Localization of Optimal Ablation Site of Idiopathic Ventricular Tachycardia from Right and Left Ventricular Outflow Tract by Body Surface ECG
Background—Idiopathic ventricular tachycardia (VT) is known to arise from the right ventricular (RV) and left ventricular outflow tracts (LVOT). However, reliable noninvasive methods to localize the optimum ablation site for VT have not been reported.
Methods and Results—Body surface maps (BSM) and 12-lead ECGs were investigated in 35 VTs from the RVOT and 5 VTs from the LVOT in which the origin was confirmed during the ablation procedure. The RVOT was classified into 8 subdivisions with the use of a 3-dimensional anatomic relation: anterior (A)–posterior (P), right (R)–left (L), and superior (S)–inferior (I). On the BSM, the following 3 indexes differentiated each location of the origin, with a diagnostic accuracy of 88% (A-P), 92% (R-L), and 77% (S-I): (1) the location of the minimum at the early-to-mid QRS (right, A; left, P), (2) the isopotential distribution in the left shoulder area after 30 ms of QRS (positive, R; negative, L), and (3) the downward moving time of the minimum at the early-to-mid QRS (≥50 ms, S; <50 ms, I). On the 12-lead ECG, (1) the QRS duration (>140 ms, A; ≤140 ms, P) and the R-wave pattern in leads II and III (RR’ or Rr’, A, R, P), (2) the QS wave amplitude in aVR and aVL (aVR≥aVL, R; aVR<aVL, L), and (3) the r-wave amplitude in V1 and V2 (high, S; low, I) localized the origin with 80%, 86% (A-P), 80% (R-L), and 66% (S-I) accuracy. R/S≥1 in lead V3 was an index suggesting the LVOT origin.
Conclusions—The origin or the optimum ablation site of idiopathic VT from RVOT and LVOT can be localized with the use of indexes obtained with a BSM or 12-lead ECG.
Catheter ablation has been accepted as curative treatment for patients with refractory tachyarrhythmias. Excellent results have recently been reported in patients with Wolff-Parkinson-White syndrome and in those with atrioventricular nodal tachycardia. However, it is generally known that complete success is not obtainable in ventricular tachycardia (VT) compared with supraventricular tachycardia because of the difficulty in localizing the origin of the VT.1 2 3 In this study, we investigated methods for estimating the origin or the optimum ablation site of VT from the right and left ventricular outflow tracts (RVOT and LVOT) by using a body surface map (BSM) or 12-lead ECG.
The subjects were 31 patients (9 men and 22 women, mean age, 36.7±13.0 years) with 35 types of nonsustained monomorphic VT originating from the RVOT in whom VT had been ablated successfully and 5 patients (4 men and 1 woman, age 15 to 58 years) with 5 types of nonsustained VT originating from the LVOT in whom the VT origins were identified by pace mapping. No patients had structural heart disease. Patients with VT showing right bundle-branch block morphology were excluded from this study.
Catheter ablation was performed with the use of radiofrequency energy in these patients between June 1992 and June 1996. All of the patients gave their written informed consent. In 29 patients, the radiofrequency energy was delivered as a continuous unmodulated sine wave at 500 to 520 kHz with an average power delivery of 15 to 50 W. In the other 2 patients, the radiofrequency energy was applied with a temperature-controlled system with a temperature set-point of 60°C. A multipolar, closely spaced (2 mm), deflectable, large-tip (4 mm) electrode catheter was used for mapping and ablation.
The optimum ablation site was determined mostly by pace mapping guided by the late coupled (>400 ms) premature ventricular contraction (PVC), which showed similar QRS morphology to each VT. When an isolated PVC was not recorded, the first beat of VT was selected as a template. The origin of PVC or VT was defined as the ablation site where the best pace mapping score was obtained and the targeted PVC and VT disappeared by a single energy delivery. For VT from the LVOT, the origin was defined as the site where the best pace mapping score was obtained and the endocardial activation time during VT was the shortest. Pace mapping was performed during sinus rhythm with bipolar cathodal stimulation at an output just greater than the diastolic threshold. The ventricle was paced at a slow rate of 120 to 140 per minute, according to the coupling interval of the PVC selected as a template. The pace mapping score was calculated in early, mid, and late QRS period of each lead; that is, the QRS segment was divided into 3 sections, and the wave morphology of each period was compared between the paced wave and the late coupled PVC. A score of 0 to 0.3 point was given, depending on the wave similarity for each period, and 0 or 0.1 point was given, depending on the similarity of QRS amplitude. If the QRS morphology and amplitude of 12 ECGs of the paced beat was perfectly similar to that of PVC, the pace mapping score was calculated as (0.3×3+0.1)×12=12 points. Successful ablation was defined as the complete absence of target VT and PVC during continuous ECG monitoring for 1 week after the ablation procedure.
Body Surface Mapping
BSM was performed in 29 patients during 30 types of PVC (26, right ventricular [RV] origin; 4, left ventricular [LV] origin) that showed the same QRS morphology as each VT of outflow origin. On the basis of the relation between the characteristics of QRS isopotential maps, QRS isointegral maps and the identified origins of VT, the indexes on the BSM for localizing the precise origin were investigated.
Because VT originating from the LVOT shows a similar morphology on an ECG to that of VT originating from the RVOT, the methods to differentiate these types of VT were evaluated after the characteristics of mapping were examined.
A BSM was recorded during PVC with the use of an integrated mapping system (VCM 3000, Fukuda Denshi Co) at the sampling interval of 1 ms. Electrodes for mapping were placed at 87 points on the anterior chest and back. On the vertical plane of the body surface, leads on the right midaxillary line, the left midaxillary line, the anterior chest, and the back were shown as lines of A, I, B to H, and J to M; on the horizontal plane, electrodes were set from rows 1 to 7 as the standard of the second intercostal space on the midsternal line (row 6) and the fifth intercostal space on the midclavicular line (row 4).4 The QRS onset was defined as the instant at which a mean-root-square voltage of the 87 ECGs (spatial magnitude) began to increase, and the QRS end was defined as the instant at which the isopotential distributions changed markedly after the spatial magnitude successively decreased. Patients in whom the preceding sinus T wave overlapped with the QRS wave of PVC because of a short coupling interval were excluded.
A 12-lead ECG was recorded during 40 types of VT or PVC. After methods for estimating the origin of VT from the 12-lead ECG were investigated on the basis of parameters obtained from the BSM, the diagnostic accuracy of the parameters consisting of the duration, morphology, amplitude, and polarity of the QRS segment was examined with the use of these ECGs, in which the late coupled PVC or the first beat of VT was recorded. The RR’ wave was defined as a triphasic R wave in which the second R peak is equal to or higher than the first R peak. The Rr’ wave was defined as a triphasic R wave in which the second R peak is lower than the first R peak or as a monophasic R wave in which a notch is present after the R peak.
Classification of Anatomic Sites
The PVC or VT origins in the RVOT were anatomically classified into 3-dimensional directions: anterior and posterior, right and left, and superior and inferior. The anterior half of the RVOT by fluoroscopy at the 60 degree left anterior oblique position was defined as the anterior side = free-wall side, and the posterior half was defined as the posterior side = septal side. When viewed by fluoroscopy at the 30 degree right anterior oblique position, the right half of the outflow was defined as the right side = posterolateral attachment side, and the left half was defined as the left side = anterior attachment side.5 The area within 1 cm of the pulmonic valve was defined as the superior side = proximal side below the pulmonic valve, and the area over 1 cm away was defined as the inferior side = distal side below the pulmonic valve (Figure 1⇓). This means that the RVOT consisted of 8 subdivisions: (1) free-wall, right, and proximal side, (2) free-wall, right, and distal side, (3) free-wall, left, and proximal side, (4) free-wall, left, and distal side, (5) septal, right, and proximal side, (6) septal, right, and distal side, (7) septal, left, and proximal side, and (8) septal, left, and distal side. The PVC or VT origins in the LVOT were not subdivided 3-dimensionally. The position of the pulmonic valve was confirmed by right ventriculography or the appearance of a ventricular potential in the distal electrogram when the ablation catheter was slowly withdrawn from the pulmonary artery. The position of the coronary cusp was confirmed by aortography or coronary angiography.
Parametric data are expressed as mean±SD. The ECG parameters in different groups were compared by Student unpaired t test. The diagnostic accuracy of each estimation method was expressed as the ratio of cases with a positive index among all cases. Percentages were compared by Pearson χ2 test or Fisher exact test, depending on the frequencies. A value of P<0.05 was considered significant.
The sites of origin of all 40 VTs are shown in Figure 1⇑. Of the 35 VTs originating from the RVOT, 8 were classified as of free-wall origin, 27 of septal, 23 of left, 12 of right, 22 of proximal, and 13 of distal origin. The QRS wave morphology during pacing at the successful ablation site was in almost complete agreement with the wave morphology during VT or PVC in all patients (mean matching score in 12-lead ECG: 11.5/12). Of the 5 VT originating from the LVOT, the identified origin was the LVOT close to the left coronary cusp in 3 VT and the LVOT close to the right coronary cusp in 2 VT. All VT showed left bundle-branch block and inferior-axis morphology on the 12-lead ECG.
Characteristics of BSM of PVCs Originating from RVOT
The origin of 26 PVCs during which BSM was recorded was as follows: 7 PVC, the free-wall side; 19 PVC, the septal side; 16 PVC, the left side; 10 PVC, the right side; 18 PVC, the proximal side; and 8 PVC, the distal side.
The following tendencies were noted on the isopotential maps. Briefly, the minimum potential appeared on the superior chest within 20 ms from QRS onset, remained on E7 or F7 between 10 and 60 ms, moved slightly downward at the mid QRS (EF5, 6), and stayed there until the QRS end. The isopotential distribution remained negative in the superior chest and back, whereas positive potentials were noted in the inferior area during the early-to-end QRS period.
The location of the minimum at the early-to-mid QRS period was useful for the differentiation of the anterior and posterior areas of the RVOT. In 5 of the 7 PVCs of free-wall origin, the minimum was fixed on the median line of the anterior chest (E line) or temporarily moved rightward, whereas in 18 of 19 PVCs of septal origin, the minimum first appeared on the median line at the early QRS, moved leftward (F line), and remained there or returned to the median line in the mid-QRS period (Figure 2⇓).
The potential distribution in the superior area of the left anterior chest after 30 ms in the early QRS period was useful for the distinction between the right and left sides of RVOT. In all PVCs of left-side origin, the superior area of the left anterior chest (F-J, 6, 7) was successively negative, whereas in 8 of the 10 PVCs of right-side origin, the same area was occupied by positive potential, although transiently (Figure 3⇓).
The time for which the minimum shifted downward during the period from the early to the end QRS was useful for the distinction between the superior and inferior sides of the RVOT. In other words, the time required for the minimum to shift from the uppermost 7 line (D, E, F-7) to the 6 line (D, E, F-6) on the map was between 50 and 80 ms in most of the PVCs originating on the proximal side below the pulmonic valve, whereas it was shorter than 50 ms in 5 of the 8 PVCs originating on the distal side below the pulmonic valve (Figure 4⇓).
In the QRS isointegral maps, there were differences in the distribution of the minimum between the PVCs of free-wall origin and of septal origin (E6 in 6 of the 7 PVCs of free-wall origin and F6 in 15 of the 19 PVCs of septal origin), but there were no differences in the potential distribution between the right origin and left origin or between the superior origin and inferior origin. In these cases, the potential was negative in the superior area of the anterior chest and the superior back and positive in the inferior area of the anterior chest and back. The minimum appeared at E6 or F6, and the maximum appeared at the lead from the left inferior chest to the left lateral chest. The subtle changes in the potential from the early-to-mid QRS observed on the isopotential maps were not reflected on the isointegral maps (Figure 5⇓).
The QRS duration was significantly longer in PVCs of free-wall origin (153.4±11.2 ms) than in the PVCs of septal origin (134.9±10.8 ms). However, there was no difference between the right origin and left origin (146.2±16.2 ms versus 136±10.4 ms) or between the superior origin and inferior origin (141.9±10.9 ms versus 135.5±18.3 ms).
Characteristics of BSMs of PVCs Originating From LVOT
Some of the VTs that were judged to originate from the RVOT were actually of LVOT origin. Figure 6⇓ shows the isopotential maps and 12-lead ECG of the PVC originating from the left coronary sinus of the aortic valve. On the isopotential maps, the minimum potential was present at E7 or F7 from the early-to-mid-QRS, and it shifted downward (6 line) after 50 ms of QRS. On the basis of these findings, the PVC origin was considered to be on the right septal side of RVOT and on the proximal side. However, the downward movement of the minimum was very slow, taking 100 ms. The prolongation of the moving time of the minimum (>80 ms) was observed in the other 3 PVCs originating from the LVOT, which was considered to be characteristic of PVC of that origin.
Table 1⇓ summarizes the indexes for localizing the origin of PVC with the left bundle-branch block and inferior-axis morphology by isopotential maps. According to this estimation method, the anterior-posterior location was differentiated by the minimum location at the early-to-mid QRS with a diagnostic accuracy of 88%. The potential distribution in the left shoulder area after 30 ms of QRS differentiated the right-left location with a diagnostic accuracy of 92%, and the downward moving time of the minimum at the early-to-mid QRS differentiated the superior-inferior location with 77% accuracy and RVOT-LVOT origin with 97% accuracy.
Estimation Method of VT and PVC Origins on 12-Lead ECG
The following relations between the origin of ventricular arrhythmias and the 12-lead ECG were deducted from the results of the isopotential maps of PVCs originating from the RVOT. Table 2⇓ shows the estimation indexes and the results, and Figure 7⇓ shows representative wave patterns of the 12-lead ECG.
Differentiation Between Free-Wall Side and Septal Side
The QRS duration and the QRS wave morphology in leads II and III was informative. If the QRS duration was >140 ms, the origin was likely to be on the free-wall side (7/8 VT). If it was ≤140 ms, the origin was likely to be on the septal side (21/27 VT) (diagnostic accuracy: 80%).
If the RR’ or Rr’ wave pattern was observed in leads II and III, the origin was on the free-wall side (8/8 VT). If the R-wave pattern was seen in leads II and III, the origin was likely to be on the septal side (22/27 VT) (diagnostic accuracy: 86%).
Differentiation Between Left Side and Right Side
The QS wave amplitude in leads aVR and aVL was useful. If the QS wave depth in lead aVL was larger than that in lead aVR, the origin was likely to be on the left side (18/23 VT). If the QS amplitude in lead aVR was equal to or larger than that in lead aVL, the origin was likely to be on the right side (10/12 VT) (diagnostic accuracy: 80%).
The QRS wave polarity of lead I was another useful index. If lead I showed negative polarity (QS, Qr, or rS wave pattern), the origin was likely to be on the left side (20/23 VT). If lead I showed positive polarity (R-wave pattern), the origin was likely to be on the right side (9 of 12 VT) (diagnostic accuracy: 83%).
Differentiation Between Superior Side and Inferior Side
The initial r-wave amplitude in leads V1 and V2 was a useful index. If the initial r-wave amplitude in leads V1 and V2 was high (≥0.2 mV in both leads), the origin was likely to be on the proximal side (14/22 VT). If the initial r-wave amplitude in V1 and V2 was low (<0.2 mV in one or both leads), the origin was likely to be on the distal side (9/13 VT) (diagnostic accuracy: 66%).
Differentiation Between LVOT Side and RVOT Side
The ratio of the R/S amplitude in lead V3 was a useful index. If the initial r-wave amplitude in leads V1 and V2 was high and if the R/S ratio in lead V3 was ≥1, the origin was likely to be in the LVOT (4/5 VT). Conversely, if the R/S ratio was <1, the origin was likely to be in the RVOT (29/35 VT) (diagnostic accuracy: 83%).
The mechanism of idiopathic right VT is considered to be automaticity or triggered activity6 7 8 in which the VT exits from same site as the origin of arrhythmia. Pace mapping is reported to be a reliable method for localizing the origin of VT.2 It has also been reported that PVC-guided pace mapping is useful as an ordinary method guided by VT.3 9 Thus, the optimum ablation site can be found with the use of body surface ECGs during VT or PVC in patients with idiopathic right VT.
We previously reported that the location of the minimum at 40 ms in the early QRS period or at the time when the minimum potential exceeds −0.5 mV on the isopotential map is a reliable index for localizing the origin of VT or PVC.10 Hayashi et al11 used isopotential distributions, and SippensGroenewegen12 13 et al used the potential distribution of the QRS isointegral map for the estimation of the origins. However, these previous methods cannot localize the origin in detail, though they suggest that the VT arises around the RVOT. We found that the origin can be localized in detail if the isopotential distributions are evaluated separately according to the 3-dimensional anatomic relation.
Localization of VT and PVC From Anterior or Posterior Side
The different minimum locations on the isopotential maps between the free-wall side and the septal side is interpreted as follows. For the free-wall origin, because the epicardial breakthrough appears on the anterior epicardial surface of the right ventricle just above the origin, the anterior chest continuously monitors the excitation front away from the RV free wall, resulting in a stable minimum on the middle-to-right area of the anterior chest. In contrast, for the septal origin, the minimum reflecting the excitation front spreading from the septum toward the left ventricle appears on the middle line at the early QRS period. After a while, an epicardial breakthrough occurs on the boundary between the free wall and the septum, which is on the left border of the RVOT, and the excitation wave front proceeds away from this breakthrough site toward the left ventricle, resulting in the leftward movement or rightward after leftward movement of the minimum.
The 12-lead ECG findings of origins in the anterior and posterior directions can be explained by the relation between the findings of the BSM and anatomic location. The shorter QRS duration in the PVC or VT of septum origin is probably caused by the excitation from the outflow septum, which spread downward over the right and left ventricles more rapidly through the septal Purkinje network than the excitation from the free wall. The positive polarity in leads II and III reflects the successive positive isopotential distribution on the lower chest. The triphasic RR’ or Rr’ wave that appeared in the PVC of free-wall origin probably reflects the longer QRS duration and the phased excitation from the RV free wall to the LV.
Localization of VT and PVC From Right or Left Side
The anterior attachment of RVOT is anatomically localized in the upper left area of the heart. Therefore, the local excitation front proceeds away from the upper area of the left anterior chest, and the negative potential area is expected to stay at that time after epicardial breakthrough in VT or PVC originating from this region. The posterolateral attachment is localized on the right side of the RVOT and at a slightly lower site than the anterior attachment. Because excitation originating from this site invariably proceeds to the superior left direction at least once, a positive-potential area is expected to appear in the superior area of the left anterior chest at the early-to-mid-QRS period.
The ECG morphology of the aVR and aVL leads are similar to those recorded from the unipolar ECGs at the right and left shoulder areas, respectively. In the PVC from the left side of the RVOT, the left upper chest leads and the aVL lead consequently show a larger QS wave than aVR because they continuously monitor the excitation front away from the left upper region of the heart. In the excitation from the right side of the RVOT, the right upper chest leads and aVR show a larger QS wave than aVL by the excitation front away from the right upper side.
Coggins et al14 reported that VT of the RV outflow septal origin showed a negative QRS complex in lead aVL, whereas that of lateral origin showed a positive QRS. In our study, every case of VT or PVC from the RVOT showed a negative QRS complex in lead aVL. We suspect that the polarity of aVL cannot be an independent index for determining the origin from the outflow tract. The polarity of aVL may be useful for the localization of VT from the inferior area of the RVOT and the middle-to-inferior region of the right ventricle.
Localization of VT and PVC From Superior or Inferior Side
The downward moving time of the minimum potential was useful for estimating the origins in the superior and inferior directions. Because the outflow tract is localized in the uppermost area of the heart, the center (the minimum), from which excitation moves away, is present in an uppermost area of the isopotential map at the early QRS period, and with the spread of excitation toward the apex, it moves slightly downward. The remaining time of the minimum in the uppermost area of the isopotential map is considered to be longer in PVC or VT, which originates from a site of the RV further up. Therefore, the remaining time of the minimum is shorter in PVCs that occur at a lower site apart from the pulmonic valve than in PVCs originating from sites further up, that is, the proximal site below the pulmonic valve.
The rapid movement of the minimum or the rapid spread of the negative-potential area from the upper anterior chest to the middle to low anterior chest makes a small r and large s wave in leads V1 and V2. Origins from the sites further up cause the extended presence of positive-potential areas in the middle anterior chest including leads V1 and V2, resulting in the higher and wider initial r wave.
Localization of VT Origins From LVOT
It is known that repetitive monomorphic VT can arise from the LVOT.15 16 However, a detailed ECG method for differentiating the LVOT origin from the RVOT origin has not been reported to date, though R-wave transition in the precordial leads appears to be useful.16
We demonstrated that the BSM is useful for differentiation. The downward moving time of the minimum potential was the longest (>80 ms) in PVCs originating around the aortic valvular cusp. The reason why more time is required in PVC from the aortic valve area than in PVC from the pulmonic valve area despite the anatomically slightly inferior location of the aortic valve compared with the pulmonic valve is probably that it takes much more time for the excitation to break through the thick LV outflow wall to reach the epicardial surface.
The ECG index of R/S amplitude ratio ≥1 in lead V3 was deducted from the longest downward moving time of the minimum potential. We did not show a method for localizing the origin of VT from various subdivisions of LVOT because the number of patients was small (n=5). Further study is required to assess the subdivisions from which VT arises and to determine which origins of VT are suitable for ablation.
The diagnostic accuracy of these ECG indexes did not differ from that of the BSM at every location (Tables 1⇑ and 2⇑). This indicates that the 12-lead ECG is still as reliable a method as BSM for the localization of the VT origin.
The ECG indexes were mainly guided by PVCs. One reason that we selected the late coupled PVC or the first beat of VT as a standard wave is that the QRS morphology of VTs and PVCs changes markedly, depending on the rate or the coupling interval.17 18 Another reason is that the clinical VT is not always inducible during the ablation procedure in many patients with nonreentrant VT from the RVOT. We had reported the superiority of the slow rate pace mapping guided by PVC compared with the rapid rate pace mapping guided by VT for successful ablation in this type of VT.17 Because our criteria are based mainly on ECG findings during PVC, attention should be paid when the ablation site is predicted by ECG during rapid VT.
The origin of PVC or VT was determined under fluoroscopy. The free-wall and septal location defined by fluoroscopy may not reflect the actual anatomic subdivision, although our definition seems feasible because most of the present VT of septal origin showed a shorter QRS duration (≤140 ms). Further studies with other anatomic classifications or other parameters on body surface electrocardiograms are required for more accurate localization.
- Received March 5, 1998.
- Revision received June 1, 1998.
- Accepted June 9, 1998.
- Copyright © 1998 by American Heart Association
Morady F, Kadish AH, DiCarlo L, Kou WH, Winston S, deBuitlier M, Calkins H, Rosenheck S, Sousa J. Long-term results of catheter ablation of idiopathic right ventricular tachycardia. Circulation. 1990;82:2093–2099.
Klein LS, Shih Ht, Hacken FK, Zipes DP, Miles WM. Radiofrequency catheter ablation of ventricular tachycardia in patients without structural heart disease. Circulation. 1992;85:1666–1674.
Kamakura S, Shimomura K, Ohe T, Matsuhisa M, Toyoshima H. The role of initial minimum potentials on body surface maps in predicting the site of accessory pathways in patients with Wolff-Parkinson-White syndrome. Circulation. 1986;74:89–96.
McAlpine WA. The right ventricle. In: Heart and coronary arteries. New York: Springer-Verlag Berlin Heidelberg; 1975: 65–86.
Buxton AE, Waxman LH, Marchrinski FE, Simson MB, Cassidy D, Josephson ME. Right ventricular tachycardia: clinical and electrophysiologic characteristics. Circulation. 1983;68:357–371.
Sung RJ, Keung EC, Nguyen NX, Huycke C. Effects of β-adrenergic blockade on verapamil-responsive and verapamil-irresponsive sustained ventricular tachycardia. J Clin Invest. 1983;81:688–699.
Wilber DJ, Baerman J, Olshansky B, Kall J, Kopp D. Adenosine-sensitive ventricular tachycardia: clinical characteristics and response to catheter ablation. Circulation. 1994;87:126–134.
Kamakura S, Aihara N, Matsuhisa M, Ohe T, Sato I, Shimomura K. The role of initial minimum potentials on body surface maps in localizing the earliest endocardial site of ectopic ventricular excitation. Circulation. 1988;78(suppl II):II-138. Abstract.
SippensGroenewegen A, Spekhorst H, van Hemel NM, Kingma JH, Hauer RNW, Janse MJ, Dunning AJ. Body surface mapping of ectopic left and right ventricular activation. QRS spectrum in patients without structural heart disease. Circulation. 1990;82:879–896.
SippensGroenewegen A, Spekhorst H, van Hemel NM, Kingma JH, Hauer RNW, de Bakker JMT, Grimbergen CA, Janse MJ, Dunning AJ. Localization of the site of origin of postinfarction ventricular tachycardia by endocardial pace mapping: body surface mapping compared with the 12-lead electrocardiogram. Circulation. 1993;88(part 1):2290–2306.
Shimizu W, Kamakura S, Aihara N, Kurita T, Emori T, Katayama K, Suyama K, Shimomura K. Difficult cases of radiofrequency catheter ablation for idiopathic ventricular tachycardia with left bundle branch block morphology and inferior axis [in Japanese]. Jpn J Cardiac Pacing Electrophysiol. 1995;11:341–345.
Kamakura S, Shimizu W, Okano Y, Suyama K, Kurita T, Aihara N, Ohe T, Shimomura K. The usefulness of pace mapping for localizing the successful ablation site in patients with idiopathic right ventricular tachycardia [in Japanese]. Electrocardiology. 1996;16:153–162.
Goyal R, Harvey M, Daoud EG, Brinkman K, Knight BP, Bahu M, Weiss R, Bogun F, Man KC, Strickberger A, Morady F. Effect of coupling interval and pacing cycle length on morphology of paced ventricular complexes; implications for pace mapping. Circulation. 1996;94:2843–2849.Characteristics of the body surface maps and 12-lead ECGs of patients with idiopathic ventricular tachycardia arising from the right ventricular or left ventricular outflow tract were investigated. The optimum ablation site could be localized accurately before catheter ablation according to the 3-dimensional classification consisting of anterior-posterior, right-left, and superior-inferior directions of the outflow tract with the use of the potential distribution of QRS isopotential maps or the QRS wave configuration of the 12-lead ECG.