Supraventricular Arrhythmias in Patients With Cardiac Sarcoidosis. Prevalence, Predictors, and Clinical Implications

Transcription

1 CHEST Original Research Supraventricular Arrhythmias in Patients With Cardiac Sarcoidosis Prevalence, Predictors, and Clinical Implications DIFFUSE LUNG DISEASE Juan F. Viles-Gonzalez, MD ; Luciano Pastori, MD ; Avi Fischer, MD ; Juan P. Wisnivesky, MD, DrPH ; Martin G. Goldman, MD ; and Davendra Mehta, MD, PhD Background: Cardiac sarcoidosis (CS) is known to be associated with congestive heart failure, conduction disorders, and tachyarrhythmias. Ventricular arrhythmias are the most feared cardiac manifestation because they often are unpredictable, may be the first manifestation of the disease, and may be fatal. The propensity for the development of supraventricular arrhythmias (SVAs) in patients with CS has not been described. The aim of this study was to assess the prevalence as well as the predictors of SVA. Methods: We retrospectively investigated 100 patients with biopsy specimen-proven systemic sarcoidosis and evidence of cardiac involvement (defined by cardiac biopsy specimen, PET scan, or cardiac MRI). The mean follow-up was years. ECG, Holter monitoring, implantable cardioverter defibrillator interrogations, or electrophysiology studies were used to document SVA. Echocardiographic data, demographics, and extracardiac involvement were recorded, and univariate and Poisson regressions were performed to compare characteristics of patients with and without documented SVA. Results: The prevalence of SVA was 32%, and atrial fibrillation was the most common arrhythmia, comprising 18% of the total burden, followed by atrial tachycardias (7%), atrial flutter (5%), and other supraventricular tachycardias (2%). Of the patients with SVA, 96% were symptomatic. Left atrial enlargement (LAE) was more frequent in the group with SVA, with an incidence of per 1,000 person-years, and it significantly increased the likelihood of SVA on multivariate analysis (risk ratio, 6.12; 95% CI, ). Diastolic dysfunction, systemic hypertension, and right atrial enlargement were predictors of SVA on univariate analysis. Left ventricular hypertrophy, right ventricular dysfunction, tricuspid valve disease, pulmonary hypertension, and pulmonary sarcoidosis were not associated with SVA on univariate analysis. Conclusions: The study systematically evaluated the frequency of SVA in a large number of patients with CS. SVA in patients with CS is frequent and associated with symptoms. LAE was clearly associated with the development of SVA in this patient population. The extent to which LAE predicts the occurrence of SVA in larger, more diverse CS populations should be evaluated prospectively. CHEST 2013; 143(4): Abbreviations: CMR 5 cardiac MRI; CS 5 cardiac sarcoidosis; EPS 5 electrophysiology study; ICD 5 implantable cardioverter defibrillator; LAE 5 left atrial enlargement; LV 5 left ventricular; LVEF 5 left ventricular ejection fraction; RAE 5 right atrial enlargement; RR 5 risk ratio; RV 5 right ventricular; SVA 5 supraventricular arrhythmia Sarcoidosis is a multisystem, granulomatous disease of unclear etiology that often is observed in individuals between the second and fifth decades of life. 1 Cardiac involvement occurs in 20% to 30% of patients and is associated with adverse prognosis. 2,3 The main cardiac manifestations are conduction system abnormalities, arrhythmias, and heart failure. 4,5 The clinical presentation is pleomorphic, ranging from completely asymptomatic to palpitations, chest pain (cardiac and noncardiac etiology), congestive heart failure, syncope, and sudden death Autopsy studies have revealed that the median survival is, 2 years following the development of cardiac signs and symptoms. 2 Therefore, early detection of any cardiac manifestation, including arrhythmias, is of major importance. Arrhythmias, including sudden death, are the most feared cardiac manifestation of sarcoidosis They often are unpredictable and may occur as the first journal.publications.chestnet.org CHEST / 143 / 4 / APRIL

2 manifestation of the disease. Data are lacking regarding the prevalence of supraventricular arrhythmias (SVAs) and risk factors for their development in patients with cardiac sarcoidosis (CS). Early reports suggested that although less common than ventricular arrhythmias, SVAs occur in about 19% of patients. 16 Understanding the role of SVAs in CS is of great importance because they may contribute to the morbidity associated with this disease. We sought to retrospectively investigate the prevalence of SVAs in a large series of patients with confirmed CS. Using data from biopsy specimens or cardiac imaging study, such as PET scan or cardiac MRI (CMR), we sought to characterize the prevalence and types of SVA. Objective documentation of SVA was obtained from ECG, Holter and event monitors, implantable cardioverter defibrillator (ICD) and pacemaker interrogations, or electrophysiology studies (EPSs). In addition, we sought to identify predictors of SVA and their clinical implications. Materials and Methods The institutional review board of the Mount Sinai Medical Center approved the study (GCO ). One hundred patients with proven CS were included in this analysis. Consecutive patients referred to the electrophysiology clinic between June 1998 and May 2011 were included. Seventy-two patients included in this analysis were also reported in prior publications by our group. All patients had biopsy specimen-proven systemic sarcoidosis and evidence of CS as defined by biopsy specimen or typical imaging findings on either CMR or PET scan. CMR findings diagnostic of CS included localized intramyocardial increased signal intensity on T2-weighted sequence indicative of edema and delayed contrast enhancement suggestive of infiltration and scarring. PET scan findings were considered positive when perfusion images showed no evidence of ischemia and fluorodeoxyglucose uptake images showed either increased or mismatched metabolic activity suggestive of infiltration or a matched decrease in metabolic activity indicative of scarring. The results of chest radiography and pulmonary function testing and echocardiography were also recorded. Patients were excluded if they had known structural cardiac disease of different etiology and did not have cardiac imaging with PET scanning or CMR and echocardiogram. Medical Record Review Standardized data forms were used to extract pertinent data from the medical record on demographics and medical history. An Manuscript received December 18, 2011; revision accepted September 8, Affiliations: From the University of Miami Leonard M. Miller School of Medicine (Dr Viles-Gonzalez), Miami, FL; Metropolitan Hospital Center (Dr Pastori), New York, NY; and Mount Sinai School of Medicine (Drs Fischer, Wisnivesky, Goldman, and Mehta), New York, NY. Funding/Support: The authors have reported to CHEST that no funding was received for this study. Correspondence to: Juan F. Viles-Gonzalez, MD, University of Miami Leonard M. Miller School of Medicine, 1120 NW 14th St, Miami, FL 33136; j.vilesgonzalez@med.miami.edu 2013 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: /chest ECG recorded at the time of office visits or EPS was interpreted. The 12-lead ECG was defined as abnormal if any of the following were present: PR duration. 200 milliseconds, QRS duration. 120 milliseconds atrial fibrillation or flutter;. 50% atrial or ventricular pacing, and evidence of right ventricular (RV) hypertrophy in the absence of pulmonary arterial hypertension. The presence of prednisone, methylprednisolone, hydrocortisone, or other immunosuppressant on the patient s medication list was considered positive for corticosteroid and immunosuppressant usage, regardless of dosage or duration. Mortality was ascertained by review of medical records and the Social Security Death Index. Person-years of follow-up were calculated as the interval from CS diagnosis to the date of death or the end of the follow-up period, whichever came first. The extent of lung involvement was assessed by three methods: chest radiograph, chest CT scan, and abnormal pulmonary function tests. Echocardiographic Parameters For each patient, all left ventricular ejection fraction (LVEF) measurements with a numeric value recorded in the institutional cardiovascular database were ascertained. Additional echocardiographic parameters, including right and left atrial size, assessment of left ventricular (LV) and RV diastolic and systolic function, LV hypertrophy, valvular disease, pulmonary hypertension, and pericardial disease, were recorded. Pulmonary hypertension was determined by conventional transthoracic echocardiography (calculated using tricuspid regurgitation jet velocity and the continuity equation). Any valvular disease except for trace regurgitation was recorded. Any degree of atrial enlargement (. 20 cm 2 ) was considered abnormal for the purpose of this analysis. Statistical Analysis Groups were defined by the presence or absence of SVA. Data are presented as mean SD. The incidence density of SVA was calculated with person-years as the denominator under the Poisson assumption to account for the variation of length of follow-up. We used the risk ratio (RR) with 95% CIs to compare the risk of SVA among study subjects. All variables were considered as candidates in multivariate analysis. The variables used in the multivariable model were age, sex, race, LVEF, LV hypertrophy and diastolic dysfunction, RV dysfunction, mitral valve disease, left atrial enlargement (LAE) and right atrial enlargement (RAE), pulmonary sarcoidosis, and pulmonary hypertension. P,.05 was considered statistically significant. All analyses were performed with SAS, version 9.0 (SAS Institute Inc) statistical software and using two-sided P values. Results Baseline clinical characteristics are listed in Table 1. The mean follow-up was years. The majority of the patients were women and black, with a mean age of 49.6 years. There were no statistically significant differences in age, race, and sex between patients with and without SVA. More than two-thirds of the patients (82%) had documented pulmonary sarcoidosis, and 56 patients (56%) had received treatment with corticosteroids or immunosuppressants. The mean LVEF was 52%, and 29% had evidence of RV systolic dysfunction. All patients underwent EPS; 12 (12%) had inducible ventricular tachyarrhythmias. Fifty-two patients had ICDs Original Research

3 Table 1 Clinical Characteristics of All Study Patients Baseline Characteristic Prevalence of SVAs Value Age, y 49.6 Female sex 60 Diastolic dysfunction 43 Mean LVEF 51 RV dysfunction 29 Left atrial enlargement 28 Right atrial enlargement 27 Mitral valve disease 16 Tricuspid valve disease 36 Pulmonary sarcoidosis 58 Immunosuppressive therapy 56 Pulmonary hypertension 30 Hypertension 46 Black race 54 Data are presented as %, unless otherwise indicated. LVEF 5 left ventricular ejection fraction; RV 5 right ventricular. The prevalence of SVAs was 32%. Atrial fibrillation was the most common arrhythmia, comprising 18% of the total SVAs followed by atrial tachycardias in 7%, atrial flutter in 5%, and atrioventricular nodal reentry tachycardia in 2%. Clinical Impact of SVAs The majority of patients with SVAs had symptoms that were attributed to the arrhythmia (96% symptomatic vs 4% asymptomatic, P,.05). The most common symptoms were palpitations and dyspnea. Therefore, only four patients had subclinical asymptomatic arrhythmias, and these were detected by ICD interrogation, Holter monitoring, or EPS. There were six deaths during the year follow-up. Four were attributed to pulmonary sarcoidosis, one to progressive congestive heart failure, and one to ventricular arrhythmias. Only two of six deceased patients had documented SVA (33.3% vs 66.7% without SVA, P 5.94) not directly related to the cause of death. Predictors of SVAs The results of the univariate analysis are presented in Table 2. All patients had an EPS. Twelve had inducible ventricular arrhythmias, but only three of them (25%) had concomitant SVA. We found no association between inducibility of ventricular arrhythmias on EPS and the prevalence of SVA. The presence of an ICD was similar between patients with and without SVA ( P 5.48). On univariate analysis, the incidence density of SVA among patients with and without LAE was and 38.3 per 1,000 person-years, respectively (RR, 6.99; 95% CI, ). Similarly, RAE was more fre- quent in patients with SVA, with an incidence density of vs 67.4 per 1,000 person-years in those without SVA (RR, 2.43; 95% CI, ) ( Fig 1 ). Diastolic dysfunction was also more prevalent in the group of patients with SVA, with an incidence density of vs 45.5 per 1,000 person-years in the group without SVA (RR, 3.46; 95% CI, ). The presence of LV hypertrophy was not more frequent in patients with SVA. The incidence density of SVA among patients with and without systemic hypertension was and 42.8 per 1,000 person-years, respectively (RR, 3.58; 95% CI, ) ( Fig 1 ). The incidence of LV or RV systolic dysfunction was not significantly higher among patients with SVA. The mean LVEF in patients with SVA was 53.5% vs 51.6% in those without SVA (RR, 0.57; 95% CI, ). The incidence density of SVA among patients with and without RV dysfunction was and 79.7 per 1,000 person-years, respectively (RR, 1.72; 95% CI, ). Only 16 patients had mitral valve disease, and in these patients, SVA was more frequent, with an incidence density of vs 77.6 per 1,000 personyears (RR, 2.47; 95% CI, ). In addition, tricuspid valve disease occurred more frequently (36%) than mitral valve disease, but it was more frequent among patients without SVA (RR, 1.81; 95% CI, ). Pulmonary sarcoidosis and pulmonary hypertension were slightly more frequent in patients without SVA (RR, 3.2; 95% CI, ) than in those with SVA (RR, 2.44; 95% CI, ) ( Fig 1 ). Multivariate Predictors of SVAs The results of the multivariate analysis are summarized in Table 3. Using a Poisson regression analysis after adjusting for all variables, LAE was the only variable associated with an increased likelihood of SVA (RR, 6.12; 95% CI, ; P ). All the other variables, including systolic and diastolic dysfunction, pulmonary sarcoidosis, RAE, mitral valve disease, systemic and pulmonary hypertension, use of steroids and immunosuppressants, age, race, and sex, lacked statistical significance. Discussion CS is well known to cause congestive heart failure and life-threatening ventricular arrhythmias The SVAs described in CS are less common than ventricular arrhythmias and include atrial ectopy, atrial tachycardia, atrial flutter, and atrial fibrillation. 8 These arrhythmias are believed to be related to atrial dilatation secondary to LV dysfunction or cor pulmonale. Less commonly, arrhythmias may be caused by direct granulomatous involvement of the atria. 6,9,10 However, the prevalence and clinical significance of SVA journal.publications.chestnet.org CHEST / 143 / 4 / APRIL

4 Table 2 Univariate Predictors of SVAs in Patients With Cardiac Sarcoidosis Risk Factor No. SVAs Follow-up, Person-y Incidence Density a RR (95% CI) Left atrial enlargement ( ) Hypertension ( ) Diastolic dysfunction ( ) Pulmonary sarcoidosis ( ) Mitral valve disease ( ) Pulmonary hypertension ( ) Right atrial enlargement ( ) Tricuspid valve disease ( ) RV dysfunction ( ) Male sex ( ) Abnormal EPS ( ) Low LVEF ( ) Nonblack race ( ) Immunosuppressive therapy ( ) EPS 5 electrophysiology study; RR 5 risk ratio; SVA 5 supraventricular arrhythmia. See Table 1 legend for expansion of other abbreviations. a Per 1,000 person-y. in this patient population are largely unknown. The present study is the first to our knowledge to systematically evaluate the frequency and clinical impact of SVA on a large number of patients with CS. In this series of patients with CS with a mean follow-up of 5.8 years, we found that (1) SVAs are prevalent; (2) atrial fibrillation is the most common SVA; (3) systemic hypertension, diastolic dysfunction, RAE, and LAE are more frequent in patients with SVA; and (4) SVAs in patients with CS tend to be associated with symptoms. Clinical Implications The finding that systemic hypertension, diastolic dysfunction, and LAE predict the occurrence of SVA is not surprising. Atrial fibrillation, the most common SVA in this series, is a left atrial arrhythmia that has been associated with all of these risk factors. Additionally, it is consistent with the pathophysiology of the LAE in this group of patients, that is, systemic hypertension with or without LV hypertrophy, leading to diastolic dysfunction and chronic elevation in left atrial pressures and consequent LAE. Of note in this cohort, LV hypertrophy did not predict SVA. The study was not designed to provide an understanding of the mechanisms leading to SVA but, rather, to describe its clinical implications. Therefore, these statements should be considered speculative and hypothesis generating. Atrial tachycardias, the second most common SVA in this group of patients, can be right or left atrial in Figure 1. Incidence of SVAs (expressed as incidence density per 1,000 person-y) with various risk factors examined in the study. EPS 5 electrophysiology study; HTN 5 hypertension; LAE 5 left atrial enlargement; LVEF 5 left ventricular ejection fraction; RAE 5 right atrial enlargement; RV 5 right ventricular; SVA 5 supraventricular arrhythmia Original Research

5 Table 3 Multivariate Predictors of SVAs in Patients With Cardiac Sarcoidosis Risk Factor RR (95% CI) P Value a Left atrial enlargement 6.12 ( ).0005 Hypertension 5.03 ( ).174 Pulmonary sarcoidosis 2.69 ( ).045 Pulmonary hypertension 1.64 ( ).540 Right atrial enlargement 1.03 ( ).968 Age 0.96 ( ).090 Mitral valve disease 0.86 ( ).799 Immunosuppressive therapy 0.85 ( ).731 Nonblack race 0.48 ( ).246 Black race 0.43 ( ).077 Diastolic dysfunction 0.32 ( ).344 Intercept 0.12 ( ).090 See Table 2 legend for expansion of abbreviations. a x 2 test. origin; hence, it is reasonable to observe some association between LAE and RAE and the prevalence of atrial tachycardias. These arrhythmias may also arise as a consequence of focal granulomatous infiltration of atrial myocardium. Atrial flutter is most commonly right atrial in origin; thus, the association with RAE is also reasonable from a pathophysiologic standpoint. Of note, all the other markers of right-side heart disease (RV dysfunction, pulmonary hypertension, tricuspid valve disease) or pulmonary sarcoidosis were not associated with SVA. Because the patient population studied was relatively small, this arrhythmia may be underrepresented, as atrial flutter occurred infrequently (only five patients ). The major findings of this study call for close attention to the role of SVA in the clinical course of patients with CS. Because SVAs occur frequently, with the most common arrhythmia being atrial fibrillation, the possibility of predicting which patients are at high risk for developing SVA based on echocardiographic findings (LAE, diastolic dysfunction) and the presence of systemic hypertension requires investigation. The importance of identifying the presence of atrial fibrillation carries additional therapeutic implications regarding thromboembolic risk in these patients. In addition, SVAs can lead to inappropriate ICD therapies; it is well known that atrial fibrillation is by far the most common cause for inappropriate ICD therapies. Nevertheless, the results of this analysis suggest that SVAs do not contribute significantly to the cause of death in patients with CS. Although the data suggest that the incidence density of SVA was less among patients receiving systemic immunosuppression, a larger prospective and controlled study would be necessary to evaluate this further. Finally, because most patients with SVA had symptoms related to the arrhythmia, screening for asymptomatic patients with Holter or event monitors does not appear to be warranted. Limitations The main limitation of this analysis arises from the retrospective nature and the lack of a control group, which in turn limits the power of the analysis. A study with a control group would make the results more definitive. Therefore, the conclusions should be considered strictly hypothesis generating. Nevertheless, the strength of the data comes from the fact that this is the first report of a large sample of patients with CS studied for SVAs and its risk factors. CS is a rare disease, making a larger study cohort difficult to establish. However, our institution is a referral center for CS, and 100 patients is among the largest series that has been published to date. Although we identified risk factors for the development of SVA, prospective validation of these findings is warranted. Additionally, we cannot exclude the possibility of selection bias because the patients in this cohort were followed at the cardiac electrophysiology clinic; thus, the likelihood of detecting an arrhythmia is likely higher than in patients followed in a different clinical setting. Finally, the low event rate (only six deaths) makes it difficult to draw any conclusions about the impact of SVA on mortality in patients with CS. Conclusions The data from this single-center study of ambulatory patients with systemic sarcoidosis and cardiac involvement indicate that the prevalence of SVA is significant and the majority of patients are symptomatic. Atrial fibrillation was the most common arrhythmia seen, occurring in 18% of patients. The association of SVA with systemic hypertension, diastolic dysfunction, LAE, and RAE in univariate analysis and LAE in multivariate analysis should be validated prospectively to establish its true predictive value. Acknowledgments Author contributions: Dr Viles-Gonzalez had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Dr Viles-Gonzalez: contributed to the study design, data analysis and interpretation, manuscript writing, and manuscript submission and revisions. Dr Pastori: contributed to the data collection and preparation of the manuscript. Dr Fischer: contributed to the manuscript writing and revision. Dr Wisnivesky: contributed to the statistical analysis and preparation of the manuscript. Dr Goldman: contributed to the echocardiographic data analysis and preparation of the manuscript. Dr Mehta: contributed to the study design, interpretation of the data, and revision of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Wisnivesky is a member of the research board of EHE International and has received a grant from GlaxoSmithKline plc and a lecture honorarium from Novartis AG. Drs Viles-Gonzalez, Pastori, Fischer, journal.publications.chestnet.org CHEST / 143 / 4 / APRIL

6 Goldman, and Mehta have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. References 1. Morimoto T, Azuma A, Abe S, et al. Epidemiology of sarcoidosis in Japan. Eur Respir J ;31(2): Roberts WC, McAllister HA Jr, Ferrans VJ. Sarcoidosis of the heart. A clinicopathologic study of 35 necropsy patients (group 1) and review of 78 previously described necropsy patients (group 11). Am J Med ;63(1): Silverman KJ, Hutchins GM, Bulkley BH. Cardiac sarcoid: a clinicopathologic study of 84 unselected patients with systemic sarcoidosis. Circulation ;58(6): Baughman RP, Teirstein AS, Judson MA, et al ; Case Control Etiologic Study of Sarcoidosis (ACCESS) Research Group. Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med ; 164 ( 10 pt 1 ): Hunninghake GW, Costabel U, Ando M, et al. ATS/ERS/ WASOG statement on sarcoidosis. American Thoracic Society/European Respiratory Society/World Association of Sarcoidosis and Other Granulomatous Disorders. Sarcoidosis Vasc Diffuse Lung Dis ;16(2): Gozo EG Jr, Cosnow I, Cohen HC, Okun L. The heart in sarcoidosis. Chest ;60(4): Mehta D, Mori N, Goldbarg SH, Lubitz S, Wisnivesky JP, Teirstein A. Primary prevention of sudden cardiac death in silent cardiac sarcoidosis: role of programmed ventricular stimulation. Circ Arrhythm Electrophysiol ;4(1): Golwala H, Dernaika T. Atrial fibrillation as the initial clinical manifestation of cardiac sarcoidosis: a case report and review of the literature [published online ahead of print January 19, 2011]. J Cardiovasc Med (Hagerstown). doi: /jcm. 0b013e328343b Kim JS, Judson MA, Donnino R, et al. Cardiac sarcoidosis. Am Heart J ;157 (1 ): Wait JL, Movahed A. Anginal chest pain in sarcoidosis. Thorax ;44 (5 ): Lam CS, Tolep KA, Metke MP, Glockner J, Cooper LT Jr. Coronary sarcoidosis presenting as acute coronary syndrome. Clin Cardiol ;32 (6 ):E67-E Winters SL, Cohen M, Greenberg S, et al. Sustained ventricular tachycardia associated with sarcoidosis: assessment of the underlying cardiac anatomy and the prospective utility of programmed ventricular stimulation, drug therapy and an implantable antitachycardia device. J Am Coll Cardiol ; 18 (4 ): Reuhl J, Schneider M, Sievert H, Lutz FU, Zieger G. Myocardial sarcoidosis as a rare cause of sudden cardiac death. Forensic Sci Int ;89 (3 ): Riezzo I, Ventura F, D Errico S, Neri M, Turillazzi E, Fineschi V. Arrhythmogenesis and diagnosis of cardiac sarcoidosis. An immunohistochemical study in a sudden cardiac death. Forensic Sci Int ;183 (1-3 ):e1-e Beckert W. Sudden cardiac death due to myocardial sarcoidosis [in German]. Z Gesamte Inn Med ;34 (17 ): Fleming HA. Sarcoidosis of the heart. Am J Med ;64 (5 ): Fujita M, Sasayama S, Hoshino T, Sakurai T, Nonogi H. Myocardial sarcoidosis with severe congestive heart failure. Jpn Circ J ;51 (4 ): Original Research