Abstract Lupus nephritis (LN) class II has generallybeen considered a mild form of LN with a good response totreatment. Although the number was small, there have alsobeen reports on later progression to class III or IV, resultingin poor renal and patient outcome. This study aims toreview cases of LN class II to analyze diVerences betweencases that progressed to class III or IV and cases that didnot. We retrospectively analyzed 15 cases of LN class IIamong 277 cases of biopsy-proven lupus nephritis diag-nosed in a tertiary medical center over about 14 years.Among the 15 patients, 5 patients progressed to class III orIV. Biopsy specimens were reviewed by a pathologistaccording to the ISN/RPS 2003 classiWcation. Response totreatment was evaluated at 6 months after treatment. OnXuorescence microscopy (IF), there was signiWcantlyhigher degree of deposition in the glomeruli of IgM, IgAand C4 in the progression group than in the non-progressiongroup. At 6 months after treatment, there was a trend towardhigher rates of complete remission in the non-progressiongroup (90%) compared with those in the progression group
(40%, p = 0.077). Five of the 15 cases of ISN/RPS 2003class II glomerulonephritis progressed to class III or IVover a mean of 5 years. The degree of immune-complexdeposition for IgM, IgA and C4 in the glomeruli wassigniWcantly higher in the progression group.
The lupus nephritis (LN) class II, recently designated asmesangial proliferative lupus nephritis in the ISN/RPS2003 criteria, has generally been considered a mild form ofLN, with a reported 10-year renal survival rate of 100% [1].However, although the number was small, there have beenreports on later progression to focal and diVuse lupusnephritis, resulting in poor renal and patient outcome [2].
Until now, neither the exact rate of nor the predictors forthe progression have been clearly revealed. Only one study,in which 9 out of 19 patients showed poor prognosis suchas progression to class III or IV, reported poor response toinitial therapy at 1 year as a predictor for the progression[3]. In accordance with this reality, the subcategories, basedon the degree of mesangial hypercellularity, of the previousWHO 1982 criteria were abandoned in the ISN/RPS 2003classiWcation criteria, because there had been no study toprovide evidence for a diVerence in prognosis according tothe degree of mesangial hypercellularity [4].
Most reports on LN class II to date have had limitationsoriginating from a small sample size, inhomogeneous deW-nition of progression (such as inclusion of membranousglomerulonephritis or death) and disproportionate numberof patients who underwent follow-up biopsy. In this
S. G. Lee · M. W. So · Y.-G. Kim · C.-K. Lee · B. Yoo (&)Division of Rheumatology, Department of Internal Medicine, University of Ulsan College of Medicine, Asan Medical Center, 388-1 Pungnap-dong, Songpa-gu, Seoul, 138-736, Republic of Koreae-mail: [email protected]
Y. M. ChoDepartment of Pathology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Republic of Korea
S. S. KimDivision of Rheumatology, Department of Internal Medicine, University of Ulsan College of Medicine, Asan Medical Center, Gangneung, Republic of Korea
123
2460 Rheumatol Int (2012) 32:2459–2464
context, we reviewed cases of LN class II diagnosed in asingle tertiary center over about 14 years to identify casesthat progressed to class III or IV and to compare clinical,laboratory and pathologic diVerences between the pro-gressed cases and cases that did not. We especially focusedon the diVerences in mesangial hypercellularity, which isnow clearly deWned as the presence of at least three mesan-gial cells per mesangial area, and small amount of membra-nous immune deposits on transmission electronmicroscopy, which is stipulated to be allowed in class II, inthe ISN/RPS 2003 criteria.
Patients and methods
Patients
We retrospectively reviewed 277 cases of biopsy-provenlupus nephritis in adults (aged ¸16 and <75) according toeither the WHO 1982 or the ISN/RPS 2003 criteria, diag-nosed in a tertiary medical center from July 1995 throughJanuary 2009. Twenty cases compatible with ISN/RPS2003 LN class II with available biopsy specimens wereidentiWed, which fulWlled the 1997 American College ofRheumatology classiWcation criteria for systemic lupuserythematosus. With 5 cases excluded, in which less than10 glomeruli were available on the biopsy specimen, a totalof 15 patients were included in this study. Initial biopsieswere performed according to generally accepted indicationsfor kidney biopsy in lupus patients in routine care settings.These included the presence of proteinuria deWned as either>300 mg/day on 24 h urine collection or >30 mg/g on albu-min–creatinine ratio in random urine sample, presence ofRBC casts or glomerular hematuria and presence of azote-mia.
Four patients had been taking hydroxychloroquine at adose of 400 mg/day at initial biopsy. Two patients had beenon prednisolone, with one patient at a dose of 7.5 mg/dayfor control of skin manifestation and the other 17.5 mg/day,who had been on tapering dose of prednisolone after receiv-ing high-dose prednisolone for lupus enteritis. None of thepatients was taking either NSAIDS or ACEI/ARB at thetime of the initial biopsy.
Among the 15 patients, 5 patients underwent follow-upbiopsy of the kidney a mean (§SD) of 58 (§17) monthsafter the initial biopsy, which revealed that the disease hadprogressed to class III (n = 3, two with class V) or IV-S(n = 2, one with class V). Indications for the follow-upbiopsy were new development of nephritic syndrome inthree cases, an increase in proteinuria in one and active uri-nary sediment without nephritic syndrome in one. Theother 10 patients were regularly followed up for a mean(§SD) of 90 (§42) months after the initial biopsy without
clinical evidence for relapse of nephritis. Two patients werelost to follow-up, with one patient 49 months and the other23 months after the initial biopsy. At the follow-up biopsyor last visit, patients were on 5 (§0) and 3 (§3) mg of pred-nisolone in the progression group and non-progressiongroup, respectively. All patients except a patient in the non-progression group were taking hydroxychloroquine indoses of 210 (§100) mg in the progression group and 240(§90) mg in the non-progression group. One patient in theprogression group and two in the non-progression groupwere taking maintenance dose of mycophenolate mofetil of1.5 g/day.
Clinical and laboratory Wndings
Records of demographic and clinical information such asblood pressure, fever, CNS manifestation, arthritis, skinrash, oral ulcer, pleuritis, pericarditis and anemia at thetime of the initial biopsy were reviewed. Disease durationwas deWned as the interval between the diagnosis of lupusand the initial biopsy. Follow-up duration was from the ini-tial biopsy to the progression or last observation. Diseaseactivity was assessed using systemic lupus erythematosusdisease activity index (SLEDAI) [5]. Baseline serologicWndings such as the presence of anti-RNP, Sm, Ro and Laantibody and laboratory Wndings at the initial biopsy suchas hemoglobin, white blood cell count, platelet count, ESR,CRP, C3, C4, anti-dsDNA Ab, creatinine, estimated GFR,albumin, 24-h urine protein, and urine RBC and proteinwere retrospectively retrieved from electronic medicalrecords.
DeWnition of complete remission
Complete remission at 6 months after treatment was deW-ned as Cr < 1.5, absence of proteinuria (<300 mg/day ornegative dip stick test) and active urinary sediment(RBC < 5/HPF, WBC < 5/HPF, no RBC cast).
Review of biopsy specimen
An experienced pathologist who was blinded to the clinicaloutcome reviewed pathology materials of initial and fol-low-up biopsies according to the ISN/RPS 2003 criteria.Pathology materials for each biopsy consisted of four spe-cial stained glass slides for light microscopy (LM) exami-nation (hematoxylin–eosin, periodic acid SchiV, periodicacid–silver methenamine and Masson trichrome staining),immunoXuorescence study (IF) for IgG, IgM, IgA, C3, C4and C1q, and transmission electron microscopy study(EM). During the review of the initial biopsies, the propor-tions of glomeruli with mesangial expansion and hypercell-ularity were assessed on LM. The activity and chronicity of
123
Rheumatol Int (2012) 32:2459–2464 2461
the lesion were assessed by NIH scoring system [6] for thefollow-up biopsies. On IF, the degree of mesangial immunedeposition was graded from 0 to 4, according to the inten-sity of Xuorescence signals [7]. Due to the relatively scantyand weak nature of immune-complex deposition in LNclass II, the peripheral immune deposition was described asabsent or present. On EM, the presence of electron-densedeposits for each case was evaluated for mesangial, subepi-thelial, subendothelial and intramembranous areas.
Statistical analysis
Continuous variables were described as a mean § SD or amedian and range depending on the distribution of values.Categorical variables were expressed as a number and per-centage. Values between groups were compared usingMann–Whitney U test and Fisher’s exact test for continu-ous and categorical variables, respectively. Two-sidedp-values of <0.05 were considered statistically signiWcant.Statistical analysis was performed using SPSS version 16.0software (SPSS, Chicago, IL).
Results
Clinical and laboratory Wndings
The mean ages of patients were 41 (§18) and 36 (§14) inthe progression and non-progression group, respectively.The proportion of women was 80% in both groups. Themedian duration of disease before the initial biopsy was 19(4–60) months in the progression group and 2.5 (1–61)months in the non-progression group (p = 0.049). Therewas no diVerence in the duration of follow-up between thetwo groups. Other clinical variables are summarized inTable 1. On laboratory Wndings, there was no signiWcantdiVerence between the two groups in C3, C4, anti-dsDNAAb and creatinine. Baseline serologic Wndings and otherlaboratory Wndings are shown in Table 2.
Pathologic Wndings
On LM, all cases showed mesangial expansion. The pro-portion (%) of glomeruli with mesangial hypercellularitywas not diVerent between the two groups (36% in the pro-gression group and 34% in the non-progression group,p = 0.902). On IF, mesangial immune deposition for IgM,IgA and C4 was signiWcantly higher in the progressiongroup than in the non-progression group (Fig. 1). In con-trast, mesangial IgG and C3 depositions were not diVerentstatistically between the two groups although two cases ofthe non-progression group showed no deposition of IgG.There were no signiWcant diVerences in the proportion of
cases with glomerular peripheral immune deposits for IgG,IgM, IgA, C3, C4 and C1q. On EM, there were no signiWcantdiVerences in the proportion of cases with subendothelial and
cases with subepithelial immune deposit between the twogroups (Table 3).
In the follow-up biopsy, 3 patients were diagnosed withclass III and 2 patients class IV-S with 3 patients havingconcomitant class V. All patients revealed increasedamount of mesangial and peripheral immune depositionscompared to those of initial biopsies in both IF and EMstudies, compatible with the progression of lupus nephritis.In no cases, were the Wndings found which are consideredto be suggestive of pauci-immune glomerulonephritis [8],including Wbrinoid necrosis without subendothelial hyper-cellularity and immune-complex deposition (Table 4).
Response to treatments
All patients started initial treatment with medium dose ofglucocorticoid (35 § 14 mg and 32 § 16 mg in the pro-gression and non-progression groups, respectively) [9].There was no signiWcant diVerence in the mean cumulativedoses of prednisolone or its equivalent between the twogroups (3.2 § 0.9 g and 3.3 § 2.0 g in the progression andnon-progression group, respectively, p = 0.903). Rate ofcomplete remission at 6 months after the initial treatmentwas 90% in the non-progression group and 40% in the pro-gression group (p = 0.077).
Discussions
In this review, we found a cumulative incidence rate of33% for the progression of LN class II to class III or IVduring a mean follow-up of 58 months. In addition, themesangial deposition of IgM, IgA and C4 on IF was signiW-cantly greater in the progression group than in the non-pro-gression group. Rate of complete remission at 6 months
after initial treatment tended to be higher in the non-pro-gression group.
In studying the progression of LN class II to higher clas-ses, it is important to exclude cases with subtle endo- orextracapillary proliferation at baseline, to deWne “progres-sion” clearly and to detect it in a sensitive way. Among themain themes of the ISN/RPS 2003 classiWcation are theemphasis on the importance of endo- or extracapillary pro-liferation, regardless of its degree, and the interpretation ofsclerotic glomeruli which are considered sequela of endo-capillary proliferative LN as focal or diVuse LN. As a
Fig. 1 IF Wndings of representative cases from the progression andnon-progression group are shown. The case of the progression grouphad grade 2 deposition for IgM and C4 and grade 3 for IgA in contrastwith the case of the non-progression group, which had grade 1 depo-sition for IgM and IgA
Table 3 Pathologic (LM, IF, EM) Wndings of the initial biopsy
a Regarding IF, because one case with images of poor quality wasexcluded, nine cases were analyzed in the non-progression group
Progression (n = 5)
Non-progression (n = 10)
p-Value
Light microscopy
Number of glomerulus 16 § 5 14 § 4 0.324
Number of glomeruli with mesangial expansion (%)
5 (100) 10 (100) 1.000
Number of glomeruli with mesangial hypercellularity (%)
36 § 32 34 § 22 0.902
ImmunoXuorescence microscopya
Mesangium
IgG 1.4 § 0.9 1.1 § 0.9 0.533
IgM 2.0 § 0 0.9 § 0.6 0.004
IgA 1.8 § 0.4 0.6 § 0.7 0.011
C3 1.8 § 0.4 1.3 § 0.7 0.095
C4 0.9 § 0.6 0 § 0 0.049
C1q 1.0 § 0.7 0.8 § 0.7 0.101
Glomerular capillary
IgG 2 (40) 5 (55.6) 1.000
IgM 2 (40) 1 (11.1) 0.505
IgA 3 (60) 1 (11.1) 0.095
C3 2 (40) 4 (44.4) 1.000
C4 1 (20) 0 (0) 0.357
C1q 0 (0) 2 (22.2) 0.505
Electron microscopy
Mesangial electron-dense deposits
5 (100) 10 (100) –
Subepithelial electron-dense deposits
5 (100) 5 (50) 0.101
Subendothelial electron-dense deposits
2 (40) 5 (50) 1.000
Intramembranous electron-dense deposits
0 (0) 0 (0) –
123
Rheumatol Int (2012) 32:2459–2464 2463
result, in the ISN/RPS 2003, the number of cases diagnosedwith class II decreases [10]. In this study, in order toinclude cases in strict accordance with the ISN/RPS 2003classiWcation, cases with less than ten glomeruli wereexcluded, which is considered the minimal number ofglomeruli to exclude focal lesions with appropriate sensi-tivity [11]. In addition, the small proportion of class IIamong patients diagnosed with lupus nephritis (15/277,5%) indirectly supports that pure cases with LN class IIwere included in this study.
In this study, the progression was deWned as class IIIor IV with or without class V. Generally, the prolifera-tive lesions of class III and IV are known to be moreimportant than lesions of class V, because prognosis isworse for and largely determined by the proliferativelesions of class III or IV rather than class V [12, 13].This is one of the reasons why the importance of capil-lary proliferation is emphasized in the ISN/RPS 2003classiWcation.
Although follow-up biopsy of the kidney was not per-formed in all cases initially diagnosed with LN class II, itis very unlikely that cases which had progressed to classIII or IV went undetected in the routine care settings,where Wndings known to be associated with relapse ofrenal diseases such as blood pressure, creatinine, C3 andC4 are regularly checked. Therefore, it can be said thatalthough the number was small, patients included in thisstudy are representative cases of LN class II, with the pro-gression group reXecting meaningful outcome and beingdetected in an appropriate way.
On IF, lupus nephritis is characterized by abundantdeposition of polyclonal IgG, C3 and C1q with variableamount of IgM, IgA and C4, as observed in this study [4].In addition, there was signiWcantly greater mesangial depo-sition of IgM, IgA and C4 in the progression group.Although there are not uniformly accepted criteria fortypical Wndings of immune-complex glomerulonephritiscompared with glomerular lesions of possibly other patho-genesis, such as pauci-immune glomerulonephritis, the 5progressed cases had pathologic Wndings that can be con-sidered consistent with immune-complex glomerulonephri-tis [8]. One of the explanations for the pathogenesis oflupus nephritis is the immune-complex glomerulonephritismodel, in which the proliferative glomerulonephritis ofclass II through IV is understood as a single spectrum ofimmune-complex glomerulonephritis, with each class rep-resenting distinct position on the spectrum [14]. Given theIF Wndings of this study, within the class II, there may beseparate groups of cases, which are distinguished by diVer-ent degrees of immune deposition and corresponding sus-ceptibility to later progression to higher classes. Because itis the basic concept of the immune-complex glomerulone-phritis theory that greater degree of immune depositionleads to more severe proliferation, the lack of diVerence inthe degree of mesangial hypercellularity between the twogroups may seem incongruent with this assumption. How-ever, due to the lack of power which originates from thesmall sample size of this study, it remains to be seenwhether diVerence in mesangial hypercellularity is a pre-dictor of the progression to more severe proliferativelesions.
This study has several limitations. Disease durationbefore the initial biopsy was longer in the progressiongroup. Longer disease duration may be associated withmore severe clinical, laboratory, pathologic Wndings, poorresponse to treatment and poor renal outcome [15], deviat-ing diVerences between the two groups against null values.However, patients included in this study had relativelyearly lupus, with median disease durations of both groupsless than two years. Together with the meticulous exclusionof subtle capillary proliferative lesions, it is very unlikelythat the diVerence in the disease duration signiWcantlyaVected the diVerence in the outcome between the twogroups.
Because of the relatively small number of class II com-pared with other classes in LN and the strict exclusion ofpotentially endo- or extracapillary proliferative lesions,only 15 cases were included in this study. As a result, dueto the high probability of type II error, lack of statisticalsigniWcance in no way guarantees the equivalence ofpopulation values. In the case of results that reachedstatistical signiWcance, because the two groups were com-pared for many variables without correction for multiple
Table 4 Pathologic (LM, IF, EM) Wndings of the follow-up biopsy
a In case number 1, the result of electron microscopy was not available
Case number 1 2 3 4 5
Light microscopy
ISN/RPS 2003 class III + V III + V IV-S IV-S + V III
Activity index 7 3 10 12 5
Chronicity index 0 5 0 1 6ImmunoXuorescence microscopyGlomerulus
IgG 2 2 1 1 1
IgM 3 1 1 1 1
IgA 3 0 1 1 2
C3 3 2 2 2 1
C4 2 1 1 1 1
C1q 2 1 1 1 1
Electron microscopya
Mesangial immune deposit + + + +
Subepithelial immune deposit + + + +
Subendothelial immune deposit ¡ + + +
Intramembranous immune deposit
+ ¡ + +
123
2464 Rheumatol Int (2012) 32:2459–2464
comparisons, the possibility of type I error cannot be com-pletely excluded.
In conclusion, ISN/RPS 2003 class II included a sub-group of cases that progressed to class III or IV. Immunedeposition of mesangial IgM, IgA and C4 was greater in theprogression group. Treatment response to medium-doseglucocorticoid tended to be lower in the progression group.Because there might be chances that the progression groupbeneWts from more intensive immunosuppressive treat-ments, further studies of larger sample sizes evaluatingother predictors for progression are needed.
ConXict of interest None of the authors declared any conXicts ofinterests.
References
1. Appel GB, Cohen DJ, Pirani CL, Meltzer JI, Estes D (1987) Long-term follow-up of patients with lupus nephritis. A study based onthe classiWcation of the world health organization. Am J Med83:877–885
2. Stankeviciute N, Jao W, Bakir A, Lash JP (1997) Mesangial lupusnephritis with associated nephrotic syndrome. J Am Soc Nephrol8:1199–1204
3. Tam LS, Li EK, Lai FM, Chan YK, Szeto CC (2003) Mesangiallupus nephritis in chinese is associated with a high rate of transfor-mation to higher grade nephritis. Lupus 12:665–671
4. Weening JJ, D’Agati VD, Schwartz MM, Seshan SV, Alpers CE,Appel GB et al (2004) The classiWcation of glomerulonephritis insystemic lupus erythematosus revisited. Kidney Int 65:521–530
5. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH(1992) Derivation of the sledai. A disease activity index for lupus
patients. The committee on prognosis studies in sle. ArthritisRheum 35:630–640
6. Austin HA 3rd, Muenz LR, Joyce KM, Antonovych TT, Balow JE(1984) DiVuse proliferative lupus nephritis: identiWcation ofspeciWc pathologic features aVecting renal outcome. Kidney Int25:689–695
7. Hill GS, Delahousse M, Nochy D, Tomkiewicz E, Remy P,Mignon F et al (2000) A new morphologic index for the evaluationof renal biopsies in lupus nephritis. Kidney Int 58:1160–1173
8. Hill GS, Delahousse M, Nochy D, Bariety J (2005) Class iv-s ver-sus class iv-g lupus nephritis: clinical and morphologic diVerencessuggesting diVerent pathogenesis. Kidney Int 68:2288–2297
9. Buttgereit F, da Silva JA, Boers M, Burmester GR, Cutolo M, JacobsJ et al (2002) Standardised nomenclature for glucocorticoid dosagesand glucocorticoid treatment regimens: current questions and tenta-tive answers in rheumatology. Ann Rheum Dis 61:718–722
10. Furness PN, Taub N (2006) Interobserver reproducibilityand application of the isn/rps classiWcation of lupus nephritis-aUK-wide study. Am J Surg Pathol 30:1030–1035
11. Corwin HL, Schwartz MM, Lewis EJ (1988) The importance ofsample size in the interpretation of the renal biopsy. Am J Nephrol8:85–89
12. Sloan RP, Schwartz MM, Korbet SM, Borok RZ (1996) Long-term outcome in systemic lupus erythematosus membranous glo-merulonephritis. Lupus nephritis collaborative study group. J AmSoc Nephrol 7:299–305
13. NajaW CC, Korbet SM, Lewis EJ, Schwartz MM, Reichlin M,Evans J (2001) SigniWcance of histologic patterns of glomerularinjury upon long-term prognosis in severe lupus glomerulonephri-tis. Kidney Int 59:2156–2163
14. Ginzler EM, Nicastri AD, Chen CK, Friedman EA, Diamond HS,Kaplan D (1974) Progression of mesangial and focal to diVuselupus nephritis. N Engl J Med 291:693–696
15. Faurschou M, Starklint H, Halberg P, Jacobsen S (2006) Prognos-tic factors in lupus nephritis: diagnostic and therapeutic delayincreases the risk of terminal renal failure. J Rheumatol 33:1563–1569
