Urinary neutrophil gelatinase-associated lipocalin (uNGAL) is known to predict the prolonged delayed graft function after kidney transplantation. We examined the relation of uNGAL with histological findings of acute tubular injury (ATI). Analyses were made in biopsies taken at 6 weeks, 3 months, and 6 months after kidney transplantation. uNGAL was measured in the spot urines, normalized to urinary creatinine excretion, and correlated to biopsy findings and clinical, laboratory, and demographic variables. Controls included healthy individuals, individuals after kidney donation and ICU patients with acute kidney failure. Renal transplant recipients without ATI did not display elevated uNGAL levels compared to the healthy controls. Transplant patients with ATI had a higher uNGAL excretion at 6 weeks than patients without ATI (27,435 versus 13,605 ng/g;
Acute kidney injury early after transplantation may arise from the donor’s condition, prolonged cold ischemia time, and early posttransplant injuries like rejection and drug toxicity. Immediate clinical correlates of this injury may be delayed graft function or a less-than-optimal glomerular filtration rate. However, acute kidney injury carries adverse long-term consequences as patients with acute tubular injury early posttransplantation have an inferior long-term allograft function [
Traditionally, serum parameters as creatinine and urine output are used to monitor kidney function in transplanted patients. However, a rise in serum creatinine already implies a significant amount of kidney damage, which limits its ability to detect impaired graft function at early stages. Recently, novel biomarkers such as Kim-1, IL-18, and neutrophil gelatinase-associated lipocalin (NGAL) have been proven useful in detecting nontransplant acute kidney damage [
Acute kidney injury primarily affects renal tubules and NGAL is produced in injured tubular epithelial cells. Therefore, we addressed the question whether urinary NGAL correlates with histologically confirmed acute tubular injury using protocol biopsies taken within the first six months after transplantation.
The study population consisted of 140 adult kidney transplant recipients randomly chosen from our protocol biopsy archive (
Demographic and clinical data of patients with and without ATI.
ATI | ATI+ | All patients |
|
|
---|---|---|---|---|
|
|
|
||
Recipient | ||||
Age (years) | 49 ± 12 (49,20–69) | 51 ± 12 (52,25–73) | 51 ± 12 (50,20–73) | |
Gender (% m/f) | 65,7/34,3 | 61,9/38,1 | 62,9/37,1 | |
Retransplanted pat. (%) | 8,6 | 10,5 | 10,0 | |
Preformed antibodies neg. (%) | 94,3 | 93,3 | 93,5 | |
Lowest s-creatinine within the first 6 weeks of tx | 160 ± 83 | 182 ± 87 | 176 ± 86 | |
Hypercholesteroemia before tx (%) | 31,4 | 37,1 | 35,7 | |
Arterial hypertension before tx (%) | 100 | 96,2 | 97,1 | |
Donor | ||||
Age (years) | 47 ± 15 (50,7−70) | 51 ± 13 (52,16–77) | 50 ± 14 (51,7–77) | |
Gender (% m/f) | 60/40 | 53,3/46,7 | 55/45 | |
Type | ||||
Deceased (%) | 82,9 | 95,2 | 92,1 | 0.029 |
Living (%) | 17,1 | 4,8 | 7,9 | |
CMV IgG positive (%) | 60 | 53,8 | 55,4 | |
s-creatinine ( |
95 ± 36 | 93 ± 44 | 93 ± 42 | |
Graft factors at tx | ||||
Cold ischemia time (h) | 14 ± 7 | 17 ± 7 | 17 ± 7 | 0.032 |
Mean number of HLA mismatches | 2,26 ± 1,6 | 2,4 ± 1,72 | 2,36 ± 1,68 | |
Initial function of the graft (%) | 68,6 | 28,6 | 38,6 | 0.000 |
Need of dialysis following tx (%) | 31,4 | 71,4 | 62,1 | 0.000 |
Immunosuppressive regimen (%) | ||||
CyA-MMF-Steroids-IL2 | 37,1 | 31 | 32,6 | |
CyA-MMF-Steroids-ATG | 0 | 1 | 0,7 | |
CyA-MMF-Steroids-no induction | 0 | 3 | 2,2 | |
CyA-MMF-IL2 | 14,3 | 31 | 26,7 | |
CyA-MMF-ATG | 0 | 1 | 0,7 | |
Other, IL2 | 28,6 | 14 | 17,8 | |
Other, ATG | 11,4 | 7 | 8,1 | |
Other, no induction | 8,6 | 12 | 11,1 |
ATI−: patients without ATI; ATI+: patients with ATI; tx: transplantation; median and range of values in brackets; CyA: cyclosporine A; MMF: mycophenolate mofetil; ATG: antithymocyte globulin; IL2: interleukin 2 receptor antagonist.
Urinary NGAL was measured in the spot urine using the Quantikine human lipocalin-2/NGAL Immunoassay (Cat.no. DLCN 20 by R&D Systems), according to the manufacturer’s protocol. Urinary NGAL excretion was normalized to urinary creatinine excretion to correct for differences in NGAL due to urine dilution. Urinary NGAL excretion is presented as the amount of urinary NGAL in ng per g of urine creatinine.
Descriptive data were reported as medians with ranges. Comparisons of categorical data between groups were performed with Fisher’s exact test and the chi-square test for two or more samples. Numerical data were compared with the Kruskal-Wallis test and the Mann-Whitney test. Linear regression analyses were performed with log transformed NGAL values as the dependent variable and the demographical, clinical, and laboratory factors that had been significantly different in univariate analyses. Variables from the univariate analyses were separately examined for the samples at 6 weeks and 6 months using the backward selection. SPSS statistical software package version 18.0 (SPSS Inc., Chicago, IL) was used for statistical analysis. Statistical significance was assumed for
First we compared the uNGAL excretion levels of renal allograft recipients without ATI to different control groups. There was no significant difference in uNGAL excretion in transplanted patients versus healthy individuals with 1 kidney (i.e., kidney donors) and healthy individuals with 2 kidneys (Figure
Urinary NGAL/creatinine ratios in transplanted patients versus other groups. Shown are the urinary NGAL/creatinine levels of the two control groups (healthy individuals with 2 (
(a) NGAL excretion in stable patients without ATI at 6 weeks, 3 months, and 6 months after transplantation. Shown is the urinary NGAL/creatinine ratio at 6 weeks (
Next, we tested the sensitivity of urinary NGAL as a biomarker to indicate ATI in transplanted patients. At six weeks, patients with ATI had urinary NGAL values that were two-fold higher than the values from patients without ATI in the biopsies, yet there was a considerable overlap of values between the two groups. At 6 months, no differences between patients with and without ATI were apparent (Figure
(a) Comparison of urinary NGAL/creatinine ratios in patients with and without ATI after kidney transplantation. Patients with ATI 6 weeks (
(a) Urinary NGAL/creatinine ratio depending on the frequency of ATI. Shown is the urinary NGAL/creatinine ratio depending on whether patients had 1 time (
In an explorative analysis, we correlated NGAL levels to allograft damage (e.g., acute tubular injury, delayed graft function, donor age) and to additional clinical, laboratory, and demographic variables (for a complete list see Supplementary Materials available online at doi:10.1155/2012/563404). The main findings of the univariate analysis are depicted in Table
Urinary NGAL levels (NGAL/creatinine; (ng/g)), and associations with various clinical, laboratory and demographic variables. Results are separately reported for the 6-week and 6-month samples. The remaining variables which were tested without significant differences are reported in the Supplementary Materials.
6 weeks |
|
6 months |
|
|||
---|---|---|---|---|---|---|
Male | Female | Male | Female | |||
Donor gender | 18,875 |
23,910 |
0.31 | 12,264 |
21,785 |
0.071 |
Recipient gender | 18,242 |
37,855 |
0.011 | 11,491 |
32,613 |
0.000 |
GFRa |
|
0.066 |
|
0.006 | ||
| ||||||
6 weeks |
|
6 months |
|
|||
Not present | Present | Not present | Present | |||
| ||||||
Delayed graft function | 18,875 ( |
21,315 |
0.218 | 17,554 |
19,354 |
0.866 |
Isometric vacuolizationb | 19,187 ( |
31,224 |
0.031 | 22,436 |
9,480 |
0.008 |
cGrade |
19,031 ( |
39,575 |
0.033 | 14,218 |
21,115 |
0.583 |
Urinary tract infection at the time of sampling | 17,901 ( |
50,548 |
0.000 | 13,760 |
64,889 |
0.014 |
Panel reactive antibodies at transplantation |
22,952 |
11,696 |
0.019 | 17,194 |
33,929 |
0.342 |
Blood transfusion before transplantation | 19,530 |
25,876 |
0.449 | 13,286 |
30,322 |
0.022 |
MMF in the initial immunosuppression | 27,842 |
19,187 |
0.245 | 24,208 |
17,875 |
0.019 |
Coronary heart disease before transplantation | 16,502 |
21,822 |
0.116 | 19,354 |
8,969 |
0.043 |
aGFR calculated according to the Cockroft and Gault formula at the time of sampling, baccording to the BANFF classification, at the time of sampling.
Isometric vacuolization and prevalence of chronic graft changes (cGrade>0) according to the current BANFF classification at 6 weeks after transplantation were associated with significantly higher uNGAL levels. However, after 6 months, uNGAL ratios were lower in patients with isometric vacuolization. In addition, patients with urinary tract infection at the time of sampling had significantly higher uNGAL ratios at 6 weeks as well as 6 months (Table
Next, linear regression analyses were performed with log transformed NGAL values to determine the contribution of the identified variables of the explorative univariate analysis to the observed NGAL levels. Significant variables from the univariate analyses were separately examined for the samples at 6 weeks and 6 months using the backward selection. For the 6-week samples (Table
Linear regression analysis of significant correlates to uNGAL excretion.
Samples from 6 weeks.
Model component | Regression coefficient | SE |
|
Tolerance |
---|---|---|---|---|
Constant | 5.029 | 0.224 | <0.001 | |
Urinary tract infection | 0.519 | 0.110 | <0.001 | 0.980 |
Isometric vacuolization | 0.272 | 0.091 | 0.004 | 1.0 |
Samples from 6 months.
Model component | Regression coefficient | SE |
|
Tolerance |
---|---|---|---|---|
Constant | 4.364 | 0.338 | <0.001 | |
Urinary tract infection | 0.335 | 0.151 | 0.03 | 0.967 |
Female recipient | 0.427 | 0.097 | <0.001 | 0.970 |
Isometric vacuolization | −0.293 | 0.099 | 0.004 | 0.994 |
SE: standard error of the regression coefficient; tolerance indicates collinearity between the variables of mode I (1.0 = no collinearity). Calculated GFR’s are based on the serum creatinine and Cockroft and Gault formula and square roots of GFR values are used for the analysis.
Since the univariate analysis indicated a relation between donor and recipient gender and NGAL levels, the possible gender combinations are additionally presented (Table
Correlation of uNGAL ratios to the different donor/recipient gender combinations.
Gender combinations |
NGAL in 6-week samplesa (ng/g) | NGAL in 6-month samplesb (ng/g) |
---|---|---|
Male/male | 14,259 | 5,670 |
|
|
|
Male/female | 37,717c | 32,893e |
|
|
|
Female/female | 37,993d | 24,050e |
|
|
|
Female/male | 21,654 | 18,454f |
|
|
Urinary NGAL/creatinine ratios at 6 weeks and 6 months after transplantation according to the four combinations of donor and recipient gender. At both time points, differences between the four combinations were significant (a
Acute tubular injury (ATI) is a frequent finding in biopsies of renal allograft recipients and is linked to an inferior long-term allograft function [
NGAL can be produced locally in the kidney by injured tubular cells and can be secreted by activated neutrophils/macrophages or inflamed vasculature [
We established the baseline uNGAL excretion in our transplanted patient cohort without findings of ATI. uNGAL levels in this group were comparable to healthy controls and remained stable 6 weeks, 3 months, and 6 months after transplantation. As proper controls for patients with a renal allograft, we collected sample material from healthy individuals with one kidney (donors with normal kidney function), in addition to healthy controls with 2 kidneys. Our observation of a tendency towards a higher NGAL excretion in healthy individuals with 2 kidneys versus 1 kidney may point to a confounding effect by the kidney mass. This finding should be confirmed with a larger cohort.
In patients with ATI, median uNGAL ratios were significantly higher at 6 weeks after transplantation than in patients without signs of ATI. While these findings may indicate a correlation of uNGAL excretion to ATI in patients early after transplantation, we could not demonstrate increased uNGAL levels in patients with ATI at 6 months after transplantation. On the contrary, median uNGAL levels in patients with ATI were even lower at 6 months compared to the controls without ATI. Further analyses showed no differences of NGAL excretion depending on the frequency of ATI or the distinction between diffuse versus focal ATI. NGAL levels were weakly inversely correlated to the eGFR; however, no link could be established between uNGAL levels and acute rises in serum creatinine at the time of ATI diagnosis. Urinary NGAL levels have been shown to correlate with histopathological alterations in IgAN patients [
Plasma NGAL measurements can be influenced by numerous clinical variables, such as hypertension [
The limitations to our study are that we present a single-centre study and that urine samples were stored at −80°C for more than a year and some protein degradation may have occurred during the freeze-thaw cycle thus, this study should be confirmed with immediate uNGAL measurements in fresh urine samples. In addition we do not have information about the uNGAL levels in our patients prior to transplantation or in the donors and some patients did not provide urine samples at all time points when protocol biopsies were taken. Thus, further studies are necessary to evaluate the diagnostic value of uNGAL measurements in patients after kidney transplantation.
In conclusion, even though NGAL may be useful in classifying patients with established AKI or glomerulonephritis in native kidneys or may be a predictor for delayed graft function and recovery early after kidney transplantation [
The authors appreciate the help of K. Worthmann and S. Zachura in the collection of sample materials. Abstracts were accepted at the ASN Renal Week and American Transplant Congress 2010. J. K. Kaufeld performed study, wrote the paper, and collected data, W. Gwinner designed the study and analyzed data, I. S. cheffner analyzed data, H. G. Haller designed study, M. Schiffer designed study.