Abstract
Background: The association between the T-peak to T-end interval (Tp-e) and ventricular arrhythmia (VA) events in cardiac sarcoidosis (CS) is unknown. The purpose of this study was to investigate whether Tp-e was associated with VA events in CS patients with implantable cardioverter defibrillators (ICDs) or cardiac resynchronization therapy defibrillators (CRT-Ds).
Methods and Results: We retrospectively studied 50 patients (16 men; mean [±SD] age 56.3±10.5 years) with CS and ICD/CRT-D. The maximum Tp-e in the precordial leads recorded by a 12-lead electrocardiogram after ICD/CRT-D implantation was evaluated. The clinical endpoint was defined as appropriate ICD therapy. During a median follow-up period of 85.0 months, 22 patients underwent appropriate therapy and 10 patients died. Kaplan-Meier analysis revealed that the probability of the clinical endpoint was 28.3% at 2 years and 35.3% at 4 years. The optimal cut-off value of the Tp-e for the prediction of the clinical endpoint was 91 ms, with a sensitivity of 72.7% and a specificity of 87.0% (area under the curve=0.81). Multivariate Cox regression analysis showed that Tp-e ≥91 ms (hazard ratio [HR] 5.10; 95% confidence interval [CI] 1.99–13.1; P<0.001) and a histological diagnosis of CS (HR 3.84; 95% CI 1.28–11.5; P=0.016) were significantly associated with the clinical endpoint.
Conclusions: Tp-e ≥91 ms was a significant predictor of VA events in patients with CS and ICD/CRT-D.
Sarcoidosis is a granulomatous disease of unknown etiology with systemic involvement including the heart, lungs, eyes, skin, and other organs.1 Patients with cardiac sarcoidosis (CS) are at high risk of ventricular arrhythmias (VAs) and progression of heart failure; therefore, implantation of implantable cardioverter defibrillators (ICDs) or cardiac resynchronization therapy defibrillators (CRT-Ds) may be required for primary/secondary prevention of sudden cardiac death (SCD).2 Long-term follow-up data regarding the incidence and efficacy of ICD therapy in CS patients implanted with ICD/CRT-D are limited. The presence of late gadolinium enhancement (LGE) on cardiac magnetic resonance imaging (MRI) has been reported to be a predictor of death and VAs in CS patients; however, gadolinium-enhanced cardiac MRI is contraindicated in patients with renal impairment or abandoned pacing leads. Thus, non-invasive electrocardiographic predictors are required; nevertheless, electrocardiographic parameters for the prediction of VA events in CS patients have not been fully evaluated.
The time interval from the peak to the end of the electrocardiographic T wave (Tp-e) is thought to reflect not only transmural dispersion of repolarization (TDR),3 but also whole-heart dispersion of repolarization (DOR).4,5 Therefore, a prolonged Tp-e has been proposed as an independent predictor of SCD in patients with long QT syndrome, Brugada syndrome, or hypertrophic cardiomyopathy.6–8 Tp-e may also be a marker of SCD in patients with CS.9
Accordingly, the aims of this study were to: (1) clarify the characteristics of VA episodes and evaluate the incidence and efficacy of ICD therapy; and (2) investigate whether Tp-e was associated with VA events in patients with CS and ICD/CRT-D.
Methods
Study Population and Ethical Considerations
In all, 135 patients diagnosed with CS between April 1999 and April 2020 were reviewed for study eligibility. We enrolled 50 patients with CS who had undergone ICD or CRT-D implantation. We confirmed that all 50 patients met either the histological or clinical diagnosis criteria of CS according to the latest guidelines from the Japanese Circulation Society (JCS).2 Even in patients without histological evidence of CS in organs other than the heart, isolated CS was diagnosed if they met the diagnostic criteria in the latest JCS guideline.2
This single-center observational retrospective cohort study was approved by the Ethics Committee of Tokyo Women’s Medical University (Approval no. 4152-R2) and was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all patients.
Defibrillator Device Implantation, Programming, and Definition of ICD Therapy
The indication for ICD/CRT-D was based on JCS guidelines and representative guidelines from the Heart Rhythm Society.10,11 Standard procedures were used for device implantation.
The ventricular fibrillation (VF) zone was programmed at 220 beats/min, and the initial therapy in the VF zone was shock therapy at ≥30 J following antitachycardia pacing (ATP) before/during charge, if available. The ventricular tachycardia (VT) zone was programmed based on the clinically documented VT rate. Multiple ATP sequences were programmed prior to the first shock therapy. In the VT zone, 2–4 sequences of burst pacing were commonly used with 8–12 pulses at 81–90% of the VT cycle length (VT-CL), whereas 2–4 sequences of burst pacing with 8–15 pulses at 75–90% of the VT-CL were used in the slow VT zone (<150 beats/min). VF episodes were defined as appropriate VAs detected in the VF zone, whereas VT episodes were defined as appropriate VAs detected in the VT zone. A successful ATP was defined as the termination of VT or VF by a programmed series of ATPs without shock therapy. Device follow-up was performed every 6–12 months at the outpatient device clinic using a remote monitoring system. The mean VT-CL and number of delivered ATPs were evaluated. VA events were divided into quartiles according to the VT-CL to compare the efficacy of ATP.
Twelve-Lead Electrocardiogram Recording and Measurement of Tp-e
The resting 12-lead electrocardiogram (ECG) was recorded at 25 mm/s with an amplitude of 10 mm/mV using a standard recorder (Nihon Kohden Corporation, Tokyo, Japan). The ECG parameters, including PQ interval, QRS duration, QT interval, corrected QT interval, and Tp-e, were measured at discharge after ICD/CRT-D implantation. Tp-e was measured from the T wave peak to T wave end at a sweep speed of 50 mm/s. The end of the T wave was defined as the intersection of the tangent to the down slope of the T wave and the isoelectric line. If a U wave was present, the T wave offset was measured as the nadir between the T and U waves. The maximum Tp-e in the precordial leads during a single beat was used in the analysis.7,12 Tp-e were measured by 2 independent electrophysiologists blinded to the clinical outcomes. Signal-averaged electrocardiography was also performed using CARDIO STAR FX-7542 (Fukuda Denshi, Tokyo, Japan) with 40–250 Hz filtering and X, Y, and Z leads. Filtered QRS (fQRS) duration, the amount of time that the fQRS complex remains below 40 µV (LAS40), and the root mean square voltage of the terminal 40 ms of the fQRS (RMS40) were automatically retrieved.
Endomyocardial Biopsy
For endomyocardial biopsy, 3–4 specimens were obtained at random from the right ventricular septum. In patients with highly suspected CS but no histopathological findings, we performed voltage mapping-guided biopsy as a second attempt at endomyocardial biopsy.13 A voltage map of the right ventricle was created using a 3-dimensional (3D) mapping system, such as CARTO (Biosense-Webster, Diamond Bar, CA, USA) or EnSite (Abbott, Minneapolis, MN, USA). Biopsy specimens were obtained from the low-voltage zone, below 1.5 mV, as indicated by the voltage map. In the absence of such a zone on the voltage map, biopsy specimens were obtained from the right ventricular septum where a fractionated potential was present or where LGE was observed on cardiac MRI. Specimens obtained by endomyocardial biopsy that showed non-caseating epithelioid granulomas were considered as a histological diagnosis of CS.
Clinical Endpoint
Long-term follow-up was performed via a review of medical records, device interrogation, and discharge summaries for all study patients. The clinical endpoint of this study was the first episode of appropriate ICD therapy. Patients were divided into 2 groups according to the presence or absence of the clinical endpoint to compare baseline characteristics.
Statistical Analysis
Categorical variables are expressed as numbers and proportions and were compared using Fisher’s exact test. Continuous variables are presented as the mean±SD or as the median with interquartile range (IQR) and were compared using the Wilcoxon exact test. Interobserver variability for measuring Tp-e was assessed using linear regression analysis. The success rate of ATP termination among VT-CL quartiles was determined using one-way analysis of variance (ANOVA). Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cut-off value of Tp-e to predict the clinical endpoint. Kaplan-Meier analysis was used to estimate the time from ICD/CRT-D implantation to the first episode of appropriate ICD therapy. The log-rank test was used to assess the significance of intergroup differences. Cox proportional hazards regression models were used to examine the factors associated with the clinical endpoint. Age, sex, other baseline variables with P<0.10 in the univariate analysis, and clinically important factors, such as amiodarone use and left ventricular ejection fraction (LVEF), were included as covariates in the multivariate analysis. Two-sided P<0.05 was considered statistically significant. Statistical analyses were performed using JMP Pro 16 software (SAS Institute Inc., Cary, NC, USA).
Results
Clinical Characteristics of the Study Cohort
Fifty patients with CS and ICD/CRT-D implantation were included in this study (Table 1). The mean patient age was 56.3±10.5 years and 68% of the study patients were women. Among the 50 patients, 48 underwent endomyocardial biopsy procedures and 2 underwent surgical biopsy during open chest surgery. CS was histologically diagnosed in 56% of the study population, whereas the others were diagnosed clinically. Fifteen patients (30%) were diagnosed with isolated CS. Six patients underwent a second endomyocardial biopsy using 3D mapping systems. Among these 6 patients, non-caseating epithelioid granulomas were identified in 4. More than 90% of all patients were treated with both β-blockers and an angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker. Amiodarone was prescribed for 40% of patients, and prednisolone was used by 90%.
Table 1. Baseline Characteristics in the Total Population and in Patients With or Without Appropriate Therapy
| |
Total
(n=50) |
Appropriate therapy |
P value |
| Yes (n=22) |
No (n=28) |
| Age (years) |
56.3±10.5 |
54.6±11.2 |
57.6±9.9 |
0.329 |
| Female sex |
34 (68.0) |
12 (54.5) |
22 (78.6) |
0.225 |
| NYHA functional class |
| II |
42 (84.0) |
17 (77.3) |
25 (89.3) |
0.373 |
| III |
7 (14.0) |
4 (18.2) |
3 (10.7) |
| IV |
1 (2.0) |
1 (4.5) |
0 (0) |
| Conduction disorder |
| Complete AVB |
22 (44.0) |
10 (45.5) |
12 (42.9) |
1.000 |
| IVCD |
3 (6.0) |
1 (4.5) |
2 (7.1) |
1.000 |
| Left bundle branch block |
1 (2.0) |
0 (0.0) |
1 (3.6) |
1.000 |
| Right bundle branch block |
11 (22.0) |
6 (27.3) |
5 (17.9) |
0.503 |
| Trifascicular block |
3 (6.0) |
1 (4.5) |
2 (7.1) |
1.000 |
| Follow-up from ICD/CRT-D implantation (months) |
85.0 [54.3–133.0] |
91.5 [55.3–148.3] |
82.0 [52.8–129.5] |
0.611 |
| Histological diagnosis |
28 (56.0) |
16 (72.7) |
12 (42.9) |
0.047 |
| Isolated cardiac sarcoidosis |
15 (30.0) |
7 (31.8) |
8 (28.6) |
1.000 |
| Involvement of other organs |
| Bone |
2 (4.0) |
1(4.5) |
1 (3.6) |
1.000 |
| Eye |
11 (22.0) |
4 (18.2) |
7 (25.0) |
0.734 |
| Liver |
1 (2.0) |
1 (4.5) |
0 (0) |
0.440 |
| Lung |
29 (58.0) |
12 (54.5) |
17 (60.7) |
0.775 |
| Muscle |
1 (2.0) |
1 (4.5) |
0 (0) |
0.440 |
| Skin |
5 (10.0) |
1 (4.5) |
4 (14.3) |
0.368 |
| Spleen |
1 (2.0) |
0 (0) |
1 (3.6) |
1.000 |
| Medication |
| Amiodarone |
20 (40.0) |
11 (50.0) |
9 (32.1) |
0.251 |
| ACEI/ARB |
45 (90.0) |
20 (90.9) |
25 (89.3) |
1.000 |
| β-blocker |
48 (96.0) |
22 (100.0) |
26 (92.9) |
0.497 |
| Methotrexate |
2 (4.0) |
1 (4.5) |
1 (3.6) |
1.000 |
| Prednisolone |
45 (90.0) |
19 (86.4) |
26 (92.9) |
0.643 |
| Atrial fibrillation |
14 (28.0) |
7 (31.8) |
7 (25.0) |
0.752 |
| Laboratory data |
| ACE (IU/L) |
15.6±11.6 |
19.0±14.5 |
12.8±7.7 |
0.068 |
| BNP (pg/mL) |
116.8 [46.8–234.6] |
169.8 [59.0–392.5] |
100.4 [44.3–202.0] |
0.309 |
| Lysozyme (μg/mL) |
8.8±4.0 |
9.7±4.7 |
8.1±3.2 |
0.183 |
Values are presented as the mean±SD, n (%) or median [interquartile range], as appropriate. ACE, angiotensin-converting enzyme; ACEI, ACE inhibitor; ARB, angiotensin II receptor blocker; AVB, atrioventricular block; BNP, B-type natriuretic peptide; CRT-D, cardiac resynchronization therapy device with a defibrillator; ICD, implantable cardioverter defibrillator; IVCD, intraventricular conduction disturbance; NYHA, New York Heart Association.
During the median follow-up of 85.0 months (IQR 54.3–133.0 months), 22 (44%) patients reached the clinical endpoint. Patients were divided into 2 groups according to the presence of the clinical endpoint. There were no significant differences in baseline characteristics between the 2 groups.
Comparison of Electrocardiographic, Echocardiographic, and Imaging Parameters Between Patients With and Without the Clinical Endpoint
After ICD/CRT-D implantation, the rate of biventricular pacing, right ventricular pacing, and intrinsic ventricular activation was 54%, 12%, and 34%, respectively. There were no significant differences in the QRS configuration and QRS duration between patients with and without clinical events (Table 2). PQ and QT intervals were similar between the 2 groups; however, Tp-e was significantly longer in patients with than without clinical endpoints (97.0±16.0 vs. 80.6±9.1 ms, respectively; P<0.001). The median period from ICD/CRT-D implantation to Tp-e measurement was 6 days (IQR 3–11 days). Interobserver variability for measuring Tp-e was assessed, and linear regression analysis revealed an acceptable correlation (y=0.98x+2.04; R2=0.98, P<0.001). There were no significant differences in the parameters of signal-averaged electrocardiography.
Table 2. Electrocardiographic Findings After Device Implantation, Echocardiographic Parameters, and Imaging Modality Findings
| |
Total
(n=50) |
Appropriate therapy |
P value |
| Yes (n=22) |
No (n=28) |
| QRS configuration |
| Biventricular pacing |
27 (54.0) |
13 (59.1) |
13 (46.4) |
0.577 |
| RV pacing |
6 (12.0) |
2 (9.1) |
4 (14.3) |
0.683 |
| Intrinsic QRS |
17 (34.0) |
7 (31.8) |
10 (35.7) |
1.000 |
| QRS duration (ms) |
135.7±28.7 |
135.3±29.9 |
136.1±28.2 |
0.923 |
| PQ interval (ms) |
169.2±32.6 |
166.7±31.4 |
171.3±34.0 |
0.634 |
| QT interval (ms) |
448.2±39.0 |
445.1±43.1 |
450.6±36.2 |
0.627 |
| QTc interval (ms) |
461.6±38.5 |
459.0±41.8 |
463.6±36.3 |
0.673 |
| Tp-e (ms) |
89.4±15.5 |
97.0±16.0 |
80.6±9.1 |
<0.001 |
| Signal-averaged electrocardiography |
| f-QRS (ms) |
175.4±44.2 |
187.9±46.3 |
166.1±41.1 |
0.124 |
| RMS40 (μV) |
15.4±12.8 |
10.9±8.5 |
18.7±14.4 |
0.053 |
| LAS40 (ms) |
53.5±35.7 |
63.6±43.1 |
46.0±27.6 |
0.126 |
| Echocardiographic parameters |
| LA dimension (mm) |
3.8±0.7 |
3.7±0.5 |
3.9±0.8 |
0.261 |
| LV end-diastolic diameter (mm) |
6.0±0.9 |
6.0±1.0 |
6.0±0.8 |
0.926 |
| LV end-systolic diameter (mm) |
4.9±1.2 |
5.0±1.3 |
4.8±1.1 |
0.643 |
| LV end-diastolic volume (mL) |
202.2±57.9 |
207.2±51.5 |
195.2±67.7 |
0.609 |
| LV end-systolic volume (mL) |
146.0±47.9 |
149.0±48.0 |
142.0±49.8 |
0.721 |
| LVEF (%) |
35.0±12.4 |
33.6±10.4 |
36.1±13.8 |
0.483 |
| Abnormalities in other imaging modalities |
| Abnormal findings in cardiac MRI |
24/25 (96.0) |
11/11 (100.0) |
13/14 (92.9) |
1.000 |
| Abnormal findings in PET |
31/39 (79.5) |
15/17 (88.2) |
16/22 (72.7) |
0.426 |
| Abnormal findings in thallium scintigraphy |
34/37 (91.9) |
16/18 (88.9) |
18/19 (94.7) |
0.604 |
Values are presented as the mean±SD or as n (%), as appropriate. f-QRS, filtered QRS duration; LA, left atrium; LAS40, duration of low-amplitude signals <40 mV; LV, left ventricular; LVEF, LV ejection fraction; MRI, magnetic resonance imaging; PET, 18F-fluoro-2-deoxyglucose positron emission tomography; RMS40, root mean square voltage of the terminal 40 ms; RV, right ventricular; Tp-e, time interval from the peak to the end of the electrocardiographic T wave.
Mean LVEF in the overall study population was 35.0±12.4%. There were no significant differences in left ventricular volumetric parameters or the LVEF between patients with and without clinical events. There were also no significant differences in the prevalence of abnormalities on cardiac MRI, scintigraphy or 18F-fluoro-2-deoxyglucose positron emission tomography between the 2 groups.
Incidence of VAs and Efficacy of Defibrillator Device Therapy
In our study population, 23 (46%) patients had ICD and 27 (54%) had CRT-D. The indication of a defibrillator device for secondary prevention was significantly higher in patients with the clinical endpoint. The use of 3-zone settings was higher in patients with events; however, there were no significant intergroup differences in the tachycardia rate programming. A lower detection threshold of tachycardia detection zone was shown in Table 3.
Table 3. Device Characteristics
| |
Total
(n=50) |
Appropriate therapy |
P value |
| Yes (n=22) |
No (n=28) |
| Device |
| ICD |
23 (46.0) |
9 (40.9) |
14 (50.0) |
0.577 |
| CRT-D |
27 (54.0) |
13 (59.1) |
14 (50.0) |
| Defibrillator indication |
| Primary prevention |
33 (66.0) |
12 (54.5) |
21 (75.0) |
0.147 |
| Secondary prevention |
17 (34.0) |
10 (45.5) |
7 (25.0) |
| Therapy zone settings |
| 2 zones |
32 (64.0) |
7 (31.8) |
25 (89.3) |
<0.001 |
| 3 zones |
18 (36.0) |
15 (68.2) |
3 (10.7) |
| Lower detection zone (beats/min) |
| VF zone |
220 [214–222] |
222 [211–224] |
220 [214–222] |
0.101 |
| Fast VT zone |
171 [150–177] |
171 [150–180] |
170 [150–173] |
0.937 |
| 180≤FVT zone<190 |
12 (24.0) |
6 (27.3) |
6 (21.4) |
0.431 |
| 170≤FVT zone<180 |
17 (34.0) |
7 (31.8) |
10 (35.7) |
| 160≤FVT zone<170 |
3 (6.0) |
1 (4.6) |
2 (7.1) |
| 150≤FVT zone<160 |
12 (24.0) |
4 (18.2) |
8 (28.6) |
| 140≤FVT zone<150 |
3 (6.0) |
1 (4.6) |
2 (7.1) |
| 130≤FVT zone<140 |
3 (6.0) |
3 (13.6) |
0 (0) |
| VT zone |
130 [116–150] |
130 [111–150] |
130 [120–145] |
0.952 |
| 150≤VT zone<160 |
5/18 (27.8) |
5 /15 (33.3) |
0/3 (0) |
0.441 |
| 140≤VT zone<150 |
2/18 (11.1) |
1 /15 (6.7) |
1/3 (33.3) |
| 130≤VT zone<140 |
3/18 (16.7) |
2 /15 (13.3) |
1/3 (33.3) |
| 120≤VT zone<130 |
3/18 (16.7) |
2 /15 (13.3) |
1/3 (33.3) |
| 110≤VT zone<120 |
4/18 (22.2) |
4 /15 (26.7) |
0/3 (0) |
| 100≤VT zone<110 |
1/18 (5.6) |
1 /15 (6.7) |
0/3 (0) |
Values are presented as the mean±SD, n (%) or median [interquartile range], as appropriate. FVT, fast ventricular tachycardia; VF, ventricular fibrillation; VT, ventricular tachycardia. Other abbreviations as in Table 1.
In all, 306 ICD therapy episodes were observed in 22 patients during a median follow-up of 85.0 months. Among these 306 episodes, 12 (3.9%) were inappropriate therapy due to atrial fibrillation (8 episodes), supraventricular tachycardia (3 episodes), and lead noise (1 episode). Among the 294 appropriate therapy episodes, 9 (3.1%) were VF and the remaining 285 (96.9%) were VT. Among the 9 VF episodes, ATP during/before charge successfully terminated 4 (44.4%) episodes, whereas the other 5 episodes were terminated by shock therapy. Among the 285 VT episodes, 247 VT episodes (86.7%) were successfully terminated by ATP, 28 episodes were terminated by shock therapy following failed ATPs, and 10 episodes spontaneously terminated before shock therapy.
VT episodes were divided into quartiles according to the VT-CL to evaluate the efficacy of ATP. The mean VT-CL was 425.3±60.6 ms (IQR 375.0–461.5 ms). The success rate of ATP in VT-CL Q1 was significantly lower (71.0%) than that in the other quartiles (Figure 1). Detailed histograms of VT-CL are shown in Figure 2. VT-CL was significantly longer in patients receiving secondary prevention than in those receiving primary prevention (430.0±57.5 vs. 394.6±71.6 ms, respectively; P<0.001). Furthermore, the VT-CL in CRT-D patients was significantly longer than that in ICD patients (434.8±59.9 vs. 377.1±36.8 ms, respectively; P<0.001).
Tp-e and the Clinical Endpoint
During the median follow-up of 85.0 months (IQR 54.3–133.0 months), 22 patients received appropriate ICD therapy and 10 patients died. Among the 10 deaths, 6 were due to heart failure and the remaining 4 were non-cardiac deaths.
The cumulative probability of appropriate ICD therapy was 28.3% at 2 years and 35.3% at 4 years (Figure 3). The median duration from ICD/CRT-D implantation to the first appropriate therapy was 15.5 months (IQR 4.5–48.5 months).
ROC curve analysis showed that the optimal cut-off value of Tp-e for predicting the clinical endpoint was 91 ms, with a sensitivity and specificity of 72.7% and 87.0%, respectively. The area under the curve (AUC) was 0.81. In multivariate Cox regression analysis, Tp-e ≥91 ms (hazard ratio [HR] 5.10; 95% confidence interval [CI] 1.99–13.1; P<0.001) and a histological diagnosis of CS (HR 3.84; 95% CI 1.28–11.5; P=0.016) were significantly associated with appropriate ICD therapy (Table 4). Kaplan-Meier analysis revealed that the cumulative probability of the clinical endpoint was significantly higher in patients with Tp-e ≥91 ms than in patients with Tp-e <91 ms (log-rank, P<0.001; Figure 4A). Similarly, histological diagnosis and male sex were significantly associated with a higher incidence of the clinical endpoint (Figure 4B,C); however, defibrillator indication was not associated with the incidence of the clinical endpoint during long-term follow-up (Figure 4D).
Table 4. Uni- and Multivariate Analysis for Predictors of Appropriate Therapy
| |
Appropriate therapy |
| Univariate analysis |
Multivariate analysis |
| HR (95% CI) |
P value |
HR (95% CI) |
P value |
| Sex, male |
2.61 (1.11–6.11) |
0.027 |
2.15 (0.84–5.49) |
0.110 |
| Age (1-year increase) |
0.99 (0.95–1.03) |
0.676 |
1.02 (0.97–1.08) |
0.371 |
| Histological diagnosis |
3.18 (1.23–8.19) |
0.017 |
3.84 (1.28–11.5) |
0.016 |
| Tp-e ≥91 ms |
5.46 (2.21–13.5) |
<0.001 |
5.10 (1.99–13.1) |
<0.001 |
| Amiodarone use |
1.96 (0.85–4.55) |
0.116 |
1.84 (0.71–4.79) |
0.208 |
| LVEF (1% increase) |
0.98 (0.95–1.02) |
0.326 |
0.99 (0.95–1.04) |
0.832 |
CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 2.
Measuring Tp-e during pacing may introduce heterogeneity in the study population because the study population included patients with CRT-D (receiving biventricular pacing) and patients with ICD (right ventricular pacing or intrinsic rhythm). We performed another analysis using Tp-e measured at the baseline intrinsic rhythm without ventricular pacing before device implantation. The optimal cut-off value of Tp-e at the baseline intrinsic rhythm for prediction of the clinical endpoint was 90 ms, with a sensitivity of 77.3% and specificity of 71.4% (AUC=0.77). Tp-e ≥90 ms was also associated with VA events (Supplementary Table).
Discussion
In this long-term follow-up study of patients with CS and ICD/CRT-D, we clarified the incidence and efficacy of ICD therapy for VAs and demonstrated that prolonged Tp-e was an independent predictor of VA events. The major findings of this study are that: (1) the incidence of the first appropriate ICD therapy was nearly 30% at 2 years after ICD/CRT-D implantation; (2) most of the appropriate ICD therapies were due to VT (97%); (3) the efficacy of ATP for the termination of VT and VF was 87% and 44%, respectively, and the efficacy of ATP for fast VT (VT-CL <375 ms) was significantly lower; and (4) prolonged Tp-e was an independent predictor of VA events, with an optimal cut-off of 91 ms.
Incidence and Characteristics of VAs, and Efficacy of ATP in Patients With CS
Patients with CS are known to be at high risk of VA and SCD. The ventricular arrhythmogenic substrate is based on the scar/fibrosis following inflammatory damage due to granulomas. Several imaging modalities have been studied for the risk stratification of SCD in CS patients. It is considered that the primary substrate of VA may be due to the amount of scar/fibrosis lesions rather than the inflammatory condition that reflects disease activity.14–16 Moreover, it has been reported that Purkinje-related VT may be likely in CS patients.17 Therefore, these fixed substrates based on scar/fibrosis or a damaged Purkinje system could be responsible for the frequent occurrence of stable sustained VT in patients with CS.
Our study included CS patients with ICD/CRT-D who were treated with optimal medical therapy and steroids. The incidence of the first appropriate therapy within 2 and 4 years was 28.3% and 35.3%, respectively, which is consistent with previous reports.18,19 However, the incidence of inappropriate ICD therapy in the present study was relatively lower than that reported recently.18,19 Most of our study population was included in the “shock reduction” era; therefore, a high-rate cut-off setting and/or long-duration/detection programming was adopted.20 This may be a reason for the low incidence of inappropriate therapy.
Another feature of this study was the evaluation of the details of each appropriate ICD therapy episode in patients with CS. This study demonstrated that most of the appropriate therapy was needed due to VT, and this result is consistent with the mechanism of VAs in CS patients with the fixed substrates resulting from scarring/fibrosis following inflammation. In addition, the VT-CL was widely distributed from 563 to 278 ms, and slow VT (VT-CL ≥400 ms) accounted for 70.5% of the VA events. In a recent systematic review reported by Papageorgiou et al, the VT-CL induced by programmed electrical stimuli in CS patients was 320–400 ms,21 which was relatively faster than that observed in the present study. However, programmed electrical stimuli may not always induce clinical VT, and they may induce multiple non-clinical fast VTs. In addition, CRT-D was indicated for more than half the patients due to an enlarged left ventricular volume and reduced LVEF. Most patients in this study who underwent cardiac MRI had evidence of scar/fibrosis, as indicated by LGE. Moreover, amiodarone was used in approximately 50% of patients who received appropriate ICD therapy. Accordingly, slow VT can be more common in CS patients in clinical settings, and these findings may explain the high success rate of ATP termination. In general, ATP can be more effective for slow than fast VTs.22 In the present study, ATP was effective, with a success rate of 44%, even in the fast VAs detected in the VF zone; therefore, scar-related re-entrant VT may be common in such VF events.
With regard to clinical implications, careful management of VA events, including medications and VT ablation in the first 2 years after ICD/CRT-D implantation, is needed in CS patients due to the high incidence of VT events; therefore, a remote monitoring system could be helpful for the early detection of VA events. Clinical VAs in CS patients with ICD/CRT-Ds are mostly scar-related VTs; these include slow VTs, for which ATP therapy can be effective. Therefore, the detection zone setting that covers slow VTs should be considered. Aggressive ATP programming should also be considered in patients with CS.
Prognostic Value of Tp-e in CS Patients With ICD/CRT-D
Tp-e is a robust predictor of SCD in patients with congenital channelopathies and several structural heart diseases.6–8,23 More recently, Tp-e, as well as the Tp-e/QT ratio, have been reported as independent predictors for long-term prognosis in patients with CS.9 Tp-e has been thought to reflect TDR;3 however, it has been reported that Tp-e can be a marker for not only TDR, but also DOR in the whole ventricle.24 In translational experimental models, the prolonged Tp-e was correlated with increased DOR in the whole ventricle enhanced by sympathetic nerve excitation.4,5 Sympathetic activation can result in shortening of the action potential duration and effective refractory period in the whole heart. Electrophysiological heterogeneity can be increased by sympathetic innervation within the scar and border zone.25 These mechanisms may contribute to the incidence of premature ventricular contraction based on the early after-depolarization or delayed after-depolarization and to the maintenance of sustained VT. Prolonged Tp-e may represent increased arrhythmogenic heterogeneity due to scar/fibrosis formation and activated sympathoexcitation. IKs
channel blockers, such as amiodarone, may affect the Tp-e. In a previous study, we revealed that amiodarone affected the repolarization phase of the ventricle.26 Amiodarone increases the corrected QT interval, but decreases the Tp-e, which reflects the dispersion of repolarization. In the present study, the rate of amiodarone use was higher in patients with VAs, although the difference was not statistically significant. The use of amiodarone may be a confounding factor between Tp-e and ICD therapy for VA events. Therefore, we included the use of amiodarone as a covariate in the multivariate analysis. In this analysis, Tp-e remained a predictor of VA events.
Tp-e can be a predictor of the occurrence and maintenance of VT in CS patients, but this does not explain the mechanism of VT or the high efficacy of ATP on VT in these patients. Tp-e for the prediction of VA events and SCD has been reported to range from approximately 95 to 115 ms not only in patients with structural heart disease, but also in the general population. In particular, the optimal cut-off point of Tp-e may be lower in patients with some disease status. Kobayashi et al. recently reported that a higher Tp-e/QT ratio predicts long-term prognosis in CS patients, and that the mean Tp-e in patients with a low Tp-e/QT ratio was 92 ms.9 In addition, the optimal cut-off of Tp-e has been reported to be 96 ms for Brugada syndrome patients and 110 ms for heart failure patients,7 but longer than 110 ms for the general population.23 The optimal cut-off of Tp-e at 91 ms in the present study may be slightly lower but acceptable.
In the present study, histopathological diagnosis of CS was also an independent predictor of VA events in CS patients. Distinct low-voltage areas in the right ventricular septum may reflect a larger scar burden and arrhythmogenic substrates. Myocardial scar/fibrosis detected by cardiac MRI is also reported to be a predictor of prognosis in CS patients;16 however, the presence of LGE on cardiac MRI was not a predictor of the clinical endpoint in the present study. Similarly, positive findings in 18F-fluoro-2-deoxyglucose positron emission tomography or scintigraphy were not associated with VA events. Scar/fibrosis formation likely plays a key role in the occurrence and maintenance of VT; however, in the present study, only qualitative analysis was performed for LGE, and most of the patients showed LGE. Quantitative analysis of the distribution and amount of scars/fibrosis/inflammation using imaging modalities may be preferred for the prediction of long-term prognosis in CS patients. Further prospective studies are warranted to establish the prognostic value of imaging modalities in CS patients with ICD/CRT-D.
Study Limitations
This study has some limitations that should be considered. First, this was a single-center retrospective cohort study with a relatively small sample size, which may have caused selection biases. Second, cardiac MRI was not performed in all study patients and only qualitative scar assessments by LGE were available in this study. The amount and distribution of scarring/fibrosis/inflammation should be quantitatively assessed by several imaging modalities in future studies. Third, the definition of VA events was based on the detection zone setting; therefore, the detailed mechanism of VAs was not evaluated by an electrophysiological study or 3D mapping in the present study. In addition, the zone setting could interact with the frequency of ICD therapy. In this retrospective study, the tachycardia detection zone was determined by the implanting physicians. Therefore, the true VT events with a tachycardia cycle length below the detection zone could not be detected, resulting in an underestimation of the number of VT events, especially in patients with ICD/CRT-D programmed with a 2-zone setting. Fourth, the stage of CS that may be associated with the occurrence of VAs was independent of the timing of ICD/CRT-D implantation. Finally, there are slight differences in guidelines for ICD indication in CS patients between the US and Japan. In the conservative Japanese guideline, the inducibility of VT/VF is required even in patients with mildly reduced LVEF who have pacemaker indication or evident LGE; however, this indication was matched only in 1 patient. Therefore, the indication of ICD in the present study can be considered to be in accordance with American Heart Association (AHA)/American College of Cardiology (ACC)/Heart Rhythm Society guidelines.
Conclusions
In CS patients implanted with ICD/CRT-D, approximately 30% of patients can experience appropriate ICD therapy due to VAs within 2 years after device implantation. Most of the VA events were slow VT, for which ATP therapy can be highly effective. Tp-e >91 ms and a histopathological diagnosis of CS were independent predictors of VA events.
Disclosures
D.Y., M.S. belong to the same endowed department established by contributions from Medtronic Japan, Boston-Scientific, Biotronik Japan, and Abbott Medical. The remaining authors have no conflicts of interest to disclose.
IRB Information
This study was approved by the Ethics Committee of Tokyo Women’s Medical University (No. 4152-R2).
Data Availability
The deidentified participant data will not be shared.
Supplementary Files
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-23-0058
References
- 1.
Newman LS, Rose CS, Maier LA. Sarcoidosis. N Engl J Med 1997; 336: 1224–1234.
- 2.
Terasaki F, Azuma A, Anzai T, Ishizaka N, Ishida Y, Isobe M, et al. JCS 2016 guideline on diagnosis and treatment of cardiac sarcoidosis: Digest version. Circ J 2019; 83: 2329–2388.
- 3.
Antzelevitch C. Transmural dispersion of repolarization and the T wave. Cardiovasc Res 2001; 50: 426–431.
- 4.
Yagishita D, Chui RW, Yamakawa K, Rajendran PS, Ajijola OA, Nakamura K, et al. Sympathetic nerve stimulation, not circulating norepinephrine, modulates T-peak to T-end interval by increasing global dispersion of repolarization. Circ Arrhythm Electrophysiol 2015; 8: 174–185.
- 5.
Vaseghi M, Yamakawa K, Sinha A, So EL, Zhou W, Ajijola OA, et al. Modulation of regional dispersion of repolarization and T-peak to T-end interval by the right and left stellate ganglia. Am J Physiol Heart Circ Physiol 2013; 305: H1020–H1030.
- 6.
Tanabe Y, Inagaki M, Kurita T, Nagaya N, Taguchi A, Suyama K, et al. Sympathetic stimulation produces a greater increase in both transmural and spatial dispersion of repolarization in LQT1 than in LQT2 forms of congenital long QT syndrome. J Am Coll Cardiol 2001; 37: 911–919.
- 7.
Castro Hevia J, Antzelevitch C, Tornés Bárzaga F, Dorantes Sánchez M, Dorticós Balea F, Zayas Molina R, et al. Tpeak-Tend and Tpeak-Tend dispersion as risk factors for ventricular tachycardia/ventricular fibrillation in patients with the Brugada syndrome. J Am Coll Cardiol 2006; 47: 1828–1834.
- 8.
Shimizu M, Ino H, Okeie K, Yamaguchi M, Nagata M, Hayashi K, et al. T-peak to T-end interval may be a better predictor of high-risk patients with hypertrophic cardiomyopathy associated with cardiac troponin I mutation than QT dispersion. Clin Cardiol 2002; 25: 335–339.
- 9.
Kobayashi Y, Nagai T, Takenaka S, Kato Y, Komoriyama H, Nagano N, et al. Long-term prognostic significance of ventricular repolarization dispersion in patients with cardiac sarcoidosis. Am J Cardiol 2021; 152: 125–131.
- 10.
Nogami A, Kurita T, Abe H, Ando K, Ishikawa T, Imai K, et al. JCS/JHRS 2019 guideline on non-pharmacotherapy of cardiac arrhythmias. Circ J 2021; 85: 1104–1244.
- 11.
Al-Khatib SM, Stevenson WG, Ackerman MJ, Bryant WJ, Callans DJ, Curtis AB, et al. 2017 AHA/ACC/HRS guideline for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. Heart Rhythm 2018; 15: e73–e189.
- 12.
Porthan K, Viitasalo M, Toivonen L, Havulinna AS, Jula A, Tikkanen JT, et al. Predictive value of electrocardiographic T-wave morphology parameters and T-wave peak to T-wave end interval for sudden cardiac death in the general population. Circ Arrhythm Electrophysiol 2013; 6: 690–696.
- 13.
Kawakatsu N, Suzuki A, Serizawa N, Suzuki T, Ejima K, Shiga T, et al. Isolated cardiac sarcoidosis diagnosed by electroanatomic voltage mapping-guided endomyocardial biopsy combined with magnetic resonance imaging and positron emission tomography. J Cardiol Cases 2016; 14: 107–110.
- 14.
Banba K, Kusano KF, Nakamura K, Morita H, Ogawa A, Ohtsuka F, et al. Relationship between arrhythmogenesis and disease activity in cardiac sarcoidosis. Heart Rhythm 2007; 4: 1292–1299.
- 15.
Okada DR, Smith J, Derakhshan A, Gowani Z, Misra S, Berger RD, et al. Ventricular arrhythmias in cardiac sarcoidosis. Circulation 2018; 138: 1253–1264.
- 16.
Muser D, Santangeli P, Liang JJ, Castro SA, Magnani S, Hayashi T, et al. Characterization of the electroanatomic substrate in cardiac sarcoidosis: Correlation with imaging findings of scar and inflammation. JACC Clin Electrophysiol 2018; 4: 291–303.
- 17.
Naruse Y, Sekiguchi Y, Nogami A, Okada H, Yamauchi Y, Machino T, et al. Systematic treatment approach to ventricular tachycardia in cardiac sarcoidosis. Circ Arrhythm Electrophysiol 2014; 7: 407–413.
- 18.
Betensky BP, Tschabrunn CM, Zado ES, Goldberg LR, Marchlinski FE, Garcia FC, et al. Long-term follow-up of patients with cardiac sarcoidosis and implantable cardioverter-defibrillators. Heart Rhythm 2012; 9: 884–891.
- 19.
Kron J, Sauer W, Schuller J, Bogun F, Crawford T, Sarsam S, et al. Efficacy and safety of implantable cardiac defibrillators for treatment of ventricular arrhythmias in patients with cardiac sarcoidosis. Europace 2013; 15: 347–354.
- 20.
Moss AJ, Schuger C, Beck CA, Brown MW, Cannom DS, Daubert JP, et al. Reduction in inappropriate therapy and mortality through ICD programming. N Engl J Med 2012; 367: 2275–2283.
- 21.
Papageorgiou N, Providência R, Bronis K, Dechering DG, Srinivasan N, Eckardt L, et al. Catheter ablation for ventricular tachycardia in patients with cardiac sarcoidosis: A systematic review. Europace 2018; 20: 682–691.
- 22.
Brown M, Kurita T, Sterns LD, Schloss EJ, Auricchio A, Zhang Y, et al. 915: ATP efficacy on terminating ventricular tachycardia by device type, indication, and ventricular median cycle length. EP Europace 2020; 22(Issue supplement_1): euaa162.343, doi:10.1093/europace/euaa162.343.
- 23.
Panikkath R, Reinier K, Uy-Evanado A, Teodorescu C, Hattenhauer J, Mariani R, et al. Prolonged Tpeak-to-tend interval on the resting ECG is associated with increased risk of sudden cardiac death. Circ Arrhythm Electrophysiol 2011; 4: 441–447.
- 24.
Opthof T, Coronel R, Wilms-Schopman FJ, Plotnikov AN, Shlapakova IN, Danilo P Jr, et al. Dispersion of repolarization in canine ventricle and the electrocardiographic T wave: Tp-e interval does not reflect transmural dispersion. Heart Rhythm 2007; 4: 341–348.
- 25.
Ajijola OA, Yagishita D, Patel KJ, Vaseghi M, Zhou W, Yamakawa K, et al. Focal myocardial infarction induces global remodeling of cardiac sympathetic innervation: Neural remodeling in a spatial context. Am J Physiol Heart Circ Physiol 2013; 305: H1031–H1040.
- 26.
Ogiso M, Suzuki A, Shiga T, Nakai K, Shoda M, Hagiwara N. Effect of intravenous amiodarone on QT and T peak-T end dispersions in patients with nonischemic heart failure treated with cardiac resynchronization-defibrillator therapy and electrical storm. J Arrhythm 2015; 31: 1–5.
