Introduction
Kidney transplantation is the most effective treatment for kidney failure, improving survival and quality of life (1,2). Up to 60%–70% of kidney transplantations originate from deceased donors in North America (3). Although organ donor organizations do not consider donor-recipient size mismatch in allocation algorithms for deceased donor kidneys (4), this is one of the elements considered in the final decision to accept or decline an offer for a deceased donor kidney for transplantation.
Small donor and/or kidney sizes relative to recipient size were associated with a higher risk of graft failure in many studies (5–678). Although information on donor kidney weight is lacking at the time an offer is made, donor and recipient heights and weights are almost always available. Body surface area (BSA) correlates with kidney size better than total body weight in both adults and children (9–1011) and is hence the anthropometric feature most likely to capture donor-recipient size mismatches that are relevant to kidney graft function and survival.
However, risk factors that are more strongly associated with graft survival may modulate the relationship between size mismatch and graft survival. For instance, glomerular sclerosis associated with older donor age (12) may further reduce the nephron mass transplanted in patients with donor-recipient size mismatch, creating greater hyperfiltration in remaining nephrons and enhancing the adverse effect of size mismatch on long-term graft function and survival. On the other hand, older recipients have shorter expected survival time and are more likely to die with a functioning graft (13). This may not leave sufficient time for the adverse effect of donor-recipient size mismatch on graft survival to become clinically manifest.
We hence hypothesized that the effect of donor-recipient size mismatch would be modulated by both donor and recipient age. The aims of this study were to determine whether the relationship between donor-recipient size mismatch and graft survival is modified by donor and/or recipient age and to assess whether young donor age could attenuate the adverse association of size mismatch on graft survival in some recipient age categories. This is a clinically relevant question when deciding whether to accept or refuse a kidney from a deceased donor for transplantation because the effect of donor-recipient size mismatch is information that is considered before accepting an organ.
Materials and Methods
This study used data from the Scientific Registry of Transplant Recipients (SRTR). The SRTR data system includes data on all donors, wait-listed candidates, and transplant recipients in the United States, 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 of the activities of the OPTN and the SRTR contractors. All first single-deceased donor kidney transplantations performed between January 1, 2000 and March 1, 2018 were included in the cohort if recipient age was 18 years and older on the day of transplantation. Recipients of previous nonkidney solid organ transplants were excluded. Patients with missing or implausible data (weight: <5 or >250 kg, height <50 or >250 cm) on donor or recipient height and weight were excluded, as well as those whose donor age was <5 years. Patients were followed from the date of transplantation until death, return to dialysis, kidney retransplantation, or March 1, 2018, whichever occurred first. The study was approved by our local ethics review board (project identification 15–615-MUHC). This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for the reporting of cohort studies (14).
The main outcome was death-censored graft loss, defined as return to dialysis or kidney retransplantation. The main exposure variable was donor-recipient BSA ratio. BSA was calculated using the Mosteller formula (BSA in meters2 = √((weight × height)/3600)) (15–161718). Size mismatch was categorized as BSA ratio <0.80, 0.80–0.89, or 0.90–0.99 versus referent category (≥1.00). We considered the possibility that the association between donor-recipient size mismatch and graft outcome may be modified by both donor and recipient age at transplantation by including a donor-recipient size mismatch by recipient age by donor age interaction term. Donor age was categorized in four groups (<40, 40–49, 50–59, and ≥60 years) on the basis of previous studies (19). Recipient age at the time of transplantation was categorized in five groups (18–30, 31–40, 41–50, 51–60, and >60 years). The following donor-related covariates were included: sex, race, history of hypertension, diabetes, terminal serum creatinine, donation after circulatory death, and stroke as cause of death. Recipient-related covariates included sex, race, cause of kidney failure, time on dialysis prior to transplantation, most recent panel reactive antibodies prior to transplant (dichotomized as 0% versus >0%), and insurance status. Donor-recipient total HLA mismatches, transplant era, and cold ischemia time were also included as covariates.
Continuous variables are reported as means and SDs or medians and interquartile ranges depending on their distributions. Categorical variables are summarized as proportions. We fit a Cox regression model with death-censored graft loss as the dependent variable and donor-recipient BSA ratio as the main independent variable of interest to represent size mismatch. We examined the proportionality of hazards by graphic method and by refitting the models and comparing the results by censoring observations at 5 years and again censoring at 10 years. The assumption of proportionality was not violated. We included a three-way interaction term for the donor-recipient size mismatch by donor age by recipient age interaction. As the latter was significant, subsequent analyses were reported stratified by donor and recipient age. The rationale for the inclusion of potential confounders in the multivariable model is found in Supplemental Appendix. The amount of missing data for other variables is reported in Table 1. Missing variables were imputed using the multiple imputation method. Missing values were replaced by the values sampled from the predicted joint distribution multiple times on the basis of the observed data (20). Our regression analysis was on the basis of ten copies of imputed data to account for uncertainty in predicting the missing values. We performed a complete patient analysis as a sensitivity analysis. We then set to evaluate whether young donor age could attenuate the adverse association of size mismatches of different magnitude on graft survival and, if so, whether this differed by recipient age category. We computed model-based estimates of 5- and 10-year death-censored graft survival probabilities (and 95% confidence intervals) stratified according to these three variables.
Table 1. -
Recipient, donor, and procedure characteristics (
n=136,321)
| Recipient/Donor/Procedure Characteristics
a
|
Donor-Recipient Body Surface Area Ratio ≥1.00, n=66,027 |
Donor-Recipient Body Surface Area Ratio 0.90–0.99, n=28,327 |
Donor-Recipient Body Surface Area Ratio 0.80–0.89, n=23,204 |
Donor-Recipient Body Surface Area Ratio <0.80, n=18,763 |
| Mean age at transplant, yr (SD) |
53 (14) |
54 (13) |
54 (12) |
52 (12) |
| Men, n (%) |
32,276 (49) |
18,976 (67) |
17,166 (74) |
14,242 (76) |
| Black race, n (%) |
20,081 (30) |
9976 (35) |
8601 (37) |
7206 (38) |
|
Cause of kidney failure, n (%)
|
|
|
|
|
| Glomerular diseases |
15,882 (24) |
5973 (21) |
4774 (21) |
4007 (21) |
| Diabetes |
17,680 (27) |
9069 (32) |
7734 (33) |
6023 (32) |
| Other or unknown |
32,421 (49) |
13,277 (47) |
10,668 (46) |
8724 (47) |
| Median time on dialysis pretransplant, mo (interquartile range) |
42 (20–68) |
41 (20–66) |
40 (20–64) |
40 (20–64) |
| Peak pretransplant panel reactive antibodies >0%, n (%) |
20,861 (32) |
7728 (27) |
6069 (26) |
4806 (26) |
|
HLA mismatches, n (%)
|
|
|
|
|
| 0–1 |
6465 (10) |
2806 (10) |
2339 (10) |
1988 (11) |
| 2–4 |
28,931 (44) |
12,347 (44) |
9980 (43) |
8021 (43) |
| 5–6 |
30,616 (46) |
13,172 (47) |
10,882 (47) |
8750 (47) |
|
Transplant date, n (%)
|
|
|
|
|
| 2000–2005 |
16,452 (25) |
7412 (26) |
6195 (27) |
5150 (27) |
| 2006–2011 |
22,338 (34) |
9744 (34) |
7834 (34) |
6469 (34) |
| 2012–2015 |
27,237 (41) |
11,171 (39) |
9175 (40) |
71,441 (38) |
| Mean cold ischemic time, h (SD) |
18 (9) |
18 (9) |
18 (9) |
18 (9) |
| Public insurance, n (%) |
49,577 (75) |
20,997 (74) |
16,958 (73) |
13,746 (73) |
| Mean donor age, yr (SD) |
42 (14) |
41 (15) |
40 (16) |
29 (20) |
| Man donor, n (%) |
46,258 (70) |
16,604 (59) |
11,320 (49) |
7488 (40) |
| Black race, n (%) |
9031 (14) |
3748 (13) |
2943 (13) |
2587 (14) |
| Stroke as cause of death, n (%) |
23,554 (36) |
10,573 (37) |
8807 (38) |
6154 (33) |
| Donation after circulatory death, n (%) |
9639 (15) |
3839 (14) |
2978 (13) |
2546 (14) |
| Terminal serum creatinine >1.5 mg/dl |
13,929 (21) |
4685 (17) |
3177 (14) |
1865 (10) |
| Donor hypertension, n (%) |
21,520 (33) |
7879 (28) |
5586 (24) |
3195 (17) |
| Donor diabetes, n (%) |
5856 (9) |
1792 (6) |
1177 (5) |
746 (4) |
| Donor positive hepatitis C status, n (%) |
1559 (2) |
935 (3) |
814 (4) |
467 (2) |
aMissing data are cause of kidney failure (n missing =69), time on dialysis before transplantation (n missing =3072), pretransplant panel reactive antibodies (n missing =6809), cold ischemic time (n missing =6198), insurance status (n missing =16), stroke as cause of death (n missing =1), donation after circulatory death (n missing =13), terminal creatinine (n missing =48), donor diabetes (n missing =687), donor hypertension (n missing =965), and HLA mismatches (n missing =24). Most recent pretransplant panel reactive antibodies originated from the Scientific Registry of Transplant Recipients (SRTR) histocompatibility database charts until 2015. Since 2015, the most recent calculated panel reactive antibodies originates from the medical history patient charts in the SRTR.
Results
Among the 292,526 patients who received a kidney transplant during the study period in the United States, 136,321 kidney transplant recipients were included in this study (Figure 1). Of the latter, 23,614 (17%) experienced death-censored graft loss over a median follow-up of 4.3 years (interquartile range, 1.9–7.8 years). The distribution of BSA ratio is provided in Supplemental Figure 1. Baseline recipient and donor characteristics are provided in Table 1, stratified by donor-recipient BSA ratio. Recipients included in the donor-recipient BSA ratio <0.8 were more likely to be men and of Black race and less likely to have a pretransplant calculated panel reactive antibodies >0%. They tended to receive transplants from younger donors who were more likely to be women and less likely to have comorbidities, such as hypertension, diabetes, elevated terminal serum creatinine, and stroke, as a cause of death.
;)
Figure 1.: Flow chart of patients.
Interaction between Donor and Recipient Age on the Relationship between Donor-Recipient Size Mismatch and Death-Censored Graft Survival
The three-way donor-recipient BSA ratio by donor age by recipient age interaction term was statistically significant (P=0.04). Supplemental Table 1 provides the multivariable hazard ratios (HRs) for donor-recipient BSA ratio on death-censored graft survival stratified by donor age in each recipient age category. There were small numbers of observations included in the categories of recipients aged 31–40 years who received a donor aged >60 years and recipients aged <30 years who received a donor aged 50–60 years. Hence, these categories have large confidence intervals and are unstable estimates.
Figure 2 provides a visual summary of the HRs for BSA ratio by donor-recipient age categories. Across donor-recipient age categories, we observed that the higher the size mismatch (or the smaller the donor-recipient BSA ratio), the greater the negative association with death-censored graft survival. The effect size of severe size mismatch (donor-recipient size ratio <0.80 versus ≥1.00) on death-censored graft survival was generally stronger with donor and recipient age. We observed no conclusive trends for the variation of modest and moderate donor-recipient size mismatches with either donor or recipient age. Univariable and multivariable HRs can be found in Supplemental Table 2. The results of the sensitivity analysis using complete cases can be found in Supplemental Figure 2 and Supplemental Table 3. Effect sizes using multiple imputation and complete patient analyses yielded similar conclusions.
Figure 2.: The effect of donor-recipient body surface area (BSA) ratio on death-censored graft failure according to donor and recipient age with multiple imputation ( n =136,321).
Absolute Differences in Long-Term Death-Censored Graft Survival According to Donor Age, Recipient Age, and Donor-Recipient Size Mismatch
To have a better grasp of the absolute differences in long-term death-censored graft survival rates associated with donor-recipient size mismatch across donor and recipient age categories, we used our multivariable model to estimate adjusted percent survival rates at 5 and 10 years post-transplant (Table 2 for 5-year death-censored graft survival and Table 3 for 10-year death-censored graft survival). In the youngest donor and recipient age categories, estimated graft survival differed by approximately 1%–2% at 5 years and 1%–3% at 10 years when the donor-recipient BSA ratio was <0.8 versus when it was ≥1. These differences were more important in older donor and recipient age categories, with differences in estimated graft survival of 4%–5% at 5 years and 8%–9% at 10 years post-transplant.
Table 2. -
Estimated 5-year death-censored graft survival by donor-recipient body surface area ratio (
n=136,321)
| Donor Age Category/Donor-Recipient Body Surface Area Ratio |
Estimated Probability for 5-yr Graft Survival from Multivariable Model
a
(95% Confidence Interval) |
| Recipients 18–30 yr, n=8654 |
Recipients 31–40 yr, n=16,418 |
Recipients 41–50 yr, n=28,436 |
Recipients 51–60 yr, n=39,142 |
Recipients >60 yr, n=43,671 |
|
Donor age <40 yr
|
|
|
|
|
|
| BSA ratio, n
|
6249 |
10,390 |
15,004 |
17,029 |
15,073 |
|
≥1.00 (reference)
|
83% (82 to 85) |
90% (89 to 91) |
92% (92 to 93) |
94% (93 to 94) |
95% (94 to 95) |
|
0.90–0.99
|
85% (83 to 86) |
90% (89 to 91) |
92% (91 to 92) |
93% (93 to 94) |
95% (94 to 95) |
|
0.80–0.89
|
85% (83 to 87) |
89% (88 to 90) |
92% (91 to 92) |
94% (93 to 94) |
94% (93 to 95) |
|
<0.80
|
83% (81 to 85) |
89% (86 to 90) |
91% (91 to 92) |
93% (92 to 94) |
94% (93 to 94) |
|
Donor age 40–49 yr
|
|
|
|
|
|
| BSA ratio, n
|
1535 |
3633 |
7059 |
9049 |
8814 |
|
≥1.00 (reference)
|
85% (84 to 87) |
88% (87 to 89) |
91% (90 to 92) |
92% (91 to 93) |
92% (91 to 93) |
|
0.90–0.99
|
82% (78 to 85) |
87% (85 to 89) |
91% (90 to 92) |
92% (91 to 92) |
93% (91 to 96) |
|
0.80–0.89
|
87% (83 to 89) |
86% (84 to 88) |
91% (89 to 92) |
91% (90 to 92) |
91% (90 to 92) |
|
<0.80
|
83% (78 to 87) |
88% (85 to 90) |
88% (86 to 90) |
90% (89 to 91) |
90% (88 to 92) |
|
Donor age 50–59 yr
|
|
|
|
|
|
| BSA ratio, n
|
779 |
2108 |
5283 |
9503 |
11,788 |
|
≥1.00 (reference)
|
82% (80 to 85) |
86% (85 to 88) |
89% (88 to 90) |
90% (89 to 91) |
90% (90 to 91) |
|
0.90–0.99
|
79% (74 to 83) |
86% (84 to 88) |
87% (86 to 89) |
90% (89 to 91) |
90% (89 to 91) |
|
0.80–0.89
|
80% (74 to 84) |
88% (85 to 90) |
88% (87 to 90) |
89% (88 to 90) |
90% (88 to 91) |
|
<0.80
|
71% (62 to 78) |
84% (81 to 87) |
88% (86 to 90) |
89% (88 to 91) |
88% (86 to 89) |
|
Donor age ≥60 yr
|
|
|
|
|
|
| BSA ratio, n
|
91 |
287 |
1090 |
3561 |
7996 |
|
≥1.00 (reference)
|
76% (67 to 83) |
83% (79 to 87) |
85% (82 to 87) |
88% (87 to 89) |
88% (87 to 89) |
|
0.90–0.99
|
73% (54 to 85) |
81% (72 to 87) |
85% (82 to 87) |
86% (84 to 87) |
87% (86 to 89) |
|
0.80–0.89
|
69% (50 to 82) |
82% (74 to 87) |
88% (85 to 91) |
86% (84 to 88) |
87% (85 to 88) |
|
<0.80
|
71% (44 to 87) |
76% (63 to 85) |
85% (80 to 89) |
84% (82 to 87) |
83% (80 to 85) |
BSA, body surface area.
aThe model calculates probability for a White man recipient of a White man donor with pretransplant calculated panel reactive antibodies of 0%, median HLA mismatch, median cold ischemic time, median time on dialysis, median recipient body mass index, and public insurance coverage in the 2012–2018 era.
Table 3. -
Estimated 10-year death-censored graft survival by donor-recipient body surface area ratio (
n=136,321)
| Donor Age Category/Donor-Recipient Body Surface Area Ratio |
Estimated Probability for 10-yr Graft Survival from Multivariable Model
a
(95% Confidence Interval) |
| Recipients 18–30 yr, n=8654 |
Recipients 31–40 yr, n=16,418 |
Recipients 41–50 yr, n=28,436 |
Recipients 51–60 yr, n=39,142 |
Recipients >60 yr, n=43,671 |
|
Donor age <40 yr
|
|
|
|
|
|
| BSA ratio, n
|
6249 |
10,390 |
15,004 |
17,029 |
15,073 |
|
≥1.00 (reference)
|
70% (68 to 72) |
81% (79 to 82) |
85% (84 to 86) |
88% (87 to 89) |
90% (89 to 91) |
|
0.90–0.99
|
72% (69 to 75) |
80% (78 to 82) |
84% (83 to 85) |
87% (86 to 89) |
90% (88 to 91) |
|
0.80–0.89
|
73% (70 to 76) |
79% (77 to 81) |
84% (83 to 86) |
88% (87 to 89) |
89% (87 to 90) |
|
<0.80
|
69% (65 to 72) |
80% (78 to 82) |
84% (82 to 85) |
86% (85 to 88) |
88% (86 to 89) |
|
Donor age 40–49 yr
|
|
|
|
|
|
| BSA ratio, n
|
1535 |
3633 |
7059 |
9049 |
8814 |
|
≥1.00 (reference)
|
73% (70 to 76) |
78% (77 to 80) |
83% (81 to 84) |
85% (84 to 86) |
85% (84 to 86) |
|
0.90–0.99
|
67% (61 to 72) |
76% (73 to 79) |
83% (81 to 85) |
84% (82 to 86) |
86% (84 to 87) |
|
0.80–0.89
|
75% (70 to 80) |
74% (70 to 77) |
82% (80 to 84) |
83% (81 to 85) |
83% (81 to 85) |
|
<0.80
|
70% (62 to 77) |
77% (73 to 81) |
78% (75 to 81) |
81% (79 to 83) |
82% (78 to 84) |
|
Donor age 50–59 yr
|
|
|
|
|
|
| BSA ratio, n
|
779 |
2108 |
5283 |
9503 |
11,788 |
|
≥1.00 (reference)
|
68% (64 to 72) |
75% (72 to 77) |
79% (78 to 81) |
81% (80 to 82) |
82% (81 to 83) |
|
0.90–0.99
|
63% (55 to 70) |
75% (70 to 78) |
76% (74 to 79) |
81% (80 to 83) |
81% (79 to 83) |
|
0.80–0.89
|
64% (55 to 71) |
77% (73 to 80) |
78% (76 to 81) |
80% (77 to 82) |
80% (78 to 82) |
|
<0.80
|
50% (38 to 61) |
71% (65 to 76) |
78% (74 to 81) |
80% (77 to 83) |
77% (74 to 80) |
|
Donor age ≥60 yr
|
|
|
|
|
|
| BSA ratio, n
|
91 |
287 |
1090 |
3561 |
7996 |
|
≥1.00 (reference)
|
58% (46 to 69) |
70% (63 to 76) |
72% (68 to 75) |
77% (75 to 80) |
78% (76 to 79) |
|
0.90–0.99
|
54% (29 to 73) |
65% (52 to 75) |
72% (67 to 77) |
73% (70 to 76) |
77% (74 to 79) |
|
0.80–0.89
|
48% (26 to 67) |
67% (55 to 76) |
78% (72 to 83) |
74% (71 to 77) |
75% (72 to 78) |
|
<0.80
|
51% (20 to 75) |
58% (40 to 72) |
73% (65 to 79) |
71% (67 to 75) |
69% (65 to 73) |
BSA, body surface area.
aThe model calculates probability for a White man recipient of a White man donor with pretransplant calculated panel reactive antibodies of 0%, with median HLA mismatch, median cold ischemic time, median time on dialysis, median recipient body mass index, and public insurance coverage in the 2012–2018 era.
We also asked whether, from a clinical perspective, young donor age could override the association between size mismatch and graft survival. We found that in recipients aged >30 years at the time of transplant, a size-mismatched kidney (donor-recipient BSA ratio <0.80) originating from a young donor (<40 years) is associated with survival probabilities that are equal to or better than those of size-matched kidneys (donor-recipient BSA ratio ≥1.00) originating from donors aged ≥40 years up to 5 and 10 years post-transplant (Figures 3 and 4). For recipients aged 18–30 years at the time of transplant, a size-mismatched kidney (donor-recipient BSA ratio <0.80) originating from a young (<40 years) donor is associated with a small decrease in estimated graft survival 5 and 10 years post-transplant compared with size-matched kidneys (donor-recipient BSA ratio ≥1.00) from donors aged 40 to 49 years, but not when the donor is older. Estimations from complete patient analyses are found in Supplemental Tables 4 and 5.
Figure 3.: Estimated 5-year death-censored graft survival by donor-recipient BSA ratio according to donor and recipient age: multiple imputation method.
Figure 4.: Estimated 10-year death-censored graft survival by donor-recipient BSA ratio according to donor and recipient age: multiple imputation method.
Discussion
We undertook this study to test the hypothesis that the association of donor-recipient size mismatch on graft survival would be modified by donor and recipient age. Here, we demonstrate that these variables indeed interact but that, overall, the association between donor age and kidney graft survival supersedes that of donor-recipient size mismatch.
Data from animal models have consistently demonstrated that reducing kidney mass causes progressive kidney damage in the remnant kidney (21). The damage is believed to be caused by hyperfiltration. The same concept can be applied to kidney recipients receiving a proportionally smaller organ: low donor kidney volume or weight has been linked to proteinuria, greater use of antihypertensive drugs, lower graft function, and higher risk of graft failure (5–678,22–2324). At the time of decision making to accept an organ from a deceased donor, the weight or volume of the kidney is generally unknown. However, weights and heights of donors and candidates are almost always available. BSA, which correlates with kidney weight (25), glomerular volume (26), and GFR (27), can thus be calculated. The relationship between donor-recipient size mismatch and graft survival is supported by most but not all studies (28–2930). The fact that not all studies concur may be due to low power (28,30) or to differences in age structures of the cohort studied if donor and/or recipient ages interact with size mismatch.
To our knowledge, we are the first to assess how size mismatch interacts with both donor and recipient ages. Given that glomerular sclerosis is greater with older age (12), we had hypothesized that (1) the magnitude of the association of size mismatch on graft survival would be larger when the donor was older, as the nephron mass transplanted would be lower, and (2) this association would be more pronounced in younger recipients as they have greater life expectancy, allowing sufficient time for hyperfiltration associated with size mismatch to become clinically manifest. Our results do show that the effect size of severe size mismatch is greater with older donor age. This is consistent with the results of a previous study that had shown that among 254 living kidney donor recipients, lower donor-recipient BSA ratio was associated with worse graft function 1 year post-transplantation when the graft came from an older (≥45 years) compared with a younger donor (31). In contrast with our a priori hypothesis, the effect size associated with donor-recipient size mismatch was not most apparent in young recipients. On the contrary, the effect size associated with moderate and severe donor-recipient size mismatch was greater with recipient age. We hypothesize that this may be explained by a higher prevalence of risk factors for nonalloimmune chronic allograft injury (hypertension, diabetes) in older recipients, which may have a synergistic influence with size mismatch in this population. In a recent study by Miller et al. (24), a larger effect size for donor-recipient size mismatch was observed in sex-mismatched donor-recipient pairs, especially in women recipients of men donors. This was attributed to a potential synergy between size mismatch and the immunologic H-Y effect. As we were interested in the interaction between donor and recipient age and size mismatch, we did not search for effect modification by sex, but we considered donor and recipient sex as confounders and included them as covariates in our analyses. Nevertheless, both our results and those of Miller et al. (24) suggest that the association between donor-recipient size mismatch can be modified by other factors, such as donor and recipient sex or age.
For clinical purposes, when the time comes to accept and discuss an offer for a deceased donor, transplant physicians and candidates must weigh the relative effect of various donor and recipient characteristics to make the best decision to accept or refuse the offer in the hope of receiving a better one. Decision aids have recently been made available to make that decision (32,33), but the potential association of donor-recipient size mismatch on graft survival was not considered. Hence, our work provides complementary information to that already available for transplant physicians and candidates. Although examining the magnitude of the association between size mismatch and graft survival in different donor and recipient age categories is interesting from a physiologic perspective, the absolute increase in the risk of graft loss is a more intuitive and clinically relevant measure to grasp the effect of various risk factors. Differences of up to 10% in estimated graft survival 10 year post-transplant were noted for the older donor and recipient age categories. We also asked whether, from a clinical perspective, young donor age could override the association between size mismatch and graft survival. We found that compared with the association of donor age with estimated long-term graft survival, the adverse effect of donor-recipient size mismatch was only modest. Accepting a size-mismatched kidney from a donor aged <40 years yielded better or equal estimated survival probabilities than a size-matched kidney from donors aged ≥40 years across all recipient age categories, except in the youngest of recipients.
Although our study has a large sample size that allowed us to test three-way interactions between donor and recipient age and size mismatch, certain limitations must be acknowledged. First, our results in young recipients who receive older donors are not interpretable, given the small number of observations in these donor-recipient age categories (recipients aged 18–30 years receiving donors aged ≥50 years or recipients aged 31–40 years receiving donors aged ≥60 years) and associated wide confidence intervals. The likelihood of a selection bias in these donor-recipient age categories is also very high, as most transplant physicians tend not to allocate organs from older donors to young recipients. Second, we were unable to study interactions between donor and recipient age with extreme forms of size disparities, as only 15% of patients were included in the lowest BSA ratio cutoff of 0.8. As in any registry study, some variables of interest have not been included. Hence, we are unable to verify some hypotheses generated by our results: for instance, whether recipient hypertension and diabetes in older recipients accentuate the adverse effect of size mismatch. Another limitation stems from uncertainties in donor weight and height values provided in the SRTR. Although both can be either estimated or directly measured, donor weight is sensitive to the fluid balance, and its measurement may be imprecise in the context of organ donation. These imprecisions are most likely to create a bias toward the null and underestimate the true effect of donor-recipient size mismatch. The use of donor-recipient BSA ratio rather than donor-recipient weight ratio adds the need for an extra calculation, but its use yielded better model fit than that of the weight ratio. Moreover, multiple free online calculators for BSA are now available. We chose to use donor-recipient BSA ratio as a reflection of the discrepancy between donor and recipient size. Although modeling ratios can lead to spurious correlations between indices that have a common component (34), this is unlikely to be the case in our work as donor and recipient BSAs do not have a common component, because we did not model the ratio per se, but rather used it to classify the severity of size mismatch, and because the dependent variable (death-censored graft survival) does not share a common component with BSA. Last, our findings are on the basis of American data that may not be universally generalizable.
In conclusion, we found that donor and recipient ages modulate the association between size mismatch and graft survival. However, the association between donor-recipient size mismatch and long-term graft survival was modest compared with the association between donor age and graft survival. In all age recipient categories except in those aged 18–30 years, young donor age superseded the negative association between donor-recipient size mismatch and graft survival. A size-mismatched young donor led to equal or better estimated long-term graft survival rates than size-matched kidneys from donors aged over 40 years.
Disclosures
R. Sapir-Pichhadze and H. Cardinal report receiving salary support from Fonds de Recherche du Québec Santé (FRQS) during the conduct of the study. All remaining authors have nothing to disclose.
Funding
This work was made possible through start-up funds allowed to H. Cardinal by FRQS and a legacy donation from Mrs. Chantal Caron, an inspiring kidney-transplant recipient we have had the privilege to care for. These funds were used to cover the cost of statistical analyses.
Acknowledgments
The data reported here have been supplied by the Hennepin Healthcare Research Institute as the contractor for the 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.
Supplemental Material
This article contains the following supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.02310220/-/DCSupplemental.
Supplemental Appendix. Rationale for the inclusion of confounders in the multivariable model.
Supplemental Figure 1. The distribution of donor-recipient body surface area ratio.
Supplemental Figure 2. The effect of donor-recipient body surface area ratio on death-censored graft failure according to donor and recipient age: complete patient analysis (119,596 observations used of 136,321).
Supplemental Table 1. The association between donor-recipient body surface area ratio and death-censored graft failure (n=136,321).
Supplemental Table 2. Effect of recipient, donor, and procedure characteristics on death-censored graft survival (n=136,321).
Supplemental Table 3. The effect of donor-recipient body surface area ratio on death-censored graft failure (complete patient analysis).
Supplemental Table 4. Estimated 5-year death-censored graft survival by donor-recipient body surface area ratio (complete patient analysis; n=119,596).
Supplemental Table 5. Estimated 10-year death-censored graft survival donor-recipient body surface area ratio (complete patient analysis; n=119,596).
References
1. Laupacis A, Keown P, Pus N, Krueger H, Ferguson B, Wong C, Muirhead N: A study of the quality of life and cost-utility of renal transplantation. Kidney Int 50: 235–242, 19968807593
2. Wolfe RA, Ashby VB, Milford EL, Ojo AO, Ettenger RE, Agodoa LY, Held PJ, Port FK: Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 341: 1725–1730, 199910580071
3. United Network for Organ Sharing: Transplant Trends, 2017. Available at:
https://unos.org/data/transplant-trends/#transplants_by_donor_type+organ+Kidney. Accessed February 20, 2018
4. Israni AK, Salkowski N, Gustafson S, Snyder JJ, Friedewald JJ, Formica RN, Wang X, Shteyn E, Cherikh W, Stewart D, Samana CJ, Chung A, Hart A, Kasiske BL: New national allocation policy for deceased donor kidneys in the United States and possible effect on patient outcomes. J Am Soc Nephrol 25: 1842–1848, 201424833128
5. Kasiske BL, Snyder JJ, Gilbertson D: Inadequate donor size in cadaver kidney transplantation. J Am Soc Nephrol 13: 2152–2159, 200212138149
6. Han SS, Yang SH, Oh YJ, Cho JY, Moon KC, Ha J, Kim YS: Graft volume as the surrogate marker for nephron number affects the outcomes of living-donor kidney transplantation. Clin Transplant 25: E327–E335, 201121395690
7. Giral M, Foucher Y, Karam G, Labrune Y, Kessler M, Hurault de Ligny B, Büchler M, Bayle F, Meyer C, Trehet N, Daguin P, Renaudin K, Moreau A, Soulillou JP: Kidney and recipient weight incompatibility reduces long-term graft survival. J Am Soc Nephrol 21: 1022–1029, 201020488949
8. Kim YS, Moon JI, Kim DK, Kim SI, Park K: Ratio of donor kidney weight to recipient bodyweight as an index of graft function. Lancet 357: 1180–1181, 200111323049
9. Nawaratne S, Brien JE, Seeman E, Fabiny R, Zalcberg J, Cosolo W, Angus P, Morgan DJ: Relationships among liver and kidney volumes, lean body mass and drug clearance. Br J Clin Pharmacol 46: 447–452, 19989833597
10. Park CW, Yu N, Yun SW, Chae SA, Lee NM, Yi DY, Choi YB, Lim IS: Measurement and estimation of renal size by computed tomography in Korean children. J Korean Med Sci 32: 448–456, 201728145648
11. Emamian SA, Nielsen MB, Pedersen JF, Ytte L: Kidney dimensions at sonography: Correlation with age, sex, and habitus in 665 adult volunteers. AJR Am J Roentgenol 160: 83–86, 19938416654
12. Glassock RJ, Rule AD: The implications of anatomical and functional changes of the aging kidney: With an emphasis on the glomeruli. Kidney Int 82: 270–277, 201222437416
13. Oniscu GC, Brown H, Forsythe JL: How old is old for transplantation? Am J Transplant 4: 2067–2074, 200415575911
14. Vandenbroucke JP, von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, Egger M; STROBE Initiative: Strengthening the reporting of observational studies in epidemiology (STROBE): Explanation and elaboration. PLoS Med 4: e297, 200717941715
15. Mosteller RD: Simplified calculation of body-surface area. N Engl J Med 317: 1098, 19873657876
16. Lam TK, Leung DT: More on simplified calculation of body-surface area. N Engl J Med 318: 1130, 19883352717
17. Roseman DA, Hwang SJ, Oyama-Manabe N, Chuang ML, O’Donnell CJ, Manning WJ, Fox CS: Clinical associations of total kidney volume: The Framingham Heart Study. Nephrol Dial Transplant 32: 1344–1350, 201727325252
18. El Edelbi R, Lindemalm S, Eksborg S: Estimation of body surface area in various childhood ages—Validation of the Mosteller formula. Acta Paediatr 101: 540–544, 201222211780
19. Metzger RA, Delmonico FL, Feng S, Port FK, Wynn JJ, Merion RM: Expanded criteria donors for kidney transplantation. Am J Transplant 3: 114–125, 200312694055
20. Rubin DB: Inference and missing data. Biometrika 63: 581–592, 1976
21. Hostetter TH, Olson JL, Rennke HG, Venkatachalam MA, Brenner BM: Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. Am J Physiol 241: F85–F93, 19817246778
22. Dick AA, Mercer LD, Smith JM, McDonald RA, Young B, Healey PJ: Donor and recipient size mismatch in adolescents undergoing living-donor renal transplantation affect long-term graft survival. Transplantation 96: 555–559, 201323838999
23. Codas R, Danjou F, Dagot C, Martin X, Morelon E, Badet L, Hanf W, Petruzzo P: Influence of allograft weight to recipient bodyweight ratio on outcome of cadaveric renal transplantation. Nephrology (Carlton) 19: 420–425, 201424802649
24. Miller AJ, Kiberd BA, Alwayn IP, Odutayo A, Tennankore KK: Donor-recipient weight and sex mismatch and the risk of graft loss in renal transplantation. Clin J Am Soc Nephrol 12: 669–676, 201728360198
25. Nyengaard JR, Bendtsen TF: Glomerular number and size in relation to age, kidney weight, and body surface in normal man. Anat Rec 232: 194–201, 19921546799
26. Kasiske BL, Umen AJ: The influence of age, sex, race, and body habitus on kidney weight in humans. Arch Pathol Lab Med 110: 55–60, 19863753571
27. White AJ, Strydom WJ: Normalisation of glomerular filtration rate measurements. Eur J Nucl Med 18: 385–390, 19911879445
28. Gaston RS, Hudson SL, Julian BA, Laskow DA, Deierhoi MH, Sanders CE, Phillips MG, Diethelm AG, Curtis JJ: Impact of donor/recipient size matching on outcomes in renal transplantation. Transplantation 61: 383–388, 19968610346
29. Lee LS, Auersvald LA, Claus EB, Bia MJ, Friedman AL, Lorber MI, Basadonna GP: Body size mismatch between donor and recipient and the development of chronic rejection in renal transplantation. Transplant Proc 29: 111, 19979122918
30. Miles AM, Sumrani N, John S, Markell MS, Distant DA, Maursky V, Hong JH, Friedman EA, Sommer B: The effect of kidney size on cadaveric renal allograft outcome. Transplantation 61: 894–897, 19968623156
31. Lv J, Guga S, Huang H, Wang R, Zhang X, Shou Z, Chen J, Wu J: The effect of donor-recipient body surface area ratio on donor age and donor glomerular filtration rate in Chinese patients undergoing a living-donor kidney transplant. Exp Clin Transplant 12: 515–521, 201425019250
32. Wey A, Salkowski N, Kremers WK, Schaffhausen CR, Kasiske BL, Israni AK, Synder JJ: A kidney offer acceptance decision tool to inform the decision to accept an offer or wait for a better kidney. Am J Transplant 18: 897–906, 2018
33. Bae S, Massie AB, Thomas AG, Bahn G, Luo X, Jackson KR, Ottmann SE, Brennan DC, Desai NM, Coresh J, Segev DL, Garonzik Wang JM: Who can tolerate a marginal kidney? Predicting survival after deceased donor kidney transplant by donor-recipient combination. Am J Transplant 19: 425–433, 201929935051
34. Kronmal R: Spurious correlation and the fallacy of the ratio standard revisited. J R Stat Soc Ser A Stat Soc 56: 379–392, 1992
