ABSTRACT
Background
There is a known recipient sex–dependent association between donor sex and kidney transplant survival. We hypothesized that donor age also modifies the association between donor sex and graft survival.
Methods
First, deceased donor kidney transplant recipients (1988–2019, n = 461 364) recorded in the Scientific Registry of Transplant Recipients, the Australia and New Zealand Dialysis and Transplant Registry and the Collaborative Transplant Study were analyzed. We used multivariable Cox regression models to estimate the association between donor sex and death censored graft loss, accounting for the modifying effects of recipient sex and donor age; donor age was categorized as 5–19, 20–34, 35–49, 50–59 and ≥60 years. Results from cohort-specific Cox models were combined using individual patient data meta-analysis.
Results
Among female recipients of donors aged <60 years, graft loss hazards did not differ by donor sex; recipients of female donors ≥60 years showed significantly lower graft loss hazards than recipients of male donors of the same age [combined adjusted hazard ratio (aHR) 0.90, 95% CI 0.86–0.94]. Among male recipients, female donors aged <50 years were associated with significantly higher graft loss hazards than same-aged male donors (5–19 years: aHR 1.11, 95% CI 1.02–1.21; 20–34 years: aHR 1.08, 95% CI 1.02–1.15; 35–49 years: aHR 1.07, 95% CI 1.04–1.10). There were no significant differences in graft loss by donor sex among male recipients of donors aged ≥50 years.
Conclusion
Donor age modifies the association between donor sex and graft survival. Older female donors were associated with similar or lower hazards of graft failure than older male donors in both male and female recipients, suggesting a better functional reserve of older female donor kidneys.
Keywords: donor age, individual patient data meta-analysis, kidney graft survival, sex differences
KEY LEARNING POINTS.
What was known:
The association between donor sex and kidney graft survival is modified by recipient sex.
Older donor age is associated with poorer graft outcomes.
This study adds:
The association between donor sex and graft survival is modified by donor age.
Female recipients: no difference in graft loss by donor sex with younger donors, but significantly lower graft loss with older female than male donors.
Male recipients: higher graft loss with younger female than male donors, but no difference by donor sex with older donors.
Potential impact:
Graft outcomes with female donors ≥50 years old are similar or better compared with male donors the same age.
Greater confidence in accepting older female donor organs.
INTRODUCTION
Kidneys from male donors are believed to offer an advantage compared with kidneys from female donors due to a greater number and size of nephrons [1–4]. However, female recipients do not appear to derive the same benefit as males from the larger nephron “dose” offered by a male donor—likely due to immune reaction against the HY-antigen found on all male tissues [5–9]. At the same time, kidneys are less susceptible to acute kidney injury [10] and ischemia reperfusion injury in a female environment, at least partly due to a protective effect of estrogen [11]. Both kidney graft survival and excess mortality are also independently associated with recipient sex [12, 13].
Donor age is among the most powerful predictors of graft survival [14]. Older donor age is associated with poorer regeneration after acute rejection, higher rates of delayed graft function and higher incidence of functional loss due to chronic changes [15, 16]. A higher senescent cell burden in older donors is believed to result in poorer capacity for repair and a profibrotic and proinflammatory milieu [16–18]; this in turn may provoke an immune response [19, 20].
While both donor sex and donor age are known to be associated with graft survival, it is not known whether the association between donor sex and graft survival is modified by donor age. We aimed to determine the independent association between donor sex and death-censored graft loss, accounting for the modifying effects of both donor age and recipient sex, in three of the largest transplant databases worldwide: the American Scientific Registry of Transplant Recipients (SRTR), the Australia and New Zealand Dialysis and Transplant Registry (ANZDATA) and the international Collaborative Transplant Study (CTS) database.
MATERIALS AND METHODS
Data sources and population
This was a retrospective cohort study of recipients of all ages of a first kidney–only deceased donor transplant recorded in the SRTR (transplants 1 January 1988 to 31 December 2018, followed until 1 June 2019), ANZDATA (transplants 1 January 1988 to 31 December 2019, followed until 31 December 2019) or CTS (transplants 1 January 1988 to 31 December 2018, followed until 28 October 2020). Recipients with a donor <5 years old were excluded due to the known poorer outcomes with very young donors [21–23].
The SRTR data system includes data on all donors, waitlisted candidates and transplant recipients in the USA, submitted by the members of the Organ Procurement and Transplantation Network (OPTN). The Health Resources and Services Administration, US Department of Health and Human Services, provides oversight to the activities of the OPTN and SRTR contractors. The ANZDATA aims to include data for all transplant recipients in Australia and New Zealand with an opt-out consent process. The ANZDATA is linked to the Australia and New Zealand Organ Donor Registry to optimize capture of transplant activity [24]. The CTS is among the largest voluntary international registries [25]. The CTS includes data from more than 400 centers. Countries with <1000 transplants, >10% loss to follow-up or <80% 1-year graft survival were excluded (9.6% of transplants) because recipients from these countries were highly likely to represent a biased sample [12, 13]. To avoid duplicates, recipients from USA, Australia and New Zealand were excluded from the CTS cohort. Countries represented in the CTS cohort are listed in the Supplementary data.
Exposure and outcome definitions
The primary exposure was donor sex. The relation between donor sex and graft failure risk differs by recipient sex [26], due at least partly to immune reaction by female recipients against the HY-antigen on male donor tissues [8]. Therefore, we considered donor–recipient sex combinations to allow us to investigate the association between donor sex and graft survival from the perspective of a female recipient separately from the perspective of a male recipient. We considered the possibility that magnitude or direction of the association between donor sex and graft survival may also differ by donor age by including a donor–recipient sex combination by donor age interaction term in all models. Donor age was categorized as 5–19 (child/adolescent), 20–34 (young adult), 35–49 (early middle-age), 50–59 (late middle-age) and ≥60 (older) years.
The primary outcome was the death-censored graft loss, defined as return to dialysis or re-transplantation, with observation censored at death.
Data harmonization
The distributions of all variables considered for inclusion in the models were examined within each cohort. Where differences between cohorts existed, new variables were generated, e.g. harmonized primary disease categories were created to allow uniform representation across cohorts. Due to significant differences in recipient race distribution between cohorts, race was dichotomized as white vs other. Panel reactive antibody (PRA) had to be classified as 0% vs >0% due to differences in the way PRA was captured between cohorts and over time.
Statistical analysis
Analyses were first performed for each cohort separately, then combined using two-stage individual patient data meta-analysis [12, 27, 28]. Descriptive statistics were combined across the three cohorts weighted for observation time.
Association between donor sex and graft survival
Cox proportional hazards models were used to estimate the association between donor sex and graft survival for female and male recipients with donors of different ages. Unadjusted analyses were followed by analyses adjusted for the all available variables that may confound the relation between donor sex and graft loss [transplant year (continuous), recipient age at transplant (piece-wise linear spline), recipient race (white/other), primary diagnosis (diabetes, hypertension, glomerulonephritis, congenital anomalies of the kidneys or urinary tract, cystic kidney disease, other), pre-transplant dialysis (yes/no), PRA (0% vs >0%), HLA mismatch (continuous), cold ischemia time (≤18, >18–24, >24–36, >36 h)]. We also adjusted for available variables that may mediate this relationship [donor–recipient weight ratio (<0.7, 0.7–<0.9, ≥0.9; not available for CTS), donor cytomegalovirus (CMV; not included for CTS or ANZDATA as it was missing in >20%), donor cerebrovascular accident (CVA), donor hypertension] because we were interested in the independent, direct association between donor sex and graft loss. Missing data were imputed using multiple imputation methods (15 imputations) based on the fully conditional specification method [29]; parameter estimates were pooled across the imputed datasets as per Rubin's rule [29]. This imputation method uses the joint distributions of all variables included in the model to impute missing values and incorporates the uncertainty introduced by missing data into the calculation of confidence intervals (CI) [29].
Adjusted hazard ratio (aHR) represent the hazards of graft loss in recipients of female donors compared with those in recipients of male donors. Proportionality was assessed by refitting the models, censoring all observation at 5 and 10 years. HR were unchanged, indicating that hazards were proportional.
In the meta-analysis, we used a random-effects model to estimate pooled HR across all three cohorts. A random-effects model assumes the observed HR estimates may vary across cohorts due to real differences in the association between donor sex and graft loss across cohorts or due to sampling variability. The random effects model synthesizes heterogeneous data as a weighted average of the effect sizes of the cohorts. Heterogeneity in outcome between cohorts was evaluated by Higgins I2 and the chi-square test of heterogeneity.
Fitted graft failure rates by donor sex–donor age–recipient sex combination
Crude failure rates may be misleading because they do not account for either the changing graft loss risk over time since transplant or for the impact of confounders. To avoid this problem, we calculated fitted cumulative failure rates for each donor sex–recipient sex–donor age combination in each cohort separately based on the adjusted Cox survival model with multiple imputations described above, at a fixed time point (10 years post-transplant) and the failure rate in recipients with a fixed profile of characteristics. The fitted rates were pooled across cohorts in a meta-analysis using a random effects model.
Sensitivity analyses
Donor:recipient weight ratio was not available in the CTS. Therefore, models for SRTR and ANZDATA were re-fitted excluding donor:recipient weight ratio. Data on donor CMV were missing in 41% of the CTS and 21.5% of the ANZDATA population, so we also refitted the model for SRTR excluding donor CMV and excluding both donor:recipient weight ratio and donor CMV. Compared with SRTR and CTS cohorts, ANZDATA is a much smaller cohort possibly contributing to greater heterogeneity; therefore, we also refitted the model excluding the ANZDATA.
Data analyses were performed using Statistical Analysis System 9.4 (SAS Institute, Cary, NC, USA) and R (version 4.2.0). The study was approved by the McGill University Health Center Research Ethics Board and the Westmead Research Ethics Board.
RESULTS
We identified 242 030 recipients recorded in the SRTR (69.3% male), 205 399 in the CTS (62.8% male) and 13 935 in ANZDATA (62.4% male) who met inclusion criteria. Overall, there were 62 647 graft losses (25.9%) over a follow-up of 5.4 ± 4.8 years in the SRTR, 50 613 graft losses (24.6%) over a follow-up of 7.1 ± 6.1 years in the CTS and 3379 graft losses (24.3%) over a follow-up of 8.5 ± 7.0 years in ANZDATA.
Patient characteristics
An overview of the characteristics of each cohort are presented in Table 1. The distribution of primary disease varied by cohort, with greater proportions in SRTR having diabetes or hypertension as the primary disease than in CTS or ANZDATA. Donors in the SRTR were younger than in ANZDATA or CTS. A larger proportion of SRTR patients underwent pre-emptive transplantation compared with the other two cohorts. The proportion of patients with a PRA >0% was lowest in CTS (19.0%), versus 30.0% in the SRTR and 49.6% in ANZDATA. Cold ischemia times were shortest in the ANZDATA.
Table 1:
Overview of patient characteristics (patient n, %).
| SRTR | CTS | ANZDATA | |
|---|---|---|---|
| Patients | 242 030 | 205 399 | 13 935 |
| Censoring | |||
| Graft loss | 62 647 (25.9) | 50 613 (24.6) | 3379 (24.3) |
| Death | 60 163 (24.9) | 40 152 (19.5) | 2921 (21.0) |
| Follow-up (years) | |||
| Mean ± SD | 6.1 ± 5.4 | 7.1 ± 6.1 | 8.5 ± 7.0 |
| Median (IQR) | 4.8 (2.0–8.9) | 6.0 (2.0–10.6) | 6.8 (2.7–12.6) |
| Donor–recipient sex combination (%) | |||
| Female–female | 38 549 (15.9) | 33 035 (16.1) | 2286 (16.4) |
| Male–female | 56 771 (23.5) | 43 595 (21.2) | 2961 (21.2) |
| Female–male | 57 093 (23.6) | 53 533 (26.1) | 3594 (25.8) |
| Male–male | 89 617 (37.0) | 75 236 (36.6) | 5094 (36.6) |
| Recipient age at transplant (years) [median (IQR)] |
51 (40–61) |
50 (38–60) |
51 (40–60) |
| Recipient race | |||
| White | 204 210 (84.4) | 131 264 (91.2) | 10 043 (72.1) |
| Other | 37 820 (15.6) | 12 615 (8.8) | 3351 (27.9) |
| Missing | 0 (0) | 61 520 (30.0) | 541 (3.9) |
| Primary disease | |||
| Diabetes | 63 674 (26.3) | 16 429 (8.4) | 1804 (13.0) |
| Hypertension | 60 647 (25.1) | 18 545 (9.5) | 823 (5.9) |
| Glomerulonephritis | 58 267 (24.1) | 66 898 (34.3) | 5889 (42.4) |
| CAKUT | 7835 (3.2) | 22 171 (11.4) | 1272 (9.2) |
| Cystic | 22 384 (9.3) | 28 081 (14.4) | 2022 (14.6) |
| Other | 28 987 (12.0) | 42 837 (22.0) | 2086 (15.0) |
| Missing | 236 (0.1) | 10 438 (5.1) | 39 (0.3) |
| Donor age (years) [median (IQR)] |
38 (23–51) |
47 (32–58) |
45 (29–57) |
| Donor age (years) | |||
| 5–19 | 40 495 (16.7) | 21 549 (10.5) | 1751 (12.6) |
| 20–34 | 66 250 (27.4) | 36 335 (17.7) | 2760 (19.8) |
| 35–49 | 69 310 (28.6) | 54 523 (26.5) | 3778 (27.1) |
| 50–59 | 45 323 (18.7) | 45 928 (22.4) | 2948 (21.2) |
| ≥60 | 20 652 (8.5) | 47 064 (22.9) | 2698 (19.4) |
| Transplant year | |||
| 1988–1994 | 41 822 (17.3) | 44 298 (21.6) | 2513 (18.0) |
| 1995–1999 | 32 306 (13.6) | 33 747 (16.4) | 1650 (11.8) |
| 2000–2004 | 34 778 (14.4) | 34 183 (16.6) | 1691 (12.1) |
| 2005–2009 | 41 843 (17.3) | 34 066 (16.6) | 1718 (12.3) |
| 2010–2013 | 43 777 (18.1) | 35 018 (17.0) | 2672 (19.2) |
| 2014–2019 | 47 504 (19.6) | 24 087 (11.7) | 3691 (26.5) |
| Cold ischemia time (h) | |||
| 0–18 | 117 398 (52.0) | 106 712 (57.9) | 10 885 (80.9) |
| >18–24 | 50 742 (22.5) | 47 614 (25.8) | 2131 (15.8) |
| >24–36 | 46 640 (20.6) | 25 920 (14.1) | 431 (3.2) |
| >36 | 11 177 (5.0) | 4010 (2.2) | 12 (0.1) |
| Missing | 16 073 (6.6) | 21 143 (10.3) | 476 (3.4) |
| Transplantation type | |||
| Pre-emptive | 18 738 (8.1) | 4 054 (2.2) | 145 (1.0) |
| With prior dialysis | 214 003 (92.0) | 177 672 (97.8) | 13 790 (99.0) |
| Missing | 9289 (3.8) | 23 673 (11.5) | 0 |
| Donor:recipient weight ratio (%) | |||
| <0.7 | 33 593 (16.1) | N/A | 1536 (11.7) |
| 0.7-<0.9 | 48 697 (23.4) | 2925 (22.4) | |
| ≥0.9 | 125 877 (60.5) | 8627 (65.9) | |
| Missing | 33 863 (14.0) | 847 (6.1) | |
| HLA mismatches | |||
| Mean ± SD | 3.9 ± 1.6 | 2.99 ± 1.42 | 3.6 ± 1.6 |
| Missing | 839 (0.4) | 21 360 (10.4) | 9 (0.1) |
| PRA (%) | |||
| 0% | 161 743 (69.6) | 118 854 (81.0) | 7016 (50.4) |
| >0% | 70 696 (30.4) | 27 844 (19.0) | 6902 (49.6) |
| Missing | 9591 (4.0) | 58 701 (28.6) | 17 (0.1) |
| Donor CMV | |||
| Positive | 147 801 (61.2) | 97 031 (60.2) | 6929 (84.3) |
| Negative | 93 642 (38.8) | 64 245 (39.8) | 1289 (15.7) |
| Missing | 587 (0.2) | 44 123 (21.5) | 5717 (41.0) |
| Donor cause of death | |||
| CVA | 83 960 (34.7) | 102 690 (56.6) | 6463 (47.8) |
| Other | 157 922 (65.3) | 78 656 (43.4) | 7048 (52.2) |
| Missing | 148 (0.1) | 24 053 (11.7) | 424 (3.0) |
| Donor hypertension | |||
| Yes | 52 378 (25.8) | 21 161 (11.6) | 2607 (22.0) |
| No | 150 865 (74.2) | 160 935 (88.4) | 9260 (78.0) |
| Missing | 38 787 (16.0) | 23 303 (11.3) | 2068 (14.8) |
IQR: interquartile range; CAKUT: congenital anomalies of the kidneys or urinary tract; N/A: not available.
Tables 2a and b (female recipient and male recipient, respectively) summarize the composition of the observed experience for recipients of female and male donors within each donor age group, pooled across all three cohorts. Among both female and male recipients across all donor age groups, a greater proportion of the observation time was contributed by recipients of male donors. Regardless of the recipient sex and across all donor ages, a greater proportion of the observation time contributed by recipients of female than male donors was for transplants where the donor:recipient weight ratio was <0.9. Conversely, a greater proportion of the observation time contributed by recipients of male than female donors represented transplants where the donor:recipient weight ratio was >0.9. Similarly, a greater proportion of the observation time for recipients of donors with cerebrovascular accident as the cause of death and with positive CMV status was contributed by recipients of female than male donors.
Table 2:
Composition of the contrasted experience, pooled across SRTR, ANZDATA and CTS, by donor age and sex, for (A) female and (B) male recipients.
| Percentage of person-years of observation | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Donor age | 5–19 years | 20–34 years | 35–49 years | 50–59 years | ≥60 years | |||||
| Donor sex | Female | Male | Female | Male | Female | Male | Female | Male | Female | Male |
| (a) Female recipients | ||||||||||
| Number of patients | 8555 | 17 633 | 12 765 | 28 116 | 22 032 | 26 971 | 17 294 | 18 377 | 13 224 | 12 230 |
| Person-years of observation | 67 771 | 142 009 | 91 575 | 207 402 | 151 765 | 186 022 | 108 230 | 113 307 | 70 938 | 64 017 |
| Graft failures | 2325 | 5042 | 3029 | 6666 | 5793 | 6812 | 4674 | 4728 | 3510 | 3303 |
| Age in years at transplant (%) | ||||||||||
| 0–12 | 7.2 | 6.5 | 3.1 | 2.6 | 1.6 | 1.2 | 0.4 | 0.5 | 0.1 | 0.2 |
| 13–24 | 11.6 | 11.8 | 8.4 | 8.5 | 5.4 | 5.3 | 3.0 | 2.6 | 1.0 | 1.0 |
| 25–44 | 36.6 | 36.3 | 37.2 | 37.9 | 35.5 | 34.9 | 26.9 | 27.2 | 15.1 | 13.7 |
| 45–59 | 32.0 | 33.5 | 36.2 | 36.4 | 40.2 | 40.8 | 46.5 | 45.4 | 36.6 | 37.2 |
| ≥60 years | 12.7 | 11.8 | 15.1 | 14.7 | 17.3 | 17.7 | 23.1 | 24.3 | 47.0 | 47.9 |
| Recipient race (%) | ||||||||||
| White | 74.9 | 74.8 | 74.6 | 74.2 | 75.0 | 74.8 | 76.0 | 75.4 | 78.7 | 79.7 |
| Others | 25.1 | 25.2 | 25.4 | 25.8 | 25.0 | 25.2 | 24.0 | 24.6 | 21.3 | 20.3 |
| Missing | 7.3 | 7.9 | 9.3 | 8.6 | 11.4 | 11.4 | 14.4 | 15.2 | 26.0 | 26.2 |
| Primary disease (%) | ||||||||||
| Diabetes | 11.7 | 12.0 | 13.5 | 12.8 | 13.2 | 12.3 | 14.2 | 13.0 | 12.3 | 12.6 |
| Hypertension | 11.6 | 12.1 | 12.8 | 13.5 | 13.2 | 13.6 | 13.6 | 14.0 | 14.2 | 13.1 |
| GN | 32.5 | 33.0 | 31.3 | 32.4 | 31.0 | 30.4 | 28.7 | 28.5 | 26.2 | 25.3 |
| CAKUT | 11.9 | 10.7 | 10.0 | 9.9 | 10.0 | 9.8 | 9.1 | 9.5 | 8.7 | 9.2 |
| Cystic | 13.0 | 13.6 | 14.2 | 14.0 | 14.8 | 16.3 | 17.3 | 17.5 | 19.6 | 21.2 |
| Other | 19.2 | 18.6 | 18.2 | 17.4 | 17.8 | 17.5 | 17.0 | 17.5 | 18.6 | 18.7 |
| Missing | 1.5 | 1.7 | 1.5 | 1.7 | 2.2 | 2.3 | 2.8 | 2.5 | 3.5 | 3.8 |
| Cold ischemia time (h) | ||||||||||
| 0–18 | 50.9 | 49.1 | 51.6 | 51.2 | 53.9 | 52.0 | 55.1 | 53.5 | 60.1 | 58.9 |
| 19–24 | 25.7 | 26.6 | 25.7 | 26.0 | 25.6 | 26.3 | 26.0 | 25.5 | 24.0 | 25.2 |
| 25–36 | 19.5 | 20.5 | 18.8 | 19.3 | 17.3 | 17.9 | 15.9 | 17.7 | 13.7 | 13.7 |
| >36 | 3.9 | 3.8 | 3.9 | 3.5 | 3.2 | 3.8 | 3.1 | 3.3 | 2.1 | 2.3 |
| Missing | 8.9 | 8.8 | 8.2 | 7.7 | 8.7 | 7.5 | 7.8 | 7.4 | 7.4 | 7.6 |
| Transplant year | ||||||||||
| 1988–1994 | 32.0 | 34.7 | 31.1 | 31.6 | 22.2 | 23.7 | 16.6 | 17.0 | 10.3 | 10.2 |
| 1995–1999 | 21.6 | 22.4 | 18.7 | 18.6 | 19.8 | 19.2 | 18.0 | 16.9 | 15.7 | 15.3 |
| 2000–2004 | 19.6 | 18.6 | 16.8 | 17.2 | 21.2 | 20.0 | 22.4 | 21.0 | 22.4 | 22.0 |
| 2005–2009 | 15.0 | 13.7 | 15.7 | 16.0 | 19.2 | 19.0 | 21.6 | 22.6 | 25.1 | 25.0 |
| 2010–2013 | 8.7 | 7.9 | 12.7 | 12.2 | 13.3 | 13.3 | 16.0 | 16.5 | 19.9 | 20.8 |
| 2014–2018 | 3.1 | 2.6 | 4.9 | 4.5 | 4.3 | 4.8 | 5.3 | 5.9 | 6.6 | 6.8 |
| Pre-emptive transplant | ||||||||||
| Yes | 7.5 | 6.8 | 6.9 | 6.5 | 6.1 | 5.9 | 5.7 | 5.6 | 4.8 | 4.5 |
| Missing | 7.1 | 6.9 | 7.5 | 6.3 | 6.7 | 6.0 | 6.1 | 5.9 | 4.9 | 5.1 |
| Donor-recipient weight ratioa | ||||||||||
| <0.7 | 40.1 | 26.3 | 13.5 | 4.4 | 10.6 | 3.2 | 10.2 | 3.1 | 9.7 | 2.4 |
| 0.7–<0.9 | 22.3 | 17.8 | 23.9 | 14.3 | 22.0 | 13.0 | 23.7 | 12.8 | 22.9 | 12.8 |
| >0.9 | 37.6 | 55.8 | 62.6 | 81.3 | 67.5 | 83.7 | 66.1 | 84.1 | 67.4 | 84.8 |
| Missingb | 47.9 | 49.3 | 50.1 | 49.2 | 54.3 | 54.5 | 56.7 | 58.5 | 69.6 | 71.0 |
| Missing (SRTR & ANZ)c | 18.1 | 20.5 | 16.1 | 16.7 | 14.4 | 13.0 | 12.2 | 10.9 | 12.4 | 11.3 |
| HLA mismatches | ||||||||||
| 0–1 | 15.4 | 15.5 | 16.8 | 15.5 | 16.5 | 16.9 | 16.5 | 15.0 | 11.5 | 11.1 |
| 2–3 | 34.7 | 35.7 | 35.6 | 35.6 | 37.7 | 38.1 | 37.7 | 38.3 | 35.6 | 37.7 |
| 4–6 | 49.9 | 48.9 | 47.6 | 48.9 | 45.8 | 45.0 | 45.8 | 46.7 | 52.9 | 51.3 |
| Missing | 3.7 | 3.7 | 3.7 | 3.7 | 4.2 | 4.0 | 5.1 | 4.6 | 6.7 | 6.8 |
| PRA | ||||||||||
| 0% | 66.4 | 66.0 | 65.3 | 64.1 | 65.1 | 65.8 | 68.0 | 67.7 | 72.5 | 71.5 |
| >0% | 33.6 | 34.0 | 34.7 | 35.9 | 34.9 | 34.2 | 32.0 | 32.3 | 27.5 | 28.5 |
| Missing | 11.6 | 11.5 | 13.4 | 12.5 | 14.8 | 13.8 | 15.5 | 16.1 | 17.5 | 17.9 |
| Donor cause of death | ||||||||||
| CVA | 13.4 | 7.7 | 32.8 | 14.4 | 66.0 | 45.0 | 75.7 | 59.8 | 81.0 | 67.5 |
| Others | 86.6 | 92.3 | 67.2 | 85.6 | 34.0 | 55.0 | 24.3 | 40.2 | 19.0 | 32.5 |
| Missing | 7.0 | 6.4 | 7.3 | 6.8 | 5.7 | 6.2 | 4.7 | 5.3 | 4.7 | 4.3 |
| CMV positive donor | ||||||||||
| Yes | 52.4 | 50.9 | 61.3 | 53.9 | 64.8 | 57.4 | 69.9 | 62.9 | 75.4 | 68.8 |
| Missing | 9.0 | 9.8 | 9.8 | 9.8 | 10.0 | 10.1 | 11.1 | 10.8 | 16.4 | 16.8 |
| Donor hypertension | ||||||||||
| Yes | 0.6 | 0.7 | 5.6 | 3.3 | 19.9 | 17.0 | 26.7 | 27.5 | 30.2 | 29.5 |
| Missing | 21.2 | 23.6 | 20.2 | 20.6 | 14.2 | 14.9 | 10.8 | 10.9 | 8.1 | 7.3 |
| (b) Male recipients | ||||||||||
| Number of patients | 11 793 | 25 814 | 19 483 | 44 981 | 33 612 | 44 996 | 26 958 | 31 570 | 22 374 | 22 586 |
| Person-years of observation | 90 449 | 205 105 | 136 614 | 327 261 | 223 773 | 302 437 | 160 860 | 185 999 | 111 167 | 111 864 |
| Graft failures | 3260 | 6747 | 4449 | 9720 | 8788 | 10 567 | 7195 | 7984 | 5938 | 5878 |
| Age in years at transplant (%) | ||||||||||
| 0–12 | 8.5 | 7.7 | 3.1 | 2.7 | 1.5 | 1.3 | 0.5 | 0.3 | 0.1 | 0.1 |
| 13–24 | 11.7 | 11.4 | 7.9 | 7.9 | 4.8 | 4.5 | 2.9 | 2.4 | 1.9 | 0.8 |
| 25–44 | 36.5 | 37.3 | 38.5 | 38.7 | 35.3 | 35.3 | 27.7 | 26.1 | 11.4 | 13.3 |
| 45–59 | 31.5 | 32.6 | 35.9 | 36.5 | 40.8 | 41.6 | 45.0 | 46.1 | 42.6 | 36.7 |
| ≥60 years | 11.8 | 11.0 | 14.7 | 14.2 | 17.5 | 17.4 | 23.9 | 25.1 | 42.4 | 49.2 |
| Recipient race (%) | ||||||||||
| White | 77.9 | 77.4 | 76.7 | 76.6 | 76.8 | 76.9 | 77.8 | 77.8 | 82.3 | 82.9 |
| Others | 22.1 | 22.6 | 23.3 | 23.4 | 23.2 | 23.1 | 22.2 | 22.2 | 17.7 | 17.1 |
| Missing | 8.7 | 8.2 | 9.6 | 9.5 | 12.0 | 12.5 | 15.1 | 16.0 | 27.6 | 26.9 |
| Primary disease (%) | ||||||||||
| Diabetes | 12.5 | 13.4 | 14.9 | 14.6 | 15.8 | 15.5 | 16.7 | 17.4 | 17.4 | 17.1 |
| Hypertension | 15.5 | 16.2 | 17.1 | 16.9 | 17.0 | 17.6 | 17.3 | 17.5 | 17.3 | 16.8 |
| GN | 33.1 | 33.4 | 32.8 | 33.7 | 33.0 | 33.0 | 30.7 | 31.1 | 30.7 | 30.4 |
| CAKUT | 11.6 | 10.3 | 8.1 | 8.1 | 6.4 | 6.5 | 5.8 | 5.4 | 4.6 | 4.9 |
| Cystic | 10.8 | 10.6 | 10.3 | 10.7 | 11.5 | 11.5 | 12.9 | 13.4 | 12.7 | 12.9 |
| Other | 16.5 | 16.1 | 16.8 | 15.9 | 16.4 | 16.0 | 16.6 | 15.2 | 17.5 | 17.9 |
| Missing | 1.7 | 2.0 | 1.9 | 1.8 | 2.4 | 2.2 | 3.0 | 2.9 | 4.6 | 4.3 |
| Cold ischemia time (h) | ||||||||||
| 0–18 | 49.1 | 50.3 | 52.0 | 50.8 | 53.6 | 52.8 | 54.9 | 54.2 | 61.0 | 60.2 |
| >18–24 | 26.5 | 26.1 | 24.8 | 26.2 | 25.7 | 26.3 | 25.6 | 26.1 | 23.5 | 24.2 |
| >24–36 | 19.9 | 19.4 | 19.4 | 19.1 | 17.3 | 17.4 | 16.2 | 16.4 | 13.2 | 13.2 |
| >36 | 4.5 | 4.2 | 3.8 | 3.9 | 3.4 | 3.6 | 3.3 | 3.2 | 2.3 | 2.4 |
| Missing | 9.6 | 8.3 | 8.5 | 7.9 | 8.5 | 8.1 | 8.2 | 7.5 | 7.6 | 7.2 |
| Transplant year | ||||||||||
| 1988–1994 | 34.2 | 36.2 | 30.2 | 32.4 | 22.2 | 23.3 | 15.9 | 16.4 | 9.5 | 9.7 |
| 1995–1999 | 22.3 | 22.6 | 19.0 | 19.0 | 20.5 | 19.7 | 17.6 | 16.7 | 14.5 | 13.7 |
| 2000–2004 | 18.5 | 17.8 | 17.2 | 17.1 | 21.1 | 19.8 | 21.9 | 20.4 | 20.5 | 21.6 |
| 2005–2009 | 14.2 | 13.6 | 15.6 | 15.5 | 18.5 | 19.0 | 22.7 | 22.8 | 26.5 | 25.1 |
| 2010–2013 | 8.2 | 7.4 | 12.9 | 11.5 | 13.4 | 13.3 | 16.5 | 17.6 | 21.8 | 22.5 |
| 2014–2018 | 2.6 | 2.4 | 5.1 | 4.4 | 4.4 | 4.9 | 5.5 | 6.1 | 7.2 | 7.5 |
| Pre-emptive transplant | ||||||||||
| Yes | 5.8 | 6.2 | 6.1 | 5.6 | 4.6 | 4.8 | 4.6 | 4.1 | 3.3 | 3.2 |
| Missing | 8.2 | 7.4 | 7.3 | 6.2 | 6.7 | 6.2 | 5.9 | 5.3 | 5.1 | 4.5 |
| Donor:recipient weight ratioa | ||||||||||
| <0.7 | 47.6 | 32.1 | 26.7 | 10.7 | 22.0 | 8.6 | 21.0 | 8.8 | 20.0 | 7.0 |
| 0.7–<0.9 | 27.0 | 24.9 | 30.0 | 25.3 | 31.0 | 23.5 | 31.3 | 23.1 | 34.1 | 24.4 |
| >0.9 | 25.4 | 43.1 | 43.3 | 64.1 | 47.0 | 68.0 | 47.7 | 68.1 | 45.9 | 68.6 |
| Missingb | 51.0 | 50.8 | 52.0 | 51.4 | 55.7 | 55.4 | 58.8 | 59.6 | 70.9 | 71.9 |
| Missing (SRTR and ANZ)c | 19.7 | 20.5 | 16.3 | 17.0 | 14.2 | 12.5 | 12.3 | 10.6 | 11.2 | 10.2 |
| HLA mismatches | ||||||||||
| 0–1 | 14.7 | 13.3 | 14.1 | 14.2 | 15.1 | 14.3 | 14.3 | 13.6 | 10.2 | 9.8 |
| 2–3 | 37.1 | 36.6 | 36.4 | 36.4 | 38.4 | 38.2 | 38.4 | 37.9 | 34.8 | 36.4 |
| 4–6 | 48.3 | 50.1 | 49.5 | 49.4 | 46.6 | 47.5 | 47.3 | 48.5 | 55.0 | 53.8 |
| Missing | 3.8 | 3.8 | 4.2 | 3.7 | 4.7 | 4.1 | 5.4 | 5.2 | 7.2 | 7.2 |
| PRA | ||||||||||
| 0% | 78.6 | 78.3 | 78.4 | 78.2 | 79.5 | 79.5 | 80.7 | 79.6 | 81.1 | 80.1 |
| >0% | 21.4 | 21.7 | 21.6 | 21.8 | 20.5 | 20.5 | 19.3 | 20.4 | 18.9 | 19.9 |
| Missing | 12.6 | 11.6 | 13.4 | 12.1 | 15.2 | 14.5 | 16.8 | 16.3 | 18.1 | 18.1 |
| Donor cause of death | ||||||||||
| CVA | 13.2 | 7.4 | 32.1 | 14.1 | 66.4 | 44.8 | 76.3 | 60.0 | 80.3 | 66.1 |
| Others | 86.8 | 92.6 | 67.9 | 85.9 | 33.6 | 55.2 | 23.7 | 40.0 | 19.7 | 33.9 |
| Missing | 6.8 | 7.6 | 6.7 | 7.5 | 5.7 | 6.6 | 4.6 | 5.5 | 4.5 | 4.6 |
| CMV-positive donor | ||||||||||
| Yes | 53.5 | 50.8 | 60.1 | 53.4 | 65.3 | 57.6 | 69.5 | 62.4 | 74.4 | 68.1 |
| Missing | 9.9 | 10.5 | 9.9 | 10.5 | 10.2 | 10.1 | 11.2 | 11.8 | 18.5 | 17.7 |
| Donor hypertension | ||||||||||
| Yes | 0.6 | 0.6 | 5.0 | 3.2 | 19.1 | 16.2 | 27.3 | 27.5 | 29.9 | 30.0 |
| Missing | 22.1 | 24.4 | 18.8 | 20.8 | 14.0 | 14.5 | 10.2 | 10.4 | 7.7 | 7.2 |
Data not available from CTS.
Proportion of missing was calculated for population from all three cohorts.
Proportion of missing was calculated for population from SRTR and ANZDATA.
Because the unit analysis was person-time, rather than person, the characteristics presented are weighted by a factor derived from the number of person-years of observation and number of events. For example, 26.2% of the person-years contributed by female recipients of female donors ≥60 years had GN as the primary kidney disease.
GN: glomerulonephritis; CAKUT: congenital anomalies of the kidneys or urinary tract.
Supplementary data, Tables S1a, b and c (SRTR, CTS and ANZDATA, respectively) summarize the composition of the observed experience for recipients of male and female donors within each donor age interval for each cohort separately.
Comparison of death-censored graft loss rates by donor sex
There was a significant interaction between donor age and donor–recipient sex combination in both SRTR (P < .0001) and CTS (P < .001) cohorts; this interaction was not significant in the smaller ANZDATA cohort (P = .34). Forest plots (Fig. 1) show the combined aHR for death-censored graft loss for recipients of a female versus male donor in each donor age category in the setting of a female recipient (Fig. 1a) and a male recipient (Fig. 1b), as well as for each cohort separately.
Figure 1:
Individual patient data meta-analyses combining data from SRTR, CTS and ANZDATA donor age-specific relative hazards of death-censored graft failure in (a) female recipients and (b) male recipients of female donor compared with male donor. Models were as adjusted for recipient age at transplant (piece-wise linear spline), recipient race (white/other), primary diagnosis (diabetes, hypertension, glomerulonephritis, congenital anomalies of the kidneys or urinary tract, cystic kidney disease, other), pre-transplant dialysis (yes/no), PRA (0% vs >0%), HLA mismatch (continuous), donor:recipient weight ratio (<0.7, 0.7–<0.9, ≥0.9; not available for CTS), transplant year (continuous), donor CMV (not included for ANZDATA as it was missing in >40%), donor CVA, donor hypertension and cold ischemia time (≤18, >18–24, >24–36, >36 h), conditioned on graft function >90 days. HR are interpreted as the risk of death censored graft loss associated with a female donor relative to that with a male donor. RE model = random effects model (pooled estimate).
As illustrated in Fig. 1, the association between donor sex and graft loss differed by recipient sex and by donor age. When the recipient was female, there were no differences in the hazards of graft loss by donor sex when the donor was <60 years. However, among female recipients with a donor ≥60 years, the hazards of graft loss were significantly lower with a female than a male donor (combined aHR 0.90, 95% CI 0.86–0.94). When the recipient was male and the donor was <50 years old, a female donor was associated with a significantly higher hazard of graft loss than a male donor: for donors 5–19 years old the aHR was 1.11 (95% CI 1.02–1.21), for donors 20–34 years old the aHR was 1.08 (95% CI 1.02–1.15) and for donors 35–49 years old the aHR was 1.07 (95% CI 1.04–1.10). There was no difference in graft loss hazards by donor sex among male recipients of a donor 50–59 years old. Among male recipients of a donor aged ≥60 years, the combined aHR suggested a slightly lower hazard of graft loss with a female than with a male donor (combined aHR 0.92, 95% CI 0.82–1.03), but this was not statistically significant.
Between cohort heterogeneity
Among both male and female recipients, the aHRs for each cohort were similar. However, there was some heterogeneity between cohorts, even after excluding the ANZDATA (Supplementary data, Table S2).
Fitted absolute graft loss rates
Fitted absolute graft loss rates were comparable across cohorts (Supplementary data, Table S3). Overall, rates were lowest in CTS and highest in ANZDATA, though differences were small. Pooled fitted absolute graft loss rates are shown in Table 3. Even where there were statistically significant differences by donor sex, absolute differences in graft loss rates by donor sex were small. Among female recipients with a donor ≥60 years, graft loss rates were only lower by 1.7 per 100 person-years of observation (pyr) with a female donor than with a male donor. Among male recipients with donors <50 years, graft loss rates were lower with a male donor than a female donor by 1.2 per 100 pyr for donors 5–19 years old, 0.5 per 100 pyr for donors 20–34 years old and 0.2 per 100 pyr for donors 35–49 years old.
Table 3:
Pooled fitted graft loss rates per 100 person-years at 10 years post-transplant.
| Female recipient | Male recipient | |||||||
|---|---|---|---|---|---|---|---|---|
| Female donor | Male donor | Female donor | Male donor | |||||
| Donor age (years) | Graft loss rate | 95% CI | Graft loss rate | 95% CI | Graft loss rate | 95% CI | Graft loss rate | 95% CI |
| 5–19 | 13.1 | 12.5–13.7 | 13.2 | 11.9–14.7 | 13.3 | 12.9–13.9 | 12.1 | 11.7–12.4 |
| 20–34 | 13.7 | 12.7–14.9 | 14.1 | 11.9–16.6 | 13.5 | 12.1–15.1 | 13.0 | 10.6–15.8 |
| 35–49 | 16.7 | 15.0–18.6 | 16.5 | 16.0–16.9 | 16.8 | 15.3–18.4 | 16.6 | 16.0–17.3 |
| 50–59 | 20.3 | 18.4–22.3 | 20.0 | 19.4–20.6 | 21.0 | 18.2–24.1 | 21.6 | 18.3–25.6 |
| 60+ | 25.4 | 24.6–26.3 | 27.1 | 26.2–28.0 | 27.2 | 26.5–28.0 | 28.5 | 25.8–31.3 |
The fitted rates were calculated for each cohort separately based on the Cox proportional hazard models with covariates held fixed as: recipient age, 51 years at transplant; recipient race, white; primary diagnosis, glomerulonephritis; donor:recipient weight ratio, ≥0.9 (not available in CTS); pre-transplant dialysis, yes; PRA, 0%; HLA mismatch, 4; transplant year, 2010; donor CVA as cause of death, no; donor hypertension, no; and cold ischemia time, ≤18 h. Fitted rates for SRTR, CTS and ANZDATA were then pooled in a meta-analysis using a random effects model.
Sensitivity analyses
For both female and male recipients across all donor age categories, the HR from models excluding donor:recipient weight ratio (Supplementary data, Table S4) were slightly higher than those from models including this variable. The results of models excluding the donor CMV were identical to those that included it (Supplementary data, Table S5). The model excluding ANZDATA showed similar estimates to the main analyses (Supplementary data, Table S2).
Comparison of recipient mortality by donor sex
The results of similar models that compared recipient mortality rates by donor sex are shown in Supplementary data, Fig. S1. There were no significant differences in mortality by donor sex, regardless of donor age.
DISCUSSION
This study combined the data of 461 364 transplants recipients from three of the largest transplant registries worldwide in an individual patient data meta-analysis to demonstrate an association between donor sex and the hazard of graft loss that differs not only by recipient sex but also by donor age. When the recipient was female, kidneys from female donors <60 years old showed similar outcomes to kidneys from male donors of the same age, whereas kidneys from female donors ≥60 years old showed lower graft loss rates than kidneys from male donors ≥60 years old. In contrast, when the recipient was male, graft loss rates were higher with female than male donors <50 years old, but did not differ by donor sex when the donor was ≥50 years old. In fact, the point estimate suggested that when the donor was ≥60 years old, female donors may be associated with lower hazards of graft loss than male donors.
Female sex and older donor age have been described as markers of inferior donor organ quality [1, 2, 14]. Our data provide evidence that the association between donor sex and graft loss differs by donor age and by recipient sex. Older female donors are not inferior to older male donors among male recipients and even provide slightly superior outcomes among female recipients. These findings, which suggest that functional reserve may be better in kidneys from older women than older men, are supported by a recent study showing sex differences in age-related GFR decline [30]. While GFR was lower in women than in men, a steeper GFR decline over time was observed in men ≥60 years compared with the same aged women [30]. An analysis of American adult kidney transplant recipients showed male donors to be associated with higher rates of delayed graft function than female donors [11]. Our data suggest that older female kidneys may have advantages even outside the female environment; mechanisms for this advantage need further exploration.
Among male recipients, female donors under 50 years old were associated with a slightly higher hazard of graft loss than male donors under 50 years old. It is likely that age-related nephron loss is minimal among young donors such that the greater nephron mass afforded by male kidneys provides an advantage. The possibility that androgens promote kidney growth [31] post-transplant differently by donor sex must also be considered. Among female recipients with donors <60 years old, we saw no association between donor sex and graft loss rate. It is likely that the advantage of greater nephron mass offered by a male donor is counterbalanced by the disadvantage of an immunologic reaction against the HY-antigen [8, 9], resulting in no net difference in outcome by donor sex. This finding is consistent with a US Renal Data System study that showed a smaller advantage associated with a male (vs female) donor for female recipients than for male recipients [26]. A UK Transplant Registry study of 1309 pediatric kidney transplant recipients (mean donor age 28–32 years) showed no independent association between donor sex and graft loss rates for either male or female recipients [32], but power was limited to detect differences. An earlier analysis from CTS including 124 911 adult-only, first kidney transplantations demonstrated a higher risk of death or graft loss in female recipients of male donors compared with female recipients of female donors and similarly a higher risk of death or graft loss in male recipients of female than male donors [1]. This study analyzed the outcome graft failure or death (including death with graft function) while our study analyzed death-censored graft survival, making a direct comparison difficult.
This is the largest and the first study investigating the modifying effects of both recipient sex and donor age on the association between donor sex and graft survival. Despite its strengths, this study has some limitations. Due to the nature of observational studies, drawing causal inferences is not possible. It is also possible our estimates of the association between donor sex and graft loss remain confounded by variables that were not available in the databases. Since donor sex is reasonably likely to be random across recipients in the setting of deceased donor transplantation (and therefore unassociated with specific recipient factors), we believe that major confounding by recipient factors is unlikely. However, confounding is not impossible and may have influenced results.
While most variables were recorded similarly across registries, there were some differences. Information on donor weight was not available in the CTS, therefore donor:recipient weight ratio could not be included in the CTS analysis. The sensitivity analyses excluding donor:recipient weight ratio from the models for SRTR and ANZDATA suggest that HR may have been overestimated in the CTS models, resulting in overestimation of pooled HR. If this was the case, then it is possible that older female donors are associated with a slightly larger advantage over older male donors among female recipients and may even have an advantage over older male donors in male recipients. Overestimation of the HR may also mean that the poorer outcomes observed for younger female than male donors among male recipients may be exaggerated. The observed heterogeneity between cohorts may be partly related to the inability to adjust for donor:recipient weight ratio in CTS. The lack of representation of many countries in the analysis is also a limitation. Lastly, reporting to the SRTR and ANZDATA is obligatory, but the CTS is a voluntary registry, therefore the possibility of reporting bias could not be excluded.
CONCLUSIONS
By combining data from three large registries, we demonstrate a modifying effect of donor age on the association between donor sex and graft survival. Among male recipients of young donor kidneys, graft loss hazards were higher with female than male donors, whereas among female recipients of young donor kidneys, there were no differences in graft loss rates by donor sex. Notably, among both male and female recipients of kidneys from donors over 60 years, graft loss rates were lower with female than male donors—though this was only significant for female recipients. Our findings point towards better functional reserve of older female donor kidneys and should result in greater confidence in accepting such organs.
Supplementary Material
ACKNOWLEDGEMENTS
Some data reported here have been supplied by the Hennepin Healthcare Research Institute (HHRI) as the contractor for the Scientific Registry of Transplant Recipients (SRTR). The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy of or interpretation by the SRTR or the US Government.
Some data reported here have been supplied by the Australia and New Zealand Dialysis and Transplant Registry (ANZDATA). The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy or interpretation of ANZDATA.
We thank the transplant registries Eurotransplant, Italian Centro Nazionale Trapianti, Organització Catalana de Trasplantaments, Nederlandse Transplantatie Stichting and UK Transplant for collaboration and data exchange with Collaborative Transplant Study (CTS) and gratefully acknowledge the generous support of the transplantation centres or hospitals that provided data for this study to CTS.
Contributor Information
Anette Melk, Children's Hospital, Hannover Medical School, Hannover, Germany.
Rizky I Sugianto, Children's Hospital, Hannover Medical School, Hannover, Germany.
Xun Zhang, Research Institute of the McGill University Health Centre, Centre for Outcomes Research and Evaluation, Montréal, QC, Canada.
Mourad Dahhou, Research Institute of the McGill University Health Centre, Centre for Outcomes Research and Evaluation, Montréal, QC, Canada.
Bernd Döhler, Institute of Immunology, Heidelberg University Hospital, Heidelberg, Germany.
Caner Süsal, Transplant Immunology Research Center of Excellence, Koç University, Istanbul, Turkey.
Ruth Sapir-Pichhadze, Research Institute of the McGill University Health Centre, Centre for Outcomes Research and Evaluation, Montréal, QC, Canada; Department of Medicine, Division of Nephrology, McGill University, Montréal, QC, Canada; Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montréal, QC, Canada.
Germaine Wong, School of Public Health, University of Sydney, Sydney, New South Wales, Australia.
Bethany J Foster, Research Institute of the McGill University Health Centre, Centre for Outcomes Research and Evaluation, Montréal, QC, Canada; Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montréal, QC, Canada; Department of Pediatrics, Division of Nephrology, McGill University, Montréal, QC, Canada.
AUTHORS’ CONTRIBUTIONS
A.M. interpreted the data, and drafted, finalized and approved the article. R.I.S. interpreted the data, and drafted, finalized and approved the article. X.Z. prepared, analyzed and interpreted the data, and revised and approved the article. M.D. prepared, analyzed and interpreted the data, and revised and approved the article. B.D. prepared, analyzed and interpreted the data, and revised and approved the article. C.S. interpreted the data, and revised and approved the article. R.S.-P. interpreted the data, and revised and approved the article. G.W. interpreted the data, and revised and approved the article. B.J.F. designed the study, acquired funding, interpreted the data, and revised, finalized and approved the article.
FUNDING
This study was supported by an operating grant from the Canadian Institutes of Health Research (PJT165832). C.S. is supported by the European Union's Horizon 2020 Research and Innovation Programme (grant no. 952 512). R.S.-P. is supported by Fonds de recherche du Quebec—Santé chercheur boursier clinicien award (grant no. 311 583). R.I.S. is partially supported by the Women in Transplantation Fellowship Award 2021–2023 (One Lambda Inc.).
DATA AVAILABILITY STATEMENT
The SRTR and ANZDATA data are publicly available. Procedures to request data are available at www.srtr.org and www.anzdata.org.au.
CONFLICT OF INTEREST STATEMENT
All authors declared no competing interests.
REFERENCES
- 1. Zeier M, Dohler B, Opelz G et al. The effect of donor gender on graft survival. J Am Soc Nephrol 2002;13:2570–6. 10.1097/01.ASN.0000030078.74889.69 [DOI] [PubMed] [Google Scholar]
- 2. Kasiske BL, Umen AJ. The influence of age, sex, race, and body habitus on kidney weight in humans. Arch Pathol Lab Med 1986;110:55–60. [PubMed] [Google Scholar]
- 3. Kim YS, Moon JI, Kim DK et al. Ratio of donor kidney weight to recipient bodyweight as an index of graft function. Lancet North Am Ed 2001;357:1180–1. 10.1016/S0140-6736(00)04377-4 [DOI] [PubMed] [Google Scholar]
- 4. Hughson M, Farris AB III, Douglas-Denton R et al. Glomerular number and size in autopsy kidneys: the relationship to birth weight. Kidney Int 2003;63:2113–22. 10.1046/j.1523-1755.2003.00018.x [DOI] [PubMed] [Google Scholar]
- 5. Kim SJ, Gill JS. H-Y incompatibility predicts short-term outcomes for kidney transplant recipients. J Am Soc Nephrol 2009;20:2025–33. 10.1681/ASN.2008101110 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Maenosono R, Nian Y, Iske J et al. Recipient sex and estradiol levels affect transplant outcomes in an age-specific fashion. Am J Transplant 2021;21:3239–55.. 10.1111/ajt.16611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Lepeytre F, Dahhou M, Zhang X et al. Association of sex with risk of kidney graft failure differs by age. J Am Soc Nephrol 2017;28:3014–23. 10.1681/ASN.2016121380 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gratwohl A, Dohler B, Stern M et al. H-Y as a minor histocompatibility antigen in kidney transplantation: a retrospective cohort study. Lancet North Am Ed 2008;372:49–53. 10.1016/S0140-6736(08)60992-7 [DOI] [PubMed] [Google Scholar]
- 9. Tan JC, Wadia PP, Coram M et al. H-Y antibody development associates with acute rejection in female patients with male kidney transplants. Transplantation 2008;86:75–81. 10.1097/TP.0b013e31817352b9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Anderson S, Eldadah B, Halter JB et al. Acute kidney injury in older adults. J Am Soc Nephrol 2011;22:28–38. 10.1681/ASN.2010090934 [DOI] [PubMed] [Google Scholar]
- 11. Aufhauser DD Jr, Wang Z, Murken DR et al. Improved renal ischemia tolerance in females influences kidney transplantation outcomes. J Clin Invest 2016;126:1968–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Vinson AJ, Zhang X, Dahhou M et al. Age-dependent sex differences in graft loss after kidney transplantation. Transplantation 2022;106:1473–84. [DOI] [PubMed] [Google Scholar]
- 13. Vinson A, Zhang X, Dahhou M et al. A multinational cohort study uncovered sex differences in excess mortality after kidney transplant. Kidney Int 2023;103:1131–43. 10.1016/j.kint.2023.01.022 [DOI] [PubMed] [Google Scholar]
- 14. Oppenheimer F, Aljama P, Asensio PC et al. The impact of donor age on the results of renal transplantation. Nephrol Dial Transplant 2004;19 Suppl 3:iii11–5. 10.1093/ndt/gfh1008 [DOI] [PubMed] [Google Scholar]
- 15. Schmitt R, Melk A. New insights on molecular mechanisms of renal aging. Am J Transplant 2012;12:2892–900. 10.1111/j.1600-6143.2012.04214.x [DOI] [PubMed] [Google Scholar]
- 16. Melk A, Schmidt BM, Braun H et al. Effects of donor age and cell senescence on kidney allograft survival. Am J Transplant 2009;9:114–23. 10.1111/j.1600-6143.2008.02500.x [DOI] [PubMed] [Google Scholar]
- 17. Braun H, Schmidt BM, Raiss M et al. Cellular senescence limits regenerative capacity and allograft survival. J Am Soc Nephrol 2012;23:1467–73. 10.1681/ASN.2011100967 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Westhoff JH, Schildhorn C, Jacobi C et al. Telomere shortening reduces regenerative capacity after acute kidney injury. J Am Soc Nephrol 2010;21:327–36. 10.1681/ASN.2009010072 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. He A, Sarwar A, Thole LML et al. Renal inflamm-aging provokes intra-graft inflammation following experimental kidney transplantation. Am J Transplant 2022;22:2529–47. 10.1111/ajt.17154 [DOI] [PubMed] [Google Scholar]
- 20. Reutzel-Selke A, Jurisch A, Denecke C et al. Donor age intensifies the early immune response after transplantation. Kidney Int 2007;71:629–36. 10.1038/sj.ki.5002098 [DOI] [PubMed] [Google Scholar]
- 21. Singh A, Stablein D, Tejani A. Risk factors for vascular thrombosis in pediatric renal transplantation: a special report of the North American Pediatric Renal Transplant Cooperative Study. Transplantation 1997;63:1263–7. 10.1097/00007890-199705150-00012 [DOI] [PubMed] [Google Scholar]
- 22. Bresnahan BA, McBride MA, Cherikh WS et al. Risk factors for renal allograft survival from pediatric cadaver donors: an analysis of united network for organ sharing data. Transplantation 2001;72:256–61. 10.1097/00007890-200107270-00016 [DOI] [PubMed] [Google Scholar]
- 23. Chesnaye NC, van Stralen KJ, Bonthuis M et al. The association of donor and recipient age with graft survival in paediatric renal transplant recipients in a European Society for Paediatric Nephrology/European Renal Association-European Dialysis and Transplantation Association Registry study. Nephrol Dial Transplant 2017;32:1949–56. 10.1093/ndt/gfx261 [DOI] [PubMed] [Google Scholar]
- 24. ANZDATA. ANZDATA —Australia and New Zealand Dialysis and Transplant Registry. Available from: https://www.anzdata.org.au/anzdata/ (1 August 2023, date last accessed).
- 25. Heidelberg University Institute of Immunology. Collaborative Transplant Study. Available from: https://www.ctstransplant.org/ (1 August 2023, date last accessed). [Google Scholar]
- 26. Tan JC, Kim JP, Chertow GM et al. Donor-recipient sex mismatch in kidney transplantation. Gend Med 2012;9:335–47.e2. 10.1016/j.genm.2012.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Crowther MJ, Riley RD, Staessen JA et al. Individual patient data meta-analysis of survival data using Poisson regression models. BMC Med Res Methodol 2012;12:34. 10.1186/1471-2288-12-34 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Thompson S, Kaptoge S, White I et al. Statistical methods for the time-to-event analysis of individual participant data from multiple epidemiological studies. Int J Epidemiol 2010;39:1345–59. 10.1093/ije/dyq063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Liu Y, De A. Multiple imputation by fully conditional specification for dealing with missing data in a large epidemiologic study. Int J Stat Med Res 2015;4:287–95. 10.6000/1929-6029.2015.04.03.7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Melsom T, Norvik JV, Enoksen IT et al. Sex differences in age-related loss of kidney function. J Am Soc Nephrol 2022;33:1891–902. 10.1681/ASN.2022030323 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Laouari D, Vergnaud P, Hirose T et al. The sexual dimorphism of kidney growth in mice and humans. Kidney Int 2022;102:78–95. 10.1016/j.kint.2022.02.027 [DOI] [PubMed] [Google Scholar]
- 32. Mudalige NL, Brown C, Marks SD. The impact of donor and recipient sex on kidney allograft survival in pediatric transplant recipients. Pediatr Nephrol 2022;37:209–16. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The SRTR and ANZDATA data are publicly available. Procedures to request data are available at www.srtr.org and www.anzdata.org.au.