Study population

206 consecutive patients with SHD admitted to a first VT ablation procedure were included between June 2018 and February 2022 (59 ± 16 years; 81% male; LVEF 36 ± 14%) with the four most common NICM. Underlying NICMs were: DCM in 117 (57%), myocarditis in 53 (26%), sarcoidosis in 17 (8%) and ARVC in 19 (9%) (Fig. 1). The baseline characteristics were significantly different with higher age, males, cardiovascular risk factors and worse LVEF among DCM patients (all p < 0.05). Most patients had already an ICD at the time of ablation (79%) with significantly higher rates of secondary preventative indication among sarcoidosis and ARVC patients (Table 1).

Fig. 1

Distribution of underlying heart disease among NICM patients (n = 206). ARVC, arrhythmogenic right ventricular cardiomyopathy; DCM, dilative cardiomyopathy

Table 1 Baseline characteristicsCatheter ablation and acute procedural outcomes

The median time between first VT or electrical storm (ES) episode and the ablation procedure was 25 days. ES was present in 38% of all cases. For pre-procedural characterization of the ventricular substrate, planning access site (endocardial/epicardial, transseptal/retrograde) and guidance of intraprocedural substrate mapping cardiac CTs and LGE-MRIs were used whenever no contraindications were present, and patients were hemodynamic stable. In total, in 25% of all patients pre-procedural imaging was obtained with significantly lower rates among DCM patients (p < 0.001).

Epicardial mapping was performed in 53% and ablation in 44%. Overall, procedural characteristics were comparable, however, DCM and ARVC patients numerically revealed more VTs inducible with longer procedural, fluoroscopy and ablation times. Complete acute procedural success was achieved in 161 (78%) of all patients with lowest rates among DCM (74%) and highest rates among ARVC patients (90%) (p = n.s.). Consecutively, among patients with DCM discharge on antiarrhythmic drugs was highest (p = 0.005) (Table 2).

Table 2 Procedural data and intraprocedural successPredictors of acute ablation success

Different NICMs per se were not negatively associated with acute ablation success (OR 1.082, 95% CI 0.496–2.360; p = 0.844) in a multivariate regression. For each VT inducible during the procedure the probability for complete short-term success decreased by the factor 0.5 (OR 0.513, 95% CI 0.368–0.716; p = 0.001). Furthermore, the presence of ES (OR 0.165, 95% CI 0.057–0.479; p = 0.001) was an independent negative predictor of complete short-term success. Age, ablation time, epicardial ablation, CL of the clinical VT and LVEF were no independent predictors of successful ablation in our model (Fig. 2).

Fig. 2

Multivariate logistic regression for the predictors of complete short-term success after catheter ablation of VT after first procedure. CL, cycle length; LVEF, left ventricular ejection fraction; NICM, non-ischemic cardiomyopathy; VT CL, ventricular tachycardia cycle length in milliseconds

Complications

During index hospitalization, in total 33 complications occurred (16%) with no difference between groups except for high ablation-induced post-interventional AV conduction system related complications (p < 0.001). All patients already had documented AV-conduction delay/disturbances and were protected by dual or biventricular pacing devices. One pericardial tamponade occurred (with short reanimation and consecutive pericardiocentesis without any sequelae). Furthermore occurred five minor pericardial effusions without hemodynamic relevance, one pneumothorax due to difficult epicardial puncture, two vascular access–related complications (one arterial bleeding with concomitant oversewing of the access site and one aneurysma spurium with compression), three pneumonia (two ventilator-associated, one due to aspiration), one stroke, three accidental RV punctures during epicardial puncture (without pericardial effusion) and one iatrogenic aortic dissection. 4 patients suffered from hemodynamic instability with consecutive cardiogenic shocks during ablation. In total, four patients died during index hospitalization (two with DCM, two with myocarditis; 30-day in-hospital mortality 1.9%). All four patients died from protracted cardiogenic shock (Table 3).

Multiple ablations until VT freedom before hospital discharge

Early intrahospital recurrence of overall VTs was observed in 35 (17%) of all patients undergoing repeat VT ablation during index hospitalization in 23 patients with no difference between NICMs (p = 0.899). Median time from first to second procedure was 5.1 ± 3.4 days. Clinical VT recurred in eight patients (23% of all in-hospital recurrences; 4% of all patients; p = 0.320).

8/23 underwent an endo/epicardial combined, 15 an endocardial only approach for the redo procedure. Acute partial ablation success was achieved in 19/23 patients (83%), complete ablation success in 14/23 patients (61%). 4 patients (1 DCM, 1 sarcoidosis, 2 myocarditis) underwent a third procedure during index hospitalization (3 with endo/epicardial, 1 endocardial approach) with 2 partial and 2 complete ablation success. In total, 31% of all patients had antiarrhythmic drugs during follow-up with highest rates among DCM patients (41%; p = 0.005).

VT recurrence after index hospitalization and predictors of success

The recurrence rate for any ventricular arrhythmia after index hospitalization during follow-up were comparable among all groups, however numerically higher among DCM and myocarditis patients (61% DCM vs. 56% myocarditis vs. 35% sarcoidosis vs. 41% ARVC; log rank p = 0.148) (Figs. 3 and 5). Overall, 10 patients (7 DCM, 3 myocarditis; p = 0.540) had long-term recurrence of the clinical VT. Among them, all 3 patients with failure of ablation procedures and inducibility of clinical VT and undergoing a second ablation procedure for VT recurrence at index hospitalization. It should be emphasized that all groups show steep slopes at the beginning of follow-up representing the high recurrence rates early after VT ablation. After multivariable adjustment for relevant cofounders the two independent predictor of VT recurrences after index hospitalization were epicardial ablation (HR 1.528, 95% CI 1.025–2.280; p = 0.038) and chronic kidney disease (CKD) (HR 1.628, 95% CI 1.091–2.428; p = 0.017). Different NICMs were only univariable associated with increased VT recurrences (HR 0.792, 95% CI 0.640–0.980; p = 0.032) (supplemental Table 1 A) (Fig. 4).

Fig. 3

Kaplan–Meier curve for different NICM illustrating A overall VT recurrences after first procedure B after multiple procedures C cardiovascular mortality D cardiovascular rehospitalizations

Fig. 4

Kaplan–Meier curve for different NICM illustrating A clinical VT recurrences after first procedure B non-clinical VT recurrences after first procedure

Second ablation procedure

Of the 101 patients with VT recurrences, 52 (51%) underwent a second ablation procedure in our hospital after a mean of 10 ± 14 months (29 DCM, 16 myocarditis, 3 sarcoidosis, 4 ARVC). 6/52 underwent second VT ablation due to clinical VT recurrence (5 DCM, 1 myocarditis).

Mean procedural (168 ± 65 min; p = 0.279), fluoroscopy (16.6 ± 10.0 min; p = 0.084) and ablation times (31.1 ± 23.7 min; p = 0.064) were comparable among all groups. Epicardial access was performed in 64% (p = 0.399) and epicardial ablation in 58% (p = 0.359) of all patients without any differences among groups. Complete ablation success was achieved in all patients with sarcoidosis and ARVC as well as in 69% among myocarditis and in 55% in DCM patients (p = 0.163). Partial success was achieved again in all sarcoidosis and ARVC, in 88% of the myocarditis and in 86% of all DCM patients (p = 0.784).

During redos, in total, seven complications occurred (13%; p = 0.06). 2 patients had pericardial effusion without tamponade (1 DCM, 1 myocarditis). 1 patient with DCM had accidental RV puncture without any sequelae and 1 ablation-induced 3rd degree AV block, 1 with sarcoidosis a pneumothorax. 1 patient with DCM suffered from hemodynamic instability with consecutive cardiogenic shocks during ablation and 1 patient with myocarditis had prolonged congestive heart failure with consecutive intrahospital death.

32% of all patients were discharged on antiarrhythmic drugs (46% DCM and 20% myocarditis; p = 0.073).

VT recurrence after second procedure

Overall recurrence rates after second procedure were 50% with highest recurrence rates among DCM patients (62% DCM vs. 38% myocarditis vs. 33% sarcoidosis vs. 25% ARVC; p = 0.259).

Three and more ablations

20 patients underwent a third VT ablation (10 DCM, 7 myocarditis, 1 sarcoidosis and 2 ARVC) after a mean of 7 ± 8 months after previous procedure. Of them, 30% had combined epi/endocardial approach (p = 0.846) with epicardial ablation in 25% (p = 0.668). Partial ablation success was achieved in 85% (80% DCM, 100% myocarditis, 0% sarcoidosis, 100% ARVC; p = 0.366) and complete ablation success in 65% of all patients (60% DCM, 71% myocarditis, 0% sarcoidosis, 100% ARVC; p = 0.366).

15 patients had 4 th VT ablation (7 DCM, 7 myocarditis and 1 ARVC) after 7 ± 10 months with epicardial approach in 6 patients. Partial success was achieved in 81% (p = 0.664) and complete success in 50% (p = 0.565).

6 patients had 5 th procedure (2 DCM, 4 myocarditis) with epi/endocardial approach in two patients after 2 ± 1 months. Complete success was achieved in 67% (2 DCM, 2 myocarditis; p = 0.221), partial success in all patients. 1 patient with DCM underwent a 6 th procedure after 10 months using an epi/endocardial approach. Here, complete ablation success was achieved (Figs. 3 and 5).

Fig. 5

Kaplan–Meier curve for different NICM with LVEF < 35% illustrating A overall VT recurrences after multiple procedures B cardiovascular mortality and for patients with LVEF > 35% illustrating C overall VT recurrences after multiple procedures D cardiovascular mortality

Long-term success rate and cardiovascular mortality

The median follow-up time was 38 ± 22 months. During follow-up period cardiovascular death occurred in 38 patients (19%). Death rates were highest among patients with DCM compared to all other cardiomyopathies (30% vs. 6% myocarditis vs. 6% sarcoidosis vs. 0% ARVC; log-rank p = 0.001).

Most patients died due to cardiogenic shock (five patients) or with heart failure exacerbation (17 patients), two due to right heart failure and three due to electrical storm. 1 patient died with pericardial tamponade, 2 with progression of cancer and 1 patient with sepsis mediated due to driveline infection of the LVAD. In seven patients, no causative reason could be identified.

In a multivariate model of our cohort, the presence of DCM was a strong predictor of cardiovascular mortality (HR 3.316, 95% CI 0.970–11.335; p = 0.046). Furthermore, better LV function (HR 0.943, 95% CI 0.912–0.975; p = 0.001) was associated with better survival (Table 4 and supplemental Table 1B; Figs. 3 and 5).

Table 4 Primary and secondary endpoints

The overall rate of freedom from any ventricular arrhythmia after multiple procedures during follow-up differed among groups (43% DCM vs. 26% myocarditis vs. 29% sarcoidosis vs. 18% ARVC; log-rank p = 0.069). Multivariable predictor of VT freedom after multiple procedures was only higher LV function (HR 0.962, 95% CI 0.943–0.981; p = 0.001).