General Information on Vasculitides & Article reviews
Vasculitides are a group of diseases, being all potentially life-threatening and/or affecting vital organs, like heart, lungs or brain, with frequent irreversible damage. With prompt and adapted treatment, the survival at 1 year exceeds 90%, thus the importance to recognize these diseases and refer patients to experienced centers for their management.
Vasculitides are characterized by inflammation of vessel walls , mainly arteries, but sometimes also veins, with or without fibrinoid necrosis and/or granulomas. They can be secondary (to several infections, but also other systemic diseases or cancers, or occur as a reaction to medications or toxic exposures, like levamisole-tainted cocaine). According to the 1994 Chapel Hill nomneclature, primary vasculitides were classified based on the size of the predominantly affected vessels:
Large vessel vasculitides affect the aorta and its major branches and include two main conditions: Giant Cell Arteritis which is seen almost exclusively in individuals older than 50 years and which can cause irreversible blindness in up to 15-20 percent of the cases, and Takayasu’s arteritis, which affects mostly women younger than 40 years-old and can cause arterial limb claudication and/or strokes.
Medium vessel vasculitides include Polyarteritis Nodosa which can affect individuals of all ages and cause infarctions of multiple organs, including the gut, kidneys, heart, muscle and nerves. Before the development of anti-hepatitis B virus vaccine, and the subsequent massive worldwide vaccination campaigns, more than half the cases of polyarteritis nodosa were due to HBV infection. In contrast, Kawasaki Disease is seen mostly in children younger than 4 years-old it has a predilection for the coronary arteries.
Small vessel vasculitides include several entities. The most "famous" ones are associated with the presence of antineutrophil cytoplasm antibodies (ANCA) in the serum of affected patients (at least some of them). These ANCA-related vasculitides include Granulomatosis with Polyangiitis (Wegener’s) , Microscopic Polyangiitis and Eosinophilic granulomatosis with polyangiitis (Churg-Strauss syndrome). These diseases can cause pulmonary-renal syndrome which is characterized by lung hemorrhage and rapidly progressive renal failure. Non-ANCA small vessel vasculitides include many different entities, like Henoch-Schönlein purpura, which is usually a self–limited disease mostly seen in children, and cryoglobulinemic vasculitis (most commonly associated with chronic hepatitis C virus infection). Anti-GBM (glomerular basement membrane) antibody disease (sometimes named Goodpasture disease when it affects lungs and kidney) has been recently included officially in the list of these vasculitides mainly affecting small sized vessel and causing renal disease (with linear deposition of antiGBM antibodies in the glomeruli) and/or alveolar hemorrhage.
Other vasculitides: Behcet's disease is a particular vasculitis that can affect vessels of all sizes, including the veins. Isolated CNS vasculitis is an extremely challenging condition as it affects the vessels of the brain diffusely and can cause various clinical manifestations. Buerger's disease (obliterans thromboangiitis) causes digital ischemia and gangrenous lesions, due to medium- and small-sized artery vasculitis and thrombosis, but also superficial vein thromboses and concerns almost exclusively smokers (classified as a medium-sized vessel vasculitis in Japan
The 2012 Revised international Chapel Hill consensus conference nomenclature of the vasculitides (link to article) now individualizes:
Large Vessel Vasculitis (LVV): Takayasu Arteritis (TAK) and Giant Cell Arteritis (GCA)
Medium Vessel Vasculitis (MVV): Polyarteritis Nodosa (PAN) and Kawasaki Disease (KD)
Small Vessel Vasculitis (SVV):
ANCA-Associated Vasculitis (AAV) including: Microscopic Polyangiitis (MPA), Granulomatosis with Polyangiitis (Wegener’s) (GPA) and Eosinophilic Granulomatosis with Polyangiitis (Churg Strauss) (EGPA)
Immune Complex SVV including: Anti-GBM Disease, Cryoglobulinemic Vasculitis, IgA Vasculitis (Henoch-Schönlein) (IgAV) and Hypocomplementemic Urticarial Vasculitis (Anti-C1q Vasculitis) (HUV).
Variable Vessel Vasculitis (VVV): Behçet's Disease (BD) and Cogan’s Syndrome (CS).
Single Organ Vasculitis (SOV): Cutaneous Leukocytoclastic Angiitis, Cutaneous Arteritis, Primary CNS Vasculitis and Isolated Aortitis.
Vasculitis Associated with Systemic Disease: Lupus Vasculitis, Rheumatoid Vasculitis and Sarcoid Vasculitis.
Vasculitis Associated with Probable Etiology: Hepatitis C Virus-Associated Cryoglobulinemic Vasculitis, Hepatitis B Virus-Associated Vasculitis, Syphilis-Associated Aortitis, Serum Sickness-Associated Immune Complex Vasculitis, Drug-Associated Immune Complex Vasculitis, Drug-Associated ANCA-Associated Vasculitis and Cancer-Associated Vasculitis.
Epidemiology - Vasculitides are rare diseases. There is no epidemiological study on vasculitides in adults in Canada, and few have been done in the United States. However, based on European, Australian, New-Zealander and Japanese studies, one can estimate the overall prevalence and annual incidence ranges of primary vasculitides as follows:
||Prevalence (per million)
||Annual Incidence (per million)
(of population >50 y-o)
|Crude estimates (no specific study)
(of population >50 y-o)
|Down to 5 in Israel in late 1960s; Up to 370 in Norway in mid 1990s; 20 in subjetcs <60 and up to 520 in those >80 y-o in Minnesota
||4.7 to 7 in the UK in early 2000s; Up to 40 in Japan (no epidemiological data for India, but probably at least the same)
||Down to 0.4 in Germany; Up to 2.6 in Minnesota in late 1970s, and 3.3 in Kuwait
||Up to >30 in late 1990s before HBV-related PAN almost disappeared
||Up to 8 in the UK, 16 in Kuwait in the late 1990s, and 77 in Alaska in late 1980s (HBV endemy)
||Acute disease (in general, but damage)
(of children <5 y-o)
|Down to 16 in Czech republic in late 1990s; Up to 2,180 in Japan; in US, 91 in Caucasian vs. 320 in Asian children in early 2000s
||Down to 23 in Paris in 2000, 30 in NYC in 1990; Up to 135 in the UK in 2013 and 160 in Sweden in 2007
||Down to almost 0 in Japan, 2.9 in Spain; Up to 11 in Australia and Minnesota; and 14 in adults in the UK between 2000-2013.
||Down to 25 In Paris in 2000; Up to 94 in Sweden in 2007
Down to 2.7 In Germany;Up to 15 in Japan, and 24 in Kuwait
||Down to 7 in Germany in 1994; Up to 22 in Australia in 2004
||Down to 0 in Japan; Up to 2.7 in the UK in late 1990s, and 4 in Minnesota in late 1970s
||Extremely wide ethnic variations (from 6 in the UK in late 1970s to 4,200 in Turkey in 2000); 100-300 in US , mainly in immigrants; 24 in European-descent vs. 175 in Asian-descent vs 350 in Norh-African- descent French population
||No precise estimation (chronic disease and wide geographical differences); around 4 in Minnesota in mid 2000s
||Only 1 study, in Minnesota
Because research on vasculitides is a rapidly moving field, we do not intend to provide here more information on pathogeny, treatments and outcomes of vasculitides. This would be rapdily outdated and potentially misleading for physicians and/or patients. Excellent reviews are regularly published and will be cited below on this page. Attending the Annual CanVasc conferences and lectures given by its members is another very valuable option to update your knowledge on this topic! (see meetings page ). You can also visit other websites on vasculitis (see links page ).
Selected News and Articles ("CanVasc Journal watch")
NOTE: Due to copyright policies, we can not directly provide here the full texts of the commented articles. Links will lead you to the Journal pages of the articles, where you will be able to get acces to them, through a single article purchase or through your own subscriber account or that of your institution. In case of major, vital and urgent need to get access to one of the articles, you still may try to contact us by email if you have no other options.
FDA approves tocilizumab for the treament of GCA
On May 22nd, 2017, the U.S. Food and Drug Administration expanded the approved use of subcutaneous Actemra (tocilizumab) to treat adults with giant cell arteritis, based on the positive outcome of the Phase III GiACTA study evaluating tocilizumab in patients with GCA (and a previous smaller study from Switzerland). Let's now what will happen in other countries, including in Canada, then at each provincial level regarding the criteria for coverage and reimbursement (i.e. the practical indications). CPx, 03 June 201.
Link to FDA: https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm559791.htm
ARChiVe's pediatric cohort of GPA and MPA.
Congratulation to David Cabral (a CanVasc core member) and colleagues (including from Canada, UK, USA, Italy, Russia and India), who report the largest cohort of children with GPA (n=183) or MPA (n=48) published thus far. The ARChiVe initiative enrolled these patients into the online RedCap registry between March 2007 and November 2015 (and it is still running!). Previous largest cohort from another group (PRINTO) reported on 56 patients.
Main findings of this first descriptive analysis (there will likely be other papers soon, including on follow-up and outcomes) shows that, using the EMA algorithm (which often led to the re-classification of the diagnoses of MPA to GPA made by the treating physicians, or, but less often, from GPA to MPA), children with MPA are younger at diagnosis than those with GPA (13.4 versus 11.2 years), have less frequent lung involvement (44% versus 74%), with less alveolar hemorrhage (15% versus 42%) but more often renal failure requiring dialysis at disease onset (25% versus 13%). Subglottic stenosis occurred in 10% of GPA patients (none in MPA, based on the EMA algorithm), and nasal collapse in 8%. Von Willebrand antigen was elevated in >60% of patients, in both groups (this dosage is not routine in adults). Most children (76%) received cyclophosphamide-based induction, 10-13% received rituximab-based induction (for new diagnosis), and 21% had plasma exchange therapy. Female represented 64% of the patients, 55% were white and mean diagnosis delay was 1.6 months for MPA and 2.1 months for GPA (shorter than in adults by almost 4 months). CPx, 10 Oct 2016.
Cabral DA et al. Comparing Presenting Clinical Features in 48 Children With Microscopic Polyangiitis to 183 Children Who Have Granulomatosis With Polyangiitis (Wegener's): An ARChiVe Cohort Study. Arthritis Rheumatol. 2016;68(10):2514-26. Link
EUVAS-EULAR recommendations for ANCA-associated vasculitides.
A few months after the publication (in November) of the first Canadian recommendations for the management of ANCA-associated vasculitides (link), the equivalent European EUVAS-EULAR updated recommendations are now out (they had to be named EULAR/ERA-EDTA for some internal reasons, out of the topic of this post, but all of us will likely refer to them as the "updated EUVAS guidelines").
Whereas most points are pretty similar, which is reassuring, there are a couple of subtle differences. The CanVasc recommendations included more points and covered more aspects of the diseases, including discussion on pregnancy and pediatric issues, whereas the EUVAS-EULAR ones do not. On the other hand, the EULAR-EUVAS recommendations mention MMF as an alternative to MTX for the treatment of limited GPA, which CanVasc did not at all (this was considered and discussed but at that time, available data were lacking to support MMF as an equivalent to MTX -- available data remains of limited convincing value, as somehow underscored between the lines in the EULAR-EUVAS explanatory text). Whereas CanVasc mentioned rituximab as an "alternative to AZA for maintenance, especially for patients with PR3-ANCA GPA", EULAR-EUVAS goes a little bit farther and lists directly AZA, MTX and rituximab (and MMF) as maintenance treatments, at the exact same levels, for patients with ANCA-associated vasculitides. This latter point sounds great to our Canadian ears and may help supporting applications for rituximab for maintenance (and hopefully its coverage for maintenance, earlier than [too] late) in Canada! The results of the ongoing RITAZAREM study should not be available before late 2018... it is a long time to wait (unless this study does not confirm the superiority of rituximab over AZA found in several retrospective studies and in the French RCT MAINRITSAN study...). - CPx, 15 July 2016.
Yates M, et al. EULAR/ERA-EDTA recommendations for the management of ANCA-associated vasculitis. Ann Rheum Dis. 2016 Jun 23. doi: 10.1136/annrheumdis-2016-209133. [Epub ahead of print]. Link
Rituximab-based is more effective than cyclophosphamide-based induction treatment in PR3-ANCA+ patients
Although this finding from the RAVE study could already be seen in Figure 2B, and Supplementary figures S2 and S6 of the 18-month results of the RAVE study, published 2 years ago (Specks et al. N Engl J Med 2013; 369:417-427), this new article on the RAVE study results makes it clearer. Beside the already known subgroup of relapsers, that of PR3-ANCA+ patients would also achieve more often complete remission at 6 months when treated with rituximab (and prednisone) for induction than with cyclophosphamide (and prednisone; 65% vs. 48%, P=0.04). At 18 months, however, that between-arm difference for the sustained complete remission rates was no longer statistically significant (it still was but only for the sub-subset of PR3-ANCA+ relapsers at enrollment).
Whether this should or not change our therapeutic choices might require further confirmation with another prospective study. However, few retrospective studies already suggested such a possible greater response to rituximab in PR3-ANCA+ patients, compared to cyclophosphamide-based induction and to MPO-ANCA+ patients. Anyhow, it further calls for relapsers, especially those PR3-ANCA+ patients, to have facilitated access to rituximab. For other patient subsets, a preference for rituximab-based induction can still be challenged as there is less supporting evidence.- December 2015, CPx.
Unizony et al. Clinical outcomes of treatment of anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis based on ANCA type. Ann Rheum Dis. 2015 Nov 30. Link.
How the classification of vasculitides can help and impact their therapeutic management
A review and perspective article from some CanVasc core members, in open access, on how the treatment of vasculitides can be decided and guided based upon their classification, how it has changed over the pat years and will likely further change in the near future.- June 2015, CPx.
Baldwin C, Carette S, Pagnoux C. Linking classification and therapeutic management of vasculitides. June 2015. Arthritis Research and Therapy. Link
Eosinophilic granulomatosis with polyangiitis (Churg–Strauss) (EGPA): Consensus Task Force recommendations for evaluation and management
The EGPA Consensus Task Force experts comprised 23 international experts from different subspecialties from Europe, USA and Canada. Following a modified Delphi quesitonnaire, they devided 22 recommendations concerning the diagnosis, initial evaluation, treatment and monitoring of EGPA patients. These recommendations will provide physicians with a practical tool for effective and individual management of their EGPA patients. - May 2015 CPx.
Groh M, Pagnoux C, Baldini C, Bel E, Bottero P, Cottin V, Dalhoff K, Dunogué B, Gross W, Holle J, Humbert M, Jayne D, Jennette JC, Lazor R, Mahr A, Merkel PA, Mouthon L, Sinico RA, Specks U, Vaglio A, Wechsler ME, Cordier JF, Guillevin L. Eosinophilic granulomatosis with polyangiitis (Churg-Strauss) (EGPA) Consensus Task Force recommendations for evaluation and management. Eur J Intern Med. 2015 May 9. pii: S0953-6205(15)00144-2. doi: 10.1016/j.ejim.2015.04.022. [Epub ahead of print] Link
Results of the Needs Assessment Questionaire for the development of CanVasc recommendations for ANCA-associated vasculitis
The results of the NAQ are now available at Open J Rheum website: http://benthamopen.com/FULLTEXT/TORJ-9-16. This work was the first step towards the development of the CanVasc recommendations for the management of ANCA-assocaited vasculitis. - April 2015, CPx.
Famorca L, Twilt M, Barra L, Bakowsky V, Benseler S, Cabral D, Carette S, Dhindsa N, Fifi-Mah A, Goulet M, Khalidi N, Khraishi M, McGeoch L, Milman N, Pineau C, Shojania K, Taylor-Gjevre R, Towheed T, Trudeau J, Yacyshyn E, Liang P, Pagnoux C; Canadian Vasculitis network (CanVasc). Development of Canadian Recommendations for the Management of ANCA-Associated Vasculitides: Results of the National Needs Assessment Questionnaire. Open Rheumatol J. 2015 Apr 14;9:16-20. doi: 10.2174/18743129014090100016. eCollection 2015. Link
ENT manifestations of EGPA
A general review on ENT manifestations of EGPA by some of the CanVasc core members in open access in the Advances in Cellular and Molecular Otolaryngology: http://www.cellmoloto.net/index.php/acmo/article/view/27181. - April 1, 2015. CPx
Renal Immune Complex Deposition in Patients with Primary ANCA-Associated Vasculitis
AAV is typically described as “pauci-immune” meaning “few immune deposits,” and there have been several reports of moderate IC deposition in the renal biopsies of patients with AAV. This descriptive retrospective study from China and published in January 2015 in a rheumatology journal describes the clinical and pathological features and outcomes of 34 patients with ANCA-associated vasculitis (AAV) and immune-complex (IC) deposition on renal biopsy. They compare these patients to 76 other patients with classic “pauci-immune” GN without such intensity of IC deposits. “Pauci-immunity” was defined by the authors as staining of 1+ or less on DIF and the absence of deposits on EM. “IC deposition” was defined as staining of 2+ or more on DIF or deposits on EM. Biopsies were read by two independent renal pathologists.
Among the “IC-patients,” 80% were MPA, 18% GPA, 2.9% EGPA with a mean age of 53. Eighty-percent (80%) were MPO-ANCA positive. Mean BVAS was 21.3 and mean FFS was 1.8. Multi-organ damage was common and renal and lung involvement was most common. 80% had GFR <60. Renal IC deposits were most commonly C3 and IgM, located in mesengial areas and capillary loops. IgA deposition was less common but still present in 1/4th of biopsies. Deposition of IgG, C4 and C1q was rare. Renal pathology in IC-patients was otherwise not statistically significantly different compared to “pauci-immune” patients; however, they had more interstitial fibrosis and glomerular cellularity.
Notably, IC-patients had worse proteinuria, were more likely to have nephrotic syndrome (30% vs 10%), and were more likely to have hypocomplementemia with low C3 compared to patients with “pauci-immune” GN. All IC-patients received induction therapy with oral glucocorticoids and immunosuppressants versus 92.1% of “pauci-immune” patients; however, fewer IC-patients (5.9%) received PLEX compared to “pauci-immune” patients (15.8%). More IC-patients remained HD dependent versus “pauci-immune” patients (100% vs 56.3%). Progression to ESRD was also higher in IC-patients at 32.4% versus 13.1%, although mortality was comparable.
This paper highlights that AAV patients may have some IC deposits on biopsy. This may be particularly true of Asian patients. However, the study lacks detail on patient comorbidities such as hypertension, diabetes or the presence of ANA, and, as there is no standardization of what IC-deposition versus pauci-immune in these patients, a comparison group of patients with real IC-mediated GN such as renal lupus or cryoglobulinemic GN would have been useful.– C. Baldwin, March 4, 2015.
Li X, Zhang W, Yu HJ, Pan XX, Shen PY, Ren H, Wang WM, Chen N. Clinical and pathological study on patients with primary antineutrophil cytoplasmic autoantibody-associated vasculitis with renal immune complex deposition. J Clin Rheumatol. 2015 Jan;21(1):3-9. Link
Pulmonary Fibrosis in ANCA-Associated Vasculitis
This French descriptive retrospective study describes main clinical features and outcomes of 49 ANCA-associated vasculitis (AAV) patients from 16 centers with concomitant pulmonary fibrosis (PF) and compares them to 65 patients previously reported in the literature. The rationale for the study is that although PF has been reported in association with AAV and in association with MPO-ANCA, this association has not been well described in non-Asian populations.
Notable key characteristics of the IPF-patients include older median age, male predominance (61% and 65%), and strong association with MPO-ANCA (88% and 94%). Interestingly, among patients in both the French cohort and the literature, the timing of the diagnosis of PF was similar, with IPF diagnosis preceding the onset of vasculitis in 45% and 54.7% of cases, being diagnosed concomitantly in 43% and 40%, and subsequently in 12% and 5.6% of cases. In the French cohort, mild peripheral eosinophilia was reported in 31% of IPF-patients. Radiographic patterns were similar between the two groups with typical UIP reported in most patients (43% of the French patients and 83% of the literature patients). Other patterns in the French Cohort included combined-PF-emphysema (21%), atypical UIP (14%), NSIP (9.5%) and fibrotic NSIP (7%). NSIP patients were diagnosed at younger age compared to typical/atypical UIP and fibrotic NSIP patients.
Median follow-up was 48 months and 45 months in the French cohort and literature, respectively. Twenty-seven percent (27%) of the French patients progressed to chronic respiratory deficiency requiring home oxygen vs 35% among the literature patients. Thirty-seven (37%) of the French patients and 56% of the literature patients died and, in both groups, the most common cause of death was chronic respiratory insufficiency – 61% in French patients and 44% in literature patients. In both groups, induction therapy with glucocorticoids (GC) alone versus glucocorticoids in combination with immunosuppressive drugs was associated with increased mortality. In the French cohort, 1, 3 and 5 year survival rates were higher in patients with both GC and cyclophosphamide/rituximab at 94% vs 73%, 94% vs 64% and 71% vs 51% and treatment with GC alone was associated with an HR for death of 2.94. This was consistent with the literature patients for whom treatment with GC alone carried a 1.7x increased risk of death (83% vs 48%) at follow-up. Other factors in the French cohort that were associated with increased risk of death included chronic respiratory insufficiency (HR 7.44), weight loss (HR 2.83), higher eosinophil counts at AAV diagnosis (HR 1.09), and older age at diagnosis of both PF and AAV (HR of 1.08 and 1.09, respectively).
This study is limited by its retrospective design, but is the largest study to date describing the association between PF and ANCA-associated vasculitis in European patients. Importantly, almost half of patients develop PF prior to the onset of vasculitis, raising the question of whether ANCA testing should be a routine aspect of the work-up for PF. Further studies are required to elucidate the relationship between AAV and PF and determine the optimal treatment strategies.– C. Baldwin, March 4, 2015.
Comarmond C, Crestani B, Tazi A, Hervier B, Adam-Marchand S, Nunes H, Cohen-Aubart F, Wislez M, Cadranel J, Housset B, Lloret-Linares C, Sève P, Pagnoux C, Abad S, Camuset J, Bienvenu B, Duruisseaux M, Hachulla E, Arlet JB, Hamidou M, Mahr A, Resche-Rigon M, Brun AL, Grenier P, Cacoub P, Saadoun D. Pulmonary fibrosis in antineutrophil cytoplasmic antibodies (ANCA)-associated vasculitis: a series of 49 patients and review of the literature. Medicine (Baltimore). 2014 Nov;93(24):340-9. Link
Rituximab for the treatment of eosinophilic granulomatosis with polyangiitis
This multicenter retrospective chart review study investigated the efficacy of Rituximab in the treatment of EGPA in 41 patients. The majority of them had relapsing or refractory disease; 5 were given the treatment first line. The dose and infusion schedules were variable with some patients receiving 375mg/m2/week for 4 weeks and others receiving 1g every two weeks in 2 doses; around half (19/41) had received only 1 induction infusion, whereas 22 received 2 or 3 infusions at either 6 months and/or 12 months intervals. The primary outcomes were complete response (defined as BVAS = 0) and partial response (defined as BVAS reduction of 50%).
Complete remission was observed in 34% and 49% at 6 and 12 months, respectively. Partial remission was observed in 49% and 39% at 6 and 12 months, respectively. The median BVAS was 11 (6-17.5) at baseline, 2 (0-6.2) at 6 months, and 1 (0-2) at 12 months. BVAS improved in 83% and 88% at 6 and 12 months, respectively. A positive ANCA was associated with increased likelihood of achieving complete remission (80%) versus negative ANCA (38%). Among the 10 renal patients, 7 achieved remission. The median daily prednisone dose reduced from 15mg at baseline to <10mg in 52% and 53% at 6 and 12 months, respectively. Only 2 patients were able to discontinue prednisone. Immunosuppressant drugs were used by 44% of patients at baseline and declined to 28% at 12 months. There was no clinical or demographic feature that was predictive of achieving remission. Similarly, there was no difference in rate of response based on dosing schedule. B cell depletion occurred in all patients. Among the 17 patients that had data, B cell reconstitution occurred in 17%. There was no difference in eosinophil counts although the levels were already low at baseline, possibly due to prednisone and other immunosuppressant drugs. Half the patients had adverse events, the most common of which were infections and infusion reactions. Among the 15 infections in 14 patients, 6 were severe. Among the 10 infusion reactions, 8 were mild and 2 were severe (one was worsening asthma). There was no leukopenia or death.
This important study is of course limited by its retrospective design, the variable dosing and infusion schedules, and its relatively short follow-up period of 12 months. However, it shows that BVAS decreases in most treated patients, although complete remission was achieved in only half of them and the majority remained on prednisone. More data, ideally from prospective studies, are needed to determine the place of rituximab in patients with (refractory) EGPA. – C.Baldwin, December 15, 2014.
AJ Mohammad, A hot, F Arnt, F Moosig, MJ Guerry, N Amudala, R Smith, P Sivasothy, L Guillevin, PA Merkel. Rituximab for the treatment of eosinophilic granulomatosis with polyangiitis (Churg-Strauss). Annals of Rheumatic Disease, 2014. Link