Thrombotic microangiopathy (TMA) is a heterogeneous disorder characterized by platelet thrombi in arterioles and capillaries and on occasion in arteries.1,2 Renal histopathologic lesions in TMA tend to take one of two broad forms with considerable overlap: (1) predominant arteriolar, and lesser arterial, involvement, with thrombi and fibrinoid necrosis, particularly in thrombotic thrombocytopenic purpura, malignant hypertension (MHT), and scleroderma; or (2) glomerular involvement, with capillary thrombi, capillary loops with double contours due to mesangial interposition, and variable mesangiolysis,3 the latter most frequently seen in the hemolytic–uremic syndromes. These morphologic lesions occur in a number of other clinical settings as well, including anti-phospholipid antibody syndrome, or as a side effect of various pharmacologic agents, and are often associated with poor renal prognosis.1–3
In immunoglobulin A nephropathy (IgAN), the most common form of primary glomerular disease worldwide,4 it has long been recognized that intrarenal arterial and arteriolar lesions, such as arteriolar wall thickening and hyalin changes, may be a prominent feature.5,6 Further, TMA has been described in IgAN in a recent study7 and attributed by the authors to severe or malignant hypertension. However, a large-scale clinicopathologic analysis focused on TMA in IgAN has not been performed. This report describes the prevalence, associated clinical features, and outcome of histologic TMA lesions found in a retrospective survey of IgAN.
This study included 128 patients, with males predominating (69.5%). Among them, 118 (92.2%) were Caucasians, and 10 (7.8%) were Asians. Mean age was 38.7 years (range, 18–78 years). Mean proteinuria was 2.47 g/d (25th to 75th percentile: 0.8–3.00 g/d), and mean estimated GFR (eGFR) was 51.2 ml/min per 1.73 m2 (25th to 75th percentile, 29–76 ml/min per 1.73 m2). All patients except one (who was pregnant and presented without TMA) received angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists, or both, in case of hypertension or persistent proteinuria. Only one patient, in the non-TMA group, had received corticosteroid therapy prior to diagnosis, and none had steroid therapy subsequent to diagnosis. No patient had other immunosuppressive therapy, either prior to or subsequent to diagnosis.
Mean follow-up was 44 months (25th to 75th percentile, 23–60 months); for those who went to ESRD, mean time from diagnosis to dialysis was 15 months (25th to 75th percentile, 1–29 months).
TMA Is a Common Feature of IgAN
Among our patients, 68 (53.1%) presented with acute or organized TMA lesions. There were no significant differences in age or sex between the patients with and without TMA. Clinical and biological characteristics of TMA patients are summarized in Table 1. Hypertension was present in 71.0% and 23.3% of patients in the TMA and the non-TMA groups, respectively (P=0.00). MHT was noted in 26% of patients in the TMA group; no patient in the non-TMA group presented with MHT. Neurologic symptoms were absent (except in patients with MHT). Compared with patients in the non-TMA group, patients with TMA had significantly higher proteinuria, lower serum albumin, higher serum creatinine, and lower eGFR at the time of the biopsy (Table 1). No possible cause for TMA (such as radiotherapy, Shiga toxin–producing bacteria infection, or drug-induced TMA) was documented in any patient. Among the 52 patients tested, there was no difference between the groups for the presence of anti-cardiolipin antibody (26.5% versus 28.5%, P<0.10) or its titer when present. Only 1 of the 39 patients tested had lupus anticoagulant (although that patient indeed had TMA). Similarly, of the 47 patients tested, only 3 patients (2 with TMA and 1 without TMA) had anti-β2 glycoprotein (anti-β2GP1) antibody. It is evident, then, that none of these factors plays a major role in TMA. Eleven (8%) patients with TMA had complete complement assays and genetic screening for complement regulatory protein gene mutations; none presented such mutations.
Notably, 20 patients presented with TMA lesions (including acute lesions) either without associated hypertension or normotensive under treatment (Table 2). Of note, most (73.9%) patients from the TMA group did not have MHT at the time of biopsy or in their medical history.
Comparisons of Patients According to the Degree of Hypertension
Comparisons were made between completely normotensive patients, patients normotensive under treatment, hypertensive patients, and those with MHT; the clinical data and the morphologic parameters are presented in Table 2. Among the 63 normotensive patients, 44 (69.8%) were treated with one or more antihypertensive agents. MHT was found in 18 (14.1%) patients who, compared with patients with less severe hypertension, presented with much more advanced renal insufficiency and with much lower eGFR, 58% of them requiring renal replacement therapy from the outset compared with 7% with lesser hypertension (P=0.00; Table 2). They also had greater proteinuria (Table 2). Importantly, there was no difference in the frequency of anti-cardiolipin antibodies between the four groups of patients. As might be anticipated, MHT biopsies disclosed greater interstitial fibrosis, greater percentage of sclerotic glomeruli, worse glomerular extracapillary proliferation, and more frequent TMA than biopsies of hypertensive patients without MHT (Supplemental Table 1). All the MHT patients (100%) presented with TMA lesions versus 65.9% of hypertensive patients (without MHT) and 31.7% of normotensive patients overall (15.8% of entirely normotensive patients and 38.6% of patients normotensive on antihypertensive therapy) (P=0.004 and P=0.0004, respectively; Table 2).
Histologic Findings in Patients with or without IgAN-Associated TMA
TMA was nearly exclusively arterial and arteriolar in location (Figure 1). Only two cases (one in the original series and one in the supplemental cases stained for CD61 [see later]) had glomerular fibrin thrombi. There was no evidence in any case for glomerular capillary endothelial swelling, double contours, or mesangiolysis. The fresh fibrinous vascular thrombi (Figures 2–4) were characterized by the presence of fibrinous material (staining bright reddish on trichrome stain as performed in our laboratory using acetic acid–formol–absolute alcohol (AFA) fixative, as opposed to the blue staining of the hyalin deposits of hyalin arteriolosclerosis) and dilation with marked distension and smoothing out of the internal elastic lamina (Figures 1 and 2). Chronic lesions were basically organized thrombi with small recanalized vascular channels and reduction or obliteration of the lumen (Figures 1 and 5–7), sometimes having an “onion-skin” appearance. The organized fibrous tissue was generally oriented in the long axis of the lumen but lacked the lamellar quality of the fibroelastotic lesions of arteriosclerosis. Focal myocyte necrosis was seen, usually in association with thrombi but sometimes separately (Figure 8).
IgAN-associated TMA was associated with more severe other vascular lesions, both in terms of reduction of lumen (arterial intimal sclerosis and arteriolar lumen reduction) and smooth muscle hypertrophy (Supplemental Figures 1–6). These differences between TMA and non-TMA cases were maintained when patients were divided into normotensive and hypertensive groups, although not all differences remained significant (Table 3) (patients with MHT were not included in this comparison because all had TMA). Consistent with these changes, hyperplasia of the juxtaglomerular apparatus was more frequent in patients with TMA (P=0.04).
In general, the biopsies with IgAN-associated TMA showed more extensive damage in terms of percentage of sclerotic glomeruli and tubulointerstitial damage (Supplemental Table 2).
The ensemble of cases was also evaluated in terms of the Oxford Classification (Supplemental Table 2). As anticipated, all of the parameters were more frequent/worse among the patients with TMA than among those without.
Staining using anti-CD61, an antiplatelet antibody, was performed for 12 recent cases of IgAN not included in the earlier main series reported here. All had evidence of either acute and/or organized TMA on routine Masson stain. Of these, 10 showed at least focal positivity on staining for CD61.
Arteries and Arterioles
In acute lesions, although sometimes platelet-rich thrombi completely filled the lumen (Figure 9A), typically platelets were present in fewer numbers, admixed in varying degrees with other elements (Figure 9B and Supplemental Figures 7 and 8), and might be present in one section of the lumen and absent in an adjacent one (Supplemental Figure 9). There frequently was staining for platelets in the media of arteries with acute lesions (Figure 9B and Supplemental Figure 8). Platelets progressively disappeared from the intima and media as lesions advanced (Figure 9C) and were generally entirely absent in organized TMA (Figure 9D).
One of the 12 cases had glomerular thrombi, recognizable on CD61 staining (Figure 9E). Another case showed platelets at the site of a presumptive area of fibrinoid necrosis (Figure 9G). The corresponding glomerulus was not identifiable on the initial Masson stain, but another glomerulus from the same case showed clear fibrinoid necrosis (Figure 9H). In addition, several cases had isolated platelets or platelet aggregates in glomerular capillary lumens in a minority of glomeruli (Figure 9F), but these lesions were not recognizable by routine microscopy.
Veins and Peritubular Capillaries
CD61 staining permitted detection of rare capillary and venous lesions unapparent on routine Masson stain. Some of these represented definite venous thrombi (Supplemental Figure 10). Others may simply represent platelet aggregates (Supplemental Figures 11 and 12).
TMA Associated with IgAN May Occur in Early and/or Mild Cases
Because TMA in IgAN has previously been reported predominantly in patients with MHT,7 it is important to point out that TMA may occur in early/mild cases: 33% (23 of 69 cases) with systolic blood pressures ≤140 mmHg; 52% in patients with diastolic pressures ≤90 mmHg; 19% with serum creatinine (SCr) ≤120 μmol/L; 16% with eGFR >60 ml/min per 1.73 m2. In morphologic terms, 16 (23.2%) of the cases of TMA occurred in patients with minimal to mild interstitial fibrosis/tubular atrophy (Oxford class 0),8,9 with 4 (5.8%) cases showing only acute TMA, 5 (7.2%) cases only organized TMA, and 7 (10.1%) cases showing both.
Conversely, however, IgAN-associated TMA rarely occurred in the absence of significant proteinuria, only 4.6% of cases having <0.5 g/24 h versus 34.5% of cases without TMA (P=0.001). Similarly, TMA was tightly associated with the presence of glomerular lesions, only two (2.9%) cases having entirely normal glomeruli by light microscopy versus 14 (24.1%) cases among the non-TMA cases (P=0.0004). However, glomerular lesions, although present in 97% of TMA biopsies, were not necessarily severe in a given biopsy.
TMA Is Associated with Bad Outcome
Table 4 presents a univariate analysis of the various clinical and vascular parameters relatable to IgAN-associated TMA with bad outcome. Among the vascular parameters, all showed significant associations with bad outcome except hyalin arteriolar deposits. As anticipated, both fibrinoid and organized TMA were strongly associated with bad outcome.
The contribution of TMA (compared with the Oxford criteria) to decline of eGFR and to bad outcome was analyzed by multiple linear regression and by Cox proportional hazards modeling, respectively (Table 5). Potential confounding factors such as mean arterial pressure, SCr, and proteinuria at the diagnosis were included. When this was done, the eGFR at diagnosis sorted as significant. Similarly, laboratory evidence for TMA sorted as significantly associated with decline of eGFR (P=0.001). However, the simple morphologic presence of TMA did not sort as significant.
Renal survival was 52.2% at 44 months among the TMA patients versus 93.5% among those without TMA (P=0.00001). However, a more telling separation comes from dividing the cases with morphologic lesions of TMA only compared with those TMA patients who had, in addition, laboratory evidence of TMA. All eight of the latter patients had a bad outcome within 6 months of presentation, with a highly significant difference between this group and those with morphologic lesions only (P=0.0002; Figure 10).
TMA was a frequently identified lesion in this study of IgAN in adults, being found in slightly more than one-half (53.1%) of our patients. This high frequency is in part attributable to the fact that our patients as a group had rather advanced disease.
A very high percentage of patients were either frankly hypertensive (48.4%) or normotensive on antihypertensive treatment (34.4%), with 18 (14%) patients presenting with MHT. This frequency of hypertension is substantially higher than that in other series8,9 and is attributable to an active hypertension clinic in our institution from which many patients were drawn. This biased recruitment of patients accounts in large part for the much poorer survivals (80% of MHT patients went to ESRD). Because our data reveal that IgAN-associated TMA increases markedly in frequency with increasing hypertension (Table 2), this accounts in large part for the very high incidence of TMA in our series.
Even taking the increased severity in our patients into account, however, it is evident that the incidence of TMA in IgAN generally is substantially higher than has previously been appreciated, as numerous examples occurred in patients who were either entirely normotensive or normotensive under therapy with normal/near-normal renal function. (In addition, we believe that our use of AFA fixative and the trichrome stain facilitates the search for these lesions, which may be inconspicuous on other stains.)
The only other study looking specifically at TMA in IgAN7 found MHT in 6 of 10 patients studied, with severe hypertension in another 3 patients, and favored the hypothesis that the TMA was the consequence of the MHT, the MHT itself being the consequence of advanced parenchymal lesions. This was a plausible theory for patient sample of that study, particularly given that MHT occurs in 7%–15% of IgAN.7,10,11 The association between MHT and TMA is well recognized, both in spontaneous12,13 and drug-induced MHT,13,14 the assumption being that the TMA is due to pressure-induced endothelial disruption.13 (In support of the pressure-induced mechanism for TMA, TMA has only been described in severe/malignant hypertension, not in mild to moderate essential hypertension.) However, our series essentially refutes the hypothesis that the TMA in IgAN is due to MHT. The frequency of TMA did indeed increase markedly in frequency with increasing blood pressure, leading to the conclusion that increasing blood pressure is a major aggravating factor. But it seems unlikely to be the sole cause of IgAN-associated TMA, as 20 of 69 (29%) cases occurred in patients with systolic pressures <140 mmHg at the time of biopsy, levels at which TMA has not been described in essential hypertension.
Nor did IgAN-associated TMA necessarily develop in a setting of advanced parenchymal lesions, 19% occurring in patients with an SCr <120 µmol/L and 23.9% occurring in patients with minimal to mild (Oxford class 0) interstitial fibrosis/tubular atrophy. It thus appears clear that TMA can precede the development of glomerulosclerosis and interstitial fibrosis rather than being a consequence of it, a sequence that has been suggested by others for vascular lesions in general in IgAN.5 Thus, neither hypertension nor advanced parenchymal lesions are necessary prerequisites to the development of TMA.
By contrast, the appearance of TMA in the biopsy did appear to be tightly linked to the presence of glomerular lesions and proteinuria. Only three (4.6%) TMA cases had proteinuria <0.5 g/24 h as opposed to 19 (34.5%) cases without TMA. Further, the frequency of TMA increased with increasing proteinuria, from 13.6% for cases <0.5 g/24 h to 80% for cases >3.0 g/24 h (P=0.00). Similar considerations held for the association of TMA with overt glomerular lesions. Only two (2.9%) TMA biopsies had normal glomeruli by light microscopy compared with 14 (24.1%) biopsies without TMA (P=0.0004). Any explanation of the mechanism(s) of TMA in IgAN must take its association with glomerular lesions and proteinuria into account.
Although CD61 staining revealed glomerular lesions to be slightly more extensive than appreciated on routine microscopy, IgAN-associated TMA remains a primarily arterial/arteriolar lesion. In this regard, it resembles scleroderma,15,16 and particularly, the kidney of malignant hypertension. A recent report of 21 patients with MHT-associated TMA found arterial or arteriolar lesions in all but glomerular thromboses in none, with only 6 patients showing laboratory evidence for TMA.17
The question remained whether some of the lesions thought to represent TMA on routine microscopy might instead represent simply banal hyalin arteriolosclerosis rather than fibrinoid material, despite the marked differences in staining on Masson stain as performed in our laboratory—blue for the former, bright red for the latter. Staining for anti-CD61, an antiplatelet antibody, largely put this question to rest, revealing that the acute lesions were extensively positive for platelets, although staining varied from artery to artery (Figure 9, A and B and Supplemental Figures 7–9). The platelet staining here corresponds in large part with that seen in other situations.18
Other arterial and arteriolar lesions of IgAN have been reported as being associated with other clinical and histologic poor prognostic factors5,19,20 and even potentially independently associated with the degradation of renal function.21 In our study, both fibrinoid and organized TMA, as well as other vascular lesions, particularly arteriolar lumen size, were significant on univariate analysis (Table 4).
However, multiple linear regression of rate of decline of eGFR and Cox proportional hazards modeling of outcome both show similar results (Table 5). Laboratory evidence of TMA sorts as a significant factor in eGFR decline and bad outcome in both models, but simple morphologic TMA does not. We have in effect a “tip of the iceberg” effect, with all eight patients with thrombotic tendencies severe enough to lead to laboratory manifestations going on to bad outcome. Those with only morphologic TMA, the “underwater” part, nonetheless had a substantially greater frequency of bad outcome (42.1% versus 11.3%, P=0.0004). The only other study looking at IgAN-associated TMA found that all of the patients for whom follow-up data were available evolved to terminal renal insufficiency within a year of diagnosis of TMA.7
Notably, in our study, TMA appears to be associated with worse lesions of arteriosclerosis, particularly striking in the normotensive patients, where possible confounding effects of hypertension can be excluded from consideration (Table 3). Although the evaluation of the arterial/arteriolar lesions was simply a semiquantitative estimate, the differences in Table 3 are sufficiently great that they seem likely to reflect a real link between TMA and vascular sclerosis. Obviously, however, extensive morphometric studies will be required to confirm this result.
The causes of TMA in IgAN are uncertain. Certain statements can be made from our analysis. First, although TMA clearly increases markedly in frequency with increasing blood pressure (Table 2), it may appear early, in situations ruling out both severe hypertension or advanced parenchymal damage and renal insufficiency as necessary to its development. However, glomerular lesions and proteinuria are integral elements of the setting in which TMA develops.
Anti-phospholipid syndrome antibodies (anti-cardiolipin, anti-β2GP1 antibodies, or lupus anticoagulant) have also been described in IgAN.22,23 However, in our series, these antibodies were present in a minority of cases, and there was no significant difference in frequency of TMA between cases with and without these antibodies. Thus, they clearly do not play a role in causation in the majority of cases.
Mutations of complement factor H (CFH) and complement factor I (CFI) and membrane cofactor protein genes have been associated with TMA and kidney involvement.2 No genetic abnormalities were identified in 11 patients from our series chosen for their severe TMA. This mitigates against the possible influence of the regulation of alternative pathway in this disease. In addition, recently, Edey et al.24 have reported the absence of mutations of CFH in a large series of patients with IgAN.
Other possible mechanisms for TMA exist that our study cannot address. First is possible alteration and/or diminution of function of vascular endothelial growth factor (VEGF).25,26 Inhibitors of VEGF, are known to lead to proteinuria regularly and less frequently to TMA.27–29 It is known that aberrantly glycosylated IgA downregulates synthesis of VEGF in mesangial cells,30 and there is diminution of podocyte staining for VEGF in IgAN.31 Anti-endothelial cell antibodies are another possibility to be considered. A study from the 1980s found a 32% incidence of anti–endothelial cell antibodies in IgAN compared with 4% in controls and 9% in other glomerular diseases. Little had been done since in this area until a recent study32 found anti–endothelial cell antibodies in 34 of 75 (45.3%) patients with IgAN (24 of the 34 having MHT) compared with 3 of 19 patients with primary MHT (P=0.02).
This study has several limitations. It is retrospective and observational and will need to be validated with a prospective cohort of patients. Further, the evaluation of prognosis was rendered difficult by the variable nature of the treatment received.
In conclusion, we have shown in this study that the lesions of TMA are frequent and severe in IgAN and have a poor prognosis. They increase in frequency with both increasing blood pressure and proteinuria. Lesions of TMA are particularly associated with MHT, but their frequent presence in patients who are normotensive either naturally or under antihypertensive therapy indicates that they are not the result of the MHT. The causative factors responsible for this TMA remain to be determined. We believe that these lesions should be systematically sought on renal biopsy, so that the TMA may be addressed therapeutically, with the future goal being to optimize treatment for this lesion when it occurs in IgAN.
All the adult (>18 years) patients diagnosed with IgAN from January 2002 to January 2008 at the Pathology Department of the Hôpital Européen Georges Pompidou (Paris, France) were enrolled in this study. These biopsies came from four different medical centers. The diagnosis was based on the presence of predominant IgA and C3 deposits in the mesangium. Patients with SLE, Henoch–Schönlein purpura, chronic liver disease, or HIV infection were excluded, as well as patients whose renal biopsy specimen contained less than eight glomeruli. Clinical and laboratory data including age, gender, blood pressure, number of antihypertensive agents used, immunosuppressive therapy, proteinuria, hematuria, familial history of IgAN, SCr, and presence of anti-cardiolipin antibody and lupus anticoagulant (defined by spontaneously prolonged activated partial thromboplastin time and abnormal specific lupus anticoagulant test). Anti-β2GP1 antibodies were collected at the time of renal biopsy and at the end of follow-up (or institution of renal replacement therapy). Seven patients were lost to follow-up shortly after biopsy and were not included in the outcome analysis. The following definitions were used. (1) Normotension: Systolic blood pressure <140 mmHg and diastolic blood pressure <90 mmHg. (2) Hypertension: Systolic blood pressure >140 mmHg or diastolic blood pressure >90 mmHg, or the need for antihypertensive medication to maintain pressures below these levels (this latter group considered separately in some analyses). (3) MHT: Marked elevation of blood pressure (mean in this study, 193/111 mmHg), obligatorily associated with central nervous system symptoms, such as blurred vision, headaches, nausea, vomiting, or papilledema. (4) Laboratory evidence of TMA: Association of anemia and/or thrombocytopenia, low haptoglobin, presence of schizocytes, elevated lactate dehydrogenase. (5) Bad outcome: Persistent doubling of SCr or requirement for renal replacement therapy (RRT). The glomerular filtration rate was estimated (eGFR) with the simplified modification of diet in renal disease formula.33
The renal biopsies were processed for light microscopy and direct immunofluorescence. Tissue for histology was fixed in AFA and processed and stained by standard methods. Six-micrometer sections were stained for immunofluorescence study with FITC-conjugated antibodies specific for human IgG, IgM, IgA, C1q, C3, κ and λ light chains, and fibrinogen (DAKO, Carpinteria, CA). All biopsy slides were re-reviewed by two senior pathologists (D. Nochy and G.S. Hill) without knowledge of clinical outcomes. The biopsies were graded according to the Oxford classification of IgAN.8,9 TMA lesions were described as (1) “acute,” with fibrin deposits, or (2) as “organized,” with evident fibrosis and recanalization and narrowing of the lumen at the arterial and arteriolar levels. TMA lesions were also classified according to location: arterial, arteriolar, or glomerular. The severity of interstitial cell infiltration and tubular atrophy was semiquantitatively scored on a scale of 0–4+. Interstitial fibrosis was also estimated as a percentage of the renal parenchyma involved. In a separate analysis performed by one pathologist (G. S. Hill), arteries and arterioles were evaluated semiquantitatively for global estimation of arteriosclerosis on a scale of 0–4+ (0, no lesions; 1+, minimal recognizable intimal sclerosis with or without mild recognizable medial fibrosis; 2+, intimal sclerosis with <25% luminal occlusion with or without mild medial fibrosis; 3+, intimal fibrosis with <50% lumenal occlusion with definite medial fibrosis and smooth muscle atrophy; 4+, advanced lesions with >50% luminal occlusion-marked medial lesions); arterial intimal sclerosis on a scale of 0–4+ (0, none; 1+, recognizable intimal sclerosis but no luminal compromise; 2+, intimal sclerosis with <25% luminal occlusion; 3+, 25%–50% occlusion; 4+, >50% occlusion); smooth muscle hypertrophy on a scale of 0–2+ (0, absent; 1, recognizable, minimal to mild; 2, moderate to severe); size of arteriolar lumen on a scale of 0–4+ (0, total occlusion; 1, marked narrowing; 2, definite narrowing; 3, normal diameter; 4, dilated); and hyalin deposits in arteries and arterioles on a scale of 0–2+ (0, absent; 1, present, small, nonocclusive of lumen; 2, present, extensive, and/or impinging on lumen).
Twelve recent cases of IgAN, not included in the original series, all having either acute fibrinoid and/or organized TMA were stained with anti-CD61, an anti-platelet antibody (Y2/51; DAKO). Three cases of TMA of other causes (cocaine-induced, hemolytic–uremic syndrome) and five cases of IgAN without TMA were used as confirmatory positive and negative controls, respectively.
Complement Assays and Genetic Screening
Analyses were performed using EDTA plasma samples at the immunology laboratory of the Hôpital Européen Georges Pompidou. Plasma concentrations of CFH and CFI were measured by ELISA, and concentrations of C4, C3, and complement factor B were determined by nephelometry (Dade Behring, Deerﬁeld, IL). Membrane expression of CD46 was analyzed on granulocytes from patients using phycoerythrin-conjugated antibodies (clone MEM258; Serotec, Oxford, UK). All CFH, membrane cofactor protein, and CFI exons were sequenced as previously described.34
Results were expressed as numerical values and percentages for categorical variables. Continuous variables are expressed as mean (25th to 75th percentiles) because the majority had non-Gaussian distribution. Comparisons were based on Fisher’s exact test for categorical data and the t test for normally distributed continuous data. For non-Gaussian–distributed parameters, we used the nonparametric Mann–Whitney U test to compare continuous variables and the Wilcoxon test to compare two paired groups. The associations of the Oxford criteria with decline in eGFR were evaluated by standard multiple linear regression analysis and with outcome by Cox proportional hazards modeling. P<0.05 was regarded as statistically significant.
Published online ahead of print. Publication date available at www.jasn.org.
This article contains supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2010111130/-/DCSupplemental.
1. Tsai HM: Advances in the pathogenesis, diagnosis, and treatment of thrombotic thrombocytopenic purpura. J Am Soc Nephrol 14: 1072–1081, 2003
2. Noris M, Remuzzi G: Atypical hemolytic-uremic syndrome. N Engl J Med 361: 1676–1687, 2009
3. Droz D, Nochy D, Noël LH, Heudes D, Nabarra B, Hill GS: Thrombotic microangiopathies: renal and extrarenal lesions. Adv Nephrol Necker Hosp 30: 235–259, 2000
4. Donadio JV, Grande JP: IgA nephropathy. N Engl J Med 347: 738–748, 2002
5. Wu J, Chen X, Xie Y, Yamanaka N, Shi S, Wu D, Liu S, Cai G: Characteristics and risk factors of intrarenal arterial lesions in patients with IgA nephropathy. Nephrol Dial Transplant 20: 719–727, 2005
6. Feiner HD, Cabili S, Baldwin DS, Schacht RG, Gallo GR: Intrarenal vascular sclerosis in IgA nephropathy. Clin Nephrol 18: 183–192, 1982
7. Chang A, Kowalewska J, Smith KD, Nicosia RF, Alpers CE: A clinicopathologic study of thrombotic microangiopathy in the setting of IgA nephropathy. Clin Nephrol 66: 397–404, 2006
8. Cattran DC, Coppo R, Cook HT, Feehally J, Roberts IS, et alWorking Group of the International IgA Nephropathy Network and the Renal Pathology Society: The Oxford classification of IgA nephropathy: rationale, clinicopathological correlations, and classification. Kidney Int 76: 534–545, 2009
9. Roberts IS, Cook HT, Troyanov S, Alpers CE, Amore A, et alWorking Group of the International IgA Nephropathy Network and the Renal Pathology Society: The Oxford classification of IgA nephropathy: pathology definitions, correlations, and reproducibility. Kidney Int 76: 546–556, 2009
10. Subías R, Botey A, Darnell A, Montoliu J, Revert L: Malignant or accelerated hypertension in IgA nephropathy. Clin Nephrol 27: 1–7, 1987
11. Chen Y, Tang Z, Yang G, Shen S, Yu Y, Zeng C, Chen H, Liu ZH, Li LS: Malignant hypertension in patients with idiopathic IgA nephropathy. Kidney Blood Press Res 28: 251–258, 2005
12. van den Born BJ, Honnebier UP, Koopmans RP, van Montfrans GA: Microangiopathic hemolysis and renal failure in malignant hypertension. Hypertension 45: 246–251, 2005
13. Vaughan CJ, Delanty N: Hypertensive emergencies. Lancet 356: 411–417, 2000
14. Bakir AA, Dunea G: Drugs of abuse and renal disease. Curr Opin Nephrol Hypertens 5: 122–126, 1996
15. D’Angelo WA, Fries JF, Masi AT, Shulman LE: Pathologic observations in systemic sclerosis (scleroderma). A study of fifty-eight autopsy cases and fifty-eight matched controls. Am J Med 46: 428–440, 1969
16. Traub YM, Shapiro AP, Rodnan GP, Medsger TA, McDonald RH Jr, Steen VD, Osial TA Jr, Tolchin SF: Hypertension and renal failure (scleroderma renal crisis) in progressive systemic sclerosis. Review of a 25-year experience with 68 cases. Medicine (Baltimore) 62: 335–352, 1983
17. Zhang B, Xing C, Yu X, Sun B, Zhao X, Qian J: Renal thrombotic microangiopathies induced by severe hypertension. Hypertens Res 31: 479–483, 2008
18. Galindo M, Gonzalo E, Martinez-Vidal MP, Montes S, Redondo N, Santiago B, Loza E, Pablos JL: Immunohistochemical detection of intravascular platelet microthrombi in patients with lupus nephritis and anti-phospholipid antibodies. Rheumatology (Oxford) 48: 1003–1007, 2009
19. Daniel L, Saingra Y, Giorgi R, Bouvier C, Pellissier JF, Berland Y: Tubular lesions determine prognosis of IgA nephropathy. Am J Kidney Dis 35: 13–20, 2000
20. Katafuchi R, Vamvakas E, Neelakantappa K, Baldwin DS, Gallo GR: Microvascular disease and the progression of IgA nephropathy. Am J Kidney Dis 15: 72–79, 1990
21. Rauta V, Finne P, Fagerudd J, Rosenlöf K, Törnroth T, Grönhagen-Riska C: Factors associated with progression of IgA nephropathy are related to renal function—a model for estimating risk of progression in mild disease. Clin Nephrol 58: 85–94, 2002
22. Sinniah R, Gan HC, Yoon KH: Primary antiphospholipid antibody syndrome and mesangial IgA glomerulonephritis. Am J Nephrol 21: 134–140, 2001
23. Silva MF, Pimentel FL, Faria MS, Carvalho-Costa AE, Nunes JP: IgA nephropathy and antiphospholipid syndrome. Nephron 83: 95–96, 1999
24. Edey M, Strain L, Ward R, Ahmed S, Thomas T, Goodship TH: Is complement factor H a susceptibility factor for IgA nephropathy? Mol Immunol 46: 1405–1408, 2009
25. Sartelet H, Toupance O, Lorenzato M, Fadel F, Noel LH, Lagonotte E, Birembaut P, Chanard J, Rieu P: Sirolimus-induced thrombotic microangiopathy is associated with decreased expression of vascular endothelial growth factor in kidneys. Am J Transplant 5: 2441–2447, 2005
26. El Karoui K, Vuiblet V, Dion D, Izzedine H, Guitard J, Frimat L, Delahousse M, Remy P, Boffa JJ, Pillebout E, Galicier L, Noël LH, Daugas E: Renal involvement in Castleman disease. Nephrol Dial Transplant 26: 599–609, 2011.
27. Frangié C, Lefaucheur C, Medioni J, Jacquot C, Hill GS, Nochy D: Renal thrombotic microangiopathy caused by anti-VEGF-antibody treatment for metastatic renal-cell carcinoma. Lancet Oncol 8: 177–178, 2007
28. Eremina V, Jefferson JA, Kowalewska J, Hochster H, Haas M, Weisstuch J, Richardson C, Kopp JB, Kabir MG, Backx PH, Gerber HP, Ferrara N, Barisoni L, Alpers CE, Quaggin SE: VEGF inhibition and renal thrombotic microangiopathy. N Engl J Med 358: 1129–1136, 2008
29. Nochy D, Lefaucheur C, Hill G: VEGF inhibition and renal thrombotic microangiopathy. N Engl J Med 359: 206, author reply 206–207, 2008
30. Amore A, Conti G, Cirina P, Peruzzi L, Alpa M, Bussolino F, Coppo R: Aberrantly glycosylated IgA molecules downregulate the synthesis and secretion of vascular endothelial growth factor in human mesangial cells. Am J Kidney Dis 36: 1242–1252, 2000
31. Hill GS, Karoui KE, Karras A, Mandet C, Duong Van Huyen JP, Nochy D, Bruneval P: Focal segmental glomerulosclerosis plays a major role in the progression of IgA nephropathy. I. Immunohistochemical studies. Kidney Int 79: 635–642, 2011
32. Jiang L, Zhang JJ, Lv JC, Liu G, Zou WZ, Zhao MH, Zhang H: Malignant hypertension in IgA nephropathy was not associated with background pathological phenotypes of glomerular lesions. Nephrol Dial Transplant 23: 3921–3927, 2008
33. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D Modification of Diet in Renal Disease Study Group: A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 130: 461–470, 1999
34. Le Quintrec M, Lionet A, Kamar N, Karras A, Barbier S, Buchler M, Fakhouri F, Provost F, Fridman WH, Thervet E, Legendre C, Zuber J, Frémeaux-Bacchi V: Complement mutation-associated de novo thrombotic microangiopathy following kidney transplantation. Am J Transplant 8: 1694–1701, 2008