Echocardiography. 2019;36:837–843. wileyonlinelibrary.com/journal/echo © 2019 Wiley Periodicals, Inc.
|
837 Received: 28 November 2018|
Revised: 2 March 2019|
Accepted: 6 March 2019DOI: 10.1111/echo.14331
B E S T P A P E R
Evaluation of the potential predictors of embolism in patients
with left atrial myxoma
Macit Kalçık MD
1| Emrah Bayam MD
2| Ahmet Güner MD
3| Ayhan Küp MD
3|
Semih Kalkan MD
3| Mahmut Yesin MD
4| Mustafa O. Gürsoy MD
5|
Sabahattin Gündüz MD
3| Süleyman Karakoyun MD
6| Mehmet Özkan MD
3,71Department of Cardiology, Hitit University Faculty of Medicine, Çorum, Turkey 2Department of Cardiology, Umraniye Training and Research Hospital, Istanbul, Turkey 3Department of Cardiology, Kosuyolu Kartal Heart Training and Research Hospital, Istanbul, Turkey 4Department of Cardiology, Kars Harakani State Hospital, Kars, Turkey 5Department of Cardiology, Izmir Katip Çelebi University Atatürk Training and Research Hospital, Izmir, Turkey 6Department of Cardiology, Faculty of Medicine, Kars Kafkas University, Kars, Turkey 7Division of Health Sciences, Ardahan University, Ardahan, Turkey Correspondence Macit Kalçık, MD, Hitit University Faculty of Medicine, Çorum, Turkey. Email: macitkalcik@yahoo.com Introduction: Cardiac myxomas are the most common primary intracardiac tumors. Although myxomas are histologically benign, they are potentially dangerous due to potential risk of systemic and cerebral embolism. In this study, we aimed to investi-gate the potential predictors of embolism in patients with left atrial myxoma.
Methods: This single- center retrospective study enrolled 93 patients (mean age:
52.9 ± 15.3 years, female: 70 [75.3%]) with left atrial myxomas between 2014 and 2018. The patients were classified into two groups (embolic vs nonembolic) to inves-tigate possible predictors of embolism. Demographic, laboratory, and echocardio-graphic parameters were recorded into a dataset and compared between patients with and without embolism.
Results: The study population was composed of 13 (14%) patients in embolic (11
cerebrovascular and 2 peripheral) and 80 (86%) patients in nonembolic group. Demographic and laboratory parameters were similar between the groups. Tumor sizes were significantly higher in the embolic group than in the nonembolic group (5.59 ± 1.08 vs 4.29 ± 0.61; P = 0.001). By multivariate analysis, increased tumor size, increased left atrial diameter, and the presence of atrial fibrillation and irregular tumor surface were identified as independent predictors of embolism. In ROC curve analyses, tumor size above 4.6 cm predicted embolism with a sensitivity of 77% and a specificity of 73% (AUC: 0.858; 95% CI: 0.752–0.964; P < 0.001).
Conclusion: The presence of atrial fibrillation, irregular tumor surface, increased
tumor size, and increased left atrial diameter is associated with increased risk of em-bolism in patients with left atrial myxoma. Early surgery should be scheduled for such patients due to increased potential for embolism. K E Y W O R D S cardiac imaging, cardiac surgery, echocardiography, embolism, left atrial myxoma, transesophageal echocardiography
1 | INTRODUCTION
Cardiac myxomas are the most common primary heart tumors and account for 40% of benign cardiac tumors with approximately 75%– 90% occurring in the left atrium.1 Most left atrial myxomas arise from the interatrial septum and grow into the atrial cavity, but may also have multiple foci and involve valvular tissue.2 The differential diagnosis of cardiac myxomas includes vegetations, thrombi, and other cardiac tumors.3 Histopathological examination is necessary for the final diagnosis.
Manifestations of myxomas include at least one of the clas-sical triad of obstructive, constitutional, and embolic symptoms. Obstructive cardiac manifestations which result from mitral valve obstruction by the tumor (Figure 1A,B) include dyspnea, dizziness, syncope, palpitations, and heart failure. Constitutional and systemic findings include nonspecific symptoms such as fever, fatigue, weight loss, myalgia, and muscle weakness. Occasionally, cardiac myxomas may manifest with neurological symptoms related to a systemic or cerebral embolic event.4
Embolism is a major complication occurring in 30%–50% of patients with cardiac myxomas and is closely associated with mor-tality. The central nervous system is affected in approximately 50% of embolic episodes. Embolic events may also involve the extremities, visceral organs, and coronary arteries.5 The specific risk factors contributing to embolism in cardiac myxomas are still unclear. In this study, we aimed to investigate potential predictors of embolism in patients with left atrial myxoma by comparing the patient and tumor characteristics between the embolic and non-embolic groups.
2 | METHODS
2.1 | Study population
Between May 2014 and January 2018, 93 patients with left atrial masses who were treated surgically and histopathologically diag-nosed with cardiac myxoma were enrolled in this single- center study. Patients with severe left ventricular systolic dysfunction, end- stage liver disorders, renal insufficiency, chronic inflammatory diseases, connective tissue diseases, and malignancies were excluded from the study. The patients who experienced stroke, transient ischemic attacks, coronary embolism, or any systemic embolic events wereincluded in the embolism group. The patients were classified into embolic and nonembolic groups in order to investigate possible pre-dictors of embolism.
All patients underwent transthoracic (TTE) and transesopha-geal echocardiography (TEE) for precise preoperative diagnosis. Preoperative coronary angiography was performed to investigate coronary artery disease in patients >50 years old. Stroke, transient ischemic attack, or other systemic embolic events were diagnosed by computed tomography scan and magnetic resonance imaging. Cardiac myxoma diagnosis was confirmed by histopathological as-sessment of surgical specimens.
Complete blood count and blood chemistry panel were carried out in all patients at the time of admission. All demographic, labora- tory, and echocardiographic parameters were recorded into a data-set and compared between embolic and nonembolic groups. The study protocol was approved by the local ethics committee of the hospital in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines.
2.2 | Echocardiography
All patients underwent TTE performed by using Vivid 5 echocardi-ography device (GE Vingmed Ultrasound AS, Horten, Norway) and 3.2- mHz adult probe with the patient in the left lateral decubitus position. In all patients, the left ventricular posterior wall thickness (PWT), interventricular septal thickness (IVST), left ventricular end- systolic diameter (LVESD), left ventricular end- diastolic diameter (LVEDD), and left atrial diameter (LAD) were measured on the par-asternal long- axis view. Left ventricular ejection fractions (LVEF) of the patients were calculated by using biplane Simpson's method.
Following TTE, two- dimensional (2D) and real time three- dimensional (RT- 3D) TEE images were performed in all patients by using an X7- 2t transducer on an iE33 ultrasound machine (Philips Medical Systems, Andover, MA, USA). The largest tumor size was measured by 2DTEE between 0° and 180° angles in mid- esophageal short- and long- axis views (Figure 2A). The characteristic features of the myxomas such as size, surface appearance, location, and attachment were determined by TEE examinations (Figure 2B). Cardiac myxomas which were attached directly by its base to the atrial wall were classified as sessile, whereas mobile myxomas which were attached to the atrial wall with a stalk were classified as pedunculated. Left atrial myxomas arising from interatrial septum F I G U R E 1 Transthoracic (A) and two- dimensional transesophageal echocardiogram (B) showing a large left atrial myxoma prolapsing through the mitral orifice into the left ventricle during diastole (A) (B)
were called as typical or septal; on the other hand, myxomas arising from other sides of the left atrium or valvular tissues were called atypical or extraseptal. Cardiac myxomas were divided into two groups with regard to surface characteristics. Tumors with regular borders were classified as smooth, which was characterized by a round shape and a solid surface. The others which were charac-terized by a soft and irregular shape and a gelatinous surface with cauliflower appearance were classified as villous.6 The echocardiography findings derived from official clinical reports and the cine images of the TTE and TEE studies of the patients were evaluated by two different cardiologists and analyzed off line. In the embolic group of patients with left atrial myxoma, the embolic events had occurred before echocardiographic evaluation and before clinical presentation. However, the reviewers were unaware of the patients’ clinical history or embolic status and a consensus has been reached for each case regarding the evaluation of the tumor characteristics.
2.3 | Statistical analysis
Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 19.0. (IBM Corp. Armonk, NY, USA). Descriptive statistics were reported as mean ± standard deviation for continu-ous variables with normal distribution or median (25th–75th per-centiles) values for continuous variables without normal distribution and as frequency with percentages for the categorical variables. The Shapiro–Wilk and Kolmogorov–Smirnov tests were used to test the normality of the distribution of continuous variables. Categorical variables were compared with chi- square or Fisher exact tests. Student's t test or Mann–Whitney U test was used to compare con-tinuous variables as appropriate. The significance level was accepted as P < 0.05 in all statistical analyses. A logistic regression analysis was performed in order to identify any independent associates of embolism. A receiver operating characteristic (ROC) curve analysis was performed to evaluate the sensitivity, specificity, area under the curve (AUC), and confidence interval (CI) of tumor size to predict embolism.
3 | RESULTS
A total of 93 patients (mean age: 52.9 ± 15.3 years, female: 70 [75.3%]) with left atrial myxoma were enrolled in this single- center study. The study population was composed of 13 (14%) patients in the embolic (11 cerebrovascular and 2 peripheral) and 80 (86%) pa-tients in the nonembolic group. The demographic characteristics of the study population were presented in Table 1. Age and gender dis-tribution were similar between the groups. There was no significant difference between the embolic and nonembolic groups in terms of systolic and diastolic blood pressure measurements and heart rate values. The prevalences of hypertension, diabetes mellitus, dyslipi-demia, smoking status, coronary artery disease, and valvular heart disease were similar between the groups. The prevalence of atrial fibrillation was numerically higher in the embolic group; however, this difference was not statistically significant. There was also no significant difference in terms of the leading symptoms between the embolic and nonembolic groups (Table 1).
Routine serum biomarkers such as glucose, blood urea nitro-gen, creatinine, uric acid, alanine aminotransferase, aspartate aminotransferase, total cholesterol, high density lipoprotein, low density lipoprotein, triglyceride, and complete blood count parameters were similar between the patients and the controls (Table 1). Among the echocardiographic parameters, LVEF, LVESD, LVEDD, IVST, and PWT were found to be similar between the groups. However, the LAD was significantly higher in the embolic group than in the nonembolic group (44.5 ± 6.8 vs 40.1 ± 5.4 mm;
P = 0.014; Table 2). The greatest tumor size assessed by TEE ex-amination was significantly higher in the embolic group than in the nonembolic group (5.59 ± 1.08 vs 4.29 ± 0.61 cm; P = 0.001; Figure 3A).
The study population consisted of 70 (75.3%) sessile and 23 (24.7%) pedunculated myxomas, and 56 (60.2%) septal and 37 (39.8%) extraseptal left atrial myxomas were observed. The dis-tributions of morphologic characteristics between the groups are included in Table 3. There was no significant difference regarding the appearance and the location of the myxomas between the em-bolic and nonembolic groups. However, the prevalence of irregular (villous) surface morphology was significantly higher in the embolic group (8 [61.5%] vs 26 [32.5%]; P = 0.044). Possible univariate predictors of embolism were entered into multivariate logistic regression analyses. By multivariate analysis, increased tumor size, increased LAD, and the presence of atrial fibrillation and irregular tumor surface were identified as inde-pendent predictors of embolism (Table 4). In ROC curve analyses, tumor size above 4.6 cm predicted embolism with a sensitivity of F I G U R E 2 Two- dimensional transesophageal echocardiography showing an example measurement of the diameters of a large left atrial myxoma (A). Three- dimensional transesophageal echocardiography revealing the irregular surface of a sessile left atrial myxoma arising from interatrial septum (B) (A) (B)
77% and a specificity of 73% (AUC: 0.858; 95% CI: 0.752–0.964;
P < 0.001; Figure 3B). The AUC presented here was for the tumor
size as continues variable.
4 | DISCUSSION
In this single- center study, we have focused on the potential pre- dictors of embolism in patients with left atrial myxoma. The over-all rate of embolization in the study population was 14%; however, our major finding was that structural and anatomical tumor char-acteristics highly influence the risk of embolization. The greatest tumor sizes were significantly higher in the embolic group than in the nonembolic group. Increased tumor size, increased LAD, and the presence of irregular surface morphology and atrial fibrillation were found to be the independent predictors of embolism in patients with left atrial myxoma. Cardiac myxomas are intracavitary benign tumors, which may be localized in any of the four cardiac chambers. They are more common in the atria, especially arise from the interatrial septum at the border of the fossa ovalis in the left atrium, usually appear in middle age, and are more frequent in women.1,2 Most cases are sporadic; however, familial cardiac myxomas may occur, in associ-ation with conditions such as the Carney complex.7 Approximately 10% of the patients with atrial myxoma show no clinical signs. The clinical presentation of cardiac myxomas is usually divided into cardiac, embolic, and systemic manifestations. Cardiac manifesta-tions may be related to tumor size, because as they grow, they can progressively obstruct cardiac chambers mimicking valve disease.6
Currently, echocardiography remains the key examination tool for the diagnosis of atrial myxomas. It enables the diagnosis and de-termines the localization, shape, and size of the tumor and its various connections with the adjacent cardiac structures. TTE is the most commonly used, but TEE has enhanced specificity and sensitivity.8 Computed tomography scan and especially cardiac magnetic reso-nance imaging also provide complementary information on myxo-mas.9 However, these investigations should be reserved for cases in which the diagnosis or characterization of the tumor remains unclear after an echocardiographic evaluation.
Surgical resection of cardiac myxomas is curative in most pa-tients, and relapse of the tumor and recurrent embolism is relatively rare.10 Although the timing of surgery may vary depending on the clinical condition of the patient, a cardiac myxoma is usually con-sidered to be a definite indication for urgent surgery to prevent embolism.11
Embolism, one of the complications of myxoma, is associated with high morbidity and mortality. Embolisms occur in 30%–40% of cases and may involve any visceral organ; however, cerebral arteries are most commonly affected.12 Left atrial myxoma should always be included in the differential diagnosis in case of multiple cerebral in-farcts, especially in young individuals.13 This phenomenon is related to the migration of the tumor fragments, or thrombi and vegetations adherent to the tumor surface.14 TA B L E 1 Comparison of demographic and laboratory parameters between patients with and without embolism Embolism (+) n = 13 Embolism (−) n = 80 P value Demographic parameters Age, y 50 (36–62) 56 (45–65) 0.345 Gender—female, n (%) 10 (76.9) 61 (76.3) 0.921 Body mass index, (kg/ m2) 27.3 ± 2.8 27.4 ± 4.9 0.990 Diastolic blood pressure (mm Hg) 78.6 ± 14.6 79.6 ± 12.2 0.807 Systolic blood pressure (mm Hg) 128.7 ± 22.1 129.9 ± 21.6 0.854 Heart rate (beats/min) 82.8 ± 15.3 82.9 ± 13.2 0.983 Hypertension, n (%) 3 (23.1) 25 (31.2) 0.551 Diabetes mellitus, n (%) 1 (7.6) 12 (15) 0.527 Dyslipidemia, n (%) 2 (15.4) 14 (17.5) 0.851 Smoking status, n (%) 3 (23.1) 14 (17.5) 0.547 Atrial fibrillation, n (%) 4 (30.7) 9 (11.3) 0.060 Coronary artery disease, n (%) 3 (23.1) 15 (18.8) 0.714 Valvular heart disease, n (%) 5 (38.5) 37 (46.3) 0.601 Leading symptoms, n (%) Dyspnea 5 (38.5) 34 (42.5) 0.632 Palpitation 4 (30.7) 28 (35) Angina 2 (15.4) 11 (13.8) Syncope 2 (15.4) 7 (8.7) Laboratory parameters Glucose, (mg/dL) 97.4 ± 11.1 95.6 ± 10.5 0.755 BUN, (g/dL) 28.5 ± 9.4 32.5 ± 11.3 0.264 Creatinine, (g/dL) 0.83 ± 0.19 0.79 ± 0.21 0.463 AST, (U/L) 21 (17.75–26) 20.5(18–23.75) 0.563 ALT, (U/L) 21 (13.75– 29.25) 21.5 (16–29.5) 0.797 Uric acid (mg/dL) 4.91 ± 1.0 4.80 ± 1.0 0.723 LDL (mg/dL) 119.7 ± 19.8 126.5 ± 21.9 0.214 HDL (mg/dL) 41 (34–49) 43 (38–53) 0.391 Triglycerides (mg/dL) 167.9 ± 74.7 171.7 ± 87.4 0.771 Total cholesterol (mg/dL) 198.4 ± 21.9 205.7 ± 23.2 0.163 Hemoglobin (g/dL) 13.47 ± 1.65 14.02 ± 1.68 0.759 White blood cell, (×103/ dL) 7.81 ± 1.95 8.37 ± 1.78 0.294 Platelet, (×103/dL) 252.7 ± 43.9 246.4 ± 51.3 0.483 MPV, (fL) 8.7 (7.5–9.5) 8.1 (7.6–9.2) 0.418 ALT = alanine aminotransferase; AST = aspartate aminotransferase; BUN = blood urea nitrogen; HDL = high- density lipoprotein; LDL = low- density lipoprotein; MPV = mean platelet volume.
The role of tumor size in myxomas as a risk factor for embolism was inconsistent in previous studies. Lee et al5 reported that there were no significant differences in vascular risk factors, the ejec-tion fraction, the left atrial diameter, and the tumor size between the embolic and nonembolic groups. Similarly, He et al15 reported that tumor size did not differ significantly between the embolic and nonembolic groups; however, apparently large myxomas (>25 mm2) were found to be associated with a higher risk of embolic events in the univariate analysis. Dias et al16 reported that the tumor size, atrial fibrillation, and valve impairment were not associated with the embolic events. On the other hand, Liang et al showed that pa-tients who developed embolic events had significantly larger cardiac myxomas and left atrial diameters.17 In the present study, we have found that increased tumor size was an independent risk factor for embolism. Since our study included only the patients with left atrial myxomas, the inconsistency between our study and the literature may come from methodological issues such as patient selection. Increased tumor size may precipitate embolism due to larger inter-active area between the myxoma and the blood circulation. Studies with larger sample sizes are needed to confirm this association.
Polypoid myxomas are usually compact and show little ten-dency toward spontaneous fragmentation. The less common vil-lous or papillary myxomas have a surface that consists of multiple fine or very fine villous extensions.6 One particular characteristic of these tumors is the release of metalloproteinase and enzymes which continuously degrades the extracellular matrix, therefore creates an imbalance between the process of synthesis and tissue
fragmentation.18 These characteristics explain why obstructive heart failure is usually associated with solid tumors while embolic events represent the most common clinical feature of fragile papil-lary myxoma.19 Consistent with established results reported in the literature,15 our study confirmed that the surface of the myxoma was an important risk factor contributing to embolism. Irregular surface contributes to tumor fragmentation and a larger interactive area which may result in embolism.
Echocardiographic parameters Embolism (+) Embolism (−) P value
LAD, (mm) 44.5 ± 6.8 40.1 ± 5.4 0.014 LV EF (%) 61 ± 6.1 63.3 ± 6.3 0.314 LVESD, (mm) 29 ± 2.3 28 ± 5.3 0.606 LVEDD, (mm) 45.3 ± 2.4 45.7 ± 6.7 0.851 IVST, (mm) 10 (9–11) 10 (9–11.5) 0.877 PWT, (mm) 10 (9–10) 10 (10–11) 0.765
IVST = interventricular septal thickness; LAD = left atrial diameter; LVEDD = left ventricular end- diastolic diameter; LVEF = left ventricular ejection fraction; LVESD = left ventricular end- systolic diameter; PWT = posterior wall thickness. TA B L E 2 Comparison of echocardiographic parameters between patients with and without embolism F I G U R E 3 The box- plot graph comparing the tumor sizes between patients with and without embolism (A) Receiver operating characteristic curve revealing the area under the curve for tumor size to predict embolism (B) TA B L E 3 Comparison of morphologic characteristics of left atrial myxomas between patients with and without embolism Morphologic
characteristics Embolism (+) n = 13 Embolism (- ) n = 80 P value
Tumor size, (cm) 5.59 ± 1.08 4.29 ± 0.61 0.001 Appearance Sessile, n (%) 9 (69.2) 61 (76.3) 0.586 Pedunculated, n (%) 4 (30.8) 19 (23.7) Location Typical (septal), n (%) 7 (53.8) 49 (61.2) 0.613 Atypical (extraseptal), n (%) 6 (56.2) 31 (38.8) Surface Irregular or villous, n (%) 8 (61.5) 26 (32.5) 0.044 Regular or smooth, n (%) 5 (38.5) 54 (67.5)
Besides increased tumor size and irregular tumor surface, in-creased LAD and the presence of atrial fibrillation were the other predictors of embolism in our study. Although the minority (30%) of patients with embolism had atrial fibrillation in this study, it was debatable whether embolism occurred due to atrial fibrillation or tumor embolism. Since both increased tumor size and atrial fibril-lation were independent predictors of embolism in this study, it could not be wrong to hypothesize that stasis from obstruction by the tumor mass may lead to embolic left atrial appendage thrombus formation in such patients.
4.1 | Study limitations
The primary limitation was that our study was a nonrandomized and single- center study with a relatively small number of patients. Secondly, the lack of computerized tomography scan and magnetic resonance imaging in all study population may have masked sev-eral patients with asymptomatic embolisms. In addition, despite a consensus has been reached for each case during the evaluation of echocardiographic images, still there may be some subjective issues regarding the evaluation of the tumor characteristics. Furthermore, we have performed multivariate analyses with several covariates as potential predictors of systemic embolic events. However, there were only 13 embolic events in the study population. Such a small number of embolic events were one of the major limitations of our study. Due to this concern, the results of our study should be con-sidered with caution.
5 | CONCLUSION
Thromboembolism is a serious complication of left atrial myxomas. The risk of embolism may be associated with the increased tumor size, increased LAD, and the presence of irregular surface morphol-ogy and atrial fibrillation in patients with left atrial myxoma. Early surgery may be scheduled in such patients due to increased potential for embolism. Larger clinical trials are necessary to confirm the find-ings of the present study. CONFLIC T OF INTEREST All of the authors have no conflict of interest. ORCIDMacit Kalçık https://orcid.org/0000-0002-8791-4475
REFERENCES
1. Butany J, Nair V, Naseemuddin A, et al. Cardiac tumours: diagnosis and management. Lancet Oncol. 2005;6(4):219–228.
2. Larrieu AJ, Jamieson WR, Tyers GF, et al. Primary cardiac tu-mors: experience with 25 cases. J Thorac Cardiovasc Surg. 1982;83:339–348.
3. Lee HS, Kim HK, Park EA, et al. Left atrial intramural hematoma after removal of atrial myxoma: cardiac magnetic resonance in the differential diagnosis of intra- cardiac mass. J Cardiovasc Ultrasound. 2014;22:205–208.
4. Meng Q, Lai H, Lima J, Tong W, Qian Y, Lai S. Echocardiographic and pathologic characteristics of primary cardiac tumors: a study of 149 cases. Int J Cardiol. 2002;84:69–75.
5. Lee SJ, Kim JH, Na CY, Oh SS. Eleven years’ experience with Korean cardiac myxoma patients: focus on embolic complications. Cerebrovasc Dis. 2012;33(5):471–479.
6. Reynen K. Cardiac myxomas. N Engl J Med. 1995;333(24):1610–1617. 7. Ozdemir N, Kaymaz C, Ozveren O, et al. Coincidence of spo-radic left atrial myxoma and benign Schwannoma of the foot erroneously diagnosed as Carney complex. Echocardiography. 2003;20(4):385–386.
8. Pérez de Isla L, de Castro R, Zamorano JL, et al. Diagnosis and treat-ment of cardiac Myxomas by transesophageal echocardiography. Am J Cardiol. 2002;90:1419–1421. 9. Araoz PA, Mulvagh SL, Tazelaar HD, Julsrud PR, Breen JF. CT and MR imaging of benign primary cardiac neoplasms with echocardio-graphic correlation. Radiographics. 2000;20:1303–1319. 10. Shah IK, Dearani JA, Daly RC, et al. Cardiac myxomas: a 50- year ex-perience with resection and analysis of risk factors for recurrence. Ann Thorac Surg. 2015;100:495–500.
11. Elbardissi AW, Derani JA, Daly RC, et al. Survival after resec-tion of primary cardiac tumors: a 48- year experience. Circulation. 2008;118:7–15.
12. Lee VH, Connolly HM, Brown RD Jr. Central nervous system mani-festations of cardiac myxoma. Arch Neurol. 2007;64(8):1115–1120. 13. Pinede L, Duhaut P, Loire R. Clinical presentation of left atrial
cardiac myxoma. A series of 112 consecutive cases. Medicine (Baltimore). 2001;80:159–172.
14. Boutayeb A, Mahfoudi L, Moughil S. Atrial myxoma: from diagnosis to management. Clin Surg. 2017;2:1498.
15. He D-K, Zhang Y-F, Liang Y, Ye S-X, Wang C, Kang B, Wang Z-N. Risk factors for embolism in cardiac myxoma: a retrospective anal-ysis. Med Sci Monit. 2015;22(21):1146–1154.
16. Dias RR, Fernandes F, Ramires FJ, et al. Mortality and embolic po-tential of cardiac tumors. Arq Bras Cardiol. 2014;103(1):13–18. 17. Yin L, Wang J, Li W, et al. Usefulness of CHA2DS2-VASc
scor-ing systems for predictscor-ing risk of perioperative embolism in pa-tients of cardiac myxomas underwent surgical treatment. Sci Rep. 2016;6:39323. TA B L E 4 Multivariate regression analysis showing independent predictors of embolism OR 95% CI P value Age 1.059 0.875–1.050 0.365 Increased LAD 1.216 1.044–1.417 0.012 Decreased LVEF 1.136 0.828–1.059 0.293 Presence of AF 9.957 2.648–31.486 0.012 Presence of VHD 1.332 0.027–4.098 0.391 Increased tumor size 6.564 2.150–20.037 0.001 Pedunculated appearance 2.105 0.424–10.464 0.363 Typical tumor location 0.517 0.118–2.256 0.380 Irregular tumor surface 5.490 1.994–12.625 0.017 AF = atrial fibrillation; CI = confidence interval; LAD = left atrial diame- ter; LVEF = left ventricular ejection fraction; OR = odds ratio; VHD = val-vular heart disease.
18. Orlandi A, Ciucci A, Ferlosio A, Pellegrino A, Chiariello L, Spagnoli LG. Increased expression and activity of matrix metallopro-teinases characterize embolic cardiac myxomas. Am J Pathol. 2005;166(6):1619–1628. 19. Swartz MF, Lutz CJ, Chandan VS, Landas S, Fink GW. Atrial myxo-mas: pathologic types, tumor location, and presenting symptoms. J Card Surg. 2006;21(4):435–440. How to cite this article: Kalçık M, Bayam E, Güner A, et al. Evaluation of the potential predictors of embolism in patients with left atrial myxoma. Echocardiography. 2019;36:837–843. https://doi.org/10.1111/echo.14331
