Nephrol Dial Transplant (2007) 22: 3646–3651
doi:10.1093/ndt/gfm487
Advance Access publication 17 August 2007
Original Article
Importance of renal mass on graft function outcome after
12 months of living donor kidney transplantation
João Batista Douverny, José Carlos Baptista-Silva, José Osmar Medina Pestana and Ricardo Sesso
Division of Nephrology, Federal University of São Paulo, São Paulo, Brazil
Abstract
Background. Few studies have directly measured
the kidney weight and investigated donor
parameters related to it. The aim of this study was
to evaluate the kidney weight and its relationship
to creatinine clearance (CrCl) after 12 months
post-transplantation.
Methods. A total of 123 recipients of renal transplantation from living donors were evaluated. Demographic
and anthropometric data from donors and recipients
were collected in the pre-operative phase. Data about
kidney weight were obtained through kidney measurement using an electronic weighing machine at the
moment of transplantation. Glomerular filtration
rate (GFR) was estimated through CrCl (modification
of diet in renal disease formula) at the 1st, 6th, 12th
and 18th month post-transplantation.
Results. The mean value of kidney weight was 170 31 g (166.4 29.2 g in women and 177.5 32.5 g in
men). The kidney weight had a correlation with the
donor’s BMI (r ¼ 0.43, P < 0.001) and with the CrCl on
the 12th month (r ¼ 0.31, P ¼ 0.001). Using multiple
linear regression, the kidney weight could be predicted
through the BMI and donor’s gender (R2 ¼ 0.21;
P < 0.01). The CrCl after 12 months had a significant
correlation with the graft weight/recipient weight ratio
and with the donor age (R2 ¼ 0.22; P < 0.01).
Conclusion. The kidney weight can be estimated using
the donor’s gender and BMI. The kidney weight
significantly influences the CrCl 12 months after
transplantation.
Keywords: graft function; kidney transplantation;
kidney weight; living donor; renal mass
Correspondence to: Dr João Batista Douverny. Av. Imperatriz
Leopoldina 303 apt. 64, São Paulo, SP, Brazil, 09770-271.
Email: [email protected]
Introduction
The amount of functioning nephron has been implicated as a factor that influences the development of
chronic rejection of kidney allografts. Reductions in
nephron number below 50% induce glomerular hypertension and hyperfiltration in the remaining units,
resulting in graft injury [1].
The nephron mass of kidney donors has been
recognized as a predictive factor for the kidney
transplant outcome. Experimental reduction of the
nephron mass, related to compensatory mechanisms,
affects size and graft function. The hyperfiltration that
occurs in the remaining nephrons can also promote
albuminuria and glomerulosclerosis [2].
In experimental studies, animals submitted to
bilateral nephrectomy presented with a high level of
proteinuria, cellular infiltration and glomerulosclerosis
in single allograft transplantation with reduced renal
mass. On the other hand, recipients with two kidneys
showed markedly reduced indices of allograft injury,
independent of whether the second kidney was native
or transplanted [3].
These findings provide clear evidence that the cycle
of gradual nephron loss characteristic of renal mass
ablation occurs in kidney transplantation and contributes significantly to renal injury. In order to
prevent the hyperfiltration, it is important to have an
adequate number of nephrons to supply the metabolic
demands of the recipient, as well as to guarantee an
adequate mass of nephrons in case of rejection
episodes, nephrotoxicity due to drugs or other injuries.
As the number of nephrons of the kidney cannot be
measured ‘in vivo’, the best marker for renal mass
seems to be the renal weight [4].
The few studies that directly determined the weight
of the kidney through an electronic weighing scale,
related it to the weight of the recipient or with the
survival rate of the renal graft [5–7]. The aim of this
study was to directly evaluate the weight of the kidney
of living donors at the moment of the renal transplant,
verify the donor factors related to it and to investigate
ß The Author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please email: [email protected]
Renal mass and graft function
its influence on the graft function 12 months
post-transplantation.
Subjects and methods
Initially, a total of 178 patients who had undergone living
renal transplantation were evaluated between 2001 and 2003,
in the Hospital of the Federal University of São Paulo.
Fifty-five patients were excluded from the study due to loss
of follow-up (n ¼ 29), period of follow-up which was less than
1 year after transplantation (n ¼ 21) and death caused by
infections (n ¼ 3) or vascular thromboses in the graft (n ¼ 2)
during the first year after transplant.
Clinical and laboratory data concerning donors and
recipients were obtained in two phases. The first phase
consisted of information obtainment before and during
surgery. In the second phase, data from recipients were
collected from medical records during the follow-up. All of
the data from recipients and donors, including height,
weight, age, sex, kinship, blood group and other information,
were obtained during the pre-operative period.
Some data related to the recipients were also obtained,
including the time of dialysis, dialysis modality, renal disease
aetiology, associated diseases and medications used. The
body mass index (BMI) and body surface area (BSA) were
calculated through the use of mathematical formulas:
BMI ¼ weight (kg)/height (m)2 and BSA ¼ 0.007184 weight (kg)0.425 height (cm)0.725.
During the surgical procedures, data from the donor
kidney were collected, including weight, volume, length and
width. In addition, at the time of each follow-up period
(the 1st, 6th, 12th and 18th month post-transplantation)
data were collected from the medical records of recipients,
including serum creatinine, age and body weight. Serum
creatinine was measured by an automatized method using the
Jaffé reaction (Hitachi 912 System, Roche Diagnostic
Corporation, Hitachi Ltd., Tokyo, Japan).
The laboratory of clinical analyses (University Hospital do
Rim) performed all the laboratory examinations. Data about
immunosuppressive drugs, rejection episodes and their
treatment and biopsies performed were also collected from
patients’ medical records. Creatinine clearance (CrCl) was
estimated by the GFR using the abbreviated modification
of diet in renal disease equation (MDRD-GFR) [8], GFR
(ml/min/1.73 m2) ¼ 186 (Scr)1.154 (Age)0.203 (0.742 if
female) (1.210 if black). The age and serum creatinine
values applied to the formula were obtained at each
follow-up period (the 1st, 6th, 12th and 18th month).
Kidney weight and volume
The process of measurement of the weight, volume, length
and width of the kidney was performed immediately after the
donor’s nephrectomy. After nephrectomy, the kidney was
perfused via catheterization of the renal artery with a
perfusion solution (EurocolinsÕ or SoltranÕ —Kidney perfusion solution BaxterÕ ) until a white colouration of kidney
was reached and the perfusion solution left the renal vein.
Then, the excess peri-renal adiposity was removed from the
kidney, except from the renal pelvis area, lymphatic vases
and ureter. The measurements were performed by the same
surgeon (co-author J.C.B.-S), in sterile conditions, and
3647
always using the same electronic weighing scale
(FILIZOLLAÕ BP6), with a precision of 2 g, maximum
load of 6 kg and minimum load of 50 g. The device was
calibrated and surveyed monthly by Inmetro (Brazilian
Institute of Weights and Measures).
At the moment of the weighing process, the kidney
consisted of the renal mass (peri-renal adiposity free), renal
capsule, pelvis, artery and renal veins, ureter and lymphatic
vases. Then, the kidney was held by a cardiac strip,
suspended by its extremities and weighed on the electronic
weighing scale. From the measured weight, 2.0 g was
deducted due to the cardiac strip, resulting in the final
weight of the kidney. The length and width measures were
performed through a metallic, stainless, flexible and sterile
ruler with a millimetre scale.
Statistical analysis
Data of quantitative variables are summarized through
descriptive measures (mean SD). In order to investigate
associations between qualitative variables and CrCl, the
ANOVA with repeated measures was used. The Tukey test
was used for comparisons between two by two groups. The
Pearson’s correlation coefficient, with its respective 95%
confidence interval (95% CI), was used to evaluate the
strength of the association between quantitative variables
and CrCl.
Multiple linear regression was used to investigate the
factors related to kidney weight and to CrCl at the 12th
month. In this analysis, the independent variables with
P < 0.20 (in univariate analysis) were tested, but only those
with statistical significance were kept in the final model. In all
statistical analyses, a level of significance of P < 0.05 was
adopted.
Results
During the period of examination, among the 178
recipients of renal transplant, 55 were excluded, leaving
123 patients for the analysis. The recipients had a mean
age of 35 13 years, weight of 61 17 kg, height of
1.61 0.1 m and BMI of 23.2 5.0 kg/m2. The donors
had a mean age of 44 9 years (range: 23–64), weight
of 69 12 kg, height of 1.62 0.1 m and BMI of
26.4 4.0 kg/m2. Donors were predominantly of
female gender (65%) and recipients of male gender
(67%). Donors and recipients were predominantly
of white race (80%).
Regarding ABO blood groups, the most frequent
types observed among donors were ‘A’ and ‘O’ (27
and 65%, respectively). The corresponding figures
for donors were 39 and 46%, respectively. The most
frequent HLA typing was haplo-identical (55%),
followed by HLA-identical (33%). Siblings, mothers
and fathers corresponded to 56, 19 and 13% of
the donors. The most frequent previous dialysis
therapy was haemodialysis (86%) and the mean time
on dialysis was 20.4 16.7 months. The most frequently used immunosuppressive drugs in the last
patient evaluation were prednisone, azathioprine and
cyclosporine (98.4, 76.4 and 66.7%, respectively).
3648
J. B. Douverny et al.
Table 1. Kidney graft weight, volume and relationships with donor
and recipient characteristics
Table 2. Correlations between CrCl at 12 months and clinical and
demographic characteristics
Variables
Mean SD
Range
Variables
r
95% CI
Kidney
Kidney
Kidney
Kidney
Kidney
Kidney
Kidney
Kidney
170 31
157 33
3.1 1.2
2.5 0.4
2.8 1.2
2.3 0.5
7.6 2.0
7.0 2.0
(94–254)
(55–242)
(1.2–8.5)
(1.4–4.2)
(1.0–8.4)
(0.8–3.9)
(3.4–14.4)
(2.3–14.3)
Donor age (years)
Recipient age (years)
Dialysis time (months)
Kidney weight (g)
Kidney volume (cm3)
Recipient weight (kg)
Kidney weight/Recipient
weight (g/kg)
Recipient BMI (kg/m2)
0.332
0.356
0.006
0.305
0.288
0.285
0.397
0.481;
0.501;
0.183;
0.135;
0.117;
0.440;
0.237;
0.165
0.191
0.171
0.457
0.442
0.114
0.536
<0.001
<0.001
0.951
0.001
0.001
0.001
<0.001
0.179
0.345; 0.002
0.047
200
300
weight (g)
volume (cm3)
weight/recipient weight (g/kg)
weight/donor weight (g/kg)
volume/recipient weight (cm3/kg)
volume/donor weight (cm3/kg)
weight/recipient BMI (g/kg/m2)
weight/donor BMI (g/kg/m2)
P-value
60
120
CrCl at 12 months
CrCl
58
56
54
52
50
0
3
6
9
12
15
100
80
60
40
20
18
0
Time (months)
50
100
Fig. 1. Mean SE of CrCl post-transplantation.
150
250
Kidney weight (g)
Fig. 2. Linear regression between CrCl and kidney weight.
Graft characteristics such as weight and volume and
their relations to donor and recipient parameters are
shown in Table 1.
Figure 1 shows the average profile of CrCl during
the 18 months of follow-up. There was no statistically
significant change in CrCl over time (P ¼ 0.593). No
significant difference was detected between the 1st
and the 6th month (P ¼ 0.209); between the 1st and the
12th month (P ¼ 0.641) and between the 1st and the
18th month (P ¼ 0.567). As the mean values between
the 12th and the 18th month were similar, we decided
to use the former period to perform the comparisons
with clinical and demographic variables, due to the
greater number of patients with available data at that
time and the stability of the renal function compared
with the 18th month.
The correlations between the CrCl at 12 months
and the clinical and demographic variables of recipients and donors are shown in Table 2. The CrCl at
12 months was positively correlated with kidney
weight, kidney volume and graft’s weight/recipient’s
weight ratio. There was an inverse correlation between
CrCl and donor age, recipient age, recipient weight and
recipient BMI. There was no statistically significant
correlation between CrCl at 12 months and ABO
blood group of donors and recipients, HLA typing,
blood kinship of donor, type of dialysis treatment,
aetiology of the renal disease and combination between
race and sex of donors and recipients.
CI: 0.14–0.46, P ¼ 0.001) (Table 2 and Figure 2). This
relation was analysed categorizing graft weight
in terciles, as shown in Table 3. Mean values of CrCl
in the first tercile of kidney weight were lower than
those of the second (P ¼ 0.088) and the third tercile
(P ¼ 0.010). There was no significant difference
between the second and third terciles (P ¼ 0.651).
The kidney weight was significantly correlated with
the donor’s BMI (r ¼ 0.43; 95% CI: 0.27–0.56,
P < 0.001) (Figure 3). The mean weight of the kidneys
from female donors was lower than that from males
(166.4 29.2 g
vs
177.5 32.5 g,
respectively,
P ¼ 0.055).
Relationship between CrCl and kidney weight
Multiple linear regression for kidney weight
There was a significant correlation between graft
weight and CrCl at 12 months (r ¼ 0.31, 95%
The multiple linear regression analysis for the variables
that influenced kidney weight is shown in Table 4.
Table 3. Distribution of the patients and CrCl at 12 months
according to terciles of kidney weight
Tercile
Kidney weight
n (%)
CrCl
First tercile
Second tercile
Third tercile
156.0 g
156.1–180.0 g
>180.0 g
Total
42
42
39
123
51.6 13.8
58.1 14.7
60.9 13.6#
(34.1)
(34.1)
(31.8)
(100.0)
P < 0.09 vs First tercile.
P ¼ 0.010 vs First tercile.
#
Renal mass and graft function
3649
Table 5. Linear regression model for CrCl at 12 months
Donor Kidney Weight (g)
270
240
210
180
Variable
Coefficient
Standard error
P-value
Constant
Kidney weight/recipient
weight (g/kg)
Donor age (years)
63.198
4.059
7.174
0.967
<0.001
<0.001
0.426
0.133
0.001
2
R ¼ 0.22.
150
120
Discussion
90
15
20
25
30
35
40
Donor Body Mass Index (kg/m2)
Fig. 3. Scattered plots between donor BMI and kidney weight.
Table 4. Multiple linear regression model for kidney weight
Variable
Coefficient
Standard error
P-value
Constant
Donor BMI (kg/m2)
Donor sex (male vs female)
78.700
3.334
10.035
6.863
0.643
5.330
<0.001
<0.001
0.062
R2 ¼ 0.21.
When the donor’s BMI increases by 1 kg/m2,
the kidney weight increases by 3.33 g (P < 0.001). For
male donors, there is an addition of 10.0 g in kidney
weight. The coefficient of explanation of this model
was 21%, i.e. the donor’s BMI and gender are
responsible for 21% of the total variability in kidney
weight. The linear regression equation for kidney
weight is as follows:
weight of the kidney ðgÞ ¼ 78:700 þ 3:334 BMIdonor
þ 10:035 genderdonor ,
where BMI is expressed in kg/m2, male sex ¼ 1 and
female sex ¼ 0.
The findings reported by Brenner et al. [2] suggest that
renal allograft survival might be improved by matching
nephron mass to recipient needs. Graft weight might
serve as a useful alternative to estimate nephron mass
until better indices or measures became available.
The interest in non-immunological factors that affect
long-term allograft survival emphasizes the importance
of an adequacy of renal graft mass to recipient
metabolic demands.
Meier-Kriesche et al. [9] studied the relation between
anthropometric characteristics of donors and recipients with the graft function and recipient survival.
Renal transplant recipients with a higher BMI showed
inferior patient survival rates compared with patients
with a lower BMI. The BMI showed a very strong
association with the outcomes after renal transplantation. The extremes of very high and very low BMI were
associated with a significantly worse patient and graft
survival [9,10]. In the present study, the recipient’s
BMI had a positive and significant correlation with
CrCl at 12 months.
In renal transplantation, donor age and allograft size
are known to have an important influence on graft
outcome, reflecting the functionality of the renal mass.
Our findings confirm that donor age is an important
factor related to renal graft function. Female kidneys
tend to be smaller and seem to have less nephrons than
male kidneys. The number of glomeruli per kidney,
as well as the mean glomerular volume, are closely
correlated with kidney weight and negatively correlated with age [11]. In our study, we did not detect
a statistically significant influence of the donor’s sex
on the graft function.
Multiple linear regression for CrCl at 12 months
In this model, all variables with P < 0.20 in the
univariate analysis and other variables reported
in the literature (recipient age, rejection episodes,
acute tubular necrosis, native kidney disease,
immunosuppressive drugs, ACE inhibitors, etc.) that
could influence CrCl at 12 months were tested
(Table 5). The final regression model equation is as
follows:
CrCl ðml=min:=1:73 m2 Þ ¼
63:198 þ 4:059 kidney weight=recipient weight ðg=kgÞ
0:426 agedonor ðyearsÞ:
Kidney weight estimate
The kidney weight has been the object of several
studies due to its relation with the amount of donor
nephrons and its possible association with the outcomes of kidney transplantation [11,12]. Some studies
have tried to substitute the direct measure of the
kidney weight by other indirect methods in order to
measure the renal size, such as the ultrasonography,
magnetic nuclear resonance and three-dimensional
helicoidal tomography [13–15].
In this study, we evaluated kidney weight and
explored other factors that could predict it.
3650
Using a multiple linear regression analysis, donor’s
BMI and sex were significantly correlated with graft
weight. The mathematical expression obtained allows
the estimation of kidney weight in living donors,
without the need of sophisticated image examinations,
which could increase the costs of kidney donation
studies. The validity of the linear regression equation
obtained must be tested in other groups of patients
in future studies in order to verify its real applicability.
It is interesting to note that, in our analysis, donor
age was not included in the formula as an independent
factor that significantly influences kidney weight.
Some reasons to explain this finding are: since the
BMI (which is part of the equation) is continuously
growing with age, age is indirectly reflected by
the BMI; another reason for hiding the direct
influence of age on kidney weight may be related to
the relatively low mean age of donors (44 years) in this
study.
According to Kwon et al. [11], the metabolic demand
is represented by parameters such as weight, height,
BSA and BMI. As female donors have less muscular
mass, these parameters are lower in females compared
with males; consequently, kidney weight will also be
lower. In our study, the mean graft weight in females
was 11 g less than that in males. In the previous studies
using kidney weighing, results of graft weight were not
separated by sex, so it is impossible to compare them
with the results of this study [5–7].
In a study using the summation of magnetic
resonance images, Cheong et al. [15] reported that
the kidney volume and kidney dimensions in men were
greater than those in women. They suggested that the
magnetic resonance method was more precise than the
ultrasound ellipsoid method for kidney measurement
[13]. In our study, the mean graft weight (170 31 g)
was not significantly different from that reported by
Pourmand et al. [5] (164 g) and by Toma et al. [7]
(167 g); however, our results were lower than those
reported by Giral et al. [6] (202 g), who also performed
a direct measure of renal weight.
The discussion regarding the ideal graft weight
for reaching values of CrCl from 60 to 89 ml/min,
corresponding to stage 2 of CKD, has not been carried
out in the literature thus far. Our data indicate that
recipients with a kidney graft weight greater than 180 g
will have a mean estimated CrCl of 61 ml/min, which
would be a reasonable renal function on the basis of
the current knowledge for the prognosis.
J. B. Douverny et al.
tissue in the renal pelvis should be discounted in the
kidney weight measurement, however, this tissue
cannot be removed in the surgical procedure and its
amount is probably negligible in the final kidney
weight estimate. Although graft weighing during the
transplant surgery is more accurate, the results cannot
be used previously to the operative procedure, which
precludes the anticipated use of this information.
In our study, we performed a practical procedure
accessible to any surgeon involved with renal
transplantation.
In a study performed with 1142 small kidney grafts
in large recipients, whose relation graft weight/recipient body weight was equal to or less than 2 g/kg, Giral
et al. [6] showed that the risk of proteinuria higher than
0.5 g/day increased by 50%. Kim et al. [16], studying
the relation between graft weight and recipient body
weight (mean of 3.8 g/kg), showed that recipients with
increased ratios had better graft function. In our study,
the mean of the graft weight to recipient body weight
ratio was 3.1 1.2 g/kg. A significantly positive correlation (r ¼ 0.40) was observed between this ratio and
the CrCl at 12 months. Multiple regression analysis
adjusting for a number of variables confirmed that
graft weight/recipient weight ratio was a significant
and independent predictor of renal function at
12 months. The final model also included donor age
as a significant variable. Our finding that graft weight
influences renal graft function after 1 year is in
agreement with other studies [5,6,14,16]. Not only
graft weight but, particularly, its relation with recipient
weight were relevant factors for graft function at
12 months. In the final regression model to estimate
CrCl at 12 months, an increase of 1 g/kg in the kidney
weight/recipient weight ratio added up to 4.1 ml/min in
the CrCl.
Kidney graft weight can be estimated by donor
sex and BMI. Graft weight is an important factor
in the prediction of renal function after 12 months of
transplantation, and it can be a useful parameter to be
considered in the process of selection of recipients.
Acknowledgements. Dr R.S. received a research grant and
Dr J.B.D. received a fellowship grant from the Brazilian Research
Council (CNPq).
Conflict of interest statement. None declared.
References
Method of direct kidney weighing
There is no doubt that the best method to directly
determine graft weight and to indirectly determine the
nephron mass, is the weighing of the graft by the
surgeon himself through the use of electronic weighing
scales during the surgery, as was done in the present
study. As all measurements were performed by the
same surgeon, our estimates have a high reliability.
It should be noted that, ideally, the fat and connective
1. Azuma H, Nadeau K, Mackenzie HS et al. Nephron mass
modulates the hemodynamic, cellular, and molecular
response of the rat renal allograft. Transplantation 1997; 63:
519–528
2. Brenner BM, Milford EL. Nephrons underdosing: a programmed
cause of chronic renal enxerto failure. Am J Kidney Dis 1993; 21
[Suppl 2]: 66–72
3. Velosa JA, Offord KP, Schroeder DR. Effect of age, sex, and
glomerular filtration rate on renal function outcome of living
kidney donors. Transplantation 1995; 60: 1618–1621
Renal mass and graft function
4. Taal M, Tilney N, Mackenzie H. Renal mass: an important
determinant of late allograft outcome. Transplant Rev 1998; 12:
74–84
5. Pourmand G, Taheri M, Mehrsai AR et al. Impact of donor
nephrons mass on outcomes in renal transplantation. Transplant
Proc 2001; 33: 2828–2829
6. Giral M, Nguyen JM, Karam G et al. Impact of graft mass on
the clinical outcome of kidney transplants. J Am Soc Nephrol
2005; 16: 261–268
7. Toma H, Tanabe K, Tokumoto T et al. Time-dependent risk
factors influencing the long-term outcome in living renal
allografts: donor age is a crucial risk factor for long-term graft
survival more than 5 years after transplantation. Transplantation
2001; 72: 940–947
8. Levey AS, Greene T, Kusek J, Beck G. A simplified equation
to predict glomerular filtration rate from serum creatinine
(abstract). J Am Soc Nephrol 2000; 11: 155A
9. Meier-Kriesche HU, Arndorfer JA, Kaplan B. The impact of
body mass index on renal transplant outcomes: a significant
independent risk factor for graft failure and patient death.
Transplantation 2002; 73: 70–74
3651
10. Meier-Kriesche HU, Vaghela M, Thambuganipalle R et al. The
effect of body mass index on long-term renal allograft survival.
Transplantation 1999; 68: 1294–1297
11. Kwon OI, Kwak JY, Kang CM. The impact of gender and age
matching for long-term graft survival in living donor renal
transplantation. Transplant Proc 2005; 37: 726–768
12. Gaston RS, Hudson SL, Julian BA et al. Impact of donor/
recipient size matching on outcomes in renal transplantation.
Transplantation 1996; 61: 383–388
13. Jones TB, Riddick LR, Harpen MD et al. Ultrasonographic
determination of renal mass and renal volume. J Ultrasound Med
1983; 2: 151–154
14. Nicholson ML, Windmill DC, Horsburgh T et al. Influence of
allograft size to recipient body-weight ratio on the long-term
outcome of renal transplantation. Br J Surg 2000; 87: 314–319
15. Cheong B, Rubin MF, Muthupillai R et al. Normal values for
renal length and volume as measured by magnetic resonance
imaging. Clin J Am Soc Nephrol 2007; 2: 38–45
16. Kim YS, Moon JI, Kim DK et al. Ratio of donor kidney weight
to recipient bodyweight as an index of graft function. Lancet
2001; 357: 1180–1181
Received for publication: 2.4.07
Accepted in revised form: 26.6.07