Value of body surface mapping in localizing the site of origin of ventricular tachycardia in patients with previous myocardial infarction.

OBJECTIVES: This study examined the performance of the 62-lead body surface electrocardiogram (ECG) in identifying the site of origin of ventricular tachycardia in patients with a previous myocardial infarction. BACKGROUND: Because the accuracy of ECG localization of ventricular tachycardia using standard 12-lead recordings is restricted to the identification of rather large ventricular areas, application of multiple torso lead recordings may augment the resolving power of the surface ECG and result in more discrete localization of arrhythmogenic foci. METHODS: Thirty-two patients were selected for electrophysiologically guided ablative therapy for drug-resistant postinfarction ventricular tachycardia. In these patients, QRS integral maps of distinct monomorphic ventricular tachycardia configurations were correlated with a previously generated infarct-specific reference data base of paced QRS integral maps. Each paced pattern in the data base corresponded with ectopic endocardial impulse formation at 1 of 18 or 22 discrete segments of the left ventricle with a previous anterior or inferior myocardial infarction, respectively. Electrocardiographic localization was compared with the results obtained during intraoperative or catheter endocardial activation sequence mapping. RESULTS: Body surface mapping was performed during 101 distinct ventricular tachycardia configurations. Compared with the activation mapping data that were acquired in 64 of 101 ventricular tachycardias, body surface mapping identified the correct segment of origin in 40 (62%) of 64 tachycardias, a segment adjacent to the segment where the arrhythmia actually originated in 19 (30%) of 64 tachycardias and a segment disparate from the actual segment of origin in 5 (8%) of 64 tachycardias. With respect to infarct location, the segment of origin was correctly identified in 28 (60%) of 47 ventricular tachycardias in patients with anterior, 7 (70%) of 10 tachycardias in patients with inferior and 5 (71%) of 7 tachycardias in patients with combined anterior and inferior myocardial infarction. CONCLUSIONS: This study shows that body surface mapping enables precise localization of the origin of postinfarction ventricular tachycardia in 62% and regional approximation in 30% of tachycardias. The multiple-lead ECG may be used to guide and shorten catheter-based mapping procedures during ventricular tachycardia and to provide relevant information on the origin of tachycardias that cannot be mapped with conventional single-site mapping techniques because of unfavorable characteristics.

A quantitative method for the localization of the ventricular pre-excitation area in the Wolff-Parkinson-White syndrome using singular value decomposition of body surface potentials.

In order to localize quantitatively the site of ventricular pre-excitation, singular value decomposition (SVD) was applied to the body surface potential distributions of patients with the Wolff-Parkinson-White syndrome. The body surface potentials of sixty-two patients were recorded during sinus rhythm and pre-excitation by means of thirty electrodes placed on the anterior thoracic wall. The sites of the anomalous bundles had been determined beforehand by multicatheter electrophysiologic study. Considerable data reduction was obtained by using the SVD technique and displaying the potential distribution during the delta wave on two isofunction maps of the first two positional vectors and their corresponding two singular values (SV). A distinction was made between two types of isofunction maps. A type-S (single extreme) and a type-D (double extremes). A quantitative analysis was performed with the orientation of the zero line on the isofunction map being represented by the angle alpha or beta, and the singular values quotient (SVQ) of the two first singular values. The angle beta belonging to type D was used to subdivide this group of pre-excitation areas. The parameters SVQ and alpha belonging to type-S were illustrated in a graph on which a distinction between the various locations of the pre-excitation areas can be seen.

Adenosine induced ventricular arrhythmias in the emergency room.

While adenosine effectively terminates most supraventricular tachycardias (SVT), rare case reports have demonstrated its proarrhythmic potential, including induction of ventricular tachycardia (VT). The aim of this study was to define the proarrhythmic effects of adenosine in a large, unselected population. During a 5-year period, adenosine was used (average dose 9.7 mg) in the emergency room to manage 187 episodes of tachycardia in 127 patients. In two thirds of the cases, adenosine induced ventricular ectopy following successful termination of SVT, including premature ventricular complexes (PVC) and nonsustained VT. The adenosine induced PVCs and VT were transient and self-terminating. More than half had a right bundle branch block morphology with a superior axis that suggested an origin in the inferior left ventricular septum. In conclusion, although adenosine is commonly used in clinical practice to treat SVTs, we found that it induced PVCs and VT in two thirds of the patients. The high incidence of ventricular arrhythmias following adenosine infusion was surprising but did not require further intervention. These arrhythmias appeared to frequently originate from the inferior left ventricular septum, suggesting that this area may be particularly susceptible to the proarrhythmic effects of adenosine.

Body surface mapping during percutaneous transluminal coronary angioplasty. QRS changes indicating regional myocardial conduction delay.

Using a radiotransparent electrode array, body surface maps (BSMs) were constructed based on simultaneous recordings from 62 leads on the entire thorax before, during, and after balloon inflation during percutaneous transluminal coronary angioplasty (PTCA). Twenty-five patients were studied, and 30 angioplasties were performed; 20 patients had one-vessel disease, and five patients had two-vessel disease. In total, 15 dilations in the left anterior descending artery (LAD), seven in the right coronary artery (RCA), and eight in the left circumflex artery (LCx) were studied. For each patient, the BSM and the QRS integral map before, during, and after the inflation was compared by subtraction of recordings "during-minus-before" inflation and "before-minus-after" inflation. The subtraction was performed on the results of the QRS integral maps. The conclusions derived from the inspection of the BSMs and the difference maps show specific changes in the QRS complex during ischemia related to the corresponding ischemic segment in 21 of 25 patients in the three groups. An area of positive potentials remained present on the BSM during dilation, indicating a depolarization wave front. For the LAD group, positive potentials were seen on the anterior thorax and, for the RCA group, on the lower part of the thorax. By subtraction analysis, these changes were extracted and presented as difference maps. For the LCx group, the BSM revealed no changes in pattern but the difference map showed a difference vector pointing in a anteroposterior direction. A regional myocardial conduction delay was hypothesized as the most likely cause for the results.

Body surface mapping of ectopic left ventricular activation. QRS spectrum in patients with prior myocardial infarction.

To improve electrocardiographic localization of the site of origin of ectopic left ventricular (LV) impulse formation in the heart with prior myocardial infarction, 62-lead body surface QRS integral maps were studied during LV pacing at a total of 221 endocardial sites in 14 patients with previous anterior (AMI), inferior (IMI), lateral (LMI), or anterior and inferior (AMI/IMI) myocardial infarction. The anatomic location of each pacing site was computed using digitized biplane fluoroscopic images and plotted on standardized LV endocardial polar projections. A data base of characteristic AMI and IMI mean QRS integral maps was developed after visually selecting subgroups with nearly identical QRS integral morphology from the ectopic activation sequences produced at 110 sites in eight patients with AMI and at 66 sites in four patients with IMI. Intrasubgroup pattern uniformity and intersubgroup pattern variability were statistically verified. The endocardial pacing site locations belonging to each AMI and IMI subgroup were depicted as segments on the respective LV polar projections. In patients with AMI, a total of 18 typical mean QRS integral patterns were obtained, whereas 22 different mean total QRS integral patterns showing more substantial intersubgroup variation were acquired in patients with IMI. Posterolateral regions exhibited a relatively low electrocardiographic sensitivity (six AMI and five IMI patterns) as compared with anteroseptal regions (12 AMI and 17 IMI patterns). Total QRS integral patterns obtained at 24 sites in one patient with LMI were largely compatible with the IMI mean total QRS integral patterns, whereas the majority of total QRS integral patterns acquired at 21 sites in one patient with AMI/IMI corresponded with the AMI mean total QRS integral patterns. The results show that total body surface QRS integral maps generated during LV pacing in patients with prior myocardial infarction cluster by pattern and that each QRS integral pattern is related to a circumscribed endocardial segment of ectopic impulse formation. The relation between a given QRS integral pattern and the position and size of the corresponding paced segment is dependent on infarct location. The present infarct-specific data base of characteristic total body surface QRS integral patterns provides a clinical tool to obtain detailed electrocardiographic localization of ventricular arrhythmias in patients with previous myocardial infarction.

Body surface mapping of ectopic left and right ventricular activation. QRS spectrum in patients without structural heart disease.

The value of simultaneous 62-lead electrocardiographic recordings in localizing the site of origin of ectopic ventricular activation in a structurally normal heart was assessed by examining body surface QRS integral maps in 12 patients during left and right ventricular (LV and RV) pacing at 182 distinct endocardial sites. A data base of 38 characteristic mean integral maps was composed after visually selecting subgroups with nearly identical total QRS integral morphology and numerically evaluating intrasubgroup pattern uniformity and intersubgroup pattern variability. Corresponding endocardial pacing site locations were computed by a biplane cineradiographic method and outlined as segments on a standardized LV and RV polar projection. LV pacing resulted in 25 markedly different mean total QRS integral patterns, showing higher electrocardiographic sensitivity for anteroseptal (18 patterns) compared with posterolateral regions (seven patterns). RV pacing demonstrated 13 mean total QRS integral patterns, exhibiting less intersubgroup variation and comparatively low electrocardiographic sensitivity for the basal anterior and outflow regions. Comparison of LV with RV pacing revealed that QRS configurations produced at LV apical and LV midseptal sites closely resembled QRS configurations generated at RV apical, RV septal, and RV anterior sites, respectively. Total QRS time integral amplitudes showed considerable intrasubgroup variation but permitted global differentiation of spatially similar QRS patterns obtained during pacing at LV and RV sites. This study demonstrates that the QRS pattern of the total body surface electrocardiogram allows discrimination among 38 different LV and RV segments of ectopic endocardial impulse formation in patients with normal cardiac anatomy.

Localization of the site of origin of postinfarction ventricular tachycardia by endocardial pace mapping. Body surface mapping compared with the 12-lead electrocardiogram.

BACKGROUND: The purpose of this study was to assess the value of body surface mapping and the standard 12-lead ECG in localizing the site of origin of postinfarction ventricular tachycardia (VT) during endocardial pace mapping of the left ventricle. METHODS AND RESULTS: Simultaneous recordings of 62-lead body surface QRS integral maps and scalar 12-lead ECG tracings were obtained in 16 patients with prior myocardial infarction during a total of 26 distinct VT configurations and during subsequent left ventricular catheter pace mapping at 9 to 24 different endocardial sites. Anatomic pacing site locations were computed by means of a biplane cineradiographic method and plotted on a polar projection of the left ventricle. The QRS integral map and the QRS complexes of the 12 standard leads of each VT morphology obtained in a particular patient were compared independently with the different paced QRS integral maps and paced QRS complexes of the 12-lead ECG generated in that same patient. The stimulus site locations of the best matching paced QRS integral map and paced QRS complexes of the 12-lead ECG were indicated on the polar projection and subsequently compared with the endocardial location of the corresponding site of VT origin identified during intraoperative (surgical ablation) or catheter activation sequence mapping (catheter ablation). The localization resolution of pace mapping was established separately for each electrocardiographic technique by computing the size of endocardial areas with similar morphological features of the QRS complex. Pace mapping advocated with body surface mapping or the 12-lead ECG enabled adequate reproduction of the VT QRS morphology in 24 of 26 VTs (92%) and 25 of 26 VTs (96%), respectively. Activation sequence mapping identified the site of origin in 12 of 26 previously observed VT configurations (46%). Ten and 11 VTs were localized by activation sequence mapping and pace mapping combined with body surface mapping or the 12-lead ECG, respectively. Pace mapping applied with body surface mapping identified the site of origin correctly (distance < or = 2 cm) in 8 of 10 compared VTs (80%); an adjacent site (distance between 2 and 4 cm) or a disparate site (distance > or = 4 cm) was identified in the remaining 2 of 10 VTs (20%). Pace mapping used with the 12-lead ECG localized the site of origin correctly in 2 of 11 VTs (18%); the site of origin was identified correctly next to an additional adjacent site in 5 of 11 VTs (55%); and an adjacent site or a disparate site was found in 1 of 11 VTs (9%) and 2 of 11 VTs (18%), respectively. The difference in localization accuracy of both electrocardiographic techniques was statistically significant (P = .02). The mean size of endocardial areas where a comparable QRS morphology was obtained during pace mapping was 6.0 +/- 4.5 cm2 with the application of body surface mapping and 15.1 +/- 12.0 cm2 with the use of the 12-lead ECG. CONCLUSIONS: These results demonstrate that application of the 62-lead instead of the 12-lead ECG during endocardial pace mapping enhances the localization resolution of this mapping technique and enables more precise identification of the site of arrhythmogenesis in the majority of compared postinfarction VT episodes.