Int J Cardiovasc Imaging DOI 10.1007/s10554-016-0843-y

ORIGINAL PAPER

Detecting cardiac involvement with magnetic resonance in patients with active eosinophilic granulomatosis with polyangiitis Sehyo Yune1 • Dong-Chull Choi1 • Byung-Jae Lee1 • Jin-Young Lee2 Eun-Seok Jeon3 • Sung Mok Kim4 • Yeon Hyeon Choe4

•

Received: 24 September 2015 / Accepted: 19 January 2016 Ó Springer Science+Business Media Dordrecht 2016

Abstract Cardiac involvement is the most important prognostic factor in eosinophilic granulomatosis with polyangiitis (EGPA, Churg–Strauss syndrome). The aims of this study were to describe findings of cardiac magnetic resonance (CMR) in patients with active EGPA and to find factors associated with cardiac involvement detected by CMR that could help identify patients who would benefit from the examination. Medical records and CMR images in 16 consecutive EGPA patients (8 women and 8 men, median age of 47 years ranging from 34 to 68 years) were reviewed. Clinical features and results of laboratory tests were compared according to the presence of myocardial late gadolinium enhancement (LGE) on CMR images. The patients were followed for the development of cardiac symptoms and signs (mean follow up duration, 40.5 ± 12.8 months). Among the total of 16 patients, 8 (50 %) had myocardial LGE according to CMR, located in the subendocardial layer in 7 of them (87.5 %). The extent of LGE had a significant negative correlation with left ventricular ejection fraction (LVEF, q = -0.723, p = 0.043). The presence & Dong-Chull Choi [email protected]

of LGE was associated with larger end-systolic left ventricle internal dimension (34 vs. 28 mm, p = 0.027) and presence of diastolic dysfunction (75 vs. 0 %, p = 0.008) on echocardiography, elevated NT-proBNP (75 vs. 12.5 %, p = 0.012), and elevated CK-MB (62.5 vs. 0 %, p = 0.010) compared to the group without LGE. Only one patient (6.3 %) had cardiac symptoms before CMR and another patient (6.3 %) developed heart failure 4 years later during remission. The other 14 patients remained free from cardiac signs and symptoms during the follow-up period. In patients with active EGPA, CMR enables detection of cardiac involvement when cardiac symptoms are not present. Echocardiographic diastolic dysfunction and elevated NTproBNP or CK-MB may help identify active EGPA patients who can benefit from CMR to detect cardiac involvement without cardiac symptoms. Keywords Cardiac magnetic resonance  Eosinophilic granulomatosis with polyangiitis  Churg–Strauss syndrome  Vasculitis 1

Division of Allergy, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, South Korea

2

Center for Health Promotion, Samsung Medical Center, 81 Irwon-ro, Gangnam-gu, Seoul 06351, South Korea

3

Division of Cardiology, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, South Korea

4

Department of Radiology, Cardiovascular Imaging Center, Heart Vascular Stroke Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, South Korea

& Yeon Hyeon Choe [email protected] Sehyo Yune [email protected] Byung-Jae Lee [email protected] Jin-Young Lee [email protected] Eun-Seok Jeon [email protected] Sung Mok Kim [email protected]

123

Int J Cardiovasc Imaging

Introduction Eosinophilic granulomatosis with polyangiitis (EGPA, Churg–Strauss syndrome) is a rare form of systemic vasculitis first described by pathologists Churg and Strauss [1]. With an annual incidence of 0.5–3.7 cases per million individuals [2], EGPA is characterized by asthma, eosinophilia, and small-vessel vasculitis involving various organs [1, 3, 4]. The clinical features of EGPA vary according to the organs involved [5], although systemic symptoms such as fever, weight loss, myalgia, and arthralgia are common. Skin, paranasal sinuses, lungs, gastrointestinal tract, kidneys, and the peripheral and central nervous systems are also involved in EGPA to varying degrees. Cardiac involvement of EGPA can manifest as eosinophilic endomyocarditis, coronary vasculitis, valvular heart disease, congestive heart failure, and pericarditis [1, 6]. According to recent studies, the heart is involved in 45–62 % of all EGPA cases [7–9]. Being the leading cause of death, cardiac involvement is the most important prognostic factor in patients with EGPA [4, 5, 7]. Although early detection of cardiac involvement is critical in management of EGPA, there is no consensus with respect to cardiac evaluation in patients without cardiac symptoms or signs. Over the last decade, cardiac magnetic resonance (CMR) has emerged as a useful noninvasive method to detect structural and functional changes of the heart [10]. The use of CMR has become popular in ischemic and non-ischemic cardiomyopathies, with accumulating evidence suggesting its high sensitivity and prognostic value [11–14]. The diagnostic value of CMR in EGPA has been demonstrated in several studies [15–18]. In these studies, late gadolinium enhancement (LGE) was identified as a characteristic finding in cardiac involvement of EGPA, as well as decreased left ventricular ejection fraction (LVEF). Nevertheless, it remains unclear whether CMR is beneficial for patients with active vasculitis who have no cardiac symptoms or signs, as most of the previous studies have been conducted in patients in remission, or in those with other evidence of cardiac involvement. In the present study, we intended to evaluate the role of CMR in patients with active EGPA without cardiac symptoms by analyzing CMR data and its association with various clinical and laboratory features in newly diagnosed EGPA patients.

Methods Patients Thirty consecutive patients diagnosed with EGPA between 2009 and 2013 at our hospital were identified by

123

retrospective review of medical records. The diagnosis of EGPA was made with the presence of four or more of following criteria: (1) asthma, (2) peripheral eosinophilia [10 % on differential leukocyte count, (3) mononeuritis multiplex or polyneuropathy, (4) paranasal sinus abnormality, (5) migrating pulmonary infiltrates, and (6) biopsy evidence of extravascular eosinophils [19]. Following the routine diagnostic protocol, all patients had blood tests for complete blood cell count, serum creatinine, liver transaminases, C-reactive protein (CRP), and antimyeloperoxidase antineutrophil cytoplasmic antibody (pANCA), as well as pulmonary function tests, radiographs of the chest and the paranasal sinuses, electromyograms, and nerve conduction studies. Histologic evaluation was performed whenever a biopsy was feasible. Other diagnostic tests were added according to each patient’s clinical manifestation. Electrocardiography (ECG), echocardiography, and assays of N-terminal pro-brain natriuretic peptide (NT-proBNP), creatine kinase-MB fraction (CK-MB) and cardiac troponin I (cTnI) were performed for cardiac evaluation. The patients were followed for the development of cardiac symptoms and signs (mean follow up duration, 40.5 ± 12.8 months). CMR was also included in the diagnostic protocol, but only 16 patients agreed to undergo CMR examination. The demographic and clinical data from the 16 patients were subsequently analyzed. To assess disease activity, the Birmingham Vasculitis Activity Score (BVAS) was calculated in each patient at the time of CMR [20]. This study was approved by Institutional Review Board of Samsung Medical Center. Cardiac magnetic resonance imaging Cardiac magnetic resonance imaging was performed using a 1.5-T scanner (Magnetom Avanto; Siemens Medical Solutions, Erlangen, Germany) with a 32-channel phasedarray receiver coil during repeated breath-holds. After localization, cine images for left ventricular (LV) functional parameters were acquired using a steady-state freeprecession sequence with 8–10 contiguous short-axis slices to cover the entire LV with a slice thickness of 6 mm and a 4 mm gap. T2-weighted images were obtained in the shortaxis direction using a black-blood T2-weighted inversionrecovery fast-spin echo sequence. Standard delayed gadolinium-enhanced imaging was acquired with the phase-sensitive inversion recovery (PSIR) technique performed 10 and 15 min after injection of 0.15 mmol/kg gadobutrol (Gadovist; Bayer Healthcare, Berlin, Germany) at a rate of 3 mL/s with contiguous short-axis image acquisition of 10–12 slices of 6 mm thickness with a 4 mm interslice gap. The inversion times varied from 280 to 360 ms.

Int J Cardiovasc Imaging

CMR image analysis Qualitative CMR analyses were carried out by consensus agreement of two observers (3 and 20 years of experience in CMR, respectively) blinded to clinical information. A 17-segment model was used to localize and quantify the LGE and high T2 signal intensity (SI) [21]. LGE was detected visually and classified into subendocardial, midmyocardial, and epicardial enhancement. The extent of LGE was graded in each segment using a 5-point scale scored as follows: 0, no hyperenhancement; 1, hyperenhancement of 1–25 % of the myocardium; 2, hyperenhancement of 26–50 % of the myocardium; 3, hyperenhancement of 51–75 % of the myocardium; and 4, hyperenhancement of 76–100 % of the myocardium. The LGE score was calculated by summing up the points in all segments. Presence or absence of high T2 SI in each segment was evaluated concurrently. Echocardiography Three patients without LGE underwent echocardiography at other institutions, and were excluded when comparing echocardiographic features. The remaining 13 patients underwent comprehensive transthoracic echocardiography with commercially available echocardiography equipment (Vivid 7; General Electric Medical Systems, Horten, Norway). A routine standard echocardiography was performed in accordance with the recommendations of the American Society of Echocardiography (ASE) [22]. The LV end-systolic dimension, end-diastolic dimension, and left atrial volume were measured using a modified biplane Simpson’s method in apical two- and four-chamber views. Mitral inflow velocities and mitral annular velocities were assessed by pulsed-wave Doppler. Peak early (E) and late (A) diastolic velocities of mitral inflow were measured at the tip of mitral valve leaflets. Early (e0 ) and late (a0 ) diastolic mitral annular velocities were acquired at the septum in apical four-chamber views. Diastolic dysfunction was diagnosed based on the current recommendation of the ASE [23]. Electrocardiography A 12-lead surface ECG was recorded and interpreted by two experienced cardiologists. The ECG was classified as abnormal if it showed complete right or left bundle branch block, bifascicular block, third-degree atrioventricular block, ventricular tachycardia, or pathological Q or ST abnormalities. Statistical analysis Data are presented as mean ± standard deviation. For comparisons between the groups with and without CMR

abnormalities, categorical variables were assessed using the Chi square test and continuous variables were analyzed by Student’s t test. The correlations of the extent of LGE (LGE score and number of myocardial segments with LGE) with the levels of NT-proBNP, cTnI, CK-MB and LVEF were assessed using the Spearman rank correlation test. All data were analyzed by SPSS statistics 21.0 for windows (SPSS Inc., Chicago, IL, USA). P values less than 0.05 were considered statistically significant.

Results CMR findings were normal in 8 patients, while the other 8 (50 %) had LGE lesions that were not compatible with coronary artery territories. The comparison between the groups with and without LGE on CMR is presented in Table 1. Table 2 shows the CMR findings of the patients with LGE. LGE was located in the subendocardial layer in 6 patients, the subepicardial layer in one patient, and extensively in different layers in one patient (case 8). High T2 SI was observed in 2 patients. Functional parameters were also abnormal in many of these patients. Specifically, LV enddiastolic volume (EDV), LV end-systolic volume (ESV), and LV ejection fraction (EF) were outside of reference range for 2, 5, and 6 patients, respectively [24]. There was no segment spared from LGE. Segment 10 was the most involved, being positive in 6 patients, while segment 16 was the least involved, being positive in 2 patients. Figure 1 shows the extensive LGE in the CMR image from case 1. Table 3 shows the results of cardiac evaluation in patients with LGE on CMR. Patient 8 was the only patient who experienced dyspnea associated with bilateral pleural effusion and pericardial effusion that disappeared completely within a few days of steroid treatment, 2 weeks before the time of CMR. The other patients in this study did not have any cardiac symptoms or signs. Five patients received treatment for EGPA before CMR examinations. The only significant correlation was seen between the number of LGE segments and the LVEF (q = -0.723, p = 0.043). Although the correlation between the extent of LGE and the level of NT-proBNP was not statistically significant (q = 0.611, p = 0.108 between the LGE score and NT-proBNP, and q = 0.683, p = 0.062 between the number of LGE segments and NT-proBNP), the two patients whose NT-proBNP levels were within the normal range (\222 pg/mL) had the lowest extent of LGE both in terms of LGE scores and the number of LGE segments. Diastolic function was not determined by echocardiography at the time of CMR in two patients because of sinus tachycardia that resulted in the fusion of E and A waves. For case 2, echocardiography performed 2 months later revealed grade 2 diastolic dysfunction. In case 3, the only

123

Int J Cardiovasc Imaging Table 1 Comparison of patients with and without LGE on CMR

LGE present (n = 8)

LGE absent (n = 8)

p value 0.228

Age (years)

51 ± 12.8

43 ± 12.6

Male/Female

4/4

4/4

N/A

Treatment duration prior to CMR (days)

91.5 ± 172.2

255.1 ± 315.8

0.219

BVAS

16.9 ± 8.3

8.5 ± 7.3

0.051

Cardiac symptoms during follow-up

1 (12.5 %)

0 (0 %)

0.302

EGPA manifestation Asthma

8 (100 %)

8 (100 %)

N/A

Pulmonary infiltrates

3 (37.5 %)

7 (87.5 %)

0.039

Paranasal sinusitis

7 (87.5 %)

6 (75.0 %)

0.522

Skin

3 (37.5 %)

5 (71.4 %)

0.189

Peripheral nerve

6 (75.0 %)

6 (75.0 %)

N/A

Central nervous system

1 (12.5 %)

0 (0 %)

0.302

Kidney

1 (12.5 %)

0 (0 %)

0.302

Gastrointestinal tract

1 (12.5 %)

0 (0 %)

Treatment before CMR None

0.302 0.179

3 (37.5 %)

0 (0 %)

Corticosteroid

2 (25.0 %)

3 (37.5 %)

Immunosuppressive agents

3 (37.5 %)

5 (62.5 %)

Laboratory findings Positive p-ANCA

2 (25 %)

0 (0 %)

0.131

Eosinophil count

5507 ± 9378

648 ± 1040

0.188

Elevated NT-proBNPa

6 (75.0 %)

1 (12.5 %)

0.012

Elevated CK-MBa

5 (62.5 %)

0 (0 %)

0.010

Elevated cTnIa

2 (25.0 %)

0 (0 %)

0.155

1 (12.5 %)

0 (0 %)

0.302

Abnormal ECG

BVAS Birmingham vasculitis activity score, CK-MB creatine kinase-MB fraction, CMR cardiac magnetic resonance, cTnI cardiac troponin I, EF ejection fraction, ECG electrocardiogram, LGE late gadolinium enhancement, LV left ventricle, N/A not applicable or assessed, NT-proBNP N-terminal pro-brain natriuretic peptide, p-ANCA perinuclear anti-neutrophil cytoplasmic antibody a

Reference ranges: NT-proBNP, 0–222 pg/mL; CK-MB, 0–5 ng/mL; cTnI, 0–0.78 ng/mL

abnormal echocardiographic finding was a dilated left coronary artery os, which was attributed to multiple aneurysms of the left anterior descending artery. The aneurysmal change of coronary artery was thought to be a manifestation of EPGA, because the patient had multiple aneurysms in the brachial and popliteal arteries that were proven to be eosinophilic vasculitis by biopsy. Endomyocardial biopsy was carried out in three cases (cases 1, 2 and 8). Active eosinophilic myocarditis was confirmed in cases 1 and 2, while only subtle focal fibrosis was noted in case 8. High T2 SI was extensive in case 8 (in segments 1, 4, 5, 7, 10, 11, 12, 15, 16 and 17), but subtle in case 2 (only in segment 4) and absent in case 1 despite histologically proven active inflammation. Case 8 received corticosteroid and cyclophosphamide treatment before the biopsy and CMR, whereas cases 1 and 2 did not receive treatment by the time of biopsy and CMR. Coronary angiography was performed in case 3 to precisely evaluate

123

the aneurysmal change of coronary artery. No other patients underwent coronary angiography. Significant differences between patients with and without LGE on CMR were noted for frequency of pulmonary infiltrates as well as elevation of NT-proBNP and CK-MB (Table 1). In one patient with elevated NT-proBNP in the LGE-absent group, NT-proBNP was 248.2 pg/ml, slightly exceeding the normal limit. The levels of CK-MB and cTnI and the result of ECG and echocardiography were all within the normal range in the LGE-absent group. The BVAS was higher in the LGE-present group without statistical significance (p = 0.051). Lastly, with the exception of one patient in remission (BVAS 0) who did not show LGE on CMR, all patients had a BVAS of 5 or higher. The most frequent echocardiographic finding in patients with LGE was diastolic dysfunction (6, 75 %). This was followed by regional wall motion abnormalities in 3 patients (37.5 %), pericardial effusion in 2 patients (25 %),

Int J Cardiovasc Imaging Table 2 CMR findings in patients with LGE Case

Age

Sex

LGE score

Number of LGE segments

High T2 SI

LGE segments

LVEDV (mL/m2)

LVESV (mL/m2)

LVEF (%)

1

47

F

27

15

-

1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15, 16, 17: subendocardial

109.9a

65.8a

34.8a

2

39

M

36

17

?

All segments: subendocardial

98.6

73.6a

25.3a

84.5

a

52.1a

a

43.3

51.9a

3

34

M

4

1

-

4: subendocardial

40.4 a

4

68

F

15

9

-

2, 3, 7, 8, 9, 10, 11, 12, 14, 17: subendocardial

90.0

5

43

M

29

12

-

1, 2, 4, 5, 6, 8, 9, 10, 13, 14, 15, 17: subendocardial

59.3

25.1

57.6a

6

62

F

4

1

-

5: subepicardial

70.2

24.8

47.7a

7

66

M

9

5

-

4, 6, 7, 9, 10: subendocardial

96.7

36.0a

62.8

8

49

F

40

13

?

1, 2, 4, 6, 7, 8, 10, 12, 17: subendocardial

54.8

17.1

68.7

5: subepicardial 13, 14, 15: mid-wall CMR cardiac magnetic resonance, LGE late gadolinium enhancement, LVEDV left ventricular end-diastolic volume, LVEF left ventricular ejection fraction, LVESV left ventricular end-systolic volume, SI signal intensity a

Values out of reference range: LVEDV \99 mL/m2 in men and \80 mL/m2 in women, LVESV \34 mL/m2 in men and \26 mL/m2 in women, LVEF [60 % in men and 63 % in women [24]

All of the patients with abnormal CMR findings underwent intensive cytotoxic treatment for 12 months or longer. A patient (case 4) initially had no cardiac symptoms and was positive for p-ANCA, but developed heart failure 4 years later during remission with maintenance corticosteroids following 12-month cyclophosphamide treatment. Because there were no signs of EGPA relapse or other causes that could result in heart failure, these events were considered to be sequelae of the cardiac involvement. Another patient (case 8) recovered from symptoms of acute heart failure before the time of CMR, and never developed overt symptoms again. The other 14 patients remained free from cardiac signs and symptoms during the follow-up period (mean 40.5 months).

Discussion

Fig. 1 An example of cardiac involvement detected with CMR in a 47-year-old female (case 1). Late gadolinium enhancement is seen in the subendocardial layer of anterior, inferolateral and inferior myocardial wall on a short-axis view CMR image (indicated with white arrows)

and moderate degree of mitral regurgitation in 1 patient (12.5 %) in the LGE-present group, whereas there were no significant changes observed in the LGE-absent group. LV systolic dimension was significantly larger in the LGEpresent group (34 mm in LGE-present group vs. 28 mm in LGE-absent group, p = 0.027).

This is the first study to evaluate the use of CMR as part of the initial diagnostic protocol in patients with active vasculitis. Although the exact incidence of cardiac involvement cannot be drawn from this study, cardiac involvement were detected by CMR in 50 % of 16 patients who underwent CMR, with the majority in the active phase of EGPA without cardiac symptoms. All the 8 patients without cardiac involvement detected by CMR never developed overt cardiac symptoms over the follow up period ranging from 2 to 5 years. The CMR findings in this study consisted primarily of LGE, mostly in the subendocardium, followed by increased LV cavity volume and decreased LVEF. Previous studies

123

Int J Cardiovasc Imaging Table 3 Clinical features and laboratory results in patients with LGE Case

Age

Sex

Treatment prior to CMR (days)

NTproBNP (pg/mL)

cTnI (ng/mL)

CK-MB (ng/mL)

Echocardiography LVEF (%)

Diastolic dysfunction

RWMA

ECG

Biopsy

1

47

F

None

5455

3.020

1.28

42

Grade 2

-

T wave inversion

Eosinophilic myocarditis

2

39

M

None

3149

5.791

11.99

43

Indeterminate

?

Normal

Eosinophilic myocarditis

3

34

M

None

4

68

F

Corticosteroid (41)

5

43

M

Corticosteroid, cyclophosphamide, and azathioprine (152)

147

0.007

1.16

56

Indeterminate

-

Normal

N/D

9478

0.123

7.69

50

Grade 1

?

Normal

N/D

813

0.048

6.11

45

Grade 1

?

Normal

N/D

6

62

F

Corticosteroid (499)

43

0.014

2.59

71

Grade 2

-

Normal

N/D

7

66

M

Corticosteroid and cyclophosphamide (23)

284

0.322

7.83

58

Grade 2

-

Normal

N/D

8

49

F

Corticosteroid and cyclophosphamide (17)

3494

0.038

6.59

58

Grade 1

-

Normal

Subtle focal fibrosis without eosinophilic infiltration

APB atrial premature beat, CMR cardiac magnetic resonance, CK-MB creatine kinase-MB fraction, ECG electrocardiogram, LAE left atrial enlargement, LGE late gadolinium enhancement, LVEF left ventricular ejection fraction, N/D not done, NT-proBNP N-terminal pro-brain natriuretic peptide, RWMA regional wall motion abnormality, VPB ventricular premature beat Reference ranges; 0–222 pg/mL for NT-proBNP, 0–0.78 ng/ml for cTnI, and 0–5 ng/mL for CK-MB

[9, 18, 25] have shown similar CMR findings, although some studies have reported that the middle myocardial layer is the most common area involved with LGE [15, 26]. Similarly, a larger extent of LGE associated with lower LVEF was observed in a previous study [17]. Contrast enhancement in CMR implies disrupted myocytes in acute myocardial damage or increased extracellular space in the setting of chronic fibrosis [27]. This kind of change may be seen in any kind of cardiac diseases, such as myocardial infarction, myocarditis, myocardial infiltrative disease, and hypertrophic cardiomyopathies [28]. Subendocardial LGE is characteristic of ischemic heart disease, and therefore coronary diseases should be ruled out when it is observed. Although performing coronary artery angiography may be justified in ambiguous cases, ischemic and non-ischemic cardiomyopathies can be distinguished by experienced radiologists [29]. In the present study, echocardiography revealed various features in patients exhibiting LGE on CMR, with diastolic dysfunction being the most frequent finding. In the previous studies, echocardiography in EGPA patients identified decreased systolic function, wall motion abnormality, valvular insufficiency, pericardial effusion, pulmonary hypertension and impaired ventricular relaxation [8, 9, 30, 31]. Unfortunately, these various nonspecific echocardiographic changes often confuse clinicians. Indeed, a mild

123

degree of valvular insufficiency and diastolic dysfunction can be found among the general population, especially in the elderly [32, 33]. Therefore, minor echocardiographic change might be a clue, but cannot be sufficient for cardiac involvement of EGPA. Thus, patients with diastolic dysfunction, increased LV cavity size, or decreased LVEF in echocardiography should be considered as candidates of CMR examinations. In this study, one patient (case 8) who received corticosteroid treatment prior to CMR underwent endomyocardial biopsy only to find subtle interstitial fibrosis. Echocardiography in this patient only showed grade 1 diastolic dysfunction despite prior symptoms of heart failure. However, the patient had extensive LGE and high T2 SI on CMR. These findings may suggest that the histologic and echocardiographic changes rapidly reverse after treatment, while changes in CMR persist. Despite efforts to determine the acuity of the LGE lesions seen in CMR, high T2-signals and positron-emission tomography (PET) showed limited capability to distinguish fibrosis from active inflammation in EGPA [15]. In this study, elevated cTnI was observed only in patients with biopsy-proven eosinophilic myocarditis. However, as seen in case 4, a patient with normal cTnI and no T2 SI can later develop heart failure. Consistently, there was a case report describing that cardiomyopathy in EGPA was

Int J Cardiovasc Imaging

reversed by cytotoxic treatment after a year of corticosteroid treatment [34]. Although chronic fibrosis and active inflammation cannot be distinguished with CMR, the presence of LGE and high T2 SI should prompt careful consideration for more intensive treatment to prevent development of chronic heart failure. ANCA was negative in all patients without LGE and in 6 patients with LGE. In contrast to previous studies [8, 9, 16], ANCA negativity was not associated with cardiac involvement of EGPA. Notably, pulmonary infiltrates were more frequent in patients with normal CMR findings. There is limited data regarding the association between the frequency of pulmonary infiltrate and that of cardiac involvement in EGPA, although one study showed that negative ANCA is associated with increased cardiac involvement and lung disease [35], contradicting the result of the present study. Confirmative diagnosis of cardiac involvement in EGPA can be made by revealing eosinophilic infiltration in the myocardial interstitium or fibrosis of endocardium by histologic examination [1, 9]. Endomyocardial biopsy may be useful in identifying these changes, but considering the invasiveness of this approach, it is of limited use, especially in asymptomatic patients. Moreover, a previous study showed that endomyocardial biopsy might miss focal inflammation even in clinically evident myocarditis [36], as it represents only a small portion of the myocardium. In the cohort evaluated in our study, all patients with LGE on CMR also exhibited abnormal echocardiography results. Nevertheless, other studies have reported patients with LGE on CMR with normal echocardiography and ECG [8, 15]. Cardiac involvement of EGPA can be diagnosed using ECG, echocardiography, 24-h ECG Holter monitoring, endomyocardial biopsy, and CMR, of which CMR seems to be the most sensitive method [8, 9, 17]. Elevated NT-proBNP in patients with LGE has been observed in other studies [17, 26], but elevated CK-MB was observed only in our study. In this way, elevated CKMB appeared to reflect the active status of our patients. Since this study was retrospective in design, the data was not complete. In addition, the small number of patients is the major weakness shared with many other studies on EGPA, and may explain the inconsistency among studies, in which the number of subjects rarely exceeds 20. Lastly, there might have been selection bias in this study because not all EGPA patients underwent CMR.

Conclusions Cardiac involvement can be detected with CMR in patients in active phase of EGPA with no cardiac symptoms or signs at the time of the examination. The typical CMR

finding is LGE in the subendocardial layer. Elevated NTproBNP level and diastolic dysf-unction in echocardiography are the distinguishing features of patients with LGE. The findings of the present study may assist clinicians at interpreting CMR results in active EGPA patients, and deciding which patients should undergo CMR to detect cardiac involvement. Compliance with ethical standards Conflict of interest

None.

References 1. Churg J, Strauss L (1951) Allergic granulomatosis, allergic angiitis, and periarteritis nodosa. Am J Pathol 27(2):277–301 2. Gibelin A, Maldini C, Mahr A (2011) Epidemiology and etiology of Wegener granulomatosis, microscopic polyangiitis, Churg– Strauss syndrome and goodpasture syndrome: vasculitides with frequent lung involvement. Semin Respir Crit Care Med 32(3):264–273 3. Jennette JC, Falk RJ (1997) Small-vessel vasculitis. N Engl J Med 337(21):1512–1523 4. Lanham JG, Elkon KB, Pusey CD et al (1984) Systemic vasculitis with asthma and eosinophilia: a clinical approach to the Churg– Strauss syndrome. Medicine (Baltimore) 63(2):65–81 5. Guillevin L, Cohen P, Gayraud M et al (1999) Churg–Strauss syndrome. Clinical study and long-term follow-up of 96 patients. Medicine (Baltimore) 78(1):26–37 6. Ramakrishna G, Midthun DE (2001) Churg–Strauss syndrome. Ann Allergy Asthma Immunol 86(6):603–613 (quiz 613) 7. Moosig F, Bremer JP, Hellmich B et al (2013) A vasculitis centre based management strategy leads to improved outcome in eosinophilic granulomatosis and polyangiitis (Churg–Strauss, EGPA): monocentric experiences in 150 patients. Ann Rheum Dis 72(6):1011–1017 8. Dennert RM, van Paassen P, Schalla S et al (2010) Cardiac involvement in Churg–Strauss syndrome. Arthritis Rheum 62(2):627–634 9. Neumann T, Manger B, Schmid M et al (2009) Cardiac involvement in Churg–Strauss syndrome: impact of endomyocarditis. Medicine (Baltimore) 88(4):236–243 10. Motwani M, Kidambi A, Greenwood JP et al (2014) Advances in cardiovascular magnetic resonance in ischaemic heart disease and non-ischaemic cardiomyopathies. Heart 100(21):1722–1733 11. Patel MR, Cawley PJ, Heitner JF et al (2009) Detection of myocardial damage in patients with sarcoidosis. Circulation 120(20):1969–1977 12. Yoshida A, Ishibashi-Ueda H, Yamada N et al (2013) Direct comparison of the diagnostic capability of cardiac magnetic resonance and endomyocardial biopsy in patients with heart failure. Eur J Heart Fail 15(2):166–175 13. Kassi M, Nabi F (2013) Role of cardiac MRI in the assessment of nonischemic cardiomyopathies. Methodist Debakey Cardiovasc J 9(3):149–155 14. Neilan TG, Coelho-Filho OR, Danik SB et al (2013) CMR quantification of myocardial scar provides additive prognostic information in nonischemic cardiomyopathy. JACC Cardiovasc Imaging 6(9):944–954 15. Marmursztejn J, Guillevin L, Trebossen R et al (2013) Churg– Strauss syndrome cardiac involvement evaluated by cardiac

123

Int J Cardiovasc Imaging

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

magnetic resonance imaging and positron-emission tomography: a prospective study on 20 patients. Rheumatology (Oxford) 52(4):642–650 Mavrogeni S, Karabela G, Gialafos E et al (2013) Cardiac involvement in ANCA (?) and ANCA (-) Churg–Strauss syndrome evaluated by cardiovascular magnetic resonance. Inflamm Allergy Drug Targets 12(5):322–327 Szczeklik W, Miszalski-Jamka T, Mastalerz L et al (2011) Multimodality assessment of cardiac involvement in Churg–Strauss syndrome patients in clinical remission. Circ J 75(3):649–655 Wassmuth R, Gobel U, Natusch A et al (2008) Cardiovascular magnetic resonance imaging detects cardiac involvement in Churg–Strauss syndrome. J Card Fail 14(10):856–860 Masi AT, Hunder GG, Lie JT et al (1990) The American College of Rheumatology 1990 criteria for the classification of Churg– Strauss syndrome (allergic granulomatosis and angiitis). Arthritis Rheum 33(8):1094–1100 Flossmann O, Bacon P, de Groot K et al (2007) Development of comprehensive disease assessment in systemic vasculitis. Ann Rheum Dis 66(3):283–292 Cerqueira MD, Weissman NJ, Dilsizian V et al (2002) Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 105(4):539–542 Picard MH, Adams D, Bierig SM et al (2011) American Society of Echocardiography recommendations for quality echocardiography laboratory operations. J Am Soc Echocardiogr 24(1):1–10 Nagueh SF, Appleton CP, Gillebert TC et al (2009) Recommendations for the evaluation of left ventricular diastolic function by echocardiography. J Am Soc Echocardiogr 22(2):107–133 Chuang ML, Gona P, Hautvast GL et al (2014) CMR reference values for left ventricular volumes, mass, and ejection fraction using computer-aided analysis: the Framingham Heart Study. J Magn Reson Imaging 39(4):895–900 Bhagirath KM, Paulson K, Ahmadie R et al (2009) Clinical utility of cardiac magnetic resonance imaging in Churg–Strauss syndrome:

123

26.

27.

28.

29.

30. 31.

32.

33.

34.

35.

36.

case report and review of the literature. Rheumatol Int 29(4):445–449 Marmursztejn J, Vignaux O, Cohen P et al (2009) Impact of cardiac magnetic resonance imaging for assessment of Churg– Strauss syndrome: a cross-sectional study in 20 patients. Clin Exp Rheumatol 27(1 Suppl 52):S70–S76 Mahrholdt H, Wagner A, Judd RM et al (2002) Assessment of myocardial viability by cardiovascular magnetic resonance imaging. Eur Heart J 23(8):602–619 Hundley WG, Bluemke DA, Finn JP et al (2010) ACCF/ACR/ AHA/NASCI/SCMR 2010 expert consensus document on cardiovascular magnetic resonance: a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents. Circulation 121(22):2462–2508 Cummings KW, Bhalla S, Javidan-Nejad C et al (2009) A patternbased approach to assessment of delayed enhancement in nonischemic cardiomyopathy at MR imaging. Radiographics 29(1):89–103 Pela G, Tirabassi G, Pattoneri P et al (2006) Cardiac involvement in the Churg–Strauss syndrome. Am J Cardiol 97(10):1519–1524 Morgan JM, Raposo L, Gibson DG (1989) Cardiac involvement in Churg–Strauss syndrome shown by echocardiography. Br Heart J 62(6):462–466 Devereux RB, Hawkins I, Kramer-Fox R et al (1986) Complications of mitral valve prolapse. Disproportionate occurrence in men and older patients. Am J Med 81(5):751–758 Kuznetsova T, Herbots L, Lopez B et al (2009) Prevalence of left ventricular diastolic dysfunction in a general population. Circ Heart Fail 2(2):105–112 Lin YC, Oliveira GH, Villa-Forte A (2013) Churg–Strauss syndrome and persistent heart failure: Active disease or damage? J Clin Rheumatol 19(7):390–392 (discussion 419) Sinico RA, Di Toma L, Maggiore U et al (2005) Prevalence and clinical significance of antineutrophil cytoplasmic antibodies in Churg–Strauss syndrome. Arthritis Rheum 52(9):2926–2935 Mahrholdt H, Goedecke C, Wagner A et al (2004) Cardiovascular magnetic resonance assessment of human myocarditis: a comparison to histology and molecular pathology. Circulation 109(10):1250–1258