Preablation refers to the 30-day period before the ablation and postablation refers to the 30-day period after ablation. Each line represents a patient for whom ICD shock information was available.
In individuals with left ventricular ejection fraction (LVEF) <50%, the odds of 1- and 2-year survival free of VT recurrence, HT, or death were 48% and 36%, respectively. In those with active inflammation, the odds of 1- and 2-year survival free of VT recurrence, HT, or death were 52% and 43%, respectively. One patient had VT recurrence on the same day of the procedure (day 0); therefore, the sum of patients at time 0 in the at-risk tables is less than the total study population of 158 patients.
eTable 1. Procedural characteristics
eTable 2. Associations of clinical, imaging, and procedural characteristics with procedural success status
eTable 3. Procedural complications
eFigure. Kaplan-Meier curve of survival free of VT recurrence in the study population
Customize your JAMA Network experience by selecting one or more topics from the list below.
Siontis KC, Santangeli P, Muser D, et al. Outcomes Associated With Catheter Ablation of Ventricular Tachycardia in Patients With Cardiac Sarcoidosis. JAMA Cardiol. 2022;7(2):175–183. doi:10.1001/jamacardio.2021.4738
What outcomes are associated with ventricular tachycardia (VT) ablation in patients with cardiac sarcoidosis (CS)?
In a multicenter analysis of 158 patients with CS-associated VT undergoing catheter ablation, antiarrhythmic drug requirement and defibrillator shocks were significantly reduced, and VT storm was eliminated in 82% of patients with that presentation. LV dysfunction and myocardial inflammation were associated with adverse long-term prognosis postablation.
These observational data suggest that catheter ablation procedures can play an important role in the management of CS-associated VT in conjunction with medical therapy, including for patients with VT storm.
Ventricular tachycardia (VT) is associated with high mortality in patients with cardiac sarcoidosis (CS), and medical management of CS-associated VT is limited by high failure rates. The role of catheter ablation has been investigated in small, single-center studies.
To investigate outcomes associated with VT ablation in patients with CS.
Design, Setting, and Participants
This cohort study from the Cardiac Sarcoidosis Consortium registry (2003-2019) included 16 tertiary referral centers in the US, Europe, and Asia. A total of 158 consecutive patients with CS and VT were included (33% female; mean [SD] age, 52  years; 53% with ejection fraction [EF] <50%).
Catheter ablation of CS-associated VT and, as appropriate, medical treatment.
Main Outcomes and Measures
Immediate and short-term outcomes included procedural success, elimination of VT storm, and reduction in defibrillator shocks. The primary long-term outcome was the composite of VT recurrence, heart transplant (HT), or death.
Complete procedural success (no inducible VT postablation) was achieved in 85 patients (54%). Sixty-five patients (41%) had preablation VT storm that did not recur postablation in 53 (82%). Defibrillator shocks were significantly reduced from a median (IQR) of 2 (1-5) to 0 (0-0) in the 30 days before and after ablation (P < .001). During median (IQR) follow-up of 2.5 (1.1-4.9) years, 73 patients (46%) experienced VT recurrence and 81 (51%) experienced the composite primary outcome. One- and 2-year rates of survival free of VT recurrence, HT, or death were 60% and 52%, respectively. EF less than 50% and myocardial inflammation on preprocedural 18F-fluorodeoxyglucose positron emission tomography were significantly associated with adverse prognosis in multivariable analysis for the primary outcome (HR, 2.24; 95% CI, 1.37-3.64; P = .001 and HR, 2.93; 95% CI, 1.31-6.55; P = .009, respectively). History of hypertension was associated with a favorable long-term outcome (adjusted HR, 0.51; 95% CI, 0.28-0.92; P = .02).
Conclusions and Relevance
In this observational study of selected patients with CS and VT, catheter ablation was associated with reductions in defibrillator shocks and recurrent VT storm. Preablation LV dysfunction and myocardial inflammation were associated with adverse long-term prognosis. These data support the role of catheter ablation in conjunction with medical therapy in the management of CS-associated VT.
In patients with cardiac sarcoidosis (CS), ventricular tachycardia (VT) can result in sudden cardiac death and frequent implantable cardioverter-defibrillator (ICD) shocks. Medical therapy with immunosuppression and antiarrhythmic drugs often fail to suppress VT, and catheter ablation must be considered.1,2 Small, single-center studies have demonstrated potential benefits with catheter ablation for VT in patients with CS.2-7 Nevertheless, VT recurrence after ablation is common, which may be attributed in part to cycles of inflammation and scarring and to the complexity of the arrhythmogenic substrate that is often located in the midmyocardium or the epicardium.7 In this multicenter collaborative study, we assessed the short- and long-term outcomes associated with ablation and factors associated with these outcomes.
Patients included in this study were part of a collaborative prospective registry of CS, the Cardiac Sarcoidosis Consortium. In this analysis, we specifically included adult patients with CS per the 2014 Heart Rhythm Society criteria1 who underwent catheter ablation for a history of sustained monomorphic or polymorphic VT or ventricular ectopy-triggered ventricular fibrillation (VF) at 16 tertiary referral centers in the US, Europe, and Asia. The first ablation procedure in each patient at the participating institutions was considered the index procedure for this study. Enrollment periods varied among institutions, with the ablation procedures performed between 2003 and 2019. Additional details on ablation procedures, imaging, and data collection processes are described in the eMethods in the Supplement. Baseline data included echocardiography in all patients to assess left ventricular (LV) ejection fraction (EF), with selective use of cardiac magnetic resonance (CMR) and 18F-fluorodeoxyglucose–positron emission tomography (FDG-PET) imaging in subsets of patients to assess myocardial scarring and inflammation. We included PET results obtained within 6 months before the ablation and data from CMR obtained anytime preablation. Imaging techniques were according to standard institution-level practices. The study was approved by the Institutional Review Boards of each participating institution, and waiver of informed consent was granted owing to the retrospective nature of this study.
For each patient, the index date and start of follow-up was the date of the first VT ablation procedure. Immediate procedural outcome was defined as successful if no VT was inducible with programmed stimulation at the end of the procedure, partially successful if only nonclinical VTs were inducible, and failed if at least 1 clinical VT was inducible. VTs documented before the procedure were defined as clinical. Patients were monitored postablation until heart transplant (HT), death, or loss to follow-up. Long-term events included VT or VF recurrence (defined as sustained VT or VF documented by electrocardiography or stored ICD electrograms), HT, and death. In individuals who did not experience these outcomes, the last date the patient was known to be alive was documented. Follow-up information was obtained from clinic visits, emergency department visits, hospital admissions, and communication with referring physicians. For patients with ICDs, the number of ICD shocks in the 30 days before and after ablation was recorded.
The main outcome of interest was the composite of VT recurrence, HT, or death, whichever occurred first after the index ablation procedure. Kaplan-Meier survival analysis was used to assess event-free survival in the overall population and in subgroups defined by presence of LV systolic dysfunction and myocardial inflammation by PET. Differences in event-free survival in the subgroups were tested with log-rank testing. Cox regression analyses were used to assess association of preprocedural, procedural, and postprocedural characteristics with the composite outcome. Baseline clinical, imaging, and procedural variables (Table 1; eTable 1 in the Supplement) were considered for inclusion in the multivariable model and were selected using a backward stepwise selection approach. In addition, we included clinically important variables if they did not qualify by backward stepwise selection, such as age, sex, history of VT storm, and complete procedural success status. A Cox regression model was also used to assess whether VT recurrence after ablation was associated with subsequent death or HT. Because PET and CMR imaging were performed only in a subset of patients, the presence of preablation inflammation and LGE were considered separately in the variable selection process and fitted in separate multivariable models. In the subset of patients with ICDs and available detailed information on ICD shocks, we also compared the number of ICD shocks in the 30 days before and after ablation using the Wilcoxon signed rank test. Statistical analyses were conducted using Stata version 15 (StataCorp), and 2-tailed P values were considered significant at <.05.
This study included 158 patients (mean [SD] age, 52  years; 52 [33%] female; mean [range] patients per site, 10 [1-42]) undergoing index catheter ablation (Table 1). Thirty-six patients (23%) had received previous VT ablation attempts at other institutions. The diagnosis of CS was based on a positive endomyocardial biopsy result in 70 patients (44%), while the remaining patients had extracardiac sarcoidosis with clinical and/or imaging evidence (CMR and/or PET) of cardiac involvement. In 107 patients (68%), the diagnosis of CS preceded presentation with VT, with a median (IQR) time from CS diagnosis to VT of 827 (210-1676) days. Median (IQR) time from first presentation with VT to index catheter ablation was 329 (61-1178) days. In 144 patients (91%), the clinical arrhythmia was monomorphic VT, while 14 patients (9%) had a history of polymorphic VT or ventricular ectopy-induced VF. Sixty-five patients (41%) had a history of VT storm or incessant VT. At the time of ablation, 148 patients (94%) were taking antiarrhythmic medications, including amiodarone in 81 (51%), sotalol in 50 (32%), and mexiletine in 30 (19%), while 96 (61%) were taking corticosteroids (22 at prednisone equivalent dose >10 mg) and 51 (32%) were taking nonsteroidal immunosuppressants (26 methotrexate, 8 mycophenolate mofetil, 8 biologic agents, and 10 other) as monotherapy or in combinations.
Prior to ablation, 100 patients underwent CMR and 100 underwent PET. PET scans were performed a median (IQR) of 31 (6-154) days preablation with 46 (46%) of them within 30 days preablation. Eighty-eight patients had late gadolinium enhancement (LGE), and 65 had increased FDG uptake consistent with active inflammation. The distribution of LGE is shown in Table 1; 14 patients had right ventricular (RV) LGE. Imaging findings in the 7 patients with history of coronary artery disease (CAD) are described in the eResults in the Supplement. Table 2 indicates the characteristics of the patients with and without active inflammation on PET. Inflammation was associated with lower EF and a history of complete AV block and resuscitated cardiac arrest.
eTable 1 in the Supplement summarizes procedural details. A median (IQR) of 3 (1-4) VTs per patient were inducible with programmed stimulation at the beginning of the procedure with a median (IQR) VT cycle length of 330 (290-390) milliseconds. The procedure was successful in 85 patients (54%), partially successful in 41 patients (26%), and failed in 10 patients (6%). Postablation programmed stimulation was not performed in 22 patients (14%). Among 14 patients with polymorphic VT or VF, the procedure was successful in 5 (36%), partially successful in 3 (21%), failed in 1 (7%), and not tested in 5. eTable 2 in the Supplement reports associations of clinical, imaging, and procedural characteristics with the procedural outcome. The presence of inflammation on preablation PET scan and LGE on CMR were not associated with immediate procedural success. RV LGE was associated with noncomplete success (ie, patients in whom not all VTs were eliminated and patients in whom programmed stimulation was not performed at the conclusion of the procedure), while no other areas of LGE involvement demonstrated association with procedural success. Major procedural complications occurred in 11 patients (7%) (eTable 3 in the Supplement).
There were no differences in the number of inducible VTs, procedure duration, or RF ablation duration in patients with and without active inflammation prior to ablation. In patients with active inflammation, percutaneous epicardial ablation was more frequently performed compared with patients without inflammation (22 [34%] vs 5 [14%]; χ2 P = .04). Patients with active inflammation were more often treated with corticosteroids postablation (44 [68%] vs 10 [29%]; χ2 P < .001).
After ablation, 110 patients (70%) were treated with an antiarrhythmic drug compared with 148 (94%) preablation (χ2 P < .001), including amiodarone in 69 patients (44%). Postablation, 76 patients (48%) were treated with corticosteroids (newly initiated in 14 patients) and 44 (28%) were treated with nonsteroid immunosuppressants (newly initiated in 12 patients).
The median (IQR) follow-up period after ablation was 2.5 (1.1-4.9) years. During follow-up, 73 patients (46%) had documented VT recurrence at a median (IQR) of 12.4 (3.0-35.2) months (eFigure in the Supplement). The 1- and 2-year recurrent VT-free probabilities were 63% and 56%, respectively. Thirty-four patients (47%) with VT recurrence had at least 1 repeated ablation procedure at a median (IQR) of 8.7 (3.3-21.3) months after the index procedure. The immediate outcomes of the repeated procedures were complete success in 19 patients, partial success in 9, failure in 1, and not tested in 5. After initial and repeated procedures, 100 patients (63%) were free of recurrent VT from their last ablation to the end of follow-up. A total of 11 patients (7%) had HT, and 14 (9%) died.
Among 65 patients with a history of VT storm or incessant VT preablation, 37 patients (57%) had no recurrent VT postablation, while 16 (25%) had recurrence with only occasional VT and 12 (18%) had recurrence with VT storm or incessant VT. The median (IQR) number of ICD shocks was 2 (1-5; mean [SD], 5.1 [9.7]) in the 30 days preablation compared with 0 (0-0; mean [SD], 1 [3.7]) in the 30 days postablation (P < .001) (Figure 1).
Twenty-five patients (16%) had preablation inflammation and VT storm or incessant VT. VT recurred postablation in 12 patients (48%) at a median (IQR) of 95 (24-222) days. ICD shocks in the 30 days before and after ablation were reduced from a mean (SD) of 8.6 (7.9) to a mean (SD) of 0.14 (0.65). Six of the 25 patients died or underwent HT at a median (IQR) of 510 (156-626) days postablation (none within the first 3 months postablation).
The composite outcome of VT recurrence, HT, or death occurred in 81 patients (51%) and HT or death occurred in 24 (15%). The rates of 1- and 2-year survival free of VT recurrence, HT, or death were 60% and 52%, respectively.
Table 3 shows the unadjusted and adjusted HRs of the composite outcome of VT recurrence, HT, or death associated with clinical, imaging, and procedural variables. In multivariable analysis, history of hypertension was significantly associated with favorable long-term prognosis, whereas LVEF less than 50% and presence of inflammation were significantly associated with adverse long-term prognosis. Kaplan-Meier survival curves and log-rank P values are shown in Figure 2.
In a separate Cox model, VT recurrence as a time-varying covariate was not significantly associated with adverse long-term prognosis for HT or death. Among the 73 patients with VT recurrence, repeated ablation was not associated with a reduction in HT or death during follow-up.
In this international, multicenter registry, VT ablation was followed by significant reductions in antiarrhythmic drug requirement, ICD shocks, and recurrence of VT storm. LV systolic dysfunction and myocardial inflammation were associated with adverse long-term outcomes after ablation.
Mapping and ablation techniques for VT in structural heart disease are largely geared toward a fixed arrhythmogenic substrate with reentry as the mechanism of VT. In the presence of inflammation, other VT mechanisms may be operative8 and may require a different approach to mapping and ablation, targeting triggers of ventricular arrhythmias rather than a fixed substrate.8,9 Even in the absence of inflammation, the arrhythmogenic substrate in patients with CS is often complex, involving both ventricles in intramural and epicardial locations. Based on CMR LGE imaging, the only location associated with immediate procedural success was the presence of RV involvement where the success rate was lower than that in patients without RV LGE, consistent with prior reports.6 Regardless of LGE location, the long procedure times in this study indicate the complexity of VT ablation procedures in patients with CS.
Prior smaller studies7,10 have shown that the presence of periprocedural myocardial inflammation by PET scan was associated with worse long-term progndosis, including recurrent VT, death, or need for HT. The combination of active inflammation and VT represents a high-risk CS phenotype in which ablation procedures may only have limited or temporary effectiveness. Indeed, late VT recurrences were not infrequent in this series (median time to recurrence 12.4 months) which emphasizes the possibility of a dynamic substrate. Thus, the timing of VT ablation in patients with CS is important. In the absence of urgent or emergent indications for ablation, such as VT storm or ventricular ectopy–triggered VF refractory to medical management, data from this observational study suggest that intensification of immunosuppression, antiarrhythmic drug therapy, and surveillance with serial PET scanning may be preferable. These data are in line with an earlier report by Jefic et al,2 in which a stepwise approach was used. In more than 50% of patients, the arrhythmia was suppressed with immunosuppression as the first step. These findings were subsequently confirmed by Naruse et al,4 and this makes up the recommended approach in the recent expert consensus document on ablation of ventricular arrhythmias.11 However, if medical treatment fails, VT ablation can play an important role. In this study, ablation was effective in reducing ICD shocks and preventing recurrent VT, even in the presence of active inflammation. Furthermore, patients with preserved EF also had better long-term outcomes postablation, suggesting that VT ablations may be more beneficial if performed early in the disease process. The wide variation in the time from index VT presentation to ablation, even across highly skilled centers, may reflect uncertainty regarding the optimal timing of ablation in CS-associated VT, which should be determined on a case-by-case basis dependent on the patient and CS substrate.
In contradistinction to other forms of structural heart disease,12,13 noninducibility of VT after ablation was not associated with improved event-free survival in this study. A possible reason for this observation may be the changing arrhythmogenic milieu owing to the continued inflammation and scarring process. Thus, even if the procedure is immediately successful, some VT recurrences may occur owing to new VTs associated with disease progression.
VT recurrences occurred in almost half of the patients in this study, and almost half of these patients had repeated ablation. Similar figures have been reported in other forms of nonischemic cardiomyopathy13-15 and are indicative of the challenging arrhythmogenic substrate. However, in the even more challenging subset of this population, those with incessant VT or VT storm, most patients did not have VT recurrences. In patients with CS-associated VT with a potentially evolving substrate, repeated ablation procedures may be required for long-term VT control. More than 60% of patients in this series remained VT-free by the end of follow-up after their initial and repeated procedures.
VT is a powerful predictor of mortality in CS.16,17 Nevertheless, unlike other forms of structural heart disease,18,19 VT recurrence after ablation was not associated with subsequent HT or death, suggesting that the malignant disease course in patients with CS and ventricular arrhythmias may not only depend on the recurrence of ventricular arrhythmias. A recent population-based analysis reported that the incidence of heart failure in patients with sarcoidosis was more than 2 times higher than matched patients from the general population and was associated with high mortality.20
The ablation procedures included in this analysis were performed at tertiary centers with experience in CS management and VT ablation, which may limit the generalizability of our data. The mean follow-up time of 2.5 years is relatively short despite the long recruitment period, which reflects the referral nature of this cohort, including VT centers that may not longitudinally monitor patients referred for ablation. Furthermore, we did not have information about arrhythmia and ICD shock burden beyond 30 days preablation and postablation, limiting assessment of the long-term effect of ablation on arrhythmia burden. Procedures were performed over a span of many years with evolving mapping and ablation technologies. We included information from PET scans within 6 months preablation. However, fewer than half of the patients had scans less than 30 days prior to the ablation, which would not permit meaningful analyses. We did not have information regarding the evolution of inflammation after ablation. Effective immunosuppression with reduction in 18FDG uptake may be associated with a lower rate of adverse events during follow-up.21 Although LGE most often represents fibrosis, the presence of inflammation cannot be entirely excluded and would require combined PET/CMR imaging. In patients who had both modalities, we did not have information on their association and therefore could not rule out that LGE reflected inflammation rather than fibrosis. Few of the patients included in this study also had a history of CAD, and it is possible that CAD might have impacted LV function and abnormal imaging. However, available imaging data from CMR and PET in all but 1 patient support that abnormalities were because of CS and not CAD. The observational nature of this study and the absence of a control group do not permit definite conclusions regarding the comparative effectiveness of various management approaches.
In the series of selected patients with CS and VT presenting to specialized centers in this study, catheter ablation was completely or partially successful in 80% of patients and reduced antiarrhythmic drug requirement, ICD shocks, and recurrent VT storm. Patients with preserved LV function and those without active inflammation had superior long-term outcomes. While the observational nature of this series does not permit us to reach generalizable conclusions regarding the comparative effectiveness of different management approaches, in the setting of active inflammation and an evolving arrhythmogenic substrate, it may be prudent to prioritize immunosuppression and antiarrhythmic drug therapy over catheter ablation, which may be best reserved for the noninflammatory stage of the disease. However, ablation was still associated with favorable outcomes as a rescue approach in the setting of VT storm and active inflammation. These data support the role of ablation procedures in conjunction with medical therapy in the management of CS-associated VT.
Accepted for Publication: September 27, 2021.
Published Online: November 17, 2021. doi:10.1001/jamacardio.2021.4738
Corresponding Author: Frank M. Bogun, MD, Division of Cardiovascular Medicine, Cardiovascular Center, University of Michigan, 1500 E Medical Center Dr, SPC 5853, Ann Arbor, MI 48109-5853 (email@example.com).
Author Contributions: Drs Siontis and Bogun had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Siontis, Santangeli, Narasimhan, Zipse, Rosenthal, Chicos, Ellenbogen, Bogun, Crawford.
Acquisition, analysis, or interpretation of data: Siontis, Muser, Marchlinski, Zeppenfeld, Hoogendoorn, Narasimhan, Sauer, Zipse, Kapa, Vedantham, Rosenthal, Robinson, Patton, Murgatroyd, Chicos, Soejima, Roukoz, Sacher, Bhan, Appelbaum, Dickfeld, Mankad, Ellenbogen, Kron, Kim, Froehlich, Eagle, Bogun, Crawford.
Drafting of the manuscript: Siontis, Sauer, Bhan, Ellenbogen, Bogun, Crawford.
Critical revision of the manuscript for important intellectual content: Siontis, Santangeli, Muser, Marchlinski, Zeppenfeld, Hoogendoorn, Narasimhan, Zipse, Kapa, Vedantham, Rosenthal, Robinson, Patton, Murgatroyd, Chicos, Soejima, Roukoz, Sacher, Appelbaum, Dickfeld, Mankad, Ellenbogen, Kron, Kim, Froehlich, Eagle, Bogun, Crawford.
Statistical analysis: Siontis, Kim, Froehlich.
Obtained funding: Marchlinski.
Administrative, technical, or material support: Siontis, Narasimhan, Sauer, Soejima, Roukoz, Mankad, Ellenbogen, Kron, Froehlich, Bogun.
Supervision: Siontis, Muser, Marchlinski, Zipse, Chicos, Bhan, Mankad, Eagle, Bogun, Crawford.
Conflict of Interest Disclosures: Dr Narasimhan reports grant support from Medtronic outside the submitted work. Dr Kapa reports grants from Abbott, Boston Scientific, and Toray and personal fees from Affera, Biosig, Philips, and Pfizer outside the submitted work. Dr Vedantham reports personal fees from Merck and Roviant and grants from Amgen outside the submitted work. Dr Robinson reports personal fees from Abbott, Biosense Webster, Boston Scientific, and Medtronic outside the submitted work. Dr Roukoz reports grant support from Medtronic and consulting fees from Medtronic and Boston Scientific outside the submitted work. Dr Sacher reports personal fees from Abbott, Bayer, Biosense Webster, Boston Scientific, and Microport outside the submitted work. Dr Bogun reports grant support from the Frankel Cardiovascular Center at the University of Michigan. No other disclosures were reported.
Funding/Support: The Cardiac Sarcoidosis Consortium registry is supported by an internal grant from the Frankel Cardiovascular Center at the University of Michigan and a research grant from Biotronik.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.