonstrated beneficial effects in experimental acute kidney failure and in kidney transplantation [30, 31, 32]. Together, these studies emphasize the potential significance of hormone release capacity in assessing kidney function and in predicting transplant outcome.
In our previous publication, we demonstrated that human kidneys exhibit endocrine function during NMP [33]. Building upon this finding, the present study aims to explore whether the endocrine function of human donor kidneys during NMP serves as a pertinent determinant of kidney function post‐transplantation. To address this, we measured levels of (pro)renin, EPO, 1,25(OH)2D in the perfusate, and urodilatin in urine during NMP. Subsequently, we analyzed the correlations between hormone release rates and donor and transplant characteristics, as well as post‐transplant graft function, to evaluate the potential of hormone secretion during NMP as a predictor of post‐transplant kidney function.
2. Materials and Methods
2.1. Kidney Perfusion
Deceased donor kidneys were preserved and transported on pulsatile hypothermic machine perfusion (HMP) after procurement. DCD kidneys of donors aged ≥ 50 years underwent oxygenated HMP, while younger DCD and all DBD kidneys received non‐oxygenated perfusion. Belzer MPS solution (Bridge to Life, USA) was used as a preservation solution for HMP and was not supplemented with additional substances. The materials used for this study came from a randomized controlled trial [34], wherein kidneys were randomized to either the HMP control arm or NMP intervention group. The 28 donor kidneys included in this paper were all assigned to the NMP intervention arm, where they underwent 2 h additional, end‐ischemic NMP with a plasma‐free red blood cell‐based solution prior to implantation at the Erasmus Medical Center. The composition of the NMP perfusate is listed in Table S1 —it did not contain 25(OH)D, the precursor of 1,25(OH)2D [34]. Perfusate samples were collected at time points 0, 1, and 2 h of NMP, after which red blood cells were removed with a 2‐µm filter (Whatman). Urine samples were collected at time points 1 and 2 h during NMP. All perfusate and urine samples were snap‐frozen and stored at –80°C for prorenin/renin, EPO, vitamin D, and urodilatin measurements, respectively. The release of prorenin/renin, EPO, and vitamin D was expressed per hour and corrected for the total perfusate volume.
The study was conducted in compliance with the Declaration of Helsinki. All patients included in this study provided written informed consent, which was approved by the Erasmus Medical Center Medical Ethical Committee (MEC 2020‐0366).
2.2. Renin and Total Renin Measurement
Renin concentration in the NMP perfusate of the initial 15 donor kidneys was assessed using a commercial immunoradiometric assay (IRMA; Renin III, Cisbio, Gif‐sur‐Yvette, France), as described in our previous publication [33].
Renin levels in the NMP perfusate of 13 newly included kidneys were measured using the DSL IRMA Active Renin kit (Beckman Coulter, Inc, Brea, CA). Total renin levels were determined using this kit following the conversion of prorenin to renin. Prorenin activation was achieved by incubating samples for 72h with sepharose‐bound trypsin at 4°C. This enzymatic process cleaves off the pro‐segment, enabling recognition by the active site‐directed antibody. Prorenin levels were calculated by subtracting the measured renin levels from the total renin levels. The detection limit of the IRMA kit was 0.81 pg/mL, with a standard curve ranging from 5 to 500 pg/mL.
To assess the comparability of renin measurements between two different assay kits, a subset of samples from the previous cohort was remeasured using the Beckman Coulter kit. The renin measurements from the Cisbio and Beckman Coulter kits exhibited a strong and significant correlation (R2 = 1.00, p < 0.0001; Figure S1A). To allow comparison between the two cohorts, an adjustment factor of 1.733 was applied to the new measurements obtained with the Beckman Coulter kit.
2.3. EPO Measurement
EPO levels in the NMP perfusate of the first 15 donor kidneys were determined by the Access Erythropoietin chemiluminescence immunoassay (Beckman Coulter, Inc, Brea, CA), following the methodology detailed in our prior publication [33].
The Human EPO ELISA kit (BMS2035‐2, Thermo Fisher Scientific, Waltham, MA, USA) was used to measure EPO levels in the NMP perfusate of 13 recently procured kidneys. The assay was carried out as described by the manufacturer.
To assess the comparability of EPO measurements between two different assay kits, a subset of samples from the previous cohort was remeasured using the Thermo Fisher kit. The analysis demonstrated a strong and significant correlation between the EPO measurements obtained with the Beckman Coulter and Thermo Fisher kits (R2 = 0.88, p < 0.0001; Figure S1B). To facilitate an accurate comparison between the two cohorts, an adjustment factor of 0.488 was applied to the new measurements obtained with the Thermo Fisher kit.
2.4. Active Vitamin D Measurement
The IDS‐iSYS 1,25 VitDXP kit (Immunodiagnostics Systems, IS‐2000) was used to determine the active vitamin D levels in the kidney NMP perfusate on the IDS‐iSYS Multi‐Discipline Automated System. The assay was performed as described by the manufacturer and had a detection range of 18–360 pmol/L.
2.5. Urodilatin Measurement
The urodilatin level in the urine of the first 12 kidneys (no urine was available for measurements in three patients) was measured as described in our previous paper [33]. The additional 13 samples were measured in a similar fashion, albeit with the Wallac Victor2 1420 multilabel counter (PerkinElmer, Waltham, Massachusetts, USA). Since urine produced during NMP was removed every 30 min, we determined the urodilatin release rates for the intervals from 0.5 to 1 h and from 1.5 to 2 h based on the urodilatin levels and urine volume.
2.6. Statistical Analysis
The data analysis was performed using GraphPad Prism version 8 (GraphPad software, La Jolla, CA, USA). The data were summarized as medians with interquartile ranges (IQR) due to their non‐normal distribution. To evaluate statistically significant differences in hormone release rates during the first and second hour of NMP, the Wilcoxon matched‐pairs signed rank test was used. Additionally, the Mann‐Whitney U test was used to compare two unpaired groups.
The nonparametric Spearman correlation test was used to evaluate relationships between two variables. The resulting p values were adjusted for multiple testing using the Benjamini–Hochberg procedure using the “p.adjust” function in R studio (version 2024.04.2.746). The correlation between hormone release rates and early post‐transplant outcome was visualized in R studio using the “lm: linear models” function. For all analyses, a two‐tailed p value of less than 0.05 was considered statistically significant.
3. Results
3.1. Donor and Transplant Characteristics
This study involved 28 deceased donor kidneys. Among these, seven originated from ECD‐DBD donors, while the remainder were procured from DCD donors. Table 1 summarizes the detailed donor and transplant characteristics. The cohort included mostly male donors (20 out of 28), and the donors had a median age of 66 years. Key metrics such as oxygenation during HMP, the initial period of warm ischemia time experienced by DCD kidneys (WIT‐1), cold ischemia time (CIT), and hypertension status are detailed in Table 1.
TABLE 1.
Summary of donor characteristics and transplant characteristics of 28 donor kidneys.
| Donor characteristics | Values (N = 28) |
|---|---|
| Age (years) | 66 (IQR, 58–72) |
| Sex (male), N (%) | 20 (71.4%) |
| Donor type | |
| DBD, N (%) | 7 (25%) |
| DCD, N (%) | 21 (75%) |
| BMI (kg/m2) | 27 (IQR, 25–29) |
| Hypertension, N (%) | 9 (32%) |
| Transplant characteristics | Values (N = 28) |
|---|---|
| Oxygenated HMP | 15 (54%) |
| Non‐oxygenated HMP | 13 (46%) |
| CIT‐1 (hour) | 6.91 (IQR, 5.92–9.99) |
| CIT‐2 (hour) | 1.38 (IQR, 0.9–2.65) |
| Total CIT (hour) | 9.73 (IQR, 7.4–11.8) |
| WIT‐1 (min) | 16 (IQR, 13.5–23.5) |
| Urine production (mL) | |
| T = 60 min | 81 (IQR, 31–123) |
| T = 120 min | 93 (IQR, 53–162) |
Abbreviations: BMI, body mass index; CIT, cold ischemia time; CIT‐1 refers to the CIT that occurs before normothermic machine perfusion (NMP); CIT‐2 refers to the short CIT during static cold storage post‐NMP and before transplantation; DBD, donation after brain death; DCD, donation after circulatory death; HMP, hypothermic machine perfusion; IQR, interquartile range; WIT‐1, the initial period of warm ischemia time experienced by DCD kidneys.
3.2. (Pro)renin, EPO, and Active Vitamin D Release in Kidney Perfusate During NMP
We examined the levels of (pro)renin, EPO, and active vitamin D in perfusate fluid during 2 h of NMP in kidneys from delayed graft function (DGF, characterized by the need for dialysis within the initial week following transplantation), non‐DGF and primary non‐function (PNF, defined as absence of graft function within 90 days after transplantation) groups, as depicted in Figure 1A–D.
FIGURE 1.
Levels of different hormones secreted by the kidneys during 2 h of NMP. (A–D) Prorenin, renin, EPO, and 1,25(OH)2D in the perfusate at timepoints 0, 1, and 2 h of NMP. Black, orange, and red lines respectively represent kidneys that exhibited non‐DGF, DGF, and PNF after transplantation. Open circles and open triangles display DCD and DBD kidneys, respectively. DGF, delayed graft function; PNF, primary non‐function; DCD, donation after circulatory death; DBD, donation after brain death.
The median release rates of prorenin during the first hour (0–1 h) and entire 2 h (0–2 h) of NMP were 105 (IQR, 20–362) and 129 (IQR, 59–384) ng/h, respectively (Figure 1A), while renin release rates within the corresponding time frames were 135 (IQR, 66–336) and 140 (IQR, 59–323) ng/h, respectively (Figure 1B). Notably, when comparing the first hour of NMP with the second hour (1–2 h), there was a significant increase in the release rate of prorenin in the second hour (p = 0.0076), while the release rate of renin did not significantly change. Additionally, a positive correlation was observed between renin and prorenin secretion during 2 h of NMP (R 2 = 0.65, p < 0.0001; Figure S2).
During the initial hour of NMP, the kidneys released EPO at a median rate of 18 (IQR, 5–47) mIU/min. Over the 2 h period of NMP, the kidneys maintained a comparable EPO secretion, with a median rate of 21 (IQR, 6–62) mIU/min (Figure 1C). There was no significant difference in EPO release rates between the initial and subsequent hour of NMP.
The kidneys exhibited a median secretion rate of 1,25(OH)2D of 54 (IQR, 22–85) pmol/h in the initial hour of NMP. Over the 2 h of perfusion, this secretion rate significantly decreased to 33 (IQR, 20–57) pmol/h (p = 0.011; Figure 1D).
3.3. Urodilatin Release in Urine During NMP
Urine production was observed in 25 donor kidneys throughout both the initial and subsequent hour of NMP, with detectable levels of urodilatin (Figure 2). Urodilatin release rates in the first hour were comparable to those in the second hour (first hour: 66 [IQR, 36–119] ng/h; second hour: 54 [IQR, 27–150] ng/h). Additionally, urodilatin release rates were positively correlated with urine output in both the first and second hours of NMP (first hour: Rs = 0.58, p = 0.015; second hour: Rs = 0.67, p = 0.0015; data not shown).
FIGURE 2.
Urodilatin levels in the urine at timepoints 1 and 2 h of NMP. Each line represents a kidney, as shown in the legend. Three samples were missing for urodilatin measurements. Black, orange, and red lines respectively represent kidneys that exhibited non‐DGF, DGF, and PNF after transplantation. Open circles and open triangles display DCD and DBD kidneys, respectively. DGF, delayed graft function; PNF, primary non‐function; DCD, donation after circulatory death; DBD, donation after brain death.
3.4. Impact of Donor Type on Hormone Secretion Capacity
Following the confirmation of hormone production by kidneys during NMP, we proceeded to investigate the potential influence of donor type on the hormone release capacity of the grafts. Prorenin and renin release rates during NMP were similar between kidneys from DBD and DCD donors (Figures 3A,B and S3A,B). DBD kidneys exhibited significantly lower EPO secretion than DCD kidneys during NMP (1 h: p = 0.017; 2 h: p = 0.017; Figures 3C and S3C). Conversely, active vitamin D release rates were significantly higher in DBD kidneys compared to DCD kidneys (1 h: p = 0.0098; 2 h: p = 0.0007; Figures 3D and S3D). No significant differences were observed in urodilatin release rates between DBD and DCD kidneys (Figure 3E,F). Overall, these findings underscore the significant impact of donor type on the hormone release capacity of donor kidneys during NMP.
FIGURE 3.
Release rates of different hormones during 2 h of NMP between DCD and DBD kidneys. (A–D) Prorenin, renin, EPO, and 1,25(OH)2D release rates in the perfusate during 2 h of NMP. (E and F) Urodilatin release rates in the urine during the first hour and the second hour of NMP. Values are shown as median with interquartile range. *p < 0.05; ***p < 0.001; ns, non‐significant p > 0.05; DBD, donation after brain death; DCD, donation after circulatory death.
3.5. Effects of Donor Characteristics on Hormone Release Capacity
We next explored whether other donor characteristics affect the hormone release capacity of kidneys during NMP. First, correlation analyses between hormone release rates and donor age were performed, but no significant correlations were found (Table S2). Similarly, no significant differences in hormone release rates were observed between kidneys from male and female donors (data not shown). However, EPO release rates in kidneys from hypertensive donors were significantly lower than those from normotensive donors during 2 h of NMP (p = 0.047; Figure S4). Besides, kidneys from hypertensive donors exhibited a significantly higher release rate of active vitamin D during 1 h of NMP (p = 0.034; Figure S4). No significant differences were found in the release capacity of prorenin, renin, and urodilatin between hypertensive and normotensive donors (Figure S4).
3.6. Effects of Transplant Characteristics on Hormone Release Capacity
We investigated the association between the hormone secretion capacity of donor kidneys during NMP and various transplant characteristics. Kidneys subjected to oxygenated HMP exhibited similar hormone release rates during NMP compared to those undergoing non‐oxygenated HMP, with the exception of EPO, which was significantly higher in the oxygenated HMP group during the initial hour of NMP (data not shown). CIT‐1 showed no correlation with hormone secretion capacity during NMP (Table S3). Additionally, we assessed whether hormone secretion during NMP was associated with WIT‐1. Our data revealed no significant correlations between hormone release rates and WIT‐1 (Table S3).
3.7. Predictive Value of Hormone Release Capacity on Transplant Outcome
Next, we examined the correlation between hormone release capacity during NMP and early post‐transplant outcome. Half of the 28 included patients experienced immediate graft function, and 12 patients showed DGF (two patients undergoing only one‐time dialysis due to hyperkalemia were excluded), and two patients exhibited PNF (Table 2). To evaluate whether hormones secreted by kidneys during NMP could serve as predictive biomarkers for kidney post‐transplant outcome, we conducted a comparative analysis of hormone release capacity between DGF and non‐DGF kidneys. The two PNF kidneys were included in the DGF group for this analysis. Non‐DGF kidneys exhibited significantly higher urodilatin release rates during both the first and second hour of NMP compared to DGF kidneys (first hour: p = 0.025; second hour: p = 0.0076; Figure 4). These results demonstrate a potential predictive value of the urodilatin release capacity of kidneys during NMP for the occurrence of DGF. Release rates of 1,25(OH)2D during the initial hour of NMP in kidneys without DGF were significantly lower compared to those with DGF (p = 0.046; Figure 4), although this difference disappeared in the second hour of NMP (Figure 4). No significant differences in the release rates of prorenin, renin, and EPO during NMP were observed (Figure 4).
TABLE 2.
Primary outcomes for the 28 included donor kidneys after transplantation.
| Patient No. | DGF | DGF duration (days) | PNF |
|---|---|---|---|
| 1 | Yes | 7 | No |
| 2 | Yes | 6 | No |
| 3 | No | — | — |
| 4 | Yes | 7 | No |
| 5 | — | — | Yes |
| 6 | No | — | — |
| 7 | Yes | 4 | No |
| 8 | Yes | 5 | No |
| 9 | No | — | — |
| 10 | No | — | — |
| 11 | Yes | 4 | No |
| 12 | No | — | — |
| 13 | — | — | Yes |
| 14 | No | — | — |
| 15 | No | — | — |
| 16 | Yes | 7 | No |
| 17 | Yes | 9 | No |
| 18 | No | — | — |
| 19 | No | — | — |
| 20 | No | — | — |
| 21 | Yes | 32 | No |
| 22 | No | — | — |
| 23 | Yes | 73 | No |
| 24 | No | — | — |
| 25 | Yes | 37 | No |
| 26 | No | — | — |
| 27 | Yes | 27 | No |
| 28 | No | — | — |
Abbreviations: DGF, delayed graft function; PNF, primary non‐function.
FIGURE 4.
Comparison of hormone release capacity during NMP between DGF and non‐DGF kidneys. (A–H) Prorenin, renin, EPO, and 1,25(OH)2D release rates in the perfusate during the first hour and the complete 2 h of NMP. (I and J) Urodilatin release rates in the urine during the first and second hour of NMP. The two primary non‐function kidneys were included in the DGF group. Values are shown as median with interquartile range. *p < 0.05; **p < 0.01; ns, non‐significant p > 0.05; DGF, delayed graft function.
We conducted further investigation into the relationship between hormone release rates and the duration of DGF (defined as the period between transplantation and the last dialysis session). Urodilatin release capacity during the second hour of NMP was negatively correlated with the duration of DGF, and it showed a trend toward a negative correlation in the first hour (first hour: Rs = –0.45, p = 0.055; second hour: Rs = –0.61, p = 0.0041; Table 3; Figure 5). These findings suggest that higher urodilatin release from donor kidneys during NMP is associated with a shorter DGF duration following transplantation. There were no significant correlations between the release rates of prorenin, renin, EPO, and active vitamin D during NMP and the duration of DGF (Table 3).
TABLE 3.
The Spearman correlation analyses between hormone release rates during the first hour (0–1 h) and 2 h (0–2 h) of normothermic machine perfusion (NMP) and the duration of delayed graft function (DGF).
| Spearman correlation | DGF duration |
|---|---|
| Prorenin (NMP 1st hour) | Rs = 0.014, p = 0.96 |
| Prorenin (NMP 2 h) | Rs = 0.016, p = 0.94 |
| Renin (NMP 1st hour) | Rs = –0.13, p = 0.82 |
| Renin (NMP 2 h) | Rs = –0.11, p = 0.85 |
| EPO (NMP 1st hour) | Rs = –0.25, p = 0.77 |
| EPO (NMP 2 h) | Rs = –0.25, p = 0.64 |
| 1,25(OH)2D (NMP 1st hour) | Rs = 0.21, p = 0.77 |
| 1,25(OH)2D (NMP 2 h) | Rs = 0.14, p = 0.85 |
| Urodilatin (NMP 1st hour) | Rs = –0.45, p = 0.055 |
| Urodilatin (NMP 2nd hour) | Rs = –0.61, p = 0.0041 * |
Abbreviation: Rs, Spearman's rank correlation coefficient.
p < 0.01.
FIGURE 5.
Negative correlation between the duration of DGF and the urodilatin release rate during the second hour of NMP. The release rate and DGF duration were log‐transformed before plotting. A DGF duration of 1 day was set for non‐DGF kidneys to include those in the figure. Each dot represents a machine‐perfused kidney. DGF, delayed graft function.
3.8. Correlation of Hormone Release Capacity of Kidneys With Early Post‐Transplant Kidney Function
To further assess the potential of hormone release as a biomarker for post‐transplant outcome, we conducted correlation analyses between hormone release rates and serum creatinine levels at 1‐week post‐transplant and estimated glomerular filtration rate (eGFR) at 1 month post‐transplant. While the non‐DGF kidneys showed a significant decline in serum creatinine levels in the first week (p = 0.015), this was not observed for the DGF and PNF kidneys (Figure S5). Interestingly, EPO release capacity during 2 h of NMP was positively correlated with recipient serum creatinine levels at 1 week after transplantation in patients with non‐DGF kidneys and patients with a DGF duration of less than 7 days (Rs = 0.83, p = 0.022; Figure 6; Table 4). No significant correlations were found between the release rates of hormones during NMP and 1‐month eGFR (Table 4).
FIGURE 6.
Positive correlation between the serum creatinine levels at 1‐week post‐transplant and EPO release rate during 2 h of NMP. Each dot represents a machine‐perfused kidney. EPO, erythropoietin.
TABLE 4.
The Spearman correlation analyses between hormone release rates during the first hour (0–1 h) and 2 h (0–2 h) of normothermic machine perfusion (NMP) and serum creatinine levels at 1‐week post‐transplant (non‐DGF recipients and recipients with a DGF duration of less than 7 days) and estimated glomerular filtration rate (eGFR) at 1‐month post‐transplant.
| Spearman correlation | Serum creatinine levels at 1‐week post‐transplant | eGFR at 1‐month post‐transplant |
|---|---|---|
| Prorenin (NMP 1st hour) | Rs = 0.55, p = 0.20 | Rs = 0.066, p = 0.94 |
| Prorenin (NMP 2 h) | Rs = 0.43, p = 0.29 | Rs = 0.17, p = 0.85 |
| Renin (NMP 1st hour) | Rs = 0.49, p = 0.21 | Rs = 0.36, p = 0.42 |
| Renin (NMP 2 h) | Rs = 0.16, p = 0.66 | Rs = 0.28, p = 0.64 |
| EPO (NMP 1st hour) | Rs = 0.73, p = 0.067 | Rs = 0.17, p = 0.82 |
| EPO (NMP 2 h) | Rs = 0.83, p = 0.022 * | Rs = 0.17, p = 0.85 |
| 1,25(OH)2D (NMP 1st hour) | Rs = –0.43, p = 0.22 | Rs = –0.37, p = 0.42 |
| 1,25(OH)2D (NMP 2 h) | Rs = –0.50, p = 0.29 | Rs = –0.30, p = 0.64 |
| Urodilatin (NMP 1st hour) | Rs = 0.030, p = 0.95 | Rs = 0.31, p = 0.20 |
| Urodilatin (NMP 2nd hour) | Rs = –0.2, p = 0.58 | Rs = 0.42, p = 0.074 |
Abbreviation: Rs, Spearman's rank correlation coefficient.
p < 0.05.
4. Discussion
In our prior research, we showcased the endocrine activity of human donor kidneys during NMP [33]. Expanding on this discovery, the current investigation endeavors to assess whether hormone secretion could function as potential biomarkers for predicting early post‐transplant outcomes.
Our analysis uncovered significant differences in the secretion capacity of EPO and active vitamin D between DBD and DCD kidneys, which aligns with our previous findings on donor type influence [33]. Specifically, we observed significantly higher rates of EPO release in DCD kidneys during NMP compared to DBD kidneys. We hypothesized that this difference may stem from the longer warm ischemic period experienced by DCD donors, during which blood flow and oxygen supply to the kidneys are halted, prompting a hypoxic response that stimulates EPO production [35, 36]. However, our analysis did not reveal a direct correlation between EPO release rates and WIT‐1, suggesting that other donor factors impact the capacity of EPO secretion. Notably, oxygenated HMP kidneys exhibited elevated EPO levels during the first hour of NMP despite remaining unchanged over the entire 2‐h perfusion period. This finding appears contradictory, as EPO production is typically stimulated by hypoxia. A possible explanation is that low temperature may suppress hypoxia‐induced EPO synthesis, which is further supported by our observation that oxygenated and non‐oxygenated HMP‐preserved DCD kidneys exhibit similar EPO release rates. Interestingly, we observed higher active vitamin D release rates from DBD kidneys compared to DCD kidneys. We investigated whether this difference was correlated with other characteristics, such as CIT‐1 and WIT‐1. While CIT‐1 did not differ between DBD and DCD kidneys, WIT‐1 was markedly different—approximately 5 min for DBD kidneys versus an average of 18 min for DCD kidneys. The detected active vitamin D most likely represents leakage from the kidney, as the vitamin D precursor required for conversion to active vitamin D is absent in the perfusate. Given the prolonged WIT‐1 experienced by DCD kidneys, we hypothesize that vitamin D is already released during this period, resulting in reduced subsequent release during perfusion compared to DBD kidneys.
The primary aim of our study was to evaluate whether the endocrine activity of donor kidneys during NMP could predict post‐transplant outcomes. While our data indicated that kidneys with DGF released significantly more active vitamin D during the first hour of NMP compared to non‐DGF kidneys, there was no correlation between the release rate of active vitamin D during NMP and the duration of DGF, serum creatinine levels at 1‐week post‐transplant, or eGFR at 1 month post‐transplant. Interestingly, the EPO release rate was positively correlated to the serum creatinine levels at 1‐week post‐transplant. Such correlation with eGFR was not present at 1 month after transplantation. This indicates that EPO release during perfusion can predict post‐transplant kidney function shortly after transplantation. Longer after transplantation, confounding factors such as donor‐specific antibody formation, calcineurin inhibitor nephrotoxicity, hypertension, and progression of donor‐derived lesions are more likely to interfere with kidney function [37, 38, 39]. Overall, our results suggest an association between the release rates of the hormones detected in the perfusate and early post‐transplant kidney function.
Several studies underscore the significance of urine output as a key quality indicator for donor kidneys [11, 12, 13]. Urodilatin, a natriuretic peptide produced by the kidneys, promotes sodium excretion and maintains fluid balance by increasing urine production. Our data demonstrate a significant positive correlation between urodilatin release rates and urine output during both the first and second hours of NMP. Notably, urodilatin release rates were significantly higher in non‐DGF kidneys and negatively correlated with the duration of DGF. Additionally, urodilatin release rates during the second hour of NMP tended to correlate positively with eGFR at 1‐month post‐transplant. Collectively, these findings suggest that higher urodilatin release during NMP is predictive of improved post‐transplant outcomes. Real‐time monitoring of urodilatin levels offers valuable insights into organ viability prediction of graft function and may be a tool for donor kidney selection, ultimately improving post‐transplant outcome.
This study has several limitations. The use of different assay kits for renin and EPO measurements may introduce variability. However, to ensure comparability, we performed calibration by remeasuring a subset of samples from the initial cohort using the new assay kits, confirming a strong correlation between kits, and applying an adjustment factor. The small sample size of kidneys reduces both the statistical power and the generalizability of the results. Additionally, the lack of long‐term post‐transplant outcome data limits the clinical relevance of the findings. Substantial variation in donor and transplant characteristics also complicates the ability to draw clear conclusions across different donor types. Moreover, the study did not include kidneys at the extremes of quality, such as those rejected for transplant, nor did it include kidneys from non‐ECD‐DBD donors or younger, healthier donors. This lack of diverse kidney inclusion limits the ability to fully assess hormone release profiles across a broader spectrum of kidney quality during NMP.
Taken together, we identified a correlation between urodilatin release capacity and post‐transplant kidney function, suggesting that urodilatin secretion levels could serve as a functional marker for predicting post‐transplant outcome. Although active vitamin D release differentiated DGF from non‐DGF kidneys during the first hour of NMP, its predictive value for early post‐transplant outcome remains uncertain due to the lack of significant differences in active vitamin D release rates during 2 h of NMP between DGF and non‐DGF kidneys, as well as the absence of correlations between active vitamin D release rates and DGF duration or eGFR at 1 month post‐transplant. Assessing the urodilatin release capacity of donor kidneys during NMP could provide a valuable criterion for determining the suitability of suboptimal donor kidneys for transplantation. This approach may help avoid the transplantation of high‐risk kidneys, thereby reducing the likelihood of poor outcomes. Furthermore, incorporating this assessment could expand the donor pool by preventing the unnecessary discard of suboptimal donor kidneys. Ultimately, this strategy may lead to the development of a more comprehensive quality assessment procedure for donor kidneys, with hormone release capacity included as a key parameter.
In conclusion, our study indicates that the release capacity of urodilatin, EPO, and, to a lesser extent, active vitamin D from kidneys during NMP may serve as potential biomarkers for predicting early post‐transplant outcomes.
Author Contributions
Hui Lin and Karim Bousnina designed and performed experiments, analyzed results, and wrote the article. Julia S. Slagter, Yitian Fang, Iacopo Cristoferi, and Ingrid M. Garrelds performed experiments and reviewed the article. A. H. Jan Danser provided expertise and reviewed the article. Marlies E. J. Reinders reviewed the article. Robert C. Minnee and Martin J. Hoogduijn designed the experiments, provided expertise and feedback, and reviewed the article.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting Information
Acknowledgments
The authors would like to thank all organ donors and their families.
Funding: The authors acknowledge funding support (Organ Transplantation: making unsuitable organs suitable; FS.OT2023) from Convergence Health & Technology, an alliance between Erasmus Medical Centre Rotterdam, Erasmus University Rotterdam, and Delft University of Technology. The authors received grants from the Dutch Kidney Foundation (grant no. 19OI09 “Laser speckle imaging: a real‐time tool for viability assessment if extended criteria donor kidneys during normothermic machine perfusion”) and Astellas Pharma (grant no. AG2021/1130).
Hui Lin and Karim Bousnina contributed equally to this study.
Robert C. Minnee and Martin J. Hoogduijn are joint senior authors.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Hosgood S. A., van Heurn E., and Nicholson M. L., “Normothermic Machine Perfusion of the Kidney: Better Conditioning and Repair?,” Transplant International 28, no. 6 (2015): 657–664. [DOI] [PubMed] [Google Scholar]
- 2. Jing L., Yao L., Zhao M., et al., “Organ Preservation: From the Past to the Future,” Acta Pharmacologica Sinica 39, no. 5 (2018): 845–857. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Hosgood S. A., Callaghan C. J., Wilson C. H., et al., “Normothermic Machine Perfusion versus Static Cold Storage in Donation After Circulatory Death Kidney Transplantation: A Randomized Controlled Trial,” Nature Medicine 29, no. 6 (2023): 1511–1519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ravikumar R., Leuvenink H., and Friend P. J., “Normothermic Liver Preservation: A New Paradigm?,” Transplant International 28, no. 6 (2015): 690–699. [DOI] [PubMed] [Google Scholar]
- 5. De Beule J. and Jochmans I., “Kidney Perfusion as an Organ Quality Assessment Tool—Are We Counting Our Chickens Before They Have Hatched?,” Journal of Clinical Medicine 9, no. 3 (2020): 879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. van Leeuwen L. L., Leuvenink H. G. D., Olinga P., et al., “Shifting Paradigms for Suppressing Fibrosis in Kidney Transplants: Supplementing Perfusion Solutions With Anti‐Fibrotic Drugs,” Frontiers in Medicine (Lausanne) 8 (2021): 806774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Summers D. M., Watson C. J., Pettigrew G. J., et al., “Kidney Donation After Circulatory Death (DCD): State of the Art,” Kidney International 88, no. 2 (2015): 241–249. [DOI] [PubMed] [Google Scholar]
- 8. Schaapherder A., Wijermars L. G. M., de Vries D. K., et al., “Equivalent Long‐Term Transplantation Outcomes for Kidneys Donated After Brain Death and Cardiac Death: Conclusions From a Nationwide Evaluation,” EClinicalMedicine 4‐5 (2018): 25–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hamelink T. L., Ogurlu B., De Beule J., et al., “Renal Normothermic Machine Perfusion: The Road Toward Clinical Implementation of a Promising Pretransplant Organ Assessment Tool,” Transplantation 106, no. 2 (2022): 268–279. [DOI] [PubMed] [Google Scholar]
- 10. Dumbill R., Knight S., Hunter J., et al., “Prolonged Normothermic Perfusion of the Kidney—A Historically Controlled, Phase 1 Cohort Study,” Research Square (2024). [Google Scholar]
- 11. Markgraf W., Muhle R., Lilienthal J., et al., “Inulin Clearance During Ex Vivo Normothermic Machine Perfusion as a Marker of Renal Function,” ASAIO Journal 68, no. 9 (2022): 1211–1218. [DOI] [PubMed] [Google Scholar]
- 12. Hosgood S. A., Barlow A. D., Hunter J. P., et al., “Ex Vivo Normothermic Perfusion for Quality Assessment of Marginal Donor Kidney Transplants,” British Journal of Surgery 102, no. 11 (2015): 1433–1440. [DOI] [PubMed] [Google Scholar]
- 13. Hosgood S. A., Thompson E., Moore T., et al., “Normothermic Machine Perfusion for the Assessment and Transplantation of Declined Human Kidneys From Donation After Circulatory Death Donors,” British Journal of Surgery 105, no. 4 (2018): 388–394. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Acharya V. and Olivero J., “The Kidney as an Endocrine Organ,” Methodist DeBakey Cardiovascular Journal 14, no. 4 (2018): 305–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Lin H., Geurts F., Hassler L., et al., “Kidney Angiotensin in Cardiovascular Disease: Formation and Drug Targeting,” Pharmacological Reviews 74, no. 3 (2022): 462–505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Reid I. A., “The Renin‐Angiotensin System and Body Function,” Archives of Internal Medicine 145, no. 8 (1985): 1475–1479. [PubMed] [Google Scholar]
- 17. Lai K. N., Leung J. C., Lai K. B., et al., “Gene Expression of the Renin‐Angiotensin System in Human Kidney,” Journal of Hypertension 16, no. 1 (1998): 91–102. [DOI] [PubMed] [Google Scholar]
- 18. Sim J. J., Shi J., Calara F., et al., “Association of Plasma Renin Activity and Aldosterone‐Renin Ratio With Prevalence of Chronic Kidney Disease: The Kaiser Permanente Southern California Cohort,” Journal of Hypertension 29, no. 11 (2011): 2226–2235. [DOI] [PubMed] [Google Scholar]
- 19. Flannery A. H., Ortiz‐Soriano V., Li X., et al., “Serum Renin and Major Adverse Kidney Events in Critically Ill Patients: A Multicenter Prospective Study,” Critical Care (London, England) 25, no. 1 (2021): 294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Kullmar M., Saadat‐Gilani K., Weiss R., et al., “Kinetic Changes of Plasma Renin Concentrations Predict Acute Kidney Injury in Cardiac Surgery Patients,” American Journal of Respiratory and Critical Care Medicine 203, no. 9 (2021): 1119–1126. [DOI] [PubMed] [Google Scholar]
- 21. Souma T., Suzuki N., and Yamamoto M., “Renal Erythropoietin‐Producing Cells in Health and Disease,” Frontiers in Physiology 6 (2015): 167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Panjeta M., Tahirovic I., Sofic E., et al., “Interpretation of Erythropoietin and Haemoglobin Levels in Patients With Various Stages of Chronic Kidney Disease,” Journal of Medical Biochemistry 36 (2017): 145–152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Fishbane S. and Spinowitz B., “Update on Anemia in ESRD and Earlier Stages of CKD: Core Curriculum 2018,” American Journal of Kidney Diseases 71, no. 3 (2018): 423–435. [DOI] [PubMed] [Google Scholar]
- 24. Souma T., Yamazaki S., Moriguchi T., et al., “Plasticity of Renal Erythropoietin‐Producing Cells Governs Fibrosis,” Journal of the American Society of Nephrology 24, no. 10 (2013): 1599–1616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Bikle D. D., “Vitamin D Metabolism, Mechanism of Action, and Clinical Applications,” Chemistry & Biology 21, no. 3 (2014): 319–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Ishimura E., Nishizawa Y., Inaba M., et al., “Serum Levels of 1,25‐dihydroxyvitamin D, 24,25‐dihydroxyvitamin D, and 25‐hydroxyvitamin D in Nondialyzed Patients With Chronic Renal Failure,” Kidney International 55, no. 3 (1999): 1019–1027. [DOI] [PubMed] [Google Scholar]
- 27. Doorenbos C. R., van den Born J., Navis G., et al., “Possible Renoprotection by Vitamin D in Chronic Renal Disease: Beyond Mineral Metabolism,” Nature Reviews Nephrology 5, no. 12 (2009): 691–700. [DOI] [PubMed] [Google Scholar]
- 28. Bienaime F., Girard D., Anglicheau D., et al., “Vitamin D Status and Outcomes After Renal Transplantation,” Journal of the American Society of Nephrology 24, no. 5 (2013): 831–841. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Kalantzi M., Kalliakmani P., Papachristou E., et al., “Parameters Influencing Blood Erythropoietin Levels of Renal Transplant Recipients During the Early Post‐Transplantation Period,” Transplantation Proceedings 46, no. 9 (2014): 3179–3182. [DOI] [PubMed] [Google Scholar]
- 30. Forssmann W., Meyer M., and Forssmann K., “The Renal Urodilatin System: Clinical Implications,” Cardiovascular Research 51, no. 3 (2001): 450–462. [DOI] [PubMed] [Google Scholar]
- 31. Schramm L., Heidbreder E., Schaar J., et al., “Toxic Acute Renal Failure in the Rat: Effects of Diltiazem and Urodilatin on Renal Function,” Nephron 68, no. 4 (1994): 454–461. [DOI] [PubMed] [Google Scholar]
- 32. Horl W. H., “Natriuretic Peptides in Acute and Chronic Kidney Disease and During Renal Replacement Therapy,” Journal of Investigative Medicine 53, no. 7 (2005): 366–370. [DOI] [PubMed] [Google Scholar]
- 33. Lin H., Du Z., Bouari S., et al., “Human Transplant Kidneys on Normothermic Machine Perfusion Display Endocrine Activity,” Transplant Direct 9, no. 7 (2023): e1503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Rijkse E., Bouari S., Kimenai H., et al., “Additional Normothermic Machine Perfusion versus Hypothermic Machine Perfusion in Suboptimal Donor Kidney Transplantation: Protocol of a Randomized, Controlled, Open‐Label Trial,” International Journal of Surgery Protocols 25 (2021): 227–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Perera M. T., “The Super‐Rapid Technique in Maastricht Category III Donors: Has It Developed Enough for Marginal Liver Grafts From Donors After Cardiac Death?,” Current Opinion in Organ Transplantation 17, no. 2 (2012): 131–136. [DOI] [PubMed] [Google Scholar]
- 36. Haase V. H., “Regulation of Erythropoiesis by Hypoxia‐Inducible Factors,” Blood Reviews 27, no. 1 (2013): 41–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Kasiske B. L., Anjum S., Shah R., et al., “Hypertension After Kidney Transplantation,” American Journal of Kidney Diseases 43, no. 6 (2004): 1071–1081. [DOI] [PubMed] [Google Scholar]
- 38. Naesens M., Kuypers D. R., and Sarwal M., “Calcineurin Inhibitor Nephrotoxicity,” Clinical Journal of the American Society of Nephrology 4, no. 2 (2009): 481–508. [DOI] [PubMed] [Google Scholar]
- 39. Wiebe C., Gibson I. W., Blydt‐Hansen T. D., et al., “Rates and Determinants of Progression to Graft Failure in Kidney Allograft Recipients With De Novo Donor‐Specific Antibody,” American Journal of Transplantation 15, no. 11 (2015): 2921–2930. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.