Predictive Capacity of Pre-Donation GFR and Renal Reserve Capacity for Donor Renal Function After Living Kidney Donation


* Corresponding author: G.J. Navis,


Kidney transplantation from living donors is important to reduce organ shortage. Reliable pre-operative estimation of post-donation renal function is essential. We evaluated the predictive potential of pre-donation glomerular filtration rate (GFR) (iothalamate) and renal reserve capacity for post-donation GFR in kidney donors.

GFR was measured in 125 consecutive donors (age 49 ± 11 years; 36% male) 119 ± 99 days before baseline GFR (GFRb) and 57 ± 16 days after donation (GFRpost). Reserve capacity was assessed as GFR during stimulation by low-dose dopamine (GFRdopa), amino acids (GFRAA) and both (GFRmax).

GFRb was 112 ± 18, GFRdopa 124 ± 22, GFRAA 127 ± 19 and GFRmax 138 ± 22 mL/min. After donation, GFR remained 64 ± 7%. GFRpost was predicted by GFRb(R2= 0.54), GFRdopa(R2= 0.35), GFRAA(R2= 0.56), GFRmax(R2= 0.55)and age (R2=–0.22; p < 0.001for all). Linear regression provided the equationGFRpost= 20.01 + (0.46*GFRb). Multivariate analysis predicted GFRpost by GFRb, age and GFRmax(R2= 0.61, p < 0.001). Post-donation renal function impairment (GFR ≤ 60 mL/min/1.73 m2) occurred in 31 donors. On logistic regression, GFRb, body mass index (BMI) and age were independent predictors for renal function impairment, without added value of reserve capacity.

GFR allows a relatively reliable prediction of post-donation GFR, improving by taking age and stimulated GFR into account. Long-term studies are needed to further assess the prognostic value of pre-donation characteristics and to prospectively identify subjects with higher risk for renal function loss.


Organs from living donors provide an increasingly important resource for kidney transplantation that not only helps to reduce the shortage in donor organs, but also leads to a better outcome for the recipient than transplantation from a post-mortal donor. However, the maintenance of good donor health after kidney donation has to be ensured and more specifically, any clinically relevant renal function impairment should be prevented.

Previous studies have shown that unilateral nephrectomy in healthy living donors usually does not lead to clinically significant short- or long-term renal damage in the remaining kidney (1–4). Nevertheless, in a small number of kidney donors, signs of renal damage have been reported, in some cases even leading to the need for renal replacement therapy for the donors themselves (2,5). This face stresses the need for a meticulous and responsible donor screening procedure that ensures a reliable prognosis of residual renal function after unilateral donor nephrectomy. In fact, an accurate assessment of the remaining post-donation renal function is crucial to justify the living donation policy.

In our center, the donor-screening protocol includes determination of glomerular filtration rate (GFR) by iothalamate clearance, and measurement of stimulated GFR during infusion of low-dose dopamine and amino acids, reflecting renal reserve capacity (6). In the current study, we have evaluated the prognostic value of pre-donation GFR and the added value of stimulated pre-donation GFR for post-donation GFR. As many centers with a living donor program currently rely on creatinine-based estimates of renal function, we evaluated the predictive value of pre-donation Cockcroft-Gault and modified modification of diet in renal disease (MDRD) estimated clearances for reference. In addition, we attempted to identify predictive factors pertinent to impairment of renal function after kidney donation.

Donors and Methods

Data from 125 consecutive living kidney donors that participated in the donor screening protocol with subsequent donation were included in the present analyses. None had a history of kidney disease, diabetes, cardiovascular events or hypertension, and all were without the use of antihypertensive drugs. Physical examination did not reveal abnormal findings.

Assessment of renal function and renal reserve capacity

Renal function and renal reserve capacity were measured as part of the screening protocol at a mean of 119 ± 99 days before donation as the clearance of constantly infused 125I-iothalamate on 2 separate days as described in detail previously (6). Briefly, on the first day, baseline GFR (GFRb) and dopamine-stimulated GFR (GFRdopa) were measured. On the second day, GFR stimulated by amino acids (GFRAA) and by the combination of amino acids and low-dose dopamine (GFRmax) were measured. GFR and stimulated GFR were measured as the urinary clearance (U*V/P) of 125I-iothalamate. Simultaneous infusion of 131I-hippuran was used to correct the GFR values for inaccurate urine collection as described previously (7). The coefficient of variation for this GFR measurement is 2.2%.

The procedure was as follows. The donors were in a quiet room, in semi-supine position. After drawing a blank blood sample, a priming solution containing 0.04 mL/kg body weight of the infusion fluid (4 MBq of 131I-hippuran and 3 MBq of 125I-iothalamate per 100 mL saline) plus a bolus of 0.6 MBq of 125I-iothalamate was administered at 8.00 a.m., after which the radio-isotopes were infused at a constant rate of 12 mL/h. After an equilibration period of 2 h to allow for stable plasma concentrations, a clearance period of 2 h was conducted and used for analysis. After the GFRb assessment, 1.5–2.0 μg/kg/min of dopamine was infused and GFRdopa was measured over an additional 2 h.

For the assessment of GFR stimulated by amino acid and amino acid/dopamine combined, infusion of a 7% amino acid solution (Vamin®, Fresenius-Kabi, Den Bosch, The Netherlands) at 500 mL/6 h was started at 6.00 p.m. on the day before the second renal function measurement. It was continued overnight and throughout renal function measurement. The protocol during the second day was similar to the first day with a 2-h clearance period during amino acids alone (GFRAA), followed by a 2-h clearance period during simultaneous infusion of low-dose dopamine (GFRmax). During the measurements, adequate diuresis was maintained by oral administration of 200 mL of fluids per hour. Urine was collected by spontaneous voiding. Blood samples were drawn at start, middle and end of each clearance period. After donation, with a mean follow-up of 57 ± 16 days, GFR was measured as described above, similar to the GFRb measurement.

A creatinine-based estimate of renal function was made by Cockcroft-Gault's and the modified MDRD equations (8,9). Serum creatinine was measured from blank blood samples, drawn at the time of iothalamate GFR measurement, both before and after nephrectomy. For 63 donors, follow-up of creatinine values for longer than 1 year after donation was available for analysis.

Blood pressure

Blood pressure was measured with a semi-automated device (Dinamap® 1846, Critikon, Tampa, FL, USA).

Statistical analysis

Analyses were performed using SPSS software version 12.0 (SPSS Inc., Chicago, IL, USA). Data are given as mean ± standard deviation. Pearson's correlation coefficients were calculated to account for univariate correlations. Student's paired t-test was used to compare GFRb and stimulated GFR values, as well as post-donation to pre-donation values. The χ2 test was used for categorical variables.

Two different analyses were performed. First, we aimed to predict post-donation renal function from the pre-donation GFR measurements. This was done by univariate and multivariate linear regression analysis, and tested by analysis of variance (ANOVA). Second, we tested whether pre-donation characteristics were different between subjects that developed renal function impairment after donation, and those who did not (comparison by Students t-test). Renal function impairment was defined as a post-donation GFR below the cut-off for moderate renal function impairment (i.e. ≤60 mL/min/1.73 m2) according to the Kidney Disease Outcome Quality Initiative (K/DOQI) guidelines (10). To test for the independent contribution of different predictors of post-donation renal function impairment, logistic regression was performed. To this purpose, pre-donation GFR was divided into quartiles. Two-sided p-values less than 0.05 were considered significant.


Donor characteristics and renal function are summarized in Table 1 for the whole donor population (left) and with stratification by gender (right). Mean age at donation was 49 ± 11 years and 36% of donors were male. Mean blood pressure was 123 ± 11 / 75 ± 8 mmHg. Mean GFRb prior to donation was 112 ± 18 mL/min, corresponding to 104 ± 15 mL/min/1.73 m2 after correction for body surface area (BSA). During dopamine infusion, GFR increased by 11 ± 13 mL/min; during amino acids and amino acids plus dopamine infusion, the increases were 15 ± 11 and 26 ± 12 mL/min, respectively (all p < 0.001 vs. baseline). As anticipated, GFR was higher in men, but BSA-corrected values were similar in men and women. There were no significant differences in age, blood pressure and body mass index (BMI) between male and female donors.

Table 1.  Donor characteristics before and after one-sided donor nephrectomy. Glomerular filtration rate (GFR) was measured as iothalamate clearance. Data are expressed as mean ± standard deviation. Post-donation values are compared to pre-donation values with paired samples t-test. Gender differences are evaluated with Student's t-test. p-Values given in table: male vs. female.
 Total n = 125Male n = 45 (36%)Female n = 80 (64%)p-Value
  1. Nx = nephrectomy; BP = blood pressure; GFRdopa= dopamine-stimulated GFR; GFRAA= amino acids stimulated GFR; GFRmax= maximal stimulated GFR by infusion of both dopamine and amino acids; NS = not significant; MDRD = modification of diet in renal disease equation; CG = renal function estimated by Cockcroft-Gault's formula.

  2. GFR corrected for body surface area.

  3. ap < 0.001 post-donation vs. pre-donation.

  4. bp < 0.01 post-donation vs. pre-donation.

  5. cp < 0.05 post-donation vs. pre-donation.

  6. dBorderline significance: p = 0.07 post-donation vs. pre-donation.

Characteristics before Nx
 Age at Nx (years)49 ± 1151 ± 1448 ± 10NS
 BP systolic/diastolic (mmHg)123 ± 11 / 75 ± 8125 ± 11 / 76 ± 8122 ± 11 / 74 ± 8NS / NS
 Body mass index (kg*m−2)25 ± 425 ± 325 ± 4NS
 Body surface area (m2)1.88 ± 0.182.03 ± 0.121.80 ± 0.15<0.001
 GFR (mL/min)112 ± 18120 ± 22108 ± 15<0.01
 GFR (mL/min/1.73 m2)104 ± 15102 ± 17104 ± 14NS
 GFRdopa (mL/min)127 ± 21131 ± 26120 ± 18<0.05
 GFRAA (mL/min)132 ± 19133 ± 23124 ± 16<0.05
 GFRmax (mL/min)143 ± 22144 ± 26134 ± 18<0.05
 Cockcroft-Gault (mL/min)91 ± 2096 ± 2288 ± 19<0.05
 Simplified MDRD (mL/min)76 ± 1076 ± 1174 ± 9NS
 Serum creatinine (μmol/L)84 ± 1396 ± 979 ± 8<0.001
Characteristics after Nx
 BP systolic/diastolic (mmHg)127 ± 13b / 78 ± 7a130 ± 13c / 80 ± 7b125 ± 14d / 78 ± 8cNS / NS
 GFR (mL/min)72 ± 11a75 ± 13a70 ± 10a<0.05
 GFR (mL/min/1.73 m2)66 ± 10a64 ± 10a67 ± 9aNS
 Cockcroft-Gault (mL/min)67 ± 16a70 ± 18a66 ± 14aNS
 Simplified MDRD (mL/min)54 ± 8a54 ± 9a54 ± 7aNS
 Decrease in GFR (%)34 ± 737 ± 735 ± 8NS
 Decrease in CG (%)25 ± 727 ± 625 ± 7<0.05
 Decrease in MDRD (%)29 ± 730 ± 628 ± 7<0.05
 Serum creatinine (μmol/L)114 ± 19a133 ± 16a103 ± 11a<0.001

After nephrectomy, GFR dropped to 64 ± 7% of its pre-donation value, with a corresponding rise in serum creatinine of 35 ± 11% (both p < 0.001). Cockcroft-Gault estimated clearance fell to 76 ± 7% of its pre-donation value. Estimated clearance by the modified MDRD equation fell to 71 ± 7% of its pre-donation value. These decreases were significantly smaller than the decrease in GFR (decrease in GFR vs. decrease in Cockcroft-Gault or MDRD estimated clearances: both p < 0.001). Blood pressure increased slightly but significantly, by 4 ± 11% (systolic, p = 0.006) and 6 ± 14% (diastolic, p < 0.001).

Prediction of post-donation renal function

Individual values of pre- and post-donation GFR are given in Figure 1 (left panel), showing a strong and significant positive correlation (adjusted R2= 0.54, p < 0.001). On univariate analysis, other predictors for post-donation GFR were GFRdopa (adjusted R2= 0.35), GFRAA (adjusted R2= 0.56) and GFRmax (Figure 1, right panel, adjusted R2= 0.55; p < 0.001 for all). The numerical value of post-donation GFR was approximately equal to GFRmax divided by two, as can be derived from regression equation (Figure 1). Finally, age was negatively correlated to post-donation GFR (adjusted R2=–0.22, p < 0.001). On multivariate analysis, post-donation GFR was best predicted by the model including GFRb, GFRmax and age as independent variables (adjusted R2= 0.61, p < 0.001).

Figure 1.

Renal function before and after unilateral nephrectomy. Left panel: X-axis pre-donation baseline GFR; Y-axis: post-donation GFR. Linear regression: GFRafter  donation= 20.01 + (0.46 * GFRbefore  donation). Right panel: X-axis: GFR maximally stimulated by simultaneous infusion of low-dose dopamine and amino acids solution; Y-axis: post-donation GFR. Linear regression:GFRafter  donation= 17.54 + (0.39 * GFRmax  before  donation).

Pre-donation Cockcroft-Gault clearance predicted post-donation GFR with an adjusted R2 of 0.46 (p < 0.001) and post-donation Cockcroft-Gault clearance with an adjusted R2 of 0.85 (p < 0.001). For the modified MDRD equation, prediction of post-donation renal function was less strong. Pre-donation MDRD predicted post-donation GFR with an adjusted R2 of 0.28 (p < 0.001) and post-donation MDRD with an adjusted R2 of 0.59 (p < 0.001).

Renal function impairment after donation

After donation, renal function impairment, defined as a GFR of ≤60 mL/min/1.73 m2, occurred in 31 of 125 donors, i.e. 25%. Pre- and post-donation characteristics of our population with a break-up by post-donation GFR at/below or above 60 mL/min/1.73 m2 are given in Table 2. The group with renal function impairment after donation appeared to consist of relatively more men. However, this was not significant (χ2; p = 0.16). The group with post-donation renal function impairment was significantly older and had a higher BMI. Its GFR prior to donation was significantly lower, and serum creatinine higher compared to the group without renal function impairment. Moreover, the drop in GFR after donation was significantly larger: 38 ± 7% vs. 35 ± 8% (p < 0.01). Before donation, systolic blood pressure was slightly but significantly higher in the group with renal function impairment. This difference was no longer present after donation.

Table 2.  Donor characteristics, with the population stratified for post-donation renal function impairment, according to K/DOQI guidelines. Data are expressed as mean ± standard deviation.
 GFR ≤ 60 (mL/min/1.73 m2) (n = 31, 45% male)GFR > 60 (mL/min/1.73 m2) (n = 94, 33% male)p-Value (t-test)
  1. Nx = nephrectomy; GFRdopa= dopamine-stimulated GFR; GFRAA= amino acids stimulated GFR; GFRmax= maximal stimulated GFR by infusion of both dopamine and amino acids; NS = not significant; MDRD = modification of diet in renal disease equation; CG = renal function assessed by Cockcroft-Gault's formula.

  2. Student's t-test for donors with renal impairment vs. group without renal impairment after donation.

  3. GFR corrected for body surface area.

  4. ap < 0.001 post-donation vs. pre-donation.

  5. bp < 0.01 post-donation vs. pre-donation.

  6. cp < 0.05 post-donation vs. pre-donation (paired t-test).

Characteristics before Nx
 Age at Nx (years)59 ± 946 ± 10<0.001
 BP systolic/diastolic (mmHg)127 ± 12 / 74 ± 8122 ± 11 / 75 ± 8<0.05 / NS
 Body mass index (kg*m−2)27 ± 524 ± 3<0.01
 GFR (mL/min)99 ± 12117 ± 18<0.001
 GFR (mL/min/1.73 m2)88 ± 10109 ± 13<0.001
 GFRmax (mL/min)124 ± 15143 ± 22<0.001
 Delta GFRmax– GFR25 ± 1026 ± 13NS
 Cockcroft-Gault (mL/min)83 ± 2193 ± 20<0.05
 Simplified MDRD (mL/min)69 ± 878 ± 10<0.001
 Serum creatinine (μmol/L)90 ± 1582 ± 11<0.01
Characteristics after Nx
 BP systolic/diastolic (mmHg)128 ± 16 / 78 ± 7c126 ± 13b / 79 ± 8bNS / NS
 GFR (mL/min)61 ± 8a75 ± 10a<0.001
 GFR (mL/min/1.73 m2)54 ± 5a70 ± 7a<0.001
 Cockcroft-Gault (mL/min)60 ± 17a70 ± 15a<0.01
 Simplified MDRD (mL/min)48 ± 6a56 ± 6a<0.001
 Decrease in GFR (%)38 ± 735 ± 8<0.05
 Decrease in CG (%)28 ± 625 ± 7<0.05
 Decrease in MDRD (%)31 ± 628 ± 70.06
 Serum creatinine (μmol/L)125 ± 23a110 ± 16a<0.01

Logistic regression was performed to analyze for the independent contribution of the different factors to post-donation renal function impairment. The model with age, BMI and pre-donation GFR (included as quartiles) provided a Nagelkerke R2 of 0.53, with p < 0.001. Increasing age and BMI had odds ratios for renal function impairment of 1.14 and 1.16 per year and per kg*m−2, respectively. For decreasing quartiles of pre-donation GFR (above 124 mL/min, 112–123 mL/min, 100–111 mL/min and below 99 mL/min), the odds ratios to develop renal function impairment after donation were: 2.99, 9.87 and 24.14, respectively. Gender was not a significant risk factor for renal function impairment. Figure 2 illustrates the logistic regression model in a simplified form, containing age and GFR as factors associated with obtaining renal function impairment. To quantify the predictive power of the model, we classified the donor as having an increased risk for renal function impairment when the predicted probability from this model was at or above 50%, which was the case in 24 donors. Thus, the model correctly identified 65% of the 31 donors with post-donation renal function impairment and 96% of the 94 donors without post-donation renal function impairment. Renal reserve capacity had no added value in predicting post-donation renal function impairment.

Figure 2.

Predicted probability of renal function impairment after donation, defined as GFR ≤ 60 mL/min/1.73 m2, according to K/DOQI guidelines (10). The population is divided into quartiles of baseline GFR prior to donation. The probability is calculated by means of a logistic regression model including baseline GFR and age, with Nagelkerke R2= 0.49 and p < 0.001. The sensitivity and specificity of this model were 61% and 93%, respectively.

Follow-up data in subgroup

The GFR measurements reported here were obtained after a relatively short follow-up. To test whether the post-donation renal function values were representative for renal function after longer follow-up, we compared creatinine-values at the time of GFR measurement with those obtained during long-term follow-up. This was performed in the subgroup of 63 subjects in whom outpatient follow-up for longer than 1 year was available. Mean duration of follow-up for this group since the nephrectomy was 161 ± 154 weeks. Data are given in Table 3, showing the long-term follow-up data and the values obtained during the GFR measurements (i.e. 57 ± 16 days after nephrectomy). Both serum creatinine and the Cockcroft and MDRD-estimated clearances remained unaltered during the observation period.

Table 3.  Characteristics of the 63 living kidney donors with long-term follow-up data of at least 1 year after one-sided nephrectomy. Data are expressed as mean ± standard deviation.
 (n = 63, 41% male)
  1. Nx = nephrectomy; GFR = glomerular filtration rate assessed as iothalamate clearance; MDRD = modification of diet in renal disease equation.

  2. GFR corrected for body surface area.

  3. *p < 0.001 characteristics after donation vs. pre-donation values (paired t-test).

Characteristics before Nx
 GFR (mL/min)113 ± 21
 GFR (mL/min/1.73 m2)105 ± 16
 Cockcroft-Gault (mL/min)89 ± 21
 Simplified MDRD (mL/min)76 ± 10
 Serum creatinine (μmol/L)85 ± 13
Characteristics after Nx
 GFR (mL/min)72 ± 13*
 GFR (mL/min/1.73 m2)67 ± 11*
 Cockcroft-Gault (mL/min)67 ± 17*
 Simplified MDRD (mL/min)55 ± 9*
 Serum creatinine (μmol/L)114 ± 21*
≥1 year after Nx
 Cockcroft-Gault (mL/min)67 ± 16*
 Simplified MDRD (mL/min)54 ± 10*
 Serum creatinine (μmol/L)115 ± 23*


In this study, pre-donation GFR significantly predicted post-donation renal function, accounting for 54% of the variance in post-donation GFR. The predictive value for post-donation renal function increased to 61% of the variance by also taking age and maximally stimulated GFR into account. Post-donation renal function impairment occurred in 25% of the donors; risk factors were older age, higher BMI and particularly a lower pre-donation GFR to start with.

Pharmacological stimulation of the kidney by dopamine and amino acids is known to elicit a considerable rise in GFR. This rise in GFR is mediated by a rise in renal blood flow as well as altered glomerular pressure, and is often referred to as renal filtration reserve capacity (6). Both GFRAA and GFRmax strongly correlated with post-donation GFR, and the numerical value of post-donation GFR was approximately equal to GFRmax divided by two. However, the predictive power—as estimated from the adjusted R2 of the regression equations—for GFRAA(R2= 0.56) and GFRmax(R2= 0.55) hardly exceeded that of unstimulated GFR (R2= 0.54). The prediction of post-donation GFR from pre-donation GFR could somewhat be improved by adding both age and GFRmax to the multivariate model, but the difference was small, and accordingly, the added value of stimulated GFR appears to be limited.

We measured GFR by the clearance of 125I-iothalamate, i.e. the gold standard for measurement of renal function. As many centers rely on creatinine-based renal function assessment, we also tested the predictive effect of both pre-donation Cockcroft-Gault and modified MDRD estimated clearances for post-donation GFR, and for post-donation estimated clearances. Significant correlations were found here. In particular, pre-donation Cockcroft-Gault clearance provided a strong predictor for post-donation Cockcroft-Gault clearance. However, it should be considered that the renal function estimate by the Cockcroft-Gault or MDRD equation is subject of debate. An additional finding of clinical relevance was that both the Cockcroft-Gault and the modified MDRD estimated clearances significantly and considerably underestimated the drop in renal function after donation. This is most likely due to the systematic error in these equations that underestimate true renal function particularly in the normal and high range, but not in the lower range of GFR (11). The unreliability of renal function equations to estimate renal function in the normal and high range has been emphasized in several recent papers (12–15). The current data illustrate its impact in the context of kidney donation.

For the prospective donor, a main purpose of screening is to provide an individual assessment of the risks conferred by living kidney donation. We sought to identify the risk factors for renal function impairment in the individual. Thirty-one donors had post-donation renal function impairment. These subjects were older, had higher systolic blood pressure, a higher BMI, were more often male, and in particular, had a lower pre-donation GFR. On logistic regression, the significant risk factors were lower GFR, higher BMI and older age, without an independent contribution of gender. It may seem intuitively obvious that subjects with a lower GFR to start with have a greater risk to end up below a certain cut-off. However, it may be relevant that the donors who ended up with renal function impairment had a significantly larger drop in renal function after donation. This suggests that a modest but specific renal vulnerability may be involved also in these subjects, possibly related to their higher age and/or BMI. At any rate, renal reserve capacity before donation did not discriminate between subjects that ended up with renal function impairment and those who did not—and thus is not a suitable tool to uncover such a possible vulnerability to post-donation renal function loss.

We defined renal function impairment as a GFR of ≤60 mL/min/1.73 m2, the DOQI guideline cut-off for renal function impairment of moderate severity (or Chronic Kidney Disease stage 3) (10). We adopted this cut-off to comply with the common frame of reference. It should be noted however, that the clinical significance of a GFR ≤60 mL/min/1.73 m2 in a single healthy kidney may not be similar to that of the same level of renal function in a subject with two diseased kidneys. Therefore, the cut-off used here may not be applicable in this population of healthy kidney donors. The clinical significance would need to be ascertained in prospective studies, addressing its prognostic value for both future renal function loss and for cardiovascular morbidity. However, the same uncertainty would apply to any other cut-off level as well.

A limitation of our study is the relatively brief duration of follow-up after kidney donation. Compensatory mechanisms like renal hypertrophy may not be fully achieved at 2 months after donation (16,17). Therefore, and to study longer-term effects of donation on renal function, we tested whether serum creatinine was stable between the short-term and long-term measurements in those donors in whom a long-time serum creatinine was available. Whereas mean serum creatinine was stable over the observation period in this subset of donors, it would be of interest to have a more extensive prospective long-term follow-up. Long-term studies have shown that, in general, kidney donation is a safe procedure with a favorable long-term renal outcome, and only a minority of donors develop clinically significant renal damage. However, these studies are hampered by the lack of data on more precise renal function assessment, since only data on serum creatinine are given and no GFR measurements. Although it is unknown whether donors with a post-donation GFR of less than 60 mL/min/1.73 m2 have a higher risk to progress to renal insufficiency, we might have identified a category at risk for decline in renal function. Further follow-up of this cohort is necessary to provide a better assessment of long-term risk in these donors and to see whether preventive measures, such as rigorous blood pressure control are needed.

We found a slight yet significant rise in systolic and diastolic blood pressure at 2 months after donation. Its significance for long-term renal and cardiovascular prognosis in this selected population is unsure. It has become increasingly clear that slight changes in blood pressure, within the range formerly considered as normal, can have impact on long-term cardiovascular and renal risk (18,19). Whether this also applies to a population that has been pre-selected to donate a kidney, and thus has a more favorable risk profile than average (20), has not been established. Previous studies have shown contradictory results on this matter, with reports varying from a drop in blood pressure short-term after donation, to either a similar prevalence in hypertension or a higher prevalence of hypertension long-term after donation compared to the ‘normal’ population (2,4,21,22). Of note, both pre- and post-donation systolic blood pressure in our population would be considered pre-hypertensive by current standards (19), which underlines the need for proper long-term follow-up concerning both renal function and cardiovascular morbidity.

In the United States and some countries in Europe, there is no formal follow-up program for living kidney donors (23). We believe that especially for the donors that are at risk for renal function impairment after donation, a follow-up program would be beneficial. Especially in an era, where organ shortage prompts for acceptance of donors that are older, have higher BMI and/or in whom blood pressure is not optimal, the importance of pre-donation risk identification and post-donation follow-up becomes apparent.

In conclusion, pre-donation GFR by iothalamate clearance provides a reasonable prediction of renal function after donation. The predictive power is slightly enhanced by taking into account maximally stimulated GFR and age. Pre-donation GFR and age, together with BMI, are predictors of the risk for renal function impairment after donation. Stimulated GFR did not add to the predictive power in this study. Accurate renal function measurement before donation allows a relatively reliable estimate of post-donation renal function early after donation. However, further studies with longer follow-up are needed to assess the prognostic value of pre-donation characteristics on long-term, and to prospectively identify subjects with a higher risk for renal function loss.