(Circulation. 1996;94:1041-1048.)
© 1996 American Heart Association, Inc.
Articles |
the Cardiac Arrhythmia Service, New England Medical Center Hospital, Divisions of Cardiology and Surgery, Tufts University School of Medicine, Boston, Mass.
Correspondence to N.A. Mark Estes III, MD, Director, Arrhythmia Service, Tufts/New England Medical Center, 750 Washington St, Boston, MA 02111. E-mail N.A.Estes@Card@NEMC.
| Abstract |
|---|
|
|
|---|
12%). Reconstruction of the left ventricle may reduce mortality. Methods and Results Reconstruction of the left ventricle with a pericardial patch, or endoaneurysmorrhaphy, was performed with mapping-guided subendocardial resection for recurrent ventricular tachycardia in 25 patients over a 5-year period. Postoperatively, electrophysiological studies were conducted to assess the results of surgery, which were further evaluated during long-term follow-up with survival analyses. The study included 25 patients, 60±9 years of age, with coronary artery disease, discrete left ventricle aneurysms, and malignant ventricular tachyarrhythmias. Left ventricular ejection fraction was 24±6% preoperatively. Left ventricular endocardial mapping, endocardial resection, and endoaneurysmorrhaphy were performed in all patients. There was no operative or postoperative (30-day) mortality. Postoperative ventricular tachycardia was induced in 2 of the 25 patients (8%); left ventricular function increased to 32±9% (range, 19% to 52%). At a mean follow-up of 37±16 months (range, 6 to 65 months), there had been 6 deaths, including 1 sudden cardiac death, 2 congestive heart failure deaths, and 3 noncardiac deaths. Analysis of multiple variables failed to identify predictors of postoperative inducibility, sudden cardiac death, cardiac death, or total mortality.
Conclusions Endoaneurysmorrhaphy with a pericardial patch combined with mapping-guided subendocardial resection frequently cures recurrent ventricular tachycardia with low operative mortality and improvement of ventricular function. Long-term follow-up demonstrates low sudden cardiac death rates.
Key Words: ablation aneurysm mapping surgery tachyarrhythmias
| Introduction |
|---|
|
|
|---|
12%),7 30 with most postoperative deaths from cardiogenic shock. Postoperative arrhythmia induction by programmed stimulation is reported in 20% to 30% of patients; however, clinical recurrence of tachycardia is prevented in most (
90%) patients with this approach. Although the technique of left ventricular reconstruction was described previously,40 41 42 43 44 45 no prior series have reported on the technique of left ventricular reconstruction with an autologous pericardial patch, or endoaneurysmorrhaphy, in conjunction with mapping-guided subendocardial resection. In this study, we give the results of this technique in 25 patients with drug refractory ventricular tachycardia. Clinical outcomes are analyzed by life-table methodology, and the influence of clinical, hemodynamic, and electrophysiological factors on postoperative arrhythmia inducibility, long-term arrhythmia control, and mortality is analyzed by stepwise logistic regression techniques. | Methods |
|---|
|
|
|---|
Preoperative Evaluation
The initial drug-free electrophysiological study was performed with patients in a fasting, nonsedated state. All drugs with antiarrhythmic activity had been discontinued for at least 5 drug half-lives. Three quadripolar, 6F electrode catheters were introduced through the femoral vein under local anesthesia by the Seldinger technique and positioned in the high right atrium, across the tricuspid valve for recording from the His bundle, and at the right ventricular apex. Programmed stimulation consisted of single and double ventricular extrastimuli during sinus rhythm and three basic drive cycle lengths followed by triple extrastimuli during sinus rhythm and ventricular pacing of the right ventricular apex and outflow tract with and without isoproterenol.46 Sustained ventricular tachycardia was defined as tachycardia lasting >30 seconds or requiring termination because of hemodynamic compromise.
All patients in this series underwent catheterization of the right and left sides of the heart, coronary angiography, and biplane left ventriculography. Preoperative mapping was performed during ventricular tachycardia in the electrophysiology laboratory with use of a deflectable-tip steerable catheter introduced through the femoral or brachial artery and moved to multiple endocardial sites, including at least 1 apical, 3 septal, 4 anterior, and 4 inferoposterior wall sites under fluoroscopic guidance.19 21 Unstable angina, >75% stenosis of the left main coronary artery, and hemodynamic instability were considered contraindications to mapping in the electrophysiology laboratory. Left ventricular mapping was considered incomplete if 1 apical, 2 septal, 3 anterior, and 3 posterior wall sites could not be recorded. The ventricular tachycardia configuration was considered different if there was a distinctive bundle-branch configuration or if the frontal-plane QRS axis was >90° with the same bundle-branch configuration.
Intraoperative Mapping and Surgical Techniques
Intraoperative mapping was performed with patients under general anesthesia with normothermic cardiopulmonary bypass. A previously placed right ventricular endocardial electrode catheter was used for arrhythmia induction by use of the previously described protocol. A 12-lead surface ECG was obtained during the ventricular tachycardia with recordings from 52 bipolar pairs with a sock electrode from the epicardium. Monomorphic ventricular tachycardia that allowed localization by recording >90% of bipolar electrode pairs was considered to be completely mapped. Once a complete map of all morphologies of ventricular tachycardia had been obtained, an incision was made through the middle portion of the aneurysm (Fig 1A
), and a balloon also containing 52 electrode pairs was placed in the cavity of the left ventricle and inflated to 15 mm Hg. The sites of endocardial origin of the ventricular tachycardia were marked with a suture.
|
The surgical techniques included resection of the local subendocardial scar (Fig 1B
) at the endocardial sites identified as the location of origin of the ventricular tachycardia. A 2-mm-deep subendocardial resection was performed at the site of earliest diastolic activity during ventricular tachycardia with a 1- to 2-cm circumferential margin. In patients with septal or papillary muscle origin of the tachycardia, cryothermia was used as a supplement around the margins of the resected tissue. These cryothermal lesions were delivered around the margins of the subendocardial resection by freezing to -60°C for 2 minutes with a 15-mm cryoprobe.
After completion of the subendocardial resection, programmed ventricular stimulation was repeated. If ventricular tachycardia was persistently inducible, endocardial mapping and subendocardial resection were repeated. When the mapping-guided subendocardial resection was complete and ventricular tachycardia was no longer inducible, a pericardial patch was sutured circumferentially on the endocardial surface of the left ventricle (Fig 1C
). This pericardial patch was harvested before epicardial mapping and treated during mapping and ablation of the ventricular tachycardia. Treatment consisted of immersing the pericardium in a 0.625% solution of glutaraldehyde (Polyscientific) for 10 minutes. After glutaraldehyde treatment, the pericardium was washed in three sequential 3-minute-long saline baths. This autologous pericardial patch was placed at the margin of the normal and fibrotic tissues and sutured so that concentric left ventricle geometry was maintained. Plication but not resection of the aneurysm was performed around the pericardial patch (Fig 1D
). The aorta was then cross clamped, and additional surgical procedures were performed. These included coronary artery bypass grafting in 20 patients, mitral valve repair in 1 patient, and aortic valve replacement in 1 patient.
Postoperative Evaluation
All patients were monitored continuously for 24 to 48 hours in a surgical intensive care unit and then in a step-down unit during the postoperative period. One week after surgery, programmed ventricular stimulation was repeated with triple extrastimuli at the right ventricular apex and outflow tract with and without isoproterenol. In those patients in whom sustained ventricular tachycardia was inducible, an automatic implantable cardioverter-defibrillator (ICD) was inserted.
Clinical Follow-up
After hospital discharge, the patients were followed at regular intervals by their referring cardiologists and in our Arrhythmia Center every 3 months for the first year and every 6 months thereafter. Any symptoms of arrhythmia recurrence were noted and evaluated with a Holter or a loop monitor.
Statistical Analysis
All continuous variables are reported as mean±SD. Log-rank tests were used to determined statistically significant (P<.05) differences of single variables. Continuous variables were analyzed with Cox proportional hazard models. The analyzed variables included age (
65 years or <65 years), sex, diabetes, number of antiarrhythmic drug trials, prior amiodarone use, ejection fraction >25% or <25%, clinical history of cardiac arrest requiring both cardiopulmonary resuscitation and cardioversion or defibrillation, recent (within 2 months) myocardial infarction, history of syncope, New York Heart Association functional class of heart failure symptoms, location of myocardial infarction, number of ventricular tachycardia morphologies, tachycardia cycle length, location of aneurysm, location of earliest activation, incomplete mapping, number of sites mapped, aorta cross-clamp time, cardiopulmonary bypass pump time, coronary bypass grafting, use of intra-aortic balloon, and postoperative ejection fraction. All variables initially were analyzed separately to identify those that had a predictive value for postoperative arrhythmia inducibility, sudden death, or nonfatal ventricular tachycardia recurrence. Curves for survival and recurrent ventricular tachycardia were constructed by use of standard survival analysis.47 48
| Results |
|---|
|
|
|---|
|
Preoperative Catheterization and Electrophysiological Mapping
Cardiac catheterization, coronary angiography, and left ventriculography revealed left main coronary artery disease in 1 patient, single-vessel disease (left anterior descending) in 7 patients, double-vessel disease in 10 patients, and triple-vessel disease in 7 patients. The left ventricular dyskinetic segment was located anteriorly in 24 patients and inferiorly in 1 patient. Preoperative electrophysiological mapping during ventricular tachycardia was completed in 14 patients.
Intraoperative mapping was considered complete for all ventricular tachycardia configurations seen clinically or induced by programmed stimulation in 15 patients. In 2 patients in whom no sustained ventricular tachycardia could be induced intraoperatively, reproducible monomorphic nonsustained ventricular tachycardia was mapped. Results of preoperative and intraoperative mapping were concordant with the same ventricular tachycardia configuration induced at both mappings in all patients. After initial mapping and subendocardial resection, multiple attempts to reinduce ventricular tachycardia were made in all patients. Ventricular tachycardia was still inducible in 4 patients; subsequently, mapping was performed, followed by further subendocardial resection until ventricular tachycardia could no longer be induced in these patients.20 Subendocardial resection was performed in all patients at the sites of earliest diastolic activity defined on the endocardial surface of the left ventricle with a 2-cm margin of tissue removed from the earliest activation site.
All patients underwent left ventricular reconstruction with pericardial patch (Fig 1
). Twenty-three had reconstruction with autologous pericardium, and 2 had reconstruction with bovine pericardium when glutaraldehyde was temporarily unavailable. During normothermic cardiopulmonary bypass and after epicardial mapping, ventriculotomy was performed through a longitudinal incision parallel to the interventricular septum at the apex of the aneurysm. Thrombus, if present, was evacuated, and blood was aspirated from the left ventricular cavity. Endocardial mapping was then performed by use of a 56-bipolar-pair electrode balloon. Mapping-guided subendocardial resection and cryoablation were subsequently performed as described above. After identification of the healthy myocardium where it bordered the scarred aneurysm, ventricular endoaneurysmorrhaphy was undertaken with an elliptical pericardial patch pretreated with glutaraldehyde. By use of a continuous suture with 3-0 polypropylene, the patch was attached to the edge of the healthy myocardium where it bordered the scar so that it replaced the nonviable tissue (Fig 1C
). The ventriculotomy was repaired with either felt buttresses or direct suture. Cold cardioplegia was then achieved in 20 patients requiring coronary artery bypass grafting. Additionally, aortic valve replacement was performed in 1 patient, and mitral valve annuloplasty with a Carpentier ring was performed in 1 patient.
Of the 25 patients, 2 had no inducible arrhythmia to allow epicardial mapping in the operating room. One patient had only incomplete epicardial mapping. The remaining 22 had complete epicardial mapping. After ventriculotomy and insertion of the electrode balloon, 7 additional patients no longer had inducible sustained arrhythmia. The remaining 15 underwent complete endocardial mapping. Of the 23 patients with intraoperative ventricular tachycardia, there were an average of 2.2 ventricular tachycardia morphologies separated by more than 3 cm. The septum was the site of origin in 16 patients, the apex in 10, the base in 4, the lateral wall in 2, and the inferior wall in 1. Epicardial mapping was concordant with endocardial mapping in patients with endocardial origins other than the septum. In patients with septal origins of ventricular tachycardia, epicardial breakthrough was relatively late and could occur anteriorly, apically, or even posteriorly.
The average bypass pump time was 179 minutes, and the average cross-clamp time was 30 minutes (10 patients had no aortic cross clamp; the remaining 15 had an average cross-clamp time of 50 minutes). All patients had limited subendocardial resection at the site of earliest activation. Cryoablation at the margins of the subendocardial resection was used in 23 patients. Three patients required intra-aortic balloon pumps for 1 to 3 days after bypass surgery.
Postoperative Electrophysiological Study
All patients underwent repeated electrophysiological evaluation
1 week after surgery. Ventricular tachycardia was inducible in 2 of 25 patients (Table 2
). Both patients received automatic ICDs. Neither of these patients had any inducible sustained tachycardia intraoperatively to allow complete endocardial mapping at the time of surgery. Preoperative mapping also was not performed in these patients. The remaining 23 patients had no inducible arrhythmias at the right ventricular apex and outflow tract with and without isoproterenol. In 2 patients, procainamide was given after the postoperative electrophysiological study to treat atrial fibrillation. Both patients' arrhythmias had been noninducible before the administration of procainamide, and these 2 patients were retested to ensure no inducible ventricular tachycardia after administration of the drug.
|
Clinical Follow-up
During the long-term follow-up period of 37±16 months (range, 6 to 65 months), there had been 6 deaths, including 1 of renal carcinoma at 6 months, 1 of sepsis at 27 months, 1 of congestive heart failure after myocardial infarction at 28 months, 1 of sudden death at 34 months, 1 of congestive heart failure at 44 months, and 1 of respiratory failure at 52 months. Of the 2 patients who received an ICD, 1 had a defibrillator shock at 62 months. This patient did not have symptoms before this shock. At 66 months, he had another shock preceded by lightheadedness.
The mean left ventricular ejection fraction increased from 24±6% before surgery to 32±9.3% after surgery. Class III or IV congestive heart failure developed in 5 patients, 1 of whom underwent an orthotopic heart transplant 25 months after the endoaneurysmorrhaphy. One patient developed severe mitral regurgitation and required mitral valve replacement 5 months after the initial endoaneurysmorrhaphy. One patient's symptoms improved to class II with medical therapy.
Survival analysis (Fig 2
) demonstrates that at 1 year the incidences of sudden cardiac death, cardiac mortality, and total mortality were 0%, 0%, and 4%, respectively. At 2 years, the incidences were 0%, 0%, and 4%. At 3 years, they increased to 6%, 11%, and 19%. At 4 years, the incidences were 6%, 21%, and 28%. At 5 years, they were 6%, 21%, and 46%.
|
Analysis of multiple clinical, hemodynamic, angiographic, and electrophysiological variables failed to identify any predictors of postoperative ventricular tachycardia inducibility, arrhythmia recurrence, or sudden cardiac death. Noncardiac mortality was increased in those patients presenting with rhythm other than ventricular tachycardia (P=.01). There was also a trend toward increased total mortality in those patients presenting with a rhythm other than ventricular tachycardia (P=.08). Congestive heart failure in the follow-up was seen more commonly in the patients <65 years of age (P=.03) and in those patients with lower preoperative ejection fractions (P=.04). Because there was no end point of operative mortality, predictors of these events were not identified.
| Discussion |
|---|
|
|
|---|
25%. In the prior series, the mean ejection fraction ranged from 23% to 38%. Two investigative groups selected patients by quantitative measures of the nonaneurysmal section of the left ventricle.14 28 Poor systolic function of the nonaneurysmal ventricular segments has been shown to be a strong and, in one series, independent predictor14 of postoperative mortality. In one series, postoperative cardiogenic shock despite treatment with intravenous pressers and intra-aortic balloon counterpulsation resulted in death within 30 days in 11% of patients. In another series, all deaths5 were observed in patients with left ventricular ejection fractions <25%.22 Other risk factors for operative mortality seen in other series were congestive heart failure class,11 28 low ejection fraction,13 25 29 35 48 bypass time,9 13 28 and age.11 21 24 36 Because we had no operative deaths, we cannot comment on risk factors for this event. The 8% incidence of postoperative inducibility of sustained ventricular tachycardia in this series compared favorably with the 6% to 50% range reported in the literature.2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Prior analyses of factors contributing to postoperative inducibility in 100 consecutive patients in one series have demonstrated that there is only a 56% success rate when no mapping could be performed, 69% cure rate when >50% of ventricular tachycardias could be mapped, and 88% success rate when all ventricular tachycardia configurations were mapped.5 In our series, one patient who had ventricular tachycardia induced after surgery had no inducible arrhythmia intraoperatively to allow mapping; in the other patient, only epicardial mapping could be performed as a result of noninducibility after left ventriculotomy. Overall, only 2 patients had neither epicardial nor endocardial mapping performed, 1 had only endocardial mapping, and 7 had epicardial mapping alone in our series. Our observation that complete epicardial and endocardial mapping could be completed in only 15 of 25 patients is consistent with prior reports of technically adequate mapping in 53%14 and 63%21 of patients. It is evident that the inability to induce and adequately map a tachycardia intraoperatively precludes identification of the earliest site of endocardial activation. Inability to perform complete mapping was most often due to an inability to reinduce the ventricular tachycardia after ventriculotomy. This inability to complete endocardial mapping mandates reliance on preoperative catheter mapping or a nondirected surgical procedure.6 7 11 28 31 39 48 Whether mapping-guided surgery is essential in lowering the risk of inducible ventricular or recurrent arrhythmias is not clear from the literature. Of the studies referenced here, most primarily used mapping-guided techniques with a postoperative inducibility rate of 3% to 37%.* Other investigators10 26 31 47 48 did not use mapping and had similar postoperative inducibility rates of 4% to 37%. Variations in the percentage of patients mapped, the completeness of mapping, different definitions of noninducibility, and the lack of randomized trials make direct comparisons between the two techniques difficult.
Four prior series have reported freedom from sudden cardiac death in follow-up with an average follow-up of 18,7 25,14 41,36 and 43 months.10 Arrhythmia recurrence was late in our series. One patient died suddenly at 34 months, and another patient had ICD shocks at 62 and 66 months. This incidence of sudden cardiac death and arrhythmia recurrence compares favorably to those of most surgical series with previous reports of sudden death in up to 18%28 and arrhythmia recurrence in up to 20%14 of patients.
Previously published series of patients treated with ICDs have reported 1-year sudden cardiac death incidence rates of 0% to 3%.49 50 51 Furthermore, 3-year sudden cardiac death rates for ICD patients are from 8% to 11%,49 51 and 5-year sudden cardiac death rates are from 4% to 17%.35 50 51 Thus, our incidences of sudden cardiac death at 1 (0%), 3 (6%), and 5 (6%) years compare favorably to those in the ICD series. The incidence of recurrent arrhythmias is dramatically lower than in those series in which patients were treated with ICDs. Cardiac mortality in ICD series at 1 year ranges from 6% to 10%; at 3 years, from 13% to 28%; and at 5 years, from 24% to 55%.49 50 51 Thus, our 1-, 3-, and 5-year cardiac mortality figures of 0%, 11%, and 21% are lower than the reported ICD series cardiac mortality rates. Total mortality in the ICD series at 1 year is 6% to 16%; at 3 years, 15% to 38%; and at 5 years, 26% to 65%.49 50 51 In our current series, 1-, 3-, and 5-year total mortality rates are 4%, 19%, and 46%, respectively. Thus, although sudden cardiac death and cardiac deaths are decreased with endoaneurysmorrhaphy compared with ICDs, total mortality remains high in long-term follow-up, largely related to the age, myocardial dysfunction, and comorbid conditions of this population.
Left ventricular reconstruction techniques were described previously and used initially by Jatene40 in 1985 and subsequently by Cooley,41 who introduced endoaneurysmorrhaphy, and other investigators42 43 44 45 in patients undergoing left ventricular aneurysm resection or repair for hemodynamic reasons. These investigators, however, used either a woven Dacron patch or a bovine pericardial patch instead of the autologous pericardial patch used in our series. Benefits of using a biological material such as bovine or human pericardium include the ease of suturing and the theoretical decrease in clot formation. Although there is little biological difference between bovine and human pericardium, autologous human pericardium is readily available without the expense of bovine pericardium. However, the difference in patch material is less important than the combination of mapping-guided subendocardial resection and left ventricular reconstruction. Only one limited series has reported on the use of endoaneurysmorrhaphy as curative surgery in patients with malignant ventricular arrhythmias.52 Our series is the first to report endoaneurysmorrhaphy with autologous pericardium in combination with mapping-guided subendocardial resection and cryothermic ablation to cure patients with ventricular tachycardia.
Conclusions
Endoaneurysmorrhaphy with an autologous pericardial patch combined with mapping-guided subendocardial resection in patients with discrete ventricular aneurysms frequently cures recurrent ventricular tachycardia without operative mortality. Ventricular function and functional status usually are improved. Long-term follow-up demonstrates low sudden cardiac death rates, but as follow-up increases, mortality increases in this population, largely as a result of congestive heart failure and noncardiac causes.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received October 24, 1995; revision received March 6, 1996; accepted March 13, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
U. Sartipy, A. Albage, E. Straat, P. Insulander, and D. Lindblom Surgery for Ventricular Tachycardia in Patients Undergoing Left Ventricular Reconstruction by the Dor Procedure Ann. Thorac. Surg., January 1, 2006; 81(1): 65 - 71. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Pirk, J. Bytesnik, J. Kautzner, P. Peichl, V. Vancura, K. Lefflerova, I. Skalsky, and V. Vinduska Surgical ablation of post-infarction ventricular tachycardia guided by mapping in sinus rhythm: long term results Eur. J. Cardiothorac. Surg., August 1, 2004; 26(2): 323 - 329. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. O. von Oppell, D. Milne, A. Okreglicki, and R. N. Scott Millar Surgery for ventricular tachycardia of left ventricular origin: risk factors for success and long-term outcome Eur. J. Cardiothorac. Surg., November 1, 2002; 22(5): 762 - 770. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Nitta, M. Mitsuno, C. K. Rokkas, R. Lee, R. B. Schuessler, and J. P. Boineau Cryoablation of ventricular tachycardia guided by return cycle mapping after entrainment J. Thorac. Cardiovasc. Surg., February 1, 2001; 121(2): 0249 - 258. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Bourke, J. Gray, C. J. Hilton, S. S. Furniss, S. Khan, J. M. McComb, and R. W.F. Campbell Identifying patients at low risk of death from cardiac failure after operation for postinfarct ventricular tachycardia Ann. Thorac. Surg., February 1, 1999; 67(2): 404 - 410. [Abstract] [Full Text] [PDF] |
||||
![]() |
Surgery for Ventricular Tachycardia Journal Watch Cardiology, November 1, 1996; 1996(1101): 18 - 18. [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |