Donor Kidney Volume and Outcomes Following Live Donor Kidney Transplantation

Authors


Abstract

Pre-donation kidney volume and function may be crucial factors in determining graft outcomes in kidney transplant recipients. We measured living donor kidney volumes by 3D helical computed tomography scanning and glomerular filtration rate (GFR) by 125I-iothalamate clearances in 119 donors, and correlated these values with graft function and incidence of acute rejection at 2 years post-transplantation. Kidney volume strongly correlated with GFR (Pearson r= 0.71, p < 0.001). Body size and male gender were independent correlates of larger kidney volumes, and body size and age were predictors of kidney function. The effects of transplanted kidney volume on graft outcome were studied in 104 donor-recipient pairs. A transplanted kidney volume greater than 120 cc/1.73 m2 was independently associated with better estimated GFR at 2 years post-transplant when compared to recipients of lower transplanted kidney volumes (64 ± 19 vs. 48 ± 14 mL/min/1.73 m2, p < 0.001). Moreover, recipients of lower volumes had a higher incidence of acute cellular rejection (16% vs. 3.7%, p = 0.046). In conclusion, kidney volume strongly correlates with function in living kidney donors and is an independent determinant of post-transplant graft outcome. The findings suggest that (1) transplantation of larger kidneys confers an outcome advantage and (2) larger kidneys should be preferred when selecting from otherwise similar living donors.

Introduction

Significant improvements in short-term renal allograft outcome have been achieved over the past decade primarily because of more potent immunosuppressive regimens and the increasing use of living donors (1). Nevertheless, long-term graft survival remains suboptimal. The quality of the donated organ is a well established factor that influences graft fate after deceased donor renal transplantation (2,3). The impact of organ quality in long-term graft function in living kidney transplantation is less well understood. Moreover, because of the increasing demand for organs, there is a growing number of living donor exchange programs where fair trade of organs with similar characteristics is essential for their long-term success. Potential markers of donor organ quality in living kidney donation may include donor kidney volume and function (4–6). It has been hypothesized that the size and pre-transplant glomerular filtration rate (GFR) of a donated kidney influence post-transplant outcome (7–9). Although animal data support this concept (10–12), data in humans are controversial and mainly derive from indirect observations in deceased donor kidney transplantation (4,13–17). Moreover, inferences from epidemiological observations, in which donated grafts from male to female fare better than grafts from female to male, have been attributed to differences in size but without careful study of direct size measurements (18–22). This issue has not been rigorously studied in living donor kidney transplantation, a rapidly growing source of organ donation. Similarly, controversial evidence suggests that higher pre-donation GFR positively correlates with post-transplant graft outcome (5,23,24).

With a growing number of potential recipients on the waiting list for a relatively fixed and scarce supply of deceased grafts, living donor kidney transplantation has become a vital source of allografts. Identifying those living donor characteristics associated with poorer long-term outcome may aid the clinician in the management of renal transplant patients with the ultimate goal of maximizing the lifespan of allografts. The objectives of our study were (1) to examine the association between live donor kidney volumes as measured by 3D CT scanning and function as measured by 125I-iothalamate clearances (iGFR) and (2) to evaluate whether larger donated graft volume and higher iGFR were associated with better estimated long-term graft function and outcomes.

Materials and Methods

This study is a retrospective chart review approved by the Internal Review Board at the Cleveland Clinic Foundation (CCF). We studied all living donors and their organ recipients who underwent a first renal transplantation between 1998 and 2002 at CCF, and who underwent both pre-operative imaging using helical computed tomography (CT) with 3D reconstruction and iGFR measurement by 125I-iothalamate urinary clearances as part of their donor evaluation (n = 119). Recipients who were lost to follow-up (n = 2), younger than 18-years-old (n = 4), or who had lost their graft from any cause within 2 years of post-transplantation (n = 9) were excluded because we were interested in analyzing the anatomical (volume) and functional (GFR) effects of the donated kidney on adult recipient long-term graft function irrespective of other immune and non-immune-mediated factors. Two patients lost their graft from thrombotic microangiopathy, 3 from recurrent/de novo glomerular disease, 1 from BK nephropathy, 1 secondary to death with functioning graft, 1 from rejection and 1 due to primary nonfunction. Baseline demographic, laboratory and clinical information were collected for all study subjects and entered in a database. Initial immunosuppressive therapy consisted of a calcineurin inhibitor (CNI)-based regimen (n = 60) or a regimen free of CNI (n = 44). Cyclosporine or tacrolimus were given as part of a dual regimen with steroids or a triple regimen with steroids and mycophenolate mofetil or azathioprine. A CNI-free regimen consisted of sirolimus and steroids with or without mycophenolate mofetil. Clinically evident biopsy-proven acute cellular rejection (ACR) was defined by Banff criteria (25), and delayed graft function (DGF) was defined as the need for dialysis treatment in the first 7 days after transplantation.

Pre-operative kidney volumes were measured by one of two radiologists. These studies were performed on a 16-slice helical CT scanner (Sensation 16, Siemens Medical Solutions, Forchheim, Germany). Images were obtained prior to contrast and after the administration of 150 mL of iodinated contrast media during the corticomedullary and parenchymal phases of enhancement. Slice images reconstructed at 1 mm intervals and displayed in multi-planar reformations were used for measurements. The absolute renal volumes (V) of each kidney were estimated using the ‘prolate ellipsoid’ method (Equation 1) (26,27) where measured dimensions of length (L), width (W) and thickness (T) are used in the model:

image(1)

The length was the longest craniocaudal length and the width and thickness were the measured maximum diameters taken in an oblique axial plane orthogonal to the long axis of the kidney. The total kidney volume for each donor was defined as the sum of the left and right kidney volumes, and was used to assess the relationship with donor GFR. We also measured renal cortical thickness for each kidney by measuring the cortical border to the corticomedullary phase. An average of three values was used to minimize measurement error. The volume of the single transplanted kidney was then corrected for recipient body surface area (BSA) (28) to calculate the adjusted transplanted graft volume (Equation 2), in a similar fashion to the correction routinely performed when measuring GFR by radioisotope methods. This adjustment is needed in order to assess the effects of the transplanted graft volume in a particular recipient:

image(2)

125I-iothalamate urinary clearances for each donor were performed at the CCF Renal Function Laboratory as previously described (29). The results are routinely reported as GFR corrected for the donor BSA (mL/min/1.73 m2). Radioisotope clearance of iothalamate represents the total kidney clearance and does not allow for single kidney GFR measurement. To calculate the transplanted graft GFR, we used the donor GFR without the correction for donor BSA (mL/min). It was then assumed that individual kidney contribution to the overall donor GFR was proportional to its contribution to the total kidney volume (Equation 3):

image(3)

Finally, analogous to the approach used to calculate the transplanted kidney volume, we adjusted the uncorrected donated GFR (mL/min) to recipient BSA (mL/min/1.73 m2) (Equation 4):

image(4)

For each transplant recipient, GFR was calculated at 6, 12 and 24 months post-transplantation by using the abbreviated 4-variable Modification of Diet in Renal Disease (MDRD) equation (30).

To facilitate the analysis and interpretation of the results, the recipients' data were divided into subgroups according to transplanted graft volume < or ≥120 cc/1.73 m2, transplanted graft GFR < or ≥55 mL/min/1.73 m2 and donor age < or ≥45 years. The cutoff values were obtained from the approximate medians of transplanted volume (122 cc/1.73 m2), GFR (53 mL/min/1.73 m2) and age (44-years-old) in order to maintain the subgroups equal with regard to sample size. Statistical analyses performed using the data as continuous variables did not affect the overall results.

Statistical analysis

Statistical analysis was performed by using SPSS software version 11.5 (Chicago, IL). Data are expressed as mean ± SD (minimum-maximum) or as n (%) when appropriate. Categorical variables were analyzed by chi-squared test or Fisher's exact test when indicated, student t-test and/or ANOVA test to compare means for parametric data and Mann-Whitney U-test for nonparametric variables. To further characterize the effects of different donor variables and recipient factors on 2-year graft function and incidence of ACR at 2 years post-transplantation, we performed univariable and stepwise multi-variable linear and logistic regression analyses between these outcomes at 2 years and all other donor and recipient variables. Statistical significance was defined as p-values <0.05.

Results

Correlation between kidney volume and GFR

Baseline donor characteristics are presented in Table 1. Mean donor age was 43 ± 9 years, 59.7% of studied subjects were female and 84.9% were non-African American. Total kidney volumes ranged from 170 to 455 cc with a mean value of 274 cc (Figure 1). The donor absolute GFR, uncorrected for donor BSA, ranged between 76 and 182 mL/min with a mean of 118 mL/min. As shown in Figure 1, males had significantly larger kidney volumes and GFR than females. Figure 2 shows a strong correlation between the total kidney volume (cc) and absolute donor GFR (mL/min) (Pearson r= 0.71, p < 0.001). Although the correlation between renal cortical thickness and absolute GFR was statistically significant (Pearson r= 0.362, p = 0.001), this correlation was much weaker than that observed between kidney volumes and GFR, and thus further association of this variable with graft outcome was not pursued. Male gender and BSA (or body mass index) were associated with larger kidney volume and higher absolute GFR by univariable analysis (Table 2). After controlling for other factors in a multi-variable model younger age but not male gender remained independently associated with higher GFR, while male gender alone but not younger age remained independently associated with larger kidney volume. BSA also remained independently associated with donor kidney volume and GFR.

Table 1. Overall donor characteristics
  1. All data are expressed as n (%) or mean ± SD (minimum-maximum).

  2. *GFR: glomerular filtration rate as measured by 125I iothalamate clearances.

  3. Adjusted to donor BSA.

Number of studied donors119
Age (years)43 ± 9 (24–60)
≥45 years old56 (47.1)
Female gender71 (59.7)
African American race18 (15.1)
Body surface area (m2)1.86 ± 0.22 (1.42–2.43)
Serum creatinine (mg/dL)0.86 ± 0.18 (0.30–1.40)
Absolute GFR (mL/min)*118 ± 22 (76–182)
Adjusted GFR (mL/min/1.73 m2)110 ± 18 (79–166)
Absolute total kidney volume (cc)274 ± 54 (170–455)
Adjusted total kidney volume (cc/1.73 m2)254 ± 37 (174–364)
Absolute total cortical thickness (mm)10.7 ± 1.5 (6.9–15.1)
Adjusted total cortical thickness (mm/1.73 m2)10.1 ± 1.4 (7.1–14.8)
Figure 1.

Histograms depicting total donor kidney volume (A) and unadjusted donor iothalamate GFR (B) for males (in black) and females (in white). The mean ± SD GFR and total kidney volumes for males were 129 ± 21 mL/min/1.73 m2 and 312 ± 50 cc/1.73 m2, respectively, and for females were 111 ± 21 mL/min/1.73 m2 and 249 ± 41 cc/1.73 m2, respectively.

Figure 2.

Plot showing the correlation between absolute iothalamate GFR (unadjusted for recipient's BSA) and total kidney volumes.

Table 2. Donor factors independently associated with larger total kidney volume and higher absolute GFR by multi-variable linear regression analysis
 UnivariableMulti-variable
Beta*p-ValueBeta*P-Value
  1. If body mass index is used instead of body surface area as a measurement of recipient size, the overall results are similar with the exception of the effects of male gender on donor GFR, which remain as an independent predictive factor in the multi-variable linear regression analysis.

  2. *For continuous variables, beta equals change in kidney volume (cc/1.73 m2) or GFR (mL/min/1.73 m2) per unit increase in that variable; and for categorical variables, beta equals change in kidney volume (cc/1.73 m2) or GFR (mL/min/1.73 m2) with variable present or absent.

  3. Compared to non-African American race (87 Caucasians, 2 Asians, 2 other).

Total kidney volume (cc)
 Age (per 10 years of age)−1.180.829 
 Male gender63.53<0.00123.870.020
 African American race−3.820.784 
 Body surface area (per 0.1 unit)17.13<0.00113.43<0.001
Total absolute GFR (mL/min)
 Age (per 10 years of age)−5.500.014−4.230.029
 Male gender17.72<0.0010.390.936
 African American race8.700.130 
 Body surface area (per 0.1 unit)5.59<0.0015.35<0.001

Correlation between transplanted kidney volume, GFR and donor age with post-transplant recipient graft function

Table 3 shows baseline recipient characteristics. Mean recipient age was 46 ± 12 years, 60.6% being male and 85.6% non-African American. The mean transplanted volume and GFR were 125 ± 32 cc/1.73 m2 and 54 ± 12 mL/min/1.73 m2, respectively. Post-transplant graft function estimated by the 4-variable MDRD equation was analyzed for the three different subgroups (Table 4 and Figure 3). Recipients of transplanted kidneys with adjusted graft volumes <120 cc/1.73 m2 had significantly lower estimated GFR at 6 months, 1 year and 2 years post-transplantation compared with those patients that received grafts with larger volumes (Figure 3A). Similarly, recipients with transplanted adjusted GFR <55 mL/min/1.73 m2 were more likely to have lower post-transplant estimated GFR at those same time points (Figure 3B). Recipients of kidneys from donors older than 45 years compared to younger donors had similar transplanted GFR. However, those recipients of grafts from donors younger than 45-years-old had statistically higher estimated GFR at 6 months, 1 and 2 year post-transplantation (Figure 3C).

Table 3. Overall recipient characteristics
  1. All data are expressed as n (%) or mean ± SD (minimum-maximum).

  2. *ACR or DGF not leading to graft loss.

  3. Adjusted for recipient BSA.

Number of studied recipientsn = 104
Age (years)46 ± 12 (20–75)
Male gender63 (60.6)
African American race15 (14.4)
Body surface area (m2)1.95 ± 0.22 (1.38–2.46)
Weight (kg)86 ± 16 (43–119)
Type of living donation—genetically related/unrelated71 (68.3)/33 (31.7)
HLA mismatches3.0 ± 1.7 (0–6)
Acute cellular rejection*10 (9.6)
Delayed graft function*5 (4.8)
Calcineurin inhibitor-free regimen44 (42.3)
Adjusted transplanted GFR (mL/min/1.73 m2)54 ± 12 (31–85)
Adjusted transplanted kidney volume (cc/1.73 m2)125 ± 32 (68–244)
Adjusted transplanted cortical thickness (mm/1.73 m2)4.8 ± 1.1 (2.8–9.9)
Table 4. Estimated GFR in different subgroups
 Transplanted kidney volume* (cc/1.73 m2)Transplanted GFR* (mL/min/1.73 m2)Donor age (years)
<120≥120<55≥55<45≥45
  1. *Adjusted for recipient BSA: (cc/1.73 m2) for transplanted kidney volume and (mL/min/1.73 m2) for transplanted GFR.

  2. p < 0.001.

  3. Mean ± SD (mL/min/1.73 m2).

  4. §p < 0.01.

Transplanted iGFR*46 ± 761± 1046 ± 664 ± 855 ± 1153 ± 12
MDRD eGFR at 6 months53 ± 1566 ± 1555 ± 1666 ± 1565 ± 1754 ± 15
MDRD eGFR at 12 months48 ± 1361 ± 1750 ± 1561 ± 1659 ± 1750 ± 14§
MDRD eGFR at 24 months48 ± 1464 ± 1951 ± 1764 ± 1762 ± 1951 ± 16§
Figure 3.

Graft function as measured by MDRD equation at 6, 12 and 24 months in recipients of transplanted kidney volumes < and ≥120 cc/1.73 m2 (A), recipients of iGFR < and ≥55 mL/min/1.73 m2 (B), and recipients of organs from donors younger than 45 years or older (C).*p < 0.001 for each time point. p = NS for the first two time points and p < 0.01 for the last two time points.

Univariable and multi-variable linear regression analyses using estimated GFR at 24 months post-transplantation as the dependent variable were performed to identify independent donor and recipient factors associated with better graft function (Table 5). By univariable analysis, donor age greater than 45 years was associated with worse graft function (−5.32 mL/min/1.73 m2 lower estimated GFR per 10 years of donor age, p = 0.008). Conversely, when the cohort was arbitrarily divided into low versus high donor kidney volume based on the median of ∼120 mL/min/1.73 m2, a high transplanted kidney volume (and GFR) correlated with better estimated GFR (15.5 mL/min/1.73 m2 for recipients of transplanted kidney volume ≥120 cc/1.73 m2, p = <0.001, and 12.65 mL/min/1.73 m2 for recipients of transplanted GFR ≥55 mL/min/1.73 m2, p = <0.001). Linear regression analysis confirmed a statistically significant correlation between transplanted kidney volume as a continuous variable and 2-year graft GFR (β= 0.13, p = 0.028). A similar finding was observed when donor age and transplanted GFR were treated as continuous variables (data not shown). Donor gender and BSA also correlated with graft function in a univariable model. As might be anticipated, a history of biopsy-proven ACR was statistically associated with worse graft outcome (−15.39 mL/min/1.73 m2, p = 0.011); however CNI-based regimen at time of transplantation and history of DGF were not. When all statistically significant variables were entered into a multi-variable model (either transplanted graft volume or GFR was entered in the model, but not both, due to colinearity), only donor age, transplanted kidney volume, transplanted GFR and a history of ACR remained significantly associated with graft function at 2 years post-transplant.

Table 5. Multi-variable linear regression analysis showing independent variables associated with estimated GFR at 2 years (mL/min/1.73 m2)
 UnivariableMulti-variable
Beta*p-ValueBeta*p-Value
  1. If all included and excluded patients are considered for analyses, the results and interpretation of the data remain similar.

  2. *For continuous variables, beta equals change in kidney volume (cc/1.73 m2) or GFR (mL/min/1.73 m2) per unit increase in that variable; and for categorical variables, beta equals change in kidney volume (cc/1.73 m2) or GFR (mL/min/1.73 m2) with variable present or absent.

  3. Compared to non-African American race (84 Caucasians, 2 Asians, 3 other).

  4. Either transplanted kidney volume or transplanted GFR was entered in the multi-variable linear regression analyses at the time due to colinearity.

Donor factors
 Age (per 10 years of age)−5.320.008−5.290.005
 Male gender7.320.048−4.620.339
 African American race8.440.112 
 BSA (per 0.1 unit)1.700.0430.280.787
 Transplanted kidney volume >120 cc/1.73 m215.50<0.00115.03<0.001
 Transplanted GFR >55 mL/min/1.73 m212.65<0.0019.300.020
Recipient factors
 Age (per 10 years of age)0.100.947 
 Male gender−5.390.146 
 African American race9.290.071 
 Living related donation (vs. unrelated)1.280.744 
 History of ACR−15.390.011−12.350.029
 History of DGF−3.910.645 
 Use of CI-based regimen−5.800.113 
 HLA mismatches−2.050.069 

Effects of donor-recipient gender combinations on graft outcome

As expected and shown in Table 1 and Figure 1, males had higher average kidney volumes and GFR, while females had lower body sizes and thus, lower BSA. Figure 4 shows relationships between donor and recipient gender at the time of transplantation and GFR. Initial GFR for male grafts transplanted into females was 64 ± 13 versus 49 ± 9 mL/min/1.73 m2 (p < 0.001) for female grafts transplanted into males (Figure 4A). The estimated GFR at 2 years post-transplantation was still higher for female recipients of male grafts than for male recipients of female grafts (65 ± 22 vs. 53 ± 15 mL/min/1.73 m2, p = 0.047) (Figure 4B). Univariable and multi-variable analysis including all donor-recipient combinations and transplanted kidney volume showed that the observed difference in graft GFR can be accounted for by the transplanted volume and not by some other gender effect (data not shown).

Figure 4.

Effects of potential donor-recipient gender combinations on initial transplanted GFR (A) and graft GFR at 2 years post-transplantation (B). Error bars showing mean and 95% CI.

Effects of graft volume on susceptibility to ACR

Figure 5 shows the relationship between adjusted transplanted volume and incidence of rejection. ACR occurred in 8/50 (16.0%) recipients with a transplanted kidney volume <120 cc/1.73 m2 versus only 2/54 (3.7%) recipients of larger graft volumes (p = 0.046 by Fisher's exact test). Univariable analyses showed that a history of DGF (OR = 8.48, p = 0.031) and a transplanted graft volume <120 cc/1.73 m2 (OR = 4.95, p = 0.050) were the only variables significantly associated with higher incidence of clinically evident ACR within 2 years post-transplantation. Moreover, in a multi-variable logistic regression analysis transplanted kidney volume <120 cc/1.73 m2 (OR = 10.2, p = 0.035) and a history of DGF (OR = 9.00, 0.048) remained statistically significant.

Figure 5.

Effects of adjusted transplanted kidney volume on risk of acute cellular rejection within 2 years of transplant.

Discussion

Our work adds to the current literature in several ways. The first observation was a strong correlation between measured donor kidney volume and GFR, suggesting that in healthy individuals kidney mass translates into a larger number of nephron units, and thus higher GFR. This concept is not necessarily true in diseased kidneys. Body size (either by BSA or BMI) was the most important independent predictor factor for kidney volume and GFR. Among other donor characteristics investigated, only younger age was associated with higher GFR, and only male gender was an independent determinant of larger kidney volumes. The negative effect of increasing age on GFR is consistent with previous reports in which measured GFR decreased in potential kidney donors (31). In contrast, in this cohort of patients the size of adult kidneys seems not to diminish as normal individuals age despite a decline in GFR. This is consistent with the findings by Kasiske et al. who reported that body size was the major determinant of kidney size in normal subjects (32).

Second, larger transplanted kidney volume adjusted for recipient's BSA (cc/1.73 m2) was independently associated with better 2-year graft function. The impact of graft volume on graft function appeared stronger than other known predictive variables like pre-transplant GFR, age of the donor and a history of acute rejection. In humans, influence of donated organ size on graft outcome has been more extensively studied in deceased donor kidney transplantation and derived from indirect observations (4,13–16,33). Although the published data on this matter are somewhat controversial because of different measurement methods, most of the studies were able to demonstrate a positive association between larger donated organs and better recipient outcomes. More recently, Saxena et al. reported a positive study on the effects of donated kidney volumes as measured by magnetic resonance imaging on 1-year graft function in a small cohort of living kidney transplant recipients; however, other potential living donor and recipient factors known to independently impact on graft outcome were not analyzed (6). Donor age and pre-transplant GFR have also been a matter of investigation by other groups, and with a few exceptions (5), a positive correlation between these variables and post-transplant graft function was observed (21,23,24).

The mechanisms by which graft size impacts on transplant outcome have long been a matter of investigation by the transplant community. Several studies in animal models were aimed at demonstrating the effects of graft size on non-immune- and immune-mediated injury to the transplanted organ. Experiments in rat models using isografts in nephrectomized and nonnephrectomized animals demonstrated that the size of the graft (9,11,12) and the overall amount of functioning renal mass (34,35) were important determinants of chronic graft injury irrespective of antigen-mediated insult. These authors suggested that the mechanisms by which graft size affects graft outcome are intra-glomerular hypertension and hyperfiltration leading to nephron loss and decrease in graft function.

The third observation derived from this study is that our data provide an explanation for the ‘gender effect.’ The initial higher transplanted GFR seen in male to female donation compared to that of female to male donation is still maintained after 2 years. These data are consistent with the previous reports on the effects of gender combinations (18,19,22) and suggest that the observed gender effect is at least partially due to the larger donated volume and GFR of male compared with female kidneys.

Lastly, our data suggest that the volume of the donated graft influences the development of immune-mediated graft injury. Recipients of a transplanted kidney volume larger than 120 cc/1.73 m2 were less likely to develop ACR requiring treatment, a finding that was independent of other factors known to predispose patients to rejection. This result is consistent with one other published study of deceased donor graft recipients in which the incidence of acute and chronic graft rejection was increased in recipients of lower ‘nephron mass’ (36). Due to the lack of protocol biopsies we cannot exclude the possibility that both groups mount a similar degree of alloreactive effector or regulatory mononuclear response which may be expressed differently as clinical rejection in recipients of low graft volume and subclinical rejection in those of larger graft volume. In such a scenario it is possible that a larger nephron mass reserve in the latter group allows serum creatinine to remain within normal values, while in those patients with lower graft volume, a higher creatinine would prompt a biopsy that uncovered subclinical rejection. Studies in murine models lend credence to the hypothesis that graft size affects susceptibility to immune mediated injury (37,38). He et al. demonstrated that despite similar T-cell anti-donor reactivity, graft loss due to immune-mediated injury occurred more rapidly and with more intensity in recipients of smaller grafts than in those who received larger grafts. The physiologic basis of this relationship remains obscure but others have suggested that the renal mass supplied by the donor exerts an immune-modulatory effect by affecting surface molecule expression and cellular infiltration in response to progressive structural changes observed in response to low transplanted nephron mass (37).

Several limitations of our study need to be noted. First is the relatively short follow-up time of 2 years, limiting our ability to state with certainty that donated kidney volume predicts long-term graft survival in living renal transplantation beyond this time period. Early graft function as determined by serum creatinine at 6 months and 1 year post-transplantation is a good predictor of longer term graft survival (39), and therefore, our findings may predict outcomes beyond 2 years. We also recognize that calculated GFR by estimation equations to assess graft function have limitations. However, creatinine-based methods specifically the MDRD equation used herein, have performed better than other formulas in renal transplantation (40–42) and this approach is commonly used in the research literature to estimate graft function. Finally, although our data demonstrate a correlation between renal volume as measured by 3D CT scanning and late GFR, we acknowledge that we did not formally demonstrate a correlation between 3D CT scanning and actual kidney weight. Thus, we cannot definitively state whether this method will be ideal for routine clinical use.

In conclusion, our study provides strong evidence that the size of living donor kidneys can have a significant and independent impact on post-transplant outcome. These results have important clinical implications (1) donor kidney size and GFR should be considered as predictive factors for graft outcome in living kidney transplantation, (2) recipients of small kidneys may benefit from closer graft surveillance and potentially from renoprotective maneuvers/drugs and (3) the size of donor kidneys should be included in the factors considered when choosing among otherwise similar potential donors.

Acknowledgment

Only internal funds were used to prepare this manuscript.

Ancillary