A UK-wide trial of the Banff classification of renal transplant pathology in routine diagnostic practice.

between centres, especially in research projects. While there are good reasons also to adopt it in routine Background. The Ban classification of renal transplant pathology has gained wide support since its introduc- diagnostic practice, further refinement is necessary before an improvement in the accuracy of diagnosis tion in 1993. There have been several studies which have tested its usefulness in the context of research- can be demonstrated. oriented centres. We sought to evaluate its use in a wider context. diagnosis; rejection; Ban Methods: We recruited pathologists from all but one of the renal transplant centres in the UK. Sections were circulated from 21 selected, ‘dicult’ cases, in all of which the clinical question was confirmation or Introduction exclusion of acute rejection, and in all of which a definite diagnosis had been obvious from the sub- The Ban classification of renal transplant pathology sequent clinical course. Participants were asked first to was introduced in 1993 [1 ]. It attempts to identify the diagnose or exclude acute rejection by their usual changes which may be seen in dysfunctional renal approach, then to apply the Ban classification. No allografts, to define a grading system for these changes, clinical information was given beyond the time since and to synthesise the results into a numeric classificaengraftment, in order to confine the evaluation to the tion. Its aims are to improve the accuracy of diagnosis, morphological features present in the sections. At especially the clinically important diagnosis of acute the end of the study the subjective impressions of rejection; to improve the reproducibility of that diathe participants were sought using a structured gnosis; and to unify the nomenclature. questionnaire. The ideal study to evaluate such a system is probably Results. Using the Ban classification produced no impossible to carry out. Even if one restricts the study detectable dierence in the number of ‘correct’ dia- to the confirmation or exclusion of acute rejection, one gnoses when compared with a conventional approach, would need a very large number of cases, representative irrespective of whether the ‘correct’ diagnosis is based of the whole spectrum of a routine workload (which on retrospective clinical information or on the con- of course varies between institutions), all with clinically sensus opinion of the pathologists involved, and irre- validated ‘correct’ diagnoses, with the same slides being spective of where in the Ban schema one applies a studied by a large number of practising pathologists. ‘cut-o ’ for the diagnosis of acute rejection. However, Several studies of the accuracy and reproducibility the reproducibility of the diagnoses was improved. The of the Ban classification have already been published results suggest that in the Ban classification the best [2‐5 ] but all have compromised this ideal by using a ‘cut-o ’ point for the diagnosis of acute rejection is small group of highly motivated, research-orientated between Ban category 3 and category 4, although in renal histopathologists to provide the Ban scores for this dicult area we found a large improvement in their cases. This approach avoids the fatigue which diagnostic accuracy if input of clinical information may be encountered by less dedicated pathologists and occurs. ensures a complete data set. These studies have validConclusions. The improved reproducibility justifies the ated the use of the Ban classification as a tool to use of the Ban classification to harmonise approaches unify diagnostic criteria when used by such dedicated individuals, especially in the context of a research