- Email: [email protected]
QRS fragmentation in standard ECG as a diagnostic marker of arrhythmogenic right ventricular dysplasia–cardiomyopathy
QRS fragmentation in standard ECG as a diagnostic marker of arrhythmogenic right ventricular dysplasia– cardiomyopathy Stefan Peters, MD,* Martina Trümmel, MD,† Brigitte Koehler, MD† From the *Asklepios Harzkliniken GmbH Goslar, Cardiology and Intensive Care, †Klinikum Quedlinburg, Cardiology. BACKGROUND Epsilon potentials in right precordial leads are reliable diagnostic electrocardiographic (ECG) criteria of arrhythmogenic right ventricular dysplasia-cardiomyopathy (ARVD/C). Sensitivity of epsilon potentials can be enhanced by highly amplified and modified ECG recording technique. Nevertheless, in many cases the definition of epsilon potentials remains difficult. OBJECTIVE To overcome these limitations, the value of QRS fragmentation in a standard 12-lead ECG was analyzed in 360 patients with ARVD/C (176 men, mean age 47.3 ⫾ 13.7 years). METHODS Analysis of QRS fragmentation of the whole collective of patients was compared with the detection of epsilon potentials in highly amplified right precordial and modified limb leads in a subgroup of 207 patients. Fifty-two phenotypically and genotypically nonaffected subjects from systematic family screening in 10 families with known plakophilin-2 and desmoplakin mutations served as a control group.
Introduction Clinical diagnosis of arrhythmogenic right ventricular dysplasia– cardiomyopathy (ARVD/C) is based on standard1 and modified2 diagnostic criteria, including depolarization and repolarization abnormalities in a standard electrocardiogram (ECG). Depolarization abnormalities are traditionally defined as right precordial QRS prolongation ⱖ 110 ms and right precordial epsilon potentials.1 Typical epsilon potentials are located between the end of the QRS complex and the beginning of the T-wave.3 Modified diagnostic criteria also include prolongation of the S-wave upstroke in right precordial leads ⱖ55 ms.4 In many cases, prolonged S-wave upstroke is caused by fragmentation of the end of the QRS complex, not always clearly distinguishable from epsilon potentials. This is the reason why Cox et al5 described Identification of ARVD/C patients according to ISFC/ESC criteria was supported by Prof. Helmut Klein and Prof. Hans-Joachim Trappe at the Medical School Hannover, Germany; by Prof. Gert-Hinrich Reil, Dr. Carl Artur Hartwig, Dr. Jörg Troester, and Dr. Bettina Weber at the Municipal Hospital Oldenburg, Germany; by Prof. William McKenna and Prof. Michael Davies at the St. George’s Medical School London, United Kingdom; and by Prof. Helmut Klein, Dr. Angelo Auricchio, and Dr. Christoph Geller at the University Hospital Magdeburg, Germany. Address reprint requests and correspondence: Dr. Stefan Peters, Asklepios Harzkliniken GmbH Goslar, Cardiology and Intensive Care, Kösliner Straße 12, 38642 Goslar, Germany. E-mail address: [email protected]. (Received May 26, 2008; accepted July 9, 2008.)
RESULTS QRS fragmentation could be found in a total of 306 of 360 patients (85%); 2.09 ⫾ 1.8 fragmented QRS complexes (range 1 to 7) could be found per patient. Fragmented QRS complexes in only 1 right precordial lead were found in 106 cases. In 190 cases, QRS fragmentation was present in more than 1 lead, including all 12 standard leads. Epsilon potentials in highly amplified right precordial and modified limb leads could be found in a total of 159 cases (77%). Typical epsilon potentials in highly amplified right precordial leads could be found in 97 cases (47%). CONCLUSION QRS fragmentation in ARVD/C has a high diagnostic value similar to epsilon potentials by a highly amplified and modified recording technique. KEYWORDS ARVD/C; QRS fragmentation; Epsilon potentials (Heart Rhythm 2008;5:1417–1421) © 2008 Heart Rhythm Society. All rights reserved.
terminal activation delay of ⱖ55 ms as a new ECG criterion of ARVD/C. In 2006, epsilon-like potentials that occur because of fragmentation at the beginning, on top, or at the end of the QRS complex (“pre-, top- and postsilons”) in different leads were presented by Zhang et al6 as typical ECG findings in ARVD/C. The value of QRS fragmentation for diagnostic purposes should be analyzed in ARVD/C in comparison to epsilon potentials in highly amplified and modified ECG recording techniques.
Methods ECGs in 50 mm/s paper speed, 10 mm/mV amplitude, and 40 Hz and 50 Hz filtering technique of 360 patients (176 men, mean age 47.3 ⫾ 13.7 years) with ARVD/C according to traditional1 and modified2 diagnostic criteria were analyzed with regard to QRS fragmentation at the beginning, on top, or at the end of the QRS complex in the standard 12 ECG leads. The phenomenon of QRS fragmentation was defined as deflections at the beginning of the QRS complex, on top of the R-wave, or in the nadir of the S-wave similar to the definition in coronary artery disease7 in either 1 right precordial lead or in more than 1 lead including all standard ECG leads. Figure 1 shows a representative example of an ECG recording of QRS fragmentation in
1547-5271/$ -see front matter © 2008 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2008.07.012
1418
Figure 1 A representative electrocardiograph recording with QRS fragmentation in the beginning of the QRS complex in lead aVF, on top of the R-wave in lead III, and in the nadir of the S-wave in leads II, V2, and V3.
different leads with all deflections at the above-mentioned positions. The first author collected patients from 2 primary and 3 tertiary academic medical centers from 1986 to 2007. These centers were the Medical School Hannover (from 1986 to 1989), the Municipal Hospital of Oldenburg (from 1989 to 1993), St. George’s Medical School London (1994), the University Hospital of Magdeburg (from 1993 to 1997), and the Hospital of Quedlinburg, Academic Teaching Hospital of the University Hospital of Magdeburg (from April 1997 to 2007). The diagnosis of ARVD/C was independently made by several colleagues in the medical centers mentioned above according to ISFC/ESC Task Force criteria published in 1994. Task Force criteria were retrospectively used in those patients diagnosed before 1994; 104 patients had at least two major diagnostic criteria, 233 patients had 1 major criterion and at least 2 minor criteria, and 23 patients had 4 minor Task Force criteria. All patients had standard ECG, echocardiography, Holter monitoring, and right ventricular angiography. Highly amplified (20 mm/ mV) precordial leads and modified limb leads with positioning of the right arm lead over the xiphoid process, the left arm lead on the manubrium sternum, and the left leg lead between V4 and V5 were additionally used in 207 cases. Cardiac magnetic resonance imaging (MRI) was done in 15 cases and endomyocardial biopsies were taken in 15 other cases. Clinical characteristics of the patients in part8 and the results of highly amplified and modified leads were already published.9 Two hundred and sixty patients had newly developed symptoms or were identified by systematic family screening. Only 80 patients had symptoms and disease manifestation for many years. Clinical manifestations were sudden cardiac arrest in 22 cases (6%), sustained ventricular tachycardia in 98 cases (27%), nonsustained ventricular tachycardia in 67 cases (19%), and unexplained syncope in 56 patients (16%). The diagnosis of ARVD/C was made by systematic family screening of symptomatic probands in 83 asymptomatic cases (23%).
Heart Rhythm, Vol 5, No 10, October 2008 A comparison of QRS fragmentation analysis in 360 patients and epsilon potential analysis in highly amplified and modified leads in 207 patients was done. In addition, the prevalence and extent of QRS fragmentation was analyzed separately in those 280 patients with clinical symptoms and 80 asymptomatic affected individuals diagnosed as having ARVD/C by systematic family screening. In the whole cohort of 360 patients, standard ECG parameters such as epsilon potentials in normal recording technique, right precordial QRS prolongation ⱖ110 ms in V1–V3, and right precordial T-wave inversions beyond V1 and modified parameters such as the QRS duration ratio between right (V1–V3) and left precordial leads (V4 –V6) ⱖ 1.2, prolonged S-wave upstroke ⱖ55 ms and terminal activation delay in right precordial leads ⱖ55 ms were documented. Finally, a possible relation between mortality and arrhythmic event data and the number of leads with fragmentation was analyzed. Fifty-two phenotypically and genotypically nonaffected discriminated from systematic family screening (30 men, mean age 42.4 ⫾ 8.3 years) in 9 families with mutations in the gene encoding for plakophilin-2 and 1 family with mutations in the gene encoding for desmoplakin served as a control group already used in the publication on the value of different depolarization criteria in the diagnosis of ARVD/C.9 ECG evaluation was done by 2 cardiologists (M.T., B.K.) blinded for the ECGs with regard to the diagnosis of ARVD/C. Statistical analysis included description of values and prevalence expressed as percentage.
Results QRS fragmentation could be found in a total of 306 of 360 patients (85%) with 2.09 ⫾ 1.8 leads (range 1 to 7) involved. The distribution of the number of leads involved per patient is shown in Figure 2. A total of 106 patients with typical ARVD/C without echocardiographic left ventricular involvement revealed fragmentation in 1 right precordial lead. Patients with fragmentation in 2 leads (right precordial and/or inferior) had in a majority of cases slight left ventricular contraction abnormalities in the inferior and posterolateral region. Typical QRS fragmentation in inferior leads is shown in Figure 3. Patients with fragmentation in 3, 4, and 6 leads revealed left ventricular involvement not only in inferior and posterolateral but also in the apical region. Three patients with QRS fragmentation in 5 leads had no left ventricular involvement, but a severely dilated and globally hypokinetic right ventricle. Three patients with QRS fragmentation in 7 leads were characterized by typical ARVD/C and severely hypokinetic left ventricles. A correlation between cardiac arrest, sustained and nonsustained ventricular tachycardia, and unexplained syncope and the number of leads involved in QRS fragmentation could not be found (Table 1). Fragmentation was characterized by deflections at the beginning of the QRS complex in 57 cases (19%), at the top of the R-wave in 77 cases (25%), and in the nadir of the
Peters et al
QRS Fragmentation
1419
No. of patients
120
n=116: typical ARVD/C without left ventricular involvement
100
n=105: typical ARVD/C with slight LV involvement inferior and posterolateral
80 60
n=53: typical ARVD/C with LV disease
40
n=19: ARVD/C with LV disease n=3: severe RV disease
20
n=7: ARVD/C + LV disease n=3: severe LV disease
0 1
2
3
4
5
6
7 No. of leads
Figure 2 Distribution of the number of leads involved in ARVD/C with the phenomenon of QRS fragmentation and characterization of ARVD/C without left ventricular involvement. ARVD/C ⫽ arrhythmogenic right ventricular dysplasia/cardiomyopathy.
S-wave in 251 cases (82%). QRS fragmentation was present in 2 nonaffected from the control group in only 1 nonright precordial lead (4%). Standard and modified ECG findings at standard recording technique are summarized in Table 2. In those 15 patients with long-term symptoms who had cardiac MRI several years ago without analysis of late enhancement, a relationship between myocardial scar location and QRS fragmentation could not be found. QRS fragmentation was present in 233 of 277 symptomatic cases (84%) with ARVD/C and in 73 of 83 asymptomatic affected cases (88%) mainly diagnosed by systematic family screening. Epsilon potentials in highly amplified right precordial and modified limb leads could be found in a total of 159 of 207 cases (77%). In highly amplified right precordial leads, epsilon waves could be documented in 97 cases (47%). In the control group, epsilon potentials could be excluded by highly amplified and modified ECG technique.
inition. Diagnostic difficulties in the differentiation between prolonged S-wave upstroke4 and epsilon potentials can be overcome by the analysis of terminal activation delay as proposed by Cox et al5 and QRS fragmentation. Additionally, QRS fragmentation also includes the observation of fragmented QRS complexes in all 12 ECG leads, especially in inferior leads already described as a typical ARVD/C finding.9 A correlate to this ECG finding can be found in electrophysiological studies with potential fragmentation at
Discussion The present article describes QRS fragmentation not only in right precordial leads as a new diagnostic criterion of ARVD/C with similar prevalence as epsilon potentials analyzed in highly amplified and modified leads. Before QRS fragmentation can be accepted as a diagnostic criterion of ARVD/C, its value needs to be studied and validated prospectively in a larger series of patients. Because the number of patients in the disease and control groups are rather small, statistical analysis was limited to sensitivity. To define specificity, it would be necessary to compare the prevalence of QRS fragmentation in ARVD/C and in patients with other forms of right ventricular tachycardias or patients with right heart failure. However, the prevalence of other ECG markers of ARVD/C in these clinical settings such as localized right precordial QRS prolongation, prolonged S-wave upstroke, and terminal activation delay was low.5 Nevertheless, this new ECG criterion seems to be a marker of ARVD/C independent from epsilon potential def-
Figure 3 QRS fragmentation in inferior leads II, III, and aVF in ARVD/C. Abbreviations as in Figure 2.
1420
Heart Rhythm, Vol 5, No 10, October 2008
Table 1 Correlation of serious arrhythmic events and number of electrocardiograph leads involved with the phenomenon of QRS fragmentation
Clinical findings
No QRS fragmentation (n ⫽ 54)
Sudden cardiac arrest (n ⫽ 22) Syncope (n ⫽ 56) Sustained VT (n ⫽ 98) Nonsustained VT (n ⫽ 67)
n n n n
⫽ ⫽ ⫽ ⫽
QRS fragmentation 1 lead (n ⫽ 116)
4 10 10 20
n n n n
⫽ ⫽ ⫽ ⫽
8 19 35 21
QRS fragmentation 2 leads (n ⫽ 105) n n n n
⫽ ⫽ ⫽ ⫽
6 16 29 16
QRS fragmentation 3 leads (n ⫽ 53) n n n n
⫽ ⫽ ⫽ ⫽
4 10 20 9
QRS fragmentation 4 to 7 leads (n ⫽ 13) n n n n
⫽ ⫽ ⫽ ⫽
0 1 4 1
VT ⫽ ventricular tachycardia.
different locations within the right ventricle10 and epsilon potentials in modified limb leads. Paper speed of ECG writing (25 or 50 mm/s) seems to be a critical issue in the analysis of QRS duration in right and left precordial leads to define right precordial QRS prolongation ⱖ110 ms and the phenomenon of localized right precordial QRS prolongation defined as QRS duration in (V1 ⫹ V2 ⫹ V3)/(V4 ⫹ V5 ⫹ V6) ⱖ 1.2 in cases with a QRS duration in right precordial leads of ⱖ100 ms.8 Summarizing QRS prolongation, prolonged S-wave upstroke, terminal activation delay, and epsilon potentials the concept of QRS fragmentation will probably solve the dilemma between traditional1 and modified2 diagnostic ECG criteria independent from paper speed used for ECG writing, although not specifically tested in this analysis. In comparison with localized right precordial QRS prolongation with a sensitivity of 98%,7 prolonged S-wave upstroke and terminal activation delay QRS fragmentation seems to be a reliable ECG marker of ARVD/C. Lower sensitivity of QRS fragmentation is possibly because in some cases very small deflections within the QRS complex cannot be detected by standard ECG recording technique. Analysis of QRS fragmentation in highly amplified leads would possibly lead to statistical equivalence. Another phenomenon that must be discussed in detail is the distribution of QRS fragmentation in only 1 right pre-
ARVD/C (n ⫽ 360)
Control group (n ⫽ 52)
n ⫽ 353 (98%)
0
n ⫽ 300 (83%)
n ⫽ 2 (4%)
n ⫽ 305 (85%)
n ⫽ 2 (4%)
cordial lead in more than one-third of all patients and QRS fragmentation in more than 1 lead in the other patients. From standard and modified ECG parameters, morphological aspects of the right ventricle, and clinical findings, no data exist to explain this difference. From a former observational study of ECG findings in ARVD/C, it is known that QRS prolongation can be found not only in right precordial leads but also in inferior and inferolateral leads because of deflections at the end of the QRS complex.8 In these cases, contraction abnormalities of the left ventricle in the inferior and posterolateral region could be detected in a high percentage.8 This suggests that the involvement of the left ventricle in cases of ARVD/C provides certain differences in ECG findings, including QRS fragmentation. A correlation between the number of leads involved in QRS fragmentation and the severity of arrhythmic events could be excluded. Even asymptomatic affected patients with ARVD/C diagnosed by systematic family screening do not present with a lower incidence of QRS fragmentation. Interestingly, QRS fragmentation was first observed in asymptomatic family members of symptomatic ARVD/C probands (personal communication, Guy Fontaine, CARDIOSTIM meeting 2006, Nice, France). In conclusion, QRS fragmentation in different leads of the standard ECG can be found in a high percentage of patients with typical ARVD/C similar to epsilon potentials only detected by a highly amplified and modified ECG recording technique. For the diagnosis of ARVD/C, it seems no longer necessary to use a highly amplified and modified recording technique if a standard ECG is analyzed carefully with regard to ECG correlates of potential fragmentation. To screen ECGs for suspicious ARVD/C, it seems appropriate first to look for right ventricular T-wave inversions and QRS fragmentation mainly in right precordial leads and then to differentiate between typical epsilon potentials, prolonged S-wave upstroke, and terminal activation delay.
n ⫽ 204 (57%)
n ⫽ 3 (6%)
Limitations
n ⫽ 81 (23%) n ⫽ 270 (75%)
0 0
n ⫽ 306 (85%)
n ⫽ 2 (4%)
Table 2 Distribution of traditional and modified ECG parameters of ARVD/C in the whole cohort of 360 patients ECG parameters Localized right precordial QRS prolongation Prolonged S-wave upstroke ⱖ 55 ms Terminal activation delay ⱖ 55 ms Right precordial T-wave inversions ⬎ V1 Epsilon potentials Right precordial QRS duration ⱖ 110 ms QRS fragmentation
ARVD/C ⫽ arrhythmogenic right ventricular dysplasia-cardiomyopathy; ECG ⫽ electrocardiographic.
Especially in modified limb leads, it was difficult to differentiate between typical epsilon potentials between the end of the QRS complex and the beginning of the T-wave and prolongation of the S-wave upstroke. This limitation of the ECG analysis increases the value of QRS fragmentation and terminal activation delay as diagnostic criteria of ARVD/C.
Peters et al
QRS Fragmentation
In general, QRS fragmentation is a frequent observation in coronary artery disease. To diagnose ARVD/C by QRS fragmentation, significant coronary artery disease must be excluded by imaging techniques, for example, coronary angiography.
References 1. McKenna WJ, Thiene G, Nava A, et al. Diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Task Force of the Working Group Myocardial and Pericardial Diseases of the European Society of Cardiology and the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. Br Heart J 1994;71:215–218. 2. Peters S. Advances in the diagnostic management of arrhythmogenic right ventricular dysplasia-cardiomyopathy. Int J Cardiol 2006;113:4 –11. 3. Marcus FI, Fontaine G. Arrhythmogenic right ventricular dysplasia/cardiomyopathy: a review. Pacing Clin Electrophysiol 1995;18:1298 –1314. 4. Nasir K, Bomma C, Tandri H, et al. Electrocardiographic features of arrhythmogenic right ventricular displasia/cardiomyopathy according to dis-
1421
5.
6. 7.
8.
9.
10.
ease severity. A need to broaden diagnostic criteria. Circulation 2004;110:1527–1534. Cox MG, Nelen MR, Wilde AA, et al. Activation delay and VT parameters in arrhythmogenic right ventricular dysplasia/cardiomyopathy: toward improvement of diagnostic ECG criteria. J Cardiovasc Electrophysiol 2008;19:775–781. Zhang L, Riera ARP, Pu J, et al. Are epsilon waves in ARVD “the right precordial leads only” phenomena? Europace 2006;8 Suppl 1:265/4. Das MK, Khan B, Jacob S, et al. Significance of a fragmented QRS complex versus a Q wave in patients with coronary artery disease. Circulation 2006;113: 2495–2501. Peters S, Trümmel M. Diagnosis of arrhythmogenic right ventricular dysplasiacardiomyopathy: value of standard ECG revisited. Ann Noninvasive Electrocardiol 2003;8:238 –245. Peters S, Trümmel M, Koehler B, et al. The value of different electrocardiographic depolarization criteria in the diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. J Electrocardiol 2007;40:34 –37. Fontaine G, Guiraudon G, Frank R, et al. Stimulation studies and epicardial mapping in ventricular tachycardia: study of mechanism and selection for surgery. In: Kulbertus HE (ed) Reentrant Arrhythmias. Lancaster, UK. MTP Press Ltd, 1997;334 –350.
Introduction Clinical diagnosis of arrhythmogenic right ventricular dysplasia– cardiomyopathy (ARVD/C) is based on standard1 and modified2 diagnostic criteria, including depolarization and repolarization abnormalities in a standard electrocardiogram (ECG). Depolarization abnormalities are traditionally defined as right precordial QRS prolongation ⱖ 110 ms and right precordial epsilon potentials.1 Typical epsilon potentials are located between the end of the QRS complex and the beginning of the T-wave.3 Modified diagnostic criteria also include prolongation of the S-wave upstroke in right precordial leads ⱖ55 ms.4 In many cases, prolonged S-wave upstroke is caused by fragmentation of the end of the QRS complex, not always clearly distinguishable from epsilon potentials. This is the reason why Cox et al5 described Identification of ARVD/C patients according to ISFC/ESC criteria was supported by Prof. Helmut Klein and Prof. Hans-Joachim Trappe at the Medical School Hannover, Germany; by Prof. Gert-Hinrich Reil, Dr. Carl Artur Hartwig, Dr. Jörg Troester, and Dr. Bettina Weber at the Municipal Hospital Oldenburg, Germany; by Prof. William McKenna and Prof. Michael Davies at the St. George’s Medical School London, United Kingdom; and by Prof. Helmut Klein, Dr. Angelo Auricchio, and Dr. Christoph Geller at the University Hospital Magdeburg, Germany. Address reprint requests and correspondence: Dr. Stefan Peters, Asklepios Harzkliniken GmbH Goslar, Cardiology and Intensive Care, Kösliner Straße 12, 38642 Goslar, Germany. E-mail address: [email protected]. (Received May 26, 2008; accepted July 9, 2008.)
RESULTS QRS fragmentation could be found in a total of 306 of 360 patients (85%); 2.09 ⫾ 1.8 fragmented QRS complexes (range 1 to 7) could be found per patient. Fragmented QRS complexes in only 1 right precordial lead were found in 106 cases. In 190 cases, QRS fragmentation was present in more than 1 lead, including all 12 standard leads. Epsilon potentials in highly amplified right precordial and modified limb leads could be found in a total of 159 cases (77%). Typical epsilon potentials in highly amplified right precordial leads could be found in 97 cases (47%). CONCLUSION QRS fragmentation in ARVD/C has a high diagnostic value similar to epsilon potentials by a highly amplified and modified recording technique. KEYWORDS ARVD/C; QRS fragmentation; Epsilon potentials (Heart Rhythm 2008;5:1417–1421) © 2008 Heart Rhythm Society. All rights reserved.
terminal activation delay of ⱖ55 ms as a new ECG criterion of ARVD/C. In 2006, epsilon-like potentials that occur because of fragmentation at the beginning, on top, or at the end of the QRS complex (“pre-, top- and postsilons”) in different leads were presented by Zhang et al6 as typical ECG findings in ARVD/C. The value of QRS fragmentation for diagnostic purposes should be analyzed in ARVD/C in comparison to epsilon potentials in highly amplified and modified ECG recording techniques.
Methods ECGs in 50 mm/s paper speed, 10 mm/mV amplitude, and 40 Hz and 50 Hz filtering technique of 360 patients (176 men, mean age 47.3 ⫾ 13.7 years) with ARVD/C according to traditional1 and modified2 diagnostic criteria were analyzed with regard to QRS fragmentation at the beginning, on top, or at the end of the QRS complex in the standard 12 ECG leads. The phenomenon of QRS fragmentation was defined as deflections at the beginning of the QRS complex, on top of the R-wave, or in the nadir of the S-wave similar to the definition in coronary artery disease7 in either 1 right precordial lead or in more than 1 lead including all standard ECG leads. Figure 1 shows a representative example of an ECG recording of QRS fragmentation in
1547-5271/$ -see front matter © 2008 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2008.07.012
1418
Figure 1 A representative electrocardiograph recording with QRS fragmentation in the beginning of the QRS complex in lead aVF, on top of the R-wave in lead III, and in the nadir of the S-wave in leads II, V2, and V3.
different leads with all deflections at the above-mentioned positions. The first author collected patients from 2 primary and 3 tertiary academic medical centers from 1986 to 2007. These centers were the Medical School Hannover (from 1986 to 1989), the Municipal Hospital of Oldenburg (from 1989 to 1993), St. George’s Medical School London (1994), the University Hospital of Magdeburg (from 1993 to 1997), and the Hospital of Quedlinburg, Academic Teaching Hospital of the University Hospital of Magdeburg (from April 1997 to 2007). The diagnosis of ARVD/C was independently made by several colleagues in the medical centers mentioned above according to ISFC/ESC Task Force criteria published in 1994. Task Force criteria were retrospectively used in those patients diagnosed before 1994; 104 patients had at least two major diagnostic criteria, 233 patients had 1 major criterion and at least 2 minor criteria, and 23 patients had 4 minor Task Force criteria. All patients had standard ECG, echocardiography, Holter monitoring, and right ventricular angiography. Highly amplified (20 mm/ mV) precordial leads and modified limb leads with positioning of the right arm lead over the xiphoid process, the left arm lead on the manubrium sternum, and the left leg lead between V4 and V5 were additionally used in 207 cases. Cardiac magnetic resonance imaging (MRI) was done in 15 cases and endomyocardial biopsies were taken in 15 other cases. Clinical characteristics of the patients in part8 and the results of highly amplified and modified leads were already published.9 Two hundred and sixty patients had newly developed symptoms or were identified by systematic family screening. Only 80 patients had symptoms and disease manifestation for many years. Clinical manifestations were sudden cardiac arrest in 22 cases (6%), sustained ventricular tachycardia in 98 cases (27%), nonsustained ventricular tachycardia in 67 cases (19%), and unexplained syncope in 56 patients (16%). The diagnosis of ARVD/C was made by systematic family screening of symptomatic probands in 83 asymptomatic cases (23%).
Heart Rhythm, Vol 5, No 10, October 2008 A comparison of QRS fragmentation analysis in 360 patients and epsilon potential analysis in highly amplified and modified leads in 207 patients was done. In addition, the prevalence and extent of QRS fragmentation was analyzed separately in those 280 patients with clinical symptoms and 80 asymptomatic affected individuals diagnosed as having ARVD/C by systematic family screening. In the whole cohort of 360 patients, standard ECG parameters such as epsilon potentials in normal recording technique, right precordial QRS prolongation ⱖ110 ms in V1–V3, and right precordial T-wave inversions beyond V1 and modified parameters such as the QRS duration ratio between right (V1–V3) and left precordial leads (V4 –V6) ⱖ 1.2, prolonged S-wave upstroke ⱖ55 ms and terminal activation delay in right precordial leads ⱖ55 ms were documented. Finally, a possible relation between mortality and arrhythmic event data and the number of leads with fragmentation was analyzed. Fifty-two phenotypically and genotypically nonaffected discriminated from systematic family screening (30 men, mean age 42.4 ⫾ 8.3 years) in 9 families with mutations in the gene encoding for plakophilin-2 and 1 family with mutations in the gene encoding for desmoplakin served as a control group already used in the publication on the value of different depolarization criteria in the diagnosis of ARVD/C.9 ECG evaluation was done by 2 cardiologists (M.T., B.K.) blinded for the ECGs with regard to the diagnosis of ARVD/C. Statistical analysis included description of values and prevalence expressed as percentage.
Results QRS fragmentation could be found in a total of 306 of 360 patients (85%) with 2.09 ⫾ 1.8 leads (range 1 to 7) involved. The distribution of the number of leads involved per patient is shown in Figure 2. A total of 106 patients with typical ARVD/C without echocardiographic left ventricular involvement revealed fragmentation in 1 right precordial lead. Patients with fragmentation in 2 leads (right precordial and/or inferior) had in a majority of cases slight left ventricular contraction abnormalities in the inferior and posterolateral region. Typical QRS fragmentation in inferior leads is shown in Figure 3. Patients with fragmentation in 3, 4, and 6 leads revealed left ventricular involvement not only in inferior and posterolateral but also in the apical region. Three patients with QRS fragmentation in 5 leads had no left ventricular involvement, but a severely dilated and globally hypokinetic right ventricle. Three patients with QRS fragmentation in 7 leads were characterized by typical ARVD/C and severely hypokinetic left ventricles. A correlation between cardiac arrest, sustained and nonsustained ventricular tachycardia, and unexplained syncope and the number of leads involved in QRS fragmentation could not be found (Table 1). Fragmentation was characterized by deflections at the beginning of the QRS complex in 57 cases (19%), at the top of the R-wave in 77 cases (25%), and in the nadir of the
Peters et al
QRS Fragmentation
1419
No. of patients
120
n=116: typical ARVD/C without left ventricular involvement
100
n=105: typical ARVD/C with slight LV involvement inferior and posterolateral
80 60
n=53: typical ARVD/C with LV disease
40
n=19: ARVD/C with LV disease n=3: severe RV disease
20
n=7: ARVD/C + LV disease n=3: severe LV disease
0 1
2
3
4
5
6
7 No. of leads
Figure 2 Distribution of the number of leads involved in ARVD/C with the phenomenon of QRS fragmentation and characterization of ARVD/C without left ventricular involvement. ARVD/C ⫽ arrhythmogenic right ventricular dysplasia/cardiomyopathy.
S-wave in 251 cases (82%). QRS fragmentation was present in 2 nonaffected from the control group in only 1 nonright precordial lead (4%). Standard and modified ECG findings at standard recording technique are summarized in Table 2. In those 15 patients with long-term symptoms who had cardiac MRI several years ago without analysis of late enhancement, a relationship between myocardial scar location and QRS fragmentation could not be found. QRS fragmentation was present in 233 of 277 symptomatic cases (84%) with ARVD/C and in 73 of 83 asymptomatic affected cases (88%) mainly diagnosed by systematic family screening. Epsilon potentials in highly amplified right precordial and modified limb leads could be found in a total of 159 of 207 cases (77%). In highly amplified right precordial leads, epsilon waves could be documented in 97 cases (47%). In the control group, epsilon potentials could be excluded by highly amplified and modified ECG technique.
inition. Diagnostic difficulties in the differentiation between prolonged S-wave upstroke4 and epsilon potentials can be overcome by the analysis of terminal activation delay as proposed by Cox et al5 and QRS fragmentation. Additionally, QRS fragmentation also includes the observation of fragmented QRS complexes in all 12 ECG leads, especially in inferior leads already described as a typical ARVD/C finding.9 A correlate to this ECG finding can be found in electrophysiological studies with potential fragmentation at
Discussion The present article describes QRS fragmentation not only in right precordial leads as a new diagnostic criterion of ARVD/C with similar prevalence as epsilon potentials analyzed in highly amplified and modified leads. Before QRS fragmentation can be accepted as a diagnostic criterion of ARVD/C, its value needs to be studied and validated prospectively in a larger series of patients. Because the number of patients in the disease and control groups are rather small, statistical analysis was limited to sensitivity. To define specificity, it would be necessary to compare the prevalence of QRS fragmentation in ARVD/C and in patients with other forms of right ventricular tachycardias or patients with right heart failure. However, the prevalence of other ECG markers of ARVD/C in these clinical settings such as localized right precordial QRS prolongation, prolonged S-wave upstroke, and terminal activation delay was low.5 Nevertheless, this new ECG criterion seems to be a marker of ARVD/C independent from epsilon potential def-
Figure 3 QRS fragmentation in inferior leads II, III, and aVF in ARVD/C. Abbreviations as in Figure 2.
1420
Heart Rhythm, Vol 5, No 10, October 2008
Table 1 Correlation of serious arrhythmic events and number of electrocardiograph leads involved with the phenomenon of QRS fragmentation
Clinical findings
No QRS fragmentation (n ⫽ 54)
Sudden cardiac arrest (n ⫽ 22) Syncope (n ⫽ 56) Sustained VT (n ⫽ 98) Nonsustained VT (n ⫽ 67)
n n n n
⫽ ⫽ ⫽ ⫽
QRS fragmentation 1 lead (n ⫽ 116)
4 10 10 20
n n n n
⫽ ⫽ ⫽ ⫽
8 19 35 21
QRS fragmentation 2 leads (n ⫽ 105) n n n n
⫽ ⫽ ⫽ ⫽
6 16 29 16
QRS fragmentation 3 leads (n ⫽ 53) n n n n
⫽ ⫽ ⫽ ⫽
4 10 20 9
QRS fragmentation 4 to 7 leads (n ⫽ 13) n n n n
⫽ ⫽ ⫽ ⫽
0 1 4 1
VT ⫽ ventricular tachycardia.
different locations within the right ventricle10 and epsilon potentials in modified limb leads. Paper speed of ECG writing (25 or 50 mm/s) seems to be a critical issue in the analysis of QRS duration in right and left precordial leads to define right precordial QRS prolongation ⱖ110 ms and the phenomenon of localized right precordial QRS prolongation defined as QRS duration in (V1 ⫹ V2 ⫹ V3)/(V4 ⫹ V5 ⫹ V6) ⱖ 1.2 in cases with a QRS duration in right precordial leads of ⱖ100 ms.8 Summarizing QRS prolongation, prolonged S-wave upstroke, terminal activation delay, and epsilon potentials the concept of QRS fragmentation will probably solve the dilemma between traditional1 and modified2 diagnostic ECG criteria independent from paper speed used for ECG writing, although not specifically tested in this analysis. In comparison with localized right precordial QRS prolongation with a sensitivity of 98%,7 prolonged S-wave upstroke and terminal activation delay QRS fragmentation seems to be a reliable ECG marker of ARVD/C. Lower sensitivity of QRS fragmentation is possibly because in some cases very small deflections within the QRS complex cannot be detected by standard ECG recording technique. Analysis of QRS fragmentation in highly amplified leads would possibly lead to statistical equivalence. Another phenomenon that must be discussed in detail is the distribution of QRS fragmentation in only 1 right pre-
ARVD/C (n ⫽ 360)
Control group (n ⫽ 52)
n ⫽ 353 (98%)
0
n ⫽ 300 (83%)
n ⫽ 2 (4%)
n ⫽ 305 (85%)
n ⫽ 2 (4%)
cordial lead in more than one-third of all patients and QRS fragmentation in more than 1 lead in the other patients. From standard and modified ECG parameters, morphological aspects of the right ventricle, and clinical findings, no data exist to explain this difference. From a former observational study of ECG findings in ARVD/C, it is known that QRS prolongation can be found not only in right precordial leads but also in inferior and inferolateral leads because of deflections at the end of the QRS complex.8 In these cases, contraction abnormalities of the left ventricle in the inferior and posterolateral region could be detected in a high percentage.8 This suggests that the involvement of the left ventricle in cases of ARVD/C provides certain differences in ECG findings, including QRS fragmentation. A correlation between the number of leads involved in QRS fragmentation and the severity of arrhythmic events could be excluded. Even asymptomatic affected patients with ARVD/C diagnosed by systematic family screening do not present with a lower incidence of QRS fragmentation. Interestingly, QRS fragmentation was first observed in asymptomatic family members of symptomatic ARVD/C probands (personal communication, Guy Fontaine, CARDIOSTIM meeting 2006, Nice, France). In conclusion, QRS fragmentation in different leads of the standard ECG can be found in a high percentage of patients with typical ARVD/C similar to epsilon potentials only detected by a highly amplified and modified ECG recording technique. For the diagnosis of ARVD/C, it seems no longer necessary to use a highly amplified and modified recording technique if a standard ECG is analyzed carefully with regard to ECG correlates of potential fragmentation. To screen ECGs for suspicious ARVD/C, it seems appropriate first to look for right ventricular T-wave inversions and QRS fragmentation mainly in right precordial leads and then to differentiate between typical epsilon potentials, prolonged S-wave upstroke, and terminal activation delay.
n ⫽ 204 (57%)
n ⫽ 3 (6%)
Limitations
n ⫽ 81 (23%) n ⫽ 270 (75%)
0 0
n ⫽ 306 (85%)
n ⫽ 2 (4%)
Table 2 Distribution of traditional and modified ECG parameters of ARVD/C in the whole cohort of 360 patients ECG parameters Localized right precordial QRS prolongation Prolonged S-wave upstroke ⱖ 55 ms Terminal activation delay ⱖ 55 ms Right precordial T-wave inversions ⬎ V1 Epsilon potentials Right precordial QRS duration ⱖ 110 ms QRS fragmentation
ARVD/C ⫽ arrhythmogenic right ventricular dysplasia-cardiomyopathy; ECG ⫽ electrocardiographic.
Especially in modified limb leads, it was difficult to differentiate between typical epsilon potentials between the end of the QRS complex and the beginning of the T-wave and prolongation of the S-wave upstroke. This limitation of the ECG analysis increases the value of QRS fragmentation and terminal activation delay as diagnostic criteria of ARVD/C.
Peters et al
QRS Fragmentation
In general, QRS fragmentation is a frequent observation in coronary artery disease. To diagnose ARVD/C by QRS fragmentation, significant coronary artery disease must be excluded by imaging techniques, for example, coronary angiography.
References 1. McKenna WJ, Thiene G, Nava A, et al. Diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Task Force of the Working Group Myocardial and Pericardial Diseases of the European Society of Cardiology and the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. Br Heart J 1994;71:215–218. 2. Peters S. Advances in the diagnostic management of arrhythmogenic right ventricular dysplasia-cardiomyopathy. Int J Cardiol 2006;113:4 –11. 3. Marcus FI, Fontaine G. Arrhythmogenic right ventricular dysplasia/cardiomyopathy: a review. Pacing Clin Electrophysiol 1995;18:1298 –1314. 4. Nasir K, Bomma C, Tandri H, et al. Electrocardiographic features of arrhythmogenic right ventricular displasia/cardiomyopathy according to dis-
1421
5.
6. 7.
8.
9.
10.
ease severity. A need to broaden diagnostic criteria. Circulation 2004;110:1527–1534. Cox MG, Nelen MR, Wilde AA, et al. Activation delay and VT parameters in arrhythmogenic right ventricular dysplasia/cardiomyopathy: toward improvement of diagnostic ECG criteria. J Cardiovasc Electrophysiol 2008;19:775–781. Zhang L, Riera ARP, Pu J, et al. Are epsilon waves in ARVD “the right precordial leads only” phenomena? Europace 2006;8 Suppl 1:265/4. Das MK, Khan B, Jacob S, et al. Significance of a fragmented QRS complex versus a Q wave in patients with coronary artery disease. Circulation 2006;113: 2495–2501. Peters S, Trümmel M. Diagnosis of arrhythmogenic right ventricular dysplasiacardiomyopathy: value of standard ECG revisited. Ann Noninvasive Electrocardiol 2003;8:238 –245. Peters S, Trümmel M, Koehler B, et al. The value of different electrocardiographic depolarization criteria in the diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. J Electrocardiol 2007;40:34 –37. Fontaine G, Guiraudon G, Frank R, et al. Stimulation studies and epicardial mapping in ventricular tachycardia: study of mechanism and selection for surgery. In: Kulbertus HE (ed) Reentrant Arrhythmias. Lancaster, UK. MTP Press Ltd, 1997;334 –350.