Most of the respondents (85%) are using the classification with another 5% using the classification with their own modifications. 15% also give the WHO class in addition only 4 out of 55 do not use the classification. Their reasons for this included the complexity of the system, the fact that their clinicians preferred another system and that they did not think there was evidence that the new system had any advantages.
There were a number of comments and I have tried to summarise these into categories as below. In each case the number in brackets shows the number of people who raised the point.  

 Problems
-Difficulty with determining whether global sclerosis is due to SLE or not (7)
-Misleading to clinicians for biopsies with only chronic lesions to be included as class III or class IV (2). Having chronic lesions in Class III and IV is confusing (3)
-Confusion or scepticism of clinicians (5)
-Undervalues EM (3)
-Activity and chronicity indices should be included (4)
-Less common lesions (e.g collapsing, MCD) not covered (2)
-Should include interstitial and vascular lesions (4)
-Distinction of IV-G and IV-S difficult with chronic lesions (2)
-Need evidence for the division into IV-G and IV-S
-Segmental sclerosis occurs in membranous GN – does that also make it class III or IV?
-What about capsular lesions less than a crescent
-Although segmental lesions are important the cut off of 50% is arbitrary
-Cut off of 90% for class VI is too high
-Cumbersome e.g. 6 subdivisions of Class IV or too complicated (3)
-Difficult to assess 50% criterion for Class VNeeds to be something more than A/C and less than activity index

Selected comments
-In class III and IV, I am careful to communicate the proportion of active and chronic lesions as part of the diagnosis.  When a biopsy qualifies for III or IV because of the chronic component (IIIA/C, IIIC, IVA/C, and IVC, I do not put class III or IV in the diagnosis but include it in a note with a disclaimer (the patient qualifies because of the proportion of scarred glomeruli, and there are no active lesions ….).
-My main problem concerns knowing how to deal with globally sclerotic glomeruli (GSS). For instance, does the presence of GSS in an otherwise class II lesion automatically push this to a class III (C). Perhaps, in some instances it should and perhaps in others it should not. If there are fewer than 10% GSS, then maybe it should be attributed to aging or within the normal limits, but greater than this percentage would suggest that the scarring was due to lupus nephritis, perhaps? It seems that there is no easy answer for this, and I would be curious how you deal with this situation. Along those same lines, the presence of GSS can also complicate categorizing class IV lesions. Do the GSS factor into the determination of whether there are more segmental or globally involved glomeruli when you are deciding whether one is dealing with a class IV-S vs class IV-G? The distinction between IV-S and IV-G seems more feasible when a biopsy is predominantly active but the presence of significant chronicity makes this distinction much more difficult.
-I have concerns that it may be misleading to clinicians to classify biopsies showing only sclerotic lesions as Class III or IV since these inactive lesions do not warrant aggressive immunosuppressive therapy. Also, since segmental sclerosis may develop in membranous lupus nephritis, I think it may be impossible to sort out Class III (C), Class IV-S(C), or Class IV-G (C) plus Class V from pure Class V with secondary sclerosing lesions. e.g of misleading - if Bx had 10 glomeruli, 8 globally sclerotic, 1 with segmental sclerosis and the last one normal, this would be a Class IV. And surprisingly there is not an iota of activity!!!!! The connotation of Class IV has always and still is bad active disease.
I am in the camp that believes that segmental proliferative lupus is different than diffuse / global proliferative lupus. Placing a lesion with 45% segmental lesions in one class, and placing a case with 55% otherwise identical segmental lesions in another class with cases that are morphologically, and often immunohistological different, is fundamentally illogical and unscientific.
-I think it is much better than WHO - and much simpler to use. The clinicians in Toronto are satisfied with it and I like it!. I think the category of IV-S is very useful and probably has real clinical meaning. I like that a normal biopsy does not classify as Class 1.
-I think it undervalues the importance of EM. I feel that EM often adds to the diagnosis in lupus, especially in deciding whether a patient has class V, and in diagnosiing TMA. I think EM should be included as necessary in the classification, since I feel that centres, such as in the developing world that don't use EM, are providing suboptimal renal pathology service.
-I also feel that something should be done to improve the A - A/C - C designation. Could a definitition for what is considered a normal number of globally sclerosed glomeruli for age be made? This way the designation A/C would have more meaning. 
-Cases with a number of segmental and global lesions are problematic.  Judging whether chronic lesions are due to lupus also problematic, and counting chronic lesions as involved glomeruli probably ups the class and may result in overtreatment; detailed comments are in order.  I often give both new and old (1995 WHO)classification system.
-The nephrologists act on the basis of discussions at biopsy meetings, not on the basis of the classification. Putting chronic
sclerosing lesions into classes III and IV is potentially confusing to the nephrologists - decisions to treat are based on activity not on segmental or global sclerosis - I have to explain each time the difference between class IV (C) and class IV (A) and without discussion of each case, inappropriate treatment may be given.

REPLIES RECEIVED
n= 54

1. In which country do you practice?
Canada 5, Italy 1, Mexico 3, Netherlands 1, South Korea 1, UK 5, US 28, Venezuela 1, Spain 5, Croatia 1, Colombia 1, Turkey 1, Iran 1

2. Do you use the new ISN/RPS classification when reporting biopsies of lupus glomerulonephritis?
Yes   47  (8 also give WHO classification)
No  4
Yes with personal modifications 3
Comments
-Cases with heavy proteinuria, and lesions clase III o IV and subepithelial deposits, and/or extensive spikes, in less than 50% of glomeruli and/or less than 50% of the glomerular tuft in >50% of glomeruli, I comment about this feature and suggest combined class IV (or III) with class V lesions.
-Modified WHO 1995 in combination with ISN/RPS. For example, I do not use the old class I. In ISN Class III/IV I do not sub-classify. Instead, I describe in the report the details on glomerulosclerosis, e.t.c.
-I use the ISN/RPS classification, but in my report it is listed after the 1995 WHO classification. I do this because many of the nephrologists I work with (most are in private practice) are either unfamiliar with the ISN/RPS and its differences from the WHO, or they simply do not like the ISN/RPS classification. They regard it as a ‘pathologist’s thing’ not relevant to their practice
- In class III and IV, I am careful to communicate the proportion of active and chronic lesions as part of the diagnosis.  When a biopsy qualifies for III or IV because of the chronic component (IIIA/C, IIIC, IVA/C, and IVC, I do not put class III or IV in the diagnosis but include it in a note with a disclaimer (the patient qualifies because of the proportion of scarred glomeruli, and there are no active lesions ….).

3. Do you find any problems in applying the ISN/RPS classification (please feel free to go in to as much detail as you like!)
-All class 3 and class 4's end up being A/C or C. I can't tell if a globally sclerosed glomerulus is due to lupus or not, so i consider any globally sclerosed glom as due to lupus
- Need better categories for less typical SLE Dx, e.g., MCD or collapsing GN in SLE patients, or co-existing disease (doesn’t need to be overly specific) like diabetes or HTN.
-It is rather straightforward except for few problems. Distinction between diffuse segmental and diffuse global class IV – How much of the glomerulus is involved is often subjective. Criteria for combining membranous with class III and IV is also subject to differences since membranous component may be variable on same biopsy.
-The definitions and the classification scheme itself are well  described in the paper so I think we are OK at doing the  classification (we review our cases so we try to be consistent within our group). activity and chronicity are used semi-quantitatively which helps.
Leaves out activity and chroncity indices that my nephrologists like to see, are used to seeing and like to use to gauge therapy by.  So, I am still giving an activity and chronicity index for proliferative lesions. Based on the ISN/RPS version that I last saw or read about, Class V Membranous lupus GN had 4 subtypes. VC and VD are rather misleading- placing most emphasis on the membranous component and not proliferative activity.  I am still using "mixed" focal or diffuse prolif. + mem lupus GN in these cases. "Active", "Active/ Chronic" and "Chronic"- I use these, but feel that they do not measure the extent or degree of involvement. "Global" vs "Segmental"- sometimes, I see both lesions in the same biopsy  and don't know what to say, so I use both in the bottom line but have no idea what it means prognostically.
My main problem concerns knowing how to deal with globally sclerotic glomeruli (GSS). For instance, does the presence of GSS in an otherwise class II lesion automatically push this to a class III (C). Perhaps, in some instances it should and perhaps in others it should not. If there are fewer than 10% GSS, then maybe it should be attributed to aging or within the normal limits, but greater than this percentage would suggest that the scarring was due to lupus nephritis, perhaps? It seems that there is no easy answer for this, and I would be curious how you deal with this situation. Along those same lines, the presence of GSS can also complicate categorizing class IV lesions. Do the GSS factor into the determination of whether there are more segmental or globally involved glomeruli when you are deciding whether one is dealing with a class IV-S vs class IV-G? The distinction between IV-S and IV-G seems more feasible when a biopsy is predominantly active but the presence of significant chronicity makes this distinction much more difficult.  I also have some difficulty with classifying whether a biopsy has activity, chronicity or both. How much chronicity is necessary before one designates a case as III (C) or IV (C)? Is one or 2 GSS sufficient? Or if a biopsy shows predominantly chronic changes with one or two intact glomeruli or cellular crescents? Is this a chronic or both active and chronic lesion? Perhaps, if more explicit criteria are stated, these diagnoses could be made with better intra- and interobserver reproducibility.
-It seems kind of complicated mainly with lesions type  IV with segmental activity in some glomeruli and segmental chronic lesions in others.  The resulting diagnosis is too large for rapid comprehension. It continues not considering vascular lesions. The activity and chronicity indexes are the same and clinicians in types III and IV which are the most frequent types in my hospital only consider the indexes
-Mostly easy to apply, good definitions.  But still occasional problems:
1 How much focal segmental endotheliosis is required to shift class 2 to 3  ie what is the minimum requirement for an active lesion.
2 Can we have a name for a capsular reaction that is less than a crescent?
3 Borderline cases are unavoidable (not a criticism- this is true of any categorical classicification)
- What is unclear for me is about active/chronic subclassification of the class III and IV. When assessing activity index and chronicity index, we encount acute and chronic tubulointerstial changes as well as those of glomeruli. Should I consider tubulointerstitial changes when I subclassify A/C of class III and IV too?
-I have concerns that it may be misleading to clinicians to classify biopsies showing only sclerotic lesions as Class III or IV since these inactive lesions do not warrant aggressive immunosuppressive therapy. Also, since segmental sclerosis may develop in membranous lupus nephritis, I think it may be impossible to sort out Class III (C), Class IV-S(C), or Class IV-G (C) plus Class V from pure Class V with secondary sclerosing lesions
-Distinguishing diffuse global from diffuse segmental is not always easy to do; most diffuse cases are global in my experience.
-The problems with the classification are:
1. There is no ‘other’ category or ‘nos’ category to accomodate the not infrequent lesions not included in either the WHO or ISN/RPS classifications
2. Cases with significant glomerular chronicity features are difficult to classify
-In general I don't find any problems when applying the ISN/RPS classification for classes I, II, V. For classes III and IV ( s or g) sometimes the limitation of the sample size made it difficult. Also I find difficult to classify cases after treatment - in Mexico it is very frequent to have a second or third biopsy after treatment
- I think that it is cumbersome to separate numbers of segmental proliferative and global proliferative lesions.  Does one consider only glomeruli in paraffin sections, or should one include glomeruli in frozen and plastic sections?  How does one count a non-proliferative or segmentally proliferative glomerulus with a global crescent?
I think that trying to distinguish chronic ischemic glomerular obsolescence from glomerulonephritis induced obsolescence is sometimes subjective, and difficult to apply in practice. I have difficulty with biopsies that have numerous deposits but a disproportionately mild proliferative response; I feel that assigning a class based on proliferation rather than the quantity and location of deposits may undergrade patients that are about to undergo proliferation.  
-Cases with a number of segmental and global lesions are problematic.  Judging whether chronic lesions are due to lupus also problematic, and counting chronic lesions as involved glomeruli probably ups the class and may result in overtreatment; detailed comments are in order.  I often give both new and old (1995 WHO) classification system.
-Prevailing skepticism by most clinicians.
-It is much easier to apply than previous classification systems
-No, but the nephrologists act on the basis of discussions at biopsy meetings, not on the basis of the classification. Putting chronic
sclerosing lesions into classes III and IV is potentially confusing to the nephrologists - decisions to treat are based on activity not on segmental or global sclerosis - I have to explain each time the difference between class IV (C) and class IV (A) and without discussion of each case, inappropriate treatment may be given.

4. Which system do you use? (For the 4 respondents who did not use ISN/RPS
Original WHO Classification    1
1982 revision of WHO Classification
1995 revision of WHO Classification   2
Other – please state what 1
Comments
-Original WHO Classification ( with one modification of using and additonal class IV(6) to indicate advanced disease (more than 50% sclerotic glomeruli). The one reason I do not use the later modifications is because Classs III was made segmental which I think was unnecessary. Of course the ISN/RPS turned it back to original)
-We classify the biopsies in 5 groups or Class and we add sclerosis index:  0 to 12  (fibrous crescents 1-3, tubular atrofy 1-3, sclerosed glom. 1.3 and intersticial sclerosis 1-3) and activity index:  0 to 24 (modification by Han 1990 of the Morel Maroger index 1976). (fibrinoid necrosis 0-6, new epithelioid crescents 0-6, hipercel.lularity 0-3, glomerulitis 0-3, interstitial infiltrate 0-3)

5 What is your reason for not using the ISN/RPS classification
I find it unsatisfactory compared with the system I use
My clinicians prefer another system
Other – please state what

-One respondent said ‘All of the above’
-One said ‘My clinicians prefer another system’.
-One said ‘I'll not change the classification if there is not a previous agreement with the clinicians. At the moment I am not aware of the benefit of this new classification’
-One said Too complicated.  A lot of subgroups ineach grade.  Too difficult grade of reproducibility between observers. In order to use, our pathology service would need special training in nephropathology!!

6. Whether you use the ISN/RPS classification or not, please describe how well you think it serves its purpose and how it could be improved
-I think it is much better than WHO - and much simpler to use. The clinicians in Toronto are satisfied with it and I like it!. I think the category of IV-S is very useful and probably has real clinical meaning. I like that a normal biopsy does not classify as Class 1.
I think it undervalues the importance of EM. I feel that EM often adds to the diagnosis in lupus, especially in deciding whether a patient has class V, and in diagnosiing TMA. I think EM should be included as necessary in the classification, since I feel that centres, such as in the developing world that don't use EM, are providing suboptimal renal pathology service.
I also feel that something should be done to improve the A - A/C - C designation. Could a definitition for what is considered a normal number of globally sclerosed glomeruli for age be made? This way the designation A/C would have more meaning. 
-Works very well. Standard use of the NIH activity and chronicity indices would be better than the descriptive system, especially for quickly communicating the disease status.
-It helps to provide consistency in diagnosing lupus. Whether subdividing class IV into segmental vs global is relevant should be determined. A discussion regarding the effect of treatment on lupus histology, fluor, and EM would be helpful.
We do not like the mixing of the chronic lesions with the acute  within the scheme itself. For example, Class III (A) and Class III 
(C) both come out as class III to the clinician. even with the  activity and chronicity stated, the treatment message is perhaps 
different between III (A) and III (C). Class IV is even worse in this  regard with the segmental v global and the active v chronic. We think  Class VI is problematic in that 90% is too high a percentage. Not  sure what is the right way to handle the severely chronic lupus  lesions. Obviously an area needing more than a 'working  classification'; in other words, more data needs to be gathered. Membranous lupus GN (Class V)- I think that the division into ABCD subclasses is confusing to clinicians- "mixed" III or IV+ V is understood by them best. I am not sure if it serves it's purpose, yet.  It has not been in practice long enough by enough renal pathologists to see (lesions by this classification scheme vs. outcome). Overall, I think that the new classification is confusing to many nephrologists
- I think that the ISN/RPS classification system works well, and it is easier to understand the abbreviations in comparison with the previous WHO system. I think that it would be improved if the issues mentioned in #3 could be addressed.
-It serves in the mesangial type because no doubt is left with a proliferative lesion with low activity with only mesangial proliferation. It defines better the inclusion criteria for each type or subtype. It is good for students, residents and nephrology fellows to understand all the broad spectrum of lesions that can be found in lupus. Could be improved trying to do descriptions of the microscopical findings and the diagnosis in synopsis form using a check list or guidelines for reporting. Clinicians need more information about the utility of this system for management of their patients
-I think it gives a large amount of information in more detail than the last classification, whilst still being similar enough to be understandable to the clinicians.  The clinicians were not used to being given any classification previously (last renal pathologist did not classify) and used to have to convert old report to WHO classification, many still convert the new to the old but probably find it easier.  They will get used to it. Much better than previous systems but see above
-I think the new classification serves it purpose well. From a pathologists point of view it is a logical, easy to remember classification, with clear cut -off points (in the old classification it was difficult to chose from the class IV subclasses and calling a biopsy without abnormalities a class I LN was strange) . Also, the classification itself (together with the activity/chronicity score) nicely describes what is seen in the biopsy.
-The classification works well. I have no suggestions for improvement.
-In the recommendations for reporting, there are many things that should be included. It would be helpful to provide an actual diagnosis format including description of all recommendations for a junior renal pathologist like me
-I feel that the system should clarity and place greater emphasis on the degree of activity in the biopsy because this factor is of paramount clinical significance
-In order to be realistic, a lupus classification must take into account the immuno and EM findings of the glomeruli.   Furthermore, tubulo-interstitial (and vascular) changes need to be integrated also. Integrating ALL markers of activity results in a scheme that clinicians can follow to decide on therapy.
-I feel that the schema addressed most of the important issues.  Would include criteria for adequacy of a biopsy to accurately categorize lupus GN.
-I feel it works fairly well, and I am not sure how it could be improved upon. I think the clinicians still think in terms of WHO. It unifies criteria for the classification of lupus nephritis, this, of course, will render benefits at the time of evaluation of therapies and prognosis.  I think, however, that some weigh should have been given to the interstitial changes
-I am in the camp that believes that segmental proliferative lupus is different than diffuse / global proliferative lupus. Placing a lesion with 45% segmental lesions in one class, and placing a case with 55% otherwise identical segmental lesions in another class with cases that are morphologically, and often immunohistological different, is fundamentally illogical and unscientific. The fact that the ISN/RPS must include qualifiers for class IV to distinguish the segmental from the global lesions seems itself a declaration that this is a conceptionally confused approach.
-Serves no new purpose, cumbersome (Class IV has six subdivisions!!), misleading; e.g of misleading - if Bx had 10 glomeruli, 8 globally sclerotic, 1 with segmental sclerosis and the last one normal, this would be a Class IV. And surprisingly there is not an iota of activity!!!!! The connotation of Class IV has always and still is bad active disease.
As mentioned, serves no new purpose, confusing to clinician and cumbersome. Has no advantage over original WHO Classification, refer (Reclassification of Lupus Glomerulonephritis: Back to the Future Richard J. Glassock. JASN 15 501). I think ISN/RPS Classification should be scrapped. I think the original WHO Classification with modifications is the way to go. Here are some modifications I would like to see to the Original WHO Classification and would like (expected) ISN/RPS to resolve.
I think ClassVI(6) should be added to the original WHO Classification. I believe that subendothelial deposits (if identified by EM) should be unequivocally used for classification. (The issue that everybody does not have access to  EM is understood.)
 Severe mesangial hypercellularity ( happens occasionally ) also should be added as criterion for inclusion in original WHO ClassIII/IV
Important: ( Activity and chronicity scores should be revisited and polished as an adjunct to WHO Classification.) E.g. exclude unnecessary features like interstitial inflammation, neutrophils infiltration. These just dilute from the activity score in most cases. Add EM subendothelial deposits. Make important features like crescents, necrosis, subendothelial deposits independent features to determine activity. (Clarification of last statement: Suppose there is only WHO Class 2 features and EM shows widespread subendothelial deposits, then this patient requires active treatment. We all know this, but the activity index does not reflect this. The patient may only get a mild activity index score. Another example, if one out of 10 glomeruli has a prominent segmental necrotizing lesion without crescent (yet)  the activity index would be mild only
- As I have mentioned, the changes after treatment are sometimes difficult to asses because changes do not fullfill a specific category. I think vascular and tubulointerstictial changes should be incorporated
-I think the rationale incorporated into the new categories in the ISN/RPS makes sense in terms of the biology of lupus nephritis.  I think that the application of the classification is somewhat cumbersome and more time consuming
-I am still waiting for information on segmental [S} vs.  global [G] involvement.
-I believe in the qualitative and not quantitative value of new classification
-I await outcome studies. 
- I was trained by Dr. Pirani-Pollak in Chicago in the late ‘60’s, and became extremely well acquainted with the original system.  I guess I’ve been quite successful in instilling my clinicians of +40 years, great confidence in the original classification. The system allows for segmental sclerosing lesions and for the presence of mixed acute and chronic glomerular lesions in the same glomeruli/biopsy. This is a considerable improvement over previous classification systems. The class VI category has been greatly narrowed which in my opinion is a good thing since my more aggressive clinicians have reported good responses to therapy despite my inclusion of the biopsy in this class using the older system