| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1995;110:1013-1022
© 1995 Mosby, Inc.
SURGERY FOR ACQUIRED HEART DISEASE |
Burlington, Mass.
From the Department of Thoracic and Cardiovascular Surgery and the Section of Cardiology, Lahey Clinic Medical Center, Burlington, Mass.
Received for publication June 15, 1994. Accepted for publication Dec. 29, 1994. Address for reprints: David M. Shahian, MD, Department of Thoracic and Cardiovascular Surgery, Lahey Clinic Medical Center, 41 Mall Rd., Burlington, MA 01805.
Abstract
The impact of adjuvant coronary revascularization was studied in a group of 138 recipients of an implantable cardioverter-defibrillator, all of whom had ischemic heart disease as the cause of their arrhythmias. Patients chosen for revascularization had more severe anatomic, symptomatic, or physiologic evidence of active ischemia. There were no operative deaths among 23 patients who actually underwent coronary artery bypass combined with cardioverter-defibrillator implantation; however, operative mortality by the intention-to-treat principle was 8% (2/25). Total cardiac survival was better for patients who underwent revascularization than for those patients who had "high-risk"characteristics and did not undergo revascularization. Stratified subgroup analysis demonstrated significant survival advantages favoring revascularization in patients with three-vessel or left main coronary artery disease, class III or IV angina, and an ejection fraction greater than 25%. Multivariate analysis revealed that low ejection fraction and left main coronary artery disease were independent predictors of decreased survival. (J THORACCARDIOVASCSURG1995;110:1013-22)
Sudden cardiac death (SCD) from ventricular tachycardia or fibrillation not associated with an acute myocardial infarction typically occurs in the presence of multivessel coronary disease (CAD) and myocardial scar. 
1-3 Unless survivors of sudden cardiac death have previously undergone successful coronary artery bypass grafting (CABG) or percutaneous transluminal coronary angioplasty, many will have persistent unrevascularized CAD with varying degrees of symptoms. When such patients are evaluated for implantation of a cardioverter-defibrillator (ICD), the possibility of concomitant coronary revascularization must be considered. The selection factors for adjuvant revascularization, the incremental impact on perioperative morbidity and mortality, and the effect of revascularization on long-term survival remain problematic. We investigated these issues in a series of 138 patients with malignant arrhythmias who underwent ICD with or without concomitant coronary revascularization.
METHODS
Study group
Of 171 ICD recipients, 138 had ischemic heart disease (CAD) as the major cause of arrhythmias. Several patients with minimal CAD had left ventricular dysfunction that was out of proportion to the extent of their CAD; they were thought to have idiopathic dilated cardiomyopathy as their primary cause of arrhythmias and were not included in this study.
Because of our interest in long-term follow-up, only those patients undergoing implantation of a complete ICD system were included in the study. Two patients underwent coronary revascularization and ICD patch implantation but died after the operation before undergoing complete defibrillation threshold testing or ICD generator placement. These two patients were not included in the current study, but an intention-to-treat analysis of postoperative morbidity and mortality is provided that includes them in the postoperative death data.
Fifty-four group A patients had no unrevascularized CAD because of previous or concomitant coronary revascularization. Additional analysis was performed by subdividing these patients into group A1, 23 patients who underwent concomitant revascularization, and group A2, 31 patients who had had previous revascularization. Group B included 84 patients who had unrevascularized CAD. For long-term survival analysis, group B was subdivided into 63 group B1 patients (low-risk), who typically had one- or two-vessel disease and mild symptoms, and 21 group B2 patients (high-risk), who had indications for coronary artery bypass grafting (CABG) (class III or IV angina or three-vessel disease with impaired ventricular function or positive thallium study results) but could not be revascularized because of poor surgical targets, comorbid disease, or other contraindications.
Statistical methods
Postoperative mortality included all deaths occurring in the hospital. It also included those deaths occurring at home within 30 days after the operation, unless the cause of death was clearly unrelated to the operation. The latter category included only one patient who died at home on postoperative day 27 of a bradycardia, which in retrospect was believed by our electrophysiologist to have been present but unrecognized before the operation. Postoperative mortality has been calculated both with and without this patient included in the operative death data, and survival curves include him in the SCD data.
Discrete variables were compared by means of the 
2 test or the Miettinen modification of Fisher's exact test. Continuous variables were compared by means of Student's t test or the Mann-Whitney test. Survival was analyzed by the Kaplan-Meier method with the differences between groups analyzed by Tarone-Ware testing, occasionally supplemented by the Breslow or Mantel-Haenszel tests. Survival analysis included crude mortality (deaths from all causes), total cardiac mortality (including postoperative deaths and all late cardiac deaths), SCD free survival (unwitnessed deaths or those occurring within 1 hour of onset of symptoms), and non-SCD free survival.
Surgical methods
All patients underwent preoperative electrophysiologic evaluation unless they had unstable angina or severe left main CAD. In such patients, CABG was performed, and patches and rate-sensing electrodes were placed at the time of the operation. Electrophysiologic testing was performed after recovery from the operation, and a generator was connected if appropriate. Three patients did not undergo any electrophysiologic testing. One had incessant spontaneous ventricular tachycardia that was pace terminable, one had severe left main CAD but could not undergo revascularization because of diffuse aortofemoral calcification, and one patient, after out-of-hospital cardiac arrest, was not studied because his arrhythmia had been noninducible on multiple previous attempts.
The decision to revascularize was based on the patient's clinical degree of angina, angiographic extent of disease, objective evidence of ischemia, ejection fraction (EF) and its estimated impact on surgical mortality, other comorbid conditions, and the presence of adequate surgical targets.
All patients were monitored during the operation with a Swan-Ganz catheter (Baxter Healthcare Corp., Edwards Div., Santa Ana, Calif.) and continuous mixed venous oxygen saturation, which determined the interval between arrhythmia inductions. Some patients underwent preoperative intraaortic balloon pump support, particularly if there was severe left ventricular dysfunction or extensive CAD that could not be revascularized.
Defibrillation threshold testing was carried out beginning at 15 joules and proceeding in a stepwise fashion. A 10-joule buffer between the measured defibrillation threshold and the output of the generator was considered adequate.
A one-stage combined procedure was performed whenever possible. If the patient was in stable condition in the operating room after revascularization, limited electrophysiologic testing was performed without cardiopulmonary bypass or with low-flow bypass and a full heart. A study by Blakeman and associates 4 demonstrated that despite the residual effects of cardioplegia, cooling, and operative ischemia, the testing results obtained immediately after the termination of cardiopulmonary bypass were comparable with those obtained at subsequent testing. In patients who are in unstable condition at the end of cardiopulmonary bypass, it is prudent to bury the leads in an abdominal pocket and conduct definitive testing and generator implantation after 1 week.
RESULTS
Preoperative demographics and descriptive statistics are given in Table I. There were no statistically significant differences except for a higher incidence of three-vessel or left main CAD, class III or IV angina, and previous cardiac operation in group A patients. Further stratification (Table II) reveals that the class III or IV symptoms or positive stress test results occurred almost exclusively in patients who were to undergo concomitant CABG (group A1). Patients who had had previous revascularization (group A2) had a minimal ischemic burden according to symptoms and stress testing. Overall, 12 of 30 patients (40%) with positive results of stress testing underwent concomitant coronary revascularization as opposed to only 4 of 35 patients (11.4%) who had negative stress test results (p = 0.0097). Similarly, 11 of 16 patients in the overall series (68.8%) who had class III or IV angina underwent concomitant coronary revascularization as opposed to only 8 of 112 patients (7.1%) who had class I or II angina (p = 0.0001).
|
|
Formal defibrillation thresholds were available for 127 patients, being lower for group B than for group A patients (13.4 ± 5.1 joules versus 16.6 ± 4.9 joules, p = 0.0005). There was no significant difference in lead impedance in the 82 patients in whom this variable was recorded (group A = 39.9 ± 8.7 ohms; group B = 43.6 ± 8.5 ohms; p = 0.0553).
Operative morbidity and mortality analysis (Table III) revealed a higher requirement for use of inotropic drugs in patients undergoing concomitant revascularization (p < .0001), probably because of prolonged cardiopulmonary bypass and postbypass defibrillation threshold testing, and a higher but not statistically significant incidence of atrial and ventricular arrhythmias in this group. Postoperative pleural effusions necessitating therapy and generator hematomas necessitating drainage were also more prevalent, the latter occurring exclusively in patients receiving postoperative anticoagulation. Three patients (13%) had neurologic complications after concomitant CABG and ICD—generalized encephalopathy in one instance, a transient ischemic attack in another, and a cerebrovascular accident in a third. This incidence of complications was higher than in the other two groups (p = 0.0061).
|
Crude, total cardiac, non-SCD free, and SCD free survival for groups A and B are depicted in Fig. 1. These demonstrate highly significant survival advantages for patients who had no unrevascularized CAD (group A) compared with the 21 patients with "high-risk" unrevascularized CAD (group B2) who could not undergo revascularization because of technical factors or comorbid disease. This advantage was noted in crude (p = 0.0215), total cardiac (p = 0.0086), and SCD free survival (p = 0.0131) determinations. Subgroup analysis revealed that survival was enhanced by revascularization in patients with class III or IV angina (p = 0.0268), three-vessel or left main CAD (p = 0.0395), or EF greater than 25% (p = 0.0282) (Figs. 2 to 4).
|
|
|
|
25%, >25%), it violated the proportional hazards assumptions. Therefore EF was analyzed as a continuous variable. Stepwise forward regression selected low EF (coefficient = -0.0633; standard error = 0.0271; p = 0.0118) and left main CAD (coefficient = 1.8761; standard error = 0.5303; relative risk = 6.53, [2.3 to 18.5]; p = 0.0056) as predictive variables for total cardiac mortality. The effect of EF on survival with CAD, with or without SCD, is well known. Patients in our study with left main CAD were a uniquely high-risk group. The mean EF of these seven patients was only 23.6% (range 15% to 35%). Four of the seven patients did not undergo concurrent revascularization because of profound left ventricular dysfunction (EF 18%) combined with absent symptoms and negative results of a thallium study in one patient, a diffusely calcified aorta in one patient, an EF of 15% in a third patient, and prior successful revascularization in a fourth patient. The combination of left main CAD and malignant arrhythmias is particularly ominous, especially if revascularization cannot be performed. DISCUSSION
SCD affects 300,000 to 400,000 individuals per year in the United States, with a 2-year recurrence rate of 40%. 3 Although significant multivessel CAD and a history of a remote myocardial infarction are present in most, only 15% to 30% of such patients had evidence of an acute myocardial infarction at the time of their SCD episode. 1-3,5
The most common cause of SCD in patients with atherosclerotic CAD is scar resulting from a previous infarction. This scar leads to abnormal and heterogeneous conduction and refractoriness, enhanced automaticity, and a fragmented electrogram, 3 which is the anatomic and electrophysiologic substrate of sustained monomorphic ventricular tachycardia. This reentrant circuit does not require an ischemic trigger, although ischemia may be present in some cases. Prearrest angina is not a common finding in SCD unassociated with myocardial infarction, and in pathologic studies of patients having SCD, acute coronary lesions are often absent. 1,6 Fortuitous Holter monitor recordings immediately preceding an episode of SCD have infrequently revealed premonitory ischemic S-T changes. 3 Meissner, Akhtar, and Lehmann 1 estimated that nonischemic sustained monomorphic ventricular tachycardia accounts for 20% to 40% of sudden tachyarrhythmic cardiac deaths unrelated to acute infarction and that these ventricular tachycardias often degenerate into ventricular fibrillation.
Despite the predominance of myocardial scar as an anatomic and physiologic substrate, ischemia may be a cofactor in some cases of SCD. For example, the circadian variability of SCD correlates with the increased morning incidence of acute ischemic cardiac events. 2,3,7 Meissner, Akhtar, and Lehmann 1 estimated that one to two thirds of SCD survivors have both an arrhythmogenic scar and ischemia. Pure ischemic SCD occurs in only 10% to 15% of patients with CAD and is associated with a higher incidence of angina, noninducibility, absence of a myocardial scar, and favorable response to revascularization. 1
Some inferences regarding the association among ischemia, revascularization, and SCD can be made from several prospective randomized trials conducted in the United States and Europe. Holmes and associates 8 reported on the prevalence of SCD in the Coronary Artery Surgery Study (CASS) registry. SCD occurred in 5.2% of 6260 medically treated patients compared with 1.8% of 7216 surgically treated patients. The difference was statistically significant for patients with two- and three-vessel disease and no congestive heart failure, but was particularly striking in patients with two-vessel disease and congestive heart failure (98% surgical versus 83% medical, p < 0.001) and three-vessel disease with congestive heart failure (91% surgical versus 69% medical, p < 0.001). Ten-year follow-up of the randomized 780 patients in the Coronary Artery Surgery Study revealed a higher percentage of myocardial infarction and SCD in medically treated patients, but the difference did not reach statistical significance. 9 In the European Prospective Randomized Coronary Surgery Study, 10 which excluded patients with poor left ventricular function, the cumulative risk of SCD at 8 years was 3% in the surgically treated group and 9% in the medically treated group (p = 0.00046).
Several nonrandomized studies also have noted the impact of coronary revascularization on survival after SCD. Wilber and associates 5 performed a multivariate analysis of risk factors for recurrent cardiac arrest in 166 survivors of SCD not associated with myocardial infarction. Absence of coronary revascularization carried a relative risk of 4.2 (p = 0.05). Every and associates 11 reported on 265 patients resuscitated from SCD caused by ventricular fibrillation, 32% of whom underwent CABG and 68% of whom were treated medically. Preoperatively, the surgically treated group had a 58% incidence of angina compared with 35% in the medically treated group (p < 0.01), a higher EF (51.7% versus 41.2%, p < 0.001), and a lower incidence of acute myocardial infarction (17% versus 32%, p < 0.001). CABG reduced the risk of recurrent SCD (risk ratio 0.48, p < 0.04) and total cardiac mortality (risk ratio 0.65, p = 0.10). Bolooki and associates 12 found that concomitant revascularization increased the shock-free survival of ICD recipients during a 6-year follow-up compared with patients who did not undergo revascularization (41% versus 21%). The rate of subsequent SCD was lower (2.39% per patient-year versus 3.5% per patient-year).
Physiologic tests for myocardial ischemia provide additional information supporting some relationship between ischemia and SCD. In a report from the Coronary Artery Surgery Study, Weiner and colleagues 13 studied the incidence of myocardial infarction and SCD over 7 years, comparing patients with silent as opposed to symptomatic ischemia during exercise testing. Compared with results in control patients, who had a 98% 7-year survival free of myocardial infarction and SCD, the 74% 7-year survival of patients with asymptomatic ischemic changes and 77% survival of patients with symptomatic ischemic changes on exercise testing were substantially lower (p<0.001). However, it is not entirely clear what proportion of SCDs had an arrhythmic origin. In the subset of patients with three-vessel coronary disease, patients with symptomatic ischemia had a 93% 7-year SCD free survival as compared with 79% in the asymptomatic group (p = 0.001). Despite these data, it must be acknowledged that when SCD survivors are monitored during exercise, most studies have demonstrated a low incidence of exercise-induced ischemia and arrhythmias. 14-16
Finally, several investigators have focused on patients with malignant rhythms other than sustained monomorphic VT that are more likely of a purely ischemic origin. These data suggest that primary ventricular fibrillation 17 and polymorphic ventricular tachycardia 18 are more often ischemic in origin and may respond to revascularization, but the risk of recurrent arrhythmias is not always eliminated. 19
With the advent of ICD therapy and subendocardial resection, the available treatment modalities to prevent recurrent SCD have expanded dramatically. In experienced centers, patients with mappable arrhythmias, left ventricular aneurysms, and good function of the residual myocardium may undergo subendocardial resection, with or without bypass or ICD, with a mortality under 10%. 20,21 However, many patients do not meet the criteria for "curative" therapy and are treated with an ICD. Given the inconclusive data presented on the role of ischemia in SCD, the indications for revascularization in these patients are unclear.
Some evidence suggests that revascularization may have a salutary effect in selected ICD recipients. Levine and associates 22 studied 197 patients who survived ICD implantation with or without CABG. Cox multivariate analysis revealed that revascularization was associated with later ICD discharge and improved survival after ICD discharge. Wilber and associates 23 studied 64 patients receiving ICDs. Of 49 patients with CAD, 28 underwent ICD implantation plus concomitant coronary revascularization. Lack of concomitant coronary revascularization was a univariate and multivariate (relative risk, 2.9) predictor of early ICD discharges.
Despite the evidence supporting a salutary effect of revascularization, a decision to add concomitant CABG may significantly increase the immediate perioperative risk. This increased risk may result from the longer operative time, extensive defibrillation testing at the end of cardiopulmonary bypass when the heart has already sustained ischemic injury, or from the potential pulmonary and cardiac effects of amiodarone hydrochloride (Cordarone, Wyeth-Ayerst Laboratories, Philadelphia, Pa.). 24 In the series of Elefteriades and associates, 25 operative mortality for ICD alone was 3.3% and for ICD plus CABG, 10.7%. Pinski and associates 26 reported a postoperative mortality rate of 11.8% for combined CABG plus ICD and 8.9% if the two procedures were staged. Gohn and associates 27 reported a mortality rate for isolated ICD placement of 3.3% in 183 patients and 6.8% in 88 patients who underwent concomitant heart operations (p = 0.15). In recent work from the planning phase of the CABG Patch Trial, Bigger 28 reported more favorable results in a series of 125 patients. Operative mortality was 4.8% for CABG plus ICD. Bolooki and associates 12 reported an operative mortality of 2% among 43 patients undergoing ICD and CABG and 2.9% among ICD recipients without CABG.
Infection is also more common after combined ICD-CABG, whether an entire ICD system or only patches are placed, 29 and may be the result of patch implantation in a surgical wound that has been open for many hours. Infection also may result from postbypass immunologic depression. 24
In our series, ICD recipients were chosen for revascularization primarily on the basis of the angiographic extent of CAD, objective and subjective evidence of ischemia, and availability of surgical targets. Revascularization combined with ICD placement was accomplished with low mortality and morbidity, with the exception of a higher incidence of inotropic use and neurologic events. Long-term survival was better in patients undergoing revascularization than in the "high-risk" group B2 patients.
These findings suggest that revascularization may be a valuable adjunct in the treatment of malignant arrhythmias in patients with CAD. Revascularization may decrease the nonarrhythmic complications of CAD and may favorably influence the incidence of those arrhythmias that have either a purely ischemic basis or a combination of an ischemic and a reentrant mechanism.
Our study does not address the question of how to treat patients who have primarily ischemic symptoms but who are thought to be at high risk for the development of malignant arrhythmias. 30 This subject is being investigated in the CABG Patch Trial. 28
Acknowledgments
We thank Gerald Heatley, MA, for his help with statistical analysis.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
