Management of Arrhythmias in Cardiac Sarcoidosis

ABSTRACT

The prevalence of cardiac involvement in sarcoidosis is under‐recognized and is associated with multiple complications, including conduction block, arrhythmias, and sudden death. The comparative roles of common therapies have been inadequately studied. The purpose of this review is to examine the literature regarding treatments utilized to manage arrhythmias associated with cardiac sarcoidosis.

Introduction

Sarcoidosis is a systemic granulomatous disease of unknown etiology characterized by lymphocytic accumulation and accelerated inflammation, predominantly mediated by type 1 T helper cells.1 The lungs are most commonly affected, but nearly every organ system is susceptible to infiltration. Autopsy studies estimate that ≥25% of patients with systemic sarcoidosis exhibit cardiac involvement, and up to 5% exhibit evidence of cardiac dysfunction. Patients with cardiac sarcoidosis (CS) are at increased risk of sudden cardiac death (SCD), even without evidence of prior cardiac dysfunction.2 The survival rate among patients with CS is controversial, given a large variation in reported data, but a recent outcomes study from Finland suggests that long‐term survival has improved since the 1970s and 1980s, largely due to improvements in detection and treatment.3

Diagnosis of CS can be challenging, largely due to its protean manifestations, coupled with the lack of a gold‐standard test. Histologic diagnosis remains the only definitive method for a pathologic diagnosis, although endomyocardial biopsy has a poor sensitivity of approximately 20%, due to patchy distribution of disease.2 A number of criteria have been proposed to aid in the diagnosis of CS (Table 1).4

Table 1.

2014 HRS Recommendation on Criteria for Diagnosis of CS3

1. Histological Diagnosis From Myocardial Tissue

CS is diagnosed in the presence of noncaseating granuloma on histological examination of myocardial tissue with no alternative cause identified (including negative organismal stains if applicable).

2. Clinical Diagnosis From Invasive and Noninvasive Studies

It is probable that there is CS if:

A. There is a histological diagnosis of extracardiac sarcoidosis and

B. ≥1 of the following is present:

Steroid ± immunosuppressant responsive cardiomyopathy orheart block

Unexplained reduced LVEF (<40%)

Unexplained sustained (spontaneous or induced) VT

Mobitz type II second‐ or third‐degree heart block

Patchy uptake on dedicated cardiac PET (in a pattern consistentwith CS)

Late gadolinium enhancement on CMR (in a pattern consistentwith CS)

Positive gallium uptake (in a pattern consistent with CS) and

C. Other causes for the cardiac manifestation(s) have been reasonably excluded.

Abbreviations: CMR, cardiac magnetic resonance imaging; CS, cardiac sarcoidosis; HRS, Heart Rhythm Society; LVEF, left ventricular ejection fraction; PET, positron emission tomography; VT, ventricular tachycardia.

A variety of treatment modalities are utilized for primary and secondary prevention of arrhythmias and SCD, including corticosteroids, immune‐modulatory therapies, antiarrhythmic medications, catheter ablation, implantable cardioverter‐defibrillators (ICDs), and heart transplantation. Many of these treatments are employed despite a paucity of prospective clinical trials supporting their use.5 This manuscript reviews the literature supporting treatment of arrhythmias caused by CS.

Alterations in Cardiac Conduction

Conduction Block

Sarcoidosis often affects the cardiac conduction system and causes a variety of clinical presentations ranging from normal electrocardiograms to complete atrioventricular (AV) block.6 Granuloma formation in the basal interventricular septum is thought to be the major mechanism of conduction disorders, with the location and burden of disease of the AV nodal arterial supply, the AV node itself, and/or the His‐Purkinje system determining the degree of dysfunction.7, 8, 9, 10

Complete AV block and bundle branch block are among the most common conduction abnormalities seen in patients with CS. The lifetime risk of developing AV block and bundle branch block is 26% to 67% and 12% to 61%, respectively.11 It is estimated that CS is responsible for up to 19% of all new‐onset unexplained AV block in adults age <55 years.12 Recently, Takaya et al showed that CS patients with an initial presentation of high‐degree AV block had a similar mortality over median follow‐up of 34 months compared with those with initial presentation of ventricular tachycardia (VT) or congestive heart failure (P = 0.877).13

AV nodal dysfunction is more common during active phases of disease when granuloma formation and surrounding edema are evolving. Inflammation of the myocardium is more likely to cause AV‐His dysfunction compared with periods of inactive disease, during which myocardial scar is predominant and predisposes patients to ventricular arrhythmias.8

Atrial Arrhythmias

Few studies have investigated the prevalence of atrial arrhythmias in patients with CS. Villes‐Gonzalez et al reported that 32% of CS patients developed a supraventricular arrhythmia during a mean 5.8‐year follow‐up, and 23% of these arrhythmias were due to either atrial fibrillation or flutter.14 These findings are corroborated by other studies.7 The lifetime risk of developing atrial fibrillation or flutter in the general population has been estimated at 26% in men and 23% in women, similar to the prevalence in the CS population.15

Proposed mechanisms of arrhythmogenesis involve substrate heterogeneity caused by granuloma formation, which is more common and more dense in the left atrium,16 or elevated atrial pressures. One study showed that the only significant risk factor associated with atrial arrhythmias in CS patients is left atrial enlargement, with a hazard ratio of 6.12 (P < 0.01).14 A recent study has suggested that elevated ventricular diastolic pressures and remodeling of the tricuspid annulus caused by pulmonary hypertension is a significant risk factor for atrial flutter in CS patients.16 Zipse et al demonstrated that non–atrial fibrillation atrial arrhythmias are caused by triggered activity in 11% of cases, abnormal automaticity in 47% of cases, and reentry in 42% of cases.7

Ventricular Arrhythmias and Sudden Cardiac Death

Ventricular arrhythmias are a common and life‐threatening manifestation of CS and may be an early or presenting feature of the disease. The most common ventricular arrhythmia is monomorphic VT.8 The prevalence of ventricular arrhythmias in CS patients has been estimated at 23%, and some long‐term follow‐up studies suggest that the prevalence is as high as 50%.17 One autopsy study of patients with biopsy‐proven CS and evidence of cardiac dysfunction showed that SCD was the most common cause of death, occurring in 67% of patients.18

Risk factors for ventricular tachyarrhythmias include infranodal conduction disease, ventricular pacing, decreased left ventricular ejection fraction (LVEF), and inducible ventricular arrhythmias.19 In addition, the presence of late gadolinium enhancement (LGE) on cardiac magnetic resonance imaging (CMR) and myocardial fluorodeoxyglucose (FDG) activity on positron emission tomography (PET) scan are both independent predictors for the development of ventricular arrhythmias (Figure 1).20, 21 A recent retrospective study of patients with a LVEF >35% and CMR‐proven CS suggested that delayed gadolinium enhancement involving >3.5% of LV mass was 90% sensitive and 97% specific for identifying patients with CS. Multifocal delayed enhancement had a positive predictive value of 77% and a negative predictive value of 100% for future events of VT and/or ventricular fibrillation.22

Figure 1.

Representative short‐axis (A, C) and horizontal long‐axis (B, D) cardiac FDG PET (top) and magnetic resonance images (bottom) in a patient with CS. Notice the presence of matching increased FDG activity (A, B) and LGE (C, D) at the inferior septum, anterior‐septum, lateral wall, and apex (arrows). Abbreviations: CS, cardiac sarcoidosis; FDG, fluorodeoxyglucose; LGE, late gadolinium enhancement; PET, positron emission tomography. Courtesy of Charles Steenbergen, MD, PhD, Johns Hopkins Hospital, Baltimore, Maryland.

Granulomas may be the substrate for reentry circuits or the focus for abnormal automaticity. Electrophysiology studies have demonstrated transient entrainment in monomorphic VT in CS patients, suggesting that macroreentrant circuits are the most common mechanism leading to VT.23 This finding was corroborated by other studies that suggest that up to 68% of monomorphic VT is caused by reentry circuits, similar to what is found in postinfarction VT.24

Studies differ with regard to VT inducibility in active vs inactive phases of disease. Furushima et al showed that most reentrant VTs were associated with active periods of disease, suggesting that the slow conduction zone may be growing during both active and inactive areas of disease.24 However, Mezaki et al reported that VT inducibility was not associated with disease activity.25

Treatment

Corticosteroids

Corticosteroids are the mainstay of therapy for the treatment of cardiac events caused by CS. Nuclear studies have shown an 80% correlation between new‐onset AV block and positive gallium‐67 uptake, suggesting that conduction abnormalities develop during the inflammatory phase of disease, which supports immunosuppression as a valid therapy.8 Multiple case series and meta‐analyses following CS patients for up to 7 years found that 47% to 57% patients treated with corticosteroids regained normal AV conduction or first‐degree AV block, compared with no recovery in untreated individuals.8, 9, 10

Patients with CS not treated with corticosteroids develop progressively worsening systolic function compared with treated patients. Kato et al showed that patients who initially presented with AV block with a normal LVEF did not have a significant decrease in LVEF if treated with corticosteroids, whereas untreated patients showed a marked decline in LVEF (60.5% ± 6.4% vs 37.6% ± 17.3; P < 0.005) over an average of 6.7 years.9 Other studies suggest that early treatment with corticosteroids slows the progression of LV dysfunction, and this effect is more pronounced when treatment is begun early in the disease process.26

The effect of corticosteroids on the burden of VT has been inconsistent likely due to the heterogeneity of studies. As previously mentioned, the mechanism of VT in the majority of CS patients is reentry caused by myocardial scar, which is less susceptible to improvement with corticosteroids. Some studies of patients with ICDs showed that patients receiving steroids did not have a reduction in ventricular arrhythmias.8, 27 Conflicting data have shown a reduction in the incidence of VT. Kato et al reported that 1 of 7 patients treated with corticosteroids experienced VT, but VT occurred in 8 of 13 untreated patients.9 Futamatsu et al showed a reduction in episodes of VT with corticosteroids, but patients in this study were concurrently treated with amiodarone.28 Yodogawa et al showed a reduction in ventricular premature contractions and nonsustained VT in patients with LVEF >35%.10

Immunomodulatory Therapy

Alternatives to corticosteroids are often considered in patients who experience severe side effects or refractory disease, although the efficacy of these therapies in CS is poorly understood. The most widely used immunosuppressive agent utilized for CS is methotrexate, due to efficacy in treating extracardiac sarcoidosis.29 One prospective study examined CS patients treated with low‐dose corticosteroids and methotrexate and found statistically significant sparing of LVEF (60.7 ± 14.3% vs 44.5 ± 13.8%; P = 0.04) and lower serum N‐terminal prohormone brain natriuretic peptide (NT‐proBNP) levels compared with patients treated with corticosteroids alone up to 5 years after initiation of therapy.29 In approximately one‐quarter of these patients, corticosteroids were successfully weaned. Leflunomide, a methotrexate analogue, has been proposed as an alternative to methotrexate with less toxicity, but clinical data supporting its use are limited.30

Azathioprine has been tested in number of small open‐label case series and has shown promise as a steroid‐sparing agent for treatment of CS.31 Cyclophosphamide was shown in one case report to suppress CS in a patient refractory to steroids,32 but these results have not been replicated in the literature.

In addition, infliximab has shown to be efficacious in refractory sarcoidosis, including multiple case reports of CS and a randomized, double‐blinded clinical trial of patients with extrapulmonary sarcoidosis.33 However, infliximab has been shown to worsen congestive heart failure and is prohibited for use in patients with New York Heart Association class III and IV symptoms.33 Adalimumab has been reported to be effective in sarcoidosis with less chance of allergic side effects, but it has not been studied in CS.34 Etanercept is not recommended due to multiple reports of lack of efficacy, side‐effect profile, and reports of activation of sarcoidosis compared with other TNF‐α inhibitors.35

Other immune‐modulatory therapies have been considered for adjunct treatment of CS based on documented efficacy in pulmonary or extrapulmonary sarcoidosis, including antimalarial agents such as chloroquine and hydroxychloroquine; mycophenolate mofetil; cyclosporine; thalidomide; and pentoxifylline.36

Antiarrhythmic Drugs

The use of antiarrhythmic drugs in conjunction with other therapies is common, but the efficacy of these treatments has been poorly studied. Due to the variability in presentation of CS, antiarrhythmic therapy is chosen on an individual basis.

The most common medications used include β‐blockers and class III agents, with amiodarone and sotalol used most frequently. Class Ic drugs are often avoided in patients with CS due to the presence of structural heart disease, although case series have reported that multidrug therapy with flecainide has been efficacious in suppression of right‐sided ventricular arrhythmias in patients with CS.37

Catheter Ablation

Medical therapy is effective in suppressing VT in approximately 50% of patients with CS.2 Catheter ablation is often used in refractory cases of VT, although efficacy is highly variable. In one case series, Koplan et al found that 6 of 8 patients had recurrent VT after catheter ablation, 4 of whom were referred for cardiac transplantation.38 Jefic et al studied a similar population, albeit with a higher LVEF and higher prevalence of lesions localized in the right ventricle (RV). Patients were noted to have a decrease in arrhythmic events from 271 ± 363 to 4 ± 9.7 after ablation, and all patients showed a decrease or elimination of VT during follow‐up of 19.8 ± 19.6 months.39 They found that right‐ventricular VTs were more likely to be successfully ablated than left‐ventricular VTs, possibly because therapeutic transmural lesions are easier to accomplish in the thinner‐walled RV.39

Little has been reported regarding ablation of atrial arrhythmias and CS. One recent study that examined outcomes in this setting found that catheter ablation was immediately successful in 7 of 9 patients, 2 of whom had recurrent atrial arrhythmias after mean follow‐up of 1.8 ± 1.9 years.16

Implantable Cardioverter‐Defibrillators

Use of ICDs is common in patients with CS due to increased risk of life‐threatening ventricular arrhythmias. Implementation of ICD is indicated for secondary prevention, but use in primary prevention often requires clinical judgment. Multiple retrospective observational studies have reported clinical outcomes of ICD therapy. The annual incidence of appropriate shocks in CS patients with an ICD placed for primary prevention of SCD is 10% to 15%, a 3‐fold higher incidence than other primary‐prevention ICD trials.17

Risk stratification for SCD is difficult. Reported risk factors for SCD in CS patients include decreased LVEF and RVEF, presence of LGE on CMR,21 myocardial FDG activity on PET,20 ventricular pacing, and a history of syncope.19 A recent study suggests that isolated CS may be a risk factor for SCD, as 9 of 13 (69.2%) patients with isolated CS received appropriate ICD therapy, compared with 75 of 222 (33.8%) patients with cardiac and extracardiac sarcoidosis (P = 0.0192).40

Multiple studies have shown an increased risk of SCD in CS patients with decreased systolic function.19 Schuller et al showed that decreased LVEF (odds ratio [OR]: 6.52, 95% confidence interval [CI]: 2.43‐17.5), decreased RVEF (OR: 6.73, 95% CI: 2.69‐16.8), and symptomatic heart failure (OR: 4.33, 95% CI: 1.86‐10.1) are each independent risk factors for appropriate ICD therapy.41

Late gadolinium enhancement on CMR and FDG activity on PET have become important predictors of adverse cardiovascular events, including ICD therapies, in CS patients. Greulich et al studied 155 patients with suspected CS who underwent CMR.21 Compared with patients without LGE (n = 114), they observed that patients with LGE (n = 39) had a significantly higher frequency of VT (0 vs 6 cases; P < 0.0001), aborted SCD (0 vs 4 cases; P < 0.0001), ICD shocks (0 vs 7 cases; P < 0.0001), and deaths (1 vs 3 cases; P < 0.0001) during a median follow‐up of 2.6 years.21 Similarly, Blankstein et al observed that, compared with patients with normal cardiac FDG/perfusion PET (n = 47), patients with abnormal myocardial FDG uptake and/or perfusion (n = 71) had a significantly higher number of VT events requiring device‐related therapies (4 vs 27 events) after a median 1.5 years of follow‐up.20

The 2008 American College of Cardiology/American Heart Association/Heart Rhythm Society device guidelines acknowledge these difficulties with risk stratification of SCD in patients with CS, stating that “[c]onsideration should be given to symptoms such as syncope, heart failure status, [left ventricular] function, and spontaneous or induced ventricular arrhythmias at electrophysiological study to make individualized decisions regarding use of the ICD for primary prevention of SCD.”5 The 2014 Heart Rhythm Society consensus statement recommendations for ICD implantation in CS patients are summarized in Table 2.4, 42

Table 2.

2014 HRS Recommendations for ICD Implantation in CS Patients

Class I	ICD implantation is recommended in patients with CD and ≥1 of the following:
	1. Spontaneous sustained ventricular arrhythmias, including prior cardiac arrest
	2. LVEF ≤35%, despite optimal medical therapy and a period of immunosuppression (if there is active inflammation)
Class IIa	ICD implantation can be useful in patients with CS, independent of ventricular function, and ≥1 of the following:
	1. An indication for permanent pacemaker implantation
	2. Unexplained syncope or near‐syncope, felt to be arrhythmic in etiology
	3. Inducible sustained ventricular arrhythmias (>30 seconds of monomorphic VT or polymorphic VT) or clinically relevant VF
	If VT ablation is planned, an indicated ICD should be implanted after ablation.
Class IIb	ICD implantation may be considered in patients with LVEF in the range of 36%–49% and/or an RVEF <40% despite optimal medical therapy for heart failure and a period of immunosuppression (if there is active inflammation).
Class III	ICD implantation is not recommended in patients with no history of syncope, normal LVEF/RVEF, no LGE on CMR, a negative EP study, and no indication for permanent pacing. However, these patients should be closely followed for deterioration in ventricular function.
	ICD implantation is not recommended in patients with ≥1 of the following:
	1. Incessant ventricular arrhythmias
	2. Severe NYHA class IV heart failure

Abbreviations: CD, cardiac dysfunction; CMR, cardiac magnetic resonance imaging; CS, cardiac sarcoidosis; EP, electrophysiologic; HRS, Heart Rhythm Society; ICD, implantable cardioverter‐defibrillator; LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; RVEF, right ventricular ejection fraction; VF, ventricular fibrillation; VT, ventricular tachycardia.

Conclusion

Cardiac involvement is an under‐recognized and poorly understood manifestation of sarcoidosis associated with high morbidity and mortality. The presentation of CS can be highly variable, ranging from asymptomatic electrocardiographic changes to SCD. Studies investigating treatment modalities including corticosteroids, immune‐modulatory therapies, antiarrhythmic medications, and catheter ablation are often contradictory due to small and heterogeneous populations. Implantable cardioverter‐defibrillators are often placed to reduce the risk of SCD, although predictive risk markers in primary prevention are unclear. Further research is needed to evaluate the role and efficacy of these treatments to prevent the high morbidity associated with CS.