Signal‐Averaged ECG in the Diagnostic Workup for Arrhythmogenic Cardiomyopathy: Insights From the Nordic ARVC Registry
Abstract
Background
The diagnostic role of signal‐averaged ECG (SAECG) in arrhythmogenic right ventricular cardiomyopathy (ARVC) has lately been questioned. We assessed the value of SAECG‐derived late ventricular potentials (LP) in ARVC diagnosis and its association with disease manifestations.
Methods and Results
Patients with definite ARVC diagnosis or genotype‐positive family members who underwent SAECG were included in register‐based observational study (n=357, mean age 41 years, 47% female, 43% probands). LP and terminal activation duration (TAD) were defined by Task Force Criteria 2010. We assessed the association of TAD and LP with structural RV abnormalities and ventricular tachycardia (VT), defined as sustained VT, appropriate implantable cardioverter‐defibrillator shock, aborted cardiac arrest, or sudden cardiac death, at diagnosis. LP were documented in 210 patients (59%) and abnormal TAD in 66 patients (18%). Each of the SAECG parameters was significantly associated with definite ARVC diagnosis in receiver‐operator characteristics curve analysis with area under the curve between 0.67 and 0.74. Exclusion of SAECG from diagnostic workup led to reclassification of 37 patients (16%) from definite to borderline ARVC (13 probands, 9 of whom had prevalent VT). Ninety patients (25%) had history of VT. LP, but not TAD, were associated with VT (adjusted odds ratio [ORadj], 2.42 [95%CI, 1.07–5.48]). LP had lower specificity (72% versus 97%) but higher sensitivity (71% versus 25%) for association with RV structural abnormalities than TAD.
Conclusions
In the Nordic ARVC cohort SAECG‐derived LP are associated with VT and structural RV abnormalities and were critical for ascertainment of ARVC diagnosis in 16% of patients with narrow QRS complexes, including 8% of all probands.
Nonstandard Abbreviations and Acronyms
- ARVC
- arrhythmogenic right ventricular cardiomyopathy
- LGE
- late gadolinium enhancement
- LP
- ventricular late potentials
- SAECG
- signal‐averaged ECG
- TAD
- terminal activation duration
- TFC
- Task Force Criteria
Clinical Perspective
What Is New?
•
Exclusion of signal‐averaged ECG ‐derived parameters from arrhythmogenic right ventricular cardiomyopathy task force criteria would lead to reclassification of 16% of patients, including 8% of probands with definite arrhythmogenic right ventricular cardiomyopathy category and narrow QRS complex, thus potentially delaying diagnosis and initiation of screening of family members.
What Are the Clinical Implications?
•
Signal‐averaged ECG remains useful diagnostic tool for the patients with narrow QRS complex and suspected arrhythmogenic right ventricular cardiomyopathy, which cannot be fully replaced by terminal activation duration demonstrating much less sensitivity in this patient subgroup.
Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a genetically determined heart muscle disorder with an autosomal dominant inheritance.1 The clinical presentation of the disease varies greatly, and even early‐stage patients are at increased risk of sudden cardiac death.2 Diagnostic approaches in ARVC are based on Task Force Criteria revised in 2010 (TFC2010).
Morphological changes in ARVC include fibrofatty infiltration and subsequent destruction of normal myocytes.3 These factors create a substrate for both ventricular dysfunction and electrical instability leading to life‐threatening ventricular arrhythmia.3, 4 Ventricular late potentials (LP) are low‐amplitude signals occurring from areas of delayed conduction in the ventricles and can be registered using signal‐averaged ECG (SAECG).5 The presence of LP is associated with both ARVC diagnosis and the risk of ventricular arrhythmias.6, 7 Upon revision of the TFC in 2010, LP were considered as a moderately sensitive and highly specific minor criterion for ARVC diagnosis in patients with QRS complexes not exceeding 110 ms.2, 6
However, in recent years, the value of SAECG in diagnostic workup has been debated, and the diagnostic value of minor depolarization criteria was suggested to be primarily related to terminal activation duration (TAD) in lead V1 measurable using standard 12‐lead ECG.8 In part, the controversies may be related to the limited availability of the SAECG technique, which has likely contributed to its less common use in the diagnostic workup.
In the Scandinavian countries SAECG is widely used, enabling the evaluation of this diagnostic modality within a relatively large cohort of patients with ARVC. Using data accumulated in the Nordic ARVC Registry, we sought to assess the impact of SAECG‐derived LP and TAD on diagnostic workup of ARVC and to explore the relation of these parameters to structural and functional changes of right ventricle and ventricular arrhythmias.
METHODS
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Study Population
The Nordic ARVC Registry is an ongoing prospective registry initiated in June 2010 by tertiary referral centers in Denmark, Finland, Norway, and Sweden. The registry included patients diagnosed with ARVC and their family members, carrying disease‐causing genetic variants, who were under follow‐up care at the time the registry was launched, and then prospectively recruited newly diagnosed patients and genotype‐positive family members.9, 10 Data collection for this study ended in September 2022. A proband was defined as the first affected individual in a family from whom cascade family screening was initiated. The registry captured baseline clinical characteristics, data regarding ECG, SAECG, imaging, genetic variants, and other relevant information specific to ARVC diagnostic criteria as proposed in TFC2010.2 Information on genetic variants in the registry included ARVC‐associated variants of PKP2, DSC2, DSG2, DSP, JUP, and TMEM43.
For the current study, all patients with either definite ARVC diagnosis by TFC2010 or carriers of pathogenic genetic variants regardless of ARVC phenotype who underwent SAECG exam as a part of diagnostic workup were included. Patients with a QRS of 110 ms or longer were excluded, as being ineligible for LP assessment per TFC2010.2 The patient selection process is schematically shown in Figure 1.
Figure 1. Study design and structure.
ARVC indicates arrhythmogenic right ventricular cardiomyopathy; CMR, cardiac magnetic resonance; LP, ventricular late potentials; RV, right ventricular; SAECG, signal averaged ECG; TFC, Task Force Criteria; and VT, ventricular tachycardia. See the main text for the information of TFC and TFC‐LP scores counting.
Control Group
The control group was comprised of patients referred to SAECG examination between May 2000 and October 2023 due to suspected ARVC, who did not have the family history of ARVC and in whom the definite ARVC diagnosis was not confirmed by further clinical investigation and the initial findings were explained by other reasons. Similarly to the study group, QRS duration shorter than 110 ms was considered the eligibility criterion for interpretation of SAECG results.
ECG and SAECG Parameters – Description and Assessment
We collected data of 3 SAECG‐derived parameters: filtered QRS duration, low amplitude signals, and root mean square amplitude of the last 40 ms. Each of these parameters was extracted from the registry and analyzed both as continuous and binary (normal/abnormal) variables. For the binary assignment we used the cutoff values provided in TFC2010: filtered QRS duration ≥114 ms; low amplitude signals ≥38 ms; root mean square amplitude of the last 40 ms ≤20 mv. Ventricular LP were considered to be present if at least 1 of these parameters was abnormal.
TAD was measured in lead V1 and considered abnormal at ≥55 ms.
Imaging Data Collection and Interpretation
The association between depolarization abnormality indices (SAECG parameters and TAD) and structural and functional right ventricular (RV) abnormalities was studied in a subgroup of patients who underwent cardiac magnetic resonance imaging (CMR) (n=267). RV end‐diastolic volume index, RV ejection fraction (RVEF), and the presence of late gadolinium enhancement (LGE) were assessed. The cutoff values were chosen based on the minor imaging criteria defined by TCF2010 (RV end‐diastolic volume index ≥100 mL/m2 for men and 90 mL/m2 for women, and RVEF <45%).2
Left ventricular (LV) dysfunction was considered to be present if the LV ejection fraction (LVEF) was <50% as assessed by either CMR or echocardiography. When several imaging exam results were available for the same patient, we chose the imaging exam closest in time to the SAECG recording, with preference to CMR when available.
Definition of Life‐Threatening Ventricular Arrhythmias
Ventricular arrhythmias in the registry are reported if they were verified on standard ECG or captured by implantable cardioverter‐defibrillator. Ventricular tachycardia (VT) was defined as either sustained VT (lasting >30 s), appropriate implantable cardioverter‐defibrillator therapy (antitachycardia pacing or shock), aborted cardiac arrest, or sudden cardiac death.
TFC Diagnostic Score and Modified TFC Scores: Description and Implementation
ARVC phenotype expression was presented as the TFC2010 diagnostic score, calculated by assigning 2 points for a major diagnostic criterion and 1 point for a minor diagnostic criterion from different diagnostic categories. The modified scoring systems were introduced to account for the number of patients in whom ascertainment of ARVC diagnosis depended on the presence of LP or TAD:
1.
TFC‐LP: TFC2010 score in which LP was not included,
2.
TFC‐TAD: TFC2010 score in which TAD was not included.
Both modified scoring systems otherwise used the same definitions as presented for TFC 2010.
To describe the profile of patients who would be reclassified from definite to borderline diagnostic category if SAECG results not considered, the subgroup of patients with definite ARVC who had TFC score 4 and TFC‐LP score 3 was defined as “definite LP‐dependent.” This group was compared with patients who had both TFC and TFC‐LP score of 4, that is, had definite diagnostic category regardless of SAECG findings (“definite LP‐independent”), as well as patients with TFC and TFC‐LP scores of 3, that is, borderline ARVC regardless of SAECG findings (“borderline LP‐independent”). See Figure 1, Analysis #3, for the scheme of group formation.
To assess whether the diagnostic value of SAECG is limited to the subgroup of patients who did not undergo the complete set of diagnostic techniques we compared the proportion of definite LP‐dependent patients between the complete workup cohort and all other study participants (incomplete workup cohort). Complete workup cohort was defined as patients who underwent all standard diagnostic modalities apart from SAECG: ECG, echocardiography, 24‐hour ambulatory ECG monitoring, CMR imaging, and genetic evaluation.
Statistical Analysis
Statistical analysis was performed using IBM SPSS Statistics for Windows 29.0.1.0 (IBM Corporation, Armonk, NY, USA). Data were presented as mean±SD for continuous variables or count (percentage) for categorical variables. Paired t tests were used for comparison of continuous variables, and categorical variables were evaluated with the χ2 test of Fisher's exact test, as appropriate. Missing data were removed from the analysis and data availability reported.
We assessed the sensitivity and specificity of the 3 SAECG parameters and TAD for association with abnormal RVEF, LGE, or RV end‐diastolic volume index.
C‐statistics and receiver‐operator characteristics curve analysis were used to assess the association of each of the SAECG parameters as continuous variables, with CMR abnormalities (RV enlargement, RVEF decrease, LGE, or any abnormal CMR finding) and TFC‐LP score ≥4 (definite ARVC by TFC2010 irrespective of SAECG findings).
To assess the association between TAD and LP and the prevalence of VT at the time of ARVC diagnosis, multivariate logistic regression analysis with adjustment for age, sex, proband status, definite ARVC status, reduced LVEF (<50%) and major imaging criteria was used. Statistical significance was considered as P<0.05.
Ethical Approval
Informed consent was obtained from all subjects in the study. The study complies with the Declaration of Helsinki. The registry was approved by regional ethics committees in all participating countries (for Denmark the Ethics Committee, the Capital Region of Denmark (H‐19043226); for Finland Ethics Committee Dnro 307/13/03/01/11 and Dnro HUS/729/2019; for Norway Ethics Committee 2009/2299; for Sweden Swedish Ethical Review Authority (2010/568 and 2017/485)).
RESULTS
Patient Population
The registry included 575 patients either with definite ARVC by TFC2010 or genotype‐positive family members. Native QRS duration ≥110 ms on the standard ECG was observed in 48 patients who were excluded as not eligible for SAECG. Of the remaining 527 patients, SAECG was performed in 357 patients, who comprised the cohort with ARVC (238 with definite, 51 with borderline, and 68 with possible ARVC diagnosis, Figure 1). The cohort with ARVC was compared with 125 control subjects. Clinical and genetic characteristics and SAECG parameters of the study populations are presented in Table 1. Control subjects had similar sex distribution though they were slightly younger than patients with ARVC. The presence of TFC criteria in the group with ARVC is shown in Table 2.
| All patients with ARVC, n=357 | Nondefinite ARVC, n=119 | Definite ARVC, n=238 | Controls, n=125 | |
|---|---|---|---|---|
| Age at diagnosis or evaluation, y | 41±16§ | 37±16† | 42±9# | 37±15 |
| Female sex, n (%) | 166 (46) | 72 (61)‡ | 95 (40) | 63 (50) |
| Mutation positive, n (%) | 272 (76) | 119 (100)‡ | 154 (64) | NA |
| PKP2 positive, n (%) | 176 (49) | 81 (68)‡ | 95 (40) | NA |
| DSC2 positive, n (%) | 11 (3) | 7 (6)† | 4 (2) | NA |
| DSG2 positive, n (%) | 49 (14) | 16 (13) | 33 (14) | NA |
| DSP positive, n (%) | 34 (10) | 12 (10) | 22 (9) | NA |
| JUP positive, n (%) | 1 (0.3) | 0 (0) | 1 (0.4) | NA |
| TMEM43 positive, n (%) | 8 (2) | 5 (4) | 3 (1) | NA |
| Probands, n (%) | 155 (43)¶ | 7 (6)‡ | 148 (62)* | 125 (100) |
| Diabetes, n (%) | 7 (2) | 4 (3) | 3 (1) | NA |
| Hypertension, n (%) | 30 (8) | 6 (5) | 24 (10) | NA |
| Ischemic heart disease, n (%) | 11 (3) | 0 (0)† | 11 (5) | NA |
| Implantable cardioverter‐defibrillator, n (%) [primary/secondary prevention] | 170 (48) [67/103] | 7 (6)‡ [5/2] | 163 (58) [62/101] | NA |
| Left ventricular ejection fraction <50%, n (%) | 31 (9) | 0 (0)‡ | 31 (13) | NA |
| SAECG parameters | ||||
| fQRSd, ms±SD | 114±21¶ | 104±14‡ | 119±22* | 104±12 |
| LAS40, ms±SD | 41±19¶ | 32±12‡ | 45±21* | 34±11 |
| RMS40, μV±SD | 29±21§ | 40±21‡ | 23±19* | 36±31 |
| fQRSd abnormal, n (%) [missing data, n] | 159 (45)¶ [1] | 21 (18)‡ [0] | 138 (58)* [1] | 17 (14) [0] |
| LAS40 abnormal, n (%) [missing data, n] | 162 (45)¶ [1] | 21 (18)‡ [1] | 141 (59)* [0] | 34 (27) [0] |
| RMS40 abnormal, n (%) [missing data, n] | 159 (45)¶ [1] | 21 (18)‡ [0] | 138 (58)* [1] | 30 (24) [0] |
| Ventricular late potentials, n (%) | 210 (59)¶ | 35 (30)‡ | 175 (74)* | 39 (31) |
| TAD abnormal, n (%) | 66 (18) | 1 (1)‡ | 65 (27) | NA |
ARVC indicates arrhythmogenic right ventricular cardiomyopathy; fQRSd, filtered QRS duration; LAS40, low amplitude signals; NA, data not available; RMS40, root mean square amplitude of the last 40 ms; SAECG, signal‐averaged ECG; and TAD, terminal activation duration.
P values for the difference between nondefinite and definite ARVC: †P<0.05; ‡P<0.001; P values for the difference between all ARVC and controls: §P<0.05; ¶P<0.001; P values for the difference between definite ARVC and controls: #P<0.05; *P<0.001.
| All patients with ARVC, n=357 | Nondefinite ARVC, n=119 | Definite ARVC, n=238 | |
|---|---|---|---|
| Family history major, n (%) | 293 (82) | 119 (100) | 174 (73)‡ |
| Family history minor, n (%) | 4 (1) | 0 (0) | 4 (2) |
| RV structural abnormalities major, n (%) | 148 (41) | 0 (0) | 148 (62)‡ |
| RV structural abnormalities minor, n (%) | 14 (4) | 0 (0) | 14 (6)† |
| Repolarization abnormalities major, n (%) | 112 (31) | 0 (0) | 112 (47)‡ |
| Repolarization abnormalities minor, n (%) | 47 (13) | 4 (3.3) | 43 (18)‡ |
| Depolarization abnormalities major, n (%) | 20 (6) | 0 (0) | 20 (8)‡ |
| Depolarization abnormalities minor, n (%) | 202 (57) | 35 (29) | 167 (70)‡ |
| Arrythmia major, n (%) | 55 (15) | 0 (0) | 55 (23)‡ |
| Arrythmia minor, n (%) | 156 (44) | 12 (10) | 144 (61)‡ |
P values for the difference between definite and nondefinite ARVC: †P<0.01; ‡P<0.001. ARVC indicates arrhythmogenic right ventricular cardiomyopathy; RV, right ventricular; and TFC, Task Force Criteria.
CMR was performed in the vast majority of the ARVC and control groups and demonstrated RV enlargement in nearly half of subjects in each group, though expectedly lower prevalence of LGE‐positive tests and reduced RVEF in the control group (Table 3).
| All patients with ARVC, n=357 | Controls, n=125 | P value | |
|---|---|---|---|
| CMR performed, n (%) | 267 (75) | 104 (83) | 0.06 |
| RV end diastolic volume index enlarged*, n (%) [missing data, n] | 112 (54) [60] | 42 (46) [13] | 0.2 |
| RV ejection fraction <45%, n (%) [missing data, n] | 86 (39) [48] | 8 (9) [10] | <0.001 |
| Late gadolinium enhancement detected, n (%) [missing data, n] | 115 (43) [0] | 12 (12) [1] | <0.001 |
ARVC indicates arrhythmogenic right ventricular cardiomyopathy; and CMR, cardiac magnetic resonance.
*
≥90 mL/m2 for women; ≥100 mL/m2 for men.
SAECG and CMR
All 3 SAECG parameters (filtered QRS duration, low amplitude signals and root mean square amplitude of the last 40 ms) were significantly associated with RV structural and functional abnormalities with similar C‐statistics values (Table 4). Receiver‐operator characteristics curves are presented in Figure 2.
| RVEDVI >100/90 mL/m2 (men/women), n=112 (54%) | RVEF <45%, n=86 (39%) | LGE presence, n=115 (43%) | Any RV abnormality, n=139 (71%) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sensitivity | Specificity | C‐statistics | Sensitivity | Specificity | C‐statistics | Sensitivity | Specificity | C‐statistics | Sensitivity | Specificity | C‐statistics | |
| TAD >55 ms, n=53 (20%) | 27 | 87 | 30 | 89 | 26 | 84 | 25 | 97 | ||||
| Ventricular late potentials, n=161 (60%) | 75 | 60 | 77 | 54 | 74 | 50 | 71 | 72 | ||||
| Filtered QRS duration >114 ms, n=123 (46%) | 63 | 76 | 0.70 (0.63–0.77) | 63 | 71 | 0.72 (0.65–0.80) | 56 | 62 | 0.59 (0.52–0.66) | 55 | 84 | 0.69 (0.61–0.76) |
| Low amplitude signals >38 ms, n=126 (47%) | 65 | 72 | 0.67 (0.59–0.74) | 66 | 67 | 0.70 (0.62–0.77) | 56 | 59 | 0.61 (0.54–0.68) | 59 | 83 | 0.68 (0.60–0.76) |
| Root mean square amplitude of the last 40 ms <20 μV, n=124 (47%) | 61 | 70 | 0.69 (0.61–0.76) | 66 | 68 | 0.72 (0.65–0.79) | 62 | 65 | 0.65 (0.58–0.71) | 58 | 84 | 0.74 (0.67–0.82) |
ARVC indicates arrhythmogenic right ventricular cardiomyopathy; LGE, late gadolinium enhancement; RV, right ventricle; RVEDVI, right ventricular end diastolic volume index; RVEF, right ventricular ejection fraction; SAECG, signal‐averaged ECG; and TAD, terminal activation duration. See explanation in main text.
Figure 2. ROC curves for SAECG parameters as predictors of RV structural abnormalities in patients with ARVC.
See Table 2 for C‐statistics values. ARVC indicates arrhythmogenic right ventricular cardiomyopathy; fQRSd, filtered QRS duration; LAS40, low amplitude signals; LGE, late gadolinium enhancement; ROC, receiver‐operator characteristics; RV, right ventricle; RVEDVI, right ventricular end‐diastolic volume index; RVEF, right ventricular ejection fraction; and SAECG, signal‐averaged electrocardiography.
TAD demonstrated high specificity, but low sensitivity, for association with RV structural abnormalities, whereas LP had lower specificity but much higher sensitivity (Table 4). Patients with both positive LP and abnormal TAD did not significantly differ from patients with positive LP alone with regard to reduced RVEF (66 versus 47%, P=0.07), RV enlargement (69 versus 69%, P=1.0) or LGE by CMR (60 versus 50%, P=0.3).
Modified TFC Scores Performance
Per conventional TFC 2010, 238 patients were classified as definite ARVC (4 or more points), 51 patients were classified as borderline (3 points), and 68 patients were classified as possible ARVC. Exclusion of LP (TFC‐LP) led to reclassifying 37 patients (16%) from definite to the borderline ARVC category. Exclusion of TAD (TFC‐TAD) led to reclassification of 2 patients (1%) from definite to borderline ARVC category (Figure 3).
Figure 3. Change in the diagnostic category from the conventional Task Force Criteria 2010 to the modified diagnostic scores that do not include either terminal activation delay or SAECG‐derived ventricular late potentials.
Only 1 patient with definite ARVC was to be reclassified if TAD was not counted, compared with 37 patients who would no longer fit into the definite diagnostic category, if the SAECG results were not considered. LP indicates ventricular late potentials; SAECG, signal‐averaged electrocardiography; TAD, terminal activationd delay; and TFC, Task Force Criteria.
Receiver‐operator characteristics curve analysis for the SAECG parameters value for discrimination of patients with definite ARVC diagnosis from controls showed their significant association with moderate power and area under the curve between 0.67 and 0.74. AUC for each of the SAECG parameters did not differ significantly between different genotype groups: genotype negative patients, PKP2, DSG2, and DSP carriers. DSC2, JUP, and TMEM43 carriers were not analyzed separately due to the low prevalence of these genotypes in the study cohort (Figure 4).
Figure 4. Genotype‐specific ROC curve analysis of the association between SAECG parameters and ascertainment of clinical ARVC diagnosis by TFC2010 with SAECG results not accounted for in the diagnostic score (TFC‐LP, see the Methods section for more explanation).
ARVC indicates arrhythmogenic right ventricular cardiomyopathy; fQRSd, filtered QRS duration; LAS40, low amplitude signals; LP, ventricular late potentials; ROC, receiver‐operator characteristics; SAECG, signal‐averaged electrocardiography; TFC, Task Force Criteria.
Patients with Definite ARVC Diagnosis Dependent on the SAECG Findings
A definite ARVC diagnosis appeared to be dependent on the SAECG findings in 37 patients. This LP‐dependent group included 13 probands (35%), of whom 9 (69%) had prevalent VT and 7 (54%) survived cardiac arrest. In 4 of the LP‐dependent probands, genetic evaluation initiated upon ascertainment of clinical ARVC diagnosis resulted in identification of disease‐causing genetic variants suitable for family screening, which, in turn, led to confirming the genotype‐positive status in 3 additional family members. One other family member was diagnosed with ARVC during family screening initiated upon ascertainment of the clinical ARVC diagnosis in a genotype‐negative proband from the LP‐dependent group. Eleven of the 37 LP‐dependent patients (30%) had prevalent VT at the time of ARVC diagnosis. The representation of the TFC diagnostic criteria among the patients of LP‐dependent group is shown in Table 5.
| Crieria group | No. (%) | |
|---|---|---|
| Imaging | Major | 7 (19) |
| Minor | 2 (5) | |
| Repolarization | Major | 0 (0) |
| Minor | 11 (30) | |
| Depolarization | Major | 0 (0) |
| Minor | 37 (100) | |
| Family | Major | 29 (78) |
| Minor | 0 (0) | |
| Arrhythmia | Major | 1 (3) |
| Minor | 24 (65) | |
LP indicates ventricular late potentials.
Thirty‐three patients had definite ARVC with the minimally required 4‐point score independent of LP, that is, SAECG findings did not affect these patients' diagnostic category (LP‐independent definite). This group included 19 probands (58%), of whom 11 had prevalent VT (58%), and 3 survived cardiac arrest (16%). No significant difference was observed between LP‐dependent and LP‐independent definite patients with ARVC regarding proband status (35% versus 58%, respectively, P=0.09), the presence of any imaging criteria (24% versus 39%, P=0.2), and VT history (27% versus 33%, P=0.6).
The borderline LP‐independent group included 17 patients, 5 of whom were probands (29%). Comparing this group to the LP‐dependent group revealed higher VT prevalence (27% versus 0%, P=0.022) and the presence of imaging TFC criteria (24% versus 0%, P=0.044) in the LP‐dependent patients with definite ARVC.
In a subanalysis of the prevalence of LP‐dependent patients with definite ARVC with regard to the completeness of the diagnostic workup, ECG, echocardiography, and genetic evaluation were reported in all patients, whereas CMR was not included in the diagnostic workup in 90 patients and 24‐hour ambulatory ECG was not performed at the time of diagnosis in 65 patients. Therefore, the complete workup group comprised 216 patients who were compared with the remaining 141 patients (incomplete workup group). Complete and incomplete workup groups had similar proportions of patients with definite ARVC diagnosis (148/216 [69%] versus 90/141 [64%] respectively, P=0.4). Similarly, there was no statistically significant difference between the complete and incomplete workup groups in terms of the proportion of LP‐dependent patients with definite ARVC (21/148 [14%] versus 16/90 [18%], P=0.5).
SAECG and VT
By the time of diagnosis, VT was reported in 90 patients (25%). LP were associated with VT in the adjusted logistic regression analysis (adjusted odds ratio [ORadj]=2.42 [95% CI, 1.07–5.48]; sensitivity 86%, specificity 50%), while TAD was not associated with VT (ORadj, 0.88 [95% CI, 0.45–1.72]; sensitivity 33%, specificity 87%).
In a subgroup of LP‐positive patients, the presence of abnormal TAD was associated with prevalence of VT (50% versus 32%, P=0.022).
DISCUSSION
In a large cohort of Scandinavian patients with ARVC who had SAECG performed as part of their diagnostic workup, we showed a significant association of SAECG parameters with structural and functional RV abnormalities assessed by CMR. Comparison of LP and TAD showed that LP had higher sensitivity, but lower specificity regarding CMR‐derived RV structural abnormalities. We found that LP, but not TAD, were associated with prevalent VT regardless of age, sex, and proband status. Exclusion of SAECG from the diagnostic workup led to reclassification of 16% of patients with definite ARVC diagnosis into the borderline category, of whom 35% were probands and 29% had prevalent VT. These patients differed significantly from those classified as borderline independently of LP with regard to VT prevalence and imaging characteristics. We did not find any association between the prevalence of LP‐dependent definite ARVC diagnosis and the completeness of diagnostic workup, which suggests that SAECG adds diagnostic information rather than reflecting the abnormalities that could be revealed by other diagnostic modalities.
The SAECG technique was first introduced in 1977 and was subsequently extensively refined.11 The initial enthusiasm regarding this tool later weaned, and many centers do not currently include it in the ARVC diagnostic workup. Two recent studies analyzing the value of SAECG for diagnosing ARVC cast doubt on its actual usefulness.12, 13
Our data support the previously reported association between SAECG parameters and structural RV abnormalities.6, 13, 14 Each of the SAECG indices and LP showed similar C‐statistics values and moderate sensitivity and specificity for association with CMR‐assessed RV abnormalities. As expected, for each SAECG measurement, specificity prevailed over sensitivity, whereas the use of the cumulative index (LP) led to an increase in sensitivity along with a decrease in specificity. Notably, TAD demonstrated very low sensitivity, but good specificity, for detecting structural RV abnormalities. This corresponds with the results of previous work showing that absence of standard ECG‐derived ARVC criteria does not rule out the presence of structural abnormalities.15 Thus, our findings do not support the use of TAD as an easily obtainable equivalent of LP, although both indices likely reflect the same electrophysiological substrate.
In order to study LP's contribution to ARVC diagnosis, we assessed the number of patients who would no longer fit into the definite diagnostic category if LP were excluded from the diagnostic score. The modified diagnostic TFC‐LP score that we used in our work was designed in a manner similar to that used by Pearman et al.13 In the Canadian cohort, which enrolled 90 patients with definite ARVC compared with 238 in our study, the diagnosis was dependent of LP in 10 patients, which corresponds to 10% of patients with definite ARVC compared with 16% in our study. Our findings indicate that in a significant proportion of patients with ARVC their formal classification into the definite or the borderline diagnostic category depends on the SAECG results, thus influencing further diagnostic and management strategy. We also found that the phenotype of LP‐dependent patients similar to that of patients with LP‐independent definite ARVC and more severe than the phenotype of patients with borderline ARVC. These findings indicate that the impact of SAECG on the ARVC diagnostic workup is still noticeable, and excluding SAECG from TFC would potentially delay ARVC diagnosis and initiation of family screening.
Pearman et al.13 investigated a Canadian cohort of patients with ARVC and showed that SAECG was more likely to be abnormal in patients with ARVC with prior ventricular arrhythmia than in patients without, consistent with previous reports.16 We further confirmed and elaborated this association in our cohort, showing it to be independent of age, sex, proband status, and presence of imaging criteria.
It is important to note that our findings are only valid for patients with a narrow QRS on the standard ECG, which is typically observed in patients with less prominent structural RV abnormalities.17 QRS prolongation in patients undergoing diagnostic workup for ARVC makes SAECG irrelevant, which explains the relatively low prevalence of TAD in our cohort and the less severe phenotype compared with the patients who were excluded from this study (Table S1). The prevalence of abnormal TAD in the entire cohort of 575 patients was 22% and reached 31% in patients with definite ARVC, which is in the range reported by others.18, 19
Limitations
This is a multinational register‐based observational study in which we cannot completely exclude heterogeneity among participating centers regarding locally accepted diagnostic approach and equipment used for SAECG registration. The TAD duration was assessed as a part of clinically driven evaluation using TFC2010 criteria. Data regarding SAECG parameters and TAD collected during diagnostic workup were retrieved from the registry as the components of TFC2010 diagnostic score assessed by local principal investigators without reassessment of ECG indices by the core laboratory.
CONCLUSIONS
In the largest to date multinational cohort with ARVC examined with SAECG, LP were significantly associated with definite ARVC diagnosis and key clinical and structural ARVC characteristics. LP were critical for ascertaining an ARVC diagnosis in 16% of patients with narrow QRS with definite ARVC, including 8% probands, in whom they were required for triggering family screening and genetic evaluation. Although the diagnostic impact of LP varies depending on SAECG availability, our data support their use as a sensitive noninvasive diagnostic tool. TAD, an ECG‐based minor depolarization criterion, had lower sensitivity as an indicator of RV structural abnormalities compared with LP in patients with narrow QRS complexes undergoing ARVC diagnostic workup.
Sources of Funding
This study was supported by the Swedish Heart Lung Foundation (#20200674 to Pyotr G. Platonov and Aleksei A. Savelev), by grants from the Swedish state under the agreement between the Swedish government and the county councils, the ALF agreement, and by donation funds at Skåne University Hospital. Eivind W. Aabel received Research Council of Norway grant (#288438). Alex Hørby Christensen is supported by The Independent Research Fund Denmark (Grant 0134‐00363B) and The Novo Nordisk Foundation, Denmark (NNF20OC0065799). Henrik K. Jensen is supported by grant from Novo Nordisk Foundation, Denmark (NNF18OC0031258). Tiina Heliö is supported by grants from Finnish Foundation for Cardiovascular Research, Aarne Koskelo Foundation, Governmental Subsidy (EVO grants). Kristina H. Haugaa is supported by the Norwegian Research council, ProCardio #309762, GENE POSITIVE #288438, and EMPATHY #298736.
Footnotes
This article was sent to Suha Bachir, MD, MS, Assistant Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at Supplemental Material
For Sources of Funding and Disclosures, see page 10.
Copenhagen University Hospital, Helsinki University Hospital, and Aarhus University Hospital are European Reference Network for Rare and Low Prevalence Complex Diseases of the Heart (ERN GUARD‐Heart) affiliated hospitals.
Supplemental Material
Table S1
- Download
- 139.33 KB
References
1.
Haugaa KH, Haland TF, Leren IS, Saberniak J, Edvardsen T. Arrhythmogenic right ventricular cardiomyopathy, clinical manifestations, and diagnosis. Europace. 2016;18:965–972.
2.
Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA, Calkins H, Corrado D, Cox MG, Daubert JP, et al. Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the task force criteria. Eur Heart J. 2010;31:806–814.
3.
Choudhary N, Tompkins C, Polonsky B, McNitt S, Calkins H, Mark Estes NA, Krahn AD, Link MS, Marcus FI, Towbin JA, et al. Clinical presentation and outcomes by sex in arrhythmogenic right ventricular cardiomyopathy: findings from the north American ARVC registry. J Cardiovasc Electrophysiol. 2016;27:555–562.
4.
Basso C, Corrado D, Marcus FI, Nava A, Thiene G. Arrhythmogenic right ventricular cardiomyopathy. Lancet. 2009;373:1289–1300.
5.
Gatzoulis KA, Arsenos P, Trachanas K, Dilaveris P, Antoniou C, Tsiachris D, Sideris S, Kolettis TM, Tousoulis D. Signal‐averaged electrocardiography: past, present, and future. J Arrhythm. 2018;34:222–229.
6.
Kamath GS, Zareba W, Delaney J, Koneru JN, McKenna W, Gear K, Polonsky S, Sherrill D, Bluemke D, Marcus F, et al. Value of the signal‐averaged electrocardiogram in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Heart Rhythm. 2011;8:256–262.
7.
Nasir K, Rutberg J, Tandri H, Berger R, Tomaselli G, Calkins H. Utility of SAECG in arrhythmogenic right ventricle dysplasia. Ann Noninvasive Electrocardiol. 2003;8:112–120.
8.
Bourfiss M, Hauer R. Current challenges in the diagnosis of arrhythmogenic cardiomyopathies. In: Brunckhorst C, Saguner AM, Duru F, eds Current Concepts in Arrhythmogenic Cardiomyopathy. 2nd ed. Cardiotext Publishing; 2021:73–87.
9.
Baturova MA, Haugaa KH, Jensen HK, Svensson A, Gilljam T, Bundgaard H, Madsen T, Hansen J, Chivulescu M, Christiansen MK, et al. Atrial fibrillation as a clinical characteristic of arrhythmogenic right ventricular cardiomyopathy: experience from the Nordic ARVC registry. Int J Cardiol. 2020;298:39–43.
10.
Borgquist R, Haugaa KH, Gilljam T, Bundgaard H, Hansen J, Eschen O, Jensen HK, Holst AG, Edvardsen T, Svendsen JH, et al. The diagnostic performance of imaging methods in ARVC using the 2010 task force criteria. Eur Heart J Cardiovasc Imaging. 2014;15:1219–1225.
11.
Zhang L, Liu L, Kowey PR, Fontaine GH. The electrocardiographic manifestations of arrhythmogenic right ventricular dysplasia. Curr Cardiol Rev. 2014;10:237–245.
12.
Bosman LP, Cadrin‐Tourigny J, Bourfiss M, Aliyari Ghasabeh M, Sharma A, Tichnell C, Roudijk RW, Murray B, Tandri H, Khairy P, et al. Diagnosing arrhythmogenic right ventricular cardiomyopathy by 2010 task force criteria: clinical performance and simplified practical implementation. Europace. 2020;22:787–796.
13.
Pearman CM, Lee D, Davies B, Khan H, Tadros R, Cadrin‐Tourigny J, Roberts JD, Sanatani S, Simpson C, Angaran P, et al. Incremental value of the signal‐averaged ECG for diagnosing arrhythmogenic cardiomyopathy. Heart Rhythm. 2023;20:224–230.
14.
Jongman JK, Zaidi A, Muggenthaler M, Sharma S. Relationship between echocardiographic right‐ventricular dimensions and signal‐averaged electrocardiogram abnormalities in endurance athletes. Europace. 2015;17:1441–1448.
15.
Stokke MK, Castrini AI, Aneq MÅ, Jensen HK, Madsen T, Hansen J, Bundgaard H, Gilljam T, Platonov PG, Svendsen JH, et al. Absence of ECG task force criteria does not rule out structural changes in genotype positive ARVC patients. Int J Cardiol. 2020;317:152–158.
16.
Turrini P, Angelini A, Thiene G, Buja G, Daliento L, Rizzoli G, Nava A. Late potentials and ventricular arrhythmias in arrhythmogenic right ventricular cardiomyopathy. Am J Cardiol. 1999;83:1214–1219.
17.
Laredo M, Tovia‐Brodie O, Milman A, Michowitz Y, Roudijk RW, Peretto G, Badenco N, Te Riele ASJM, Sala S, Duthoit G, et al. Electrocardiographic findings in patients with arrhythmogenic cardiomyopathy and right bundle branch block ventricular tachycardia. Europace. 2023;25:1025–1034.
18.
Bae MH, Kim JH, Jang SY, Park SH, Lee JH, Yang DH, Park HS, Cho Y, Chae SC. Changes in follow‐up ECG and signal‐averaged ECG in patients with arrhythmogenic right ventricular cardiomyopathy. Pacing Clin Electrophysiol. 2014;37:430–438.
19.
Cox MG, van der Zwaag PA, van der Werf C, van der Smagt JJ, Noorman M, Bhuiyan ZA, Wiesfeld AC, Volders PG, van Langen IM, Atsma DE, et al. Arrhythmogenic right ventricular dysplasia/cardiomyopathy: pathogenic desmosome mutations in index‐patients predict outcome of family screening: Dutch arrhythmogenic right ventricular dysplasia/cardiomyopathy genotype‐phenotype follow‐up study. Circulation. 2011;123:2690–2700.
Information & Authors
Information
Published In
Copyright
© 2025 The Author(s). Published on behalf of the American Heart Association, Inc., by Wiley. This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made.
Versions
You are viewing the most recent version of this article.
History
Received: 8 July 2024
Accepted: 27 January 2025
Published online: 7 March 2025
Published in print: 18 March 2025
Keywords
Subjects
Authors
Disclosures
Jesper H. Svendsen reports to be a member a Medtronic advisory board; Tiina Heliö is board member of ERN GuardHeart as the representative of Helsinki University Hospital, has research Collaboration with Blueprint Genetics Quest, and received Bristol Myers Squibb lecturer, member of Advisory Board fees; Aevar Ulfarsson received speaker fees from Pfizer, Abbott, and Orion pharma; Jesper H. Svendsen received research grants (institutional) from Medtronic, speaker fee from Medtronic and fee for being a member of advisory board (Medtronic); Henrik K. Jensen received lecture fees from Abbott Denmark, Amgen Denmark, and Biosense Webster Europe; Kristina H. Haugaa has received speakers' honoraria from Bristol Myers Squibb; Pyotr G. Platonov is a member of Advisory Board at Tenaya Therapeutics and received speaker fees from Pfizer. The remaining authors have no disclosures to report.
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.