Catheter Ablation in Patients With Multiple and Unstable Ventricular Tachycardias After Myocardial Infarction
Short Ablation Lines Guided by Reentry Circuit Isthmuses and Sinus Rhythm Mapping
Background— Extensive lines of radiofrequency (RF) lesions through infarct (MI) can ablate multiple and unstable ventricular tachycardias (VTs). Methods for guiding ablation that minimize unnecessary RF applications are needed. This study assesses the feasibility of guiding RF line placement by mapping to identify a reentry circuit isthmus.
Methods and Results— Catheter mapping and ablation were performed in 40 patients (MI location: inferior, 28; anterior, 7; and both, 5) with an electroanatomic mapping system to measure the infarct region and ablation lines. The initial line was placed in the MI region either through a circuit isthmus identified from entrainment mapping or a target identified from pace mapping. A total of 143 VTs (42 stable, 101 unstable) were induced. An isthmus was identified in 25 patients (63%; 5 with only stable VTs, 5 with only unstable VTs, and 15 with both VTs). Inducible VTs were abolished or modified in 100% of patients when the RF line included an isthmus compared with 53% when RF had to be guided by pace mapping (P=0.0002); those with an isthmus identified received shorter ablation lines (4.9±2.4 versus 7.4±4.3 cm total length, P=0.02). During follow-up, spontaneous VT decreased markedly regardless of whether an isthmus was identified. VT stability and number of morphologies did not influence outcome.
Conclusions— A 4- to 5-cm line of RF lesions abolishes all inducible VTs in more than 50% of patients. Less ablation is required if a reentry circuit isthmus is identified even when multiple and unstable VTs are present.
Received March 9, 2001; revision received May 23, 2001; accepted May 25, 2001.
Patients with ventricular tachycardia (VT) after infarction often have multiple reentry circuits with an average of 3 to 4 different inducible VTs.1 In addition, induced VTs are often unstable, preventing extensive mapping to localize all VT circuits. VT can be unstable for mapping because of hemodynamic collapse, frequent changes from one VT to another, or inability to reproducibly induce the VT. Recently, Marchlinski and colleagues2 demonstrated that an extensive set of ablation lesions in or around areas of scar during sinus rhythm can ablate multiple and unstable VTs. The extent of radiofrequency (RF) ablation lines required for success, however, is unknown. It is theoretically desirable to limit the length of RF lines to the minimum required for success, minimizing the risk of damage to functioning myocardium and the creation of potentially thrombogenic endocardial lesions. However, for ablation over a small area to be effective, precise mapping to identify narrow isthmuses in the VT circuit is required.1 Although precise mapping is often not possible when VT is unstable, isthmuses can still be confirmed in some patients by finding potential target areas during sinus rhythm and then assessing the region only briefly during induced VT.3 The purpose of this study is to (1) assess the feasibility of guiding RF line placement by limited mapping to attempt to identify reentry circuit isthmuses, (2) relate the length of RF ablation lines to the effect on inducible VTs, and (3) determine whether the presence of multiple morphologies of inducible VT and unstable VTs reduces the efficacy of these approaches.
Between October 1997 and June 2000, a total of 60 patients were referred for RF ablation of monomorphic VT associated with myocardial infarction. Patients with multiple morphologies of VT or unstable VT were not excluded. Data from 40 consecutive patients (38 men; age 66.6±10.9 years; MI location: 28 inferior, 7 anterior, and 5 both; ejection fraction 28.9±10.0%) who had catheter ablation with an electroanatomic mapping system (CARTO, Biosense Webster, Inc) for construction of ventricular electrogram voltage maps were reviewed. All patients gave informed consent, and procedures were performed according to protocols approved by the Brigham and Women’s Hospital Human Subject Protection Committee.
Our methods have been described previously.1 Ventricular mapping was performed with 7F steerable catheters with either a 4- or 3.5-mm electrode tip (Navi Star or Thermo-Cool, Biosense Webster). Bipolar electrograms were recorded on the electroanatomic mapping system (filtered at 10 to 400 Hz) and a separate digital system (filtered at 30 to 500 Hz; Prucka Engineering Inc). Pace mapping and entrainment mapping used unipolar pacing from the distal electrode with an initial current strength of 10 mA and pulse width of 2 ms.1
The sequence of mapping and ablation is shown in Figure 1. The left ventricle was mapped during sinus rhythm to identify the low-voltage (<1.5 mV) area of infarction.2 The catheter was then placed at a region where pace mapping produced a QRS morphology similar to a previously induced VT and/or where the S-QRS interval exceeded 40 ms, indicative of conduction delay.4 VT was then initiated by programmed stimulation. Entrainment was performed at the site, followed immediately by application of the first RF lesion if the site was a reentry circuit isthmus, defined as entrainment with concealed fusion with an S-QRS interval <70% of the VT cycle length and a postpacing interval—VT cycle length difference ≤30 ms.1 If the site was not an isthmus and VT was stable, further mapping continued during VT. If an isthmus was not identified for a stable VT, other reentry circuit sites, such as outer or inner loops,1 were targeted. If the induced VT was unstable, RF ablation, rapid pacing, or cardioversion terminated it. Further mapping was then performed during sinus rhythm, and the initial ablation region was selected based on the pace-mapping QRS morphology and conduction delay (S-QRS interval >40 ms).1,4 If the initial VT was incessant (8 patients), the sinus rhythm map was obtained after VT was terminated by ablation; ablation sites were excluded from the subsequent voltage map.
After the initial ablation target site was selected, additional RF lesions were applied during sinus rhythm, extending approximately parallel to the border zone of the infarct over 1 to 2 cm until pacing at 10 mA at 2-ms stimulus strength failed to capture in that region. If the target site was within 2 to 3 cm of the mitral annulus, lesions were extended to the mitral annulus to interrupt a potential submitral isthmus.5,6 Ablation lesions were placed only in areas of low voltage (<1.5 mV).
After completion of the initial set of RF lesions, programmed stimulation (1 to 3 extrastimuli after a 600-ms and then 400-ms basic drive from the right ventricular apex and outflow tract) was repeated. If any monomorphic VT was inducible, the mapping and ablation process was repeated with extension of the initial RF line unless entrainment of the new VT demonstrated that the region was not involved in the VT circuit.
Ablation lesions were created with RF current (250 or 500 kHz) delivered for up to 2 minutes between the distal electrode of the mapping catheter and a cutaneous adhesive electrode with a maximum power of 50 W (EP Technologies. Inc or Stockhert GMBH). In 32 patients, ablation was performed with a 4-mm-tip electrode with power titrated to a 5- to 10-Ω decrease in impedance (21 patients) or a maximum temperature of 60°C to 65°C (11 patients). In 8 patients, ablation was performed with an investigational saline-irrigated 3.5-mm-tip electrode (Thermo-cool, Cordis-Webster), with power titrated to a 5- to 10-Ω decrease in impedance and maximum temperature of 45°C. In 4 patients in whom VT remained inducible and originated from a region where an ablation line had been completed with a 4-mm standard ablation catheter, additional lesions were applied along the line with another saline-irrigated RF ablation catheter (Chilli Cool Catheter, Cardiac Pathways) with up to 50 W titrated to a temperature of 36°C to 40°C with a decrease in impedance of 5 to 10Ω.7 These lesions, which could not be tagged on the electroanatomic mapping system for measurement, were not included in the area and length calculations but are described in more detail below.
The procedure ended when no monomorphic VT was inducible, hemodynamically stable VT was inducible but the VT circuit could not be found on the endocardium, or only hemodynamically untolerated VT that was faster than any of the previously tolerated or spontaneous VTs was inducible. The acute outcome of the ablation procedure was defined as follows: success—no monomorphic VT was inducible; modified—monomorphic VT was inducible but was different and faster than VTs induced at the beginning of the procedure; and failure—VT inducible at the beginning of the procedure remained inducible.
After ablation, antiarrhythmic drugs were discontinued in 15 patients. In 25 patients, previously ineffective drugs were continued either as required by a study protocol (8 patients), because the patient had been receiving amiodarone chronically without toxicity (12 patients), or because a VT remained inducible (5 patients). All patients received anticoagulation with warfarin for more than 1 month or aspirin chronically. Follow-up was obtained in July 2000 after a median of 260 days (range 16 to 945). No patient was lost to follow-up.
The infarct was considered to be an ellipse. Area was calculated as: πr1 r2 (r1 indicates longer radius; r2, shorter radius). Circumference was calculated as 2π SQRT (r12+r22/2). The total length of all RF ablation lines was measured as the linear distance between the ends of each line and by summing all lines in an individual patient. Procedure time was the interval from entry into the electrophysiology laboratory to patient transport out of the room.
Continuous data are expressed as mean±1 SD. Groups were compared with Fisher’s exact test or ANOVA as appropriate. Survival analyses were performed with the Kaplan-Meier method and compared with log-rank analysis. Multivariate logistic regression analysis was performed with SAS (Statistical Software version 6.12, SAS Institute).
The 40 patients had 143 different VTs (average of 3.6±2.1 VTs per patient, range 1 to 8) inducible with a mean cycle length of 391±71 ms (range 263 to 690 ms). All VTs were stable and allowed mapping in 7 (17.5%) of 40 patients; all VTs were unstable in 13 patients (32.5%); both stable and unstable VTs were present in 20 patients (50%). Sinus rhythm maps were defined from an average of 121.3±52.1 points (range 48 to 262 points) per patient. Left ventricular infarcts (<1.5 mV) had an area of 38.6±34.6 cm2 (range 6.4 to 205.4 cm2) and circumference of 21.4±6.0 cm (range 9.4 to 51.0 cm).
Reentry Circuit Isthmus Identified
An isthmus was identified to guide placement of the initial RF line in 25 patients (Table 1): 5 patients with stable VT, 15 with stable and unstable VTs, and 5 with only unstable VTs. An initial line 3.5±1.2 cm in length (11.6±5.7 lesions) was placed through the isthmus (Figure 2). After the initial line was placed, the same VT remained inducible in 8 patients (after 10.5±4.5 lesions over 3.9±1.4 cm), a different VT was inducible in 4 patients (after 15.3±11.4 lesions over 4.7±1.3 cm), and no VT was inducible in 13 patients (after 11.2±3.9 lesions over 3.6±1.3 cm). Additional lesions (9.6±5.6) extended the initial line in 8 patients and created a new RF line at a different location in 4 patients. After these additional lesions, no VT was inducible in 18 of the 25 patients, who had a mean of 2.7±1.8 (range 1 to 8) different morphologies of VT after a total line length of 4.3±2.0 cm created with 18.1±11.2 RF applications. In the remaining 7 patients, inducible VT was modified by a total ablation line length of 6.3±2.8 cm with 18.4±5.8 RF applications.
For the 25 patients with an isthmus identified, follow-up was 288±224 days (range 6 to 735 days). Antiarrhythmic drugs were discontinued in 11 patients. No VT recurred in 18 (72%) of 25 patients. In 7 patients with incessant VT, no VT recurred in 6, and isolated episodes were terminated by a defibrillator in 1 patient. For the 18 patients without incessant VT, the frequency of VT episodes decreased from 9.4±5.3/mo for the 2 months preceding ablation to 0.1±0.3/mo for the 6 months after ablation (P=0.0004).
No Reentry Circuit Isthmus Identified
In 15 patients, no reentry circuit isthmus was identified (Table 2). A stable VT was present in 7 patients; 8 patients had only unstable VT (Table 2 and Figure 3). After the initial RF line guided by pace mapping (16.1±7.7 lesions over 5.2±2.9 cm), no VT was inducible in 4 patients (after 14.0±5.3 lesions over 3.3±1.9 cm), the same VT was again inducible in 5 patients (after 13.8±4.0 lesions over 6.9±3.6 cm), and a different VT was inducible in 5 patients (after 22.4±7.4 lesions over 5.5±4.6 cm). One patient was not tested after ablation because of a peripheral arterial complication. In 9 of 10 patients who had inducible VT, 6.9±3.1 additional lesions were placed to extend the initial line. At the conclusion of the procedure, no VT was inducible in 5 patients who had a mean of 3.0±1.9 different VTs (range 1 to 6) after a total line length of 6.6±5.6 cm with 21.4±10.9 RF lesions. VT was modified in 3 patients (line length 8.5±4.5 cm with 19.3±8.4 lesions), and VT remained in 7 patients (line length 7.4±3.8 cm; RF lesions 25.3±12.0).
Follow-up was 349±286 days (range 16 to 945 days). Antiarrhythmic drugs were discontinued in 4 patients. Eight patients (53%) had a recurrence of some type of VT. The frequency of VT episodes decreased from 11.9±11.1 for the 2 months preceding ablation to 0.4±0.7/mo for the subsequent 6 months (P=0.0029).
Comparison of Isthmus-Guided and Sinus Rhythm—Guided Approaches
Patients for whom an isthmus was identified were similar to patients without an isthmus identified (Table 3) but were more likely to have all VTs ablated or modified (100% compared with 53%, P=0.0002) and received a shorter RF line (4.9±2.4 versus 7.4±4.3 cm; P=0.02). Procedure time and fluoroscopy times were similar for the 2 groups. Fewer patients who had isthmus-guided ablation experienced recurrence of VT (28% versus 53%), but this difference did not reach statistical significance (P=0.11; Figure 4A). The reduction in frequency of VT episodes was similar (Table 3 and Figure 4B).
In 12 patients (10 with an isthmus identified, 2 with no isthmus identified), a saline-irrigated ablation system was used. Exclusion of these patients did not alter the results; the isthmus-identified group received fewer total RF lesions (17.0±7.8 versus 24.6±10.1, P=0.03) and shorter ablation lines (2.8±0.9 versus 5.0±2.5 cm, P=0.03) than the group with no isthmus identified. In a logistic regression model incorporating ejection fraction, stability of VT, fastest VT cycle length, amiodarone use, and use of cooled RF ablation, identification of an isthmus was independently associated with abolition of inducible VT (odds ratio 4.9, 95% CI 1.1 to 22.6; P=0.04) and with a trend toward decreased risk of any VT recurrence (odds ratio 3.37, 95% CI 0.8 to 14.1; P=0.1).
There were no procedure-related deaths. Four procedure-related complications (2 in each group) included iliac artery dissection, femoral artery pseudoaneurysm, embolism to lower leg from a diseased iliac artery, and a retroperitoneal hematoma. No patient had clinically apparent aggravation of heart failure.
During follow-up, 9 patients (22.5%) died; 5 of these deaths were of noncardiac causes unrelated to the procedure and 3 were due to congestive heart failure (125, 208, and 771 days after ablation, respectively). Sudden death occurred in 1 patient who refused placement of an implantable cardioverter-defibrillator despite discontinuation of amiodarone because of liver toxicity. Two patients with chronic heart failure underwent heart transplant (88 and 16 days after ablation). The 1-year mortality rate was not statistically different in groups with isthmus identified (12.0%) and no isthmus identified (26.7%; P=0.15).
In 7 patients (17.5%), only stable VTs were induced. These patients had fewer VTs induced (1.1±0.4 VTs) than did patients with only unstable VT (3.8±1.9) or both stable and unstable VT (4.3±2.9; P=0.0006). An isthmus was identified more frequently in patients with stable VT (20 [74.1%] of 27) versus patients with unstable VT only (5 [38.5%] of 13; P=0.03). Patients with at least 1 stable VT tended to have fewer recurrences than those with only unstable VT (Table 4), but the differences were not statistically significant.
Ablation of infarct-related VT during stable sinus rhythm, guided by delineation of the infarct region from sinus rhythm electrograms, maintains hemodynamic stability while ablation of unstable VTs and multiple VTs can potentially be achieved. However, how best to guide placement of ablation lesions is unclear. Ablation over the entire infarct region or around the entire infarct border is difficult owing to the large infarct size.2,8,9 Extensive lesions may not be desirable because the risk of complications, including damage to functioning myocardium, may increase. A method of targeting the ablation lines to a segment of the infarct region is therefore of interest.
In the present study, sinus rhythm mapping was combined with limited mapping during VT. A reentry circuit isthmus could be identified for at least 1 VT in 62.5% of patients. For patients without a stable VT, this was accomplished by placing the mapping catheter at a candidate site based on mapping during sinus rhythm and then inducing and terminating VT promptly.3 A single line through an isthmus identified for one VT abolished multiple VTs in 11 of 25 patients. Extension of this line abolished all VTs in another 5 patients. Other studies have also observed multiple morphologies of VT originating from one area of the infarction.5,6 Thus, finding an isthmus for a single VT can help limit RF ablation lines even when multiple VTs and unstable VTs are present. Ablation of all inducible VTs was achieved in 57.5% of these patients; recurrences were markedly reduced during follow-up with an average RF line length of 6.0±3.5 cm. We cannot exclude the possibility that finding an isthmus was simply a marker for a subendocardial VT circuit. Thus, a pace-mapping—guided approach may also have been successful in these patients, although the RF ablation may have been more extensive.
In 15 of 40 patients, an isthmus could not be identified owing to VT instability or failure to find the isthmus on the endocardium. A worse outcome was anticipated in this group, but more extensive ablation guided by pace-mapping findings often achieved a reduction in spontaneous VT.
As in all studies of VT ablation in humans, the patient population was selected on the basis of referrals for the procedure. This study was not a randomized comparison of RF line length or RF line placement. We did not try to induce VT after every RF application, and thus the minimum line length required was not determined.
RF application over a region 4 to 5 cm in length that includes a reentry circuit isthmus or likely exit based on pace mapping achieves ablation of all inducible VTs in more than 50% of patients regardless of the presence of multiple VTs and unstable VTs. Testing inducibility after creation of a 3- to 4-cm line through such target areas is a reasonable approach to guide creation of ablation lines.
This study was supported in part by an unrestricted educational grant from Biosense Webster, Inc.
Dr Stevenson has received funding from Biosense Webster, Inc, which supported this study in part by an unrestricted educational grant.
Stevenson WG, Friedman PL, Kocovic D, et al. Radiofrequency catheter ablation of ventricular tachycardia after myocardial infarction. Circulation. 1998; 98: 308–314.
Marchlinski FE, Callans DJ, Gottlieb CD, et al. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation. 2000; 101: 1288–1296.
Wilber DJ, Kopp DE, Glascock DN, et al. Catheter ablation of the mitral isthmus for ventricular tachycardia associated with inferior infarction. Circulation. 1995; 92: 3481–3489.
Miller JM, Tyson GS, Hargrove WC, et al. Effect of subendocardial resection on sinus rhythm endocardial electrogram abnormalities. Circulation. 1995; 91: 2385–2391.
Bourke JP, Campbell RW, Renzulli A, et al. Surgery for ventricular tachyarrhythmias based on fragmentation mapping in sinus rhythm alone. Eur J Cardiothorac Surg. 1989; 3: 401–406.