Abstract
ABSTRACT. There have been conflicting reports that kidneys from small donors may be at increased risk for late graft failure if they are transplanted into large recipients. Data from the United States Renal Data System was used to study all first cadaver kidney transplantations performed during the years 1994 to 1999. Donor and recipient body surface area (BSA) combinations were included along with other patient and transplant characteristics in a Poisson analysis of factors associated with early (in the first 4 mo) and late (≥4 mo) graft failure. The numbers of large (BSA >2.2 m2) and medium size (BSA 1.6 to 2.2 m2) recipients that received kidneys from small (BSA <1.6 m2) donors are less than expected (χ2 = 118.09; P < 0.0001), suggesting that transplant centers may be refusing some kidneys on the basis of donor-recipient size differences. Large recipients who received kidneys from small donors made up 1.5% of the population and had a 43% (95% CI, 17 to 75%; P = 0.0004) increased risk of late graft failure compared with medium-size recipients who received kidneys from medium-size donors (53.4% of the population). Medium-size recipients who received kidneys from small donors made up 12.0% of the population and had a 16% (95% CI, 6 to 26%; P = 0.0012) increased risk of late graft failure. Disparities in recipient and donor size had similar adverse affects on mortality. Effects of recipient obesity (body mass index) and donor gender on late graft survival were no longer statistically significant after the effects of donor and recipient body size were taken into account. In conclusion, the relative size of the donor and recipient should possibly be taken into account when choosing kidneys for transplantation.
The theory that compensatory increases in glomerular capillary pressures and flows contribute to chronic progressive kidney damage has played a prominent role in nephrology. We know that reducing kidney mass causes progressive injury in the remaining remnant kidney in animal models (1,2). We also know that compensatory increases in glomerular capillary pressure, flow, and filtration, e.g., “hyperfiltration”, are associated with progressive injury in the rat remnant kidney (3). However, direct evidence to support this hypothesis in humans still does not exist.
In the past decade, several investigators suggested that hyperfiltration injury from “inadequate nephron dosing” may cause progressive injury in transplanted human kidneys. According to this theory, a large patient who receives a small kidney may develop some of the same compensatory changes originally described in the rat 5/6 nephrectomy model (4,5). Although it is not possible to directly measure glomerular hemodynamics in humans, the hypothesis can be indirectly tested if we assume that large individuals have a greater renal functional capacity than smaller individuals. Body size, measured as body surface area (BSA), correlates with glomerular volume (6,7), kidney weight (6,7), and GFR (8,9). Therefore, a reasonable test of the hypothesis is to examine the effects on long-term graft survival of placing a kidney from a small (low BSA) donor into a large (high BSA) recipient.
Studies in individual transplant centers have produced conflicting results. Some investigators have reported reduced graft survival in situations in which the size of the donor kidney would be expected to be reduced in proportion to the size of the recipient (10,11). Other single-center studies found no effect of donor-recipient size mismatching on graft survival (12–17). However, many of these studies may have lacked adequate statistical power to test the hypothesis rigorously. Studies using data collected by the United Network for Organ Sharing (UNOS) have shown that graft survival is generally reduced in recipients with greater body weight, body mass index (BMI), and/or BSA (18,19). Unfortunately, UNOS did not begin collecting data on donor size until 1994, so the effects of donor-recipient size mismatching on long-term outcomes has never been examined. At the 16th International Congress of the Transplantation Society in 1996, Cho et al. (20) reported that size mismatching was associated with decreased 1-yr graft survival. However, the duration of follow-up was short, and this analysis did not adjust for potential confounding effects from donor and recipient age or gender or other factors (20).
We used data from UNOS and the United States Renal Data System (USRDS) to examine the effects of donor-recipient size mismatching on long-term patient and graft survival after kidney transplantation. We particularly investigated whether the effect of size mismatching was simply due to an effect of obesity in the recipient per se or whether it could have been due to effects of size mismatching that were independent of obesity. In addition, we hypothesized that the effects of hyperfiltration injury should be manifest in the late posttransplant period.
Materials and Methods
Study Population
We used data from the USRDS to study all first cadaver kidney transplantations performed during the years of 1994 to 1999 (n = 32,083). We included donors and recipients of all ages. We excluded patients transplanted with organs other than the kidney and patients who were transplanted with more than one kidney.
Patient and Transplant Characteristics
We calculated donor and recipient BSA using the formula of Mosteller (21): BSA (m2) = ([Wt × Ht] ÷ 3600)1/2, where Wt was body weight in kilograms and Ht was height in centimeters. We calculated donor and recipient BMI (kg/m2) = Wt (kg) ÷ Ht(m)2(2). Other study variables included donor and recipient gender, donor and recipient age, recipient race and ethnicity, prior time on dialysis, primary cause of kidney failure, cold ischemia time, panel-reactive antibody status, number of human leukocyte antigen (HLA) mismatches, education (highest level achieved), and work ability.
Statistical Analyses
We examined the effects of donor and recipient BSA combinations on outcomes using interval Poisson analysis. We first examined the effects of BSA and other variables on short-term outcomes, i.e., during the first 4 mo. We next examined the effects on long-term outcomes for patients who had a functioning graft at 4 mo. Each analysis was adjusted for multiple patient and transplant characteristics. The outcomes that we examined in separate analyses were graft failure (return to dialysis, re-transplantation, or death with a functioning graft), death-censored graft failure (return to dialysis or re-transplantation), and death with a functioning graft. All analyses were carried out using the statistical software package SAS version 8.2 (SAS Institute, Inc., Cary, NC). All results were considered statistically significant if P < 0.05.
Results
There were 17,194 (53.4%) medium-size recipients with BSA 1.6 to 2.2 m2 who received kidneys from medium-size donors with BSA 1.6 to 2.2 m2 (Figure 1). However, there were 3864 medium-size recipients (12%) with BSA 1.6 to 2.2 m2 who received kidneys from smaller donors with BSA <1.6 m2, and 486 large recipients (1.5%) with BSA >2.2 m2 who received kidneys from smaller donors with BSA <1.6 m2 (Figure 1). The distribution of recipient and donor sizes is not random, and the pattern suggests that there is more than the expected recipient and donor size matching. For example, only 13.0% of large recipients received kidneys from small donors, whereas 21.3% of small recipients received kidneys from small donors (Figure 2). Some of the effect of donor-recipient size matching may be due to an effect of donor-recipient age matching, because donor size and age correlate, e.g., kidneys from donor <18 yr are generally smaller than kidneys from adults. The significance of the effect of donor-recipient size matching (χ2 = 118.09; P < 0.0001) was reduced somewhat (χ2 = 43.13; P < 0.0001) by taking donor age into account.
Figure 1. The numbers of recipients and donors with similar or different body surface area (BSA). Small, medium, and large recipients correspond to BSA <1.6 m2, BSA 1.6 to 2.2 m2, and BSA >2.2 m2, respectively. S, small donor (BSA <1.6 m2); M, medium-size donor (BSA 1.6 to 2.2 m2); L, large donor, (BSA >2.2 m2).
Figure 2. The percentages of small, medium, and large BSA recipients with small, medium, and large BSA donors. Numbers in the vertical bars are the actual percentages of different donor sizes (total = 100%) for each recipient BSA category. The proportions were significantly different by χ2 square, indicating that donors and recipients have to some extent (greater than chance) been matched according to size.
Donor and recipient BSA had no statistically significant effects on graft failure, death-censored graft failure, or mortality during the first 4 mo after transplantation. (Figures 3 to 5). In contrast, for those surviving at least 4 mo with a functioning graft, the adjusted risk for subsequent graft failure was increased 43% for large recipients of kidneys from small donors (Figure 3, Table 1⇓). For medium-size recipients of kidneys from small donors, the adjusted risk of late graft failure was increased 16% (Figure 3, Table 1). Similarly, for those surviving at least 4 mo with a functioning graft, the relative risk for death-censored graft failure was 1.37 (95% confidence interval [CI], 1.04 to 1.81) for large recipients of kidneys from small donors (Figure 4). For those surviving at least 4 mo with a functioning graft, the adjusted relative risk for death was 1.49 (1.12 to 1.98) for large recipients of kidneys from small donors and 1.20 (1.06 to 1.37) for medium-size recipients of kidneys from small donors (Figure 5).
Figure 3. The effects of donor and recipient BSA combinations on graft failure during the first 4 mo after transplantation (left panel) and limited to patients whose grafts functioned at least 4 mo (right panel). Vertical bars are each adjusted Poisson relative risks compared with the reference group of medium donor and recipient BSA (middle bar). Vertical lines are 95% confidence intervals (CI), where failure to cross 1.00 indicates a statistically significant difference compared with the reference group. Small, medium, and large recipients correspond to BSA <1.6 m2, BSA 1.6 to 2.2 m2, and BSA >2.2 m2, respectively. S, small donor (BSA <1.6 m2); M, medium-size donor (BSA 1.6 to 2.2 m2); L, large donor, (BSA >2.2 m2).
Figure 4. The effects of donor and recipient BSA combinations on death-censored graft failure during the first 4 mo after transplantation (left panel) and limited to patients whose grafts functioned at least 4 mo (right panel). See legend to Figure 3 for a detailed explanation.
Figure 5. The effects of donor and recipient BSA combinations on mortality during the first 4 mo after transplantation (left panel) and limited to patients whose grafts functioned at least 4 mo (right panel). See legend to FIgure 3 for a detailed explanation.
The independent effects of donor and recipient body size on late (>4 mo) graft failurea
(Continued)
Recipient obesity (BMI ≥30 kg/m2) was associated with a slightly higher rate of graft failure after the first 4 mo (relative risk. 1.08 [1.01 to 1.15]; P = 0.0230) adjusted for the same covariates shown in Table 1 but excluding donor and recipient BSA. However, after taking recipient and donor BSA into account, the effect of obesity on graft failure was not statistically significant (Table 1). Donor obesity had no effect on the rate of late graft failure, even when donor and recipient BSA were excluded (adjusted relative risk, 1.01 [0.94 to 1.09]; P = 0.7407). The effect of recipient and donor BSA on graft failure was not affected by including the nonsignificant effects of BMI in the same analysis (Table 1). Indeed, after removing donor and recipient BMI (from the analysis of patients surviving at least 4 mo with a functioning graft), the adjusted relative risks for graft failure were 1.47 (1.21 to 1.78), P < 0.0001, for large recipients of kidneys from small donors, and 1.15 (1.06 to 1.26), P = 0.0015, for medium-size recipients of kidneys from small donors.
Donor gender had an effect on the rate of late graft failure that was not seen when differences in donor and recipient BSA were taken into account. When donor and recipient BSA were not taken into account, the risk of graft failure associated with kidneys from male (compared with female) donors was significantly less, (relative risk, 0.93 [0.88 to 0.98]; P = 0.0042). However, when donor and recipient BSA differences were taken into account, there was no effect of donor gender (Table 1).
The effect of donor age on the rate of late graft failure was virtually identical whether or not donor and recipient BSA differences were taken into account. Without accounting for donor and recipient BSA differences, the adjusted relative risks of donor age (reference, 18 to 44 yr) were 1.00 (0.92 to 1.08), P = 0.9336, for age <18; 1.36 (1.27 to 1.44), P < .0001, for age 45 to 59; and 1.79 (1.64 to 1.95), P < .0001, for age ≥60 yr. These donor age effects were not different from those from a model that also took donor and recipient BSA differences into account (Table 1).
Discussion
The results of this study suggest that a donor kidney can be too small for a large recipient and that donor-recipient size disparity is a major risk factor for late allograft failure. However, as in any retrospective analysis, a statistical association does not prove causality, and the mechanism(s) that may explain the association cannot be determined from these results. Interestingly, a large donor-recipient size mismatch was a stronger risk factor for death with function than for death-censored graft failure. However, this is not necessarily inconsistent with a role for glomerular hemodynamic changes (hyperfiltration), proteinuria, or other factors in causing the increased risk for graft failure associated with a small kidney. Indeed, acute rejection, treatment with additional immunosuppression, and other correlates of poor graft function could explain why transplanting a kidney that is too small for a large recipient increases mortality.
A large amount of circumstantial evidence has been used to support the theory that hyperfiltration from inadequate nephron mass contributes to late graft failure (4,5,19). However, one of the best tests of the hyperfiltration hypothesis is to determine whether placing a small kidney in a large recipient is associated with an increased risk of late graft failure. This is because normal kidney size and functional capacity both correlate with BSA (6–9).
Obesity also correlates with BSA, and obese individuals have larger kidneys and proportionally greater GFR (22–24). However, graft survival may be compromised in obese patients for reasons unrelated to hyperfiltration in a small donor kidney. For example, obese patients are more prone to wound infections, and this could lead to a higher incidence of bacterial sepsis in the setting of immunosuppression (25–28). Cardiovascular disease complications may also be more frequent in obese transplant recipients (29). Thus, it is important to distinguish effects of obesity, probably best measured as BMI, from effects of donor-recipient size disparities.
The results of the present analysis suggest that much of the adverse effects of obesity can be explained by donor-recipient size mismatches. Indeed, the effects of BMI on late graft failure were no longer statistically significant when the effects of donor and recipient size differences were taken into account. In other words, obese individuals who receive a kidney that is large enough may not be at increased risk of late graft failure. Similarly, the effects of donor gender on the rate of late graft failure were no longer significant after the effects of donor and recipient size differences were taken into account. This suggests that a major part of the donor gender effect (30–32) may be attributable to donor-recipient size mismatching.
There are a number of potential reasons why previous studies investigating the effects of donor-recipient size mismatching have produced conflicting results. We conducted a MEDLINE search for studies that have investigated the effects of donor-recipient size mismatching on kidney graft survival (Table 2). We identified 11 published studies, but 6 had not yet been published in a peer-reviewed journal, and many details of the analysis in these 6 studies were missing (10,13,16,20,33,34). There were 8 single-center studies, (10–17) and it is likely that many of these single-center studies lacked adequate statistical power to test the hypothesis. Indeed, 4 studies included fewer than 200 patients (10,14,15,17). Many of the studies failed to use a multivariate analysis to eliminate possible confounding from the effects of other variables (Table 2). In fact, none of the 4 studies reporting a statistically significant effect of donor-recipient size disparity on graft survival performed a multivariate analysis (10,11,13,20). In addition, 7 studies used recipient body weight rather than BSA as a measure of recipient size (10,13,15–17,33,34). Most studies failed to distinguish the effects of donor-recipient size mismatches on early versus late graft failure. Thus, differences in the results of studies indirectly testing the hypothesis that donor-recipient size mismatching contributes to late graft failure can be explained by differences in sample size and study design.
Studies of the effect of donor and recipient size differences on graft survivala
We can only speculate what mechanisms may explain the association we observed between donor-recipient size mismatches and graft failure. As originally hypothesized, compensatory changes in glomerular capillary pressures and flows could lead directly or indirectly to progressive kidney damage. For example, compensatory hemodynamic changes may lead to proteinuria, which could be injurious. Indeed, some small, uncontrolled, observational studies suggest that donor-recipient size mismatches may be associated with a higher incidence of proteinuria (35,36). Compensatory hemodynamic changes could also be a pro-inflammatory stimulus leading to alloantigen-dependent kidney damage. Alternatively, it is not difficult to imagine that donor-recipient size disparities and the resulting higher levels of serum creatinine in the recipient may prompt physicians to either increase (to prevent rejection) or decrease immunosuppressive medications (to treat calcineurin inhibitor toxicity). Thus, there are many plausible mechanisms for the effects of donor-recipient size disparities on graft survival.
The results of this study suggest that centers may be turning down offers of kidneys because the donor is small in comparison to the potential recipient (Figure 2). Our results suggest that this may be reasonable because the magnitude of the possible effect of size mismatches on graft survival exceed, or are at least comparable to, those of HLA matching and other important risk factors for graft survival (Table 1). Additional studies with longer follow-up are clearly needed to confirm these results. Additional studies are also needed to better define the threshold, and circumstances associated with the adverse effects of donor-recipient size disparities. The results of such studies may help determine the extent to which donor and recipient size should be taken into account in allocating cadaveric kidneys for transplantation. Similar studies are also needed for living donors and recipients.
Acknowledgments
The data reported here have been provided by the United States Renal Data System, Minneapolis, Minnesota. 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 the US Government. Portions of this work were presented at the American Society of Nephrology and International Congress of Nephrology’s joint World Congress of Nephrology, October 13 to 17, 2001, in San Francisco, CA. The authors wish to thank Katie Durhham, Christopher Forsberg, Benjamin French, Grant Riewe, and Victoria Shultz from St. Olaf College in Northfield, MN, under the tutelage of Dr. Ted Vessey and Dr. Michael Kahn, for their contributions to this analysis.
- © 2002 American Society of Nephrology
References
Jump to section
More in this TOC Section
Cited By...
- Donor Age, Donor-Recipient Size Mismatch, and Kidney Graft Survival
- A Kidney Graft Survival Calculator that Accounts for Mismatches in Age, Sex, HLA, and Body Size
- Donor-Recipient Weight and Sex Mismatch and the Risk of Graft Loss in Renal Transplantation
- Donor Quality in the Eye of the Beholder: Interactions between Nonimmunologic Recipient and Donor Factors as Determinants of Graft Survival
- Kidney and Recipient Weight Incompatibility Reduces Long-Term Graft Survival
- The Clinical Importance of Nephron Mass
- Adult Hypertension and Kidney Disease: The Role of Fetal Programming
- Impact of Graft Mass on the Clinical Outcome of Kidney Transplants