ARTHRITIS & RHEUMATISM Vol. 50, No. 8, August 2004, pp 25692579 DOI 10.1002/art.

20433 2004, American College of Rheumatology

Antiphospholipid Syndrome Nephropathy in Patients With Systemic Lupus Erythematosus and Antiphospholipid Antibodies
Prevalence, Clinical Associations, and Long-Term Outcome
Maria G. Tektonidou,1 Flora Sotsiou,2 Lidia Nakopoulou,1 Panayiotis G. Vlachoyiannopoulos,1 and Haralampos M. Moutsopoulos1
Objective. To evaluate the prevalence, clinical associations, and outcome of antiphospholipid syndrome (APS) nephropathy in patients with systemic lupus erythematosus (SLE) and antiphospholipid antibodies (aPL) and in SLE patients without aPL. Methods. Kidney biopsy specimens obtained from 81 patients with aPL (18 of whom had APS) and 70 patients without aPL were retrospectively examined for the presence of APS nephropathy. Clinical and serologic data obtained at the time of kidney biopsy and during a mean followup of 7 years were recorded. In cases for which serial kidney biopsy specimens were available, the evolution of APS nephropathy was examined. Results. APS nephropathy existed in 39.5% of patients with aPL, compared with only 4.3% of patients without aPL. APS nephropathy was associated with both lupus anticoagulant and anticardiolipin antibodies. Among aPL-positive SLE patients, APS nephropathy was found in two-thirds of those with APS and in one-third of those without APS. A strong association between APS nephropathy and the presence of arterial thrombosis and livedo reticularis was noted. Patients with APS nephropathy had a higher frequency of hypertension and elevated serum creatinine levels at the time of kidney biopsy but did not have a higher frequency of
1 Maria G. Tektonidou, MD, Lidia Nakopoulou, MD, Panayiotis G. Vlachoyiannopoulos, MD, Haralampos M. Moutsopoulos, MD, FACP, FRCP: National University of Athens, Athens, Greece; 2Flora Sotsiou, MD: Evagelismos Hospital, Athens, Greece. Address correspondence and reprint requests to Maria G. Tektonidou, MD, Department of Pathophysiology, School of Medicine, National University of Athens, 75 Mikras Asias Street, 115 27 Athens, Greece. E-mail: balts@otenet.gr. Submitted for publication November 26, 2003; accepted in revised form April 13, 2004.

renal insufficiency, end-stage renal disease, or death at the end of followup. Serial kidney biopsy specimens were available from 11 patients and showed progression of APS nephropathy lesions. During followup, manifestations of APS (especially arterial thromboses) developed more frequently in the SLE/non-APS patients with APS nephropathy than in those without APS nephropathy. Conclusion. Among patients with SLE, APS nephropathy occurs almost exclusively in those with aPL, suggesting an important role of aPL in the pathogenesis of APS nephropathy. Patients with APS nephropathy develop hypertension, raised serum creatinine levels, and progression of histologic lesions, all of which are associated with a poor renal outcome. Manifestations of APS also tend to develop in these patients. APS nephropathy should be included in the APS classification criteria, and the use of appropriate anticoagulant therapy should be tested. The antiphospholipid syndrome (APS) is a multisystem disease characterized by arterial and venous thromboses, pregnancy morbidity, and the presence of antiphospholipid antibodies (aPL), namely anticardiolipin antibodies (aCL) and lupus anticoagulant (LAC) (1,2). APS is classified as primary or classified as secondary when it occurs in the context of other underlying disorders, mainly systemic lupus erythematosus (SLE) (3). Antiphospholipid antibodies are detected in 30 40% of patients with SLE; arterial or venous thrombosis will develop in nearly one-third of these patients (1,4). Thrombosis may occur at any vascular site and in any organ system. However, the intrarenal vascular involvement in association with APS was poorly recognized until recently. A possible explanation is that kidney
2569

2570

TEKTONIDOU ET AL

biopsies were rarely performed in patients with primary APS, and the majority of studies in SLE focused on immune complex glomerulonephritis rather than renal microangiopathy. The most commonly reported intrarenal vascular lesion in patients with APS (primary or secondary) (510) and in patients with SLE and aPL (1117) is thrombotic microangiopathy (TMA), which is characterized by the presence of fibrin thrombi in glomeruli and/or arterioles. In a recent multicenter retrospective study, Nochy et al examined kidney biopsy specimens obtained from 16 patients with primary APS (18). All of the patients had vasoocclusive lesions characterized by acute thromboses (TMA) and chronic vascular lesions such as fibrous intimal hyperplasia (FIH) of interlobular arteries, recanalizing thrombi in arteries and arterioles, fibrous occlusions, or focal cortical atrophy (FCA). Daugas et al observed that the above-mentioned histologic lesions, defined as APS nephropathy, exist also in SLEespecially in SLE patients with secondary APS independently of lupus nephritis (19). An association of APS nephropathy with APS (mainly with arterial thrombosis and abortions) and LAC, but not with aPL, was observed. In both of these studies, systemic hypertension was the main clinical manifestation of APS nephropathy. The purpose of the current study was to evaluate the following: 1) the presence of APS nephropathy in kidney biopsy specimens obtained from SLE patients with or without aPL, in order to assess any association between APS nephropathy and aPL, 2) clinical and laboratory associations with APS nephropathy, 3) the renal prognosis of patients with APS nephropathy, 4) the evolution of APS nephropathy in patients who underwent repeated kidney biopsies, and 5) any tendency of SLE/non-APS patients with APS nephropathy to develop APS during long-term followup. PATIENTS AND METHODS
Patients. The study group comprised all patients with SLE and biopsy-proven renal involvement (n 151) who were followed up on a regular basis at the Department of Pathophysiology and for whom at least 2 measurements of aPL were performed before or at the time of kidney biopsy. The patients were categorized as those with aPL (n 81), including those with APS (n 18), and those without aPL (n 70). All of the patients fulfilled at least 4 of the American College of Rheumatology criteria for the diagnosis of SLE (20) at the time of kidney biopsy. APS was diagnosed according to the Sapporo criteria (2). Patients who had aPL but did not have APS were defined as SLE/non-APS patients with aPL. The criteria for exclusion from the study were other potential causes for renal microangiopathy, such as systemic

sclerosis, malignant hypertension, thrombotic thrombocytopenic purpura, hemolytic uremic syndrome, postpartum renal failure, preeclampsia, diabetic nephropathy, human immunodeficiency virus infection, chemotherapy, or cyclosporine therapy. The kidney biopsies were performed from 1987 to 2002 in patients who had either proteinuria (urinary protein concentration 500 mg/dl), abnormal urinary sediment, or an elevated serum creatinine level. The biopsy specimens were retrospectively studied, using light microscopy, by 2 renal pathologists who had no knowledge of the clinical and laboratory features of the patients. The World Health Organization (WHO) classification of lupus glomerulonephritis (21) and the WHO activity and chronicity index scores (22) were reevaluated. The immunofluorescence microscopy findings as described in the initial reports of the kidney biopsies were accepted. As in previously published reports (18,19), APS nephropathy was diagnosed when at least 1 of the following lesions was detected: TMA, characterized by the presence of fibrin thrombi in arterioles and/or glomeruli (acute lesion), or myofibroblastic intimal cellular proliferation leading to intimal thickening of interlobular arteries [FIH]), organized thrombi with or without recanalization, fibrous arterial and arteriolar occlusion, and subcapsular zone with FCA (chronic lesions). In cases of APS nephropathy for which serial kidney biopsy specimens were available, the evolution of histologic lesions was examined. The following demographic, clinical, and serologic data obtained at the time of the first kidney biopsy were collected from the medical records: age, sex, duration of SLE (time from the diagnosis of SLE to the end of followup), time from the kidney biopsy to the end of followup, pregnancy morbidity as defined by the Sapporo criteria, arterial or venous thrombosis, pulmonary embolism, arthritis, malar or discoid rash, mouth ulcers, photosensitivity, serositis, Raynauds phenomenon, livedo reticularis, systemic hypertension (systolic blood pressure 140 mm Hg, diastolic blood pressure 90 mm Hg), raised serum creatinine level (1.4 mg/dl), proteinuria, nephrotic syndrome (urinary protein concentration 3 gm/24 hours), leukopenia (white blood cell count 4,000/mm3), thrombocytopenia (platelet count 100,000/mm3), autoimmune hemolytic anemia, hyperlipidemia, low C3 or C4 levels, antibodies to extractable nuclear antigens (ENA) Ro, La, Sm, and U1 nRNP, anti-DNA antibodies, LAC, and IgG and IgM aCL. Treatment with cyclophosphamide, azathioprine, hydroxychloroquine, or corticosteroids, the duration of therapy with steroids, aspirin, coumarin, antihypertensive drugs, and estrogens, and the patients use of cigarettes or alcohol were also recorded. The duration of followup for each patient was measured from the time of the kidney biopsy to the last visit or the beginning of dialysis. The frequency of patient visits was every 34 months. The end of followup was December 2002. The following parameters were recorded during this period: arterial or venous thrombosis, pregnancy morbidity, serum creatinine levels, and renal insufficiency (creatinine clearance 75 ml/minute) at the end of followup, end-stage renal disease (ESRD) requiring dialysis or renal transplantation, and death. Kidney biopsy. The renal tissues obtained by needle biopsy were fixed in Bouins fluid or in 10% neutral buffered formalin, gradually dehydrated, and embedded in paraffin.

ROLE OF aPL IN PATHOGENESIS OF APS NEPHROPATHY

2571

Table 1. Clinical and laboratory associations with APSN in patients with systemic lupus erythematosus* Patients without APSN (n 116) APS Lupus anticoagulant Anticardiolipin antibodies Livedo reticularis Hypertension Increased serum creatinine level 6 (5.2) 4 (3.4) 51 (44.0) 29 (25) 42 (36.2) 19 (16.3) Patients with APSN (n 35) 12 (34.3) 16 (45.7) 30 (85.7) 20 (57.1) 27 (77.1) 17 (48.5) Univariate analysis, OR (95% CI) 9.6 (3.328.1) 23.6 (12.843.5) 7.7 (2.821.1) 4.0 (1.88.8) 4.1 (1.610.6) 2.6 (1.06.5) Multivariate analysis, OR (95% CI) Not selected 11.46 (3.141.7) 5.7 (1.817.8) Not selected 4 (1.58.6) Not selected

* Except where indicated otherwise, values are the number (%) of patients. Odds ratios (ORs) are based on the risks of antiphospholipid syndrome nephropathy (APSN), given the absence or presence of the clinical and laboratory findings. 95% CI 95% confidence interval. P 0.01. P 0.05.

Paraffin sections (23) were stained with eosin and hematoxylin, periodic acidSchiff (PAS), silver methenamine, Massons trichrome stain, and elasticVan Gieson stain. Small portions of fresh renal tissue were snap-frozen and stored at 80C. Four-micrometer cryostat sections were incubated with fluorescein-conjugated rabbit antisera against human IgG, IgA, IgM, C1q, C3, C4, and light chains, and fibrinogen and were examined by direct immunofluorescence. All of the antisera were purchased from Dako Cytomation (Glostrup, Denmark). In the cases in which TMA was demonstrated on light microscopy, additional paraffin sections were studied for fibrinogen deposits by immunohistochemistry. After dewaxing and dehydration, paraffin sections were transferred to Tris buffered saline (TBS) and subjected to antigen retrieval in a microwave oven for 15 minutes. After rinsing in distilled water for 5 minutes, the sections were incubated with peroxidaseblocking solution (3% H2O2) for 10 minutes, rinsed in distilled water for 5 minutes, and washed in TBS for 10 minutes. Furthermore, they were incubated with polyclonal anti-human fibrinogen serum for 30 minutes at room temperature (1:1,500; Dako) and washed in TBS for 10 minutes. EnVision horseradish peroxidase (Dako) was then applied for 30 minutes, followed by a TBS bath for 10 minutes. Sections were incubated with diaminobenzidine tetrahydrochloride for 10 minutes and rinsed in distilled water for 5 minutes. PAS was used as a counterstain. Control sections were treated with nonimmune gamma globulin. Laboratory evaluation. Antinuclear and anti-DNA antibodies, ENA, C3, and C4 were assessed according to standard methods. Determination of IgG and IgM antibodies to aCL was performed with an enzyme-linked immunosorbent assay, as previously described (23). Cardiolipin (50 g/ml) in ethanol (Sigma, St. Louis, MO) was used as antigen on polystyrene microtiter plates (Nunc, Naperville, IL), which were left to dry overnight at 4C. After washing with phosphate buffered saline, the nonspecific binding sites were blocked by 10% bovine serum in phosphate buffered saline. LAC was assayed by activated thromboplastin time, kaolin clotting time, or dilute Russells viper venom time (24). The detection of LAC in medium or high titers of aCL at least 2 times, 6 weeks apart (2), before or at the time of kidney biopsy, was required. Statistical analysis. The associations between categorical variables were tested using Fishers exact test or the chi-square test when required. In addition, the odds ratios

(ORs) with the corresponding 95% confidence intervals (95% CIs) were determined using binary logistic regression. Variables that were statistically significant in univariate models were considered in multivariate regression models built on backward elimination according to the likelihood regression criteria. For continuous variables, the comparisons were carried out using the t-test for 2 independent samples or the Mann-Whitney U test for non-normal data. P values less than 0.05 were considered significant. The analysis was performed using advanced SPSS software version 11 (2001; SPSS, Chicago, IL).

RESULTS Prevalence of APS nephropathy in SLE patients with or without aPL. This series consisted of 151 white patients with SLE (132 women and 19 men). The mean (SD) age of patients at the time of kidney biopsy was 31.56 13.18 years (range 966 years), and the mean (SD) duration of followup was 7.07 3.7 years (range 115 years). The mean (SD) duration of disease before the kidney biopsy in SLE patients with aPL and those without aPL was 5.3 4.1 years and 4.47 3.78 years, respectively (P 0.24). In aPL-positive patients, positivity persisted during the entire followup period. The frequency of lupus-related manifestations (arthritis, rash, mouth ulcers, photosensitivity, serositis, Raynauds phenomenon, leukopenia, thrombocytopenia, autoimmune hemolytic anemia) did not differ between aPLpositive and aPL-negative patients. In addition, hypertension, serum creatinine levels, proteinuria, nephrotic syndrome, or renal insufficiency, and treatment with cyclophosphamide, azathioprine, hydroxychloroquine, or corticosteroids, as well as the duration of corticosteroid treatment were not statistically different between patients who were aPL positive and those who were aPL negative (P 0.05). APS nephropathy was diagnosed in 32 (39.5%) of 81 SLE patients with aPL, compared with only 3 (4.3%)

2572

TEKTONIDOU ET AL

Table 2. Demographic, clinical, and laboratory associations with APSN in patients with SLE and antiphospholipid antibodies* Characteristic Age at biopsy, mean SD years SLE duration, mean SD years Time from renal biopsy to end of followup, mean SD years Arterial thrombosis Stroke Pulmonary embolism Abortion Venous thrombosis Livedo reticularis Hypertension 1serum creatinine level Proteinuria Nephrotic syndrome Lupus anticoagulant Anticardiolipin antibodies Patients without APSN (n 49) 29.5 13.3 10.4 6.7 6.3 3.7 4 (8.2) 2 (4.1) 1 (2.0) 1 (2.0) 3 (6.1) 16 (32.7) 17 (34.7) 8 (16.3) 44 (89.8) 9 (18.4) 4 (8.2) 49 (100) Patients with APSN (n 32) 34.3 13 12.5 7.3 7.1 4.1 11 (34.4) 8 (25.0) 5 (15.6) 3 (9.4) 3 (9.4) 19 (59.4) 25 (78.1) 14 (43.8) 28 (87.5) 11 (34.4) 16 (50.0) 30 (93.8) Univariate analysis, OR (95% CI) 5.9 (1.720.7) 7.8 (1.539.8) 8.9 (180.1) 5.0 (0.55.0) 1.6 (0.38.4) 3.0 (1.27.6) 6.7 (2.418.7) 4.0 (1.411.2) 0.8 (0.23.2) 2.3 (0.86.5) 12 (3.541.5) Multivariate analysis, OR (95% CI) Not selected Not selected Not selected Not selected 6.7 (2.121.2) Not selected 8 (2.130.3)

* Except where indicated otherwise, values are the number (%) of patients. Odds ratios (ORs) are based on the risks of antiphospholipid syndrome nephropathy (APSN), given the absence or presence of the clinical and laboratory findings. SLE systemic lupus erythematosus; 95% CI 95% confidence interval. P 0.01.

of the 70 SLE patients without aPL (P 0.01). The prevalence of APS nephropathy in SLE patients with secondary APS and in SLE/non-APS patients with aPL was 67% and 31.7%, respectively. A strong association between APS nephropathy and APS, LAC, aCL, and livedo reticularis was noted (Table 1). Twenty-three of 151 patients with SLE (15.2%) had acute APS nephropathy, while 31 of 151 patients (20.5%) had chronic APS nephropathy. The mean (SD) disease duration before kidney biopsy in patients with and those without chronic APS nephropathy was 6.45 5.12 years and 4.72 4.68 years, respectively (P 0.125). The most common histologic lesions were TMA, FIH, and FCA. All of the

patients with chronic APS nephropathy had FIH. Four patients had only acute lesions. The group of SLE patients without aPL included a very small number of patients with APS nephropathy (n 3); therefore, this group was not considered in the subsequent statistical analyses. Two of these 3 patients had high IgG aCL titers during followup but only on one measurement, which was not reproduced on several subsequent examinations. Demographic, clinical, laboratory, and histologic characteristics of SLE patients with aPL and APS nephropathy. Table 2 shows the demographic, clinical, and laboratory characteristics of SLE patients with aPL

Figure 1. Multiple intraglomerular thrombi in a patient with antiphospholipid antibodies and World Health Organization class IV lupus nephritis. A, Thrombi appear yellow by silver methenamine stain. B, Immunohistochemical analysis revealed fibrinogen within thrombi (periodic acidSchiff stained). (Original magnification 100.) Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

ROLE OF aPL IN PATHOGENESIS OF APS NEPHROPATHY

2573

who were diagnosed with or without APS nephropathy at the time of kidney biopsy. The mean age, duration of lupus, and mean time from renal biopsy were not statistically different between the 2 groups. A significant association between APS nephropathy and arterial thrombosis (especially stroke), pulmonary embolism, livedo reticularis, and LAC was documented. Patients with pulmonary embolism were not included in the group with venous thrombosis, because they had no findings of deep vein thrombosis on Doppler venous ultrasonography and/or venography. No statistically significant correlation was noted between APS nephropathy and venous thrombosis, pregnancy morbidity, lupus clinical characteristics (arthritis, rash, mouth ulcers, photosensitivity, serositis, Raynauds phenomenon), and serologic findings (leukopenia, thrombocytopenia, autoimmune hemolytic anemia, anti-DNA antibodies, ENA antibodies, low C3 or C4 levels, aCL, hyperlipidemia) (P 0.05). There also was no difference between patients with and those without APS nephropathy regarding treatment with azathioprine (P 0.572), hydroxychloroquine (P 0.785), cyclophosphamide (P 0.058), aspirin (P 0.471), or corticosteroids (P 0.805), and the duration of steroid therapy (P 0.67). The mean SD duration of aspirin therapy in patients with and those without APS nephropathy was 3.0 1.4 years and 3.3 1.8 years, respectively (P 0.56). The duration of anticoagulant therapy in SLE patients with secondary APS was 7.6 4.3 years. Acute APS nephropathy was observed in 21 patients, and chronic APS nephropathy was demonstrated in 28 patients (Figures 15). Seventeen patients had both acute and chronic lesions. FIH and FCA were the most frequently observed chronic vascular lesions, detected in 28 and 17 patients, respectively. No association between APS nephropathy and the WHO classes of lupus nephritis was detected. The kidney biopsy specimens obtained from patients with APS nephropathy (n 32) were classified according to WHO criteria as follows: class II, 4 specimens (12.5%); class III, 8 specimens (25%); class IV, 9 specimens (28.1%); class V, 9 specimens (28.1%); and class VI, 2 specimens (6.9%) (2 6.1, P 0.20). The WHO activity and chronicity index scores also were not different between the 2 groups (P 0.31 and P 0.06, respectively). No association between chronic APS nephropathy and the WHO chronicity index score was observed (P 0.057). Renal prognosis of SLE patients with aPL and APS nephropathy. At the time of kidney biopsy, the patients with APS nephropathy more frequently had systemic hypertension and raised serum creatinine levels

Figure 2. Thrombotic microangiopathy in a patient with antiphospholipid antibodies and World Health Organization class Vc lupus nephritis. A, Thrombotic occlusion of a hilar arteriole accompanied by smooth muscle cell hyperplasia and occlusive changes of another arteriole (arrow) (silver methenamine stained). B, Occlusion of intraglomerular part of afferent arteriole by thrombus (arrow) (silver methenamine stained). C, The same thrombus containing fibrinogen (arrow) in immunohistochemical analysis (periodic acidSchiff stained). (Original magnification 100.) Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

than did the patients without APS nephropathy (Table 2). The mean SD duration of hypertension in patients with and those without APS nephropathy was 6.96 4.5 years and 6.25 4.5 years, respectively (P 0.62).

2574

TEKTONIDOU ET AL

Figure 3. Fibrous intimal hyperplasia (FIH) in a patient with systemic lupus erythematosus and antiphospholipid antibodies but without the antiphospholipid syndrome. A, Focal cellular FIH (Massons trichrome). B, Concentric cellular FIH of arcuate arteries (silver methenamine stained). C, FIH of an interlobular artery (elasticvan Gieson stained). D, FIH of fibrotic arterioles (elasticvan Gieson stained). (Original magnification 100.) Color figure can be viewed in the online issue, which is available at http://www.arthristisrheum.org.

Figure 4. Occlusive lesions of an interlobular artery in a patient with World Health Organization class V lupus nephritis (silver methenamine stained; original magnification 100). Color figure can be viewed in the online issue, which is available at http:// www.arthritisrheum.org.

Figure 5. Focal cortical atrophy in a patient with systemic lupus erythematosus and antiphospholipid syndrome, showing ischemic and wrinkled glomeruli surrounded by intracapsular fibrosis, tubular atrophy with thyroidization, and occlusive fibrotic arteriolar changes (silver methenamine stained; original magnification 40). Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

ROLE OF aPL IN PATHOGENESIS OF APS NEPHROPATHY

2575

Table 3. Histologic, clinical, and laboratory associations with hypertension in patients with SLE and antiphospholipid antibodies* Patients without hypertension (n 39) 5 (12.8) 13 (33.3) 3 (7.7) 3 (7.7) 12 (30.8) 4 (10.2) 4 (10.2) 27 (69.2) 8 (20.5) Patients with hypertension (n 42) 16 (38.1) 15 (35.7) 14 (33.3) 19 (45.2) 23 (54.8) 16 (38.1) 6 (14.3) 24 (57.1) 12 (28.6) Univariate analysis, OR (95% CI) 4.2 (1.412.9) 1.1 (0.71.8) 6.0 (3.011.9) 9.9 (2.637.3) 2.7 (1.16.8) 5.2 (1.617.5) 0.6 (0.22.4) 1.3 (0.44.0) Multivariate analysis, OR (95% CI) Not selected Not selected 9.6 (2.116.7) Not selected Not selected

Characteristic Acute APSN Chronic APSN FCA Increased serum creatinine level Livedo reticularis Lupus anticoagulant WHO class II IIIIV VVI

* Values are the number (%) of patients. Odds ratios (ORs) are based on the risks of hypertension given the absence or presence of the clinical, laboratory, and histologic findings. SLE systemic lupus erythematosus; 95% CI 95% confidence interval; APSN antiphospholipid syndrome nephropathy; FCA focal cortical atrophy; WHO World Health Organization. P 0.05. P 0.01.

Hypertension was found to be associated with acute APS nephropathy, livedo reticularis, and LAC, but not with the WHO class of lupus glomerulonephritis (Table 3). The mean SD serum creatinine level in patients with and those without APS nephropathy was 1.33 0.51 mg/dl and 0.98 0.45 mg/dl, respectively (P 0.024). The mean SD creatinine level in the patients with APS nephropathy who had raised serum creatinine levels at the time of kidney biopsy was 1.58 0.18 mg/dl. Elevated creatinine levels at the time of kidney biopsy were associated with chronic APS nephropathy and

LAC (Table 4). The occurrence of proteinuria or nephrotic syndrome was not different between patients with and those without APS nephropathy (Table 2). At the end of followup, no difference was detected between the 2 groups in the frequency of renal insufficiency and ESRD (OR 2.0 [95% CI 0.65.9], P 0.26 and OR 0.3 [95% CI 0.043.4], P 0.33, respectively). Renal insufficiency at the end of the followup period was strongly associated with the presence of hypertension and a raised serum creatinine level at the time of kidney biopsy (OR 6.6 [95% CI 2.312.7], P

Table 4. Histologic, clinical, and laboratory associations with raised serum creatinine levels at the time of kidney biopsy in patients with SLE and antiphospholipid antibodies* Patients without raised serum creatinine (n 59) 12 (20.3) 13 (22.0) 7 (11.9) 26 (44.1) 10 (17.0) 7 (11.9) 40 (67.8) 12 (20.3) Patients with raised serum creatinine (n 22) 9 (40.9) 15 (68.2) 10 (45.5) 9 (40.9) 10 (45.5) 3 (13.6) 11 (50.0) 8 (36.4) Univariate analysis, OR (95% CI) 2.7 (0.97.8) 4.8 (1.815.3) 6.2 (2.019.6) 0.9 (0.32.4) 4.0 (1.411.8) 0.6 (0.12.9) 1.9 (0.66.3) Multivariate analysis, OR (95% CI) Not selected 5.8 (1.719.8) Not selected

Characteristic Acute APSN Chronic APSN FCA Livedo reticularis LAC WHO class II IIIIV VVI

* Except where indicated otherwise, values are the number (%) of patients. Odds ratios (ORs) are based on the risks of raised serum creatinine given the absence or presence of the clinical, laboratory, and histologic findings. SLE systemic lupus erythematosus; 95% CI 95% confidence interval; APSN antiphospholipid syndrome nephropathy; FCA focal cortical atrophy; LAC lupus anticoagulant; WHO World Health Organization. P 0.01. P 0.05.

2576

TEKTONIDOU ET AL

Table 5. Evolution of histologic lesions on sequential kidney biopsy specimens from 11 patients with SLE and APSN* Lesion status at followup Histologic lesion Thrombotic microangiopathy Fibrous intimal hyperplasia Arterial occlusive lesions Focal cortical atrophy Interstitial fibrosis Global glomerular sclerosis Tubular atrophy Hyalinosis of arterioles Concentric (onion skinlike) fibrosis of arterioles Mural myocytic hyperplasia of arterioles None 1 2 8 3 2 0 2 4 8 11 Same 0 3 0 0 0 0 2 0 1 0 Better 10 0 0 1 1 0 0 2 0 0 Worse 0 0 0 4 8 8 7 2 0 0 New lesions 0 6 3 3 0 3 0 3 2 3

* Values are the number of patients. The mean SD time interval between the first and the last biopsy was 6.3 3.4 years. SLE systemic lupus erythematosus; APSN antiphospholipid syndrome nephropathy. No lesions detectable on either the initial or sequential kidney biopsy specimens. Lesions that were not detected on the initial kidney biopsy specimen.

0.009 and OR 8.7 [95% CI 2.718.6], P 0.001, respectively). Four patients with APS nephropathy and 3 without APS nephropathy died. Evolution of histologic lesions in SLE patients with aPL and APS nephropathy. During the followup period, 11 of 32 patients with APS nephropathy had serial kidney biopsies performed (8 patients had 2 procedures and 3 patients had 3 procedures). The mean SD time between the first biopsy and the last biopsy was 6.3 3.4 years (range 212 years). Table 5 shows the number of patients in whom the status of lesions was the same, better, or worse at followup, as well as the number of patients with new histologic lesions on sequential kidney biopsy specimens and the number of patients with no detectable lesions on either the initial or sequential specimens. Clinical evolution to APS in SLE/non-APS patients with aPL. During the followup period, thrombosis developed in 7 (35%) of 20 patients with APS nephropathy, compared with 4 (9%) of 43 patients without APS nephropathy (P 0.04). The patients with APS nephropathy had thrombosis at the arterial level (4 had stroke, 2 had pulmonary embolism, and 1 had acute myocardial infarction), although deep vein thrombosis developed in the patients without APS nephropathy. Among patients with APS nephropathy, no difference was found between those with thrombosis (n 7) and those without thrombosis (n 13) regarding the presence of additional risk factors for thrombosis such as nephrotic syndrome, hypertension, chronic renal failure, hyperlipidemia, treatment with estrogens or corticosteroids, and the duration of corticosteroid treatment (P 0.05). None of these 7 patients with thrombosis

smoked cigarettes or abused alcohol. Four of the 7 patients were receiving aspirin treatment at the time of thrombosis, but none was being treated with coumarin. During followup, thrombosis did not develop in any of the SLE patients without aPL. DISCUSSION In the present study, we examined the prevalence and the long-term outcome of APS nephropathy in 151 SLE patients with or without aPL. This report is the first to address the histologic evolution of APS nephropathy lesions on serial kidney biopsy specimens and the clinical evolution to APS in SLE/non-APS patients with APS nephropathy. APS nephropathy was documented independently of lupus glomerulonephritis in approximately two-thirds of SLE patients with secondary APS and one-third of SLE/non-APS patients with aPL but in only 3 of 70 SLE patients without aPL. APS nephropathy was strongly associated with both aCL and LAC, suggesting that aPL play a direct role in the development of APS nephropathy. In this study, the sample size was quite small; therefore, the results of the multivariate analyses should be interpreted with caution. Unavoidably, our study could not have had enough power to detect modest differences in clinical, histologic, and laboratory findings between the groups, yet we were able to determine strong associations with large ORs for specific clinical, laboratory, and histologic parameters. The most frequently observed pathologic lesions were TMA, FIH, and FCA. TMA is characterized by fibrin thrombi in the glomeruli and/or small arterioles in

ROLE OF aPL IN PATHOGENESIS OF APS NEPHROPATHY

2577

the absence of vascular immune deposits or inflammatory cells. The appearance of TMA lesions is similar to the changes observed in thrombotic thrombocytopenic purpura and the hemolytic uremic syndrome, entities occasionally occurring in the context of SLE (25,26). When making a diagnosis of APS nephropathy, the clinical context must be considered rather than relying only on biopsy criteria. Patients with SLE who had features of the above-mentioned disorders were excluded from the current study. In SLE, TMA has been associated with the presence of LAC and/or aCL but not with the WHO class of lupus nephritis. In several studies, it has also been associated with a worse renal prognosis (13,17,26). TMA was the main renal vascular lesion observed in sporadic cases of primary APS. DAgati et al reported the presence of TMA in the kidney biopsy specimens obtained from 2 patients with primary APS and 1 patient with SLE (5). Kincaid-Smith and coworkers (11) reported the occurrence of TMA and FIH in 12 women with aPL and postpartum renal failure (8 with primary APS, 4 with SLE). Amigo et al documented the presence of TMA, FIH, and glomeruli ischemia in 5 patients with primary APS and renal involvement (6). In 1999, a group of French investigators described for the first time the presence of APS nephropathy in patients with primary APS (18). In 2002 the same group described the presence of APS nephropathy in patients with SLE, especially those with secondary APS (19). The latter elegant study examined 24 SLE patients with APS, 52 SLE/non-APS patients with aPL, and a small number of patients without aPL (n 20). APS nephropathy was observed in 63%, 22%, and 15% of these patients, respectively. An association of APS nephropathy was noted with APS and LAC but not with aPL. The discrepancy between these results and ours, which showed a strong correlation of APS nephropathy with both LAC and aCL, may be explained by the fact that APS was associated with LAC but not with aCL in the above-mentioned series. In our study, only 2 of 32 aPL-positive patients with APS nephropathy were negative for aCL. Concerning the renal prognosis of the patients with APS nephropathy, a high prevalence of hypertension was found, which is consistent with previous reports (5,11,14,16,18,19,26). Regarding the possibility that features of APS nephropathy are secondary to hypertension, it could be argued that the histologic lesions described above developed in several patients without hypertension. In addition, hypertension was found to be associated with acute and chronic APS nephropathy,

livedo reticularis, and LAC, but not with the WHO class of lupus nephritis. An association of APS nephropathy with raised serum creatinine levels but not with proteinuria, nephrotic syndrome, renal insufficiency, or ESRD was also documented. In previous studies, the prevalence of proteinuria, nephrotic syndrome, chronic renal failure, or elevated serum creatinine levels in patients with SLE and/or APS and renal vascular lesions was variable (1719,2628). Given that progression of renal histologic findings in patients with SLE has been correlated with a poor renal prognosis, the evolution of APS nephropathy findings on repeated kidney biopsy specimens was examined. A progression of the acute thrombotic lesions to chronic proliferative, obstructive, and fibrotic forms was observed. The presence of TMA in the first biopsy specimen was usually followed by the appearance of chronic lesions such as FIH, arteriolar occlusions, FCA, or sclerotic lesions in the second or third biopsy specimen. Daugas et al reported that the higher age of patients with chronic APS nephropathy suggests that the chronic lesions develop later, probably originating as acute APS nephropathy (19). Kincaid-Smith et al reported biopsy specimens showing fibrin thrombi in glomerular arterioles and interlobular arteries at the time of acute renal episodes, and ischemic glomeruli with double contours, arteries narrowed by recanalizing thrombi, and cellular or fibroelastic intimal proliferation in specimens obtained at repeated biopsies months or years later (11). In the literature, the presence of hypertension and elevated creatinine levels at the time of kidney biopsy is associated with a worse renal outcome. In our series, however, no difference was observed in the frequency of renal insufficiency or ESRD at the end of followup between patients with and those without APS nephropathy. This discrepancy could be explained by the fact that patients with APS nephropathy had mild renal insufficiency at the time of kidney biopsy (the mean SD creatinine level in patients with raised serum creatinine levels at the time of kidney biopsy was 1.58 0.18 mg/dl), which was probably reversible with the immunosuppressive and/or anticoagulant treatment. Twelve (37.5%) of 32 patients with APS nephropathy had secondary APS; these patients had been receiving anticoagulant therapy continuously from the time of kidney biopsy. Anticoagulant therapy could influence the renal outcome. The fact that only 1 of 11 patients with progression of histologic lesions on sequential kidney biopsy specimens had secondary APS (this patient had been receiving anticoagulation treatment) may support

2578

TEKTONIDOU ET AL

the above hypothesis. Unfortunately, the small number of patients prevents drawing any definite conclusion regarding the potential role of anticoagulant therapy on renal prognosis. Better-powered prospective studies are warranted to examine this issue. A strong association between APS nephropathy and APS-related manifestations such as arterial thromboses and livedo reticularis was found in this series. It is well known that the manifestations of APS are characterized by recurrent episodes (29,30) in the absence of therapy. Hypothesizing that APS nephropathy may represent an additional feature of APS, we investigated any tendency of the SLE/non-APS patients with APS nephropathy to develop APS manifestations during a mean followup period of 7 years. It was observed that among SLE/non-APS patients with aPL, those with APS nephropathy developed thromboses more frequently than did those without APS nephropathy (P 0.04). Moreover, patients who had this characteristic renal smallartery vasculopathy developed arterial thromboses, while those without APS nephropathy developed venous thromboses. Previous studies have demonstrated that the arterial thromboses in APS tend to be followed by more arterial thromboses, and that the venous thromboses tend to be followed by more venous thromboses (3032). Thus, the strong association of APS nephropathy with aPL and the tendency of the patients with APS nephropathy to develop thrombotic events (especially at the same vascular level) suggests that APS nephropathy should be considered as a manifestation of APS, and the role of appropriate anticoagulant treatment (32,33) should be tested. In conclusion, APS nephropathy occurred in 39.5% of SLE patients with aPL but in only isolated SLE patients without aPL, suggesting an important role of aPL in the pathogenesis of APS nephropathy. Among SLE patients with aPL, APS nephropathy can be detected in two-thirds of those with APS and in one-third of those without APS, independently of lupus nephritis. Arterial thromboses (especially stroke), pulmonary embolism, livedo reticularis, aCL, and LAC are strongly associated with APS nephropathy. Patients with APS nephropathy develop hypertension, raised serum creatinine levels, and progression of histologic lesions in serial kidney biopsy specimens; these features are associated with a worse renal prognosis. In patients with APS nephropathy, manifestations of APS, especially arterial thromboses, also tend to develop during the course of the disease. Renal pathologists should carefully examine for the presence of APS nephropathy on kidney biopsy specimens obtained from SLE patients with aPL. Clini-

cians should be aware of this histologic entity in order to propose a rationale for treatment and to maintain close clinical and laboratory followup. We suggest that APS nephropathy should be included in the APS classification criteria, and that the use of anticoagulant therapy should be examined. REFERENCES
1. Hughes GR. The antiphospholipid syndrome: ten years on. Lancet 1993;341:3424. 2. Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999;42: 130911. 3. Asherson RA, Kant KS. A primary antiphospholipid syndrome? J Rheumatol 1988;15:17426. 4. McNeil HP, Hunt JE, Krilis SA. Antiphospholipid antibodies: new insights into their specificity and clinical importance. Scand J Immunol 1992;36:64752. 5. DAgati V, Kunis C, Williams G, Appel GB. Anti-cardiolipin antibody and renal disease: a report of three cases. J Am Soc Nephrol 1990;1:77784. 6. Amigo MC, Garcia-Torres R, Robles M, Biochicchio T, Reyes PA. Renal involvement in primary antiphospholipid syndrome. J Rheumatol 1992;19:11815. 7. Piette JC, Cacoub P, Wechsler B. Renal manifestations of the antiphospholipid syndrome. Semin Arthritis Rheum 1994;23: 35766. 8. Domrongkitchaiporn S, Cameron EC, Jetha N, Kassen BO, Sutton RA. Renal microangiopathy in the primary antiphospholipid syndrome: a case report with literature review. Nephron 1994;68: 12832. 9. Becquemont L, Thervent E, Rondeau E, Lacave R, Mougenot B, Sraer JD. Systemic and renal fibrinolytic activity in a patient with anticardiolipin syndrome and renal thrombotic microangiopathy. Am J Nephrol 1990;10:2548. 10. Kleinknecht D, Bobrie G, Meyer O, Noel LH, Callard P, Ramdane M. Recurrent thrombosis and renal vascular disease in patients with a lupus anticoagulant. Nephrol Dial Transplant 1989;4:8548. 11. Kincaid-Smith P, Fairley KF, Kloss M. Lupus anticoagulant associated with renal thrombotic microangiopathy and pregnancyrelated renal failure. Q J Med 1988;68:795815. 12. Glueck HI, Kant KS, Weiss MA, Pollak VE, Miller MA, Coots M. Thrombosis in systemic lupus erythematosus: relation to the presence of circulating anticoagulants. Arch Intern Med 1985;145: 138995. 13. Kant KS, Pollak VE, Weiss MA, Glueck HI, Miller AN, Hess EV. Glomerular thrombosis in systemic lupus erythematosus: prevalence and significance. Medicine (Baltimore) 1981;60:7186. 14. Farrugia E, Torres VE, Gastineau D, Michet CJ, Holley KE. Lupus anticoagulant in systemic lupus erythematosus: a clinical and renal pathological study. Am J Kidney Dis 1992;20:46371. 15. Frampton G, Hicks J, Cameron JS. Significance of antiphospholipid antibodies in patients with lupus nephritis. Kidney Int 1991;39:122531. 16. Hughson MD, Nadasdy T, McCarty GA, Sholer C, Min KW, Silva F. Renal thrombotic microangiopathy in patients with systemic lupus erythematosus and the antiphospholipid syndrome. Am J Kidney Dis 1992;20:1508. 17. Bridoux F, Vrtovsnik F, Noel C, Saunier P, Mougenot B, Lemaitre V, et al. Renal thrombotic microangiopathy in systemic lupus

ROLE OF aPL IN PATHOGENESIS OF APS NEPHROPATHY

2579

18. 19. 20. 21. 22. 23.

24.

25.

erythematosus: clinical correlations and long-term renal survival. Nephrol Dial Transplant 1998;13:298304. Nochy D, Daugas E, Droz D, Beaufils H, Grunfeld JP, Piette JC, et al. The intrarenal vascular lesions associated with primary antiphospholipid syndrome. J Am Soc Nephrol 1999;10:50718. Daugas E, Nochy D, Huong du LT, Duhaut P, Beaufils H, Caudwell V, et al. Antiphospholipid syndrome nephropathy in systemic lupus erythematosus. J Am Soc Nephrol 2002;13:4252. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:12717. Churg J, Sobin LH. Renal disease. In: Churg J, Sobin LH, editors. Classification and atlas of glomerular disease. 1st ed. Tokyo-New York: Igaku-Shoin; 1982. p. 12749. Austin HA III, Muenz LR, Joyce KM, Antonovych TA, Kullick ME, Klippel JH, et al. Prognostic factors in lupus nephritis: contribution of renal histologic data. Am J Med 1983;75:38291. Manoussakis MN, Gharavi AE, Drosos AA, Kitridou RC, Moutsopoulos HM. Anticardiolipin antibodies in unselected autoimmune rheumatic disease patients. Clin Immunol Immunopathol 1987;44:297307. Brandt JT, Triplett DA, Alving B, Scharrer I, on behalf of the Subcommittee on Lupus Anticoagulant/Antiphospholipid antibody of the Scientific and Standardisation Committee of the ISTH. Criteria for the diagnosis of lupus anticoagulant: an update. Thromb Haemost 1995;74:118590. Cecere FA, Yoshinoya S, Pope RM. Fatal thrombotic thrombocytopenic purpura in a patient with systemic lupus erythematosus: relationship to circulating immune complexes. Arthritis Rheum 1981;24:5503.

26. Banfi G, Bertani T, Boeri V, Faraggiana T, Mazzucco G, Monga G, et al. Renal vascular lesions as a marker for poor prognosis in patients with lupus nephritis. Am J Kidney Dis 1991;18:2408. 27. Hughson MD, McCarty GA, Brumback RA. Spectrum of vascular pathology affecting patients with the antiphospholipid syndrome. Hum Pathol 1995;26:71624. 28. Magil AB, McFadden D, Rae A. Lupus glomerulonephritis with thrombotic microangiopathy. Hum Pathol 1986;17:1924. 29. Rosove MH, Brewer MC. Antiphospholipid thrombosis: clinical course after the first thrombotic event in 70 patients. Ann Intern Med 1992;117:3038. 30. Finazzi G, Brancaccio V, Moia M, Ciaverella N, Mazzucconi MG, Schinco PC, et al. Natural history and risk factors for thrombosis in 360 patients with antiphospholipid antibodies: a four year prospective study from the Italian registry. Am J Med 1996;100: 5306. 31. Tektonidou MG, Ioannidis JP, Boki KA, Vlachoyiannopoulos PG, Moutsopoulos HM. Prognostic factors and clustering of serious clinical outcomes in antiphospholipid syndrome. Q J Med 2000; 93:52330. 32. Ruiz-Irastorza G, Khamashta MA, Hunt BJ, Escudero A, Cuadrado MJ, Hughes GR. Bleeding and recurrent thrombosis in definite antiphospholipid syndrome: analysis of a series of 66 patients treated with oral anticoagulation to a target international normalized ratio of 3.5. Arch Intern Med 2002;162:11649. 33. Khamashta MA, Cuadrado MJ, Mujic F, Taub NA, Hunt BJ, Hughes GR. The management of thrombosis in the antiphospholipid syndrome. N Engl J Med 1995;332:9937.