Journal Logo


Cardiac Involvement in Churg-Strauss Syndrome

Impact of Endomyocarditis

Neumann, Thomas MD; Manger, Bernhard MD; Schmid, Michael MD; Kroegel, Claus MD, PhD; Hansch, Andreas MD; Kaiser, Werner A. MD, MS; Reinhardt, Dirk MD; Wolf, Gunter MD; Hein, Gert MD; Mall, Gerhard MD; Schett, Georg MD; Zwerina, Jochen MD

Author Information
doi: 10.1097/MD.0b013e3181af35a5
  • Free



Churg-Strauss syndrome (CSS) is a necrotizing small-vessel vasculitis with extravascular granulomas and eosinophilic infiltrates.4 The size of the targeted vessels and the occurrence of antineutrophil cytoplasmic antibodies (ANCA) led to its classification as an ANCA-associated systemic vasculitis. However, several features of this disease render CSS unique as compared to Wegener granulomatosis and microscopic polyangiitis: 1) virtually all patients report a history of bronchial asthma and/or chronic sinusitis ("allergic granulomatous angiitis"); 2) eosinophils are usually found in high numbers in the peripheral blood and tissue biopsies, which also leads to the classification of CSS in the group of hypereosinophilic disorders; and 3), organ manifestations are different from the other ANCA-associated vasculitides: cardiac; peripheral nerve; and ear, nose, and throat system involvement are frequent, while renal disease is rare.15

The treatment and prognosis of an individual CSS patient depends on the type and severity of organ manifestations.6 Guillevin et al7 identified 5 factors associated with poor prognosis (the "Five Factor Score" [FFS]) requiring intensive immunosuppression in these patients, with cardiomyopathy being 1 of the factors. The heart clearly is a target organ in CSS, affecting about one-third of cases. Importantly, approximately 50% of deaths in CSS patients are directly attributed to cardiac involvement and occur mainly within the first few months following diagnosis, underlining its prognostic importance.3,6 However, the type of cardiac disease may range from mild to severe, finally causing substantial morbidity and mortality. If cardiac biopsy is performed, eosinophilic infiltration of the cardiac tissue is the dominant finding.1

Cardiomyopathy has long been recognized as the clinical correlate of endomyocardial fibrosis for most of the hypereosinophilic disorders. Idiopathic hypereosinophilic syndrome, chronic eosinophilic leukemia, and tropical endomyocardial fibrosis all share Löffler endocarditis as the most severe cardiac manifestation causing high morbidity and mortality.13 The prevalence, clinical pattern, and impact of Löffler endocarditis in CSS patients are unclear, but its presence has been reported anecdotally.2 Herein, we report a cohort of CSS patients in whom detailed cardiac studies have been performed.


Study Population and Clinical Procedures

We collected data retrospectively by reviewing medical records. Two rheumatology and 1 pulmonary medicine departments of tertiary referral centers (University of Erlangen-Nuremberg and University of Jena) and 1 cardiology department of a secondary referral center (Evangelisches Krankenhaus, Hamm) contributed patients to this study. We retrieved all medical records and radiology and pathology reports of clinically diagnosed CSS patients from 1982 to 2008. Patients were entered in the study if data from initial presentation and at least 1 follow-up visit were available. Initially, 51 patients were screened. Two patients were excluded (diagnosis not confirmed and missing data), resulting in 49 patients available for analysis. All patients fulfilled the 1990 American College of Rheumatology criteria11 and/or the Chapel Hill definition8 of CSS.

At diagnosis, patients had routine laboratory investigations including complete blood count, serum creatinine, liver functions tests, C-reactive protein, serum immunoglobulin E (IgE) levels and ANCA testing using indirect immunofluorescence and enzyme-linked immunosorbent assay (ELISA). Some data were missing: C-reactive protein levels in 5 patients, serum IgE levels in 3 patients, and ANCA in 3 patients. Chest X-ray was performed in every patient, and at least 1 diagnostic biopsy was performed in all but 2 patients. Other investigations were performed depending on the clinical presentation of the individual patient. Number, location, and histopathologic diagnosis were recorded.

Cardiac involvement of CSS was assumed in patients who had either clinical signs of acute cardiac failure, laboratory findings indicative for myocardial damage, or abnormalities in cardiac imaging such as echocardiography or cardiac MRI. Causes other than vasculitis had to be excluded.

To investigate cardiac involvement, all patients had laboratory investigations of creatinine kinase, and a 12-lead electrocardiography (ECG) was recorded at diagnosis. Elevated creatinine kinase levels lead to analysis of troponin I. Echocardiography was performed in every patient with clinically suspected impaired cardiac function. Results were considered abnormal when regional or global systolic dysfunction, valvular abnormalities, pulmonary hypertension, or pericardial effusions were found.

Contrast agent-enhanced cardiac MRI was performed in 12 patients using a 1.5-T scanner (Siemens Medical Solutions, Erlangen, Germany) with a cardiac-dedicated phased-array coil. The image acquisition was ECG-triggered by standard software. Studies consisted of steady-state-free precession sequences and T2-weighted breath-hold sequences in all patients. Short-axis, vertical long-axis, and 4-chamber views were obtained. Steady-state-free precession sequences were performed to assess regional wall motion abnormalities.

Late gadolinium-enhanced MRI was performed 10 minutes after the administration of 0.1 mmol/kg gadolinium-DTPA (gadopentetate meglumine; 0.1 mmol/kg; Schering, Berlin, Germany) and was obtained in all patients. Gadolinium contrast medium was administered intravenously. Inversion recovery prepared breath-hold images were acquired. Short-axis, vertical long-axis, and 4-chamber views were obtained. Regional differences in right and left ventricular wall enhancement were measured and localized according to the 17-segment model10 (Figure 1). Two independent radiologists experienced in contrast-enhanced cardiac MRI evaluated the studies. Endomyocardial biopsies were performed for diagnostic purposes in 11 patients.

Topography of cardiac involvement in CSS. Circumferential polar plot demonstrating left ventricular segmentation in 17 myocardial segments10 (A) and the distribution of late gadolinium enhancement per segment in 12 patients with cardiac involvement of Churg-Strauss syndrome (B).

Finally, we retrieved information on immunosuppressive treatment and outcome until the latest follow-up for all patients. The occurrence of severe disease flares was defined as the necessity to increase the dose of or to re-introduce immunosuppressive drugs in addition to corticosteroids.

Statistical Analysis

Data derived from the CSS cohorts were entered into a database and analyzed by SPSS 16.0 statistical software. Group data are expressed as mean values ± SD. Differences between groups were statistically evaluated by the chi-square test or the Mann-Whitney U test as appropriate. Organ involvement was calculated as percentages of affected individuals. All tests were 2-sided, and p values less than 0.05 were considered statistically significant.


Clinical and Laboratory Characteristics

We analyzed 49 patients with CSS. Detailed clinical characteristics of these patients are depicted in Table 1. The mean age was 43 years (range, 18-69 yr) with a balanced sex ratio (23 male, 26 female patients). Symptom duration before diagnosis was 0.7 years (95% confidence intervals, 0.4-1.0). All but 3 patients had asthma, and migratory lung infiltrates were found in 27 patients (55%). Renal involvement was diagnosed in 6 patients (12%), 25 patients (52%) had peripheral neuropathy, and involvement of the central nervous system was diagnosed in 3 patients (6%). Gastrointestinal manifestations were found in 9 patients (18%).

Characteristics of CSS Patients According to the Presence of Cardiac Involvement

CSS patients had elevated C-reactive protein levels (mean, 39.9 ± 44.2 mg/L), high serum IgE levels (mean, 849 ± 1044 U/L), and severe peripheral blood eosinophilia (mean absolute count, 6333 ± 7197/μL; mean relative count, 34.8% ± 19.4%). ANCA were found in 7 patients: 5 patients had p-ANCA with specificity for myeloperoxidase and 2 patients had c-ANCA specific for proteinase 3. Biopsies were performed in 46 of 49 patients. CSS was biopsy-proven in 39 of 46 patients (85%). Extravascular eosinophil infiltrates were seen in 36 (78%), vasculitis in 16 (35%), and granulomas in 6 (13%) biopsy specimens.

Cardiac Involvement

In 22 patients (45%), cardiac involvement was diagnosed by clinical signs of acute cardiac failure due to impaired left ventricular function or rhythm abnormalities and mostly confirmed by cardiac imaging. In 2 patients, the cardiac manifestation was a major flare of disease after initial remission. All but 1 patient were previously asymptomatic for cardiac function and had normal cardiopulmonary capacity in daily life activities. One patient had pre-existing coronary heart disease with preserved cardiac function. Compared to patients without cardiac involvement, patients with cardiac involvement had a shorter interval of CSS-specific symptom duration until initial diagnosis was made, while organ involvement was equally distributed in both groups. All patients with cardiac involvement were ANCA-negative (p < 0.05). Serum IgE and C-reactive protein levels were not different between these 2 groups, whereas eosinophilic blood count was statistically significantly higher in patients with cardiac manifestations of CSS than in those without (mean, 9947 ± 9092 vs. 3657 ± 3710 eosinophils/μL, respectively; p < 0.001). Patients with cardiac involvement had higher FFSs than patients without cardiac involvement. These findings are summarized in Table 1.

ECG changes were found in 17 patients: right or left bundle branch block was identified in 5 patients, ST and T-wave abnormalities in 14 patients. In 2 patients, ventricular arrhythmias occurred, whereas atrioventricular blocks were not found at all in this patient cohort. On echocardiographic examination, pericardial effusions were detectable in 9 patients, mostly circumferential. Mitral valve insufficiencies were found in 6 patients (3 mild, 3 moderate), aortic valve insufficiencies in 3 patients (all mild), and tricuspid valve insufficiencies in 7 patients (4 mild, 3 moderate). Cardiac impairment due to valve destruction could be excluded in these patients. Left ventricular ejection fraction (LVEF) was reduced in 11 patients: mild (LVEF 40%-55%) in 6 patients, moderate (LVEF 30%-39%) in 1 patient, severe (LVEF <30%) in 4 patients. In 6 patients, pulmonary hypertension was detected, and in 4 patients thrombotic structures were observed within the ventricular chambers (3 right chamber, 1 left and right chamber). Fourteen patients had elevated cardiac markers (troponin I).

Cardiac MRI was performed in 12 of the 22 patients with cardiac involvement. Late gadolinium enhancement was present in 10 patients (mean, 9.6 segments per patient; range, 2-17 segments). Most enhancing lesions were located in the apical and mid-cavity left ventricular segments (segment 7-17 [73%]) (Figures 1 and 2). All patients had late gadolinium enhancement in the subendocardial layers suggesting Löffler endocarditis, 1 patient in intramyocardial layers. Cardiac thrombus formation could be identified in 2 patients (1 right chamber, 1 both chambers), 1 of which was not identified on echocardiography (Figure 3). Coronary angiography to exclude coronary artery disease was performed in 13 patients. Twelve patients had completely inconspicuously coronary arteries, whereas 1 had pre-existing coronary artery disease with 2-vessel disease that did not appear different from previous coronary catheterizations before the onset of CCS.

Cardiac MRI with typical signs of cardiac involvement in CSS. Short axis delayed-enhanced image of left ventricle after injection of gadolinium. Subendocardial late enhancement demonstrates fibrosis at the left ventricle (arrows) (A). Short axis phase-sensitive T1-weighted contrast-based late enhancement sequence (PSIR). Subendocardial late enhancement (subendocardial band-like hyperintensity) due to endomyocardial fibrosis in the left ventricle (arrows) and into myocardium (arrowhead) (B). Four-chamber view (C) and 2-chamber view (D) phase-sensitive T1-weighted contrast-based late enhancement sequence (PSIR) subendocardial late enhancement due to endomyocardial fibrosis in the apex of the left ventricle (arrowhead) and in the right ventricle (arrow). Four-chamber noncontrast T2-weighted (TR/TE, 41/1) view demonstrates pericardial effusions (arrows) and pleural effusions (arrowhead) (E). Short axis noncontrast T2-weighted (TR/TE, 41/1) view shows circumference pericardial effusions (F).
Cardiac MRI with typical signs of cardiac involvement in CSS. Short axis noncontrast view (TIRM; TR/TE, 1620.5/60) demonstrates anteroseptal (arrows) and lateral (double arrow) and subendocardial inferior (arrowhead) myocardial edema as a sign of myocardial inflammation (A). Two-chamber (B) and 4-chamber (C) noncontrast views (TIRM; TR/TE, 1620.5/60) demonstrate apical (arrowhead) and subendocardial inferior (arrow, B) and lateral (arrow, C) myocardial edema as a sign of myocardial inflammation. Noncontrast T2-weighted (TR/TE, 1620.5/60) 4-chamber view shows myocardial edema of the septum (upper arrow) and small pleural effusion (lower arrow) (D). Four-chamber view, phase-sensitive T1-weighted contrast-based late enhancement sequence (PSIR) shows subendocardial late enhancement (double arrow) due to endomyocardial fibrosis in the apex and apical thrombus of the right (arrow) and of the left ventricle (arrowhead) (E). Two-chamber view, phase-sensitive T1-weighted contrast-based late enhancement sequence (PSIR) shows apical thrombus of the left ventricle (arrowhead) (F).

To confirm suspected Löffler endocarditis in these patients, endomyocardial biopsies were performed in 11 patients. Acute eosinophilic endomyocarditis was seen in 2 patients, and biopsy was nondiagnostic in 2 patients. Typical histologic signs of Löffler endocarditis with the formation of fibrotic tissue could be observed in 7 patients (Figure 4). Compared with patients with cardiac involvement other than Löffler endocarditis, patients with Löffler endocarditis had little or no pericardial effusions, more severe left heart failure, elevated cardiac enzymes, and thrombus formation (Table 2).

Endomyocardial biopsies in CSS. Left ventricular endomyocardial biopsy, hematoxylin and eosin staining, primary magnification 2.5x and 20x: active myocardial inflammation with eosinophils illustrating the inflammatory stage of Löffler endomyocarditis (A and B). Left ventricular endomyocardial biopsy, hematoxylin and eosin staining, primary magnification 2.5x and 20x: severe endomyocardial fibrosis corresponding to the fibrotic stage of Löffler endomyocarditis (C and D).
Abnormal Cardiac Findings in 22 Patients With CSS

Severity, Treatment, and Outcome

The mean follow-up of patients was 3.9 ± 4.6 years. According to the FFS, 22 patients (44.9%) had good prognosis (FFS = 0) and 27 patients (55.1%) had poor prognosis (FFS ≥ 1). Cardiac involvement led to a high prevalence of reduced cardiac function at presentation, as described above. The diagnosis of Löffler endocarditis was associated with an even more pronounced reduced left ventricular function (mean LVEF, 43.6% ± 15.4%) compared to patients with cardiac involvement other than endomyocarditis (mean, 57.1% ± 7.6%; p = 0.047). During follow-up, cardiac function improved with immunosuppressive treatment in both groups (mean, 17%; range, [min]26% to + 96%).

All patients were initially treated with high-dose oral corticosteroids; 42 patients (85.7%) received additional immunosuppression (22 cyclophosphamide, 13 azathioprine, 6 methotrexate, 4 mycophenolate, 14 interferon-α, 2 rituximab, 2 plasmapheresis). Additional immunosuppression was more often used in patients with a FFS ≥ 1 (25 of 27 patients) than in patients with a good prognosis (17 of 22 patients). Remission was achieved in all patients initially. Severe disease flares were reported in 11 patients (22%). No relationship was found between the appearance of relapses and initial CSS activity as described by FFS. Two patients died after 5.2 and 8.3 years, respectively, due to cardiac failure attributed to the underlying disease.


In the current study, we demonstrate the prevalence and impact of cardiac manifestations in CSS. Our results show that 1) cardiac involvement is a frequent manifestation in CSS, 2) it affects the peri-, myo- and endocardial tissue, and 3) it causes impaired left ventricular function at diagnosis. Endomyocardial disease (Löffler endocarditis) may represent the most severe form of cardiac involvement, similar to other hypereosinophilic disorders.

Cardiac involvement associated with CSS is heterogeneous, and previous studies have reached different conclusions. However, the different clinical presentation among referral centers is not surprising: nephrology departments usually see a higher prevalence of ANCA-positive patients and therefore also see more frequent neurologic and renal disease. The latter typically is rare in rheumatology departments. Cardiac involvement has been associated with the absence of ANCA.15,18 This is also reflected in our cohort, where a higher prevalence of cardiac disease according to the lower prevalence of ANCA has been found. However, the overall presence of ANCA in our patient cohort was lower than in previous studies.15,18

Investigation of the incidence and clinical appearance of cardiac disease in CSS is important, as associated morbidity and mortality are attributed to disease-associated heart failure. Clinical data from large CSS cohorts revealed pericardial effusion and the clinical pattern of cardiomyopathy as the most common cardiac features of CSS.9,15 Cardiac valve abnormalities, granulomatous myocarditis, coronary vasculitis, and other clinical features have all been described but are mainly restricted to single case reports. However, the underlying pathogenesis of cardiomyopathy is still unknown. Herein, we provide detailed evidence of the importance of screening CSS patients for cardiac disease, which was present in 44% of our patients. Though the outcome was generally excellent, disease-related death occurred in 2 patients, both due to underlying CSS-associated cardiac disease. Importantly, although heart function improved in most patients under treatment, in some patients heart disease did not respond to therapy.

CSS patients with a pericardial effusion but preserved systolic left ventricular function had an excellent outcome and maintained cardiac function at follow-up. However, 27% of patients presented with severe cardiac manifestations, associated with reduced cardiac function in the majority of cases. These patients had endomyocarditis as the underlying cause. ECG and echocardiography could not demonstrate specific findings for this condition in these patients. In 3 individuals cardiac thrombus formation as an indirect sign of endocardial involvement was diagnosed, but no endomyocardial thickening, as previously described.12

Importantly, cardiac MRI suggested endomyocarditis in our patients with cardiomyopathy and was confirmed by endomyocardial biopsy. Delayed contrast enhancement in cardiac MRI has been shown to occur in the setting of suspected endomyocarditis and is associated with active endomyocarditis or endomyocardial fibrosis.10 Late wash-out of contrast agents is thought to be caused by interstitial fibrosis with increased extracellular space. In contrast to scar formation after myocardial infarction, late gadolinium enhancement is not restricted to certain regions of coronary perfusion in endomyocarditis. We note that signs of myocarditis were found in only 2 patients, possibly due to early biopsy, while the remaining showed subendocardial involvement similar to classical Löffler endocarditis.

The current study exactly defines the segmental localization of late gadolinium enhancement in patients with endomyocarditis. We were able to demonstrate that apical and mid-cavity segments are most frequently involved. This is in contrast to the pattern of cardiac involvement in sarcoidosis, where late gadolinium enhancement is predominantly detected in basal segments, and in acute myocarditis, where late gadolinium enhancement is mainly restricted to subepicardial and midmyocardial structures.19,20 Cardiac thrombus formation was found in 3 patients. Thus, endomyocarditis seen in our cohort clinically behaves similar to classical Löffler endocarditis associated with hypereosinophilic syndrome. This is in accordance with our finding that cardiac involvement in CSS is associated with the highest eosinophil counts at diagnosis.

Löffler endocarditis has long been recognized as the most severe clinical presentation significantly contributing to morbidity and mortality in other hypereosinophilic disorders such as tropical endomyocardial fibrosis and idiopathic hypereosinophilic syndrome (HES). Endomyocardial fibrosis develops in 3 stages: an acute necrotic stage with myocyte necrosis, followed by cardiac thrombus formation in both ventricles and finally resulting in endomyocardial fibrosis causing the clinical picture of restrictive cardiomyopathy.5 Recently it has been demonstrated that hypereosinophilic syndrome represents a heterogeneous group of disorders including myeloproliferative disorders such as the FIP1L1-PDGFRA associated form (M-HES), the lymphocytic variant of hypereosinophilic syndrome associated with clonal T-cell expansion (L-HES) and other yet to be defined subgroups.14,16,17 While it is unclear why all patients with hypereosinophilic syndrome patients do not develop cardiac disease, earlier studies identified risk factors that were consistent with clinical features of the M-HES variant: high serum vitamin B12 levels, splenomegaly, cytopenia, myelofibrosis, and dysplastic eosinophils.

Compared to what has been referred to as idiopathic hypereosinophilic syndrome, endomyocarditis seems to be less common in CSS patients. However, as described above, the patients with hypereosinophilic syndrome most severely affected are those with chronic eosinophil leukemia. These patients also show prolonged eosinophilia due to poor treatment responsiveness before imatinib treatment, which is usually not the case in patients with CSS. Thus, the development of cardiac disease may depend on the length and magnitude of eosinophilia in affected patients. We demonstrate in the current study that patients with cardiac involvement have significantly higher eosinophil counts compared with those who do not have cardiac disease. This is also supported by the first report4 of CSS patients published in 1951: autopsy studies revealed severe heart involvement in most patients including a significant amount of endomyocardial fibrosis, which the authors attributed to eosinophilic inflammation and secondary changes due to vasculitic damage.

All of our patients fulfilled the American College of Rheumatology criteria11 for CSS, and all but 3 patients had asthma and most of the patients had clearly elevated serum IgE levels. In addition, the majority suffered from chronic rhinosinusitis further underlining the "allergic state" of these patients. Histopathologic evidence of CSS in our cohort was comparable to that of other reported CSS cohorts.15 Thus, we believe the patients included in the current study represent what is currently considered CSS and, hence, data from the current study provide valid information about the occurrence of cardiac manifestations in patients with CSS. All patients described in this study were followed by specialized departments, which may bias the analysis and contribute to the high proportion of cardiac involvement due to more severe disease. The strengths of the study are the detailed cardiac examinations performed including cardiac MRI and endomyocardial biopsy, especially as echocardiography was not able to demonstrate specific findings in most of these patients.

In conclusion, we found a high prevalence of cardiac involvement in CSS patients, and endomyocarditis represents the most severe form of cardiac disease. High eosinophil counts and a negative ANCA test are associated with the presence of cardiac disease. Cardiac MRI seems to be a sensitive diagnostic tool for detecting endomyocarditis. The detection of endomyocarditis in CSS patients requires early and aggressive treatment.


1. Agard C, Rendu E, Leguern V, Ponge T, Masseau A, Barrier JH, Trochu JN, Hamidou MA, Guillevin L. Churg-Strauss syndrome revealed by granulomatous acute pericarditis: two case reports and a review of the literature. Semin Arthritis Rheum. 2007;36:386-391.
2. Alter P, Maisch B. Endomyocardial fibrosis in Churg-Strauss syndrome assessed by cardiac magnetic resonance imaging. Int J Cardiol. 2006;108:112-113.
3. Bourgarit A, Le Toumelin P, Pagnoux C, Cohen P, Mahr A, Le Guern V, Mouthon L, Guillevin L, French Vasculitis Study Group. Deaths occurring during the first year after treatment onset for polyarteritis nodosa, microscopic polyangiitis, and Churg-Strauss syndrome: a retrospective analysis of causes and factors predictive of mortality based on 595 patients. Medicine (Baltimore). 2005;84:323-330.
4. Churg J, Strauss L. Allergic granulomatosis, allergic angiitis, and periarteritis nodosa. Am J Pathol. 1951;27:277-301.
5. Fauci AS, Harley JB, Roberts WC, Ferrans VJ, Gralnick HR, Bjornson BH. NIH conference. The idiopathic hypereosinophilic syndrome. Clinical, pathophysiologic, and therapeutic considerations. Ann Intern Med. 1982;97:78-92.
6. Guillevin L, Cohen P, Gayraud M, Lhote F, Jarrousse B, Casassus P. Churg-Strauss syndrome. Clinical study and long-term follow-up of 96 patients. Medicine (Baltimore). 1999;78:26-37.
7. Guillevin L, Lhote F, Gayraud M, Cohen P, Jarrousse B, Lortholary O, Thibult N, Casassus P. Prognostic factors in polyarteritis nodosa and Churg-Strauss syndrome. A prospective study in 342 patients. Medicine (Baltimore). 1996;75:17-28.
8. Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL, Hagen EC, Hoffman GS, Hunder GG, Kallenberg CG, McCluskey RT, Sinico A, Rees AJ, van Es LA, Waldherr R, Wiik A. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum. 1994;37:187-192.
9. Lanham JG, Elkon KB, Pusey CD, Hughes GR. Systemic vasculitis with asthma and eosinophilia: a clinical approach to the Churg-Strauss syndrome. Medicine (Baltimore). 1984;63:65-81.
10. Mahrholdt H, Goedecke C, Wagner A, Meinhardt G, Athanasiadis A, Vogelsberg H, Fritz P, Klingel K, Kandolf R, Sechtem U. Cardiovascular magnetic resonance assessment of human myocarditis: a comparison to histology and molecular pathology. Circulation. 2004;109:1250-1258.
11. Masi AT, Hunder GG, Lie JT, Michel BA, Bloch DA, Arend WP, Calabrese LH, Edworthy SM, Fauci AS, Leavitt RY, Lightfoot RW Jr, McShan DJ, Mills JA, Stevens MB, Wallace SL, Zvaifler NJ. The American College of Rheumatology 1990 criteria for the classification of Churg-Strauss syndrome (allergic granulomatosis and angiitis). Arthritis Rheum. 1990;33:1094-1100.
12. Morgan JM, Raposo L, Gibson DG. Cardiac involvement in Churg-Strauss syndrome shown by echocardiography. Br Heart J. 1989;62:462-466.
13. Ogbogu PU, Rosing DR, Horne MK III. Cardiovascular manifestations of hypereosinophilic syndromes. Immunol Allergy Clin North Am. 2007;27:457-475.
14. Roufosse F, Cogan E, Goldman M. Lymphocytic variant hypereosinophilic syndromes. Immunol Allergy Clin North Am. 2007;27:389-413.
15. Sable-Fourtassou R, Cohen P, Mahr A, Pagnoux C, Mouthon L, Jayne D, Blockmans D, Cordier JF, Delaval P, Puechal X, Lauque D, Viallard JF, Zoulim A, Guillevin L. Antineutrophil cytoplasmic antibodies and the Churg-Strauss syndrome. Ann Intern Med. 2005;143:632-638.
16. Simon HU, Plotz S, Simon D, Seitzer U, Braathen LR, Menz G, Straumann A, Dummer R, Levi-Schaffer F. Interleukin-2 primes eosinophil degranulation in hypereosinophilia and Wells' syndrome. Eur J Immunol. 2003;33:834-839.
17. Simon HU, Plotz SG, Dummer R, Blaser K. Abnormal clones of T cells producing interleukin-5 in idiopathic eosinophilia. N Engl J Med. 1999;341:1112-1120.
18. Sinico RA, Di Toma L, Maggiore U, Bottero P, Radice A, Tosoni C, Grasselli C, Pavone L, Gregorini G, Monti S, Frassi M, Vecchio F, Corace C, Venegoni E, Buzio C. Prevalence and clinical significance of antineutrophil cytoplasmic antibodies in Churg-Strauss syndrome. Arthritis Rheum. 2005;52:2926-2935.
19. Smedema JP, Snoep G, van Kroonenburgh MP, van Geuns RJ, Dassen WR, Gorgels AP, Crijns HJ. Evaluation of the accuracy of gadolinium-enhanced cardiovascular magnetic resonance in the diagnosis of cardiac sarcoidosis. J Am Coll Cardiol. 2005;45:1683-1690.
20. Yelgec NS, Dymarkowski S, Ganame J, Bogaert J. Value of MRI in patients with a clinical suspicion of acute myocarditis. Eur Radiol. 2007;17:2211-2217.
© 2009 Lippincott Williams & Wilkins, Inc.