Clinicopathological characteristics and outcomes of patients with crescentic lupus nephritis
2009; Elsevier BV; Volume: 76; Issue: 3 Linguagem: Inglês
10.1038/ki.2009.136
ISSN1523-1755
AutoresFeng Yu, Ying Tan, Gang Liu, Suxia Wang, Wan-zhong Zou, Ming‐Hui Zhao,
Tópico(s)Renal Diseases and Glomerulopathies
ResumoThere are few clinicopathologic and outcome data on patients with crescentic lupus nephritis, therefore, we determined factors of the disease by retrospectively reviewing the records of 327 patients diagnosed with lupus nephritis. Of these, 152 cases were regrouped as class IV-G, including 33 patients with crescentic glomerulonephritis. Significantly, all patients with crescentic glomerulonephritis had acute kidney injury as compared with only about a quarter of the patients without the disease. On pathological evaluation, activity scores, chronicity indexes, relapse rates, and the frequency of positive serum anti-neutrophil cytoplasmic antibody (ANCA) were each significantly higher, whereas complete remission rates and renal outcomes, over a mean follow-up of 4 years, were significantly poorer in patients with crescentic glomerulonephritis. Our study shows that crescentic glomerulonephritis was not rare in patients with lupus nephritis and that their long-term outcome was poor. The precise role of ANCA in the pathologic course of crescentic lupus nephritis remains to be determined. There are few clinicopathologic and outcome data on patients with crescentic lupus nephritis, therefore, we determined factors of the disease by retrospectively reviewing the records of 327 patients diagnosed with lupus nephritis. Of these, 152 cases were regrouped as class IV-G, including 33 patients with crescentic glomerulonephritis. Significantly, all patients with crescentic glomerulonephritis had acute kidney injury as compared with only about a quarter of the patients without the disease. On pathological evaluation, activity scores, chronicity indexes, relapse rates, and the frequency of positive serum anti-neutrophil cytoplasmic antibody (ANCA) were each significantly higher, whereas complete remission rates and renal outcomes, over a mean follow-up of 4 years, were significantly poorer in patients with crescentic glomerulonephritis. Our study shows that crescentic glomerulonephritis was not rare in patients with lupus nephritis and that their long-term outcome was poor. The precise role of ANCA in the pathologic course of crescentic lupus nephritis remains to be determined. Renal involvement is common in patients with systemic lupus erythematosus (SLE). The clinical presentation of lupus nephritis is highly variable, ranging from mild asymptomatic proteinuria to rapidly progressive glomerulonephritis, and the occurrence of kidney disease is the most important predictor of morbidity and mortality in patients with SLE.1.Appel G.B. Radhakrishnan J. D'Agati V.D. Secondary glomerular disease.in: Brenner B.M. Brenner & Rector's The Kidney. 8th ed. WB Saunders, Philadephia2008: 1067-1146Google Scholar Concerning the classification of lupus nephritis, a new classification was proposed by the International Society of Nephrology and Renal Pathology Society (ISN/RPS) in 20032.Weening J.J. D'Agati V.D. Schwartz M.M. et al.The classification of glomerulonephritis in systemic lupus erythematosus revisited.Kidney Int. 2004; 65: 521-530Abstract Full Text Full Text PDF PubMed Scopus (971) Google Scholar to rectify some of the problems that have arisen over the earlier years. Diffuse global lupus nephritis (class IV-G) was defined as more than 50% of glomeruli display endocapillary or extracapillary glomerulonephritis. In this class, patients with crescentic glomerulonephritis, defined as more than 50% of glomeruli, had crescent formation on light microscopy, and always presented with rapidly progressive glomerulonephritis clinically. However, the ratio of crescentic glomerulonephritis in patients with lupus nephritis class IV-G is not clear and the pathogenesis remains inconclusive. Owing to the relatively limited earlier studies on crescentic lupus nephritis, the purpose of this study was to assess the clinical and pathological features, outcome and related autoantibody profile of crescentic glomerulonephritis in a cohort of Chinese patients with lupus nephritis. Among the 327 patients with lupus nephritis in Peking University First Hospital, 152 cases were reclassified as class IV-G. In total, 33 of the 152 patients (21.7%) also fulfilled the diagnosis of crescentic glomerulonephritis. In patients with crescentic glomerulonephritis, 28 were female and 5 were male with an average age of 30.9±10.3 (14 to 57) years at presentation. All the patients presented with rapidly progressive glomerulonephritis clinically. The mean value of serum creatinine was 3.74±2.68 mg/dl (range 1.7–11.4 mg/dl) on diagnosis. On renal pathological evaluation of patients with crescentic glomerulonephritis, the average percentage of glomeruli with crescents was 66.23±16.76% (range 51.2–95.7%). For constituent of crescents, the average ratio of cellular, fibro-cellular, and fibrous crescents was 60.8±35.9% (range 0–100%), 29.5±30.6% (range 0–100%), and 9.7±20.9% (range 0–100%), respectively. The 33 patients with crescentic glomerulonephritis received intravenous methylprednisone pulse (500–1000 mg for three days) therapy. The majority of patients completed induction treatment with oral prednisone (0.8–1 mg/kg per d or equivalent for 4–6 weeks and tapered gradually to a maintenance 5–10 mg/d) (33/33) and monthly intravenous cyclophosphamide (600–800 mg/month) (30/33) for at least 6 consecutive months. The other three patients received mycophenolate mofetil. Eight patients achieved complete remission, 16 had partial remission, and nine failed to respond. Among the 24 patients with remission, eight patients reached secondary end point (two with end-stage renal disease and six with doubling of serum creatinine) during follow-up. The median time between remission and end point was 18 months (2–141 months). Among the nine patients with treatment failure, eight patients reached secondary end point (five with end-stage renal disease and three with doubling of serum creatinine). The proportion of patients reaching renal end point was significantly lower in remission group than in the non-remission group (8/24 vs 8/9, P=0.007, Figure 1). In the remission group, nine patients had a relapse. The proportion of patients reaching renal end point was significantly lower in the non-relapse group than in the relapse group (8/9 vs 0/15, P<0.001, Figure 1). We further compared the clinical and pathological characteristics of patients with and without crescentic glomerulonephritis in class IV-G. The clinical and laboratory features of patients in the two groups were listed in Table 1.Table 1Comparisons of clinical data and laboratory data between patients with lupus nephritis class IV-G with and without crescentic glomerulonephritisCrescentic glomerulonephritisNon-crescentic glomerulonephritisP-valueClinical dataNumber of patients33119Gender (male/female)5/2821/980.736Age (mean±s.d.) (years)30.9±10.331.5±10.80.774The time between presentation/diagnosis of lupus nephritis and biopsy/treatment (mean±s.d. and range) (months)16.8±28.8 (0.5–120)14.2±23.4 (1–110)0.653Hypertension (blood pressure≥140/90 mm Hg), n (%)21 (63.6)68 (57.1)0.503Systolic blood pressure (mean and range) (mmHg)140 (110–160)140 (110–150)0.768Diastolic blood pressure (mean and range) (mmHg)80 (80–90)80 (70–90)0.912Fever, n (non-infectious) (%)8 (24.2)31 (26.1)0.833Malar rash, n (%)19 (57.6)57 (47.9)0.432Photosensitivity, n (%)6 (18.2)20 (16.8)0.8Oral ulcer, n (%)8 (24.2)32 (26.9)0.827Alopecia, n (%)10 (30.3)34 (28.6)0.831Arthralgia, n (%)21 (63.6)54 (45.4)0.077Serositis, n (%)7 (21.2)22 (18.5)0.803Neurological disorder, n (%)4 (12.1)13 (10.9)0.765Anemia, n (%)32 (97)103 (86.6)0.122Leukocytopenia, n (%)11 (33.3)70 (58.8)0.011Thrombocytopenia, n (%)11 (33.3)46 (38.7)0.686Hematuria, n (%)33 (100)111 (93.3)0.202Leukocyturia (non-infection), n (%)28 (84.8)76 (63.9)0.021Nephrotic syndrome, n (%)25(75.8)84 (70.6)0.665Acute kidney injury, n (%)33 (100)32 (26.9)<0.001SLEDAI (mean±s.d.)19.85±5.3519.35±5.310.637Laboratory dataNumber of patients33119Hemoglobin (mean±s.d.) (g/l)84.61±19.4695.03±21.970.015Urine protein (mean±s.d.) (g/24 h)6.25±3.545.85±3.890.598Serum creatinine (mean±s.d.) (mg/dl)3.74±2.681.61±1.57<0.001Creatinine clearance rate (mean±s.d.) (ml/min)25.83±26.2864.52±31.26<0.001Serum albumin (mean±s.d.) (g/dl)2.54±0.672.71±0.420.563Positive ANA, n (%)33 (100)119 (100)1.0Positive anti-ds-DNA, n (%)25 (75.8)91 (76.5)1.0Positive anti-Sm, n (%)5 (15.2)27 (22.7)0.471Positive anti-SSA, n (%)14 (42.4)53 (44.5)0.846Positive anti-SSB, n (%)6 (18.2)15 (12.6)0.403Positive anti-RNP, n (%)4 (12.1)36 (30.3)0.044Anti-cardiolipin antibody, n (%)1/20 (5)7/78 (9)1.0C3 (mean±s.d.) (g/l)0.38±0.160.38±0.161.0ANA, anti-nuclear antibodies; RNP, ribonucleoprotein; SLEDAI, systemic lupus erythematosus disease activity index; SSA, Sjögren's syndrome A antigen; SSB, Sjögren's syndrome B antigen. Open table in a new tab ANA, anti-nuclear antibodies; RNP, ribonucleoprotein; SLEDAI, systemic lupus erythematosus disease activity index; SSA, Sjögren's syndrome A antigen; SSB, Sjögren's syndrome B antigen. There was no significant difference in demographic features (gender and age) and the time between presentation/diagnosis of lupus nephritis and biopsy/treatment between the two groups. In clinical data, there was a significantly lower proportion of leukocytopenia (P=0.011), a higher proportion of leukocyturia (non-infectious) (P=0.021), and a higher proportion of acute kidney injury (P<0.001) in the crescentic group. In laboratory findings, there was a significantly lower hemoglobin level (P=0.015), higher value of serum creatinine (P<0.001), lower level of creatinine clearance rate (P<0.001), and lower ratio of positive serum anti-ribonucleoprotein antibody (P=0.044) in patients with crescentic glomerulonephritis. In total, 10 of the 33 patients (30.3%) in the crescentic glomerulonephritis group were anti-neutrophil cytoplasmic antibodies (ANCA)-positive, including nine with p-ANCA and one with c-ANCA by indirect immunofluorescence and seven with anti-myeloperoxidase (MPO) antibodies by enzyme-linked immunosorbent assay (ELISA). Only three of the 119 patients (2.5%) without crescentic glomerulonephritis were ANCA-positive including two with p-ANCA and one with c-ANCA by indirect immunofluorescence and one with anti-MPO antibodies by ELISA. The difference in ratio of positive serum ANCA was significant (P<0.001). One of the 33 patients in the crescentic glomerulonephritis group was anti-GBM antibody positive and none of the 119 patients without crescentic glomerulonephritis was anti-GBM antibody positive. However, the difference in ratio of positive serum anti-GBM antibody was not significant (P=0.217). The renal histopathological features of patients with and without crescentic glomerulonephritis were listed in Table 2. In light microscopic examination, the average number of biopsied glomeruli between patients with and without crescentic glomerulonephritis was comparable. According to the 2003 revised ISN/RPS classification, the lupus nephritis class IV-G was further divided into three subgroups according to activity and chronicity of the renal lesions, that is, active lesion only (IV-G(A)), both active and chronic lesions (IV-G(A/C)), and chronic lesions only (IV-G(C)). There was a lower ratio of subgroup IV-G(A) and a higher ratio of subgroup IV-G(A/C) in patients with crescentic glomerulonephritis than in those without crescentic glomerulonephritis (P<0.001, P<0.001, respectively). In patients with crescentic glomerulonephritis, there was a significantly higher score of interstitial inflammation, activity indexes, tubular atrophy, interstitial fibrosis, and chronicity indexes (CI) (P=0.002, P<0.001, P<0.001, P<0.001, P<0.001, respectively) compared with those without crescentic glomerulonephritis besides a significant difference in cellular crescentic score. There was no significant difference in other indexes between the two groups.Table 2Comparison of renal pathological data between patients with lupus nephritis class IV-G with and without crescentic glomerulonephritis: light microscopy; direct immunofluorescence and electron microscopyCrescentic glomerulonephritisNon-crescentic glomerulonephritisP-valueLight microscopyNumber of biopsies33119Number of glomeruli (mean±s.d.)24.1±8.525.4±11.90.167% of cases with sub-class A42.479<0.001% of cases with sub-class A/C54.521<0.001% of cases with sub-class C3.100.217% of cases with fibrinoid necrosis3 (1/33)5 (6/119)1.0% sclerotic glomeruli (median and range)0, 0–270, 0–600.174AI score (mean±s.d.)13.48±3.0910.22±3.09<0.001Endocapillary hypercellualrity (mean±s.d.)2.88±0.492.93±0.280.416Cellular crescents (mean±s.d.)4.82±1.791.55±1.46<0.001Karyorrhexis/fibrinoid necrosis (mean±s.d.)1.27±1.101.45±1.180.431Subendothelial hyaline deposits (mean±s.d.)1.27±0.911.66±1.040.052Interstitial inflammation (mean±s.d.)1.88±0.931.40±0.710.002Glomerular leukocyte infiltration (mean±s.d.)1.12±0.821.41±1.470.278CI score (mean±s.d.)4.85±2.172.86±1.77<0.001Glomerular sclerosis (mean±s.d.)0.61±0.700.39±0.600.087Fibrous crescents (mean±s.d.)0.67±0.920.12±0.35<0.001Tubular atrophy (mean±s.d.)1.82±0.851.24±0.66<0.001Interstitial fibrosis (mean±s.d.)1.76±0.831.12±0.69<0.001Direct immunofluorescence and electron microscopyNumber of biopsies33119Immunofluorescence parameters% of cases with capillary loop deposition971000.217% of cases with mesangial deposition69.781.50.153IgG (mean±s.d.)1.82±0.982.14±0.910.075IgA (mean±s.d.)1.52±1.061.95±0.870.017IgM (mean±s.d.)1.27±0.981.66±0.950.039C3 (mean±s.d.)2.42±1.002.67±0.710.11C1q (mean±s.d.)1.76±0.972.2±0.830.01Fibrin (mean±s.d.)0.55±0.750.73±0.820.244Electron microscopy parameters% of cases with subendothelial deposition63.670.60.524% of cases with subepithelial deposition69.771.40.831% of cases with mesangial deposition78.876.51.0% of cases with intramembranous deposition33.325.20.379A, active lesions; AI, activity indexes; C, chronic lesions; CI, chronicity indexes. Open table in a new tab A, active lesions; AI, activity indexes; C, chronic lesions; CI, chronicity indexes. On evaluation of immunofluorescence parameters, there was no significant difference in the locations of immunoglobulin deposition between the two groups. However, the average intensity of IgG, IgA, IgM, C3, C1q, and fibrin was lower in patients with crescentic glomerulonephritis and the difference in IgA, IgM and C1q reached statistical significance. In electron microscopic parameters, there was no significant difference in the locations of electron dense deposits between the two groups. The data of treatment and outcome of patients with and without crescentic glomerulonephritis were shown in Table 3.Table 3Comparison of treatments between patients with lupus nephritis class IV-G with and without crescentic glomerulonephritis; comparison of renal outcomes between patients with lupus nephritis class IV-G with and without crescentic glomerulonephritisCrescentic glomerulonephritis n=33 Number of patients (%)Non-crescentic glomerulonephritis n=119 Number of patients (%)P-valueComparison of treatmentsInduction treatment Methylprednisone impulse33 (100)56 (47.1)<0.001 Oral prednisone33 (100)119 (100)1 Cyclophosphamide i.v.30 (90.9)111 (93.3)0.705 Mycophenolate mofetil3 (9.1)8 (6.7)0.705Maintenance treatment Oral prednisone33 (100)119 (100)1.0 Cyclophosphamide i.v.31 (94)94 (79)0.069 Azathioprine0 (0)3 (2.5)1.0 Mycophenolate mofetil1 (3)7 (5.9)1.0 Leflunomide1 (3)15 (12.6)0.196Treatment response Complete remission8 (24.2)82 (68.9)<0.001 Partial remission16 (48.5)30 (25.2)0.01 Treatment failure9 (27.3)7 (5.9)<0.001The time between initiation of treatment and final remission (mean±s.d.) (months)7.17±1.866.62±1.780.174Urine protein at remission (mean±s.d.) (g/24 h)1.21±0.360.52±0.18<0.001Serum creatinine at remission (mean±s.d.) (mg/dl)1.65±0.860.82±0.35<0.001Creatinine clearance rate at remission (mean±s.d.) (ml/min)45.77±12.3473.25±11.64<0.001Comparison of renal outcomesDuration of follow-up (mean±s.d.) (months)47.4±48.354.3±54.80.523Cumulative dose of cyclophosphamide (mean±s.d.) (g)7.35±2.647.96±3.790.453Use of ACE inhibitor18 (55)75 (63)0.376 Use of benazepril13 (13/18,72.2%)62 (62/75,82.7%)0.314 Use of fosinopril5 (5/18,27.8%)13 (13/75,17.3%)0.314Control of blood pressure≤130/80 mm Hg26 (79)99 (83)0.558Relapse rate9 (9/24,37.5%)17 (17/112, 15.2%)0.012The time between remission to relapse (mean and range) (months)17 (2–141)51 (14–152)0.112ACE, angiotensin-converting enzyme; i.v., intravenous. Open table in a new tab ACE, angiotensin-converting enzyme; i.v., intravenous. There was no significant difference in treatment algorithm between the two groups, except that there was a significantly higher number of patients using intravenous methylprednisone pulse in the crescentic glomerulonephritis group (P<0.001). The proportion of complete remission was significantly lower and the proportions of partial remission and treatment failure were significantly higher in the crescentic group (P<0.001, P=0.01, P<0.001, respectively). The cumulative incidences of the first complete remission, partial remission, and total remission were shown in Figure 2. At the time of remission, there was a significantly higher level of proteinuria (P<0.001), a higher value of serum creatinine (P<0.001), and a lower level of creatinine clearance rate (P<0.001) in the crescentic group. During maintenance therapy, nine patients in the crescentic group (eight with nephritic relapse and one with proteinuric relapse) and 17 patients in the non-crescentic group (nine with nephritic relapse and eight with proteinuric relapse) had a relapse. The relapse rate was significantly higher in the crescentic group (37.5% vs 15.2%, P=0.012, Figure 3). All of the relapses occurred on maintenance therapy. All the nine patients in the crescentic group were with cyclophosphamide compared with 11 patients with cyclophosphamide, one with azathioprine, and five with leflunomide in non-crescentic group. After relapses, all of the patients increased the dosage of corticosteroids, 12 patients reused monthly pulses of cyclophosphamide, and three patients changed to mycophenolate mofetil. Five and 12 patients achieved remission again, four and five patients presented with treatment failure in crescentic and non-crescentic groups, respectively. Most patients in the two groups used ACE (angiotensin-converting enzyme) inhibitor and blood pressure was controlled well, and there was no significant difference between the two groups. Regarding ANCA, seven (both p-ANCA positive) of the 10 patients with positive ANCA by indirect immunofluorescence in the crescentic group remained positive and the other three patients turned negative. The titers of anti-MPO antibodies of the seven positive patients in the crescentic group decreased during follow-up, but only two patients turned negative. In the non-crescentic group, p-ANCA in one of the three patients turned negative and the other two patients remained positive by indirect immunofluorescence. The titer of anti-MPO antibodies of one patient decreased and remained positive in the non-crescentic group. Repeated kidney biopsy was carried out in 10 patients, including one in the crescentic group and the other nine in the non-crescentic group (details in Table 4).Table 4Profile of patients with lupus nephritis class IV-G who underwent repeat biopsiesPatientInterval betweenRenal pathology typeActivity indexesChronicity indexesNo.Reason for rebiopsybiopsies (months)PrePostPrePostPrePost1Relapse7IVIV1115332Relapse15IVIV117993Relapse38IVV+III56014Relapse35IVIV+V79225Relapse10IVIV1214336Re-evaluation12IVIII+V133117Re-evaluation9IVIII+V122218Relapse39IVIV89029Relapse11IVIV+V782210Relapse18IVIV6923Patient 1 was in crescentic group and patients 2∼10 were in non-crescentic group. Open table in a new tab Patient 1 was in crescentic group and patients 2∼10 were in non-crescentic group. During follow-up (average for nearly 4 years), there was no significant difference between the two groups regarding long-term survival. In the crescentic group, all the patients survived and only one patient died of severe infection in those without crescentic glomerulonephritis. Regarding long-term renal outcome, patients with crescentic glomerulonephritis had a significantly poorer renal outcome (P<0.001, Figure 4). Nine (27.3%) patients had a doubling of serum creatinine and seven (21.2%) reached end stage renal disease in crescentic group. In patients without crescentic glomerulonephritis, nine patients (7.6%) had a doubling of serum creatinine and four (3.4%) reached end-stage renal disease. As patients in the crescentic group clearly had a worse serum creatinine and creatinine clearance at baseline with higher CI in the biopsy, we further carried out an adjusted proportional hazard analysis accounting for those baseline differences and confirmed that patients with crescentic glomerulonephritis had a significantly poorer renal outcome (Table 5). We further compared the value of remission and relapse (Figure 4) determining the overall renal survival between the crescentic and non-crescentic groups. We found that both remission and relapse could affect renal survival besides the pathological difference. As there was no secondary end point occurring in patients without relapse in both two groups, the similar analysis was not done.Table 5Comparison of renal outcomes between patients with lupus nephritis class IV-G with and without crescentic glomerulonephritis (unadjusted and adjusted baseline indexes)Unadjusted Cox hazard analysisP-valueHR (95%CI)Group 1.5 mg/dl8.678(1.867–30.327)Adjusted chronicity indexes and creatinine clearance rate Cox hazard analysisGroup0.004 Crescentic1 Non-crescentic0.301 (0.134–0.677)Chronicity indexes1.364 (1.134–1.641)0.001Creatinine clearance rate0.697 ≥80 ml/min1 <80 ml/min1.369 (0.281–6.674) Open table in a new tab Renal involvement is common in SLE and there are a number of different pathological phenotypes of lupus nephritis. Although crescent formation is common in lupus nephritis especially in the background of proliferative glomerular lesions, the ratio of ‘true’ crescentic glomerulonephritis in lupus nephritis is not clear and the pathogenesis remains inconclusive. In our study, we found that crescentic glomerulonephritis was not rare in lupus nephritis, which accounted for 10.1% of total biopsy-proven lupus nephritis and 21.7% of lupus nephritis class IV-G. In the literature, Sumethkul V et al.3.Sumethkul V. Chalermsanyakorn P. Changsirikulchai S. et al.Lupus nephritis: a challenging cause of rapidly progressive crescentic glomerulonephritis.Lupus. 2000; 9: 424-428Crossref PubMed Scopus (12) Google Scholar showed that lupus nephritis with crescentic glomerulonephritis accounted for 51.6% (32/62) of all patients with biopsy-proven various crescentic glomerulonephritis. Therefore, lupus nephritis with crescentic glomerulonephritis should be paid more attention. In comparison with lupus nephritis class IV-G without crescentic glomerulonephritis, it was not surprising that patients with crescentic glomerulonephritis had a significantly higher proportion of acute kidney injury and a lower hemoglobin level because of the presence of rapidly progressive glomerulonephritis clinically. We also found a lower proportion of positive serum anti-ribonucleoprotein antibodies in the crescentic group. Although the difference is marginal, a further study with a larger sample size is needed to clarify the clinical significance. On histopathological evaluation, besides the difference in ratio of crescent formation, we found that there were significantly higher scores of interstitial inflammation, activity indexes, tubular atrophy, interstitial fibrosis, and CI in patients with crescentic glomerulonephritis than in those without crescentic glomerulonephritis. These results suggested that both acute and chronic lesions were prominent in patients with crescentic glomerulonephritis. More interestingly, we found that the average intensity of IgG, IgA, IgM, C3, C1q, and fibrin were lower in the crescentic group by immunofluorescence and the average intensity of IgG, IgA, and IgM was <2+. Combined with the immunofluorescence presentation and a higher ratio of the subclass IV-G(A/C) by light microscopy, crescentic glomerulonephritis in lupus nephritis seemed more like ‘pauci-immune’ crescentic glomerulonephritis because of ANCA-associated vasculitis. Therefore, we further detected serum ANCA and anti-GBM antibodies in all of the patients with lupus nephritis class IV-G. The result was so striking that nearly 1/3 of patients with crescentic glomerulonephritis were ANCA-positive, compared with only 2.5% of patients without crescentic glomerulonephritis. We used Hep-2 cell and paraformaldehyde-fixed neutrophils to allow for the distinction between anti-nuclear antibodies and p-ANCA. For further detailed analysis of the 10 patients with positive ANCA in patients with crescentic glomerulonephritis, we could not find a history of taking special medications such as hydralazine, propylthiouracil, etc. Therefore, drug-induced SLE with multi-autoantibodies including ANCA was unlikely. More importantly, we found that 8 out of the 10 patients presented with ‘pauci immune’ by immunofluorescence and no significant subendothelial deposits by electron microscopy. Therefore, they might be diagnosed simultaneously as renal involvement of ANCA-associated vasculitis. However, in this study, only a few patients had necrosis and there was no difference between the two groups, which is less consistent with vasculitis. The explanation was the following: first, the fibrinoid necrotic lesions could be found in both lupus nephritis and ANCA-associated vasculitis, but were not diffuse; second, owing to the focal and segmental nature, the percentage of glomeruli with fibrinoid necrosis was dependent on the number of sections of renal tissue: the higher the number of sections, the higher the percentage of necrosis; third, the sample size, especially of patients with crescents was small. These might be attributed to the comparable percentage of fibrinoid necrosis between the two groups in our study. We also noticed a lower proportion of leukocytopenia in the crescentic group, which might be explained by the existence of leukocytosis in patients with ANCA-associated vasculitis, but this needs further study. It is known that 15–20% of patients with SLE have detectable serum ANCA.4.Sen D. Isenberg D.A. Antineutrophil cytoplasmic antibodies in systemic lupus erythematosus.Lupus. 2003; 12: 651-658Crossref PubMed Scopus (68) Google Scholar Indeed, isolated cases have been reported in which ANCA seems clearly to be implicated as a major pathogenetic factor and might play a role in the development of necrotizing vasculitis and crescent formation in lupus nephritis.5.Arahata H. Migata K. Izumoto H. et al.Successful treatment of rapidly progressive lupus nephritis associated with anti-MPO antibodies by intravenous immunoglobulins.Clin Rheumatol. 1999; 18: 77-81Crossref PubMed Scopus (24) Google Scholar, 6.Marshall S. Dressler R. D'Agati V. Membranous lupus nephritis with antineutrophil cytoplasmic antibody-associated segmental necrotizing and crescentic glomerulonephritis.Am J Kidney Dis. 1997; 29: 119-124Abstract Full Text PDF PubMed Scopus (40) Google Scholar, 7.Masani N.N. Imbriano L.J. D'Agati V.D. et al.SLE and rapidly progressive glomerulonephritis.Am J Kidney Dis. 2005; 45: 950-955Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar Some authors had found a correlation between the presence of serum ANCA and the formation of crescents in patients with lupus nephritis,8.Lee S.S. Lawton J.W. Chan C.E. et al.Antilactoferrin antibody in systemic lupus erythematosus.Br J Rheumatol. 1992; 31: 669-673Crossref PubMed Scopus (59) Google Scholar an association between p-ANCA, such as anti-cathepsin G antibodies, and the development of lupus nephritis.9.Chin H.J. Ahn C. Lim C.S. et al.Clinical implications of antineutrophil cytoplasmic antibody test in lupus nephritis.AmJ Nephrol. 2000; 20: 57-63Crossref PubMed Scopus (45) Google Scholar,10.Zhao M.H. Liu N. Zhang Y.K. et al.Antineutrophil cytoplasmic autoantibodies (ANCA) and their target antigens in Chinese patients with lupus nephritis.Nephrol Dial Transplant. 1998; 13: 2821-2824Crossref PubMed Scopus (45) Google Scholar Although several other studies have failed to find a correlation between ANCA with lupus nephritis,11.Cambridge G. Wallace H. Bernstein R.M. et al.Autoantibodies to myeloperoxidase in idiopathic and drug-induced systemic lupus erythematosus and vasculitis.Br J Rheumatol. 1994; 33: 109-114Crossref PubMed Scopus (97) Google Scholar, 12.Nishiya K. Chikazawa H. Nishimura S. et al.Anti-neutrophil cytoplasmic antibody in systemic lupus erythematosus is unrelated to clinical features.Clin Rheumatol. 1997; 16: 70-75Crossref PubMed Scopus (45) Google Scholar, 13.Kuster S. Apenberg S. Andrassy K. et al.Antineutrophil cytoplasmic antibodies in systemic lupus erythematosus.Contrib Nephrol. 1992; 99: 94-98Crossref PubMed Google Scholar, 14.Pauzner R. Urowitz M. Gladman D. et al.Antineutrophil cytoplasmic antibodies in systemic lupus erythematosus.J Rheumatol. 1994; 21: 1670-1673PubMed Google Scholar most authors have indicated that ANCA seropositivity was commonly encountered in patients with SLE, especially those with the di
Referência(s)