The Banff classification revisited
Kim Solez1and Lorraine C. Racusen2
1Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Alberta, Canada and2Department of Pathology,
Johns Hopkins Medical Institutions, Baltimore, Maryland, USA
From small beginnings in 1991, the Banff working
classification of renal allograft pathology has grown to be
a major force for setting standards in renal transplant
pathology, and is widely used in international clinical trials of
new antirejection agents. The meeting, classification, and
consensus process have unique history, and look poised to
continue for another several decades as the embodiment
of the process for setting global standards in pathology.
The Banff meetings have expanded from renal allograft
pathology to most other areas of solid organ transplantation,
and increasingly incorporate international working groups,
so that productive collaborative activity is ongoing, creating
an important dynamic process enhancing clinical success in
transplantation. On the other hand, despite the successes
of the working classifications and ongoing collaborative
efforts, there are limitations in this and other pathological
classifications, related to potential for sampling error, issues
of reproducibility when implemented globally, and lack of
formal incorporation of morphometry and molecular and
genomics approaches. Some of these problems cannot be
overcome within the realm of traditional histopathology,
and will only be solved when the classification is able to
confidently embrace genomics and molecular medicine
parameters for all common diagnoses. The smooth
integration of these newer technologies with traditional
histopathology is one of the great challenges for the future.
advance online publication, 12 December 2012;
KEYWORDS: central slide review; consensus generation; genomics; history of
transplantation; technology and future of medicine course; transplantation
The Banff Classification of Kidney Allograft Pathology had its
origin in a meeting in Banff, Canada, held on 2–4 August
1991.1This original Banff meeting in 1991 was part of
the activities of the International Society of Nephrology
Commission on Acute Renal Failure, which also included
international disaster relief. The initiative was inspired by the
then recent development of a consensus grading system for
diagnosis of rejection in cardiac allografts2led by Dr
Margaret Billingham, a key participant at the first Banff
meeting. Looking back now 21 years later at that meeting and
the working classification developed there, it is clear that the
classification,3as it has evolved since then, has made a critical
contribution to many advances in the field of transplantation
(Figure 1). That success is a testament to the efforts of an
ever-growing number of people who continued to meet every
2 years and refine and expand the classification. Beginning in
the kidney, the Banff Allograft Pathology consensus process
ultimately not only led to revisions and expansion in the
schema for kidney allograft pathology but also extended to
development of classifications of allograft pathology in the
liver, pancreas, and composite tissue grafts,4–6as well as
contributing to advances in the existing classifications of the
heart and lung allograft pathology.
Counterbalancing the successes of the classification are its
limitations—limitations that apply to any biopsy- and
histopathology-based classification system. These include
inherent potential for sampling error, suboptimal reprodu-
cibility when implemented globally, lack of widespread
application of morphometry with true quantification, and
lack of formal integration of molecular and genomics data
into the classification. These are issues to be addressed in the
ongoing international Banff conferences and related activ-
ities, described below.
The need for a new classification was quite obvious in 1991
when the Banff Classification of Allograft Pathology began.
Most people learning kidney transplant pathology had no
direct mentor and were learning from outdated textbooks.
Although a few individual classifications for renal allograft
rejection had been developed, none were in general use. The
new consensus classification for heart allograft rejection had
just been published2and served as a model for the process. A
group of individuals who had published in the field met to
discuss the literature, cardinal histopathological findings, and
& 2012 International Society of Nephrology
Correspondence: Kim Solez, Department of Laboratory Medicine and
Pathology, University of Alberta, 8440 112th Street, Edmonton, Alberta,
Canada T6G 2B7. E-mail firstname.lastname@example.org
Received 31 May 2012; revised 31 July 2012; accepted 23 August 2012
potential categories of rejection that would be clinically
relevant for guiding therapy.
The new classification used lesion scoring and guidelines
to provide rigor in the evaluation of renal allograft pathology
and the assignment of biopsies into diagnostic categories.
Nonrejection pathology in the allograft was also considered
and described in some detail, and this continues to be a focus
in ongoing Banff meetings.
THEMES OF THE BANFF MEETINGS
A previous article on the history of the Banff Classification
focused on the participants.1Here, we will concentrate on the
ideas espoused by the Banff meetings (Table 1) and the Banff
consensus process. The flavor of the Banff meetings from the
beginning was one of flexibility and openness to the ideas of
others. At the original meeting of 20 people in 1991, the
kidney transplant pathology classification that emerged was
very different from any of the drafts that individuals had
created before the consensus discussions began. It was a true
creation of the meeting itself, based on international
expertise, the existing literature, and facilitated discussions.
The classification was designed as a dynamic working
document that could be modified as the need for future
changes was demonstrated.
At the second meeting in 1993, a similar effort was
initiated to create a classification for liver transplant
pathology, and this became a major focus of the Third Banff
Conference held in July 1995.4The third conference had four
times as many attendees as the first conference and was
conducted with an air of positive expectation that could not
have been anticipated in 1991. The 1995 meeting also
brought Banff lesion scoring into line with the CADI scoring
system,7so there was no difference between the two
By the time of the 1995 meeting, the original Banff
Classification of Renal Allograft Pathology was in wide use,
and there had been extensive studies showing its clinical
validity and reproducibility.8–12Extrapolating from these
studies, it seemed likely that the classification had already
resulted in improvement in patient care. The Banff
classification had been endorsed by the FDA and other
regulatory agencies, and had enabled the use of objective
histological endpoints for international clinical trials of new
antirejection agents and other scientific studies, a process that
continues to this day.
Table 1|Banff Allograft Pathology Meetings since 1991 with key themes
(days)LinksKey subjects debated
Classification established, lesion scoring, diagnostic categories, physician-led consensus
Liver classification, chronic rejection, first presentation on molecular pathology approaches
Pancreas classification, glomerulitis, first international medical meeting on CD-ROM, first
Banff conference with microscope sessions. Lesion scoring normalized with CADI.
Merging of Banff and CCTT classifications, establishing basis for current Banff classification,
post-transplant lymphoproliferative disorder, first Banff conference with posters
Protocol biopsies, chronic rejection, and viral diseases, clinical practice guidelines.
First conference supported by an NIH grant.
AMR, donor biopsies, genomics, CAN, heart transplantation
C4d, macrophages, tolerance, accommodation, immunodepletion
1997 Banff, Canada5ISN
1999Banff, Canada5ISN, NKF, NIH
La Coruna, Spain
20056NKFGenomics and molecular markers, B cells, chronic allograft injury with elimination of CAN,
establishment of criteria for chronic rejection
Protocol biopsies, transcriptome, mechanisms of rejection, ptc grading, new total
inflammation score; working groups for v-lesion, genomic integration, pancreas and
composite tissue rejection schemas
Viruses, quality assurance, AMR in kidney, heart, and pancreas, liver allograft
accommodation, endothelial cells, surrogate markers. Working groups.
Sensitized patient, C4d, isolated v-lesion, the future, genomics, glomerulitis, epithelial
injury/epithelial mesenchymal transformation, operational tolerance monitoring in liver
Abbreviations: AMR, antibody-mediated rejection; CAN, chronic allograft nephropathy; CAP, Canadian Association of Pathologists, and Future of Pathology/Laboratory
Medicine in Canada Consortium; ISN, International Society of Nephrology; NKF, National Kidney Foundation (US); NIH, National Institutes of Health (US); ptc, peritubular
capillary; UofA, University of Alberta.
Banff articles per year
1996 1995 1994
Figure 1|Number of Banff articles in transplantation per year.
The peak of 73 articles in 2005 was the combined effect of increasing
interest in antibody-mediated rejection and viral disease, and the
controversy surrounding the term ‘chronic allograft nephropathy’.
The distribution into categories was Kidney-Clinical (human) 611,
Kidney-Experimental 40, Liver-Clinical (human) 48, Liver-Experimental
8, Pancreas-Clinical (human) 4, All Organs-Clinical (human) 2,
Composite Tissue-Clinical (human) 3, and Heart-Clinical (human) 1.
There are 38 articles thus far in 2012 to July 31, which extrapolate to
65 for the year making 2012 the second highest year for Banff articles.
K Solez and LC Racusen: Banff classification revisited
Interestingly, in 1995, an energetic attempt was mounted
to remove glomerulitis—the ‘g’ lesion (Figure 2)—from the
classification, as the lesion appeared to have little clinical
significance, based on the existing literature at that time. This
was successfully resisted, and today glomerulitis is recognized
as an important sign of microcirculatory alloinflammation
and is an integral part of the classification. With the recent
recognition of C4d-negative antibody–mediated rejection
(AMR), the presence of ‘the g lesion’ is now sometimes the
main evidence supporting that diagnosis along with peri-
tubular capillaritis (Figure 3).13The 1995 meeting also saw
establishment of precepts
classification: ‘no changes should be made unless supported
by well conducted studies from two different investigative
At the 1997 meeting, the most useful and validated parts
of the Banff Classification were merged together with the
most useful parts of the CCTT classification,15which had
been developed as a simplified version of the Banff criteria,
defining tubulointerstitial, vascular, and severe rejection with
clinical correlations. These were merged to create a new
classification: Banff ’97.16There have been subsequent
modifications and additions, but the classification today is
still built on the framework established at the 1997 meeting.
A CD-ROM was created capturing video high points of the
Although AMR was rarely diagnosed at the time of the
inaugural Banff meeting, development of the classification,
and the 1997 modified schema, by the mid-to-late 1990s,
characteristic histological changes had been identified and
AMR was being increasingly recognized in renal allografts.
With the addition of immunohistological staining for C4d to
the pathologists’ tool box,18definite criteria enabling
diagnosis of AMR could be established. On the basis of the
growing literature and consensus discussions, an addition to
the Banff classification outlining criteria for diagnosis of
AMR was published.19In the years since then, AMR in other
solid organ allografts has been discussed at the Banff
meetings, and criteriafor
published (e.g., in pancreas).20
guiding changesin the
As immunosuppressive strategies became more effective,
acute rejection was no longer a common cause of graft loss,
and investigators began to focus on late allograft changes and
fibrosis in the allograft. The 1997 Banff working classification
had guidelines for semiquantitative assessment of fibrosis in
allografts and moved from labeling all chronic changes in the
allograft ‘chronic rejection’ to a new concept, ‘chronic
allograft nephropathy’. Chronic allograft nephropathy in-
cluded a subgroup because of chronic rejection, but with
recognition of the other numerous causes of fibrosis in the
allograft. In the Banff meeting in 2001, there was extensive
discussion of fibrosis and atrophy in the allograft, with
nephropathy, while correlating with graft outcomes, in itself
became a blanket term that was almost meaningless in its lack
of specificity. This topic remained a major issue at
subsequent Banff meetings, and by the Banff meeting in
2005 and 2007, precise definitions of lesions led to
elimination of the term chronic allograft nephropathy22to
be replaced by specific diagnoses, including chronic rejection,
both cell- and antibody-mediated.23Only a relatively small
subset of patients remained with interstitial fibrosis and
tubular atrophy of undefined cause.
A major advance of the 2009 meeting was the creation of
the Banff working groups, which aim to address issues in
transplantation pathology. These include the significance of
isolated intimal arteritis without tubulointerstitial inflamma-
tion (the ‘isolated v-lesion’), accurate scoring of fibrosis,
detection and accurate scoring of acute and chronic
glomerular lesions, incorporation of molecular pathology
into diagnostic schemas, a possible new classification system
for polyomavirus nephropathy, quality assurance in histology
and immunohistology, and diagnosis of AMR in kidney
independent of C4d. These international Banff working
groups collect data from multiple centers in areas where
Banff criteria are problematic, and validate the results,
ultimately leading to refinement and improvement of the
working classification, thus ensuring that the Banff process is
evidence-based and continues to lead to improvements in
patient care and management.24,25At the most recent Banff
Figure 2|Glomerulitis. One of the key lesions in antibody-mediated
Figure 3|Peritubular capillaritis. Another key lesion in antibody-
K Solez and LC Racusen: Banff classification revisited
meeting in 2011, the criteria for diagnosis of AMR in the
kidney were revisited with the recognition that one or more
diagnostic criteria may be absent in cases that subsequently
developed chronic microvascular injury.13,24,25Banff working
groups are currently studying this issue and developing
recommendations for a change in the working classification.
LIMITATIONS OF THE CLASSIFICATION
From the beginning, the Banff classification has included
semiquantitative lesion grading, which made it simple and
quick to apply. However, some have emphasized the
histopathology scoring system, is another limiting factor
but seems highly dependent on the group in which the
reproducibility is tested. Initial studies on reproducibility in
the early years of the classification were reasonably
encouraging, with things looking less favorable as studies
expanded more widely.8–12,27It is fascinating that with
fibrosis scoring, the same first author who demonstrated
excellent reproducibility in fibrosis scoring in the native
kidney28demonstrated much poorer reproducibility in the
transplanted kidney where participants were more widely
scattered.29Ultimately, reproducibility is influenced by true
biological variability, by the exact ways in which lesions are
scored, and by the experience of the pathologists performing
Early on in deliberations at the Banff meetings, standar-
dization and streamlining of morphometry techniques were a
focus, so that they could be routinely applied to diagnostic
material with quantitative results obtained in a timely
manner. However, such strategies have never been developed
for standardized application to biopsy assessment. This is an
issue currently being addressed by the graft fibrosis Banff
working group. Other problems include the persistence of the
‘borderline’ category, when inflammation is below threshold
to confidently diagnose acute cellular- or AMR using current
histopathological criteria. Diagnostic thresholds are set to
avoid overtreatment with attendant toxicities but may reduce
sensitivity. Integration of molecular and genomics data into
the classification remains an ideal and a working group focus,
although implementation of this throughout the world may
be problematic, especially in under-resourced areas.
Whether one understands the failure to disseminate and
standardize morphometry, genomics, and molecular studies
as a failure of medical politics and vision, or as simple
economics, it is clear that it may have retarded progress to
an ideal diagnostic paradigm. The success of recent mor-
phometry, genomics, and molecular studies in experimental
models and in the native kidney may ultimately lead to
standardization and implementation in individual transplant
centers worldwide, and these strategies remain a major focus
of the ongoing Banff meetings.
A recent article related to the Banff classification suggests
that tubulitis is severely underdetected by conventional
microscopy.30In this study, in which lymphocytes in the
tubular epithelium, which define tubulitis, were identified by
immunostaining for CD3, the authors conclude that tubulitis
is missed very frequently, but the Banff classification seems to
be ‘calibrated’ to allow for this and it does not seriously affect
the identification of clinically significant acute rejection.
Immunostaining is therefore not indicated in routine practice
because (by Banff criteria) it would result in overdiagnosis of
rejection. Intimal arteritis can indicate acute rejection even if
extremely mild and the ‘isolated v-lesion’ can be clinically
meaningful. The importance of isolated intimal arteritis/
‘isolatedv-lesion’ in the
inflammation is an ongoing focus of a multicenter study
organized by a Banff working group, and will continue to be
debated. One recent presentation suggests that patients with
this lesion may actually have a worse prognosis than when the
arteritis is associated with tubulointerstitial inflammation.31
Two recent articles relating molecular changes to Banff
lesion scoring directly show the potential in the long run for
molecular studies to greatly improve the classification.32,33
THE BANFF CONSENSUS PROCESS—WHAT HAS BEEN
The Banff meetings are designed for optimal interaction, with
many small group sessions and working group meetings, and
attendees feel very much a part of the intellectual process. It
is also an excellent forum for junior and senior participants
to mingle, and for organization of collaborative studies
around issues discussed at the Banff meetings. Meetings are
open to all, and poster sessions that frequently feature the
work of young investigators are an important component of
the meetings. Following consensus discussions and literature
review, numerous formal multiauthored publications have
resulted from the Banff meetings. The Banff working groups
were established in 2009. The work of many of these groups
has been presented at meetings in 2012, with full-length
papers expected in 2013. It appears that the Banff
collaborative process will be productive well into the future.
As shown in Figure 1, there are now 717 articles in the
transplantation literature that reference the Banff process.
Activities of and interest in the Banff Allograft Pathology
consensus process continue to grow. These 717 articles are an
underestimation of the impact of the Banff Allograft
Pathology consensus process, as they include only those
articles where Banff was specifically listed as a key word or
search term. Of the 717 articles, most articles are in journals
with impact factor of 3.2 or higher, and one-third have
impact factors of 6–9. The Banff meeting reports and main
meeting papers have been cited over 4300 times in the
medical literature. The 1999 paper ‘‘The Banff 97 Working
Classification of Renal Allograft Pathology’’,16an update of
the original classification, is a citation classic in the field
having been cited 1764 times. This paper reflects a basic
philosophy of the Banff process—that this and other Banff
schemas are working classifications intended to be con-
tinuously refined and updated as the field of solid organ
transplantation evolves (e.g., Demetris et al.34).
K Solez and LC Racusen: Banff classification revisited
The strong interest in global consensus generation and
electronic communications that began with the Banff process
led to the establishment of the cyberNephrology and
cyberMedicine initiatives and eventually to the creation of
the Technology and Future of Medicine course at the
University of Alberta.35
It can be difficult to focus on the future in the field of
medicine when the problems of the present already seem so
challenging. It is heartening to see presentations about the
Banff classification begin to be more future oriented as seen
in this slide from a May 2012 presentation by Thangamani
Muthukumar (Figure 4). We are fortunate that the next
(2013) Banff meeting will be in Brazil where the application
of strategic foresight to science is official government
planned in Canada, Turkey, and Spain in 2015, 2017, and
2019, respectively, is available on line.
about future meetings
All the authors declared no competing interests.
1. Solez K. History of the Banff classification of kidney allograft pathology as it
approaches its twentieth year. Curr Opin Organ Transplant 2010; 15: 49–51.
2. Billingham ME, Cary NR, Hammond ME et al. A working formulation for
the standardization of nomenclature in the diagnosis of heart and lung
rejection: Heart Rejection Study Group. The International Society for
Heart Transplantation. J Heart Transplant 1990; 9: 587–593.
3. Solez K, Axelsen RA, Benediktsson H et al. International standardization of
criteria for the histologic diagnosis of renal allograft rejection: the Banff
working classification of kidney transplant pathology. Kidney Int 1993; 44:
4.An International Panel. Banff schema for grading liver allograft rejection:
an international consensus document. Hepatology 1997; 25: 658–663,
5.Drachenberg CB, Odorico J, Demetris AJ. Banff schema for grading
pancreas allograft rejection: working proposal by a multi-disciplinary
international consensus panel. Am J Transplant 2008; 8: 1237–1249.
6.Cendales LC, Kanitakis J, Schneeberger S et al. The Banff 2007 working
classification of skin-containing composite tissue allograft pathology. Am
J Transplant 2008; 8: 1396–1400.
7. Isoniemi HM, Krogerus L, von Willebrand E et al. Histopathological
findings in well-functioning, long-term renal allografts. Kidney Int 1992;
Solez K, Hansen HE, Kornerup HJ et al. Clinical validation and
reproducibility of the Banff classification of renal allograft pathology.
Trans Proc 1995; 27: 1009–1011.
Gaber LW, Moore LW, Alloway RR et al. Correlation between Banff
classification, acute renal rejection scores and reversal of rejection.
Kidney Int 1996; 49: 481–487.
Marcussen N, Olsen TS, Benediktsson H et al. Reproducibility of the Banff
classification of renal allograft pathology: inter- and intra-observer
variation. Transplantation 1995; 60: 1083–1089.
Rush DN, Jeffery JR, Gough J. Sequential protocol biopsies in renal
transplant patients. Clinico-pathological correlations using the Banff
schema. Transplantation 1995; 59: 511–514.
Gaber LW, Moore LW, Gaber AO et al. Correlation of histology to clinical
rejection reversal: a thymoglobulin multicenter trial report. Kidney Int
1999; 55: 2415–2422.
Sis B, Jhangri GS, Riopel J et al. A new diagnostic algorithm for antibody-
mediated microcirculation inflammation in kidney transplants. Am J
Transplant 2012; 12: 1168–1179.
Solez K, Benediktsson H, Cavallo T et al. Report of the Third Banff
Conference on Allograft Pathology (July 20-24, 1995) on classification and
lesion scoring in renal allograft pathology. Transplant Proc 1996; 28:
Colvin RB, Cohen AH, Saiontz C et al. Evaluation of pathologic criteria for
acute renal allograft rejection: reproducibility, sensitivity, and clinical
correlation. J Am Soc Nephrol 1997; 8: 1930–1941.
Racusen LC, Solez K, Colvin RB et al. The Banff 97 working classification of
renal allograft pathology. Kidney Int 1999; 55: 713–723.
Solez K, Racusen L. Fourth Banff Conference on Allograft Pathology CD-
ROM. Churchill Livingstone: Philadelphia, PA, 1999; http://
Collins AB, Schneeberger EE, Pascual MA et al. Complement activation
in acute humoral renal allograft rejection: diagnostic significance
of C4d deposits in peritubular capillaries. J Am Soc Nephrol 1999; 10:
Racusen LC, Colvin RB, Solez K et al. Antibody-mediated rejection
criteria—an addition to the Banff 97 classification of renal allograft
rejection. Am J Transplant 2003; 3: 708–714.
Drachenberg CB, Torrealba JR, Nankivell BJ et al. Guidelines for the
diagnosis of antibody-mediated rejection in pancreas allografts-updated
Banff grading schema. Am J Transplant 2011; 11: 1792–1802.
Racusen LC, Solez K, Colvin R et al. Fibrosis and atrophy in the renal
allograft- interim report and new directions. Am J Transplant 2002; 2:
Solez K, Colvin RB, Racusen LC et al. Banff 05 meeting report: differential
diagnosis of chronic allograft injury and elimination of chronic allograft
nephropathy (CAN). Am J Transplant 2007; 7: 518.
Banff 2015 update
• No tubulitis
• No glomerulitis
• No arteritis
• No interstitial infiltrates
• No BM duplication
• Normal blood vessels
• No tubular atrophy
• No interstitial fibrosis
• Interstitial infiltrates
• BM duplication
• Abnormal blood vessels
• Tubular atrophy
• Interstitial fibrosis
Normal allograft pathomics
Abnormal allograft pathomics
Figure 4|Presentation slide by Thangamani Muthukumar in the style of Wired Magazine’s FOUND: Artifacts from the Future
speculating on what the state of science will be at the Banff 2015 meeting. ‘Pathomics’ is a term from a 2007 editorial by Robert Colvin.36
BM, basement membrane.
K Solez and LC Racusen: Banff classification revisited
23.Solez K, Colvin RB, Racusen LC et al. Banff 07 classification of renal allograft Download full-text
pathology; update and future directions. Am J Transplant 2008; 8: 753.
Sis B, Mengel M, Haas M et al. Banff ’09 meeting report: antibody
mediated graft deterioration and implementation of Banff working
groups. Am J Transplant 2010; 10: 464–471.
Mengel M, Sis B, Haas M et al. Banff 2011 meeting report: new concepts in
antibody-mediated rejection. Banff meeting report writing committee.
Am J Transplant 2012; 12: 563–570.
Howie AJ. Problems with BANFF. Transplantation. 2002;73(9):1383-4, a
comment on McCarthy GP, Roberts IS. Diagnosis of acute renal allograft
rejection: evaluation of the Banff 97 Guidelines for Slide Preparation.
Transplantation 2002; 73: 1518–1521.
Furness PN, Taub N, Assmann KJ et al. International variation in histologic
grading is large, and persistent feedback does not improve
reproducibility. Am J Surg Pathol 2003; 27: 805–810.
Farris AB, Adams CD, Brousaides N et al. Morphometric and visual
evaluation of fibrosis in renal biopsies. J Am Soc Nephrol 2011; 22:
Farris AB, Chan S, Climenhaga J et al. The Banff Fibrosis Trial for
reproducibility and clinical significance of visual assessment of interstitial
fibrosis in kidney biopsies. American Transplant Congress 2012, abstract #185.
Elshafie M, Furness PN. Identification of lesions indicating rejection in
kidney transplant biopsies: tubulitis is severely under-detected by
conventional microscopy. Nephrol Dial Transplant 2012; 27: 1252–1255.
31.Chang JH, Mohan S, Crew RJ et al. Acute antibody mediated rejection with
isolated endarteritis in kidney transplants portend worse prognosis.
American Transplant Congress 2012, abstract #62.
de Freitas DG, Sellare ´s J, Mengel M et al. The nature of biopsies with
‘borderline rejection’ and prospects for eliminating this category. Am J
Transplant 2012; 12: 191–201.
Moulavi D, Hajiloo M, Sander J et al. Combining gene expression and
interaction network data to improve kidney lesion score prediction. Int J
Bioinform Res Appl 2012; 8: 54–66.
Demetris A, Adams D, Bellamy C et al. Update of the International
Banff Schema for Liver Allograft Rejection: working recommendations
for the histopathologic staging and reporting of chronic rejection.
An International Panel. Hepatology 2000; 31: 792–799.
Solez K, Hales M, Katz SM. Electronic strategies for information and
research: cyberNephrology/cyberMedicine in the emerging world. Kidney
Int Suppl 2005; 98: S89–S94.
Colvin RB. Getting out of flatland: into the third dimension of microarrays.
Am J Transplant. 2007;7(12):2650-1. Comment on: Mueller TF, Einecke G,
Reeve J, et al. Microarray analysis of rejection in human kidney
transplants using pathogenesis-based transcript sets. Am J Transplant
2007; 7: 2712–2722.
Nehme CC, Santos MM, Filho LF. Challenges in communicating the
outcomes of a foresight study to advise decision-makers on policy and
strategy. Sci Public Policy 2012; 39: 245–257.
K Solez and LC Racusen: Banff classification revisited