- Email: [email protected]
Renal biopsy findings predicting outcome in scleroderma renal crisis
Human Pathology (2009) 40, 332–340
www.elsevier.com/locate/humpath
0riginal contribution
Renal biopsy findings predicting outcome in scleroderma renal crisis☆ Ibrahim Batal MD a , Robyn T. Domsic MD b , Aaron Shafer MD c , Thomas A. Medsger Jr. MD b , Lawrence P. Kiss MD a , Parmjeet Randhawa MD a , Sheldon Bastacky MD a,⁎ a
Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Division of Rheumatology and Clinical Immunology, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA c Division of Gynecologic Oncology, University of North Carolina, Chapel Hill, NC 27522, USA b
Received 29 May 2008; revised 1 August 2008; accepted 1 August 2008
Keywords: Scleroderma renal crisis; Vascular thrombosis; Glomerular ischemic collapse; C4d deposits
Summary Scleroderma renal crisis is irreversible in some patients despite aggressive treatment. This study was designed to identify pathologic prognostic features in scleroderma renal crisis. We retrospectively reviewed the pathology and the clinical records of 17 patients who underwent kidney biopsies during scleroderma renal crisis (group A, recovered renal function [n = 7]; group B, remained in renal failure or died [n = 10]). Multiple histologic features were assessed semiquantitatively (0-3) or as percentages. C4d staining of peritubular capillaries and small vessels was assessed semiquantitatively (0-3) in patients with scleroderma (n = 11), normotensive (n = 10), and hypertensive (n = 12) nonscleroderma native kidney controls. The percentage of thrombosed vessels (25.1 ± 21.0 versus 5.6 ± 12.3, P = .045) and the severity of glomerular ischemic collapse (2.9 ± 0.3 versus 1.4 ± 0.8, P = .001) were significantly higher in group B than in group A. Also, group B patients tended to have more severe acute tubular injury and vascular fibrinoid changes. The peritubular capillary C4d score in patients with scleroderma, normotensive controls, and hypertensive controls were 1.1 ± 0.9, 0.3 ± 0.7, and 0.3 ± 0.5, respectively (P = .018, scleroderma versus other controls). Small vessel C4d score was higher in scleroderma compared to normotensive but not hypertensive controls. Within scleroderma samples, a significantly higher peritubular capillary C4d score (1.6 ± 0.7 versus 0.3 ± 0.5, P = .024) but not small vessel score was found in group B compared to group A. This tended to be associated with peritubular capillary leukocyte margination. Vascular thrombosis, severe glomerular ischemic collapse, and peritubular capillary C4d deposits in scleroderma renal crisis kidney biopsies correlated with increased risk of failure to recover renal function. © 2009 Elsevier Inc. All rights reserved.
1. Introduction ☆
This work was presented at the 2008 annual United States and Canadian Academy of Pathology (USCAP) meeting (Denver, CO). ⁎ Corresponding author. E-mail address: [email protected] (S. Bastacky). 0046-8177/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2008.08.001
Systemic sclerosis is a multisystem disorder affecting small blood vessels and connective tissues. Its standard classification is as diffuse cutaneous or limited cutaneous variants [1]. Scleroderma renal crisis (SRC) is a
Renal biopsy findings predicting outcome in scleroderma renal crisis complication of systemic sclerosis that occurs in up to 10% of patients. SRC is typically characterized by the new onset of hypertension (although few patients remain normotensive) and acute renal failure, with or without evidence of microangiopathic hemolytic anemia or thrombocytopenia [2,3]. The mortality associated with SRC has diminished significantly because of early diagnosis and angiotensin-converting enzyme inhibitor therapy [4,5]. Nevertheless, some patients are refractory to angiotensinconverting enzyme inhibitor therapy and remain on dialysis or die [2,6,7]. Although many studies have been conducted to investigate its pathogenesis, the etiology of systemic sclerosis remains incompletely understood [8]. Some investigators believe that systemic sclerosis is a disorder of cell-mediated immunity [9,10]; some contend that antibody-mediated injury plays an important role [11], whereas others stress the importance of T cell–B cell interaction [1]. Patients with systemic sclerosis have expansion in naive B cells, decrease in memory B cells, and increase in CD19 expression. This may result in antibody production through inhibition of B-cell peripheral tolerance [12]. C4d is a classic pathway complement degradation product. Its presence is associated with antibody-mediated injury in renal allografts, and it has been shown to independently predict unfavorable renal allograft outcome [13-15]. Activated C4 is cleaved to C4a and C4b. C4b binds to amino or hydroxyl groups and is then converted to C4d, which can covalently bind to the peritubular capillary (PTC) basement membrane [16]. C4d staining pattern in SRC has not been investigated yet. This study was designed to identify potential clinical and pathologic prognostic factors and to investigate C4d deposition in SRC.
2. Materials and methods 2.1. Study design We retrospectively reviewed clinical and surgical pathology records of 17 patients with scleroderma who underwent native renal biopsies during SRC between 1990 and 2007, with the last clinical follow-up in October 2007. Fifteen patients with SRC were seen at the University of Pittsburgh Medical Center and met the original classification criteria proposed by the American College of Rheumatology [17] for systemic sclerosis and the recent revisions [18]. The remaining 2 patients had their biopsy specimens referred from an outside institution with the diagnosis of systemic sclerosis and worsening renal function. Ten specimens derived from nonneoplastic areas of kidneys resected for renal neoplasms and 12 from non-SRC patients with hypertension (5 idiopathic malignant hypertension and 7 chronic benign hypertension) were used as controls for C4d staining in native kidneys.
333
2.2. Histologic evaluation Standard hematoxylin and eosin, periodic acid–Schiff, silver, and trichrome sections were evaluated. Histologic changes in all 4 renal anatomic compartments (glomeruli, tubules, interstitium, blood vessels) were assessed and scored semiquantitatively (0-3) or recorded as percentages. The glomerular alterations assessed included glomerular ischemic collapse, glomerular basement membrane (GBM) double contours (tram tracking), and juxtaglomerular apparatus (JGA) prominence. These changes were graded on a semiquantitative scale (0-3) according to the severity of the most affected nonsclerotic glomerulus. Glomerular ischemic collapse was graded as no (0), mild (1), moderate (2), or severe (3) glomerular collapse based on the most severely affected glomeruli. Glomerular double contour was graded from 0 to 3 when double contour lesion was seen in less than 10%, 11% to 25%, 26% to 50%, or more than 50% of capillary loops, respectively. JGA status was graded on a scale (0-3) corresponding to inconspicuous JGA, mild hyperplasia, moderate hyperplasia, and severe hyperplasia, respectively. Global glomerulosclerosis and glomerular thrombosis were recorded as a percentage of the number of affected glomeruli. Tubular pathologic alterations were assessed. Acute tubular injury was graded on a semiquantitative scale based on the most affected area, corresponding to no injury (0), wrinkling of tubular basement membrane or tubular epithelial vacuolization (1), attenuation of tubular epithelium (2), and necrosis (3), respectively. Tubular atrophy was graded on a semiquantitative scale according to the cortical surface area involved by tubular atrophy: no atrophy (0), and atrophy involving up to 25% (1), 26% to 50% (2), and more than 50% (3) of cortical tubular profiles, respectively. Interstitial fibrosis was graded on a semiquantitative scale based on the cortical surface area involved, corresponding to less than 5% (0), 6% to 25% (1), 26% to 50% (2), and more than 50% (3), respectively. Vascular abnormalities were separated into acute (vascular myxoid and fibrinoid changes), subacute (onion skin lesions), and chronic (fibroelastic intimal thickening, muscular thickening, and hyalinosis). Each abnormality was graded on a semiquantitative severity scale (0-3) based on the most severely affected vessel. Fibrinoid change was evaluated according to extension of this lesion within vascular wall as no fibrinoid alteration (0), fibrinoid change involving the intima up to the muscularis (1), through the muscularis (2), and the entire wall thickness (3). Onion skin lesion was evaluated according to the severity of proliferative endarteropathy and concentric fibrosis within the involved arteriole/ interlobular artery as no (0), early (1), well established (2), or advanced (3) onion skinning lesion. Fibroelastic intimal thickening and myxoid changes were graded on a scale (0-3) corresponding to no significant narrowing, narrowing of up to 25%, 26% to 50%, and more than 50% of luminal cross-sectional surface
334
I. Batal et al.
caused by fibrosis or accumulation of myxoid material, respectively. Muscular thickening was graded as no (0), mild (1), moderate (2), or severe (3). Arteriolar hyalinosis was graded on a scale (0-3) corresponding to no, mild, moderate, or severe periodic acid-Schiff–positive hyaline thickening, respectively. Vascular thrombosis, considered to be an acute pathologic injury, was recorded as percentage of thrombosed arterioles/arteries. Finally, leukocyte margination within peritubular capillaries was semiquantitavely graded. The lesion was considered to be absent when less than 10% of peritubular capillaries showed leukocyte margination. It was graded as 1, 2, or 3 when more than 10% of peritubular capillaries showed leukocyte margination and the most severely affected cross-sectioned capillary contained up to 4, 5 to 10, or more than 10 white blood cells (neutrophilic and/or mononuclear), respectively. Each case was histologically evaluated by the first author (I. B.) and the senior author (S. B.). All cases were also reviewed on a multiheaded microscope to eliminate any possible discrepancy.
fields. Staining was referred to as focal when present in 10% to 50% of high-power microscopic fields, and at least one cluster of positive peritubular capillaries could be demonstrated. Lesser degree of staining was considered minimal, and absent staining was considered negative. Areas of atrophy were excluded from evaluation. C4d staining in small vessels, namely, arterioles and small arteries, was classified as diffuse, focal, minimal, or negative when more than 50%, 26% to 50%, 5% to 25%, or less than 5% of small vessels showed C4d deposits, respectively. To facilitate comparison, C4d staining was graded using a semiquantitative score for both PTC and arteries (0-3) corresponding to negative (0), minimal (1), focal (2), and diffuse (3), respectively. Glomerular capillary staining was recorded as present or absent. When present, the stain was further characterized as minimal, segmental, or global. Evaluation of C4d staining was performed by the first author (I. B.). The results were confirmed by the sixth author (P. R.) using a multiheaded microscope.
2.4. Statistics 2.3. C4d staining C4d immunostaining was performed using rabbit antihuman C4d polyclonal antibody (ALPCO Diagnostics, Windham, NH). Four-micrometer formalin-fixed, paraffinembedded sections of native kidneys were deparaffinized, placed in a pressure cooker for 20 minutes, and sequentially incubated at 37°C with TRIS-EDTA (pH 8.5), 1:50 dilution of primary antibody for 44 minutes, 1:100 dilution of a biotinylated antirabbit secondary antibody for 8 minutes, streptavidin-alkaline phosphate conjugate for 8 minutes, fast red or brown A-naphthol for 8 minutes, and fast red or brown B-naphthol for 8 minutes. Sections were then counterstained with hematoxylin. All reagents were components of the Ventana Enhanced Alkaline Phosphatase Red or Brown Detection Kit (cat no. 760-031, Ventana Medical Systems, Tucson, AZ). PTC staining was classified as diffuse when staining was present in more than half of the high-power microscopic Table 1
Statistical comparison between different study groups was performed using the Mann-Whitney rank sum test or χ2 test (P values expressed as Fisher exact test) using Sigma Stat 2.0.3 software (SPSS Inc, Chicago, IL). P values less than .05 were considered statistically significant. The study was conducted in accordance with procedures established by the University of Pittsburgh Institutional Review Board (protocol no. 0701150).
3. Results Of the 17 patients, 13 were white and 4 were African American. Four patients were male and thirteen were female. The age range was 23 to 73 years with a mean ± SD of 51.7 ± 16.4 years. Mean ± SD diastolic blood pressure and serum creatinine (Cr) available at the time of biopsy were 102 ± 24
Demographic characteristics
Age (y) a Sex (male/female) White/African American Cr at time of biopsy (mg/dL) a Diastolic blood pressure (mm Hg) a Disease duration (years from clinical diagnosis to biopsy) a
Group A (n = 7) recovered
Group B (n = 10) HD/transplant/ death on dialysis
P
43.7 ± 16.0 1:6 4:3 4.2 ± 3.4 116 ± 13 1.4 ± 1.9
57.3 ± 15.0 3:7 9:1 5.1 ± 2.3 91 ± 25 3.1 ± 3.6
.18 .6 .25 .29 .023 .78
NOTE. Serum Cr and diastolic blood pressure at the time of biopsy were not available for 1 group B patient. Disease duration information was available for 7 group A patients and 4 group B patients. Abbreviation: HD, hemodialysis. a Data presented as mean ± SD.
Renal biopsy findings predicting outcome in scleroderma renal crisis mm Hg and 4.7 ± 2.8 mg/dL, respectively (n = 16). The follow-up period ranged between 51 days and 13.1 years with a mean ± SD of 3.0 ± 3.9 years. All the patients with a follow-up period less than 6 months were the patients who died with renal failure due to severe disease except one patient who was on dialysis for 3 months at the end of follow-up period. Mean ± SD of disease duration from first diagnosis of scleroderma to biopsy, when information was available (n = 11), was 2.0 ± 2.6 years. Patients were divided into 2 groups based on renal function status: group A (recovered renal function) and group B (persistent renal failure requiring dialysis or transplant, or death with renal failure). There were no significant differences in age, sex, race, presenting serum Cr, and disease duration between groups A and B patients. Diastolic blood pressure was significantly higher in group A (116 ± 13 mm Hg) compared to group B (91 ± 25 mm Hg) (P = .023) (Table 1). Serum autoantibody results were available in 16 patients. Of these, scleroderma-associated antibodies were identified in 6 patients in group A including RNA polymerase III (n = 2), U1 RNP (n = 1), U3 RNP (n = 1), Th/To (n = 1), and PL7 (n = 1). In 6 patients in group B, sclerodermaassociated autoantibodies were detected including RNA polymerase III (n = 3), Scl70 (n = 1), p-ANCA (n = 1), and smooth muscle (n = 1). Histologic features were compared between groups A and B (Table 2). Glomerular ischemic collapse scores (Fig. 1) were significantly higher in group B (2.9 ± 0.3) compared to group A (1.4 ± 0.8; P = .001). The other parameters assessed (global glomerulosclerosis, tram tracking, juxtaglomerular apparatus hyperplasia) did not differ significantly between the groups. Although it did not reach statistical significance, none of the patients with glomerular capillary thrombosis (n = 3, all in group B) (Fig. 2) recovered renal function. The acute tubular injury scores tended to be higher for group B Table 2
Fig. 1 Severe glomerular ischemic collapse in a SRC patient who failed to recover renal function (methenamine silver, original magnification ×400).
(1.8 ± 0.6) compared to group A (1.1 ± 0.7; P = .11), whereas chronic tubular atrophy scores were similar. Also, the interstitial fibrosis score was similar in group A versus group B. The proportion of thrombosed vessels (Fig. 3) was significantly higher in patients who did not recover renal function (25.1 ± 21.0 group B versus 5.6 ± 12.3 group A; P = .045). Noteworthy, an area of infarction was noted in one biopsy with no detectable thrombosis. This likely represented a manifestation of thrombosis or critical vascular narrowing in an unsampled larger artery. When the percentage of infarcted areas was considered in evaluation together with the percentage of thrombosed vessels, the difference became even greater (28.5 ± 19.2 group B versus 5.6 ± 12.3 group A, P = .017). The fibrinoid change score tended to be higher in group B but did not achieve statistical
Histopathologic comparison between groups A and B
Glomerular
Tubular Interstitium Vascular pathology
a
335
Lesions
Group A (n = 7) a
Group B (n = 10) a
P
Global sclerosis (%) Thrombosis (%) Ischemic collapse (0-3) Tram tracking (GBM double contour) (0-3) JGA hyperplasia (0-3) Acute injury (0-3) Atrophy (0-3) Fibrosis (0-3) Thrombosed vessels (%) Fibrinoid changes (0-3) Myxoid changes (0-3) Onion skinning (0-3) Fibroelastic intimal thickening (0-3) Muscular thickness (0-3) Hyalinosis (0-3) Peritubular capillaritis (0-3)
9.7 0.0 1.4 0.1 1.3 1.1 0.4 0.7 5.6 0.0 1.6 1.1 1.3 0.4 0.3 0.7
13.8 ± 13.9 1.4 ± 3.2 2.9 ± 0.3 0.3 ± 0.7 1.5 ± 0.1 1.8 ± 0.6 0.4 ± 0.7 0.8 ± 0.8 25.1 ± 21.0 0.7 ± 1.0 1.6 ± 1.3 0.7 ± 1.0 1.8 ± 0.8 1.2 ± 1.0 0.5 ± 1.1 1.1 ± 1.1
.3 .32 .001 .84 .7 .11 .8 .96 .045 .094 1.0 .52 .35 .14 .96 .6
Data are presented as mean ± SD.
± 13.0 ± 0.0 ± 0.8 ± 0.4 ± 1.1 ± 0.7 ± 0.5 ± 0.5 ± 12.3 ± 0.0 ± 1.5 ± 1.2 ± 1.1 ± 0.5 ± 0.5 ± 0.8
336
I. Batal et al.
Fig. 4 Arterial fibrinoid change observed in a SRC patient who failed to recover renal function. In addition, ischemic glomerular collapse and acute tubular injury are seen in this field (hematoxylin and eosin, original magnification ×200). Fig. 2 Glomerular capillary thrombosis in a SRC patient who failed to recover renal function (methenamine silver, original magnification ×600).
significance (0.7 ± 1.0 versus 0 ± 0; P = .094). Yet, all patients with fibrinoid change (n = 5, all in group B) (Fig. 4) developed refractory renal failure. No differences between the 2 groups were found for vascular myxoid change, onion skin lesions, fibroelastic intimal thickening, muscular hypertrophy, or arteriolar hyalinosis. The mean score of leukocyte margination in PTC was higher in group B (1.1 ± 1.1) compared to group A (0.7 ± 0.8), but this was not significantly different. To further investigate if the presence or absence of the manifestations of acute endothelial injury within biopsy can affect prognosis, we again treated vascular thrombosis/ infarction, fibrinoid changes, glomerular thrombosis, and
Fig. 3 Arterial thrombosis in a SRC patient who failed to recover renal function (methenamine silver, original magnification ×100).
vascular myxoid changes as noncontinuous variables and reevaluated using the Fisher exact test. A significant difference in the renal outcome was identified for vascular thrombosis/infarction (2/7 group A versus 9/10 group B, P = .035), and fibrinoid change (0/7 group A versus 5/10 group B, P = .044) but not glomerular thrombosis (0/7 group A versus 3/10 group B, P = .23) or myxoid change (4/7 group A versus 7/10 group B) (Fig. 5). Paraffin tissue blocks for 11 scleroderma patients (group A [n = 4], group B [n = 7]) were available for immunohistochemical evaluation. These biopsies, as well as 10 normotensive and 12 hypertensive non-SRC specimens retrieved from our files to serve as controls, were stained with C4d immunohistochemical stain. The scores were
Fig. 5 Arterial myxoid changes observed in a SRC patient who recovered renal function (methenamine silver, original magnification ×200).
Renal biopsy findings predicting outcome in scleroderma renal crisis Table 3
337
Composition of C4d staining in peritubular capillaries and arteries in SRC patients versus controls
PTC
Small vessels
Diffuse Focal Minimal Negative Score (0-3) Diffuse Focal Minimal Negative Score (0-3)
Normotensive controls (n = 10)
Hypertensive controls (n = 12)
All SRC (n = 11)
0 (0%) 1 (10%) 1 (10%) 8 (80%) 0.3 ± 0.7 0 (0%) 2 (20%) 1 (10%) 7 (70%) 0.5 ± 0.9
0 (0%) 0 (0%) 3 (25%) 9 (75%) 0.3 ± 0.5 6 (50%) 1 (8%) 2 (17%) 3 (25%) 1.8 ± 1.4
1 (9%) 2 (18%) 5 (46%) 3 (27%) 1.1 ± 0.9 ⁎ 5 (46%) 3 (27%) 2 (18%) 1 (9%) 2.1 ± 1.0 †
NOTE. Scores are presented as mean ± SD. ⁎ P = .018, PTC C4d score (Kruskal Wallis). P = .048 and P = .024, SRC patients versus normotensive controls and SRC patients versus non-SRC hypertensive controls (Mann-Whitney). † P = .008, Small vessels C4d score (Kruskal Wallis). P = .005 and P = .025 SRC patients versus normotensive controls and non-SRC hypertensive controls versus normotensive controls (Mann-Whitney).
semiquantitated for both PTC and small vessels as described earlier (Table 3). There was a significant difference comparing PTC C4d score between SRC patients (Fig. 6) (1.1 ± 0.9) and both normotensive controls (0.3 ± 0.7, P = .048) and hypertensive non-SRC controls (0.3 ± 0.5, P = .024) (Table 3). Diffuse staining was only observed in SRC cases (n = 1), whereas focal staining was observed additionally in one normotensive control. C4d deposition in small arteries and arterioles in SRC patients was significantly greater compared to normotensive controls (2.1 ± 1.0 versus 0.5 ± 0.9; P = .005) but not hypertensive non-SRC controls (1.8 ± 1.4), which also showed greater staining compared to the normotensive group (P = .025). Within the hypertensive non-SRC control group, diffuse C4d deposition in small vessels was observed more frequently
in acute malignant hypertension samples (4/5 idiopathic malignant hypertension versus 2/7 benign hypertension). C4d staining patterns within PTC as well as small vessels were compared between SRC groups A and B (Table 4). Although the sample size was small, PTC C4d score was significantly higher in group B (1.6 ± 0.7) versus group A (0.3 ± 0.5) (P = .024); moreover, diffuse and focal PTC stain was exclusively seen in group B. No difference was observed when small vessel C4d staining was evaluated (2.1 ± 1.2 [group B] versus 2.0 ± 0.8 [group A]). PTC leukocyte margination was then evaluated in all specimens stained with C4d. The mean PTC leukocyte margination score was highest in SRC patients (0.94 ± 0.97) and lowest in non-SRC normotensive controls (0.1 ± 0.3, P = .028). The non-SRC hypertensive controls had a mean value of 0.6 ± 0.7 and was not significantly different when compared to the aforementioned groups.
Table 4 Composition of C4d staining in peritubular capillaries and arteries in SRC patients who recovered renal function (group A) versus who failed to recover renal function (group B) Group A (n = 4) Group B (n = 7) PTC
Fig. 6 Finely granular C4d deposits in peritubular capillaries in a biopsy from a SRC patient who failed to recover renal function. Although not all peritubular capillaries stained positively, greater than 50% of the sampled biopsy area showed capillaries with C4d deposition. Hence, the biopsy was scored as diffusely positive, as suggested by the Banff 2007 classification for renal allograft pathology [19] (polyclonal C4d immunostain performed on formalin-fixed, paraffin-embedded tissue, original magnification ×400).
Diffuse Focal Minimal Negative Score (0-3) Small vessels Diffuse Focal Minimal Negative Score (0-3)
0 (0%) 0 (0%) 1 (25%) 3 (75%) 0.3 ± 0.5 2 (25%) 2 (50%) 1 (25%) 0 (0%) 2.0 ± 0.8
1 (14%) 2 (29%) 4 (57%) 0 (0%) 1.6 ± 0.7 ⁎ 4/7 (58%) 1/7 (14%) 1/7 (14%) 1/7 (14%) 2.1 ± 1.2
NOTE. Scores are presented as mean ± SD. ⁎ P = .024, PTC C4d score group A versus group B (MannWhitney).
338 Noteworthy, for SRC patients, all biopsies with diffuse and focal PTC C4d deposits (n = 3) had high PTC leukocyte margination score (2.3 ± 0.6). This score was significantly different when compared to the other SRC biopsies with negative and minimal PTC C4d score (0.9 ± 0.8, P = .041). In SRC, the PTC mononuclear component was the predominant marginating white blood cell component except for one case with focal PTC staining with a predominant neutrophilic component. All SRC patients with high PTC C4d score (2-3) had mild mononuclear interstitial inflammation. Evaluation of glomerular C4d staining revealed that the stain was completely negative in all but 4 biopsies. Two of these 4 had segmental staining in up to 3 glomeruli, whereas minimal staining was detected in the remaining 2 specimens. All of these 4 cases were associated with detectable PTC C4d deposits and 3 of 4 were in group B patients. None of the specimens that were stained for C4d (n = 11) showed glomerular double contour. Finally, in our series of SRC patients with detectable PTC C4d staining, review of the immunofluorescence did not reveal any PTC immunoglobulin staining. IgA (+1) in arteries (n = 1), IgM (+2-4) in arterioles and arteries (n = 2), and IgG (+1) in the interstitium (n = 1) were detected in the 3 patients with focal and diffuse PTC C4d staining.
4. Discussion SRC is a serious complication of systemic sclerosis that may lead to permanent renal failure or death. Some patients recover renal function while others do not. Because SRC is a clinical diagnosis, renal biopsy is not frequently performed. We investigated whether any clinical or pathologic features would predict poor renal functional status after SRC. No significant difference was found comparing age, sex, race, serum Cr at the time of biopsy, or disease duration in this small patient population. A significantly lower diastolic blood pressure was found within the group who failed to recover renal function. This is consistent with previous observations that less hypertensive or normotensive SRC patients have reduced survival, perhaps due to a delay in diagnosis and treatment [3,20]. In our study, the histologic changes reflecting the most severe acute vascular injury such as vascular thrombosis and glomerular ischemic collapse, and to a lesser extent acute tubular injury, were associated with a worse prognosis. When treated as categorical variable, the presence of fibrinoid change was also associated with a worse prognosis. Although not significantly different between groups A and B, thrombosis in glomerular capillaries was consistently associated with a poor outcome. This small retrospective study revealed that renal biopsies may not only confirm a suspected diagnosis of SRC, but more importantly, there are several pathologic
I. Batal et al. features that may have prognostic significance. Penn et al [21] in a single-center study found that acute vascular changes, namely, myxoid and fibrinoid changes and vascular thrombosis were associated with poor prognosis in SRC. These authors combined the acute changes and treated them as categorical variables. In our study, the presence of myxoid change was not associated with an adverse renal outcome and was observed frequently in patients who recovered renal function (Fig. 5). It would be of interest to know if vascular myxoid change without fibrinoid change was independently predictive of an adverse renal outcome in the patients in Penn's study. Our data suggest that vascular myxoid change in the absence of fibrinoid change does not irreversibly result in permanent vascular occlusion or destruction of the involved vessel if the SRC is successfully treated. In addition, Penn et al [22] in a recently published abstract pooled data from multiple medical centers and confirmed their previous observation. Yet, surprisingly, they found an association between chronic pathologic changes and better prognosis. In our study, we did not observe such an association. This might be attributed to the sample size, which is larger than ours, and to different criteria to evaluate chronic injury. Penn et al used a composite index of chronic damage to evaluate chronic pathologic changes. This index is calculated by a software program that combines pathologic changes in different anatomical compartments and expresses the final results as percentages [23]. We assessed chronic injury by evaluating each renal compartment individually. Our analysis does not suggest that chronic injury is a protective factor by itself. It could be a result of a milder ongoing active injury, or alternatively, an expression of previous resolving more severe acute injury. The pathogenesis of scleroderma remains obscure. Autoantibodies may play a pathogenetic role either by inducing endothelial apoptosis (antiendothelial cell antibodies) or by activating fibroblasts (antifibrillin-1 antibodies) [11]. Alternatively, activated T cells, which are usually Th-2, may produce interleukin 4 and 13, leading to fibrosis [24]. Additional evidence of a potential role of autoantibodies in SRC is its association with anti-RNP polymerase III antibodies [25,26]. Complement activation comparing systemic sclerosis patients with controls has been reported [27,28]. We assessed endothelial C4d deposition, an early classic complement pathway marker, using a polyclonal antibody immunostain. In our study, SRC biopsies showed significantly greater PTC C4d deposits compared to both normotensive non-SRC controls and hypertensive non-SRC controls. Furthermore, PTC leukocyte margination score was significantly higher in SRC compared to normotensive controls. The mean value of this score was also higher in SRC patients compared with that of hypertensive controls, but this was not significantly different. Within SRC specimens, we found significantly greater PTC C4d staining in patients who irreversibly lost renal
Renal biopsy findings predicting outcome in scleroderma renal crisis function or died on dialysis compared to patients who recovered renal function. This was accompanied by PTC leukocyte margination and mild interstitial inflammation. Li et al [29] showed that diffuse C4d deposition in PTC was detected in 7% of patients with lupus nephritis. This is comparable to our observation of diffuse PTC C4d deposition in 9% of our SRC patients. Lupus patients with diffuse PTC C4d staining showed higher clinical activity scores and lower serum complement levels. Li et al indicated that these deposits are different from that observed in antibody-mediated rejection, in the sense of being granular and associated with immunoglobulin G and immune complex deposits. Similar to lupus deposits, PTC C4d deposits in SRC were finely granular (Fig. 6); yet, unlike lupus nephritis, staining in our SRC biopsies was not associated with immune deposits. Noteworthy, staining for C4d in our laboratory tends to be finely granular rather than uniformly linear in our renal allograft biopsies. We regard this finely granular staining as real because it correlates well with presence of donor-specific antibodies in kidney transplant patients [30]. A finely granular pattern of staining has also been noted in other laboratories [31,32]. An additional study using monoclonal C4d staining on frozen tissue is needed to further characterize these deposits; however, at this time, the presence of such deposits at least raises the possibility of antibody-mediated injury in a subset of SRC patients. Finely granular GBM C4d deposits were present in a small subset of SRC. These deposits were always found in association with PTC staining and were not attributed to GBM remodeling. This also suggests that the same mechanism responsible for C4d deposition in PTC may be responsible for GBM deposits. C4d deposition in extraglomerular small vessels is commonly detected in allograft biopsies (unpublished observation). Vascular C4d deposition may represent nonspecific complement activation after endothelial injury due to hypertension, drugs, or any type of mechanical stress. The presence of more frequent C4d staining in malignant hypertension compared with chronic benign hypertension suggests an association with acute endothelial injury. The retrospective nature and small sample size of this study limit its conclusions, although this is one of the largest pathologic single-center studies performed on SRC. Larger prospective studies are needed to confirm our preliminary results and to determine if renal biopsies can be used to stratify SRC patients according to their risk of developing therapy-refractory renal failure. This could lead to optimization of therapy in SRC, with patients separated into those in whom aggressive therapy is likely to be beneficial and those in whom early consideration of renal transplant would be most appropriate. In conclusion, diastolic blood pressure during SRC presentation was significantly lower in patients who failed to recover renal function, which may reflect delayed diagnosis and more severe ongoing renal injury. Renal
339
biopsies during SRC may have prognostic value. Finally, PTC C4d deposits may be an unfavorable prognostic feature in SRC and suggest the possibility of ongoing antibodymediated injury in a subset of SRC patients.
References [1] Sakkas. New developments in the pathogenesis of systemic sclerosis. Autoimmunity 2005;38:113-6. [2] Lewandowski B, Domysławska I, Klimiuk PA, et al. Kidney crisis in systemic sclerosis. Rocz Akad Med Bialyms 2005;50(Suppl 1): 294-6. [3] Haviv YS, Safadi R. Normotensive scleroderma renal crisis: case report and review of the literature. Ren Fail 1998;20:733-6. [4] Collins DA, Patel S, Eastwood JB, et al. Favorable outcome of scleroderma renal crisis. J R Soc Med 1996;89:49-50. [5] Steen VD, Costantino JP, Shapiro AP, et al. Outcome of renal crisis in systemic sclerosis: relation to availability of angiotensin converting enzyme (ACE) inhibitors. Ann Intern Med 1990;113:352-7. [6] Charles C, Clements P, Furst DE. Systemic sclerosis: hypothesisdriven treatment strategies. Lancet 2006;367:1683-91. [7] Rhew EY, Barr WG. Scleroderma renal crisis: new insights and developments. Curr Rheumatol Rep 2004;6:129-36. [8] Chizzolini C. Update on pathophysiology of scleroderma with special reference to immunoinflammatory events. Ann Med 2007;39:42-53. [9] Haynes DC, Gershwin ME. The immunopathology of progressive systemic sclerosis. Semin Arthritis Rheum 1982;11:331-51. [10] Padula SJ, Clark RB, Korn JH. Cell mediated immunity in rheumatic disease. HUM PATHOL 1986;17:254-63. [11] Arnett FC. Is scleroderma an autoantibody mediated disease? Curr Opin Rheumatol 2006;18:579-81. [12] Sato S, Fujimoto M, Hasegawa M, et al. Altered B lymphocyte function induces systemic autoimmunity in systemic sclerosis. Mol Immunol 2004;41:1123-33. [13] Colvin RB. Antibody-mediated renal allograft rejection: diagnosis and pathogenesis. J Am Soc Nephrol 2007;18:1046-56. [14] Feucht HE, Schneeberger H, Hillebrand G, et al. Capillary deposition of C4d complement fragment and early renal graft loss. Kidney Int 1993;43:1333-8. [15] Herzenberg AM, Gill JS, Djurdjev O, Magil AB. C4d deposition in acute rejection: an independent long-term prognostic factor. J Am Soc Nephrol 2002;13:234-41. [16] Campbell RD, Gagnon J, Porter RR. Amino acid sequence around the thiol and reactive acyl groups of human complement component C4. Biochem J 1981;199:359-70. [17] Adapted from Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980;23:581-90. [18] Lonzetti LS, Joyal F, Raynauld JP, et al. Updating the American College of Rheumatology preliminary classification criteria for systemic sclerosis: addition of severe nailfold capillaroscopy abnormalities markedly increases the sensitivity for limited scleroderma. Arthritis Rheum 2001;44:735-6. [19] Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant 2008;8:753-60. [20] Helfrich DJ, Banner B, Steen VD, et al. Normotensive renal failure in systemic sclerosis. Arthritis Rheum 1989;32:1128-34. [21] Penn H, Howie AJ, Kingdon EJ, et al. Scleroderma renal crisis: patient characteristics and long-term outcomes. QJM 2007;100:485-94. [22] Penn H, Howie AJ, Stratton RJ, et al. The prognostic value of renal biopsy in scleroderma renal crisis. Arthritis Rheum 2007;56:53-4 (Abstract #3).
340 [23] Howie AJ, Ferreira MA, Adu D. Prognostic value of simple measurement of chronic damage in renal biopsy specimens. Nephrol Dial Transplant 2001;16:1163-9. [24] Sakkas LI, Chikanza IC, Platsoucas CD. Mechanisms of disease: the role of immune cells in the pathogenesis of systemic sclerosis. Nat Clin Pract Rheumatol 2006;2:679-85. [25] Chang M, Wang RJ, Yangco DT, et al. Analysis of autoantibodies against RNA polymerases using immunoaffinity-purified RNA polymerase I, II, and III antigen in an enzyme-linked immunosorbent assay. CLIM Immunol Immunopathol 1998;89:71-8. [26] Santiago M, Baron M, Hudson M, et al. Antibodies to RNA polymerase III in systemic sclerosis detected by ELISA. J Rheumatol 2007;34:1528-34. [27] Hudson M, Walker JG, Fritzler M, et al. Hypocomplementemia in systemic sclerosis—clinical and serological correlations. J Rheumatol 2007;34:2218-23.
I. Batal et al. [28] Senaldi G, Lupoli S, Vergani D, et al. Activation of the complement system in systemic sclerosis. Relationship to clinical severity. Arthritis Rheum 1989;32:1262-7. [29] Li SJ, Liu ZH, Zen CH, et al. Peritubular capillary C4d deposition in lupus nephritis different from antibody-mediated renal rejection. Lupus 2007;16:875-80. [30] Kayler LK, Kiss L, Sharma V, Mohanka R, Zeevi A, Girnita A, et al. Acute renal allograft rejection: diagnostic significance of focal peritubular capillary C4d. Transplantation 2008;27:813-20. [31] Akalin E, Dinavahi R, Dikman S, et al. Transplant glomerulopathy may occur in the absence of donor-specific antibody and C4d staining. Clin J Am Soc Nephrol 2007;2:1261-7. [32] Smith RN, Kawai T, Boskovic S, et al. Chronic antibody mediated rejection of renal allografts: pathological, serological and immunologic features in nonhuman primates. Am J Transplant 2006;6: 1790-8.
www.elsevier.com/locate/humpath
0riginal contribution
Renal biopsy findings predicting outcome in scleroderma renal crisis☆ Ibrahim Batal MD a , Robyn T. Domsic MD b , Aaron Shafer MD c , Thomas A. Medsger Jr. MD b , Lawrence P. Kiss MD a , Parmjeet Randhawa MD a , Sheldon Bastacky MD a,⁎ a
Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Division of Rheumatology and Clinical Immunology, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA c Division of Gynecologic Oncology, University of North Carolina, Chapel Hill, NC 27522, USA b
Received 29 May 2008; revised 1 August 2008; accepted 1 August 2008
Keywords: Scleroderma renal crisis; Vascular thrombosis; Glomerular ischemic collapse; C4d deposits
Summary Scleroderma renal crisis is irreversible in some patients despite aggressive treatment. This study was designed to identify pathologic prognostic features in scleroderma renal crisis. We retrospectively reviewed the pathology and the clinical records of 17 patients who underwent kidney biopsies during scleroderma renal crisis (group A, recovered renal function [n = 7]; group B, remained in renal failure or died [n = 10]). Multiple histologic features were assessed semiquantitatively (0-3) or as percentages. C4d staining of peritubular capillaries and small vessels was assessed semiquantitatively (0-3) in patients with scleroderma (n = 11), normotensive (n = 10), and hypertensive (n = 12) nonscleroderma native kidney controls. The percentage of thrombosed vessels (25.1 ± 21.0 versus 5.6 ± 12.3, P = .045) and the severity of glomerular ischemic collapse (2.9 ± 0.3 versus 1.4 ± 0.8, P = .001) were significantly higher in group B than in group A. Also, group B patients tended to have more severe acute tubular injury and vascular fibrinoid changes. The peritubular capillary C4d score in patients with scleroderma, normotensive controls, and hypertensive controls were 1.1 ± 0.9, 0.3 ± 0.7, and 0.3 ± 0.5, respectively (P = .018, scleroderma versus other controls). Small vessel C4d score was higher in scleroderma compared to normotensive but not hypertensive controls. Within scleroderma samples, a significantly higher peritubular capillary C4d score (1.6 ± 0.7 versus 0.3 ± 0.5, P = .024) but not small vessel score was found in group B compared to group A. This tended to be associated with peritubular capillary leukocyte margination. Vascular thrombosis, severe glomerular ischemic collapse, and peritubular capillary C4d deposits in scleroderma renal crisis kidney biopsies correlated with increased risk of failure to recover renal function. © 2009 Elsevier Inc. All rights reserved.
1. Introduction ☆
This work was presented at the 2008 annual United States and Canadian Academy of Pathology (USCAP) meeting (Denver, CO). ⁎ Corresponding author. E-mail address: [email protected] (S. Bastacky). 0046-8177/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2008.08.001
Systemic sclerosis is a multisystem disorder affecting small blood vessels and connective tissues. Its standard classification is as diffuse cutaneous or limited cutaneous variants [1]. Scleroderma renal crisis (SRC) is a
Renal biopsy findings predicting outcome in scleroderma renal crisis complication of systemic sclerosis that occurs in up to 10% of patients. SRC is typically characterized by the new onset of hypertension (although few patients remain normotensive) and acute renal failure, with or without evidence of microangiopathic hemolytic anemia or thrombocytopenia [2,3]. The mortality associated with SRC has diminished significantly because of early diagnosis and angiotensin-converting enzyme inhibitor therapy [4,5]. Nevertheless, some patients are refractory to angiotensinconverting enzyme inhibitor therapy and remain on dialysis or die [2,6,7]. Although many studies have been conducted to investigate its pathogenesis, the etiology of systemic sclerosis remains incompletely understood [8]. Some investigators believe that systemic sclerosis is a disorder of cell-mediated immunity [9,10]; some contend that antibody-mediated injury plays an important role [11], whereas others stress the importance of T cell–B cell interaction [1]. Patients with systemic sclerosis have expansion in naive B cells, decrease in memory B cells, and increase in CD19 expression. This may result in antibody production through inhibition of B-cell peripheral tolerance [12]. C4d is a classic pathway complement degradation product. Its presence is associated with antibody-mediated injury in renal allografts, and it has been shown to independently predict unfavorable renal allograft outcome [13-15]. Activated C4 is cleaved to C4a and C4b. C4b binds to amino or hydroxyl groups and is then converted to C4d, which can covalently bind to the peritubular capillary (PTC) basement membrane [16]. C4d staining pattern in SRC has not been investigated yet. This study was designed to identify potential clinical and pathologic prognostic factors and to investigate C4d deposition in SRC.
2. Materials and methods 2.1. Study design We retrospectively reviewed clinical and surgical pathology records of 17 patients with scleroderma who underwent native renal biopsies during SRC between 1990 and 2007, with the last clinical follow-up in October 2007. Fifteen patients with SRC were seen at the University of Pittsburgh Medical Center and met the original classification criteria proposed by the American College of Rheumatology [17] for systemic sclerosis and the recent revisions [18]. The remaining 2 patients had their biopsy specimens referred from an outside institution with the diagnosis of systemic sclerosis and worsening renal function. Ten specimens derived from nonneoplastic areas of kidneys resected for renal neoplasms and 12 from non-SRC patients with hypertension (5 idiopathic malignant hypertension and 7 chronic benign hypertension) were used as controls for C4d staining in native kidneys.
333
2.2. Histologic evaluation Standard hematoxylin and eosin, periodic acid–Schiff, silver, and trichrome sections were evaluated. Histologic changes in all 4 renal anatomic compartments (glomeruli, tubules, interstitium, blood vessels) were assessed and scored semiquantitatively (0-3) or recorded as percentages. The glomerular alterations assessed included glomerular ischemic collapse, glomerular basement membrane (GBM) double contours (tram tracking), and juxtaglomerular apparatus (JGA) prominence. These changes were graded on a semiquantitative scale (0-3) according to the severity of the most affected nonsclerotic glomerulus. Glomerular ischemic collapse was graded as no (0), mild (1), moderate (2), or severe (3) glomerular collapse based on the most severely affected glomeruli. Glomerular double contour was graded from 0 to 3 when double contour lesion was seen in less than 10%, 11% to 25%, 26% to 50%, or more than 50% of capillary loops, respectively. JGA status was graded on a scale (0-3) corresponding to inconspicuous JGA, mild hyperplasia, moderate hyperplasia, and severe hyperplasia, respectively. Global glomerulosclerosis and glomerular thrombosis were recorded as a percentage of the number of affected glomeruli. Tubular pathologic alterations were assessed. Acute tubular injury was graded on a semiquantitative scale based on the most affected area, corresponding to no injury (0), wrinkling of tubular basement membrane or tubular epithelial vacuolization (1), attenuation of tubular epithelium (2), and necrosis (3), respectively. Tubular atrophy was graded on a semiquantitative scale according to the cortical surface area involved by tubular atrophy: no atrophy (0), and atrophy involving up to 25% (1), 26% to 50% (2), and more than 50% (3) of cortical tubular profiles, respectively. Interstitial fibrosis was graded on a semiquantitative scale based on the cortical surface area involved, corresponding to less than 5% (0), 6% to 25% (1), 26% to 50% (2), and more than 50% (3), respectively. Vascular abnormalities were separated into acute (vascular myxoid and fibrinoid changes), subacute (onion skin lesions), and chronic (fibroelastic intimal thickening, muscular thickening, and hyalinosis). Each abnormality was graded on a semiquantitative severity scale (0-3) based on the most severely affected vessel. Fibrinoid change was evaluated according to extension of this lesion within vascular wall as no fibrinoid alteration (0), fibrinoid change involving the intima up to the muscularis (1), through the muscularis (2), and the entire wall thickness (3). Onion skin lesion was evaluated according to the severity of proliferative endarteropathy and concentric fibrosis within the involved arteriole/ interlobular artery as no (0), early (1), well established (2), or advanced (3) onion skinning lesion. Fibroelastic intimal thickening and myxoid changes were graded on a scale (0-3) corresponding to no significant narrowing, narrowing of up to 25%, 26% to 50%, and more than 50% of luminal cross-sectional surface
334
I. Batal et al.
caused by fibrosis or accumulation of myxoid material, respectively. Muscular thickening was graded as no (0), mild (1), moderate (2), or severe (3). Arteriolar hyalinosis was graded on a scale (0-3) corresponding to no, mild, moderate, or severe periodic acid-Schiff–positive hyaline thickening, respectively. Vascular thrombosis, considered to be an acute pathologic injury, was recorded as percentage of thrombosed arterioles/arteries. Finally, leukocyte margination within peritubular capillaries was semiquantitavely graded. The lesion was considered to be absent when less than 10% of peritubular capillaries showed leukocyte margination. It was graded as 1, 2, or 3 when more than 10% of peritubular capillaries showed leukocyte margination and the most severely affected cross-sectioned capillary contained up to 4, 5 to 10, or more than 10 white blood cells (neutrophilic and/or mononuclear), respectively. Each case was histologically evaluated by the first author (I. B.) and the senior author (S. B.). All cases were also reviewed on a multiheaded microscope to eliminate any possible discrepancy.
fields. Staining was referred to as focal when present in 10% to 50% of high-power microscopic fields, and at least one cluster of positive peritubular capillaries could be demonstrated. Lesser degree of staining was considered minimal, and absent staining was considered negative. Areas of atrophy were excluded from evaluation. C4d staining in small vessels, namely, arterioles and small arteries, was classified as diffuse, focal, minimal, or negative when more than 50%, 26% to 50%, 5% to 25%, or less than 5% of small vessels showed C4d deposits, respectively. To facilitate comparison, C4d staining was graded using a semiquantitative score for both PTC and arteries (0-3) corresponding to negative (0), minimal (1), focal (2), and diffuse (3), respectively. Glomerular capillary staining was recorded as present or absent. When present, the stain was further characterized as minimal, segmental, or global. Evaluation of C4d staining was performed by the first author (I. B.). The results were confirmed by the sixth author (P. R.) using a multiheaded microscope.
2.4. Statistics 2.3. C4d staining C4d immunostaining was performed using rabbit antihuman C4d polyclonal antibody (ALPCO Diagnostics, Windham, NH). Four-micrometer formalin-fixed, paraffinembedded sections of native kidneys were deparaffinized, placed in a pressure cooker for 20 minutes, and sequentially incubated at 37°C with TRIS-EDTA (pH 8.5), 1:50 dilution of primary antibody for 44 minutes, 1:100 dilution of a biotinylated antirabbit secondary antibody for 8 minutes, streptavidin-alkaline phosphate conjugate for 8 minutes, fast red or brown A-naphthol for 8 minutes, and fast red or brown B-naphthol for 8 minutes. Sections were then counterstained with hematoxylin. All reagents were components of the Ventana Enhanced Alkaline Phosphatase Red or Brown Detection Kit (cat no. 760-031, Ventana Medical Systems, Tucson, AZ). PTC staining was classified as diffuse when staining was present in more than half of the high-power microscopic Table 1
Statistical comparison between different study groups was performed using the Mann-Whitney rank sum test or χ2 test (P values expressed as Fisher exact test) using Sigma Stat 2.0.3 software (SPSS Inc, Chicago, IL). P values less than .05 were considered statistically significant. The study was conducted in accordance with procedures established by the University of Pittsburgh Institutional Review Board (protocol no. 0701150).
3. Results Of the 17 patients, 13 were white and 4 were African American. Four patients were male and thirteen were female. The age range was 23 to 73 years with a mean ± SD of 51.7 ± 16.4 years. Mean ± SD diastolic blood pressure and serum creatinine (Cr) available at the time of biopsy were 102 ± 24
Demographic characteristics
Age (y) a Sex (male/female) White/African American Cr at time of biopsy (mg/dL) a Diastolic blood pressure (mm Hg) a Disease duration (years from clinical diagnosis to biopsy) a
Group A (n = 7) recovered
Group B (n = 10) HD/transplant/ death on dialysis
P
43.7 ± 16.0 1:6 4:3 4.2 ± 3.4 116 ± 13 1.4 ± 1.9
57.3 ± 15.0 3:7 9:1 5.1 ± 2.3 91 ± 25 3.1 ± 3.6
.18 .6 .25 .29 .023 .78
NOTE. Serum Cr and diastolic blood pressure at the time of biopsy were not available for 1 group B patient. Disease duration information was available for 7 group A patients and 4 group B patients. Abbreviation: HD, hemodialysis. a Data presented as mean ± SD.
Renal biopsy findings predicting outcome in scleroderma renal crisis mm Hg and 4.7 ± 2.8 mg/dL, respectively (n = 16). The follow-up period ranged between 51 days and 13.1 years with a mean ± SD of 3.0 ± 3.9 years. All the patients with a follow-up period less than 6 months were the patients who died with renal failure due to severe disease except one patient who was on dialysis for 3 months at the end of follow-up period. Mean ± SD of disease duration from first diagnosis of scleroderma to biopsy, when information was available (n = 11), was 2.0 ± 2.6 years. Patients were divided into 2 groups based on renal function status: group A (recovered renal function) and group B (persistent renal failure requiring dialysis or transplant, or death with renal failure). There were no significant differences in age, sex, race, presenting serum Cr, and disease duration between groups A and B patients. Diastolic blood pressure was significantly higher in group A (116 ± 13 mm Hg) compared to group B (91 ± 25 mm Hg) (P = .023) (Table 1). Serum autoantibody results were available in 16 patients. Of these, scleroderma-associated antibodies were identified in 6 patients in group A including RNA polymerase III (n = 2), U1 RNP (n = 1), U3 RNP (n = 1), Th/To (n = 1), and PL7 (n = 1). In 6 patients in group B, sclerodermaassociated autoantibodies were detected including RNA polymerase III (n = 3), Scl70 (n = 1), p-ANCA (n = 1), and smooth muscle (n = 1). Histologic features were compared between groups A and B (Table 2). Glomerular ischemic collapse scores (Fig. 1) were significantly higher in group B (2.9 ± 0.3) compared to group A (1.4 ± 0.8; P = .001). The other parameters assessed (global glomerulosclerosis, tram tracking, juxtaglomerular apparatus hyperplasia) did not differ significantly between the groups. Although it did not reach statistical significance, none of the patients with glomerular capillary thrombosis (n = 3, all in group B) (Fig. 2) recovered renal function. The acute tubular injury scores tended to be higher for group B Table 2
Fig. 1 Severe glomerular ischemic collapse in a SRC patient who failed to recover renal function (methenamine silver, original magnification ×400).
(1.8 ± 0.6) compared to group A (1.1 ± 0.7; P = .11), whereas chronic tubular atrophy scores were similar. Also, the interstitial fibrosis score was similar in group A versus group B. The proportion of thrombosed vessels (Fig. 3) was significantly higher in patients who did not recover renal function (25.1 ± 21.0 group B versus 5.6 ± 12.3 group A; P = .045). Noteworthy, an area of infarction was noted in one biopsy with no detectable thrombosis. This likely represented a manifestation of thrombosis or critical vascular narrowing in an unsampled larger artery. When the percentage of infarcted areas was considered in evaluation together with the percentage of thrombosed vessels, the difference became even greater (28.5 ± 19.2 group B versus 5.6 ± 12.3 group A, P = .017). The fibrinoid change score tended to be higher in group B but did not achieve statistical
Histopathologic comparison between groups A and B
Glomerular
Tubular Interstitium Vascular pathology
a
335
Lesions
Group A (n = 7) a
Group B (n = 10) a
P
Global sclerosis (%) Thrombosis (%) Ischemic collapse (0-3) Tram tracking (GBM double contour) (0-3) JGA hyperplasia (0-3) Acute injury (0-3) Atrophy (0-3) Fibrosis (0-3) Thrombosed vessels (%) Fibrinoid changes (0-3) Myxoid changes (0-3) Onion skinning (0-3) Fibroelastic intimal thickening (0-3) Muscular thickness (0-3) Hyalinosis (0-3) Peritubular capillaritis (0-3)
9.7 0.0 1.4 0.1 1.3 1.1 0.4 0.7 5.6 0.0 1.6 1.1 1.3 0.4 0.3 0.7
13.8 ± 13.9 1.4 ± 3.2 2.9 ± 0.3 0.3 ± 0.7 1.5 ± 0.1 1.8 ± 0.6 0.4 ± 0.7 0.8 ± 0.8 25.1 ± 21.0 0.7 ± 1.0 1.6 ± 1.3 0.7 ± 1.0 1.8 ± 0.8 1.2 ± 1.0 0.5 ± 1.1 1.1 ± 1.1
.3 .32 .001 .84 .7 .11 .8 .96 .045 .094 1.0 .52 .35 .14 .96 .6
Data are presented as mean ± SD.
± 13.0 ± 0.0 ± 0.8 ± 0.4 ± 1.1 ± 0.7 ± 0.5 ± 0.5 ± 12.3 ± 0.0 ± 1.5 ± 1.2 ± 1.1 ± 0.5 ± 0.5 ± 0.8
336
I. Batal et al.
Fig. 4 Arterial fibrinoid change observed in a SRC patient who failed to recover renal function. In addition, ischemic glomerular collapse and acute tubular injury are seen in this field (hematoxylin and eosin, original magnification ×200). Fig. 2 Glomerular capillary thrombosis in a SRC patient who failed to recover renal function (methenamine silver, original magnification ×600).
significance (0.7 ± 1.0 versus 0 ± 0; P = .094). Yet, all patients with fibrinoid change (n = 5, all in group B) (Fig. 4) developed refractory renal failure. No differences between the 2 groups were found for vascular myxoid change, onion skin lesions, fibroelastic intimal thickening, muscular hypertrophy, or arteriolar hyalinosis. The mean score of leukocyte margination in PTC was higher in group B (1.1 ± 1.1) compared to group A (0.7 ± 0.8), but this was not significantly different. To further investigate if the presence or absence of the manifestations of acute endothelial injury within biopsy can affect prognosis, we again treated vascular thrombosis/ infarction, fibrinoid changes, glomerular thrombosis, and
Fig. 3 Arterial thrombosis in a SRC patient who failed to recover renal function (methenamine silver, original magnification ×100).
vascular myxoid changes as noncontinuous variables and reevaluated using the Fisher exact test. A significant difference in the renal outcome was identified for vascular thrombosis/infarction (2/7 group A versus 9/10 group B, P = .035), and fibrinoid change (0/7 group A versus 5/10 group B, P = .044) but not glomerular thrombosis (0/7 group A versus 3/10 group B, P = .23) or myxoid change (4/7 group A versus 7/10 group B) (Fig. 5). Paraffin tissue blocks for 11 scleroderma patients (group A [n = 4], group B [n = 7]) were available for immunohistochemical evaluation. These biopsies, as well as 10 normotensive and 12 hypertensive non-SRC specimens retrieved from our files to serve as controls, were stained with C4d immunohistochemical stain. The scores were
Fig. 5 Arterial myxoid changes observed in a SRC patient who recovered renal function (methenamine silver, original magnification ×200).
Renal biopsy findings predicting outcome in scleroderma renal crisis Table 3
337
Composition of C4d staining in peritubular capillaries and arteries in SRC patients versus controls
PTC
Small vessels
Diffuse Focal Minimal Negative Score (0-3) Diffuse Focal Minimal Negative Score (0-3)
Normotensive controls (n = 10)
Hypertensive controls (n = 12)
All SRC (n = 11)
0 (0%) 1 (10%) 1 (10%) 8 (80%) 0.3 ± 0.7 0 (0%) 2 (20%) 1 (10%) 7 (70%) 0.5 ± 0.9
0 (0%) 0 (0%) 3 (25%) 9 (75%) 0.3 ± 0.5 6 (50%) 1 (8%) 2 (17%) 3 (25%) 1.8 ± 1.4
1 (9%) 2 (18%) 5 (46%) 3 (27%) 1.1 ± 0.9 ⁎ 5 (46%) 3 (27%) 2 (18%) 1 (9%) 2.1 ± 1.0 †
NOTE. Scores are presented as mean ± SD. ⁎ P = .018, PTC C4d score (Kruskal Wallis). P = .048 and P = .024, SRC patients versus normotensive controls and SRC patients versus non-SRC hypertensive controls (Mann-Whitney). † P = .008, Small vessels C4d score (Kruskal Wallis). P = .005 and P = .025 SRC patients versus normotensive controls and non-SRC hypertensive controls versus normotensive controls (Mann-Whitney).
semiquantitated for both PTC and small vessels as described earlier (Table 3). There was a significant difference comparing PTC C4d score between SRC patients (Fig. 6) (1.1 ± 0.9) and both normotensive controls (0.3 ± 0.7, P = .048) and hypertensive non-SRC controls (0.3 ± 0.5, P = .024) (Table 3). Diffuse staining was only observed in SRC cases (n = 1), whereas focal staining was observed additionally in one normotensive control. C4d deposition in small arteries and arterioles in SRC patients was significantly greater compared to normotensive controls (2.1 ± 1.0 versus 0.5 ± 0.9; P = .005) but not hypertensive non-SRC controls (1.8 ± 1.4), which also showed greater staining compared to the normotensive group (P = .025). Within the hypertensive non-SRC control group, diffuse C4d deposition in small vessels was observed more frequently
in acute malignant hypertension samples (4/5 idiopathic malignant hypertension versus 2/7 benign hypertension). C4d staining patterns within PTC as well as small vessels were compared between SRC groups A and B (Table 4). Although the sample size was small, PTC C4d score was significantly higher in group B (1.6 ± 0.7) versus group A (0.3 ± 0.5) (P = .024); moreover, diffuse and focal PTC stain was exclusively seen in group B. No difference was observed when small vessel C4d staining was evaluated (2.1 ± 1.2 [group B] versus 2.0 ± 0.8 [group A]). PTC leukocyte margination was then evaluated in all specimens stained with C4d. The mean PTC leukocyte margination score was highest in SRC patients (0.94 ± 0.97) and lowest in non-SRC normotensive controls (0.1 ± 0.3, P = .028). The non-SRC hypertensive controls had a mean value of 0.6 ± 0.7 and was not significantly different when compared to the aforementioned groups.
Table 4 Composition of C4d staining in peritubular capillaries and arteries in SRC patients who recovered renal function (group A) versus who failed to recover renal function (group B) Group A (n = 4) Group B (n = 7) PTC
Fig. 6 Finely granular C4d deposits in peritubular capillaries in a biopsy from a SRC patient who failed to recover renal function. Although not all peritubular capillaries stained positively, greater than 50% of the sampled biopsy area showed capillaries with C4d deposition. Hence, the biopsy was scored as diffusely positive, as suggested by the Banff 2007 classification for renal allograft pathology [19] (polyclonal C4d immunostain performed on formalin-fixed, paraffin-embedded tissue, original magnification ×400).
Diffuse Focal Minimal Negative Score (0-3) Small vessels Diffuse Focal Minimal Negative Score (0-3)
0 (0%) 0 (0%) 1 (25%) 3 (75%) 0.3 ± 0.5 2 (25%) 2 (50%) 1 (25%) 0 (0%) 2.0 ± 0.8
1 (14%) 2 (29%) 4 (57%) 0 (0%) 1.6 ± 0.7 ⁎ 4/7 (58%) 1/7 (14%) 1/7 (14%) 1/7 (14%) 2.1 ± 1.2
NOTE. Scores are presented as mean ± SD. ⁎ P = .024, PTC C4d score group A versus group B (MannWhitney).
338 Noteworthy, for SRC patients, all biopsies with diffuse and focal PTC C4d deposits (n = 3) had high PTC leukocyte margination score (2.3 ± 0.6). This score was significantly different when compared to the other SRC biopsies with negative and minimal PTC C4d score (0.9 ± 0.8, P = .041). In SRC, the PTC mononuclear component was the predominant marginating white blood cell component except for one case with focal PTC staining with a predominant neutrophilic component. All SRC patients with high PTC C4d score (2-3) had mild mononuclear interstitial inflammation. Evaluation of glomerular C4d staining revealed that the stain was completely negative in all but 4 biopsies. Two of these 4 had segmental staining in up to 3 glomeruli, whereas minimal staining was detected in the remaining 2 specimens. All of these 4 cases were associated with detectable PTC C4d deposits and 3 of 4 were in group B patients. None of the specimens that were stained for C4d (n = 11) showed glomerular double contour. Finally, in our series of SRC patients with detectable PTC C4d staining, review of the immunofluorescence did not reveal any PTC immunoglobulin staining. IgA (+1) in arteries (n = 1), IgM (+2-4) in arterioles and arteries (n = 2), and IgG (+1) in the interstitium (n = 1) were detected in the 3 patients with focal and diffuse PTC C4d staining.
4. Discussion SRC is a serious complication of systemic sclerosis that may lead to permanent renal failure or death. Some patients recover renal function while others do not. Because SRC is a clinical diagnosis, renal biopsy is not frequently performed. We investigated whether any clinical or pathologic features would predict poor renal functional status after SRC. No significant difference was found comparing age, sex, race, serum Cr at the time of biopsy, or disease duration in this small patient population. A significantly lower diastolic blood pressure was found within the group who failed to recover renal function. This is consistent with previous observations that less hypertensive or normotensive SRC patients have reduced survival, perhaps due to a delay in diagnosis and treatment [3,20]. In our study, the histologic changes reflecting the most severe acute vascular injury such as vascular thrombosis and glomerular ischemic collapse, and to a lesser extent acute tubular injury, were associated with a worse prognosis. When treated as categorical variable, the presence of fibrinoid change was also associated with a worse prognosis. Although not significantly different between groups A and B, thrombosis in glomerular capillaries was consistently associated with a poor outcome. This small retrospective study revealed that renal biopsies may not only confirm a suspected diagnosis of SRC, but more importantly, there are several pathologic
I. Batal et al. features that may have prognostic significance. Penn et al [21] in a single-center study found that acute vascular changes, namely, myxoid and fibrinoid changes and vascular thrombosis were associated with poor prognosis in SRC. These authors combined the acute changes and treated them as categorical variables. In our study, the presence of myxoid change was not associated with an adverse renal outcome and was observed frequently in patients who recovered renal function (Fig. 5). It would be of interest to know if vascular myxoid change without fibrinoid change was independently predictive of an adverse renal outcome in the patients in Penn's study. Our data suggest that vascular myxoid change in the absence of fibrinoid change does not irreversibly result in permanent vascular occlusion or destruction of the involved vessel if the SRC is successfully treated. In addition, Penn et al [22] in a recently published abstract pooled data from multiple medical centers and confirmed their previous observation. Yet, surprisingly, they found an association between chronic pathologic changes and better prognosis. In our study, we did not observe such an association. This might be attributed to the sample size, which is larger than ours, and to different criteria to evaluate chronic injury. Penn et al used a composite index of chronic damage to evaluate chronic pathologic changes. This index is calculated by a software program that combines pathologic changes in different anatomical compartments and expresses the final results as percentages [23]. We assessed chronic injury by evaluating each renal compartment individually. Our analysis does not suggest that chronic injury is a protective factor by itself. It could be a result of a milder ongoing active injury, or alternatively, an expression of previous resolving more severe acute injury. The pathogenesis of scleroderma remains obscure. Autoantibodies may play a pathogenetic role either by inducing endothelial apoptosis (antiendothelial cell antibodies) or by activating fibroblasts (antifibrillin-1 antibodies) [11]. Alternatively, activated T cells, which are usually Th-2, may produce interleukin 4 and 13, leading to fibrosis [24]. Additional evidence of a potential role of autoantibodies in SRC is its association with anti-RNP polymerase III antibodies [25,26]. Complement activation comparing systemic sclerosis patients with controls has been reported [27,28]. We assessed endothelial C4d deposition, an early classic complement pathway marker, using a polyclonal antibody immunostain. In our study, SRC biopsies showed significantly greater PTC C4d deposits compared to both normotensive non-SRC controls and hypertensive non-SRC controls. Furthermore, PTC leukocyte margination score was significantly higher in SRC compared to normotensive controls. The mean value of this score was also higher in SRC patients compared with that of hypertensive controls, but this was not significantly different. Within SRC specimens, we found significantly greater PTC C4d staining in patients who irreversibly lost renal
Renal biopsy findings predicting outcome in scleroderma renal crisis function or died on dialysis compared to patients who recovered renal function. This was accompanied by PTC leukocyte margination and mild interstitial inflammation. Li et al [29] showed that diffuse C4d deposition in PTC was detected in 7% of patients with lupus nephritis. This is comparable to our observation of diffuse PTC C4d deposition in 9% of our SRC patients. Lupus patients with diffuse PTC C4d staining showed higher clinical activity scores and lower serum complement levels. Li et al indicated that these deposits are different from that observed in antibody-mediated rejection, in the sense of being granular and associated with immunoglobulin G and immune complex deposits. Similar to lupus deposits, PTC C4d deposits in SRC were finely granular (Fig. 6); yet, unlike lupus nephritis, staining in our SRC biopsies was not associated with immune deposits. Noteworthy, staining for C4d in our laboratory tends to be finely granular rather than uniformly linear in our renal allograft biopsies. We regard this finely granular staining as real because it correlates well with presence of donor-specific antibodies in kidney transplant patients [30]. A finely granular pattern of staining has also been noted in other laboratories [31,32]. An additional study using monoclonal C4d staining on frozen tissue is needed to further characterize these deposits; however, at this time, the presence of such deposits at least raises the possibility of antibody-mediated injury in a subset of SRC patients. Finely granular GBM C4d deposits were present in a small subset of SRC. These deposits were always found in association with PTC staining and were not attributed to GBM remodeling. This also suggests that the same mechanism responsible for C4d deposition in PTC may be responsible for GBM deposits. C4d deposition in extraglomerular small vessels is commonly detected in allograft biopsies (unpublished observation). Vascular C4d deposition may represent nonspecific complement activation after endothelial injury due to hypertension, drugs, or any type of mechanical stress. The presence of more frequent C4d staining in malignant hypertension compared with chronic benign hypertension suggests an association with acute endothelial injury. The retrospective nature and small sample size of this study limit its conclusions, although this is one of the largest pathologic single-center studies performed on SRC. Larger prospective studies are needed to confirm our preliminary results and to determine if renal biopsies can be used to stratify SRC patients according to their risk of developing therapy-refractory renal failure. This could lead to optimization of therapy in SRC, with patients separated into those in whom aggressive therapy is likely to be beneficial and those in whom early consideration of renal transplant would be most appropriate. In conclusion, diastolic blood pressure during SRC presentation was significantly lower in patients who failed to recover renal function, which may reflect delayed diagnosis and more severe ongoing renal injury. Renal
339
biopsies during SRC may have prognostic value. Finally, PTC C4d deposits may be an unfavorable prognostic feature in SRC and suggest the possibility of ongoing antibodymediated injury in a subset of SRC patients.
References [1] Sakkas. New developments in the pathogenesis of systemic sclerosis. Autoimmunity 2005;38:113-6. [2] Lewandowski B, Domysławska I, Klimiuk PA, et al. Kidney crisis in systemic sclerosis. Rocz Akad Med Bialyms 2005;50(Suppl 1): 294-6. [3] Haviv YS, Safadi R. Normotensive scleroderma renal crisis: case report and review of the literature. Ren Fail 1998;20:733-6. [4] Collins DA, Patel S, Eastwood JB, et al. Favorable outcome of scleroderma renal crisis. J R Soc Med 1996;89:49-50. [5] Steen VD, Costantino JP, Shapiro AP, et al. Outcome of renal crisis in systemic sclerosis: relation to availability of angiotensin converting enzyme (ACE) inhibitors. Ann Intern Med 1990;113:352-7. [6] Charles C, Clements P, Furst DE. Systemic sclerosis: hypothesisdriven treatment strategies. Lancet 2006;367:1683-91. [7] Rhew EY, Barr WG. Scleroderma renal crisis: new insights and developments. Curr Rheumatol Rep 2004;6:129-36. [8] Chizzolini C. Update on pathophysiology of scleroderma with special reference to immunoinflammatory events. Ann Med 2007;39:42-53. [9] Haynes DC, Gershwin ME. The immunopathology of progressive systemic sclerosis. Semin Arthritis Rheum 1982;11:331-51. [10] Padula SJ, Clark RB, Korn JH. Cell mediated immunity in rheumatic disease. HUM PATHOL 1986;17:254-63. [11] Arnett FC. Is scleroderma an autoantibody mediated disease? Curr Opin Rheumatol 2006;18:579-81. [12] Sato S, Fujimoto M, Hasegawa M, et al. Altered B lymphocyte function induces systemic autoimmunity in systemic sclerosis. Mol Immunol 2004;41:1123-33. [13] Colvin RB. Antibody-mediated renal allograft rejection: diagnosis and pathogenesis. J Am Soc Nephrol 2007;18:1046-56. [14] Feucht HE, Schneeberger H, Hillebrand G, et al. Capillary deposition of C4d complement fragment and early renal graft loss. Kidney Int 1993;43:1333-8. [15] Herzenberg AM, Gill JS, Djurdjev O, Magil AB. C4d deposition in acute rejection: an independent long-term prognostic factor. J Am Soc Nephrol 2002;13:234-41. [16] Campbell RD, Gagnon J, Porter RR. Amino acid sequence around the thiol and reactive acyl groups of human complement component C4. Biochem J 1981;199:359-70. [17] Adapted from Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980;23:581-90. [18] Lonzetti LS, Joyal F, Raynauld JP, et al. Updating the American College of Rheumatology preliminary classification criteria for systemic sclerosis: addition of severe nailfold capillaroscopy abnormalities markedly increases the sensitivity for limited scleroderma. Arthritis Rheum 2001;44:735-6. [19] Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant 2008;8:753-60. [20] Helfrich DJ, Banner B, Steen VD, et al. Normotensive renal failure in systemic sclerosis. Arthritis Rheum 1989;32:1128-34. [21] Penn H, Howie AJ, Kingdon EJ, et al. Scleroderma renal crisis: patient characteristics and long-term outcomes. QJM 2007;100:485-94. [22] Penn H, Howie AJ, Stratton RJ, et al. The prognostic value of renal biopsy in scleroderma renal crisis. Arthritis Rheum 2007;56:53-4 (Abstract #3).
340 [23] Howie AJ, Ferreira MA, Adu D. Prognostic value of simple measurement of chronic damage in renal biopsy specimens. Nephrol Dial Transplant 2001;16:1163-9. [24] Sakkas LI, Chikanza IC, Platsoucas CD. Mechanisms of disease: the role of immune cells in the pathogenesis of systemic sclerosis. Nat Clin Pract Rheumatol 2006;2:679-85. [25] Chang M, Wang RJ, Yangco DT, et al. Analysis of autoantibodies against RNA polymerases using immunoaffinity-purified RNA polymerase I, II, and III antigen in an enzyme-linked immunosorbent assay. CLIM Immunol Immunopathol 1998;89:71-8. [26] Santiago M, Baron M, Hudson M, et al. Antibodies to RNA polymerase III in systemic sclerosis detected by ELISA. J Rheumatol 2007;34:1528-34. [27] Hudson M, Walker JG, Fritzler M, et al. Hypocomplementemia in systemic sclerosis—clinical and serological correlations. J Rheumatol 2007;34:2218-23.
I. Batal et al. [28] Senaldi G, Lupoli S, Vergani D, et al. Activation of the complement system in systemic sclerosis. Relationship to clinical severity. Arthritis Rheum 1989;32:1262-7. [29] Li SJ, Liu ZH, Zen CH, et al. Peritubular capillary C4d deposition in lupus nephritis different from antibody-mediated renal rejection. Lupus 2007;16:875-80. [30] Kayler LK, Kiss L, Sharma V, Mohanka R, Zeevi A, Girnita A, et al. Acute renal allograft rejection: diagnostic significance of focal peritubular capillary C4d. Transplantation 2008;27:813-20. [31] Akalin E, Dinavahi R, Dikman S, et al. Transplant glomerulopathy may occur in the absence of donor-specific antibody and C4d staining. Clin J Am Soc Nephrol 2007;2:1261-7. [32] Smith RN, Kawai T, Boskovic S, et al. Chronic antibody mediated rejection of renal allografts: pathological, serological and immunologic features in nonhuman primates. Am J Transplant 2006;6: 1790-8.