| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;83:468-474
© 2007 The Society of Thoracic Surgeons
Cattedra di Cardiochirurgia, Università degli Studi di Milano, IRCCS MultiMedica, Sesto San Giovanni, Milan, Italy
Accepted for publication September 6, 2006.
* Address correspondence to Dr Pocar, Via Pompeo Litta 2, 20122 Milan, Italy (Email: marco.pocar{at}unimi.it).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
METHODS: We retrospectively analyzed 45 consecutive angina-free patients with ischemic left ventricular dysfunction (ejection fraction 
0.35) and heart failure (New York Heart Association functional class III to IV) who were selected for CABG between 1988 and 1995. Positron emission tomography was used for preoperative identification of myocardial viability.
RESULTS: The 30-day mortality was 4.4%. At a median follow-up of 117 months (longest observation, 205 months), the probability of survival at 1, 5, 10, and 15 years after CABG was 93.3%, 84%, 65%, and 44%, respectively. At multivariable analysis, a left ventricular end-diastolic pressure (LVEDP) of 25 mm Hg or more predicted a threefold increase of the hazard of death (p = 0.02), whereas a LVEDP of 20 mm Hg or more correlated with the requirement of an intraaortic balloon pump perioperatively (p = 0.04). Other independent predictors of survival were age older than 70 years and peripheral vascular disease. Cardiac events accounted for 88% of late deaths, which were primarily related to sudden death or progressive heart failure. Most patients were in New York Heart Association functional class I to II at late follow-up.
CONCLUSIONS: CABG alone yields good long-term outcome in selected angina-free patients with ischemic systolic dysfunction and advanced heart failure. However, associated diastolic impairment, reflected by elevated LVEDP, predicts reduced long-term survival despite myocardial viability.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
This retrospective study focuses on long-term survival, up to 17 years, after complete revascularization in patients with ischemic heart failure and assessment of myocardial viability. The impact of preoperative and surgical variables is also analyzed.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
No patient was supported with intravenous inotropic agents or intraaortic balloon pump (IABP) counterpulsation preoperatively, either for low cardiac output or prophylactically. Early and late mortality, respectively, denote events occurring during or after the first 3 postoperative months. Characteristics and operative variables are outlined in Table 1. This retrospective study was approved by the Institutional Review Board, and individual consent was obtained from patients at the time of follow-up data collection.
|
Left heart catheterizations were performed at a single institution, and values of intracardiac pressures refer to measurements before contrast medium injection and coronary angiography. Patients were scheduled for CABG provided that the left anterior descending branch was judged technically graftable.
Surgical Procedures
All operations were performed on moderately hypothermic (30°C core temperature) cardiopulmonary bypass with blood antegrade and retrograde cardioplegia, normothermic induction, cold maintaining doses every 20 minutes, and substrate-enriched controlled reperfusion, according to Buckberg’s principles for energy-depleted hearts. The left ventricle was vented through the right superior pulmonary vein. Stenoses of 50% or more were bypassed.
Follow-Up
Data were obtained from the patient or patient’s family and referring physicians or cardiologists directly or by phone interviews. Follow-up was complete at March 31, 1996, although 42 patients (93.3%) could be traced at June 30, 2005. Median follow-up, including hospital deaths, was 117 months (longest observation, 205 months).
Statistical Analysis
The t test was used to compare means between groups. The 
2 test, or the Fisher exact test when 2 x 2 tables had a cell with an expected frequency of less than 5, and stepwise logistic regression were used for univariable and multivariable analysis. Variables showing a trend toward statistical significance (p < 0.15) at univariable analysis were subsequently evaluated in the multivariable analyses. Survival was estimated by the Kaplan-Meier method. All 95% confidence intervals (CIs) were calculated as ± 2 SEs. Differences in probability estimates were calculated using the log-rank test. Cox’s proportional hazard model was used for multivariable analysis of time-dependent variables. Unless otherwise specified, continuous variables are expressed as mean ± SD. Data were analyzed using SPSS 11.5.1 (SPSS Inc, Chicago, IL) for Windows (Microsoft Corp, Redmond, WA). Values of p < 0.5 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
|
|
Predictors of death and low cardiac output requiring IABP are listed in Table 2. The only independent predictor of overall death other than LVEDP was age. Peripheral vascular disease (claudication, carotid occlusion or >50% stenosis, previous or planned intervention on the abdominal aorta, limb arteries or carotids) correlated with death beyond 3 months, but this was not observed when early events were included in the multivariable analysis.
|
Operative Variables and Perioperative Complications
There was no occurrence of incomplete revascularization. Saphenous vein grafts alone were constructed in 2 patients (4.4%) only. No arterial conduit other than the internal thoracic artery was used. High-dose inotropic support (intravenous dopamine at >6 µg/[kg · min], epinephrine, or phosphodiesterase inhibitors) was used in 21 patients (47%) and IABP in 13 (29%). A LVEDP of 20 mm Hg or more (29 patients, 64%) independently predicted perioperative IABP (odds ratio, 10.2; p = 0.04). Only 1 (6.7%) of 15 patients with a LVEDP of less than 20 mm Hg versus 12 (41%) of 29 with a LVEDP of 20 mm Hg or more required IABP. Two of the latter required delayed sternal closure for hemodynamic instability. Mechanical respiratory support was prolonged beyond 48 hours in 11 instances (24%). In 9 patients (20%), perioperative renal dysfunction developed, defined as a serum creatinine rise of more than 2 mg/dL or more than 1.5 baseline value in patients with preexisting renal dysfunction.
Adverse Late Cardiovascular Events
Two patients underwent reoperation dictated by recurrence of heart failure. Both were alive at late follow-up, but clearly represented failures of primary CABG. One underwent transplantation 56 months after CABG. In another, secondary mitral regurgitation developed and the patient was reoperated on twice, at 36 and 84 months after CABG. Valve dysfunction was initially corrected with ring annuloplasty (Carpentier-Edwards Classic, Edwards Lifesciences, Irvine, CA) and subsequently with implantation of a mechanical prosthesis (St. Jude Medical, Inc, St. Paul, MN).
Three patients underwent percutaneous coronary intervention, whereas 4 required implantation of a permanent pacemaker. Among the latter, atriobiventricular resynchronization was applied in 3 patients, and a defibrillator was implanted in 2. Extracardiac vascular events were recorded in 4 patients, including permanent stroke in 2, carotid endarterectomy in 1, and endovascular descending thoracic aortic aneurysm exclusion in 1.
Among 40 operative survivors traced in 2005 (including late deaths), 18 (45%), 12 (30%), 3 (7.5%), and 7 (17%) were in NYHA I, II, III, and IV, respectively, at time of follow-up or death. No independent predictor could be outlined, but a LVEDP of 25 mm Hg or more correlated with a worse late functional status at univariable analysis (p = 0.01), whereas no patient with a LVEDP of 25 mm Hg or more improved to NYHA I. Furthermore, all late sudden deaths occurred in NYHA I or II patients. Among 23 survivors at late follow-up, NYHA class was I in 12 patients, II in 8, III in 2, and IV in 1. At echocardiography, the ejection fraction was 0.386 ± 0.054 (p < 0.001), whereas at least moderate mitral regurgitation was documented in 5 patients, of whom 3 were in NYHA II, and 1 each was in III and IV. Irrespective of mitral dysfunction at late follow-up, regurgitation was also outlined in case of cardiac resynchronization (n = 3) or mitral surgery (n = 1). Finally, angina developed in 9 patients, dictating percutaneous revascularization in 3, and 2 patients died of myocardial infarction.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Survival and Predictors of Death
This retrospective analysis confirms that good long-term survival—beyond 10 years after CABG—can be anticipated in patients with advanced ischemic heart failure and residual viability, irrespective of concurrent angina. In our experience, the probability of survival at 15 years for the entire cohort was 44%, which rose to 51% in patients with a preoperative LVEDP of less than 25 mm Hg. This is noteworthy when considering that during the same period, most angina-free patients with cardiomyopathy and ischemic heart failure were referred for cardiac transplantation (2 patients actually were on the waiting list), whereas survival rates obtained with transplantation during the same surgical era approach 50% at 10 years [14]. However, a true comparison is impossible because transplant registries also include more critical patients (eg, circulatory support) that account for a higher mortality during the first 6 to 12 months.
Conversely, outcome is more disappointing in patients with a preoperative LVEDP of 25 mm Hg or more. Although LVEDP independently correlated with survival in the Coronary Artery Surgery Study [2], this has not been elucidated specifically in case of left ventricular dysfunction and heart failure, which is often associated with higher baseline LVEDP. Extent of viability could not predict outcome and was not correlated with LVEDP in our analysis.
Interestingly, impaired flow reserve and correspondingly decreased contractile reserve, particularly in the left anterior descending and circumflex coronary arteries, has been outlined in idiopathic dilated cardiomyopathy, indicating microvascular dysfunction [15]. Because flow is predominantly diastolic in left coronary territories, a similar pathophysiologic condition is also possible in ischemic cardiomyopathy after CABG and could explain the predictive role of LVEDP for adverse events despite complete revascularization and myocardial viability. In other words, increased LVEDP seems related to the degree of left ventricular remodeling, when excluding non-elective patients.
In this context, medical therapy for heart failure and coronary artery disease, and the respective impact on recommended guidelines, has changed substantially since the period of the present study. ß-blocker therapy, which was particularly uncommon, has recently been reported to improve diastolic function in heart failure with severe diastolic and systolic impairment, both on an ischemic or nonischemic basis [16]. Four-year survival of more than 80% has been documented with ß-blockers, although the Carvedilol Or Metoprolol European Trial (COMET) also included NYHA II patients enrolled a decade later [17]. On theoretic grounds, wider use of ß-blockers might improve survival after CABG in systolic heart failure patients with elevated LVEDP and represents an intriguing issue. Conversely, improved medical therapy might reduce the relative benefits of CABG, but a true comparison between treatments performed in different periods is not feasible.
These considerations stress the importance of echocardiography and the assessment of diastolic filling patterns [18], suggesting that more aggressive approaches with respect to additional procedures might be justified in patients with a higher LVEDP. In this context, improved survival has been documented with left ventricular restoration, cardiac resynchronization, and implantable defibrillators [19–22], but respective benefits are under investigation in controlled trials. Other procedures, such as cardiac constraint and stem cell transplantation, are also being introduced in clinical practice. Finally, although reports addressing functional mitral regurgitation are controversial, the latter and increased left atrial volume index adversely affect survival after myocardial infarction [23–26]. Ongoing trials, such as the Surgical Treatment for IschemiC Heart failure (STICH) and the PET And Recovery following Revascularization-2 (PARR-2) trials, will further clarify benefits of ventricular restoration and cost-effectiveness of PET-guided CABG [7, 27].
Excluding older age, the only other independent predictor of mortality was coexisting peripheral vascular disease, even though this could be outlined only excluding early events. The predictive value of peripheral arterial disease is noteworthy when considering that the study does not comprise diabetic patients, in whom diffuse arteriosclerosis is more prevalent, since the latter did not undergo PET viability testing during the same period. More specifically, only 9 patients with impaired glucose tolerance were included. Similarly, poor coronary targets are more common in diabetic patients and correlate with worse outcome [8]. It may be inferred that survival might be less satisfactory in correspondingly similar diabetic patients.
Early Mortality
Elevated LVEDP did not independently predict early mortality in our population; however, this could be related to the small number of events and a correlation could be outlined at univariable analysis. Interestingly, both patients who died perioperatively had a LVEDP of 25 mm Hg or more, whereas a LVEDP of 20 mm Hg or more predicted the requirement of perioperative IABP at multivariable analysis. This is in accordance with studies correlating echocardiographic diastolic filling patterns with operative risk in heart failure patients undergoing CABG [28]. It may be inferred that preoperative left ventricular unloading, including IABP, should reduce the hazard of postcardiotomy low cardiac output, which was the cause of both hospital deaths. Similarly, newer-generation phosphodiesterase-3 inhibitors or calcium sensitizer agents (eg, levosimendan) represent adjunctive pharmacologic tools [29].
Causes of Late Death
Most late deaths (88%) were cardiac related. Sudden death (47%) and end-stage heart failure (40%) represented the leading causes of death, which further emphasizes the potential role of implantable defibrillators and cardiac resynchronization therapy in this population. Consequently, we now implant a left ventricular epicardial lead intraoperatively in patients eligible for resynchronization therapy. Of interest is that an increased risk of sudden death has been observed in diabetic patients, including patients with impaired glucose tolerance, showing a progressively higher risk in relation to the severity of diabetes [30]. This further suggests that our results are possibly biased by the exclusion of diabetic patients. Conversely, outcome might be more favorable in current years owing to the advances in medical therapy for heart failure. Finally, acute myocardial infarction accounted for two late deaths, and 3 patients underwent percutaneous coronary intervention during follow-up. Although we were unable to outline the benefits of bilateral internal mammary artery grafting, complications were not increased.
Late Functional Status
Thirty of 40 operative survivors were in NYHA I or II at late follow-up or death. Even when considering heart failure as a plausible cause of death in 3 patients lost from follow-up, this results in a NYHA class I or II in 30 (70%) of 43 survivors. A LVEDP of 25 mm Hg or more correlated with a worse outcome and precluded recovery to NYHA I. The incidence of sudden death in NYHA I and II patients, and progressive mitral valve dysfunction in patients in whom heart failure recurred, indicate the importance of cardioverter defibrillators and cardiac resynchronization. Stress echocardiography was seldom performed to evaluate functional mitral insufficiency at time of the study, and the importance of moderate regurgitation has become progressively evident. The relative prevalence of mitral dysfunction in patients with recurrent heart failure may thus indicate an underestimation of preoperative regurgitation, particularly in patients with elevated LVEDP.
Ejection fraction at echocardiography was improved in late survivors, and this was confirmed in a small group of PET-selected patients in whom segmental wall motion was reassessed with radionuclide angiography 6 months postoperatively [10]; however, the lack of improvement has not been associated with reduced survival [9]. Finally, angina developed in 9 patients and was successfully treated percutaneously or medically.
Limitations
Although the selection criteria exclude important confounding variables, the major drawbacks are inherent to the retrospective nature of the study. Specific echocardiographic variables, recently outlined as independent survival predictors after myocardial infarction (sphericity and left ventricular end-systolic volume indexes), were inconstantly specified, especially in earlier records. This dictated the policy to analyze systolic performance data (ejection fraction, regional wall motion) derived from radionuclide preoperative assessment. Although elevated LVEDP is more likely in dilated hearts, this does not allow the comparison with postoperative echocardiographic data focusing on volume changes or with reports addressing associated ventricular restoration surgery.
Conclusion
CABG alone yields good long-term outcome in angina-free patients with ischemic cardiomyopathy and advanced heart failure who are selected on the basis of residual viability. However, associated diastolic impairment, reflected by elevated LVEDP, predicts reduced long-term survival despite myocardial viability. Because PET cannot be routinely proposed, this report indirectly stresses the role of rest and stress test echocardiography.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
