Report of the Participants’ Meeting

of the National Renal Pathology EQA Scheme

held in the Castle Leazes Hall of Residence

University of Newcastle Medical School.

1pm on 5th July 2005.

Professor Furness opened the meeting by summarising its objectives and inviting participants to comment on the running of the scheme and suggest any improvements, either at the meeting or in private afterwards.

The cases in circulation R and S were discussed as follows.

R230

The consensus diagnosis in this case was fibrillary glomerulonephritis.  It was agreed that “non-amyloid monoclonal immunoglobulin deposition disease” was also correct, indeed arguably was a better diagnostic category in view of the evidence of deposition of monoclonal protein and the patient having chronic lymphocytic leukaemia.  There was a discussion as to whether immunotactoid glomerulopathy should or should not be regarded as a separate diagnosis to fibrillary glomerulonephritis.  It was agreed that for the purpose of personal scoring immunotactoid glomerulopathy should be regarded as correct. 

Diagnostic categories 7 to 10 were regarded by the meeting as incorrect.  “Light chain deposition disease” was specifically considered and was taken to indicate light chain nephropathy, which is a different diagnosis.

R231

The clear consensus was that this was a case of acute tubulo-interstitial nephritis.  A high proportion of participants had suggested this was probably due to the drugs given.

The diagnostic category “interstitial nephritis +/- thrombotic micro-angiopathy” was specifically discussed by the meeting.  No-one present at the meeting felt there was evidence of a thrombotic micro-angiopathy in this case and therefore that diagnosis should be accorded half marks.

R232

The most popular diagnosis in this case was immunotactoid glomerulonephritis, which was also the diagnosis of the submitting pathologist.  There was, however, a discussion as to what should be done with the other diagnostic categories.  Specifically, an unqualified diagnosis of cryoglobulinaemia was problematic, since although immunotactoid glomerulonephritis can be associated with cryoglobulinaemia, there was actually no evidence of cryoglobulinaemia in this case.  There was further uncertainty as to what was meant by “monoclonal immunoglobulin deposition disease”.  This could imply non-amyloidotic fibrillar monoclonal immunoglobulin deposition disease, light chain nephropathy or heavy chain nephropathy.  After lengthy discussion it was decided that this case should not be used for personal scoring.

R233

This was a straightforward case of membranous glomerulonephritis, as identified by all the participants.

The suggested diagnosis of “IgA myeloma” was discussed.  Despite the fact that the diagnosis of myeloma was correct, this was regarded by the meeting as incorrect diagnosis because IgA myeloma does not typically cause mesangial IgA deposits and there was no evidence that the monoclonal protein in the casts was of IgA type.  The deposition of IgA in the glomeruli was regarded by the meeting as incidental. 

R235

A large proportion of the participants had already indicated that this case was not suitable for personal analysis because of the poor quality of the submitted material.  It was also notable that the most popular diagnosis in the EQA scheme was HIV nephropathy – despite the patient having a low viral load and the biopsy not showing typical collapsing glomerulopathy.  The submitting pathologist had made a diagnosis of minimal change nephropathy, which had subsequently proved steroid sensitive.  It was agreed that this case should not be used for personal scoring. 

S236

This case was submitted from overseas by a participant who is no longer a member of the EQA scheme.  It was submitted for education and interest only.  The meeting agreed that this was an immune complex mediated glomerulonephritis, with features raising the possibility of lupus nephropathy, but a more specific diagnosis was not possible without further clinicopathological correlation, especially lupus serology and hepatitis status.  Unfortunately such further information was not available.  The case was deleted for personal scoring purposes.

S237

This was a 29 year old woman with lupus nephropathy.  The vast majority of participants had correctly identified the presence of lupus.  It was agreed that all diagnoses of lupus nephropathy class IV would be regarded as correct.  Lupus class II was specifically discussed by the meeting and was regarded as incorrect in view of the clear evidence of capillary loop involvement.

S238

This was a straightforward case of IgA nephropathy, as identified correctly by all participants.

S239

This 29 year old male was known to have AIDS but had stopped taking antiretroviral therapy prior to developing nephrotic syndrome and acute renal failure.  The consensus diagnosis was HIV-associated nephropathy.  The meeting agreed that given the clinical information, collapsing glomerulopathy could be regarded as an equally correct diagnosis.  The diagnosis “segmental glomerulonephritis ?collapsing” should be accorded half marks for personal scoring and other diagnoses will be regarded as incorrect.

Subsequent information indicated that re-starting antiretroviral therapy had produced an improvement in renal function and a decrease in proteinuria, though the patient has subsequently been lost to follow-up.

The minority diagnosis of ‘IgA nephropathy and microangiopathy’ was specifically discussed.  In view of the high proportion of participants who had suggested the concurrent presence of hypertensive damage it was agreed that the mention of micro-angiopathy probably referred to the hypertensive damage.  This diagnosis was therefore to be merged with IgA nephropathy for personal scoring purposes.  The other suggested diagnoses in this case were regarded as incorrect. 

S241

The vast majority of participants agreed that this was a case of lupus nephropathy, though there was a very worrying spread in terms of further classification of the changes.  Although the consensus was lupus class IV, a significant proportion suggested class III and almost 14% suggested class II.  After discussion it was agreed that this case should not be used for personal scoring purposes.

After agreeing how to use the cases for personal analysis Professor Furness repeated the invitation to suggest improvements or amendments to the scheme.  None were received.

Professor Furness proceeded to describe preliminary results from the UK-wide study of reproducibility of the new WHO/RPS classification of lupus nephritis.  A simple statistical analysis had produced a highly significant result indicating that the new classification produces considerably more reproducible results than the old WHO classification.  However, the data has been reviewed by the statistician Dr Nick Taub who sounded a note of caution on the grounds that the new classification appeared to be producing a higher proportion of diagnosis of lupus class IV.  This change in the distribution of the classes could produce a spurious impression of improved reproducibility on the basis of the simple statistical tests used.  (The validity of this argument can be illustrated by considering an extreme situation;  if a new classification ensured that every case was classified into a single category, the reproducibility of that classification would inevitably be 100%).   He has undertaken to perform a more sophisticated statistical analysis, but the results of that analysis are not yet available. 

Professor Furness also discussed the proposal to carry out a randomised study of participant performance in EQA by telepathology against EQA performance by conventional light microscopy.  The meeting supported this proposal and further details will follow under separate cover.

The meeting closed at 2.15pm.