YAG Laser Photocoagulation of Ventricular Tachycardia Without Ventriculotomy in Patients After Myocardial Infarction
Background Surgical ablation of ventricular tachycardia (VT) after myocardial infarction has been reported by different endocardial approaches. The ventriculotomy may increase mortality of the procedure.
Methods and Results We report on nine patients who suffered from recurrent VT in the late post–myocardial infarction period. Significant stenoses were detected in all patients. The mean left ventricular ejection fraction was 43.1±8.3%. Left ventricular scar (n=9) was seen. The mean NYHA class was 2.2±0.4. Sustained VT (mean cycle length, 293±52 ms) occurred spontaneously (n=9) and could be induced reproducibly. Catheter mapping detected a prematurity of −42±13 ms in six patients. Clinical VT was inducible during surgery in seven patients. Middiastolic potentials were detected from the epicardial surface (n=3), and premature potentials were found (n=8 with prematurity of −108±46 ms). Application of neodymium/yttrium/argon/garnet (Nd:YAG) laser energy to early epicardial activation terminated the arrhythmia (n=7). Ventriculotomy was not performed. Seven patients have been free of VT for a mean follow-up period of 17±11 months; one patient relapsed and was treated with an implantable cardioverter-defibrillator, as was a second patient with inducible VT after surgery.
Conclusions Surgical Nd:YAG laser photocoagulation of VT on the epicardial surface of the heart in post–myocardial infarction patients without ventriculotomy is safe and has a high success rate. At the present time, this method is recommended in patients with sustained and tolerated VT who need bypass surgery. This is the first report on epicardial laser ablation of VT in post–myocardial infarction VT.
The origin of ventricular tachyarrhythmias (VTs) is located in the subendocardial layers of myocardium in the vast majority of post–myocardial infarction patients.1 Therefore, the surgical treatment proposed was subendocardial excision,2 encircling endocardial ventriculotomy,3 endocardial cryoablation,4 or endocardial laser ablation1 5 6 7 of the arrhythmogenic substrate.8 All these surgical procedures require a ventriculotomy,2 3 4 7 9 which probably contributes to the morbidity and mortality of the surgical ablation.10 Downar et al10 avoided a ventriculotomy by delivery of DC shocks through an endocardial balloon electrode. Ablation of subendocardial layers by epicardial cryosurgery seems possible but resulted in large epicardial effects. However, laser energy lends itself to deep tissue penetration. Epicardial laser photocoagulation has been successful in some patients, but all had ventriculotomy either for other VT morphologies or as part of a mapping procedure.9 11
This article describes the first known cure of VT in patients after myocardial infarction by epicardial neodymium/yttrium/argon/garnet (Nd:YAG) laser ablation without left ventriculotomy.
Nine male patients between 38 and 70 years of age were admitted to the University hospitals of Bonn or Marburg. They had survived anterior (n=5), lateral (n=2), or posterobasal (n=3) myocardial infarctions, 1 to 16 years earlier. Echocardiographic investigation revealed dilatation of the left ventricle (one, patient 1, with a thrombus) in an anteroseptal (n=3, patients 1, 2, and 3), anterolateral (n=2, patients 4 and 6), or calcified anterior basal (n=1, patient 5) akinetic region. The left ventricular angiogram showed an anteroapical and septal (n=4, patients 1, 2, and 3), lateral (n=3, patients 4, 6, and 8), or posterior (n=2, patients 7 and 9) akinesia with calcification (n=3, patients 1, 2, and 5). The mean left ventricular ejection fraction was measured to be 43.1±8.3%, and left ventricular end-diastolic pressure was increased between 13 and 19 mm Hg at rest and between 14 and 24 mm Hg after contrast overload. Significant stenosis of the left anterior descending coronary artery was observed in 7 patients, of the first marginal branch in 4 patients, and of the right coronary artery in 3 patients. A critical stenosis in the left main artery was detected in 2 patients. The patients were treated with nitroglycerin, converting enzyme inhibitors, acetylsalicylic acid, and diuretics and were in NYHA class 2 (n=7) or 3 (n=2) during drug treatment (Table 1⇓). Antiarrhythmic drugs (n=2 to 5; mean, 2.4) were ineffective in suppression of sustained VT.
After spontaneous occurrence of monomorphic VT with a mean cycle length of 293±52 ms (n=9) (Fig 1⇓) with left (n=4) or right (n=6) bundle branch block pattern and left (n=5), inferior (n=2), or right (n=3) axis deviation, an electrophysiological study was performed. Spontaneous tachycardias had a cycle length between 230 and 400 ms and were hemodynamically well tolerated (n=6) or unstable (n=3). In all patients, the clinical VT could be induced reproducibly by double or triple extrastimuli (S1S2, S2S3, or S3S4) during paced ventricular rhythm (S1S1) (Fig 2⇓) and could not be suppressed with sotalol (up to 480 mg/d) and/or mexiletine (600 mg/d). The termination of VT was successful with programmed double extrastimuli or burst pacing. Endocardial catheter mapping (n=6) detected early endocardial activation in the anterior paraseptal region of the left ventricular free wall (patients 1 through 3 in the tables, n=3), in the anterior basal region (n=1, patient 5 in the tables), or in the posterolateral area (n=2, patients 4 and 6 in the tables). The regional endocardial electrograms in this area during tachycardia started −30 to −60 ms (mean, −42±13 ms) before the onset of the surface QRS complex (Fig 3⇓, Table 2⇓). Catheter mapping was impossible in 3 patients (patients 7 through 9) who did not tolerate the VT. High-resolution ECGs revealed late potentials by spectrotemporal mapping technique (n=6, patients 1, 2, 6, and 7 through 9) (Fig 4⇓), showed no late potentials (n=1, patient 3), or were not used in bundle branch block (n=2, patients 4 and 5). Bypass surgery, resection of the scar or aneurysm whenever possible, and map-guided endocardial laser ablation of VT was planned in the first 3 patients. After the first experience, the other 6 patients underwent surgical intervention for bypass grafts with additional epicardial laser ablation of VT without planned resection of a scar or aneurysm. The primary indication was bypass graft surgery and the second indication, the ablation of VT. The patients gave written informed consent.
After median sternotomy, epicardial mapping was performed in a normothermic state. The epicardial surface appeared to have a thin covering of myocardium through which whitish scar tissue could be seen. The clinical VTs could be induced (n=7) or were noninducible (n=2). The earliest epicardial activation was detected between the left anterior descending artery and diagonal branch within (n=1) (Fig 3⇑) or at the border of a large anterolateral and septal scar (n=3); an epicardial multipolar mapping probe indicated it to be in this region after sinus rhythm and pace mapping (n=2), in the anterolateral region within the scar (n=2), or posterobasal (n=2).7 The regional epicardial potentials during sustained VT showed a distinct middiastolic deflection −65 to 180 ms (mean, 108±46 ms) before the onset of QRS and were much earlier in this subgroup than all endocardial activations during preoperative catheter mapping (n=4, Fig 3⇑). Epicardial photocoagulation was performed with a continuous-wave Nd:YAG laser (MBB) coupled to a gas-cooled silica quartz fiber in a handheld probe. Irradiation power ranged between 50 and 80 W. The irradiated epicardial surface of the heart was 1.5×2 to 2×3 cm. During application of laser energy, the VT stopped after several seconds without extra beats or prolongation of tachycardia cycle length (n=8) (Fig 5⇓). If the same VT could be reinduced later on, the same procedure with laser coagulation in the identical region was repeated up to noninducibility. After further epicardial applications of laser energy, the arrhythmia could not be reinduced. The total duration of laser impulse application ranged between 12 and 42 seconds. Intraoperative endocardial mapping was not performed. The occluded arteries were reperfused by aortocoronary bypass grafts (left anterior descending coronary artery, n=8; marginal branch, n=7; right coronary artery, n=1; diagonal branch, n=1) (Table 3⇓). Resection of the scar was not performed after successful ablation of VT.
Electrophysiological study was performed 10 to 15 days after surgery (n=8); 1 patient refused. The clinical VT could not be induced with triple extrastimuli or fast ventricular stimulation (n=7), but nonclinical VTs were inducible in 2 patients. Late potentials could be detected in the spectrotemporal mapping high-resolution ECG in 4 patients and were abolished in 2 (Table 4⇓).
After a mean follow-up of 14±11 months, the patients were reinvestigated (n=7). One patient reported recurrence of the arrhythmia. The others were free of VT in ECG tracings, Holter monitoring, and by their own feelings. The patients were in good physical condition and were free of left heart failure or chest pain and without spontaneous tachycardia or syncope. Seven patients were free of antiarrhythmic drugs, 2 received sotalol (patient 7, because of premature contractions) or amiodarone (patient 8, because of atrial flutter), and 2 needed an implantable cardioverter-defibrillator. One patient had spontaneous relapse; in the other, VT identical to that before surgery was inducible. No spontaneous activation of the device in the follow-up period was reported. No VT could be induced with double (n=1) or triple (n=7) impulses on two different sites of the right ventricle (apex and outflow tract) with three different basic cycle lengths (600, 500, and 400 ms) (n=5). One patient refused the electrophysiological study. Late potentials could not be demonstrated in 3 patients but were seen in 4 (Fig 4⇑).
These patients are the first reported cases of surgical ablation of post–myocardial infarction VT without opening of the left heart chamber. This report lends additional support to a growing body of evidence that the arrhythmogenic substrate of ischemic postinfarction VT may be in an intramural or epicardial as well as an endocardial location.9 11 15 16 17 Our results suggest the possibility that appropriately selected patients may be amenable to surgical ablation as the optional procedure without additional risk of ventriculotomy or in cases requiring thoracotomy for coronary bypass graft surgery.
A combination of careful preoperative endocardial catheter mapping and epicardial mapping during surgery may extend the usefulness of an epicardial approach to the surgical ablation with a reduced risk. Taking advantage of the deep tissue penetration of Nd:YAG photocoagulation, ablation of arrhythmogenic substrates may be possible from the epicardium extending to near subendocardial layers. In the cases presented, most of the reentrant circuit was probably epicardial. In most patients, however, parts or all of the circuit may be subendocardial and identified by catheter mapping and/or entrainment studies. The endocardial location as determined by catheter mapping studies can be compared with the epicardial mapping data at surgery, looking for other parts of the diastolic portion of the circuit, points of earliest epicardial activation, or points at the end of the QRS or in the very first part of the diastolic interval. For example, laser photocoagulation (or cryoablation) could be delivered from the epicardium between opposite endocardial sites showing only presystolic activation and epicardial sites activating at the end of the QRS or in the first portion of the diastolic interval, a region that may overlie the middiastolic portion of the circuit. Similarly, photocoagulation could also be performed between the sites of latest and earliest epicardial activation if they are spatially within several centimeters of each other. The epicardial approach would not be applicable to patients in whom the middiastolic portion of the reentrant circuit is in the septum. Therefore, we included in this study patients with free-wall sustained VT suspected from the localization of the scar, the surface ECG of the arrhythmia, and endocardial catheter mapping.
Curative surgery of malignant ventricular arrhythmias was successful with several techniques in large series from different centers.12 13 14 The endocardial approach is the preferred method in patients with post–myocardial infarction arrhythmias. This technique depends on the ventriculotomy in most cases, which increases the risk of the procedure.10
Therefore, the question arises as to what type of arrhythmia can be treated surgically in patients with indication for bypass graft surgery avoiding ventriculotomy. Selected articles describe patients with ischemic postinfarction VT that was ablated surgically by an epicardial9 or combined epicardial and endocardial approach. Most of these patients had inferior myocardial infarctions.10 This article presents successful ablated left free wall VT in patients with anterior or lateral infarctions.
The cases presented give some information on successful epicardial laser ablation of VT: (1) During careful endocardial catheter mapping with the steerable ablation catheter, only premature endocardial signals up to a prematurity of −30 to −60 ms could be registered. Although the quality of epicardial mapping results could not be anticipated during endocardial catheter mapping, it was quite clear that information on endocardial potentials for catheter ablation was insufficient. The origin of the arrhythmia is expected to be more in the subepicardial or intramural layers than in subendocardial sites at the border of the transmural scar in this case. (2) Middiastolic activation during VT on the epicardial surface during intraoperative mapping and (3) reproducible nonextrasystolic termination of sustained VT during epicardial laser application were markers of the permanent success of surgery in this study. The transmural effects of Nd:YAG laser application are investigated elsewhere.6 This information can increase the success rate of surgery in patients with endocardial origin but early epicardial activation of VT, because our patients had late endocardial depolarizations and are not identified by endocardial mapping. (4) Noninducibility after laser coagulation of a reproducibly inducible reentrant VT in the catheter laboratory and the operating room during normothermia is an important indicator of the success of the epicardial approach in post–myocardial infarction VT. (5) Significant loss of delayed potentials after surgery by signal-averaging techniques predicts long-term success. (6) In contrast to other authors, we recommend epicardial mapping at the beginning of the mapping procedure even in patients with ischemic VT. After comparison of the earliest endocardial potentials with the best epicardial electrograms, an assumption of detected endocardial and epicardial localizations of the arrhythmogenic substrate should be made. If a subepicardial origin of the tachycardia is assumed, the epicardial approach without ventriculotomy can be successful. The ventriculotomy can be avoided in a subgroup of patients. (7) A subepicardial arrhythmogenic substrate in postinfarction VT can be an explanation for failure of endocardial catheter ablation or endocardial surgical techniques in some series.2 3 8 12 13 The Nd:YAG laser, with its known property of producing deep tissue coagulation in the myocardium, can solve this problem without ventriculotomy in selected cases. Cryoablation is not investigated in the same way. (8) An arrhythmia-free interval of several months after antiarrhythmic surgery without antiarrhythmic drugs, the noninducibility of the VT, and the abolishment of late potentials in late follow-up study are predictors of long-term suppression of the tachycardia after epicardial Nd:YAG laser photocoagulation.
A major limitation of the present study is the inability to distinguish between a subepicardial location of the reentrant VT that has been ablated by epicardial photocoagulation from a subendocardial origin of VT that would be ablated in the same way by the deep transmyocardial penetration of the laser energy that is delivered from the epicardial site. We believe that the first hypothesis is true in this subgroup of patients, because the epicardial mapping detected electrograms much better than the endocardial catheter mapping. Demonstration of the real difference between the two concepts needs simultaneous intracardial, epicardial, and intramural mapping in the operating room. This is possible with ventriculotomy, which we would avoid because of the increasing risk of the procedure. At the end of the ventriculotomy, for mapping, some of the patients would have noninducibility of VT after the procedure. On the other hand, we are able to demonstrate that epicardial laser delivery may coagulate arrhythmogenic foci after myocardial infarction. Finally, the epicardial delivery of energy for ablation of VT is not recommended with the aim of increasing the success of VT surgery but rather for reducing the risk of the procedure in a highly selected subgroup of patients.
Reprint requests to Dietrich Pfeiffer, MD, FACC, Department of Cardiology, University of Leipzig, Johannisallee 32, D-04103 Leipzig, Germany.
- Received February 19, 1996.
- Revision received July 15, 1996.
- Accepted July 30, 1996.
- Copyright © 1996 by American Heart Association
Spielman SR, Michelson EL, Horowitz LN, Spear JF, Moore EN. The limitations of epicardial mapping as a guide to the surgical therapy of ventricular tachycardia. Circulation. 1978;57:666-670.
Saksena S. Intraoperative laser ablation in malignant ventricular tachycardia. In: Lu¨deritz B, Saksena S, eds. Interventional Electrophysiology. Mt Kisco, NY: Futura Publishing Co; 1991:545-556.
Svenson RH, Selle JG, Gallagher JJ, Marroum MC, Tatsis GP, Seifert KT. Neodymium:YAG laser photocoagulation: a potentially useful method for intraoperative ablation of arrhythmogenic foci. In: Fontaine G, Scheinman MM, eds. Ablation in Cardiac Arrhythmias. Mt Kisco, NY: Futura Publishing Co; 1987:379-403.
Svenson RH, Littmann L, Gallagher JJ, Selle JG, Zimmern SH, Fedor JM, Colavita PG. Termination of ventricular tachycardia with epicardial laser photocoagulation: a clinical comparison with patients undergoing successful endocardial photocoagulation alone. J Am Coll Cardiol. 1990;15:163-170.
Littmann L, Svenson RH, Gallagher JJ, Selle JG, Zimmern SH, Fedor JM, Colavita PG. Functional role of the epicardium in postinfarction ventricular tachycardia: observations derived from computerized epicardial activation mapping, entrainment, and epicardial laser photoablation. Circulation. 1991;83:1577-1591.
Cox JL. Patient selection criteria and results of surgery for refractory ischemic ventricular tachycardia. Circulation. 1989;79(suppl I):I-163-I-177.
Hargrove WC, Josephson ME, Marchlinski FE, Miller JM. Surgical decisions in the management of sudden cardiac death and malignant ventricular arrhythmias: subendocardial resection, the automatic internal defibrillator, or both. J Thorac Cardiovasc Surg. 1989;190:923-928.
Selle JG, Svenson RH, Gallagher JJ, Littmann L, Sealy WC, Robicsek F. Surgical treatment of ventricular tachycardia with Nd:YAG laser photocoagulation. Pacing Clin Electrophysiol. 1992;15:1352-1361.
Page´ PL, Cardinal R, Shenasa M, Kaltenbrunner W, Cossette R, Nadeau R. Surgical treatment of ventricular tachycardia: regional cryoablation guided by computerized epicardial and endocardial mapping. Circulation. 1989;80(suppl I):I-124-I-134.