- Email: [email protected]
Electrogram Characteristics in Postinfarction Ventricular Tachycardia
Journal of the American College of Cardiology © 2005 by the American College of Cardiology Foundation Published by Elsevier Inc.
Vol. 46, No. 4, 2005 ISSN 0735-1097/05/$30.00 doi:10.1016/j.jacc.2005.01.064
Heart Rhythm Disorders
Electrogram Characteristics in Postinfarction Ventricular Tachycardia Effect of Infarct Age Frank Bogun, MD,* Subramaniam Krishnan, MD,† Mukarram Siddiqui, MD,† Eric Good, DO,* Joseph E. Marine, MD,† Claudio Schuger, MD,† Hakan Oral, MD,* Aman Chugh, MD,* Frank Pelosi, MD,* Fred Morady, MD* Ann Arbor and Detroit, Michigan The purpose of this study was to correlate infarct age with characteristics of the endocardial electrograms (EGM) obtained in patients undergoing mapping procedures for postinfarction ventricular tachycardia (VT). BACKGROUND Experimental studies have demonstrated that infarct age influences EGM duration in the subepicardial left ventricle (LV). The relationship between infarct age and endocardial EGM characteristics has not been investigated in patients with postinfarction VT. METHODS In a consecutive series of 23 patients with a history of remote infarction (range 1 to 31 years) and VT, endocardial LV mapping was performed with an electroanatomical mapping system (CARTO, Biosense Webster Inc., Diamond Bar, California) during sinus rhythm. Electrogram morphology and width were analyzed and correlated with infarct age. Isthmus sites of the VT re-entry circuits were identified by entrainment mapping and related to the results of substrate mapping. RESULTS There was a significant correlation between infarct age and width of the bipolar endocardial EGM during baseline rhythm in the peri-infarct zone (r ⫽ 0.84; p ⬍ 0.0001). Increasing infarct age was associated with progressive activation delays in the scar and with isolated potentials separated by an isoelectric interval, the duration of which also correlated with infarct age (r ⫽ 0.77; p ⬍ 0.001). Among all endocardial sites, the VT isthmus sites displayed the most delay and broadest EGMs during sinus rhythm. CONCLUSIONS The presence of broad, fractionated EGMs and isolated potentials indicates a healed myocardial infarction; the older the infarction, the broader the EGM width. Remodeling over time alters the electrophysiologic properties of the peri-infarct tissue. (J Am Coll Cardiol 2005;46:667–74) © 2005 by the American College of Cardiology Foundation OBJECTIVES
Myocardial infarction (MI) results in deposition of fibrous tissue and ventricular remodeling (1). Fibrous tissue invades the surviving myocardium and separates muscle bundles, resulting in slowing of impulse propagation (2). The electrophysiologic correlate of this process See page 675 is fractionation and broadening of the local bipolar electrogram (EGM) and the formation of isolated potentials separated from the main ventricular EGM by an isoelectric segment (3). In a canine infarct model, there was progressive widening of EGMs recorded in the infarct zone with increasing age of the infarct (3). The effects of infarct age and remodeling on EGM width and on the arrhythmogenic substrate have not been described in humans. From the *University of Michigan Medical Center, Ann Arbor, Michigan; and the †Henry Ford Hospital, Division of Cardiology, Detroit, Michigan. Drs. Morady and Oral are consultants for Biosense Webster Inc. (Diamond Bar, California). Manuscript received October 11, 2004; revised manuscript received January 6, 2005, accepted January 25, 2005.
To elucidate potential arrhythmogenic effects of postinfarction remodeling, this study analyzed the relationship of infarct age to EGM characteristics and to the mapping data obtained during ablation of ventricular tachycardia (VT).
METHODS Characteristics of subjects. Twenty-three consecutive patients (mean age 68 ⫾ 7 years, mean ejection fraction 0.21 ⫾ 0.11) referred for VT ablation underwent left ventricular (LV) endocardial mapping during baseline rhythm. All patients had a remote history of MI, with the first infarction being anterior in nine patients and inferior in nine patients. In five patients, the date of the infarction could not be determined by history. Ten patients subsequently developed infarctions in other vessel territories (six with prior anterior infarction then had an inferior MI, and four with prior inferior infarction then had an anterior infarction), and nine had recurrent infarctions in the same vessel territory. The infarct age was defined as the number of years from the time of the first infarction to the time of the electrophysiology procedure and ranged from 1 to 31 years, with a mean of
668
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
Abbreviations and Acronyms EGM ⫽ electrogram LV ⫽ left ventricle/ventricular MI ⫽ myocardial infarction RV ⫽ right ventricular VT ⫽ ventricular tachycardia
16.2 ⫾ 9.0 years. Sixteen patients received reperfusion and/or revascularization therapy consisting of thrombolysis in three patients, primary angioplasty in four patients, and coronary artery bypass graft surgery in nine patients. The remaining seven patients did not receive revascularization therapy. Coronary angiography revealed single vessel coronary disease in 4 patients, two-vessel disease in 8 patients, and three-vessel disease in 11 patients. In 5 of 23 patients, the infarct age could not be determined on the basis of available clinical information. These patients were not included in the statistical correlation of EGM characteristics and infarct age. Data from these patients were included in the assessment of the duration index at isthmus sites and the analysis of morphologic EGM characteristics. Symptomatic VT was first documented 12 ⫾ 9.5 years after the first MI (range 6 months to 29 years). Thirteen patients were being treated with amiodarone at the time of the electrophysiology procedure, five patients were being treated with a combination of amiodarone and procainamide (one patient) or mexiletine (four patients), and one patient was being treated with sotalol. All 23 patients were treated with an angiotensin-convertingenzyme inhibitor or angiotensin receptor blocker. Eight patients were being treated with spironolactone, and 21 patients were being treated with a beta-blocker at the time of the electrophysiology study. Sequence of data acquisition. Programmed right ventricular (RV) stimulation was performed to assess the inducibility of the clinical VT and other tachycardias. An electroanatomical map of the LV then was constructed, and areas with broad EGMs were marked. With the ablation catheter positioned at the site with the broadest EGM during sinus rhythm, VT was induced by programmed stimulation, and radiofrequency energy was delivered if the site was within the re-entry circuit. The catheter was then moved to sites with narrower EGMs. Electrophysiology study. The electrophysiology procedures were performed in the fasting state after informed consent was obtained. A quadripolar catheter with 5-mm inter-electrode spacing was advanced to the RV apex or RV outflow tract and programmed stimulation was performed with up to four extrastimuli at two basic drive cycle lengths. In 22 of 23 patients, LV access was obtained with a retrograde aortic approach. A transseptal approach was used in one patient because of bilateral iliac artery occlusions. Once the catheter was positioned in the arterial system, a bolus of 5,000 U of heparin was administered followed by
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
an infusion of 1,000 U/h to maintain an activated clotting time of approximately 250 s. Mapping and ablation. A three-dimensional mapping system (CARTO, Biosense Webster Inc., Diamond Bar, California) was used to create a voltage map during the baseline rhythm (sinus rhythm in 22 patients and RV pacing in 1 patient) with a Navistar catheter (Biosense Webster Inc.). Bipolar EGMs were recorded between the 4-mm tip electrode and a 2-mm ring electrode separated by 1 mm. The bipolar EGMs were filtered at 10 to 30 to 400 to 500 Hz. The endocardial sites with the broadest ventricular EGMs and the EGMs with the latest activation during sinus rhythm were sought. Electrogram width was measured with electronic calipers from the onset to the offset of the ventricular EGM (Fig. 1A) or, if an isolated potential was present, to the end of the isolated potential (Fig. 1B). The delay of the EGM was measured from the beginning of the QRS complex to the end of the ventricular EGM or, if an isolated potential was present, to the end of the isolated potential. In the presence of an isolated potential, the duration of the isoelectric segment separating the ventricular EGM from the isolated potential was measured from the end of the ventricular EGM to the beginning of the isolated potential. The local bipolar EGMs recorded by the distal electrode pair of the mapping catheter were categorized according to previously reported criteria (4): 1. Normal EGMs: sharp biphasic or triphasic spikes with amplitudes ⱖ3 mV and duration ⬍70 ms, and/or amplitude/duration ratio ⬎0.046 (Fig. 1A). 2. Fractionated EGMs: amplitude ⱕ0.5 mV, duration ⱖ133 ms, and/or amplitude/duration ratio ⬍0.005 (Fig. 1C). 3. Isolated potential: a potential separated from the ventricular EGM by an isoelectric segment (Fig. 1B), and/or a segment with low amplitude noise (⬍0.05 mV) of ⬎20 ms duration at a gain of 40 to 80 mm/mV. 4. The remaining EGMs were defined as abnormal (Fig. 1D). In a porcine chronic infarction model, Callans et al. (5) found that the area of contiguous bipolar EGMs of ⱕ1 mV correlated well with the infarct size determined by pathological analysis. Therefore, we defined low-voltage segments as areas with an amplitude ⬍1 mV. With a user-defined map, each endocardial site was color-coded with respect to the EGM width (Figs. 2A and 3A). The areas covered by EGM durations ⬍133 ms, 133 to 160 ms, 160 to 200 ms, and ⬎200 ms were measured (with rectangular, triangular, and trapezoid shapes covering the area of interest) and compared with the low-voltage area (Figs. 2B, 2C, 3B, and 3C). These EGM duration categories were chosen arbitrarily. Target sites for radiofrequency ablation were identified with standard mapping criteria described in prior studies (6,7). In addition, to test the hypothesis that the broadest EGMs obtained during sinus rhythm play a critical role in the substrate of inducible VT, the ablation catheter was
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
Figure 1. (A) Shown are recordings from surface leads I, aVF, and V1, and intracardiac recordings from the mapping catheter electrogram (EGM) and the right ventricular apex (RVA). The local EGM that is recorded by the mapping catheter is normal. The amplitude is 7.8 mV, and the width is 65 ms. (B) The tracings are analogous to the tracings in panel A. The local EGM recorded by the mapping catheter displays an isolated potential (oblique arrow) that is separated from the ventricular EGM by an isoelectric line of 210 ms. The EGM amplitude is 0.12 mV, and the EGM width is 357 ms. (C) The tracings are again analogous to the tracings in panel A. The recorded EGM is fractionated; the amplitude is 0.37 mV, and the width is 192 ms. (D) The tracings are again analogous to the tracings in panel A. The local EGM is abnormal. The amplitude is 0.7 mV, and the width of the EGM is 112 ms.
positioned at sites displaying the widest EGMs during sinus rhythm, and VT was induced by RV programmed stimulation. Radiofrequency energy was delivered if concealed entrainment was demonstrated at that site. Isthmus sites were defined as sites where there was concealed entrainment and where radiofrequency energy delivery resulted in termination of VT and the inability to re-induce the VT. The EGM width at isthmus sites was measured during sinus rhythm and compared with the maximal EGM width obtained during endocardial LV mapping. The EGM duration index was defined as the ratio
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
669
Figure 2. (A) This figure displays a user-defined map of a view of the septal aspect of the left ventricle (LV) and shows the LV outflow tract and part of the mitral valve annulus (MVA). The map is color-coded according to the measured electrogram (EGM) width. Three EGMs are displayed. Two of them have isolated potential (sites #1 and #3; isolated potentials are indicated by arrows) and represent the widest EGMs identified in this patient. Sites at which fragmented EGMs without isolated potentials are recorded separate them (site #2). The different EGM widths are indicated below the EGMs. Areas with isolated potentials are marked in blue, areas with fragmented electrograms are marked in white, and the area where the His was recorded is indicated as orange dots. The map was obtained in a patient who had an inferior wall myocardial infarction 30 years before the mapping study. (B) The scheme represents the same view as outlined in panel A. The cross-hatched area corresponds to areas with low-voltage EGMs and the black area represents the area where EGMs are broader than 200 ms. Areas in white correspond to sites with higher voltage EGMs (⬎1 mV). (C) This figure illustrates a voltage map of the same view shown in panels A and B. Low-voltage areas are differentiated from higher voltage areas.
670
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
of EGM width at a particular site to the maximal EGM duration in that patient during sinus rhythm. The EGM duration index was calculated at isthmus sites in order to assess the relationship of isthmus sites to areas with the maximally delayed activation during sinus rhythm. Statistical analysis. Continuous variables were indicated with the mean ⫾ standard deviation. Continuous variables were compared with Student t test, and categorical variables were compared with chi-square analysis or the Fisher exact test. Pearson correlation coefficient was calculated and indicated as r and p values. A p value of ⬍0.05 was considered significant.
RESULTS
Figure 3. (A) This figure shows a user-defined map of the infero-posterior wall of the left ventricle (LV), including the LV apex and the mitral valve annulus (MVA). As in Figure 2A, the electrogram (EGM) width is colorcoded. The patient had an inferior myocardial infarction one year before the mapping study. The EGM shown is the broadest endocardial EGM obtained in this patient and represents the site of effective ablation of the targeted ventricular tachycardia during sinus rhythm. The EGM width is 150 ms; an isoelectric segment of 26 ms separates the ventricular EGM from the isolated potential (arrow). (B) The scheme represents a view of the same area outlined in panel A. The striped area corresponds to areas with low-voltage EGMs, and the black area represents the contiguous area with an EGM width of ⬎133 ms. The white area correspond to sites with higher voltage electrograms (⬎1 mV). Black area ⫽ EMG ⬎133 ms; striped area ⫽ EGMs ⬍1 mV; white area ⫽ EGMs ⬎1 mV. (C) This figure illustrates the voltage map of the same view outlined in panels A and B in this patient, delineating the low-voltage areas from areas with higher voltage (⬎1 mV).
Electrophysiology study and mapping data. A total of 115 distinct VTs (mean cycle length 362 ⫾ 92 ms) were inducible by RV programmed stimulation (5 ⫾ 3 VTs per patient) (Table 1, Fig. 4). Forty-three VTs (mean cycle length 428 ⫾ 85 ms; range: 280 to 580 ms) were reproducibly inducible. In 32 of 43 VTs (74%), an isthmus site could be identified and the targeted VT was successfully ablated. A mean of 164 ⫾ 56 endocardial points (range 62 to 305 points) were recorded to reconstruct the LV geometry with the electro-anatomical mapping system. The low-voltage area (⬍1.0 mV) covered 63 ⫾ 35 cm2 of the LV endocardium. Contiguous areas with an EGM duration ⬎133 ms covered 19 ⫾ 18 cm2, comprising 37 ⫾ 31% of the low-voltage area. An EGM duration ⬎160 ms covered an area of 6.7 ⫾ 8.0 cm2, comprising 12 ⫾ 11% of the low-voltage area. Areas at which the EGM duration was ⬎200 ms covered 2.7 ⫾ 1.7 cm2, comprising 5 ⫾ 4% of the low-voltage area. In 1 of 23 patients, the maximal EGM width was ⬍160 ms. In 3 of 23 patients the maximal EGM duration was ⬍200 ms. Of the EGMs that had a duration ⬎133 ms, 95% had a low-voltage amplitude (598 of 630 EGMs; mean amplitude: 0.34 ⫾ 0.33 mV); 96% of EGMs ⬎160 ms had a low amplitude (342 of 358 EGMs; mean amplitude: 0.3 ⫾ 0.28 mV); and 97% of EGMs with a width ⬎200 ms had a low-voltage amplitude (178 of 183 EGMs; mean amplitude: 0.28 ⫾ 0.26 mV). Infarct age and mapping data. There was a significant positive correlation between the infarct age and the duration of the broadest endocardial EGM detected during sinus rhythm mapping (r ⫽ 0.84; p ⬍ 0.0001) (Table 1, Fig. 5). There also was a strong correlation between infarct age and maximal delay of local endocardial activation after the QRS complex during sinus rhythm (mean 318 ⫾ 91 ms; range 143 to 494 ms) in the peri-infarct zone (r ⫽ 0.80; p ⬍ 0.001). The isoelectric segment separating the ventricular EGM from isolated potentials was longer (mean 170 ⫾ 93 ms; range 29 to 344 ms) the older the infarction (r ⫽ 0.77; p ⬍ 0.001). The duration index at an isthmus site obtained during sinus rhythm was 0.9 ⫾ 0.12. The maximal EGM duration per patient did not correlate
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
671
Table 1. Demographic and Mapping Characteristics of Patients Patient No.
Age (yrs)
EF
MI Location
MIs
Duration Index of Clinical VT at Isthmus Site
Yrs After 1st MI
Max. EGM Width (ms)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
60 77 63 70 73 65 70 61 65 62 78 56 69 84 72 66 67 57 57 64 65 47 40
15 10 10 10 40 30 25 42 25 35 20 20 30 20 15 15 15 20 10 10 15 25 10
A⫹I A A⫹I A⫹I I A I I A⫹I I I A⫹I I I A⫹I A A⫹I I A⫹I A A⫹I A A⫹I
3 2 2 3 1 uk 1 1 2 2 2 uk 3 1 1 2 1 uk uk 1 2 2 2
0.99 0.99 1 1 1 0.99 0.99 N/A 1 N/A N/A 1 N/A 1 N/A 1 1 1 N/A N/A N/A N/A N/A
21 13 20 15 1 uk 2 30 uk 15 12 uk 26 8 20 20 22 uk uk 23 10 5 8
301 290 343 351 150 450 205 487 465 341 237 400 440 345 425 332 308 184 270 464 282 188 223
A ⫽ anterior infarction; EF ⫽ ejection fraction; I ⫽ inferior infarction; MI Location ⫽ localization of myocardial infarction; Max. EGM Width ⫽ maximal width of recorded electrograms; N/A ⫽ not applicable; uk ⫽ unknown; VT ⫽ ventricular tachycardia.
with the ejection fraction (r ⫽ 0.13; p ⫽ 0.7) or with baseline QRS duration of the patients (r ⫽ 0.07, p ⫽ 0.8). The maximal EGM duration was similar in the 16 patients who did undergo revascularization after infarction as compared with those that did not (315 ⫾ 92 ms vs. 367 ⫾ 113 ms; p ⫽ 0.3). There was no correlation between EGM duration at an isthmus site and VT cycle length for the VTs in which an isthmus could be identified (r ⫽ 0.1, p ⫽ 0.7). The number of MIs did not correlate with the maximal EGM duration (r ⫽ 0.17; p ⫽ 0.5). Furthermore, the maximal EGM duration was similar, whether or not patients were on amiodarone at the time of the mapping procedure (349 ⫾ 97 ms vs. 361 ⫾ 105 ms; p ⫽ 0.8). The
areas of EGM duration ⬎133 ms, ⬎160 ms, and ⬎200 ms did not correlate with infarct age (r ⫽ 0.1, ⫽ 0.03, ⫽ 0.1; p ⫽ 0.7, ⫽ 0.9, ⫽ 0.9, respectively). Infarct location did not correlate with infarct age or maximal EGM duration. In patients with prior anterior infarctions as the first infarction, the maximal EGM width was 316 ⫾ 21 ms vs. 313 ⫾ 114 ms in patients with prior inferior wall infarction (p ⫽ 0.9). Infarct age was 18 ⫾ 5 years in patients with prior anterior infarction at the time of the mapping procedure and 13 ⫾ 11 years in patients with prior inferior infarction (p ⫽ 0.18). EGM characteristics. Contiguous EGMs with a duration ⬎200 ms contained isolated potentials in 86% of cases (Tables 2 and 3, Fig. 4). Sites with isolated potentials were
Figure 4. This figure shows the distribution of electrogram (EGM) types into different electrogram width categories: ⱕ133, ⬎133, ⬎160, and ⬎200 ms. The EGM types include normal electrograms (nl EGM), abnormal electrograms (abn EGM), fractionated electrograms (frag EGM), and EGMs with isolated potentials (IP).
Figure 5. Shown is the maximal width of the local electrogram (EGM) obtained in each patient plotted against the infarct age. There is a significant association between infarct age and maximal EGM width (R ⫽ 0.84).
672
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
Table 2. Electrogram Characteristics According to Electrogram Width Number of EGMs Mean width (ms) Amplitude (mV) Normal EGMs Abnormal EGM Fractionated EGMs IP EGMs Isthmus sites
<133 ms
133–160 ms
160–200 ms
>200 ms
1,532 90 ⫾ 25 1.6 ⫾ 2.0 231/1,532 (15%) 879/1,532 (57%) 381/1,532 (25%) 41/1,532 (3%) 0/32
273 145 ⫾ 54 0.41 ⫾ 0.38 1/273 (0.4%) 18/273 (7%) 175/273 (64%) 79/273 (29%) 1/32
171 178 ⫾ 12 0.32 ⫾ 0.3 0/171 (0%) 5/171 (3%) 89/171 (52%) 77/171 (45%) 1/32
187 270 ⫾ 72 0.28 ⫾ 0.26 0/187 (0%) 2/187 (1%) 24/187 (13%) 161/187 (86%) 28/32
EGM ⫽ electrogram (at 2 of 32 isthmus sites the EGM width could not be determined); IP ⫽ isolated potential.
significantly broader and had significantly smaller amplitude compared with sites with normal, abnormal, or fractionated EGMs. The vast majority of sites with an isolated potential were within the low-voltage area. All but one of the isthmus site EGMs fell into the area containing contiguous EGMs with a duration of ⬎200 ms. Among the 23 patients, 39 contiguous areas bounded by EGMs with a duration of ⬎200 ms could be identified (2.3 ⫾ 1.5 per patient).
postinfarction model of Gardner et al. (3), isolated potentials were related to two separated myocardial strands that each gave rise to a single biphasic waveform, the muscle strands being separated by fibrous tissue. Furthermore, there was, over time, a gradual increase of nonmuscular tissue mainly composed of fibrous tissue when controls were compared with subacute, healing, and healed infarcts (3). This was most pronounced in areas where fractionated EGMs were recorded. In the present clinical study, all of the patients had a remote MI. The relationship between infarct age and EGM width in the peri-infarct zone suggests that there may be an ongoing process of collagenous fiber deposition for many years after infarction, resulting in progressive separation of individual muscle bundles and a gradual increase in maximal EGM width. The areas bounded by broad EGMs, however, did not correlate with infarct age and probably depend on variables other than only time. Postinfarct remodeling and the substrate of VT. The infarct scar is a dynamic tissue with continued collagen turnover for many years after the initial infarction (8). Continued collagen deposition is considered to be one of the major consequences of ventricular remodeling in patients with ischemic cardiomyopathy (9). There is evidence that prevention and attenuation of LV remodeling decreases the occurrence of ventricular arrhythmias and/or sudden death (10,11). Although our data do not directly prove the relationship between remodeling and ventricular arrhythmias, they support a possible role of the remodeling process in creating the substrate for re-entrant VT. The disruption of side-to-side connections between surviving myocardial strands secondary to fibrosis results in slowing of conduction due to a zig-zag course of activation (2); a milieu that promotes re-entry is thereby created. Consistent with this process, this
DISCUSSION Main findings. In patients with post-infarction VT, there was a significant correlation between infarct age and the maximal duration of endocardial EGMs in the peri-infarct zone during sinus rhythm. Critical isthmus sites in the VT re-entry circuit were likely to be located at sites with the latest activation during sinus rhythm. Most critical isthmus sites displayed EGMs that had a duration ⬎200 ms and isolated potentials. These findings are consistent with postinfarction remodeling that results in progressive electrophysiological effects in the anatomical substrate for VT. Furthermore, isolated potentials may identify surviving muscle bundles within the infarct scar that can be localized and targeted for ablation during sinus rhythm. Postinfarct remodeling and EGM characteristics. Experiments with superfused subacute, healing, and healed canine infarct hearts have demonstrated a relationship between infarct age and EGM characteristics (3). This study demonstrates that the same type of relationship occurs in postinfarction patients who develop VT. The anatomical basis of fractionated EGMs has been described in human postinfarct papillary muscles that have been harvested after cardiac transplantation. The higher the fractionation of recorded EGMs, the more that surviving myocardial muscle tracts separated by collagenous tissue were identified in the recording area (2). In the canine Table 3. Characteristics of Bipolar Electrograms
Number EGM amplitude (mV) EGM width (ms) Amplitude/duration
Normal EGM
Abnormal EGM
Fractionated EGM
IP
232 5.16 ⫾ 2.9* 58 ⫾ 13* 0.1 ⫾ 0.68*
905 1.28 ⫾ 0.72* 91 ⫾ 23* 0.02 ⫾ 0.01*
669 0.31 ⫾ 0.23 131 ⫾ 38* 0.002 ⫾ 0.003
357 0.35 ⫾ 0.34 208 ⫾ 80* 0.001 ⫾ 0.001
*p ⬍ 0.01, for individual group comparisons one against another. EGM ⫽ electrogram; IP ⫽ isolated potential.
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
study showed that critical isthmus sites of VT re-entry circuits were identified at areas showing a high EGM duration index. A relationship between areas with isolated potentials and ventricular arrhythmias has been proposed in earlier studies (12). This relationship was confirmed in this study. All of the critical isthmus sites that were identified displayed an isolated potential during sinus rhythm and relatively broad EGMs. Histologic analysis of sites with isolated potentials has demonstrated the presence of surviving myocardial bundles surrounded by collagenous tissue (3), making isolated potentials an EGM correlate of the remodeling process. Wilber et al. (13) reported that in the Multicenter Automatic Defibrillator Implantation Trial (MADIT II), the mortality risk increased as a function of time after MI. Our study provides a possible explanation for an increase in potentially lethal arrhythmias in a time-dependent manner after an MI. Postinfarction remodeling and collagen fiber deposition is a time-dependent process that eventually leads to the extent of complex anisotropy necessary for postinfarction VT. We cannot exclude a time-dependent effect of other factors on the appearance of postinfarction VT, including remodeling with compensatory hypertrophy (14,15), mechanical stretch (11,16), and alterations in the autonomous nervous system (17,18), for example. The latter factors most likely could initiate VT secondary to abnormal automaticity or triggered activity, but would be less likely to set the stage for re-entry as the only mechanism. Study limitations. This study has several limitations. First, the study was a cross-sectional analysis. A longitudinal analysis with serial mapping procedures over several years would be the preferred method for analyzing the relationship between infarct age and EGM characteristics. Second, although there was no correlation between EGM characteristics and the number of recurrent MIs, the additional infarctions also might have affected the remodeling process. A third limitation is that the width of the EGM at sites with isolated potentials included both the isolated potential and the ventricular EGMs. It is possible that these EGMs, at times, included both far-field and near-field components. In analyzing broad, complex EGMs that are often composed of multiple high-frequency deflections with isolated potentials separated by long isoelectric intervals, we chose to define them as single EGMs. Because of the potential limitations of this definition of maximal EGM width, however, additional measurements (such as the length of the isoelectric interval between the ventricular EGM and isolated potentials, and the activation delay between the onset of the QRS complex and isolated potentials) were performed. These other measures of EGM width also correlated with infarct age, thereby validating the relationship between maximal EGM width and infarct age. This study did not include a control group of postinfarction patients without inducible monomorphic VT. Therefore, whether the EGM characteristics found in this
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
673
study are specific to the substrate of re-entrant VT is unknown. The LV reconstructions were only partial, because not all possible endocardial sites were sampled. Furthermore, although the data point density was highest in areas of scar, it is possible that points with even later EGMs might have been missed. Lastly, it is possible that changes in the treatment of acute MI, over time, might have had an influence on the remodeling process and served as a confounding variable. Conclusions. In this study of patients with post-infarction VT, there was a significant correlation between infarct age and EGM width. Postinfarction ventricular remodeling with continued collagenous fiber deposition most likely accounted for these time-dependent changes in peri-infarct endocardial EGM characteristics. Critical isthmus sites for post-infarction VT are composed of the broadest EGMs mainly displaying isolated potentials with an EGM width of ⬎200 ms. Isolated potentials identify surviving myocardial bundles within the infarct scar and may be useful in identifying the critical substrate of post-infarction VT during sinus rhythm. Reprint requests and correspondence: Dr. Frank Bogun, Division of Cardiology, University of Michigan Health System, 3119 TC, 1500 E. Medical Center Dr., Ann Arbor, Michigan 481090366. E-mail: [email protected].
REFERENCES 1. Jugdutt BI, Amy RW. Healing after myocardial infarction in the dog: changes in infarct hydroxyproline and topography. J Am Coll Cardiol 1986;7:91–102. 2. de Bakker JM, van Capelle FJ, Janse MJ, et al. Slow conduction in the infarcted human heart. “Zigzag” course of activation. Circulation 1993;88:915–26. 3. Gardner PI, Ursell PC, Fenoglio JJ Jr., Wit AL. Electrophysiologic and anatomic basis for fractionated electrograms recorded from healed myocardial infarcts. Circulation 1985;72:596 – 611. 4. Josephson M. Clinical Cardiac Electrophysiology Techniques and Interpretations. Malvern, PA: Lea & Febiger, 1993. 5. Callans DJ, Ren JF, Michele J, Marchlinski FE, Dillon SM. Electroanatomic left ventricular mapping in the porcine model of healed anterior myocardial infarction. Correlation with intracardiac echocardiography and pathological analysis. Circulation 1999;100:1744 –50. 6. Bogun F, Bahu M, Knight B, et al. Comparison of effective and ineffective target sites that demonstrate concealed entrainment in patients with coronary artery disease undergoing radiofrequency ablation of ventricular tachycardia. Circulation 1997;95:183–90. 7. Bogun F, Bahu M, Knight B, et al. Response to pacing at sites of isolated diastolic potentials during ventricular tachycardia in patients with previous myocardial infarction. J Am Coll Cardiol 1997;30:505–13. 8. Willems IE, Havenith MG, De Mey JG, Daemen MJ. The alphasmooth muscle actin-positive cells in healing human myocardial scars. Am J Pathol 1994;145:868 –75. 9. Beltrami CA, Finato N, Rocco M, et al. Structural basis of end-stage failure in ischemic cardiomyopathy in humans. Circulation 1994;89: 151– 63. 10. Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study Investigators. N Engl J Med 1999; 341:709 –17.
674
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
11. Sogaard P, Gotzsche CO, Ravkilde J, Norgaard A, Thygesen K. Ventricular arrhythmias in the acute and chronic phases after acute myocardial infarction. Effect of intervention with captopril. Circulation 1994;90:101–7. 12. Klein H, Karp RB, Kouchoukos NT, Zorn GL Jr., James TN, Waldo AL. Intraoperative electrophysiologic mapping of the ventricles during sinus rhythm in patients with a previous myocardial infarction. Identification of the electrophysiologic substrate of ventricular arrhythmias. Circulation 1982;66:847–53. 13. Wilber DJ, Zareba W, Hall WJ, et al. Time dependence of mortality risk and defibrillator benefit after myocardial infarction. Circulation 2004;109:1082– 4. 14. Pahor M, Bernabei R, Sgadari A, et al. Enalapril prevents cardiac fibrosis and arrhythmias in hypertensive rats. Hypertension 1991;18:148 –57.
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74 15. Chevalier B, Heudes D, Heymes C, et al. Trandolapril decreases prevalence of ventricular ectopic activity in middle-aged SHR. Circulation 1995;92:1947–53. 16. White HD, Norris RM, Brown MA, Brandt PW, Whitlock RM, Wild CJ. Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction. Circulation 1987;76:44 –51. 17. Meredith IT, Broughton A, Jennings GL, Esler MD. Evidence of a selective increase in cardiac sympathetic activity in patients with sustained ventricular arrhythmias. N Engl J Med 1991;325:618 –24. 18. Sabbah HN, Goldberg AD, Schoels W, et al. Spontaneous and inducible ventricular arrhythmias in a canine model of chronic heart failure: relation to haemodynamics and sympathoadrenergic activation. Eur Heart J 1992;13:1562–72.
Vol. 46, No. 4, 2005 ISSN 0735-1097/05/$30.00 doi:10.1016/j.jacc.2005.01.064
Heart Rhythm Disorders
Electrogram Characteristics in Postinfarction Ventricular Tachycardia Effect of Infarct Age Frank Bogun, MD,* Subramaniam Krishnan, MD,† Mukarram Siddiqui, MD,† Eric Good, DO,* Joseph E. Marine, MD,† Claudio Schuger, MD,† Hakan Oral, MD,* Aman Chugh, MD,* Frank Pelosi, MD,* Fred Morady, MD* Ann Arbor and Detroit, Michigan The purpose of this study was to correlate infarct age with characteristics of the endocardial electrograms (EGM) obtained in patients undergoing mapping procedures for postinfarction ventricular tachycardia (VT). BACKGROUND Experimental studies have demonstrated that infarct age influences EGM duration in the subepicardial left ventricle (LV). The relationship between infarct age and endocardial EGM characteristics has not been investigated in patients with postinfarction VT. METHODS In a consecutive series of 23 patients with a history of remote infarction (range 1 to 31 years) and VT, endocardial LV mapping was performed with an electroanatomical mapping system (CARTO, Biosense Webster Inc., Diamond Bar, California) during sinus rhythm. Electrogram morphology and width were analyzed and correlated with infarct age. Isthmus sites of the VT re-entry circuits were identified by entrainment mapping and related to the results of substrate mapping. RESULTS There was a significant correlation between infarct age and width of the bipolar endocardial EGM during baseline rhythm in the peri-infarct zone (r ⫽ 0.84; p ⬍ 0.0001). Increasing infarct age was associated with progressive activation delays in the scar and with isolated potentials separated by an isoelectric interval, the duration of which also correlated with infarct age (r ⫽ 0.77; p ⬍ 0.001). Among all endocardial sites, the VT isthmus sites displayed the most delay and broadest EGMs during sinus rhythm. CONCLUSIONS The presence of broad, fractionated EGMs and isolated potentials indicates a healed myocardial infarction; the older the infarction, the broader the EGM width. Remodeling over time alters the electrophysiologic properties of the peri-infarct tissue. (J Am Coll Cardiol 2005;46:667–74) © 2005 by the American College of Cardiology Foundation OBJECTIVES
Myocardial infarction (MI) results in deposition of fibrous tissue and ventricular remodeling (1). Fibrous tissue invades the surviving myocardium and separates muscle bundles, resulting in slowing of impulse propagation (2). The electrophysiologic correlate of this process See page 675 is fractionation and broadening of the local bipolar electrogram (EGM) and the formation of isolated potentials separated from the main ventricular EGM by an isoelectric segment (3). In a canine infarct model, there was progressive widening of EGMs recorded in the infarct zone with increasing age of the infarct (3). The effects of infarct age and remodeling on EGM width and on the arrhythmogenic substrate have not been described in humans. From the *University of Michigan Medical Center, Ann Arbor, Michigan; and the †Henry Ford Hospital, Division of Cardiology, Detroit, Michigan. Drs. Morady and Oral are consultants for Biosense Webster Inc. (Diamond Bar, California). Manuscript received October 11, 2004; revised manuscript received January 6, 2005, accepted January 25, 2005.
To elucidate potential arrhythmogenic effects of postinfarction remodeling, this study analyzed the relationship of infarct age to EGM characteristics and to the mapping data obtained during ablation of ventricular tachycardia (VT).
METHODS Characteristics of subjects. Twenty-three consecutive patients (mean age 68 ⫾ 7 years, mean ejection fraction 0.21 ⫾ 0.11) referred for VT ablation underwent left ventricular (LV) endocardial mapping during baseline rhythm. All patients had a remote history of MI, with the first infarction being anterior in nine patients and inferior in nine patients. In five patients, the date of the infarction could not be determined by history. Ten patients subsequently developed infarctions in other vessel territories (six with prior anterior infarction then had an inferior MI, and four with prior inferior infarction then had an anterior infarction), and nine had recurrent infarctions in the same vessel territory. The infarct age was defined as the number of years from the time of the first infarction to the time of the electrophysiology procedure and ranged from 1 to 31 years, with a mean of
668
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
Abbreviations and Acronyms EGM ⫽ electrogram LV ⫽ left ventricle/ventricular MI ⫽ myocardial infarction RV ⫽ right ventricular VT ⫽ ventricular tachycardia
16.2 ⫾ 9.0 years. Sixteen patients received reperfusion and/or revascularization therapy consisting of thrombolysis in three patients, primary angioplasty in four patients, and coronary artery bypass graft surgery in nine patients. The remaining seven patients did not receive revascularization therapy. Coronary angiography revealed single vessel coronary disease in 4 patients, two-vessel disease in 8 patients, and three-vessel disease in 11 patients. In 5 of 23 patients, the infarct age could not be determined on the basis of available clinical information. These patients were not included in the statistical correlation of EGM characteristics and infarct age. Data from these patients were included in the assessment of the duration index at isthmus sites and the analysis of morphologic EGM characteristics. Symptomatic VT was first documented 12 ⫾ 9.5 years after the first MI (range 6 months to 29 years). Thirteen patients were being treated with amiodarone at the time of the electrophysiology procedure, five patients were being treated with a combination of amiodarone and procainamide (one patient) or mexiletine (four patients), and one patient was being treated with sotalol. All 23 patients were treated with an angiotensin-convertingenzyme inhibitor or angiotensin receptor blocker. Eight patients were being treated with spironolactone, and 21 patients were being treated with a beta-blocker at the time of the electrophysiology study. Sequence of data acquisition. Programmed right ventricular (RV) stimulation was performed to assess the inducibility of the clinical VT and other tachycardias. An electroanatomical map of the LV then was constructed, and areas with broad EGMs were marked. With the ablation catheter positioned at the site with the broadest EGM during sinus rhythm, VT was induced by programmed stimulation, and radiofrequency energy was delivered if the site was within the re-entry circuit. The catheter was then moved to sites with narrower EGMs. Electrophysiology study. The electrophysiology procedures were performed in the fasting state after informed consent was obtained. A quadripolar catheter with 5-mm inter-electrode spacing was advanced to the RV apex or RV outflow tract and programmed stimulation was performed with up to four extrastimuli at two basic drive cycle lengths. In 22 of 23 patients, LV access was obtained with a retrograde aortic approach. A transseptal approach was used in one patient because of bilateral iliac artery occlusions. Once the catheter was positioned in the arterial system, a bolus of 5,000 U of heparin was administered followed by
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
an infusion of 1,000 U/h to maintain an activated clotting time of approximately 250 s. Mapping and ablation. A three-dimensional mapping system (CARTO, Biosense Webster Inc., Diamond Bar, California) was used to create a voltage map during the baseline rhythm (sinus rhythm in 22 patients and RV pacing in 1 patient) with a Navistar catheter (Biosense Webster Inc.). Bipolar EGMs were recorded between the 4-mm tip electrode and a 2-mm ring electrode separated by 1 mm. The bipolar EGMs were filtered at 10 to 30 to 400 to 500 Hz. The endocardial sites with the broadest ventricular EGMs and the EGMs with the latest activation during sinus rhythm were sought. Electrogram width was measured with electronic calipers from the onset to the offset of the ventricular EGM (Fig. 1A) or, if an isolated potential was present, to the end of the isolated potential (Fig. 1B). The delay of the EGM was measured from the beginning of the QRS complex to the end of the ventricular EGM or, if an isolated potential was present, to the end of the isolated potential. In the presence of an isolated potential, the duration of the isoelectric segment separating the ventricular EGM from the isolated potential was measured from the end of the ventricular EGM to the beginning of the isolated potential. The local bipolar EGMs recorded by the distal electrode pair of the mapping catheter were categorized according to previously reported criteria (4): 1. Normal EGMs: sharp biphasic or triphasic spikes with amplitudes ⱖ3 mV and duration ⬍70 ms, and/or amplitude/duration ratio ⬎0.046 (Fig. 1A). 2. Fractionated EGMs: amplitude ⱕ0.5 mV, duration ⱖ133 ms, and/or amplitude/duration ratio ⬍0.005 (Fig. 1C). 3. Isolated potential: a potential separated from the ventricular EGM by an isoelectric segment (Fig. 1B), and/or a segment with low amplitude noise (⬍0.05 mV) of ⬎20 ms duration at a gain of 40 to 80 mm/mV. 4. The remaining EGMs were defined as abnormal (Fig. 1D). In a porcine chronic infarction model, Callans et al. (5) found that the area of contiguous bipolar EGMs of ⱕ1 mV correlated well with the infarct size determined by pathological analysis. Therefore, we defined low-voltage segments as areas with an amplitude ⬍1 mV. With a user-defined map, each endocardial site was color-coded with respect to the EGM width (Figs. 2A and 3A). The areas covered by EGM durations ⬍133 ms, 133 to 160 ms, 160 to 200 ms, and ⬎200 ms were measured (with rectangular, triangular, and trapezoid shapes covering the area of interest) and compared with the low-voltage area (Figs. 2B, 2C, 3B, and 3C). These EGM duration categories were chosen arbitrarily. Target sites for radiofrequency ablation were identified with standard mapping criteria described in prior studies (6,7). In addition, to test the hypothesis that the broadest EGMs obtained during sinus rhythm play a critical role in the substrate of inducible VT, the ablation catheter was
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
Figure 1. (A) Shown are recordings from surface leads I, aVF, and V1, and intracardiac recordings from the mapping catheter electrogram (EGM) and the right ventricular apex (RVA). The local EGM that is recorded by the mapping catheter is normal. The amplitude is 7.8 mV, and the width is 65 ms. (B) The tracings are analogous to the tracings in panel A. The local EGM recorded by the mapping catheter displays an isolated potential (oblique arrow) that is separated from the ventricular EGM by an isoelectric line of 210 ms. The EGM amplitude is 0.12 mV, and the EGM width is 357 ms. (C) The tracings are again analogous to the tracings in panel A. The recorded EGM is fractionated; the amplitude is 0.37 mV, and the width is 192 ms. (D) The tracings are again analogous to the tracings in panel A. The local EGM is abnormal. The amplitude is 0.7 mV, and the width of the EGM is 112 ms.
positioned at sites displaying the widest EGMs during sinus rhythm, and VT was induced by RV programmed stimulation. Radiofrequency energy was delivered if concealed entrainment was demonstrated at that site. Isthmus sites were defined as sites where there was concealed entrainment and where radiofrequency energy delivery resulted in termination of VT and the inability to re-induce the VT. The EGM width at isthmus sites was measured during sinus rhythm and compared with the maximal EGM width obtained during endocardial LV mapping. The EGM duration index was defined as the ratio
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
669
Figure 2. (A) This figure displays a user-defined map of a view of the septal aspect of the left ventricle (LV) and shows the LV outflow tract and part of the mitral valve annulus (MVA). The map is color-coded according to the measured electrogram (EGM) width. Three EGMs are displayed. Two of them have isolated potential (sites #1 and #3; isolated potentials are indicated by arrows) and represent the widest EGMs identified in this patient. Sites at which fragmented EGMs without isolated potentials are recorded separate them (site #2). The different EGM widths are indicated below the EGMs. Areas with isolated potentials are marked in blue, areas with fragmented electrograms are marked in white, and the area where the His was recorded is indicated as orange dots. The map was obtained in a patient who had an inferior wall myocardial infarction 30 years before the mapping study. (B) The scheme represents the same view as outlined in panel A. The cross-hatched area corresponds to areas with low-voltage EGMs and the black area represents the area where EGMs are broader than 200 ms. Areas in white correspond to sites with higher voltage EGMs (⬎1 mV). (C) This figure illustrates a voltage map of the same view shown in panels A and B. Low-voltage areas are differentiated from higher voltage areas.
670
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
of EGM width at a particular site to the maximal EGM duration in that patient during sinus rhythm. The EGM duration index was calculated at isthmus sites in order to assess the relationship of isthmus sites to areas with the maximally delayed activation during sinus rhythm. Statistical analysis. Continuous variables were indicated with the mean ⫾ standard deviation. Continuous variables were compared with Student t test, and categorical variables were compared with chi-square analysis or the Fisher exact test. Pearson correlation coefficient was calculated and indicated as r and p values. A p value of ⬍0.05 was considered significant.
RESULTS
Figure 3. (A) This figure shows a user-defined map of the infero-posterior wall of the left ventricle (LV), including the LV apex and the mitral valve annulus (MVA). As in Figure 2A, the electrogram (EGM) width is colorcoded. The patient had an inferior myocardial infarction one year before the mapping study. The EGM shown is the broadest endocardial EGM obtained in this patient and represents the site of effective ablation of the targeted ventricular tachycardia during sinus rhythm. The EGM width is 150 ms; an isoelectric segment of 26 ms separates the ventricular EGM from the isolated potential (arrow). (B) The scheme represents a view of the same area outlined in panel A. The striped area corresponds to areas with low-voltage EGMs, and the black area represents the contiguous area with an EGM width of ⬎133 ms. The white area correspond to sites with higher voltage electrograms (⬎1 mV). Black area ⫽ EMG ⬎133 ms; striped area ⫽ EGMs ⬍1 mV; white area ⫽ EGMs ⬎1 mV. (C) This figure illustrates the voltage map of the same view outlined in panels A and B in this patient, delineating the low-voltage areas from areas with higher voltage (⬎1 mV).
Electrophysiology study and mapping data. A total of 115 distinct VTs (mean cycle length 362 ⫾ 92 ms) were inducible by RV programmed stimulation (5 ⫾ 3 VTs per patient) (Table 1, Fig. 4). Forty-three VTs (mean cycle length 428 ⫾ 85 ms; range: 280 to 580 ms) were reproducibly inducible. In 32 of 43 VTs (74%), an isthmus site could be identified and the targeted VT was successfully ablated. A mean of 164 ⫾ 56 endocardial points (range 62 to 305 points) were recorded to reconstruct the LV geometry with the electro-anatomical mapping system. The low-voltage area (⬍1.0 mV) covered 63 ⫾ 35 cm2 of the LV endocardium. Contiguous areas with an EGM duration ⬎133 ms covered 19 ⫾ 18 cm2, comprising 37 ⫾ 31% of the low-voltage area. An EGM duration ⬎160 ms covered an area of 6.7 ⫾ 8.0 cm2, comprising 12 ⫾ 11% of the low-voltage area. Areas at which the EGM duration was ⬎200 ms covered 2.7 ⫾ 1.7 cm2, comprising 5 ⫾ 4% of the low-voltage area. In 1 of 23 patients, the maximal EGM width was ⬍160 ms. In 3 of 23 patients the maximal EGM duration was ⬍200 ms. Of the EGMs that had a duration ⬎133 ms, 95% had a low-voltage amplitude (598 of 630 EGMs; mean amplitude: 0.34 ⫾ 0.33 mV); 96% of EGMs ⬎160 ms had a low amplitude (342 of 358 EGMs; mean amplitude: 0.3 ⫾ 0.28 mV); and 97% of EGMs with a width ⬎200 ms had a low-voltage amplitude (178 of 183 EGMs; mean amplitude: 0.28 ⫾ 0.26 mV). Infarct age and mapping data. There was a significant positive correlation between the infarct age and the duration of the broadest endocardial EGM detected during sinus rhythm mapping (r ⫽ 0.84; p ⬍ 0.0001) (Table 1, Fig. 5). There also was a strong correlation between infarct age and maximal delay of local endocardial activation after the QRS complex during sinus rhythm (mean 318 ⫾ 91 ms; range 143 to 494 ms) in the peri-infarct zone (r ⫽ 0.80; p ⬍ 0.001). The isoelectric segment separating the ventricular EGM from isolated potentials was longer (mean 170 ⫾ 93 ms; range 29 to 344 ms) the older the infarction (r ⫽ 0.77; p ⬍ 0.001). The duration index at an isthmus site obtained during sinus rhythm was 0.9 ⫾ 0.12. The maximal EGM duration per patient did not correlate
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
671
Table 1. Demographic and Mapping Characteristics of Patients Patient No.
Age (yrs)
EF
MI Location
MIs
Duration Index of Clinical VT at Isthmus Site
Yrs After 1st MI
Max. EGM Width (ms)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
60 77 63 70 73 65 70 61 65 62 78 56 69 84 72 66 67 57 57 64 65 47 40
15 10 10 10 40 30 25 42 25 35 20 20 30 20 15 15 15 20 10 10 15 25 10
A⫹I A A⫹I A⫹I I A I I A⫹I I I A⫹I I I A⫹I A A⫹I I A⫹I A A⫹I A A⫹I
3 2 2 3 1 uk 1 1 2 2 2 uk 3 1 1 2 1 uk uk 1 2 2 2
0.99 0.99 1 1 1 0.99 0.99 N/A 1 N/A N/A 1 N/A 1 N/A 1 1 1 N/A N/A N/A N/A N/A
21 13 20 15 1 uk 2 30 uk 15 12 uk 26 8 20 20 22 uk uk 23 10 5 8
301 290 343 351 150 450 205 487 465 341 237 400 440 345 425 332 308 184 270 464 282 188 223
A ⫽ anterior infarction; EF ⫽ ejection fraction; I ⫽ inferior infarction; MI Location ⫽ localization of myocardial infarction; Max. EGM Width ⫽ maximal width of recorded electrograms; N/A ⫽ not applicable; uk ⫽ unknown; VT ⫽ ventricular tachycardia.
with the ejection fraction (r ⫽ 0.13; p ⫽ 0.7) or with baseline QRS duration of the patients (r ⫽ 0.07, p ⫽ 0.8). The maximal EGM duration was similar in the 16 patients who did undergo revascularization after infarction as compared with those that did not (315 ⫾ 92 ms vs. 367 ⫾ 113 ms; p ⫽ 0.3). There was no correlation between EGM duration at an isthmus site and VT cycle length for the VTs in which an isthmus could be identified (r ⫽ 0.1, p ⫽ 0.7). The number of MIs did not correlate with the maximal EGM duration (r ⫽ 0.17; p ⫽ 0.5). Furthermore, the maximal EGM duration was similar, whether or not patients were on amiodarone at the time of the mapping procedure (349 ⫾ 97 ms vs. 361 ⫾ 105 ms; p ⫽ 0.8). The
areas of EGM duration ⬎133 ms, ⬎160 ms, and ⬎200 ms did not correlate with infarct age (r ⫽ 0.1, ⫽ 0.03, ⫽ 0.1; p ⫽ 0.7, ⫽ 0.9, ⫽ 0.9, respectively). Infarct location did not correlate with infarct age or maximal EGM duration. In patients with prior anterior infarctions as the first infarction, the maximal EGM width was 316 ⫾ 21 ms vs. 313 ⫾ 114 ms in patients with prior inferior wall infarction (p ⫽ 0.9). Infarct age was 18 ⫾ 5 years in patients with prior anterior infarction at the time of the mapping procedure and 13 ⫾ 11 years in patients with prior inferior infarction (p ⫽ 0.18). EGM characteristics. Contiguous EGMs with a duration ⬎200 ms contained isolated potentials in 86% of cases (Tables 2 and 3, Fig. 4). Sites with isolated potentials were
Figure 4. This figure shows the distribution of electrogram (EGM) types into different electrogram width categories: ⱕ133, ⬎133, ⬎160, and ⬎200 ms. The EGM types include normal electrograms (nl EGM), abnormal electrograms (abn EGM), fractionated electrograms (frag EGM), and EGMs with isolated potentials (IP).
Figure 5. Shown is the maximal width of the local electrogram (EGM) obtained in each patient plotted against the infarct age. There is a significant association between infarct age and maximal EGM width (R ⫽ 0.84).
672
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
Table 2. Electrogram Characteristics According to Electrogram Width Number of EGMs Mean width (ms) Amplitude (mV) Normal EGMs Abnormal EGM Fractionated EGMs IP EGMs Isthmus sites
<133 ms
133–160 ms
160–200 ms
>200 ms
1,532 90 ⫾ 25 1.6 ⫾ 2.0 231/1,532 (15%) 879/1,532 (57%) 381/1,532 (25%) 41/1,532 (3%) 0/32
273 145 ⫾ 54 0.41 ⫾ 0.38 1/273 (0.4%) 18/273 (7%) 175/273 (64%) 79/273 (29%) 1/32
171 178 ⫾ 12 0.32 ⫾ 0.3 0/171 (0%) 5/171 (3%) 89/171 (52%) 77/171 (45%) 1/32
187 270 ⫾ 72 0.28 ⫾ 0.26 0/187 (0%) 2/187 (1%) 24/187 (13%) 161/187 (86%) 28/32
EGM ⫽ electrogram (at 2 of 32 isthmus sites the EGM width could not be determined); IP ⫽ isolated potential.
significantly broader and had significantly smaller amplitude compared with sites with normal, abnormal, or fractionated EGMs. The vast majority of sites with an isolated potential were within the low-voltage area. All but one of the isthmus site EGMs fell into the area containing contiguous EGMs with a duration of ⬎200 ms. Among the 23 patients, 39 contiguous areas bounded by EGMs with a duration of ⬎200 ms could be identified (2.3 ⫾ 1.5 per patient).
postinfarction model of Gardner et al. (3), isolated potentials were related to two separated myocardial strands that each gave rise to a single biphasic waveform, the muscle strands being separated by fibrous tissue. Furthermore, there was, over time, a gradual increase of nonmuscular tissue mainly composed of fibrous tissue when controls were compared with subacute, healing, and healed infarcts (3). This was most pronounced in areas where fractionated EGMs were recorded. In the present clinical study, all of the patients had a remote MI. The relationship between infarct age and EGM width in the peri-infarct zone suggests that there may be an ongoing process of collagenous fiber deposition for many years after infarction, resulting in progressive separation of individual muscle bundles and a gradual increase in maximal EGM width. The areas bounded by broad EGMs, however, did not correlate with infarct age and probably depend on variables other than only time. Postinfarct remodeling and the substrate of VT. The infarct scar is a dynamic tissue with continued collagen turnover for many years after the initial infarction (8). Continued collagen deposition is considered to be one of the major consequences of ventricular remodeling in patients with ischemic cardiomyopathy (9). There is evidence that prevention and attenuation of LV remodeling decreases the occurrence of ventricular arrhythmias and/or sudden death (10,11). Although our data do not directly prove the relationship between remodeling and ventricular arrhythmias, they support a possible role of the remodeling process in creating the substrate for re-entrant VT. The disruption of side-to-side connections between surviving myocardial strands secondary to fibrosis results in slowing of conduction due to a zig-zag course of activation (2); a milieu that promotes re-entry is thereby created. Consistent with this process, this
DISCUSSION Main findings. In patients with post-infarction VT, there was a significant correlation between infarct age and the maximal duration of endocardial EGMs in the peri-infarct zone during sinus rhythm. Critical isthmus sites in the VT re-entry circuit were likely to be located at sites with the latest activation during sinus rhythm. Most critical isthmus sites displayed EGMs that had a duration ⬎200 ms and isolated potentials. These findings are consistent with postinfarction remodeling that results in progressive electrophysiological effects in the anatomical substrate for VT. Furthermore, isolated potentials may identify surviving muscle bundles within the infarct scar that can be localized and targeted for ablation during sinus rhythm. Postinfarct remodeling and EGM characteristics. Experiments with superfused subacute, healing, and healed canine infarct hearts have demonstrated a relationship between infarct age and EGM characteristics (3). This study demonstrates that the same type of relationship occurs in postinfarction patients who develop VT. The anatomical basis of fractionated EGMs has been described in human postinfarct papillary muscles that have been harvested after cardiac transplantation. The higher the fractionation of recorded EGMs, the more that surviving myocardial muscle tracts separated by collagenous tissue were identified in the recording area (2). In the canine Table 3. Characteristics of Bipolar Electrograms
Number EGM amplitude (mV) EGM width (ms) Amplitude/duration
Normal EGM
Abnormal EGM
Fractionated EGM
IP
232 5.16 ⫾ 2.9* 58 ⫾ 13* 0.1 ⫾ 0.68*
905 1.28 ⫾ 0.72* 91 ⫾ 23* 0.02 ⫾ 0.01*
669 0.31 ⫾ 0.23 131 ⫾ 38* 0.002 ⫾ 0.003
357 0.35 ⫾ 0.34 208 ⫾ 80* 0.001 ⫾ 0.001
*p ⬍ 0.01, for individual group comparisons one against another. EGM ⫽ electrogram; IP ⫽ isolated potential.
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74
study showed that critical isthmus sites of VT re-entry circuits were identified at areas showing a high EGM duration index. A relationship between areas with isolated potentials and ventricular arrhythmias has been proposed in earlier studies (12). This relationship was confirmed in this study. All of the critical isthmus sites that were identified displayed an isolated potential during sinus rhythm and relatively broad EGMs. Histologic analysis of sites with isolated potentials has demonstrated the presence of surviving myocardial bundles surrounded by collagenous tissue (3), making isolated potentials an EGM correlate of the remodeling process. Wilber et al. (13) reported that in the Multicenter Automatic Defibrillator Implantation Trial (MADIT II), the mortality risk increased as a function of time after MI. Our study provides a possible explanation for an increase in potentially lethal arrhythmias in a time-dependent manner after an MI. Postinfarction remodeling and collagen fiber deposition is a time-dependent process that eventually leads to the extent of complex anisotropy necessary for postinfarction VT. We cannot exclude a time-dependent effect of other factors on the appearance of postinfarction VT, including remodeling with compensatory hypertrophy (14,15), mechanical stretch (11,16), and alterations in the autonomous nervous system (17,18), for example. The latter factors most likely could initiate VT secondary to abnormal automaticity or triggered activity, but would be less likely to set the stage for re-entry as the only mechanism. Study limitations. This study has several limitations. First, the study was a cross-sectional analysis. A longitudinal analysis with serial mapping procedures over several years would be the preferred method for analyzing the relationship between infarct age and EGM characteristics. Second, although there was no correlation between EGM characteristics and the number of recurrent MIs, the additional infarctions also might have affected the remodeling process. A third limitation is that the width of the EGM at sites with isolated potentials included both the isolated potential and the ventricular EGMs. It is possible that these EGMs, at times, included both far-field and near-field components. In analyzing broad, complex EGMs that are often composed of multiple high-frequency deflections with isolated potentials separated by long isoelectric intervals, we chose to define them as single EGMs. Because of the potential limitations of this definition of maximal EGM width, however, additional measurements (such as the length of the isoelectric interval between the ventricular EGM and isolated potentials, and the activation delay between the onset of the QRS complex and isolated potentials) were performed. These other measures of EGM width also correlated with infarct age, thereby validating the relationship between maximal EGM width and infarct age. This study did not include a control group of postinfarction patients without inducible monomorphic VT. Therefore, whether the EGM characteristics found in this
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
673
study are specific to the substrate of re-entrant VT is unknown. The LV reconstructions were only partial, because not all possible endocardial sites were sampled. Furthermore, although the data point density was highest in areas of scar, it is possible that points with even later EGMs might have been missed. Lastly, it is possible that changes in the treatment of acute MI, over time, might have had an influence on the remodeling process and served as a confounding variable. Conclusions. In this study of patients with post-infarction VT, there was a significant correlation between infarct age and EGM width. Postinfarction ventricular remodeling with continued collagenous fiber deposition most likely accounted for these time-dependent changes in peri-infarct endocardial EGM characteristics. Critical isthmus sites for post-infarction VT are composed of the broadest EGMs mainly displaying isolated potentials with an EGM width of ⬎200 ms. Isolated potentials identify surviving myocardial bundles within the infarct scar and may be useful in identifying the critical substrate of post-infarction VT during sinus rhythm. Reprint requests and correspondence: Dr. Frank Bogun, Division of Cardiology, University of Michigan Health System, 3119 TC, 1500 E. Medical Center Dr., Ann Arbor, Michigan 481090366. E-mail: [email protected].
REFERENCES 1. Jugdutt BI, Amy RW. Healing after myocardial infarction in the dog: changes in infarct hydroxyproline and topography. J Am Coll Cardiol 1986;7:91–102. 2. de Bakker JM, van Capelle FJ, Janse MJ, et al. Slow conduction in the infarcted human heart. “Zigzag” course of activation. Circulation 1993;88:915–26. 3. Gardner PI, Ursell PC, Fenoglio JJ Jr., Wit AL. Electrophysiologic and anatomic basis for fractionated electrograms recorded from healed myocardial infarcts. Circulation 1985;72:596 – 611. 4. Josephson M. Clinical Cardiac Electrophysiology Techniques and Interpretations. Malvern, PA: Lea & Febiger, 1993. 5. Callans DJ, Ren JF, Michele J, Marchlinski FE, Dillon SM. Electroanatomic left ventricular mapping in the porcine model of healed anterior myocardial infarction. Correlation with intracardiac echocardiography and pathological analysis. Circulation 1999;100:1744 –50. 6. Bogun F, Bahu M, Knight B, et al. Comparison of effective and ineffective target sites that demonstrate concealed entrainment in patients with coronary artery disease undergoing radiofrequency ablation of ventricular tachycardia. Circulation 1997;95:183–90. 7. Bogun F, Bahu M, Knight B, et al. Response to pacing at sites of isolated diastolic potentials during ventricular tachycardia in patients with previous myocardial infarction. J Am Coll Cardiol 1997;30:505–13. 8. Willems IE, Havenith MG, De Mey JG, Daemen MJ. The alphasmooth muscle actin-positive cells in healing human myocardial scars. Am J Pathol 1994;145:868 –75. 9. Beltrami CA, Finato N, Rocco M, et al. Structural basis of end-stage failure in ischemic cardiomyopathy in humans. Circulation 1994;89: 151– 63. 10. Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study Investigators. N Engl J Med 1999; 341:709 –17.
674
Bogun et al. Postinfarct Left Ventricular Remodeling and VT
11. Sogaard P, Gotzsche CO, Ravkilde J, Norgaard A, Thygesen K. Ventricular arrhythmias in the acute and chronic phases after acute myocardial infarction. Effect of intervention with captopril. Circulation 1994;90:101–7. 12. Klein H, Karp RB, Kouchoukos NT, Zorn GL Jr., James TN, Waldo AL. Intraoperative electrophysiologic mapping of the ventricles during sinus rhythm in patients with a previous myocardial infarction. Identification of the electrophysiologic substrate of ventricular arrhythmias. Circulation 1982;66:847–53. 13. Wilber DJ, Zareba W, Hall WJ, et al. Time dependence of mortality risk and defibrillator benefit after myocardial infarction. Circulation 2004;109:1082– 4. 14. Pahor M, Bernabei R, Sgadari A, et al. Enalapril prevents cardiac fibrosis and arrhythmias in hypertensive rats. Hypertension 1991;18:148 –57.
JACC Vol. 46, No. 4, 2005 August 16, 2005:667–74 15. Chevalier B, Heudes D, Heymes C, et al. Trandolapril decreases prevalence of ventricular ectopic activity in middle-aged SHR. Circulation 1995;92:1947–53. 16. White HD, Norris RM, Brown MA, Brandt PW, Whitlock RM, Wild CJ. Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction. Circulation 1987;76:44 –51. 17. Meredith IT, Broughton A, Jennings GL, Esler MD. Evidence of a selective increase in cardiac sympathetic activity in patients with sustained ventricular arrhythmias. N Engl J Med 1991;325:618 –24. 18. Sabbah HN, Goldberg AD, Schoels W, et al. Spontaneous and inducible ventricular arrhythmias in a canine model of chronic heart failure: relation to haemodynamics and sympathoadrenergic activation. Eur Heart J 1992;13:1562–72.