Scholarly article on topic 'Does magnetic resonance imaging have a role in diagnosing benign intracardiac lesions?'

Does magnetic resonance imaging have a role in diagnosing benign intracardiac lesions? Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Benign cardiac tumors" / "Hydatid disease" / "Cardiac thrombi" / "Cardiac magnetic resonance" / "Cardiac masses"}

Abstract of research paper on Clinical medicine, author of scientific article — Noha H. Behairy, Sherif Osama El Sayed Gouda

Abstract Objective In this study, transthoracic echocardiography and cardiac magnetic resonance were compared in terms of the detection, localization, characterization and extra cardiac association of cardiac lesions to specify the value of adding CMR to TTE in diagnosing benign intracardiac lesions. Materials and methods Twenty nine patients ranging in age between 7months and 80years were enrolled in the study. Twenty five patients with suspected intracardiac lesions by TTE, in addition four patients with accidentally discovered thrombi on CMR were added to the study. Results CMR detected 17 non neoplastic lesions and 11 neoplastic lesions. CMR changed the diagnosis made by TTE in five cases, showed additional findings to TTE in three cases and was superior to TTE in detecting four cases of cardiac thrombi. Conclusion CMR is superior to TTE in characterization of intracardiac lesions, detecting its site, size, vascularity, hemodynamics and multiplicity of the lesion with high tissue contrast. Thus it should be performed in all cases suspicious of cardiac masses as it could aid in patient management and guide for surgery. Moreover it can prevent unneeded cardiac surgery in cases of benign non-neoplastic lesions or in benign tumors with no hemodynamic significance.

Academic research paper on topic "Does magnetic resonance imaging have a role in diagnosing benign intracardiac lesions?"

The Egyptian Journal of Radiology and Nuclear Medicine (2013) 44, 167-174

Egyptian Society of Radiology and Nuclear Medicine The Egyptian Journal of Radiology and Nuclear Medicine

www.elsevier.com/locate/ejrnm www.sciencedirect.com

ORIGINAL ARTICLE

Does magnetic resonance imaging have a role in diagnosing benign intracardiac lesions?

Noha H. Behairy a *, Sherif Osama El Sayed Gouda b

a Department of Radiology, Kasr El Aini Hospital, Cairo University, Cairo, Egypt b Department of Cardiology, Kasr El Aini Hospital, Cairo University, Egypt

Received 9 October 2012; accepted 6 February 2013 Available online 13 March 2013

KEYWORDS

Benign cardiac tumors; Hydatid disease; Cardiac thrombi; Cardiac magnetic resonance; Cardiac masses

Abstract Objective: In this study, transthoracic echocardiography and cardiac magnetic resonance were compared in terms of the detection, localization, characterization and extra cardiac association of cardiac lesions to specify the value of adding CMR to TTE in diagnosing benign intracardiac lesions.

Materials and methods: Twenty nine patients ranging in age between 7 months and 80 years were enrolled in the study. Twenty five patients with suspected intracardiac lesions by TTE, in addition four patients with accidentally discovered thrombi on CMR were added to the study. Results: CMR detected 17 non neoplastic lesions and 11 neoplastic lesions. CMR changed the diagnosis made by TTE in five cases, showed additional findings to TTE in three cases and was superior to TTE in detecting four cases of cardiac thrombi.

Conclusion: CMR is superior to TTE in characterization of intracardiac lesions, detecting its site, size, vascularity, hemodynamics and multiplicity of the lesion with high tissue contrast. Thus it should be performed in all cases suspicious of cardiac masses as it could aid in patient management and guide for surgery. Moreover it can prevent unneeded cardiac surgery in cases of benign non-neoplastic lesions or in benign tumors with no hemodynamic significance.

© 2013 Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V.

All rights reserved.

* Corresponding author. Address: 47, Street 199, Apartment 13, Degla, Maadi, Cairo, Egypt. Tel.: +20 225203545/20 1 227781598. E-mail address: nohabehairy@gmail.com (N.H. Behairy). Peer review under responsibility of Egyptian Society of Radiology and Nuclear Medicine.

1. Introduction

Primary cardiac neoplasms are rare, occurring in only 0.0010.03% of patients in autopsy series. In all age groups, benign primary cardiac neoplasms are more common than malignant ones (1).

It is estimated that primary cardiac neoplasms are 100-1,000 times less prevalent than secondary neoplasm of the heart. The most common primary cardiac neoplasm is myxoma, which accounts for approximately half of all cases. Other benign primary tumors include papillary fibroelastoma

0378-603X © 2013 Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V. All rights reserved. http://dx.doi.Org/10.1016/j.ejrnm.2013.02.003

(the most common valvular tumor), rhabdomyoma, fibroma, hemangioma, and lipoma (1).

The diagnosis and management of primary cardiac neoplasm have been greatly facilitated by the development of non-invasive cardiac imaging. Transthoracic echocardiography (TTE) is the primary modality for imaging intracardiac disease. It provides high-resolution and real-time images (2). However, TTE suffers from several well-described limitations: restricted field of view; incomplete assessment of an invading cardiac mass due to an unfavorable patient body habits; and limited ability to perform tissue characterization (3).

A number of benign masses and normal anatomical variants can cause confusion to the inexperienced observer and must be recognized to avoid unnecessary intervention. These include intracardiac thrombus, bronchogenic and pericardial cysts, and anatomical structures, such as the Crista terminalis and moderator band (4).

Cardiac magnetic resonance (CMR) provides a noninvasive and three-dimensional assessment of masses involving the cardiac chambers, the pericardium, and the extracardiac structures. Therefore, it has become an established method for the yielding of complimentary diagnostic information and to guide surgeons in the design of an appropriate therapeutic strategy. Furthermore, magnetic resonance imaging has a soft tissue contrast superior to that of TTE allowing characterization of tumor tissue (5,6).

In this study, TTE and CMR were compared in terms of the detection, localization, characterization and extra cardiac association of cardiac lesions to specify the value of adding CMR to TTE in diagnosing benign intracardiac lesions.

2. Materials and methods

Twenty nine patients ranging in age between 7 months and 80 years (17 males and 12 females) were enrolled in the study. Twenty five patients diagnosed as having intracardiac lesions by TTE were prospectively studied. In addition four patients with intracardiac thrombi missed on TTE and detected by CMR were added to the study.

Eight were in the pediatric age group and the other twenty one were adults. The study was conducted over one year starting from May 2011 to May 2012. All patients were examined by history taking, chest X-ray and clinical examination. CMR was done within one week from TTE.

The study was approved by the ethics committee and verbal consent was obtained from the patients or their guardians.

2.1. Clinical findings

Patients experienced variable symptoms and signs of: dyspnea, murmur, respiratory distress, myocardial dysfunction, and congestive heart failure.

2.2. Transthoracic echocardiography

Two dimensional TTE was performed using Hewlett Packard (sonos 4500) with 8 or 4 mHz transducers for the pediatric age group and 2 mHz for the adult group. The site, size, mobility, multiplicity and sonographic texture of masses were documented.

2.3. Magnetic resonance imaging

All patients were imaged by a Philips Gyroscan Intera 1.5 tesla super conducting magnet using the cardiac coil and ECG triggering.

Patients were examined in supine position; ECG leads and respiratory trigger were applied and adjusted until a good signal is obtained. All patients were in sinus rhythm during the examination. Children under 5 years were sedated using a regular dose of chloral hydrate which was well tolerated with no reported side effects. Endotracheal intubation was not needed in any case.

2.3.1. CMR protocol

Multiple stack surveys providing coronal, transverse and sagittal images with consequent transverse, sagittal and coronal views were taken in the following sequences:

Axial, sagittal and coronal T1 weighted images with TR/ TE = 600-750/25-30, Axial and coronal T2 weighted images with TR/TE = 1300-1700/75-85, 156 x 256 matrix. Fat suppression images were obtained in cases of suspected fatty lesions. Gradient echo sequences bFFE in axial and coronal views were performed with TR/TE = 3.2/1.7, 215-256 x 256 matrix and 50° flip angle, 5-8 mm slice thickness, 10-14 slice numbers encompassing the heart and mediastinum with 300350 mm FOV according to the body dimensions was performed for all patients. Triple IR was done in cases with suspected myocardial abnormalities.

Post enhancement with intravenous gadolinium-based contrast material was performed for all patients (0.2 mm/kg) in transverse, sagittal and/or coronal views.

2.3.2. Evaluation and interpretation of cardiac magnetic resonance images

The site, size, mobility, multiplicity, extension and tissue characterization by studying the signal intensity of the mass at different sequences were documented.

Lesions were categorized as neoplastic and non-neoplastic. The tissue composition of a lesion was estimated based on the signal characteristics and qualitatively scored as homogenous or heterogeneous. In addition, in each sequence lesion signal intensity was assessed qualitatively as hypo-, iso-, or hyperin-tense with reference to the normal myocardium. Vascularity of the lesion was determined by the degree and pattern of contrast enhancement.

Extracardiac structures such as the lungs, mediastinum, great vessels, dorsal vertebrae and thoracic cage were visualized to detect any associated condition.

3. Results

Twenty-five patients diagnosed as having intracardiac lesion by TTE were studied in addition to four patients with myocar-dial abnormalities and accidentally discovered thrombi by CMR.

Chest X-ray demonstrated cardiomegaly, pericardial effusion, and signs of heart failure while some showed no abnormal findings. In one case of interventicular septum (IVS) fibroma there were associated vascular and vertebral anomalies.

Table 1 Site and characters of intra cardiac benign neoplastic lesions by echocardiography and MRI.

Echocardiographic findings MRI findings Diagnosis No Site of the mass

T1wIs T2wIs Contrast Special

Hyperechoic Iso High + + Rhabdomyomas 5 LV (n = 3), RV(n = 2)

Narrow stalk, hypoechoic Iso Hetero + + Mobile with low Myxomas 2 LA

signal on bFFE

Echogenic-heteroenous Iso intramural Low-- Fibromas 2 IVS

Hypoechoic mass High High-- —in fat supp. Lipoma 1 RA

Elongated, mobile Iso Low — —bFFE Fibroelastoma 1 Pulmonary valve

*Hetero = Heterogenous, iso = Isointense, IVS = Interventricular septum, LA = Left atrium, LV = Left ventricle ,RA = Right atrium,

RV = Right ventricle, bFFE = Balanced fast field echo.

Table 2 Site and characters of intra cardiac benign non-neoplastic lesions by echocardiography and MRI.

Echocardiographic findings NO MRI findings T1wIs T2wIs Contrast Special Diagnosis No Site of the mass

Hypoechoic 11 Iso Iso-low No Dark signal on Thrombus 15 LV (n = 6), RV(n = 4) RA (n = 3), LA (n = 2)

bFFE and IR

Hypoechoic cyst 2 Low High No Hydatid 2 IVS

TTE detected the site, size and number of 25 intracardiac lesions with no adequate tissue characterization of the lesions. Nevertheless, TTE missed four intracardiac thrombi which were associated with dysfunctional myocardium secondary to myocardial infarction in three cases and myocardial infiltration in a case of Loffler's eosinophilia.

CMR detected the size, site, and vascularity of 28 intracar-diac lesions and characterized them into neoplastic and non-neoplastic lesions. CMR detected 11 cases of benign tumors including: rhabdomyoma (n = 5), fibroma (n = 2), myxoma (n = 2), lipoma (n = 1) and fibroelastoma (n = 1) (Table 1).

Benign tumors were characteristically homogenous, well defined with smooth edges and no evidence of infiltration to the surrounding.

Seventeen cases of non tumoral lesions were diagnosed by CMR including 15 cases of intracardiac thrombi and two cases of hydatid cysts (Table 2).

Intracardiac thrombi were diagnosed on CMR by showing isointense signal on T1, low signal on T2 weighted images, characteristically low signal on both gradient echo and inversion recovery sequences with no evident contrast enhancement (Fig. 1). Three of these thrombi were mobile as two were attached to the chorde tendini of the left ventricle (LV) and one was attached to the interatrial septum.

Thrombi were of different sizes and locations; four were found in the right ventricle, six in the left ventricle, three in the right atrium and two within the left atrium (Table 2).

Three of the LV thrombi were on top of apical infarction with CMR showing delayed enhancement, thinning on the wall and akinesia of the anteroseptal wall and apex (Fig. 2).

Two of the RV thrombi were patients suffering from Beh-get's disease. All cases of right atrial thrombi were either propagated from a previous IVC thrombus or recurrent due to a systemic embolic condition.

One of the LA thrombi was a small thrombus lying on top of an enhanced myxoma.

Two of the patients suffering from Behcet's disease were initially discovered by CMR, since their primary finding was

an intracardiac mass by TTE while CMR detected additional findings including the presence of bilateral parahilar pulmonary artery aneurysms in one patient and a large left intrapul-monary artery aneurysm in another one, hence raising the possibility of Behcet's disease. The third patient was a known case of Behcet's disease in which CMR discovered a right ventricular subendocardial fibrosis with thrombosis on top.

TTE diagnosed two interventricular septum (IVS) cysts. CMR showed nonenhancing cysts, excluding the possibility of cystic tumors. In addition CMR detected the presence of associated hepatic cyst in one patient favoring the diagnosis of hydatid disease (Fig. 3). One of them had a positive serology test while the other was an isolated IVS hydatid cyst with -ve serology test. Both cases were confirmed by postoperative pathology.

Five cases of rhabdomyomas were diagnosed in this study showing isointense signal on T1 weighted images and relatively high signal of T2 weighted images with homogenous contrast enhancement. One of the cases showed multiple lesions within the RV of which one was seen obstructing the right ventricle outflow tract (Fig. 4). Two of these cases had associated tuberous sclerosis.

The two cases of fibromas were found in the interventricu-lar septum showing low signal on T2 weighted images with no enhancement (Fig. 5).

Myxomas were found in the left atria showing mobile low signal mass on gradient echo with homogenous contrast enhancement on post contrast study (Fig. 6).

Intraatrial lipoma was diagnosed on bases of being homogeneous with hyperintense signal on spin echo sequences, behaving similar to subcutaneous fat in all sequences (Fig. 7).

3.1. CMR changed the echocardiographic diagnosis in the following five cases

TTE diagnosed a four year old patient as having multiple masses suspicious of rhabdomyomatosis. CMR proved it to be a solitary intramural mass with a characteristic low T1

Fig. 1 (a) Left atrial thrombus showing isointense signal on T1wIs within the dilated left atrium. (b) Axial bFFE showing a characteristically dark signal of the thrombus (c) No evident of contrast enhancement within the thrombus on T1wIs.

and T2 weighted image signal with no evident contrast enhancement. Thus changing the diagnosis to a solitary IVS fibroma (Fig. 5).

In a seven month old male child CMR changed the diagnosis from pulmonary valve vegetation into papillary fibroelas-toma of the pulmonary valve. CMR depicted a small 12 mm

Fig. 2 Patient with chronic myocardial infarction showing transmural enhancement and thinning of the apex with thrombus formation on IR sequences (arrow).

elongated mobile mass at the pulmonary valve that was isointense in both T1 and T2 weighted images and low in gradient echo sequences, showing no contrast enhancement (Fig. 8).

In one case which was diagnosed as right atrial cystic lesion by TTE, CMR detected a huge pulmonary mass lesion compressing the right atrium with no evident intracardiac mass (Fig. 9).

TTE diagnosed three cases of cardiac myxomas on basis of being mobile intra atrial mass, however CMR confirmed the diagnosis of only two cases and changed the diagnosis of the third case into thrombus showing low signal on both T1 and T2 weighted images with no contrast enhancement.

CMR changed the TTE diagnosis of a mass into a thrombus which was later confirmed on surgery to be infected thrombi within the left ventricle.

CMR showed additional findings within the thorax that altered the diagnosis in two cases with intracardiac masses associated with pulmonary aneurysms, hence raising the possibility of Behcet's disease which was confirmed later by clinical and laboratory investigations.

CMR was superior to TTE in diagnosing four cases of left ventricular thrombi on top of dysfunctional myocardium which were missed by TTE.

Some of the cardiac masses were part of a systemic disease such as Behcet's disease and tuberous sclerosis, while some were associated with myocardial disease.

Fig. 3 (a) Pathologically proven intraventricular septum Hydatid cyst on T2wIs. (star within the cyst) (b) Sagittal and coronal b-FFE images show the IVS hydatid cyst with associated hepatic cyst.

Four patients had undergone surgical operation based on the combined CMR and TTE; cases of hydatid cysts, one case of myxoma, and the case of infected thrombus. All other cases have undergone follow up by TTE or CMR where some showed regression in size like rhabdomyomas and others completely disappeared like thrombi.

4. Discussion

Cardiac tumors have a broad differential diagnosis and need to be distinguished from non-neoplastic masses (7).

A relatively large review of primary cardiac neoplasms demonstrated approximately 10% mortality in myxomas, 30% mortality in non-myxoma benign tumors, and 100%

Fig. 5 Sagittal postcontrast image showing a nonenhancing intramural fibroma within the IVS.

Fig. 8 A small elongated 12 mm lesion seen at the region of the pulmonary valve in a 7 month old child on T1wIs denoting fibroelastoma.

mortality within 3 years in all patients with malignant tumors. Thus, pre-surgical diagnosis should focus on determining whether a tumor is benign or malignant (8).

TTE is considered the procedure of choice for diagnosing intracardiac tumors. This technique is noninvasive and may provide information on the size, mobility, shape, and location of cardiac masses. However, it has the limitation of poor echogenicity in 30% of patients, absent tissue characterization and poor evaluation of pericardium/extracardiac structures. Moreover, Bone and lung interference remains a major limitation of TTE and renders this real-time imaging technique suboptimal in patients with chronic obstructive pulmonary disease or narrow rib spaces (9,10).

At present, CMR is definitely one of the preferred imaging modalities in the evaluation of patients with suspected cardiac masses (9).

Spin echo sequences provide detailed morphological information of the heart, the great vessels and the adjacent structures (5).

According to Hoffmann et al. (5); the major tasks of magnetic resonance imaging in diagnosing cardiac masses are: to confirm or exclude a mass suspected by X-ray or TTE, to assess the location, mobility and its relationship to surrounding tissues, to image the degree of vascularization, which allows differentiation between benign and malignant masses, to distinguish solid from fluid lesions, and to determine tissue characteristics and the specific nature of a mass (11); all of which were applied in this study.

4.1. Rhabdomyomas

Rhabdomyomas being the commonest cardiac tumor in early childhood (12) were represented by five cases. Associated tuberous sclerosis was found in two cases of rhabdomyomas, while three of them showed regression in size on follow up examination which were the most important differential features from fibroma according to Bader et al. (13).

Fig. 9 Axial b-FFE showing a huge right pulmonary mass occupying the right lung, invading and compressing both atria. The case was misdiagnosed by TTE as RA mass.

4.2. Fibromas

Intracardiac fibromas showed the same CMR findings as reported by Kim et al. (6). None of them had associated gorlin syndrome. Although one of them was a unique case of IVS fibroma associated with vertebral and vascular anomalies (persistant left SVC, hypoplastic left pulmonary artery and interrupted hepatic segment of the IVC). The patient had normal CT and MRI of the brain, normal hands and genitalia with no basal cell carcinomas, thus gorlin syndrome was excluded.

4.3. Myxomas

Two cases of cardiac myxomas were presented in this study; both were in the left atrium in agreement with Grebenc et al.

4.4. Papillary fibroelastoma

Since papillary fibroelastoma is a rare cardiac benign tumor (6) one case was diagnosed in a seven month old male infant who had murmur on the site of pulmonary artery during examination.

4.5. Nontumoral Masses

Since nontumoral masses usually require medical (or no) treatment, accurate differentiation between the two conditions is important.

4.6. Thrombi

Thrombi can be misinterpreted as a cardiac tumor on TTE

(15). However on CMR thrombus has a characteristically low signal on gradient echo sequences with isointense or low signal on T1 weighted images and low signal on T2 weighted images with no contrast enhancement (9). Fifteen cases of intracardiac thrombi were reported in which most of them were misinterpreted as cardiac masses by TTE. However CMR was relevant as the lesions did not enhance after contrast administration and showed characteristic low signal on gradient echo and inversion recovery sequences (16).

Left ventricular (LV) thrombus formation is a frequent complication in patients with ischemic heart disease. Its recognition and appropriate treatment remain important due to the risk of systemic embolization, occurring in 613% of patients (17). TTE missed three cases of small thrombi occurred on top of dysfunctional left ventricles likely because apical thrombi are often difficult to image, which is related to insufficient image quality and problems assessing the LV apex, especially when aneurysmatic (near-field probe) (17).

Few case reports (18,19) have being published before showing the presence of cardiac thrombus associated with pulmonary aneurysms in Behcet's disease as was detected in this study. Patients with Behcet's disease showed mainly right sided thrombi similar to those detected in the study conducted by Mogulkoc et al. (20).

Although involvement of the heart by hydatid disease is rare (21), two cases of IVS hydatid cysts were included in this study. Due to the larger field of CMR over TTE we could detect the presence of associated hepatic cyst, favoring the diagnosis of hydatid disease.

This advantage of CMR also allowed us to accurately diagnose a case of huge pulmonary mass compressing the heart, which was initially misdiagnosed by TTE as cystic right atrial lesion.

Sarcomas are the second most common primary cardiac tumors after myxomas and are encountered in adults with a mean age of 40 years. The most common type of sarcomas are angiosarcomas, frequently involving the right atrium. Other sarcoma types are undifferentiated sarcomas, malignant fibrous histiocytoma and leiomyosarcoma affecting the left atrium. These tumors are diagnosed on MRI as heterogeneous invasive masses with hemorrhage, necrosis, valvular destruction, extracardiac invasion and metastases (16).

Hence, CMR provides diagnosis of benign tumors, specifies their types, location, extension and also differentiates neoplas-tic from non-neoplatic lesions; while it can suggest the presence and extension of malignant masses without accurate specification of the type of tumor.

Cardiac masses are rare but can be well characterized and often confidently diagnosed with CMR, allowing the differentiation between neoplastic and non-neoplastic masses, hence facilitating the choice for medical or surgical treatment.

5. Conclusion

CMR is superior to TTE in characterization of the tumor, detecting its site, size, vascularity, hemodynamics and multiplicity of the lesion with high tissue contrast. Thus it should be performed in all cases suspicious of cardiac masses as it could aid in patient management and guide for surgery. Moreover it can prevent unneeded cardiac surgery in cases of benign non-neoplastic lesions or in benign tumors with no hemody-namic significance.

References

(1) Burke A, Virmani R. Tumors of the heart and great vessels. In: Atlas of tumor pathology: fasc 16, ser 3. Washington, DC: Armed Forces Institute of Pathology; 1996. p. 1-98.

(2) Thomas JD, Rubin DN. Tissue harmonic imaging: why does it work? J Am Soc Echocardiogr 1998;11:803-8.

(3) Shah DJ. Evaluation of cardiac masses: the role of cardiovascular magnetic resonance. Methodist Debakey Cardiovasc J 2010;6(3):4—11.

(4) Randhawa K, Ganeshan A, Hoey ET. Magnetic resonance imaging of cardiac tumors: part 2, malignant tumors and tumorlike conditions. Curr Probl Diagn Radiol 2011;40(4):79-169.

(5) Hoffmann U, Globits S, Frank H. Cardiac and paracardiac masses current opinion on diagnostic evaluation by magnetic resonance imaging. Eur Heart J 1998;19:553-63.

(6) Kim EY, Choe YH, Sung K, et al. Multidetector CT and MR imaging of cardiac tumors. Korean J Radiol 2009;10(2):164-75.

(7) Fussen S, De Boeck B, Zellweger MJ, et al. Cardiovascular magnetic resonance imaging for diagnosis and clinical management of suspected cardiac masses and tumours. Eur Heart J 2011;32:1551-60.

(8) Beek E, Stolpen AH, Khanna G, Thompson HB. CT and CMR of pericardial and cardiac neoplastic disease. Cancer Imaging 2007;7:19-26.

(9) Bogaert J, Dymarkowski S. Cardiac masses. In: Bogaert J, Dymarkowski S, Taylor AM, editors. Clinical cardiac CMR. Berlin: Springer; 2005. p. 305-53.

(10) Alam M. Pitfalls in the echocardiographic diagnosis of intracardiac and extracardiac masses. TTE 1993;10:181-91.

(11) Amparo EG, Higgins CB, Farmer D, et al. Gated CMR of cardiac and paracardiac masses, initial experience. Am Heart J 1984;143:6-1151.

(12) Kiaffis MG, Powell AJ, Geva T. Magnetic resonance imaging evaluation of cardiac tumor characteristics in infants and children. Am J Cardiol 2002;89:1229-33.

(13) Bader RS, Chitayat D, Kelly E, et al. Fetal rhabdomyoma: prenatal diagnosis, clinical outcome, and incidence of associated tuberous sclerosis complex. J Pediatr 2003;143:620-4.

(14) Grebenc ML, Rosado de Christenson, Green CE, et al. From the archives of the AFIP. Cardiac myxoma: imaging features in 83 patients. Radiographics 2002;22:673-89.

(15) Ragland MM, Tak T. The role of TTE in diagnosing space-occupying lesions of the heart. Clin Med Res 2006;4(1): 22-32.

(16) Narin B, Arman A, Arslan D, et al. Assessment of cardiac masses: magnetic resonance imaging versus transthoracic TTE. Anadolu Kardiyol Derg 2010;10:69-74.

(17) Mollet NR, Dymarkowski S, Volders W, et al. Visualization of ventricular thrombi with contrast-enhanced magnetic resonance imaging in patients with ischemic heart disease. Circulation 2002;106:2873-6.

(18) Yoshida S, Fujimori K, Hareyama M, Nakata T. Cardiac thrombus in Behcet's disease. Chest 2001:688-9.

(19) Ernam D, Atalay F, Alp A, Hasanoglu HC. A Behcet's disease patient with intracardiac thrombus, pulmonary artery aneurysms complicating recurrent pulmonary thromboembolism. Tiiberkii-loz ve Toraks Dergisi 2006;54(2):168-71.

(20) Mogulkoc N, Burgess MI, Bishop PW. Intracardiac thrombus in Behcet's disease: a systematic review. Chest 2000;118:479-87.

(21) Karabay O, Onen A, Yildiz F, et al. The case of a cyst hydatid localized within the interatrial septum. Jpn Heart J 2004;45:703-7.