Diagnostic techniques in the work-up of renal allograft dysfunction--an update

Systematic Review of the Management of Ureteroarterial Fistulas After Ileal Conduit Urinary Diversion

OBJECTIVE. No studies or guidelines exist to direct management of ureteroarterial fistula (UAF) after ileal conduit urinary diversion in which the possible risks and complications associated with stent-graft infection from the conduit flora must be reconciled with those of open surgical repair. This study seeks to characterize the clinical presentation, pathogenesis, and optimal diagnostic and therapeutic management of this entity through a systematic review of the literature. MATERIALS AND METHODS. A systematic search of the English-language literature using the PubMed, Scopus, and ScienceDirect databases was performed: 264 abstracts were identified. From those abstracts, 32 studies comprising 40 patients with 43 UAFs were selected for analysis. Data points including demographics, clinical presentation, UAF specifications, procedural details, postprocedural complications, and clinical outcomes were reviewed. RESULTS. Predisposing factors included female sex, chronic ureteral stent placement, and past surgical intervention and irradiation for pelvic malignancy. Fistulization was overwhelmingly unilateral (95.0% of patients) and included the common iliac artery (90.7% of UAFs). Combined endovascular and endoureteral modalities presented similar outcomes compared with surgical approaches in terms of UAF-related mortality (7.1% vs 13.3%, respectively) and complication rates (28.6% vs 26.7%) during a similar median follow-up period (9.5 vs 14.0 months). Endovascular stent-graft infections were present in 14.3% of cases and represented a leading indication for reintervention after endovascular management (50.0%). CONCLUSION. Short- and intermediate-term outcomes of combined endovascular and endoureteral techniques compare favorably with those of surgical approaches in the treatment of UAF after ileal conduit urinary diversion. Although there is a relatively low stent-graft infection rate, close follow-up within the first year after the procedure is required given the propensity of complications to develop during this window. The use of postprocedural antibiotics is uncertain but is likely prudent.

Functional MRI in transplanted kidneys

Renal transplantation is the therapy of choice for patients with end-stage renal diseases. Improvement of immunosuppressive therapy has significantly increased the half-life of renal allografts over the past decade. Nevertheless, complications can still arise. An early detection of allograft dysfunction is mandatory for a good outcome. New advances in magnetic resonance imaging (MRI) have enabled the noninvasive assessment of different functional renal parameters in addition to anatomic imaging. Most of these techniques were widely tested on renal allografts in past decades and a lot of clinical data are available. The following review summarizes the comprehensive, functional MRI techniques for the noninvasive assessment of renal allograft function and highlights their potential for the investigations of different etiologies of graft dysfunction.

Immunologic monitoring in kidney transplant recipients

Transplant biopsy has always been the gold standard for assessing the immune response to a kidney allograft (Chandraker A: Diagnostic techniques in the work-up of renal allograft dysfunction-an update. Curr Opin Nephrol Hypertens 8:723-728, 1999). A biopsy is not without risk and is unable to predict rejection and is only diagnostic once rejection has already occurred. However, in the past two decades, we have seen an expansion in assays that can potentially put an end to the "drug level" era, which until now has been one of the few tools available to clinicians for monitoring the immune response. A better understanding of the mechanisms of rejection and tolerance, and technological advances has led to the development of new noninvasive methods to monitor the immune response. In this article, we discuss these new methods and their potential uses in renal transplant recipients.

Relation of resistive and pulsatility indices with graft function after renal transplant

OBJECTIVES

There are conflicting data regarding the use of some measured indices by Doppler ultrasound such as the resistive index and the pulsatility index in predicting renal allograft dysfunction. This study sought to evaluate the association of early postoperative Doppler indices and 3-month serum creatinine levels in renal transplant recipients.

MATERIALS AND METHODS

During a 1-year period, all patients who underwent renal transplant at our hospital were recruited into a prospective study. Doppler ultrasound was performed on all patients 6 days and 3 months after the transplant and the resistive index and the pulsatility index were calculated for each patient. Then, the association between these indices and 3-month outcomes of patients were investigated.

RESULTS

Thirty-eight patients including 21 men (mean age, 36.6 ± 13.1 y) were evaluated. There was a positive correlation between the resistive index and the pulsatility index at 6 days after transplant and the serum creatinine measured at the same day (P < .001 and r=0.570 for resistive index; P < .001 and r=0.547 for pulsatility index). There was also a positive correlation between the pulsatility index and the resistive index at 6 days after transplant and 3-month serum creatinine level (P = .009 and r=0.420 for resistive index; P = .009 and r=0.417 for pulsatility index). There were negative correlations between the resistive index and the pulsatility index on the sixth day after surgery and creatinine clearance measured at 6 days and 3 months after transplant.

CONCLUSIONS

This study reveals a strong-to-medium correlation between the resistive index and the pulsatility index, serum creatinine level, measured 6 days after transplant.

Color Doppler ultrasound-guided transducer compression of post biopsy bleeding of kidney transplants

BACKGROUND

Color Doppler ultrasound (CDUS) has a potential of early detection of post biopsy bleeding. We describe CDUS guidance in planning, acquisition, and, in the case of bleeding, compression of the needle tract in biopsy procedures of kidney transplants.

METHODS

Eighty-three kidney transplant biopsy procedures performed on clinical indication were performed in 71 adult patients, 25 women, mean age 51 years, using CDUS and 18-G biopsy needles. Bleeding needle tracts were compressed using CDUS guidance.

RESULTS

CDUS immediately detected blood leakage and facilitated compression of the bleeding needle tract in 34 (41%) of the 83 procedures, including 53 (25%) of 215 needle passes. In 34 bleeding procedures, the mean duration of the longest bleeding time after a needle pass was 124 seconds (median, 20 seconds; range, 3-1440 seconds). In 12 of these procedures (35%), the bleeding time was 30 seconds or more. In six procedures (18%), a bleeding of 120 seconds or more was observed. Complications included seven small hematomas. Five hematomas developed in procedures where the longest duration CD bleeding was 120 seconds or more.

CONCLUSIONS

CDUS detects bleeding and facilitates direct transducer compression of the needle tract in a substantial portion of biopsy procedures of kidney transplants. Only minor hematomas occurred.

Opportunistic posttransplantation virus infections in renal transplant recipients

BACKGROUND

Opportunistic virus infection is one of the most common complications in renal transplant (RT) recipients. Cytomegalovirus (CMV) and BK virus (BKV) are important pathogens and each of these infections affects the other. In contrast, there is only limited information on JC virus (JCV) infection and its relation to CMV infection in RT recipients. This prospective study investigated the rates of JCV and CMV infections and their risk factors and correlations.

METHODS

We studied 52 RT recipients. JCV and CMV were detected using nested qualitative polymerase chain reaction assays of urine. The clinical characteristics of JCV and CMV infection were compared and risk factors analyzed with the use of binary logistic regression.

RESULTS

JCV and CMV were detected in 40.4% and 34.6% of the RT recipients, respectively. Cyclosporine (CsA) was a risk factor for both JCV and CMV infection (odds ratio [OR] 7.187; P=.002; OR 4.182; P=.021); CMV infection was a risk factor for JCV infection (OR 3.900; P=.039).

CONCLUSIONS

JCV and CMV infections are common in RT recipients. CsA is a risk factor for both JCV and CMV infection. JCV infection is related to CMV infection.

Quantitative evaluation of acute renal transplant dysfunction with low-dose three-dimensional MR renography

PURPOSE

To assess prospectively the ability of quantitative low-dose three-dimensional magnetic resonance (MR) renography to help identify the cause of acute graft dysfunction.

MATERIALS AND METHODS

This HIPAA-compliant study was approved by the institutional review board, and written informed consent was obtained. Between December 2001 and May 2009, sixty patients with transplanted kidneys (41 men and 19 women; mean age, 49 years; age range, 22-71 years) were included. Thirty-one patients had normal function and 29 had acute dysfunction due to acute rejection (n = 12), acute tubular necrosis (ATN) (n = 8), chronic rejection (n = 6), or drug toxicity (n = 3). MR renography was performed at 1.5 T with three-dimensional gradient-echo imaging. With use of a multicompartment renal model, the glomerular filtration rate (GFR) and the mean transit time (MTT) of the tracer for the vascular compartment (MTT(A)), the tubular compartment (MTT(T)), and the collecting system compartment (MTT(C)) were calculated. Also derived was MTT for the whole kidney (MTT(K) = MTT(A) + MTT(T) + MTT(C)) and fractional MTT of each compartment (MTT(A/K) = MTT(A)/MTT(K), MTT(T/K) = MTT(T)/MTT(K), MTT(C/K) = MTT(C)/MTT(K)). These parameters were compared in patients in the different study groups. Statistical analysis was performed by using analysis of covariance.

RESULTS

There were significant differences in GFR and MTT(K) between the acute dysfunction group (36.4 mL/min ± 20.8 [standard deviation] and 177.1 seconds ± 46.8, respectively) and the normal function group (65.9 mL/min ± 27.6 and 140.5 seconds ± 51.8, respectively) (P < .001 and P = .004). The MTT(A/K) was significantly higher in the acute rejection group (mean, 12.7% ± 2.9) than in the normal function group (mean, 8.3% ± 2.2; P < .001) or in the ATN group (mean, 7.1% ± 1.4; P < .001). The MTT(T/K) was significantly higher in the ATN group (mean, 83.2% ± 9.2) than in the normal function group (mean, 72.4% ± 10.2; P = .031) or in the acute rejection group (mean, 69.2% ± 6.1; P = .003).

CONCLUSION

Low-dose MR renography analyzed by using a multicompartmental tracer kinetic renal model may help to differentiate noninvasively between acute rejection and ATN after kidney transplantation.

Interventional management of renal transplant arteriovenous fistula

Percutaneous needle biopsy has become an indispensable tool for the evaluation and management of patients with renal allograft dysfunction. But this invasive procedure is not without risk. Vascular injury in the form of arteriovenous fistula, pseudoaneurysm, or arteriocalyceal fistula may result in symptoms that require percutaneous endovascular intervention. In this article, the occurrence, detection, and treatment of biopsy-related renal transplant injury are described.

Urine proteomic profiling for biomarkers of acute renal transplant rejection

Acute allograft rejection is a serious impediment to long-term success in renal transplantation. Early detection of rejection is crucial for treatment of rejection, and can help avoid long-term effects such as chronic rejection or loss of the transplanted organ. The current diagnostic paradigm is a combination of clinical presentation, biochemical measurements (serum creatinine), and needle biopsy. There are significant efforts underway to find alternate biomarkers for early detection of acute rejection, including protein profiling of urine by mass spectrometry. One approach for protein profiling is to use affinity mass spectrometry - we describe a method for this using ProteinChips and SELDI-TOF mass spectrometry.

Noninvasive detection of acute and chronic injuries in human renal transplant by elevation of multiple cytokines/chemokines in urine

BACKGROUND

Injuries in kidney transplant is currently diagnosed by needle biopsy. A noninvasive test that sensitively detects these injuries would benefit the patients.

METHODS

Urine samples were collected from healthy controls and kidney transplant recipients. Urine samples were screened first with an antibody array consisting of 120 chemokines and cytokines and then with a multiplex beads assay. Representative parameters, including macrophage inflammatory protein-1Delta, osteoprotegerin, monokine induced by interferon-gamma (IFN), and IFN-gamma-induced protein of 10 kDa, were simultaneously determined by a quadruplex assay in urine samples from 84 patients with renal allograft injury, 29 patients with stable graft function, and 19 healthy individuals.

RESULTS

Twenty-three cytokines/chemokines were found to be elevated in urine samples of patients with acute rejection by the antibody array. The second round of screening confirmed that 11 of the 23 parameters were elevated in the patients but not in the healthy controls. Induced protein of 10 kDa and monokine induced by IFN-gamma were significantly elevated in urine samples of patients with acute renal injury, and macrophage inflammatory protein-1Delta and osteoprotegerin were significantly elevated in patients with both acute and chronic renal injuries. The combination of the four parameters had a high positive detection rate (97.6%) for renal transplant injury and could differentiate between acute and chronic injury.

CONCLUSION

These results might indicate that the present multiplex assay provides a basis to establish a noninvasive method for the diagnosis and monitoring of renal transplant injury.

Prediction of Acute Renal Allograft Rejection by Urinary Monokine Induced by IFN-γ (MIG)

Early diagnosis of acute allograft rejection (AR) is still decisive for long-term renal allograft survival. The aim of this study was to define the role of the chemokine monokine induced by IFN-gamma (MIG) (CXCL9) and IFN-gamma-inducible protein 10 (IP-10) (CXCL10) as early markers of AR in renal transplantation (NTX). In a prospective study, 69 de novo renal transplant recipients were monitored and urine samples were collected after NTX for a median of 29 d. In pH-adjusted urine, MIG and IP-10 were determined by modified ELISA. AR was clinically diagnosed in 15 of 69 recipients and confirmed by biopsy in 14 of 15 AR patients (Banff classification). Corresponding to CXCR3-positive infiltrates in renal tissue, urinary MIG was elevated in 14 of 15 AR patients with a median of 2809 pg/ml (quartile 25% and 75% = 870 and 13,000; n = 15), being significantly (P < 0.0001) different from both nonrejecting allograft patients (NO-AR) (median, 25%, and 75%: 96, 1.0, and 161, n = 54) and healthy controls (median, 25%, and 75%: 144, 19, and 208, n = 13). Urinary MIG predicted AR with a sensitivity of 93% and a specificity of 89%. In AR and NO-AR groups, MIG values correlated well with IP-10 (P < 0.001). MIG values indicated both imminent rejection and response to successful antirejection therapy. MIG was not related to intercurrent infections or other causes for impairment of renal function. In a multivariate analysis, MIG correlated best (P < 0.001) with AR from all AR-associated parameters. In conclusion, urinary MIG serves as a very sensitive and specific predictor for AR, mirrors response to antirejection therapy, and thus may contribute to improved long-term renal allograft survival.

Quantitative gene expression analysis of fractalkine using laser microdissection in biopsies from kidney allografts with acute rejection

Percutaneous biopsy of the kidney remains the gold standard to establish a diagnosis in renal diseases. Only semiquantitative assessments of gene expression on biopsies have been possible so far. We studied gene expression of the chemokine fractalkine (FKN) in 12 biopsies from laser microdissected kidney allografts that showed histologic signs of acute rejection and 10 controls. As quantified by real-time PCR, the relative tubular FKN expression increased from 1.0 [0.81 to 2.95] (median [range]) in controls to 12.44 [0.90 to 191.0] in acute rejection (P < .01); glomerular FKN expression from 1.3 [0.07 to 27.44] to 12.22 [1.32 to 50.23] (P < .05); and vascular expression, from 0.72 [0.37 to 5.11] to 7.07 [1.19 to 73.49] (P < .01). Furthermore, there was a trend toward higher glomerular FKN expression among patients with more severe rejection. Our results suggest a role of FKN in acute renal allograft rejection.

CC-chemokine RANTES is increased in serum and urine in the early post-transplantation period of human renal allograft recipients

BACKGROUND

The chemokine RANTES is a potent chemoattractant for T cells and monocytes that has been shown to enhance inflammation. The aim of our study was to investigate whether RANTES is upregulated within the early post-transplantation period that may influence short-time allograft function rate.

METHODS

Serum and urine samples from transplanted renal allograft recipients (n = 17) were obtained from specimens taken for diagnostic reasons. Four patients developed biopsy-proven rejection episodes within the first month. Time course of RANTES was studied within the first 12 days after renal transplantation using ELISA technique. Data were tested for significances between patients with rejection and without rejection, compared to healthy volunteers as controls, and correlated with clinical data.

RESULTS

In the control group RANTES concentration was 37.2 +/- 2.7 ng/ml (serum) and 8.1 +/- 1.3 pg/ml (urine), respectively. In transplanted recipients serum RANTES was significantly upregulated up to 132 +/- 28 ng/ml on day 1 after transplantation and remained elevated within the first 12 days (n = 17). Time course of urine RANTES demonstrated elevated concentrations with 754 +/- 115 pg/ml on day 1 followed by an continuous decrease to 22.3 +/- 7 pg/ml on day 12 (n = 17). No significant differences could be detected between patients with rejection and without rejection episodes.

CONCLUSIONS

In contrast to data of other urinary marker molecules (like IL-6), there are no significant differences between the rejection and non-rejection group. RANTES is therefore not suitable for early detection of rejection. Nevertheless, serum and urine RANTES concentrations were highly elevated in freshly transplanted renal allograft recipients reflecting an activated immune system.

Serum levels of macrophage-colony stimulating factor (M-CSF): a marker of kidney allograft rejection

BACKGROUND

Macrophage-colony stimulating factor (M-CSF) is the principal factor for survival of monocytes and macrophages that play an important role in allograft rejection. We studied M-CSF serum levels during successful renal transplantation and acute graft rejection.

METHODS

A total of 114 kidney allograft recipients were assessed for M-CSF levels by enzyme-linked immunosorbent assay (ELISA).

RESULTS

M-CSF serum levels were elevated in pre-transplant haemodialysis patients (611+/-355 IU/ml vs 168+/-61 in normal controls, P<0.01). Following successful renal transplantation, M-CSF decreased in the first month, stabilizing at 257+/-222 IU/ml (not significantly different from normal controls) in 52 post-transplant stable patients. There was no correlation between M-CSF level and creatinine clearance. M-CSF levels increased significantly (2-5 times) during biopsy-proven acute rejection episodes in 20 of 25 patients. All rejection episodes were successfully treated and serum M-CSF decreased rapidly to pre-rejection levels in 17/20 patients. In contrast, in five patients with cyclosporin toxicity and four patients with other causes of allograft dysfunction, M-CSF serum levels did not change.

CONCLUSIONS

M-CSF serum level might be a specific marker of acute rejection. The source of increased production during rejection warrants further investigation, with infiltrating T cells and resident kidney cells being likely candidates.

Modulation of matrix gelatinases and metalloproteinase-activating process in acute kidney rejection

Changes in matrix metalloproteinase (MMP) activities would contribute to the accumulation of extracellular matrix during acute kidney allograft rejection. MMP-2 and MMP-9 and other gelatinolytic activities were examined in the rejected graft and the urine of a rat model of acute kidney rejection (orthotopic allotransplantation from a Buffalo donor to a Wistar-Furth recipient) by either zymography or fluorescence assay. MMP-2, membrane type 1 (MT1)-MMP, and tissue inhibitor of metalloproteinase (TIMP)-2 were also examined by immunodetection. The proMMP-2 activity and protein level increased in the graft during rejection when compared with normal Buffalo kidney, whereas activated MMP-2 decreased. TIMP-2 protein levels were markedly decreased and MT1-MMP proteolytic fragments (44-40 kDa) were undetectable. This suggests an altered MT1-MMP-dependent processing of proMMP-2 into active MMP-2 due to a diminished TIMP-2 level in acute kidney rejection. In the urine the overall gelatinolytic activity decreased considerably, although activity associated with an as yet unidentified 78-kDa protein appeared 6 days after transplantation.

Gene expression analysis in microdissected renal tissue. Current challenges and strategies

The architecture and compartmentalization of the kidney has stimulated the development of an array of microtechniques to study the functional differences between the distinct nephron segments. With the vast amounts of genomic sequence data now available, the groundwork has been laid for a comprehensive characterization of the molecular pathways defining the differences in nephron function. With the development of sensitive gene expression techniques the tools for a comprehensive molecular analysis of specific renal microenvironments have been provided: Quantitative RT-PCR technologies now allow the analysis of specific mRNAs from as little as single microdissected renal cells. A more global view of gene expression regulation is a logical development from the application of large scale profiling techniques. In this review, we will discuss the power and pitfalls of these approaches, including their potential for the functional characterization of nephron heterogeneity and diagnostic application in renal disease.

Ultrasound score is more predictive than serum creatinine in assessment of cellular rejection in cynomolgus monkey renal allografts

RATIONALE AND OBJECTIVES

To investigate whether ultrasound (US), in particular the use of an ultrasound scoring system, can provide more diagnostic information than clinical parameters, such as serum creatinine, for the diagnosis and determination of the degree of cellular rejection in renal allografts in the cynomolgus monkey (Macaca fascicularis).

METHODS

Sixty-eight cynomolgus monkeys with life-supporting renal allografts were examined with a 7.5MHz linear ultrasound transducer. One-hundred fifty two-dimensional, spectral, and power Doppler examinations were performed and four ultrasound parameters, percentage increase in graft volume, cortical thickness, resistive index (RI) of the renal arcuate artery, and power Doppler (PD) scores were recorded from serial examinations. An ultrasound score was assigned to each graft based on the number of those parameters that were abnormal; a score of 1 indicated that all four were normal, and a score of 5 that all four were abnormal. Each parameter and the combined score were compared with serum creatinine values and histology and evaluated statistically using Spearman rank correlation.

RESULTS

In animals with dysfunctioning allografts (serum creatinine elevations >200 micromol/L), Spearman rank correlation showed a significant correlation between the US score and the histology score: between 200 and 500 micromol/L, r = 0.309, P = 0.046, n = 31 and if > 500 micromol/L, r = 0.486, P = 0.005, n = 30. In those same animals, no correlation could be shown between serum creatinine values and the US score or between the serum creatinine values and the histologic diagnosis. In contrast to the US score, single ultrasound parameters were not found to correlate to histologic findings.

CONCLUSION

The application of ultrasound imaging in nonhuman primate renal transplant models provides valuable information concerning the presence and severity of cellular rejection in cases of graft dysfunction and the US score has a better predictive value of histology than serum creatinine values alone.

In vitro T cell proliferation from kidney allograft biopsies with unremarkable pathology: new strategies for an old problem

Acute rejection of renal allografts is mediated by infiltrating alloreactive T cells. The goals of this study were to correlate T cell proliferation with rejection and to determine whether T cell proliferation in the absence of rejection would predict future rejection episodes. Toward this, kidney biopsies (n=100) were cultured in the presence of interleukin-2. Cultures were examined at 4, 24, and 48 hr for T cell proliferation. A strong correlation was observed between T cell proliferation at any time point and rejection. There was not a significant correlation between T cell proliferation in biopsies with no rejection and the occurrence of a rejection episode within 2 months. However, T cell proliferation after 4 hr was a better predictor of the occurrence of rejection within 2 months compared with observations after 24 and 48 hr. Therefore, a subgroup of patients with unremarkable biopsies but T cell proliferation may be at risk for rejection and warrant closer observation and possible tailoring of immunosuppression.

Quantitative gene expression analysis in renal biopsies: a novel protocol for a high-throughput multicenter application

BACKGROUND

Recent advances in gene expression analysis may add the quantification of mRNA species in renal biopsies to routine diagnostic procedures in nephrology.

METHODS

A systematic evaluation was performed on the relevant steps required to efficiently obtain cDNA from renal biopsies for high-throughput reverse transcription-polymerase chain reaction (RT-PCR) based mRNA quantification.

RESULTS

The protocol preserves mRNA integrity by a novel RNase inhibitor and allows meticulous microdissection followed by maximal RNA recovery from tissue samples. Reverse transcription was optimized to give the best yield from minimal starting material. RNA quantity and quality were systematically investigated by real-time RT-PCR and electrophoresis on a microfluidic system, respectively. The reported procedure offers high RNA preservation and increases the yield of cDNA significantly compared to former protocols.

CONCLUSION

The simplicity of biopsy material acquisition combined with the centrally performed processing makes this protocol suitable for a wide spectrum of expression analysis in diverse clinical settings.

Laser microdissection and gene expression analysis on formaldehyde-fixed archival tissue

BACKGROUND

Analysis of renal biopsies is currently based on histological recognition of typical structural patterns and immunohistological detection of protein expression alterations. Both can be performed using formaldehyde as the tissue fixative. As a consequence of recent advances in molecular medicine, mRNA expression analysis may offer an attractive option to obtain functionally relevant information. However, quantification of mRNA expression in human renal biopsies thus far has not been possible in formaldehyde-fixed tissue.

METHODS

The present study evaluated a recently reported mRNA extraction protocol. Using this approach gene expression analysis could be performed on formaldehyde-fixed archival renal tissues by laser microbeam microdissection, laser pressure catapulting and real time reverse transcription-polymerase chain reaction.

RESULTS

For an initial feasibility study, the expression of two chemokines (IP-10 and RANTES) in renal transplant rejection was examined. Induction of protein expression in allografts undergoing rejection was demonstrated for both chemokines by immunohistochemistry. The mRNA expression alterations in the defined renal compartments of glomeruli, vessels and tubulointerstitium were quantified using laser microdissection from formaldehyde-fixed, paraffin-embedded or frozen tissue sections. A pronounced increase of mRNA expression compared to controls was demonstrated for IP-10 as well as RANTES with both tissue-processing protocols.

CONCLUSIONS

Using formaldehyde as the tissue fixative, information on the disease process can now be obtained by histological, immunohistochemical and gene expression techniques. In the future this may allow the study of activated molecular programs in routine renal biopsies as well as archival tissue samples.