Abstract
Background:
Ventricular tachycardia (VT) recurrence rates remain high following ablation among patients with non-ischemic cardiomyopathy (NICM).
Objectives:
To define the prevalence of lipomatous metaplasia (LM) in patients with NICM and VT, and its association with post-ablation VT recurrence.
Methods:
From patients who had ablation of left ventricular VT, we retrospectively identified 113 consecutive NICM patients with preprocedural contrast-enhanced cardiac computed tomography (CECT), from which LM was segmented. Nested within this cohort were 62 patients that prospectively underwent CECT and cardiac magnetic resonance from which myocardial border zone (BZ) and dense late gadolinium enhancement (LGE) were segmented. A control arm of 30 NICM patients without VT with CECT were identified.
Results:
LM was identified among 57% of control patients without VT, versus 83% of patients without VT recurrence and 100% of patients with VT recurrence following ablation. In multivariable analyses, LM extent was the only independent predictor of VT recurrence, with adjusted hazard ratio (HR) 1.1 per 1 g LM increase, p<0.001. Patients with LM extent ≥ 2.5 g had 4.9-fold higher hazard of VT recurrence than those with LM < 2.5 g (p < 0.001). In the nested cohort with 32 VT recurrences, LM extent was independently associated with VT recurrence after adjustment for BZ and LGE extent (HR 1.1 per 1 g increase, p =0.036).
Conclusion:
Myocardial LM is prevalent in patients with NICM of a variety of etiologies, and its extent is associated with post-ablation VT recurrence independent of the degree of fibrosis.
Keywords: non-ischemic cardiomyopathy, lipomatous metaplasia, ventricular tachycardia
Introduction
Catheter ablation of ventricular tachycardia (VT) is often necessary as an adjunct to, or in place of, anti-arrhythmic drug therapy (AAD) in the setting of non-ischemic cardiomyopathy (NICM)(1),(2). Improved understanding of perivalvular substrates and adoption of techniques for epicardial and intramural venous mapping and delivery of deep ablation lesions have improved outcomes. However, freedom from VT recurrence remains modest, ranging from 51% to 69% (3–5). Enhanced understanding of the substrate associated with ablation failure is essential, to optimize mapping and ablation strategies, and to identify patients that may require advanced therapies (4). Fibrosis is traditionally regarded as the anatomic substrate of reentrant VT, and late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR) has been used as a surrogate of fibrosis. It has been reported that the extent of fibrosis, roughly quantified as the number of AHA segments with LGE presence, predicts arrhythmia-related outcomes in NICM patients (6). However, it is important to note that LGE is not specific for fibrosis. LGE, rather, indicates increased interstitial expansion and may indicate not only fibrosis, but also myocardial lipomatous metaplasia (LM), as well and inflammation and other infiltrative processes (7). We have shown that LM identified on contrast enhanced cardiac computed tomography (CECT), among post-infarct patients, exhibits a stronger association with VT circuitry than fibrosis (8–10). Furthermore, LM is the anatomic substrate of VT in patients with arrhythmogenic right ventricular cardiomyopathy (11). Three decades ago, Baroldi et al. reported myocardial LM in 38% of patients with valvular heart disease and 24% of patients with dilated NICM (12). Prior data also indicates the presence of progressive LM in myocarditis (13,14), and sarcoidosis (15,16). However, the prevalence of LM in patients with NICM and VT and its association with post-ablation VT recurrence are unknown. In this study, we sought to 1) define the prevalence of LM, and 2) examine the association of LM with VT recurrence, in patients with NICM and VT.
METHODS
Briefly, from patients who had ablation of left ventricular (LV) VT, a total number of 113 NICM patients with pre-ablation cardiac CECT were retrospectively identified as the full cohort (see Figure 1). Within the full cohort, the nested cohort included NICM patients from the two-center prospective Intra-Myocardial Fat Deposition and Ventricular Tachycardia in Cardiomyopathy (INFINITY) Study between November 2018 and September 2022 at the Hospital of the University of Pennsylvania and the Johns Hopkins Hospital. The INFINITY study comprises two sub-cohorts of patients with NICM and ischemic cardiomyopathy (ICM). All study participants in the nested cohort underwent a pre-procedural CMR and CECT examinations. In summary, the full cohort included NICM patients with CECT prior to the index VT ablation and the nested cohort included NICM patients with CMR and CECT prior to the index VT ablation. Additionally, a control arm of 30 NICM patients without documented VT that underwent cardiac CECT for surveillance of coronary disease, were retrospectively identified from 2020 to 2022. Investigators analyzing the imaging data were blinded to clinical data, including detailed disease history, current medications, physical examination, implantable cardioverter-defibrillator (ICD) logs, and routine clinical blood work. Ethical approvals were obtained from our respective institutional review boards for both the retrospective database and for the prospective INFINITY study. Informed written consent was obtained from each participant enrolled in the prospective study.
Figure 1 –
Patient identification flowchart. Abbreviations: NICM = non-ischemic cardiomyopathy; RV = right ventricle; LV = left ventricle; PVC = premature ventricular contraction; VT = ventricular tachycardia; NSVT = non-sustained ventricular tachycardia; VF = ventricular fibrillation; CECT = contrast-enhanced CT; CMR = cardiac magnetic resonance imaging.
Dense late gadolinium enhancement (LGE) and border zone area were segmented from the LGE-CMR image, and LM was segmented from the cardiac CECT images. Both imaging analyses were performed using ADAS 3D (ADAS 3D Medical SL, Barcelona, Spain) as previously described (8,9). On LGE images, tissue characterization of healthy myocardium (<40% maximal signal intensity), border zone tissue (BZ, 40%-60% maximal signal intensity), and dense LGE region (>60% maximal signal intensity, including scar and LM with <0 Hounsfield (HU) on corresponding CT coordinates) were performed by thresholds, which are set by default to 40 and 60% (± 5%) with respect to the maximum intensity voxel in the myocardium (a modification of the fullwidth half-maximum) as described Andreu et al (17).
On CT images, careful delineation of the subepicardial contour from epicardial adipose tissue was performed to avoid inclusion of epicardial adipose tissue as LM. Also, we referenced the continuity of the adjacent myocardium in the long axis, which can enhance myocardial border identification. Notably, epicardial adipose tissue tends to be homogeneous, however, LM is heterogeneously mixed with myocardium and fibrosis, with resultant intermediate attenuation. If the inferior wall was thin and hypo- or akinetic, the sub-epicardium was carefully delineated to avoid the diaphragm via tracing the continuity of the diaphragm in a three-dimensional manner. The interventricular septum was carefully segmented in the absence of sufficient RV contrast. In the nested cohort, we referred to the myocardial mobility and ventricular structure of steady-state free precession imaging sequence or ventricular structure of T2 imaging which can assist in identifying the local myocardial thickness and morphology. We applied an imaging noise reduction filter using ADAS 3D (ADAS 3D Medical SL, Barcelona, Spain) to avoid inclusion of noise, which exhibits negative attenuation, as LM. Usually, imaging noise appears as an isolated pixel in only one slice at an exact location (Supplemental Figure 1A). LM was identified from the cardiac CECT image as the area located between the subendocardial and subepicardial contour with attenuation value ranging from −180 to 0 HU and a continuous volume > 1 mm3 (Supplemental Figure 1B & C & D) (8,9,18).
The details of patient recruitment, acquisition protocols and imaging analysis of CMR and cardiac CT, strategy of electrophysiological study and ablation, and statistical analysis was included in the supplemental material.
RESULTS
Baseline characteristics
The patient cohort inclusion scheme is summarized in Figure 1. Of 3,052 patients retrospectively identified to have undergone VT ablation at the Hospital of the University of Pennsylvania from 2015 to 2022, 113 patients with NICM and VT had preprocedural CECT and comprised the full cohort. Among these patients, the nested cohort included 62 patients that were prospectively recruited in the NICM sub-cohort of the INFINITY study and underwent preprocedural cardiac CECT and CMR. A control arm included 30 NICM patients who never experienced VT and underwent cardiac CECT with analyzable images at the Hospital of the University of Pennsylvania from 2020 to 2022.
In the full cohort, mean age was 60 years and 75% were male. Of all patients, 58% had dilated cardiomyopathy, 12% prior myocarditis, 9% valvular heart disease, 11% hypertrophic cardiomyopathy, 9% cardiac sarcoidosis, and 3% others (2 patients with cardiac amyloidosis, and 1 patient with peripartum cardiomyopathy). There was no statistically significant difference in the incidence of documented heart failure, obstructive sleep apnea, amiodarone use, or mean LV wall thickness (median 5.6 versus 6.5 mm, p = 0.069) among patients with versus without VT recurrence. The nested cohort had similar characteristics to the full cohort. Importantly, the nested cohort exhibited similar fibrosis extent (median 61.8% versus 57.3%, p = 0.753), as well as similar proportions of border zone, and dense LGE on CMR among those with and without VT recurrence. The baseline demographics and image characteristics have been detailed in Tables 1 and 2, respectively.
Table 1.
Baseline characteristics at the time of CT imaging
| Control arm | After propensity score matching with control arm | Before propensity score matching with control arm | Before propensity score matching with control arm | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| NICM patients without VT ( N = 30) | NICM patients with VT( N = 30) | P-value | Nested cohort NICM patients with preprocedural Cardiac MRI & CECT ( N = 62) | Full cohort NICM patients with preprocedural cardiac CECT ( N = 113) | |||||||
| Total, n = 62 | VT recurrence (−), n=30 | VT recurrence (+), n=32 | P-value | Total, n = 113 | VT recurrence (−), n=63 | VT recurrence (+), n=50 | P-value | ||||
| Vital statistics | |||||||||||
| Age, years | 63±13 | 61 ±11 | 0.216 | 61 ±11 | 61 ±11 | 60 ±11 | 0.695 | 60 ±11 | 59 ±12 | 62 ±11 | 0.333 |
| Male, % | 18(60) | 23(77) | 0.267 | 50(81) | 27(84) | 23(77) | 0.328 | 85(75) | 46(73) | 39(78) | 0.542 |
| BMI, kg/m2 | 30 ±8 | 30 ±7 | 0.935 | 31 ±7 | 31 ±8 | 31 ±7 | 0.861 | 30 ±7 | 30 ±7 | 31 ±7 | 0.605 |
| SBP, mmHg | 135 ±21† | 130±17 | 0.316 | 117 ±18 | 114 ±17 | 120 ±18 | 0.201 | 119 ±18 | 119±18 | 118 ±19 | 0.732 |
| DBP, mmHg | 79 ±9† | 78 ±9 | 0.676 | 73 ±10 | 73 ±11 | 73 ±10 | 0.948 | 72 ±11 | 71 ±11 | 72 ±11 | 0.467 |
| Heart rate, beats/min | 74 ±9 | 75 ±11 | 0.745 | 73 ±12 | 74 ±10 | 73 ±14 | 0.658 | 74 ±13 | 76 ±14 | 72 ±11 | 0.165 |
| Disease history | |||||||||||
| Heart failure, % | 17(57) | 17(57) | 1.00 | 39(63) | 16(53) | 23(72) | 0.131 | 62(55) | 30(48) | 32(64) | 0.082 |
| Hypertension, % | 22(83) | 22(73) | 0.347 | 35(57) | 15(50) | 20(63) | 0.321 | 65(58) | 35(56) | 30(60) | 0.635 |
| DM, % | 7(23) | 5(17) | 0.748 | 13(21) | 5(17) | 8(25) | 0.421 | 23(20) | 12(19) | 11(22) | 0.699 |
| Afib/AFL | 6(20) | 11(37) | 0.152 | 34(55) | 17(57) | 17(53) | 0.779 | 58(51) | 34(54) | 24(48) | 0.528 |
| Pulmonary disease, % | 6(20) | 4(13) | 0.465 | 10(16) | 7(23) | 3(9) | 0.176 | 18(16) | 13(21) | 5(10) | 0.195 |
| CKD, % | 10(33) | 6(20) | 0.254 | 8(13) | 4(13) | 4(13) | 1.000 | 18(16) | 11(18) | 7(14) | 0.618 |
| Hyperlipidemia, % | 6(20)‡ | 6(20) | 1.000 | 17(27) | 11(37) | 6(19) | 0.114 | 39(35) | 27(43) | 12(24) | 0.036 |
| Hypothyroidism, % | 5(17) | 2(7) | 0.424 | 10(16) | 5(16) | 5(17) | 1.000 | 22(20) | 12(19) | 10(20) | 0.899 |
| OSA, % | 2(7)† | 2(7) | 1.000 | 12(19) | 3(9) | 9(30) | 0.055 | 24(21) | 7(14) | 17(27) | 0.094 |
| Presence of ICD prior to imaging and VT ablation | 5(17)† | 29(97) | <0.001 | 60(97) | 31(97) | 29(97) | 1.000 | 110(97) | 61(97) | 49(98) | 1.000 |
| Laboratory test | |||||||||||
| Serum creatinine, mg/dl | 1.1 ±0.4 | 1.1 ±0.3 | 0.760 | 1.1 ±0.3 | 1.1 ±0.3 | 1.1 ±0.3 | 0.915 | 1.1 ±0.3 | 1.1 ±0.4 | 1.1 ±0.3 | 0.683 |
| Anti-arrhythmic medications | |||||||||||
| Amiodarone, % | 3(10)† | 12(40) | 0.015 | 24(39) | 7(23) | 17(53) | 0.016 | 43(38) | 19(30) | 24(48) | 0.052 |
| Sotalol, % | 0(0)† | 6(20) | 0.009 | 10(16) | 6(20) | 4(13) | 0.502 | 17(15) | 9(14) | 8(16) | 0.798 |
| Mexiletine, % | 0(0)† | 3(10) | 0.076 | 3(5) | 1(3) | 2(6) | 1.000 | 7(6) | 4(6) | 3(6) | 1.000 |
| Other cardiovascular medications | |||||||||||
| Antiplatelet, % | 13(45) | 11(37) | 0.529 | 20(32) | 10(33) | 10(31) | 0.861 | 50(44) | 31(49) | 19(38) | 0.234 |
| Anticoagulation, % | 9(30) | 13(43) | 0.296 | 33(53) | 16(53) | 17(53) | 0.987 | 54(48) | 30(48) | 24(48) | 0.968 |
| Statin, % | 20(67) | 20(67) | 1.000 | 36(58) | 20(67) | 16(50) | 0.184 | 73(65) | 43(68) | 30(60) | 0.362 |
| ACEI/ARB use, % | 19(63) | 21(70) | 0.566 | 41(66) | 21(70) | 20(63) | 0.533 | 67(59) | 36(57) | 31(62) | 0.602 |
| Beta blocker, % | 21(70) | 20(67) | 0.632 | 53(86) | 26(87) | 27(84) | 1.00 | 97(86) | 52(83) | 45(90) | 0.291 |
| MRA, % | 6(20) | 8(27) | 0.542 | 19(31) | 9(30) | 10(31) | 0.915 | 31(27) | 17(27) | 14(28) | 0.904 |
| Diuretics, % | 15(50) | 15(50) | 1.000 | 30(48) | 16(53) | 14(44) | 0.450 | 53(47) | 29(46) | 24(48) | 0.835 |
| Etiologies & classification of NICM | |||||||||||
| DCM, % | 12(40) | 16(53) | 0.350 | 28(44) | 14(47) | 14(44) | 0.111 | 65(58) | 37(59) | 28(56) | 0.07 |
| Prior myocarditis, % | 1(3) | 2(7) | 11(18) | 2(7) | 9(28) | 13(12) | 3(5) | 10(20) | |||
| HCM, % | 6(20) | 5(17) | 9(15) | 6(20) | 3(9) | 12(11) | 8(13) | 4(8) | |||
| Cardiac sarcoidosis, % | 3(10) | 4(13) | 8(13) | 3(10) | 5(16) | 10(9) | 4(6) | 6(12) | |||
| VHD, % | 8(27) | 2(7) | 5(8) | 4(13) | 1(3) | 10(9) | 8(13) | 2(4) | |||
| Others, % | 0(0) | 1(3) | 2(3) | 1(3) | 0(0) | 3(4) | 2(4) | 1(2) | |||
means significantly different from the full cohort (n=113).
Continuous variables expressed as mean ± SD or median (interquartile range), as appropriate and categorical variables are expressed as number (percentage).
Abbreviation: CECT = contrast-enhanced computed tomography; VT = ventricular tachycardia; BMI = Body mass index; SBP = systolic blood pressure; DBP = diastolic blood pressure; DM = Diabetes mellitus; CKD = chronic kidney disease; OSA= Obstructive sleep apnea; Afib/AFL = atrial fibrillation/flutter; PVC = premature ventricular contraction; ICD = Implantable cardioverter-defibrillators; ACEI = angiotensin converting enzyme inhibitor; ARB = angiotensin II receptor blocker; MRA = Mineralocorticoid Receptor antagonist; DCM = Dilated cardiomyopathy ; HCM = Hypertrophic cardiomyopathy; VHD = Valvular Heart disease; NICM = non-ischemic cardiomyopathy.
Table 2.
Imaging characteristics
| Control arm | After propensity score matching with control arm | Before propensity score matching with control arm | Before propensity score matching with control arm | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| NICM without VT, n=30 | NICM with VT, n=30 | P-value | Nested cohort NICM patients with VT undergoing preprocedural Cardiac MRI & CECT (N = 62) | Full cohort NICM patients with VT undergoing preprocedural cardiac CECT ( N = 113) | |||||||
| Total, n = 62 | VT recurrence (−), n=30 | VT recurrence (+), n=32 | P-value | Total, n = 113 | VT recurrence (−), n=63 | VT recurrence (+), n=50 | P-value | ||||
| Basic cardiac MRI measurements | |||||||||||
| LVEF, % | 36±13 | 36±15 | 35±12 | 0.763 | |||||||
| *LVEDVi, ml/m2 | 98±38 | 100±50 | 0.846 | 119±36 | 118±35 | 119±37 | 0.941 | 114±42 | 110±42 | 119±42 | 0.250 |
| LVESVi, ml/m2 | 79±32 | 80±32 | 78±33 | 0.872 | |||||||
| *LV Massi, g/nf | 68±26 | 69±23 | 0.861 | 67±20 | 70±22 | 64±17 | 0.246 | 72±24 | 73±25 | 70±22 | 0.500 |
| Cardiac LGE measurements | |||||||||||
| Border zone, % | 44 (23, 58) | 43 (22, 55) | 48 (26, 59) | 0.515 | |||||||
| Dense LGE, % | 14 (7, 21) | 13 (6, 21) | 15 (9, 22) | 0.246 | |||||||
| #Presence of LGE | 62 (100) | 32 (100) | 30 (100) | 1.00 | |||||||
| Border zone & Dense LGE, % | 61 (36, 77) | 58.8 (27, 75) | 68 (41, 78) | 0.359 | |||||||
| **Fibrosis, % | 59 (33, 72) | 57 (27, 75) | 62 (35, 71) | 0.753 | |||||||
| Cardiac CECT measurements | |||||||||||
| Mean LV wall thickness, mm | 5.8 (5, 6.7) | 5.7 (5.3,6.1) | 0.938 | 5.9 (5.3, 6.8) | 6.5 (5.6,7.1) | 5.6 (4.9,6.6) | 0.028 | 5.9 (5.3,6.7) | 6.1 (5.4,6.8) | 5.6 (5.0,6.6) | 0.069 |
| Presence of LM, % | 17(57)† | 27(90) | 0.004 | 56(90) | 24(80) | 32(100) | 1.00 | 102(90) | 52(83) | 50(100) | 1.00 |
| LM, g | 0.2 (0, 0.5)† | 2.0 (0.4,7.3) | <0.001 | 1.7 (0.8, 5.3) | 0.8 (0.1,1.8) | 5.2 (1.6, 10.0) | <0.001 | 1.5 (0.5,5.0) | 0.8 (0.2,1.8) | 4.8 (1.6,8.8) | <0.001 |
means significantly different from the full cohort (n=113).
indexed LVEDV and LV mass of the full cohorts was derived from the end-diastolic phase of cardiac CECT image.
LGE is the combination of border zone and dense LGE.
Fibrosis of each patient was derived via total border zone & dense LGE % minus LM%. Continuous variables expressed as mean ± SD or median (interquartile range), as appropriate and categorical variables are expressed as number (percentage). Abbreviation: CECT = contrast-enhanced computed tomography; VT = ventricular tachycardia; LVEF = left ventricular ejection fraction; LVEDVi = left ventricular end-diastolic volume index; LVESVi = left ventricular end-systolic volume index; LV Massi = left ventricular mass index; LGE = late gadolinium enhancement; LM = lipomatous metaplasia.
Procedure characteristics
Among 113 patients, epicardial access in 40 patients (36%) was attempted and successfully achieved in 35 (31%). VT was not inducible at baseline in 16 (14%) patients. Of all patients, 78 (69.1%) required endocardial ablation, 30 (26.5%) required endocardial and epicardial ablation, and 5 (4.4%) had epicardial ablation. A total of 226 distinct VT morphologies and/or cycle lengths were induced, with a median of 2 (IQR 1-3) VTs per patient, including 76 (33.3%) with exits in the basal septum, 45 (19.9%) in the basal lateral wall, 18 (8.0%) in the mid-ventricular septum, 17 (7.5%) in the basal inferior wall, 16 (7.1%) in the mid-ventricular lateral wall, and 12 (5.3%) at the apex (only locations with >5% were noted here). To identify ablation targets, comprehensive substrate-based mapping was performed in 96% of patients and in a limited fashion in the remaining. Additional activation and entrainment mapping was performed in 57% of patients. Among 91 (81%) of patients, PES was performed at the end of the procedure. VT was no longer inducible in 58% of patients, whereas 17% of patients had inducible non-clinical VT, and 6% had inducible clinical VT. Post-procedural non-invasive programmed stimulation (NIPS) was performed in 43% of patients, including 19% with negative NIPS, 14% with inducible non-clinical VT, and 10% with inducible clinical VT.
Prevalence of LM within LGE in NICM patients with documented VT
Among all patients with NICM and VT referred for ablation, 102 (90%) had prevalent LM on pre-procedural CECT. LM was noted in every NICM patient with VT recurrence with a median LM mass 4.8 (IQR1.6, 8.8) g, and 52 out of 63 (83%) patients without VT recurrence with median LM mass 0.8 (IQR 0.2, 1.8) g, with p < 0.001. LM distribution within the myocardium overlapped with LGE distribution (Supplemental Figure 3), mixed with fibrosis and viable myocardium (Figure 2). In the nested cohort, a significantly larger mass of LM at baseline CECT was observed in patients with VT recurrence compared to patients without VT recurrence [5.2 (1.6, 10.0) versus 0.8 (0.1, 1.8) g, p < 0.001], with a similar proportion of baseline LM relative to LGE, with 5.7% (2.3%, 12.0%) versus 1.0% (0.2%, 2.5%), p < 0.001.
Figure 2 -.
Examples of myocardial lipomatous metaplasia in patients with NICM and documented VT---Example #1-#3 from the nested cohorts: Left Panel A1-C1 of LGE-CMR images highlighting both myocardial lipomatous metaplasia (LM) and fibrosis; Middle Panel A2-C2 of cardiac CECT images exhibit hypoenhanced area within the myocardium as myocardial LM; Right Panel A3-C3 of cardiac CECT images exhibit LM highlighted as the green area which corresponds to the hypoenhanced area in A2-C2 and aligns along the distribution of LGE and scatters within the area of LGE in A1-C1. Example #4-#7 from the full cohort but not nested cohorts: Panel D1-G1 of cardiac CECT images exhibit hypoenhanced area within the myocardium as myocardial LM; Right Panel D2-G2 of cardiac CECT images exhibit LM highlighted as the green area.
Abbreviations: NICM = non-ischemic cardiomyopathy; VT = ventricular tachycardia; CECT = contrast-enhanced CT; CMR = cardiac magnetic resonance imaging; LGE = late gadolinium enhancement; LM = lipomatous metaplasia.
After propensity score matching between the control arm and NICM patients with VT, the baseline clinical characteristics (Table 1), including vital statistics, disease history, serum tests, non-antiarrhythmic cardiovascular medications, and conventional LV structural measurements (Table 2) were statistically similar. However, LM prevalence [90% versus 57%, p = 0.004] and LM mass [median 2.0 (0.4, 7.3) versus 0.2 (0, 0.5) g, p<0.001] remained higher in NICM patients with VT, compared to the control arm.
LM Association with post-ablation VT recurrence
During a median follow-up of 616 (260, 1454) days, 50 post-ablation VT recurrences occurred in the full cohort. Adjusted with the competing risk of death (n = 18) in univariable analyses, history of heart failure [hazard ratio (HR) 2.1, p = 0.009], obstructive sleep apnea (HR 1.5, p = 0.068), and the use of AAD (HR 1.9, p = 0.019) were associated with VT recurrence. LM mass was significantly associated with VT recurrence in the univariable analysis, with HR 1.1 per 1g increase, p < 0.001, and remained significant in the multivariable model adjusted for heart failure, sleep apnea, and AAD use, with adjusted HR 1.1 per 1 g increase, p < 0.001 (Table 3 & Table 4). In the restricted cubic spline plot, LM mass was consistently and positively associated with VT recurrence with HR and 95% CI higher than 1 in the majority of LM mass range (Central Illustration B). LM mass also exhibited excellent discriminative performance for VT recurrence, significantly higher than the LV volume, with the area under the curve 0.830 versus 0.581, P<0.001.
Table 3.
Cox regression analysis with Competing risk method of predicting VT reoccurrence
| Nested cohort NICM patients with preprocedural Cardiac MR & CECT, N = 62 VT recurrence, n = 32 | Full cohort NICM patients with preprocedural Cardiac CECT, N = 113 VT reurrence, n = 50 | |||
|---|---|---|---|---|
| Univariable analysis HR (95% CI) | P-value | Univariable analysis HR (95% CI) | P-value | |
| Vital statistics | ||||
| Age, per 10 year increase | 1.1 (0.8, 1.6) | 0.527 | 1.2 (1.0, 1.5) | 0.106 |
| Male | 0.8 (0.3, 1.8) | 0.518 | 0.8 (0.4, 1.5) | 0.459 |
| BMI, per 10 kg/m2 increase | 1.0 (0.6, 1.5) | 0.946 | 1.1 (0.8, 1.5) | 0.677 |
| SBP, per 10 mmHg increase | 1.0 (0.8, 1.2) | 0.688 | 1.0 (0.8, 1.2) | 0.932 |
| DBP, per 10 mmHg increase | 1.2 (0.8, 1.7) | 0.360 | 1.1 (0.9, 1.5) | 0.282 |
| Heart rate, per 10 beat/min increase | 1.0 (0.7, 1.3) | 0.783 | 0.8 (0.7, 1.0) | 0.075 |
| Disease history | ||||
| Heart failure | 2.1 (1.0, 4.3) | 0.058 | 2.1 (1.2, 3.8) | 0.009 |
| Hypertension | 1.7 (0.8, 3.3) | 0.148 | 1.3 (0.8, 2.3) | 0.303 |
| Diabetes mellitus | 2.8 (1.2, 6.2) | 0.015 | 1.5 (0.8, 3.0) | 0.241 |
| Other arrhythmia (Afib/AFL or PVC) | 1.2 (0.6, 2.4) | 0.637 | 0.8 (0.5, 1.4) | 0.422 |
| Pulmonary disease | 0.5 (0.2, 1.5) | 0.223 | 0.5 (0.2, 1.2) | 0.121 |
| CKD | 1.1 (0.3, 3.2) | 0.971 | 0.9 (0.4, 2.0) | 0.788 |
| Hyperlipidemia | 0.5 (0.2, 1.3) | 0.181 | 0.5 (0.3, 1.0) | 0.036 |
| Hypothyroidism | 0.8 (0.3, 1.7) | 0.536 | 0.9 (0.5, 1.8) | 0.830 |
| OSA | 1.3 (1.1, 1.9) | 0.033 | 1.5 (0.9, 2.1) | 0.068 |
| Presence of ICD | 1.4 (0.1, 21.2) | 0.809 | 1.3 (0.1, 13.5) | 0.811 |
| Use of medications | ||||
| Amiodarone | 2.8 (1.4, 5.6) | 0.030 | 2.1 (1.2, 3.6) | 0.008 |
| Anti-platelet | 0.8 (0.4, 1.5) | 0.474 | 0.7 (0.4, 1.2) | 0.155 |
| Anti-coagulation | 1.3 (0.7, 2.6) | 0.419 | 1.4 (0.8, 2.4) | 0.247 |
| Statin | 0.9 (0.5, 1.8) | 0.783 | 1.0 (0.6, 1.6) | 0.856 |
| ACEI or ARB use | 1.1 (0.5, 2.2) | 0.833 | 1.3 (0.8, 2.4) | 0.325 |
| Beta blocker use | 0.9 (0.4, 2.6) | 0.916 | 2.0 (0.8, 5.1) | 0.153 |
| MRA | 0.7 (0.4, 1.5) | 0.399 | 0.8 (0.4, 1.3) | 0.334 |
| Diuretics | 0.9 (0.5, 1.8) | 0.760 | 1.1 (0.7, 1.9) | 0.682 |
| Basic cardiac MRI measurements | ||||
| LVEF, per 10% increase | 1.1 (0.8, 1.4) | 0.528 | ||
| LVEDVi, per 10 ml/m2 increase | 1.0 (0.9, 1.1) | 0.796 | 1.1 (1.0, 1.1) | 0.079 |
| LVESVi, per 10 ml/m2 increase | 1.0 (0.9, 1.1) | 0.702 | ||
| LV Massi, per 10 g/m2 increase | 0.9 (0.8, 1.1) | 0.235 | 1.0 (0.9, 1.2) | 0.514 |
| Cardiac LGE measurements | ||||
| Border zone, per 10% increase | 1.0 (0.9, 1.3) | 0.734 | ||
| Dense LGE, per 10% increase | 1.1 (0.8, 1.5) | 0.704 | ||
| Border zone & Dense LGE, per 10% increase | 1.0 (0.9, 1.2) | 0.682 | ||
| **Fibrosis, per 10% increase | 1.0 (0.9, 1.2) | 0.971 | ||
| Cardiac CECT measurements | ||||
| Mean LV wall thickness, per 1 mm decrease | 1.5 (1.0, 2.1) | 0.033 | 1.1 (0.9, 1.5) | 0.320 |
| LM, per 1 g increase | 1.1 (1.0, 1.1) | 0.004 | 1.1 (1.1, 1.1) | <0.001 |
Fibrosis of each patient was derived via total border zone & dense LGE % minus LM%
Abbreviation: CECT = contrast-enhanced computed tomography; VT = ventricular tachycardia; BMI = Body mass index; SBP = systolic blood pressure; DBP = diastolic blood pressure; DM = Diabetes mellitus; CKD = chronic kidney disease; OSA = obstructive sleep apnea; Afib/AFL = atrial fibrillation/flutter; PVC = premature ventricular contraction; ICD = Implantable cardioverter-defibrillators; ACEI = angiotensin converting enzyme inhibitor; ARB = angiotensin II receptor blocker; NICM = non-ischemic cardiomyopathy; LVEF = left ventricular ejection fraction; LVEDVi = left ventricular end-diastolic volume index; LVESVi = left ventricular end-systolic volume index; LV Massi = left ventricular mass index; LM = lipomatous metaplasia.
Table 4.
Multivariable model
| Nested cohort, NICM patients with preprocedural cardiac MRI & CECT, number of subjects: N= 62; VT reoccurrence, n = 32 | |||||
|---|---|---|---|---|---|
| Model 1 | Model 2 | ||||
| HR (95% CI) | P-value | HR (95% CI) | P-value | ||
| Heart failure | 2.1 (0.9, 4.7) | 0.072 | 1.6 (0.8, 3.4) | 0.18 | |
| Use of Amiodarone | 3.0 (1.5, 6.1) | 0.002 | 2.2 (1.1, 4.4) | 0.03 | |
| Model 1: LV wall thickness, per 1 mm increase | 0.6 (0.4, 0.9) | 0.011 | |||
| Model 2: LM, per 1 g increase | 1.1 (1.0, 1.1) | 0.036 | |||
| Full cohort, NICM patients with preprocedural CECT, number of subjects: N = 113; VT reoccurrence, n = 50 | |||||
| Model 1 | Model 2 | ||||
| HR (95% CI) | P-value | HR (95% CI) | P-value | ||
| Base model | Heart failure | 1.7 (0.9, 3.2) | 0.114 | 1.9 (1.0, 3.6) | 0.056 |
| Use of Amiodarone | 2.0 (1.1, 3.6) | 0.031 | 1.6 (0.8, 2.9) | 0.171 | |
| Obstructive sleep apnea | 1.4 (1.2, 1.9) | 0.019 | 1.4 (1.2, 1.8) | 0.014 | |
| Model 1: LVEDVi, per 10 ml/m2 increase | 1.0 (1.0, 1.1) | 0.353 | |||
| Model 2: LM, per 1g increase | 1.1 (1.0, 1.1) | <0.001 | |||
The imaging variable with univariable P value <0.2 was added separately to the base model, and 4 models were built.
Abbreviation: VT = ventricular tachycardia; CECT = contrast-enhanced computed tomography; LM = lipomatous metaplasia.
Central illustration -.
Prediction of post-ablation ventricular tachycardia recurrence in patients with non-ischemic cardiomyopathy--- Panel A1-3: Matching LGE-CMR (A1) and cardiac CECT (A2) images with LM highlighted with green overlay in the cardiac CECT (A3); Panel B. Restricted cubic spline model of the association between LM % and post-ablation VT recurrence. The blue line indicates the mean hazard ratio and the shaded blue region indicates the 95% confidence interval, with the confidence interval above the red line y =1 over most of the LM % range. Panel C. Survival curve from Cox proportional hazards regression model of LM % ≥ 2.5 g vs. LM % < 2.5 g, adjusted with the base model, in predicting VT recurrence (independent variables: presence of heart failure, obstructive sleep apnea, and use of anti-arrhthymics);
Abbreviations: LM = lipomatous metaplasia; VT = ventricular tachycardia.
In ROC analysis, an optimal threshold value of LM mass for predicting VT recurrence was computed as 2.5 g, with 72% sensitivity and 84% specificity. The unadjusted survival curve demonstrated that the presence of LM mass ≥ 2.5 g was strongly associated with the higher incidence of post-ablation VT recurrence, compared to the presence of LM mass < 2.5 g, with HR 5.4, p < 0.001. This association remained consistent in the multivariable model adjusted for heart failure, sleep apnea, and AAD use, with adjusted HR 4.9, p < 0.001 (Central Illustration C). Importantly, the presence of LM mass ≥ 2.5 g was consistently associated with VT recurrence during the 1st year follow-up, during the 2nd year follow-up, and after the 2nd year follow-up, respectively (Figure 3).
Figure 3 -.
Prediction of short- and long-term post-ablation ventricular tachycardia recurrence among patients with non-ischemic cardiomyopathy ---Panel A, B, C. Survival curve comparing LM mass ≥ 2.5 g vs. LM % < 2.5 g in predicting the cumulative incidence of VT recurrence during the 1st year, during the 2nd year, and >2 years after ablation, respectively.
Abbreviations: LM = lipomatous metaplasia; VT = ventricular tachycardia.
During a median follow-up of 552 (295, 1113) days, 32 VT recurrences occurred in the nested cohort. History of heart failure and AAD use were associated with VT recurrence. Function and volume measurements were unassociated with VT recurrence. Notably, the degree of fibrosis [HR 1.0 (0.9, 1.2) per 10% increase, p = 0.971], as well as border zone and dense LGE percentage, was not associated with VT recurrence. On the other hand, mean LV wall thickness was associated with VT recurrence on univariable analysis, with HR 1.5 per 1 mm decrease, p = 0.033, and in the multivariable analysis adjusted for heart failure and AAD use, with HR 1.6 per 1 mm decrease, p = 0.023. Similar to the full cohort, LM mass was an independent predictor of VT recurrence, with univariable HR 1.1 per 1 g increase, p = 0.004, and multivariable adjusted for heart failure and AAD use HR 1.1 per 1 g increase, p = 0.023 (Table 3 & 4). The ROC analysis showed that LM mass exhibited better discriminative performance for VT recurrence than the mean LV wall thickness, with the area under the curve 0.849 (0.756, 0.942) versus 0.662 (0.525, 0.799), P = 0.036.
DISSCUSSION
Main findings
The main findings of the current study were that: 1) Myocardial LM was detected in every NICM patient with post-ablation VT recurrence, in contrast to 83% of patients without post-ablation VT recurrence; 2) Despite similar LGE content, patients with VT recurrence exhibited a significantly higher amount of LM than those without VT recurrence; and 3) Patients with LM mass ≥ 2.5 g had 4.9-fold hazard of VT recurrence compared to those with LM < 2.5 g.
Prevalence of LM in NICM with versus without VT cohorts
NICM represents a heterogeneous combination of cardiomyopathies manifesting with ventricular dysfunction and heart failure symptoms, with inherent risk for sudden death (3). Sustained monomorphic VT in patients with NICM is a significant contributor to sudden death. Catheter ablation is an important adjunct or alternative to AAD for VT suppression. Our cohort characteristics, including NICM etiologies, age, and ventricular function were similar to those of prior studies of NICM patients (3,4,6,19). In contrast, Di Marco et al. reported that only 44% of NICM patients exhibited LGE, significantly lower than the 100% in the current study; however, not all patients in that study had VT (6). Piers et al. reported that 90% of NICM patients with VT exhibited LGE, while only 66% of NICM patients referred for primary ICD implantation had LGE (20). The direct association of LGE and LM prevalence and extent with VT is demonstrated in the current study. NICM patients with VT exhibit advanced myocardial disease at the tissue characterization level, with extensive LGE and LM and are substantially different from general NICM patients.
Risk of VT recurrence after ablation in NICM
Early VT recurrence after ablation in NICM patients with severe LV dysfunction is associated with mortality and heart transplantation (4). Nevertheless, data regarding the risk of recurrent VT following VT ablation in NICM patients is limited. Vergara et al. established a risk score for predicting VT recurrence based on LVEF, presence of cardiac devices, and programmed stimulation results after ablation in a mixed cohort of NICM and ICM patients (21). Distinct etiologies of NICM,(3,22,23) and inducibility of VT at the end of ablation are associated with VT recurrence (24,25). To our knowledge, no studies have investigated the prognostic value of tissue characterization in NICM patients with VT (8–10,18,26). Our group has demonstrated that LM alters the electrophysiological properties of diseased myocardium and constitutes the primary substrate of re-entrant VT in post-infarct patients. (8–10,18)
In the current study, LM was present in 57% NICM patients without VT and 90% of NICM patients with VT. Lu et al. reported only a 12.9% LM presence detected by fat-water separation sequence in CMR imaging among patients with dilated cardiomyopathy (19). Di Bella et al. reported a 15% LM presence identified as “India Ink” in functional cine CMR imaging in patients with prior myocarditis (14). Both reported significantly lower LM incidence than that detected by high-resolution CECT in the current study sample of patients with NICM with and without VT. Our study includes a higher proportion of patients with VT and demonstrates that myocardial injury in patients with NICM and VT is more severe than general NICM patients. Furthermore, the higher LM prevalence in our study is potentially attributable to CECT exhibiting higher sensitivity and specificity in detecting LM than CMR, especially when LM is small, located within a thin myocardial wall, or close to or at the apex. In our study, the myocardium in NICM patients without VT contained a small LM mass, with a median of 0.2 g that could easily be missed by less sensitive imaging modalities, noting CECT isotropic resolution of 0.45 mm versus cine and fat-water separation CMR in-plane resolution of 0.8 mm with 0.2 mm interslice gaps (27).
In the current study, a higher LM prevalence and extent was detected in NICM patients developing post-ablation VT recurrence compared to those free from recurrence, while the content of LGE exhibited no difference between the two groups. A stepwise increase in LM prevalence and a stepwise increase in LM mass was observed from the NICM patients without VT, to NICM patients with VT but no VT recurrence, to NICM patients with VT and VT recurrence, which reflects a “dose response” association. Notably, LM was independently associated with VT recurrence after multivariable adjustment. This study and findings reported by Lu et al. showed that LM is distributed within regions with LGE (19). The reparative fibrosis in NICM might be a precursor of LM, but LM is independent of fibrosis in predicting VT recurrence in the current study. Like myocardial LM in patients with prior infarction (8,28), myocardial LM in NICM is not homogeneous like epicardial adipose tissue or subcutaneous adipose tissue but heterogeneously mixed with fibrosis and residual myocardium. This structural complexity may explain the association between LM and VT recurrence following ablation. LM exhibits higher impedance than myocardial fibrosis (8), and may therefore isolate VT corridors from detection during the electrogram recording. Second, pacing at the surface shielded by LM may result in a surface 12 lead electrocardiogram QRS morphology distinct from the VT exit site because the adjacent LM might affect wavefront propagation. Third, depending upon the pacing site, programmed electrical stimulation might not be able to enter the myocardial corridor isolated by LM, resulting in VT non inducibility during the procedure. Fourth, LM may hinder the delivery to the targeted myocardium of resistive or even conductive heating with radiofrequency ablation. Finally, after the ablation procedure, LM, as well as the potentially increased LM over time, may secrete inflammatory cytokines (29), continue to alter the electrophysiological features of neighboring myocardium, and facilitate additional new VT circuit development, therefore promoting future VT recurrences. Enough LM might be needed to enhance the above features, which may explain the association between the relative mass of LM and VT recurrence. A threshold of ≥ 2.5 g LM was computed in the current study, indicating that ≥ 2.5 g LM yields a higher rate of VT recurrences.
Limitations
The current study has several limitations. The sample size of this study was modest. External validation in a larger independent cohort may refine the results. Nevertheless, we have acknowledged that this study was limited in power and that future studies with a larger cohort and ability to adjust for additional confounders are warranted. ICD systems may cause artifacts in the interventricular septum of CECT images. The patients with obvious imaging artifacts were excluded as shown in the patient identification flowchart. For the remaining patients, if artifact was noted, a lower attenuation threshold of −180 HU was utilized to filter artifacts. In the border of the artifact, we manually and carefully excluded pixels with attenuation values between −180 to 0 HU. Therefore, the LM located in the septum might be slightly underestimated.
Conclusion
Myocardial LM is commonly observed in patients with NICM, and documented VT. Myocardial LM is an independent predictor of post-ablation VT recurrence in patients with NICM.
Supplementary Material
Clinical perspectives: Competency in Medical Knowledge
Myocardial LM was found in 90% of patient with NICM and documented VT and was independently associated with post-ablation VT recurrence.
Patients with ≥ 2.5 g LM mass had 5-fold higher hazard of VT recurrence than those with < 2.5 g LM mass.
Translational outlook
Myocardial LM extent measurements may improve early risk stratification of post-ablation VT recurrence in patients with NICM.
The association of myocardial LM with post-ablation VT recurrence suggests potential mechanisms for mapping and ablation failure, which warrant further study.
Funding:
The INFINITY study is funded by NIH grant 1R01HL142893-01
Disclosure:
Dr. Nazarian is a consultant for CardioSolv and Circle CVI; and principal investigator for research funding from Biosense Webster, ImriCor, Siemens, ADAS software, and the US NIH. Dr. Marchlinski has served as consultant for Abbott Medical, Biosense Webster, Biotronik, and Medtronic Inc. Dr. Xu is an American Heart Association postdoctoral fellow (23POST909139). The University of Pennsylvania Conflict of Interest Committee manages all commercial arrangements. The remaining authors have nothing to disclose.
Abbreviations
- NICM
non-ischemic cardiomyopathy
- VT
ventricular tachycardia
- LM
lipomatous metaplasia
- INFINITY
Prospective Cohort Study of Mechanistic Associations between Intra-Myocardial Fat Deposition and Ventricular Tachycardia in Cardiomyopathy
- LGE
late gadolinium enhancement
- CMR
cardiovascular magnetic resonance
- CECT
contrast-enhanced computed tomography
- BZ
border zone
- SD
standard deviation
- IQR
interquartile range
- ICD
implantable cardioverter-defibrillator
- HU
Hounsfield unit
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Briceno DF, Gupta T, Romero J et al. Catheter ablation of ventricular tachycardia in nonischemic cardiomyopathy: A propensity score-matched analysis of in-hospital outcomes in the United States. J Cardiovasc Electrophysiol 2018;29:771–779. [DOI] [PubMed] [Google Scholar]
- 2.Quinto L, Sanchez-Somonte P, Alarcon F et al. Ventricular tachycardia burden reduction after substrate ablation: Predictors of recurrence. Heart Rhythm 2021;18:896–904. [DOI] [PubMed] [Google Scholar]
- 3.Vaseghi M, Hu TY, Tung R et al. Outcomes of Catheter Ablation of Ventricular Tachycardia Based on Etiology in Nonischemic Heart Disease: An International Ventricular Tachycardia Ablation Center Collaborative Study. JACC Clin Electrophysiol 2018;4:1141–1150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zeppenfeld K, Wijnmaalen AP, Ebert M et al. Clinical Outcomes in Patients With Dilated Cardiomyopathy and Ventricular Tachycardia. J Am Coll Cardiol 2022;80:1045–1056. [DOI] [PubMed] [Google Scholar]
- 5.Ammar A, Sharief M, Abouelmagd K, Riad O, Ibrahim M. Outcomes of catheter ablation of ventricular tachycardia in non-ischemic idiopathic dilated cardiomyopathy: A systematic review and meta-analysis. Front Cardiovasc Med 2022;9:1007392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Di Marco A, Brown PF, Bradley J et al. Improved Risk Stratification for Ventricular Arrhythmias and Sudden Death in Patients With Nonischemic Dilated Cardiomyopathy. J Am Coll Cardiol 2021;77:2890–2905. [DOI] [PubMed] [Google Scholar]
- 7.Kellman P, Hernando D, Arai AE. Myocardial Fat Imaging. Curr Cardiovasc Imaging Rep 2010;3:83–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Xu L, Khoshknab M, Berger RD et al. Lipomatous Metaplasia Enables Ventricular Tachycardia by Reducing Current Loss Within the Protected Corridor. JACC Clin Electrophysiol 2022;8:1274–1285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Xu L, Zahid S, Khoshknab M et al. Lipomatous metaplasia prolongs repolarization and increases repolarization dispersion within post-infarct ventricular tachycardia circuit cites. Europace 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sung EPA, Zhou S, Berger RD, Chrispin J, Nazarian S, Trayanova NA. Fat infiltration in the infarcted heart as a paradigm for ventricular arrhythmias. NatureCardiovascular Research 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Thiene G, Nava A, Corrado D, Rossi L, Pennelli N. Right ventricular cardiomyopathy and sudden death in young people. N Engl J Med 1988;318:129–33. [DOI] [PubMed] [Google Scholar]
- 12.Baroldi G, Silver MD, De Maria R, Parodi O, Pellegrini A. Lipomatous metaplasia in left ventricular scar. Can J Cardiol 1997;13:65–71. [PubMed] [Google Scholar]
- 13.Moura-Ferreira S, Van Cleemput J, Verbeken E, Jan B. Progressive myocardial lipomatous metaplasia following acute myocarditis. Eur Heart J Cardiovasc Imaging 2020;21:705. [DOI] [PubMed] [Google Scholar]
- 14.Di Bella G, Gentile G, Irsuti F et al. Prognostic Role of Left Ventricular Intramyocardial Fatty Metaplasia in Patients With Previous Myocarditis (MYOFAT Study). Am J Cardiol 2021;143:135–144. [DOI] [PubMed] [Google Scholar]
- 15.Saremi F, Saremi A, Hassani C et al. Computed Tomographic Diagnosis of Myocardial Fat Deposits in Sarcoidosis. J Comput Assist Tomogr 2015;39:578–83. [DOI] [PubMed] [Google Scholar]
- 16.Cha MJ, Seo JW, Oh S et al. Indirect pathological indicators for cardiac sarcoidosis on endomyocardial biopsy. J Pathol Transl Med 2020;54:396–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Andreu D, Ortiz-Perez JT, Fernandez-Armenta J et al. 3D delayed-enhanced magnetic resonance sequences improve conducting channel delineation prior to ventricular tachycardia ablation. Europace 2015;17:938–945. [DOI] [PubMed] [Google Scholar]
- 18.Sasaki T, Calkins H, Miller CF et al. New insight into scar-related ventricular tachycardia circuits in ischemic cardiomyopathy: Fat deposition after myocardial infarction on computed tomography-A pilot study. Heart Rhythm 2015;12:1508–1518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lu M, Zhao S, Jiang S et al. Fat deposition in dilated cardiomyopathy assessed by CMR. JACC Cardiovasc Imaging 2013;6:889–98. [DOI] [PubMed] [Google Scholar]
- 20.Piers SR, Everaerts K, van der Geest RJ et al. Myocardial scar predicts monomorphic ventricular tachycardia but not polymorphic ventricular tachycardia or ventricular fibrillation in nonischemic dilated cardiomyopathy. Heart Rhythm 2015;12:2106–14. [DOI] [PubMed] [Google Scholar]
- 21.Vergara P, Tzou WS, Tung R et al. Predictive Score for Identifying Survival and Recurrence Risk Profiles in Patients Undergoing Ventricular Tachycardia Ablation: The I-VT Score. Circ Arrhythm Electrophysiol 2018;11:e006730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Shirai Y, Liang JJ, Santangeli P et al. Comparison of the Ventricular Tachycardia Circuit Between Patients With Ischemic and Nonischemic Cardiomyopathies: Detailed Characterization by Entrainment. Circ Arrhythm Electrophysiol 2019;12:e007249. [DOI] [PubMed] [Google Scholar]
- 23.Kumar S, Romero J, Mehta NK et al. Long-term outcomes after catheter ablation of ventricular tachycardia in patients with and without structural heart disease. Heart Rhythm 2016;13:1957–63. [DOI] [PubMed] [Google Scholar]
- 24.Okubo K, Gigli L, Trevisi N et al. Long-Term Outcome After Ventricular Tachycardia Ablation in Nonischemic Cardiomyopathy: Late Potential Abolition and VT Noninducibility. Circ Arrhythm Electrophysiol 2020;13:e008307. [DOI] [PubMed] [Google Scholar]
- 25.Arya A, Bode K, Piorkowski C et al. Catheter Ablation of Electrical Storm Due to Monomorphic Ventricular Tachycardia in Patients with Nonischemic Cardiomyopathy: Acute Results and Its Effect on Long-Term Survival. Pace 2010;33:1504–1509. [DOI] [PubMed] [Google Scholar]
- 26.Pouliopoulos J, Chik WW, Kanthan A et al. Intramyocardial adiposity after myocardial infarction: new implications of a substrate for ventricular tachycardia. Circulation 2013;128:2296–308. [DOI] [PubMed] [Google Scholar]
- 27.Xu L, Desjardins B, Witschey WR, Nazarian S. Noninvasive Assessment of Lipomatous Metaplasia as a Substrate for Ventricular Tachycardia in Chronic Infarct. Circ Cardiovasc Imaging 2023;16:e014399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.de Bakker JM, van Capelle FJ, Janse MJ et al. Slow conduction in the infarcted human heart. ‘Zigzag’ course of activation. Circulation 1993;88:915–26. [DOI] [PubMed] [Google Scholar]
- 29.Mazurek T, Zhang LF, Zalewski A et al. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation 2003;108:2460–2466. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.