Arrhythmias in cardiac sarcoidosis

Contact address: Alexandru B. Chicos, MD Associate Professor, Cardiac Electrophysiology, Bluhm Cardiovascular Institute, Feinberg School of Medicine, Northwestern University, Chicago, USA. E-mail: achicos@nm.org 1 Department of Cardiac Electrophysiology, Bluhm Cardiovascular Institute, Feinberg School of Medicine, Northwestern University, Chicago, USA. sis was 1.3% for women and almost 1% for men. However, the disease is encountered worldwide and is probably more prevalent than thought, including in Central and Eastern Europe. Deubelbeiss et al4 found a prevalence of sarcoidosis in Switzerland that was higher than assumed based on previous international estimates. The prevalence of lifetime and currently active sarcoidosis were 121 and 44 per 100.000 inhabitants, respectively. The mean annual incidence of sarcoidosis was 7/100.000. An Italian study5 found a prevalence of sarcoidosis estimated at 48/100,000 in the Parma region for the years 2010-2013, much higher than previously thought, and as high as 196/100,000 in some areas. A study6 in Columbus, Ohio, found an unexpectedly high prevalence, at least 50/100,000 and possibly as high as 200/100.000, in population that was predominantly Causian (White, 74%). In the Romanian Registry for Interstitial Lung Diseases (REGIS)1,2, 144 patients were enrolled over 3 years, of which 28 were diagnosed with sarcoidosis, most with mediastinal and/ or pulmonary involvement. Authors concluded that, in Romania, interstitial lung diseases, including sarcoidosis, are under-diagnosed and under-reported, and that diagnosis is often made in advanced stages of disease despite availability of diagnostic tools, such as bronchoalveolar lavage and biopsy, in dedicated centers.


INTRODUCTION
Cardiac sarcoidosis (CS) is an under-diagnosed and under-treated etiology of cardiac arrhythmias and sudden death. This article reviews, using a series of case vignettes, some of the clinical presentations in cardiology and the challenges in diagnosis and treatment, particularly focused on arrhythmia.
detrimental to embark on long-term immunosuppression without confi dence in the diagnosis. Diagnostic sets of criteria have been proposed by the Japanese Circulation Society (JCS) 8 , Heart Rhythm Society (HRS) 9 , and the World Association for Sarcoidosis and Other Granulomatous Disorders (WASOG) ( Table 1 and Table 2).
The majority of cases are diagnosed by typical cardiac involvement combined with extra-cardiac biopsy. Cardiac biopsy has a relatively low-yield, but can be useful in selected patients. Tissue diagnosis should always be considered. Other causes of non-caseating granulomas should be excluded. There are some situations when the pattern of involvement is very characteristic, for example bilateral hilar lymphadenopathy with increased FDG-PET uptake in the lymph nodes as well as patchy myocardial uptake. These are exceptions when, if tissue diagnosis cannot be obtained, a presumptive diagnosis of cardiac sarcoidosis can be made. The updated Japanese criteria 8 allow for clinical diagnosis, in selected situations, in the absence of tissue diagnosis.

PATHOPHYSIOLOGY OF ARRHYTHMIAS IN CARDIAC SARCOIDOSIS
The pathophysiology of arrhythmias in cardiac sarcoidosis is related to the infl ammatory infi ltrates. "Early lesions" consist of lymphocytic myocarditis. Lesions containing granulomas (non-necrotizing) along with lymphocytic myocarditis have been termed "intermediate". In the chronic stage, areas of active infl ammation have caused myocyte loss and repair fi brosis, and tom onset to diagnosis of cardiac sarcoidosis was 16.1 +/-34 months 7 . Access to advanced cardiac imaging such as MRI and FDG-PET is one limitation. Assuming a prevalence of currently active cases of sarcoidosis in Romania of 10-50/100.000, this would translate to ~2.000 to 10.000 patients with active sarcoidosis. An estimated 5% of them could have manifest cardiac involvement (100-500 patients) and up to 50% (up to 1.000-5.000 patients) may have apparently asymptomatic cardiac involvement. It is therefore essential to screen for cardiac involvement if a diagnosis of sarcoidosis involving any organ system is made, and also to consider and look for sarcoidosis if the patient presents with unexplained cardiac symptoms or fi ndings. This is particularly important because cardiac sarcoidosis is treatable, but, if left untreated, it can lead to signifi cant morbidity and death from arrhythmia or progressive myocardial damage.

CLINICAL MANIFESTATIONS AND DIAGNOSIS
Sarcoidosis is a systemic disease, and can involve any organ system. It can present with a broad spectrum of symptoms, or it can be asymptomatic and diagnosed incidentally 8 . Its cardiac manifestations depend on the location, extent and degree of infl ammation, and include conduction abnormalities, ventricular arrhythmia and heart failure. Figure 1 illustrates this in a patientfriendly schematic representation. For the diagnosis of cardiac sarcoidosis, it is important to have keep a high index of suspicion and try as much as possible to obtain tissue diagnosis. It is diffi cult and potentially

CASE VIGNETTE #1:
34-year old athletic Caucasian male presents with new exertional dyspnea and lightheadedness. His ECG is shown in Figure 2. Note the markedly prolonged PR interval to ~360 ms (1 st degree atrioventricular (AV) block). Figure 3 shows rhythms strips obtained during exercise, showing a sudden drop in heart rate to approximately half. The strips show 2 nd degree type II AV block resulting in a dropped 7 th QRS, followed by 2:1 AV block at peak exercise, associated with marked lightheadedness. Careful examination of the rate and the T waves identifi es blocked P waves on top of the T waves, which look slightly different when compared to T waves during 1-to-1 AV conduction. These conduction abnormalities are, of course, markedly abnormal in a 34 year old and they will prompt further investigations. One of these investigations should be a cardiac have been largely replaced by scar. However, different lesions and areas of lesions can be at different stages, and they evolve over time. Thus, the substrate for arrhythmia is very complex. It consists of areas with different degrees of active infl ammation (acute, chronic, low-grade, high-grade), fi brosis and scar. Moreover, the arrhythmic substrate is dynamic: infl ammation waxes and wanes, fl ares up. Importantly, myocardial reactivations frequently go undetected until arrhythmia or left ventricular systolic dysfunction become apparent. The location of lesions can be patchy, mesocardial, epicardial or endocardial. Lesions have a border zone that is typically irregular and heterogenous, and can also involve local Purkinje fi bers. This substrate for ventricular arrhythmia is very different from the classical scar we see, for example, due to chronic myocardial infarction. With increased awareness, access to advanced imaging and earlier screening and diagnosis, it is likely that the patient population is changing from the cohorts reported >10 years ago -patients might

Diagnostic Guidelines for Cardiac Sarcoidosis Clinical fi ndings defi ning cardiac involvement
Cardiac fi ndings should be assessed based on the major criteria and the minor criteria. Clinical fi ndings that satisfy the following 1) or 2) strongly suggest the presence of cardiac involvement. 1) Two or more of the fi ve major criteria (a) to (e) are satisfi ed.
2) One in the fi ve major criteria (a) to (e) and two or more of the three minor criteria (f) to (h) are satisfi ed. Criteria for cardiac involvement 1. Major criteria a) High-grade atrioventricular block (including complete atrioventricular block) or fatal ventricular arrhythmia (e.g., sustained ventricular tachycardia, and ventricular fi brillation) b) Basal thinning of the ventricular septum or abnormal ventricular wall anatomy (ventricular aneurysm, thinning of the middle or upper ventricular septum, regional ventricular wall thickening) c) Left ventricular contractile dysfunction (left ventricular ejection fraction less than 50%) or focal ventricular wall asynergy d) 67Ga citrate scintigraphy or 18F-FDG PET reveals abnormally high tracer accumulation in the heart e) Gadolinium-enhanced MRI reveals delayed contrast enhancement of the myocardium 2. Minor criteria f) Abnormal ECG fi ndings: Ventricular arrhythmias (nonsustained ventricular tachycardia, multifocal or frequent premature ventricular contractions), bundle branch block, axis deviation, or abnormal Q waves g) Perfusion defects on myocardial perfusion scintigraphy (SPECT) h) Endomyocardial biopsy: Monocyte infi ltration and moderate or severe myocardial interstitial fi brosis Diagnostic guidelines for cardiac sarcoidosis 1) Histological diagnosis group (those with positive myocardial biopsy fi ndings) Cardiac sarcoidosis is diagnosed histologically when endomyocardial biopsy or surgical specimens demonstrate non-caseating epithelioid granulomas (See Note (6)). 2) Clinical diagnosis group (those with negative myocardial biopsy fi ndings or those not undergoing myocardial biopsy) The patient is clinically diagnosed as cardiac sarcoidosis (1) when epithelioid granulomas are found in organs other than the heart, and clinical fi ndings strongly suggestive of the above-mentioned cardiac involvement are present; or (2) when the patient shows clinical fi ndings strongly suggestive of pulmonary or ophthalmic sarcoidosis; at least 2 of the fi ve characteristic laboratory fi ndings of sarcoidosis (Table below); and clinical fi ndings strongly suggest the above-mentioned cardiac involvement. Characteristic laboratory fi ndings of sarcoidosis 1) Bilateral hilar lymphadenopathy 2) High serum angiotensin-converting enzyme (ACE) activity or elevated serum lysozyme levels 3) High serum soluble interleukin-2 receptor (sIL-2R) levels 4) Signifi cant tracer accumulation in 67Ga citrate scintigraphy or 18F-FDG PET 5) A high percentage of lymphocytes with a CD4/CD8 ratio of >3.5 in BAL fl uid granulomas and stains and cultures were negative for fungi and acid-fast bacilli (AFBs).
This case illustrates the need to investigate cardiac conduction abnormalities, especially AV block, if no obvious explanation is presenting itself. In a Finnish series of 133 patients aged 18 to 55 years who received pacemakers for second-or third-degree AV block, 72 patients had initially unexplained AV block. Of these, cardiac sarcoidosis was found in 19% 11 . In MRI, as well as other tests looking for sarcoidosis. This patient had mediastinal lymphadenopathy and a small area of focal myocardial late Gadolinium enhancement (LGE) on cardiac MRI in the inferolateral wall ( Figure  4). The MRI did not detect any lesions that explain the AV block. Involvement of the AV conduction system can be due to microscopic lesions that are below the detection threshold of our current imaging modalities. A transbronchial lung biopsy showed non-caseating  Causes of AV block should be investigated in patients younger than 60 or 65 years old, if no obvious cause is present, such as aortic valve replacement, aortic valvular extensive calcifi cation, acute myocarditis or endocarditis etc (see Table 3). We routinely obtain a cardiac MRI prior to device insertion if pos-another series of 32 patients from Ontario, Canada aged 18-60 years who presented with unexplained 2 nd or 3 rd degree AV block and no previous history of sarcoidosis in any organ, 34% were diagnosed with cardiac sarcoidosis, all of which also had extra-cardiac involvement 12 .

CASE VIGNETTE #2
36 year old previously healthy Caucasian male presents with exertional palpitations. He is sent for a treadmill stress test. A 12-lead ECG obtained during exercise is shown ( Figures 5 and 6). Note frequent PVCs with right bundle branch block-like morphology and superior axis suggesting a focus in the inferior left ventricle. The PVCs are relatively narrow and differential diagnosis at this point could include idiopathic, benign fascicu-sible, and if the MRI suggests cardiac sarcoidosis, we discuss with the patient the option of a pacemaker versus defi brillator. MRI can also be obtained after device insertion; we usually allow at least 2-3 months after device placement and use software to fi lter the artifact introduced by the metallic generator can. The devices are turned to asynchronous mode during MR imaging and parameters and programming are checked before and after MRI.  lar PVCs. Note, however, the prominent inferior Q waves, with Q amplitude greater than 1/3 rd of the R wave. This suggests the presence of inferior wall scar. In Figure 6 we see a burst of paroxysmal VT with variable cycle length and somewhat pleomorphic QRS. This is suggestive of a focal tachycardia due to increased automaticity and is not a typical presentation of left posterior fascicular VT. The echocardiogram showed normal left ventricular ejection fraction (LVEF). A cardiac MRI, shown in Figure 7, also showed a normal LVEF of 61%, but showed areas of mid-myocardial and epicardial delayed enhancement in infero-septum and LV inferior wall from mid-chamber to the apex, also with involvement of the adjacent inferior wall of the right ventricle. He had bilateral hilar lymphadenopathy and transbronchial biopsy was positive for non-caseating granulomas (stains and cultures were negative for fungi and AFBs).

CASE VIGNETTE #3:
57 year old Caucasian female presents with syncope. Her presenting ECG is shown in Figure 8. Note the bifascicular block (right bundle branch block and left  ght ventricles. Transbronchial biopsy showed non-caseating granulomas (stains and cultures were negative for fungi and AFBs). She was aggressively treated with immunosuppressive therapy with PET-guided titration (see below) and, 8 more years later, her LVEF is 29% and has not had any more sustained VT/VF or ICD shocks. This case presents a sequence of events that is quite common and illustrates the importance of screening for cardiac sarcoidosis in patients presenting with unexplained heart block (see Table 3). It also illustrates the role of anti-infl ammatory, immunosuppressive therapy in the treatment of ventricular arrhythmias in the setting of active myocardial infl ammation, and the increase in LVEF that can frequently be expected if therapy is instituted before permanent damage and scar are present.
anterior fascicular block) and prolonged PR interval. In the hospital, transient complete heart block was noted on telemetry and a dual-chamber pacemaker was implanted. 18 months later, she complained of shortness of breath. Pacemaker interrogation showed that she was 100% paced in the right ventricle and her LVEF was 21%. A coronary angiogram was normal. Her pacemaker was upgraded to a biventricular ICD. 6 months later, she received 2 appropriate ICD shocks for ventricular fi brillation and an Electrophysiology second opinion was obtained. Her LVEF was now 15%. An FDG PET-CT was obtained, after preparatory diet to suppress myocardial glucose uptake and switch myocyte metabolism to fatty acids (protein and fat diet, no carbohydrates for 24-72 hours). This showed extensive foci of hypermetabolic lymphadenopathy in the mediastinum and hilar regions bilaterally, as well as extensive patchy uptake in both left and ri- conference held at the hospital. The ECG is shown in Figure 9. The ECG showed non-specifi c, but quite notable, T wave inversions in the inferior and lateral leads and she was referred for further evaluation. Her echocardiogram showed normal LVEF=55%, but severe hypokinesis of the inferolateral LV wall was no-

CASE VIGNETTE #4
45 year old Hispanic female who has been treated by Ophthalmology for stable sarcoid uveitis. She was asymptomatic, but a screening ECG was obtained according to practice guidelines discussed in the quarterly multidisciplinary, interdepartmental sarcoidosis  of positive FDG-PET in patients with ARVC has been reported by other investigators, as well 13 . This case illustrates that not all positive myocardial FDG-PET scans represent cardiac sarcoidosis. Differential diagnosis includes some genetic cardiomyopathies, which may have areas of necrosis triggering local infl ammation; other myocarditis etiologies, including viral, lymphocytic, giant cell myocarditis, etc. Importantly, recent radiofrequency ablation, in the preceding 3-6 months, can confound the picture by ted. A cardiac MRI was obtained, which showed LGE wall thinning and severe hypokinesis in the apical and mid-inferolateral LV wall. LVEF measured by MRI was 42%. A 3-lead outpatient cardiac monitor also showed frequent PVCs, ~5,000 in 48 hours, with a superior axis QRS (Figure 10), compatible with a focus at the location of LGE on MRI. This case illustrates the importance of screening for cardiac involvement in patients with known sarcoidosis involving other organ systems (see Table 4).

CASE VIGNETTE #5:
A 42 year old Caucasian male sees his doctor for screening because his mother has congestive heart failure and an idiopathic, non-ischemic cardiomyopathy. His LVEF is 40-45% by echocardiogram. A cardiac MRI showed multiple areas of patchy LGE (interventricular septum, anterolateral LV wall). An FDG-PET CT was also obtained. See Figure 11: there was increased FDG uptake in the same areas noted on MRI. Familial or household clusters of sarcoidosis have been reported. However, this patient had no lymphadenophy and no extra-cardiac FDG uptake. A genetic test was obtained and it revealed a pathogenic desmoplakin mutation, consistent with a diagnosis of ARVC (arrhythmogenic right ventricular cardiomyopathy). The possibility

ROLE OF IMPLANTABLE DEFIBRILLATORS
Once a diagnosis of cardiac sarcoidosis is made, it is important to assess the patient's risk of sudden death and whether an implantable cardioverter-defi brillator (ICD) is indicated (see Table 5).
causing increased FDG uptake in the ablated areas. Careful clinical correlation should be made. Isolated cardiac sarcoidosis is relatively rare. In a international registry, 32 of 275 (12%) patients with cardiac sarcoidosis had isolated cardiac involvement 7 -and this is likely an over-estimate. This is one of the reasons why tissue diagnosis is important. It has been reported that bronchoaleolar lavage and transbronchial biopsy are frequently positive even if the PET scan does not show increased uptake.

ROLE OF ELECTROPHYSIOLOGY STUDY (EPS) IN RISK STRATIFICATION
The role of programmed ventricular stimulation was investigated in a study 14 (Figure 7), however, note that two imaging modalities show different things. Late Gadolinium enhancement (LGE) represents scar or late infl ammation, while FDG uptake is a functional imaging modality and refl ects metabolic activity of the infl ammatory cells. We commonly see areas of LGE that are negative for FDG uptake and areas of positive FDG uptake without LGE.
This case illustrates that cardiac sarcoidosis is dynamic and unpredictable, and close surveillance for response to therapy and monitoring for reactivation is necessary.
This patient went on to have multiple ICD shocks for VF and VT despite aggressive medical therapy with high doses prednisone (40-60 mg/day) and methotrexate, in addition to amiodarone 200-400 mg/day and mexiletine 150 mg TID. His arrhythmias did subside eventually and LVEF improved to 25%. It has been reported that cardiac sarcoidosis is highly arrhythmogenic, more so than fi xed scar (such as healed myocardial infarction), likely due to the multifocal, infl ammatory, evolving nature of the proarrhythmic substrate. In a cohort 16 of 235 patients followed for 4.2±4.0 years, 36% received appropriate ICD therapy. One should have a relatively lower threshold for recommending ICD in these patients. Many or most patients will probably benefi t from an ICD, including those with normal LVEF but who have signifi cant LGE on MRI or positive EP study (see Table 5). It may be easier to identify those who most likely will not benefi t from ICD -see Table 6.
cardiac sarcoidosis was diagnosed based on biopsyproven extracardiac sarcoidosis in combination with abnormal PET or MR consistent with cardiac involvement 9 . In this cohort of patients with probable CS and preserved LV and RV function, a positive EPS was found to be predictive of sudden cardiac death or ventricular arrhythmias, while a negative EPS was associated with low incidence of these outcomes.

CASE VIGNETTE #6 (FOLLOW-UP OF THE SAME PATIENT DESCRIBED IN CASE #2)
The patient underwent an EP study, which was technically negative for inducible sustained VT (only "nonsustained" VT was induced, up to 19 seconds). He was then treated with prednisone, initially 60 mg daily starting in January 2010, followed by a slow taper until to October 2012. He was asymptomatic, competing in triathlons. He returns in September 2013 after a lapse in follow up. He described becoming tired perhaps more easily, but still training for triathlon. ICD interrogation showed that he had 10 seconds of VF with aborted ICD shock ( Figure 12). It was also noted that the sensed R wave amplitude had decreased (to 4-5 mV, down from 12 mV). The sensed R waves had been stable at ~12 mV until February 2013. A transthoracic echocardiogram was obtained in September 2013, which showed LVEF had now decreased to 15% (from 65%). A myocardial FDG-PETCT in October 2013 showed extensive ventricular myocardial uptake ( Figure 13). costeroid are the mainstay of therapy. Steroid-sparing agents are frequently added as the steroids are tapered down. More detailed discussions of immunosuppression for cardiac sarcoidosis can be found elsewhere 18 . Long-term immunosuppressive therapy is a complex treatment requiring specialized skill and should be directed by physicians with expertise and experience, usually rheumatologists, pulmonologists with experience managing pulmonary sarcoidosis, or cardiologists with experience managing heart transplant patients. I will mention here the potential role of FDG-PET guidance in titrating and adjusting immunosuppressive therapy. This is done in the idea that suppression of FDG uptake correlates with suppression of infl ammation, which might prevent progressive myocardial damage and facilitate healing of reversible injury 19 . Figure 14 shows serial FDG-PETCT images obtained in one patient. He was initially treated with prednisone 60 mg daily (some reports suggest that initial prednisone dose of 30 mg daily might be equally good).

TREATMENT OF VENTRICULAR ARRHYTHMIAS IN CARDIAC SARCOIDOSIS
In addition to ICD, we should always consider the following tools that are available to treat VT in these patients. Even if an ICD is inserted, it obviously does not actually solve the VT problem, and if VT is left untreated, the patient is exposed to the potential of multiple ICD shocks.

ROLE OF ABLATION FOR VENTRICULAR ARRHYTHMIAS IN CARDIAC SARCOIDOSIS
Ablation is not a default fi rst line therapy in many or most patients with cardiac sarcoidosis. The overall strategy has to be tailored according to the clinical scenario and underlying mechanisms (see Table 8).
Most series of VT ablation published in the literature are small. Patients who underwent ablation presented with incessant VT or VT storm (10-30%), most frequently the arrhythmias were scar-related -and multiple morphologies were inducible. Other noted VT mechanisms were Purkinje fi ber-related, and included bundle branch reentry and microreentry, as well as non-reentrant Purkinje foci.
The initial scan demonstrates myocardial infl ammation at the base of LV, localized predominantly to the anterior, septal and inferior walls of the left ventricle. Complete suppression is achieved on the third scan and maintained on the 4 th scan. The post-scan immunosuppressive regimen, sequence in treatment series, and date are listed in association with each scan.
In a series of 42 patients 20 , we found that using FDG-PET scans to monitor myocardial infl ammation and to titrate immunosuppressive therapy may be benefi cial in patients with cardiac sarcoidosis (results presented in Table 7). Complete suppression of myocardial infl ammation was associated with low risk of adverse outcomes and death during follow up. Conversely, failure to obtain complete suppression of myocardial infl ammation was associated with increased risk of adverse outcomes and death. Surveillance FDG-PET scanning can identify patients who might have steroid-resistant form of disease and require additional or alternative agents. In conclusion, immunosuppression plays an important antiarrhythmic role in cardiac sarcoidosis when active myocardial infl ammation is present. Arrhythmia-free course seems to correlates with the ability to suppress infl ammation.
The value of FDG-PET-guided immunosuppression has to be balanced against its high cost and signifi cant radiation dose if repeated scans are obtained (the -Regular -Sustained -Monomorphic  Role of FDG-PET in identifying active infl ammation and guiding treatment tential benefi t for ICD. The ICD may be life-saving, but it does eliminate the arrhythmia, and these patients have a high rate of ICD shocks, so more needs to be done if arrhythmias are present. Antiarrhythmic drug therapy is important. Amiodarone is the mainstay of antiarrhythmic therapy, especially acutely, or in patients with decreased LVEF, and in most patients with active infl ammation while allowing time for immunosuppressive therapy to work. The patient should be monitored closely for liver toxicity and other side effects of amiodarone. This can be later switched to sotalol -particularly in younger patients and if the arrhythmias subside, in order to avoid the toxicity of long-term amiodarone. In all patients with cardiac sarcoidosis presenting with arrhythmia, it is important to assess for active myocardial infl ammation. If it is present, fi rst line of therapy should include immunosuppression and antiarrhythmic drug therapy. Ablation can be considered if necessary, in situation such as VT storm unresponsive to medical therapy or recurrent VF with an identifi able PVC trigger (see Table 10). If no active myocardial infl ammation is present, antiarrhythmic drug therapy and ablation are useful tools, in addition to careful programming of the ICD (maximizing anti-tachycardia pacing, increasing lower rate limits to preempt post-PVC pauses, etc.)

CONCLUSION
Cardiac sarcoidosis is an underdiagnosed disease with protean manifestations. The prevalence of cardiac sarcoidosis in Romania is not known, but a signifi cant number of patients might be currently under the care of cardiologists, cardiac electrophysiologists and other medical specialists. Identifying and treating these patients could impact signifi cantly their clinical course, morbidity and mortality. Once a diagnosis is made, cardiac sarcoidosis has a dynamic and unpredictable course. Diagnosis and treatment requires a multidisciplinary team of medical specialists in cardiology, cardiac electrophysiology, pulmonology, rheumatology and others. Arrhythmias in cardiac sarcoidosis have a variety of mechanisms and are associated with poor prognosis and signifi cant risk of death. Arrhyth-We investigated the role of VT ablation in a multicenter study from international Cardiac Sarcoidosis Consortium 22 . This is a multicenter study of 158 patients (age 52±11 years, 33% female) who underwent VT ablation. The median time from cardiac sarcoidosis diagnosis to VT was 827 days (interquartile range 210-1,676). There was a high rate of VT recurrence, especially in patients with FDG-PET evidence of active infl ammation. In this multicenter registry refl ecting real-world practice internationally, catheter ablation treatment of cardiac sarcoidosis -related VT remained challenging and rates of recurrence and repeat procedures were high. Survival curves are shown in Figure  15 and conclusions are summarized in Table 9.
To summarize, the management of ventricular arrhythmias in patients with cardiac sarcoidosis should follow a stepwise approach tailored to the individual patient.
First, one has to assess the risk of sudden death, even in asymptomatic patients, and consider the po- Table 9. Summary of results of VT ablation in cardiac sarcoidosis (CS). A report from the Cardiac Sarcoidosis Consortium 22  VT storm was eliminated in 82% of patients  ICD shocks were signifi cantly reduced  During median 2.5-year follow-up, 81 (51%) patients experienced VT recurrence, heart transplantation or death  LV dysfunction and infl ammation in pre-procedural 18F-FDG PET were signifi cantly associated with adverse prognosis  Appropriately timed ablation procedures remain important in the management of CS-related VT in conjunction with medical therapy. Confl ict of interest: none declared.