| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2002 European Association of Cardio-Thoracic Surgery
Beneficial effects of coronary revascularization in patients with ischaemic left ventricular dysfunction with and without anginal symptoms
a Nuclear Cardiology, CNR Institute of Clinical Physiology, Via Moruzzi 1, 56124 Pisa, Italy
* Corresponding author. Tel.: +39-50-315-2153; fax: +39-50-315-2151 Received February 4, 2002; received in revised form April 2, 2002; accepted April 10, 2002
Therapy for ischaemic congestive heart failure has been well documented in patients with angina. The goal of this study was to compare the benefit of revascularization in patients with and without chest pain. A series of 180 patients with ischaemic heart failure symptoms (New York Heart Association III–IV class) and low ejection fraction (28±9%) were recruited and followed for 3 years. Group A, 97/180 patients, had chest pain. Group B, 83/180 patients, did not have angina. The two groups did not differ with respect to known determinants of postinfarction prognosis. The relative presence of viable tissue versus scar was defined by Thallium-201 uptake. Intraoperative mortality was 5 and 7% in Groups A and B (P=not significant); in particular, in both groups, it was lower when only patients with mostly viable myocardium were considered. At 6 months, the presence of viable myocardium was highly predictive of improvement of heart failure symptoms and wall motion abnormalities. At 3 years, revascularized patients of Group A with mostly viable myocardium had a survival of 89% compared to 87% for corresponding Group B patients (P=not significant). In conclusion, similarly to patients with angina, patients with left ventricular dysfunction, maintained viability and without anginal symptoms may benefit from coronary revascularization.
Key Words: Ischemic heart failure; Viability; Prognosis; Revascularization
Therapy for congestive heart failure is directed at both relief of symptoms and preventing further deterioration of function. In patients with severe ischaemic left ventricular dysfunction [1–4] coronary revascularization improves symptoms and provides a better outcome than medical therapy, with similar and occasionally better survival than that observed in transplant populations. However, most patients in these studies had angina rather than heart failure as their major symptom [1–4]. On the other hand, in patients with ischaemic heart failure but without limiting angina, previous studies suggested that coronary artery by-pass grafting (CABG) [5,6] is associated with a poor short term outcome, with a perioperative mortality rate as high as 15–20%. Bonow [7] recently suggested that, besides improvement in chest pain, reduction in heart failure symptoms should be used to assess the clinical benefit of revascularization. This approach was supported by Bax et al. [8] who reported that patients with four or more viable segments on FDG SPECT had improvement in left ventricular ejection fraction and heart failure symptoms postoperatively. Moreover, Pagano et al. [9] provided evidence of better outcome after CABG also in severe ischaemic dysfunctioning patients with mostly viable myocardium detected with FDG PET and without anginal symptoms. However, both studies enrolled small cohorts of patients, and thus the observations cannot be extended to all patients with cardiac heart failure without angina. The goal of this study was to evaluate the effect of revascularization in heart failure patients with and without angina pectoris. Data regarding the effectiveness of revascularization in patients without chest pain is important since multivessel coronary artery disease often leads to CABG [3,9–14].
The study population consisted 180 consecutive patients who were scheduled for revascularization between January 1, 1995 and December 31, 1997. Patients were recruited from 3 medical centers. All patients met the following criteria: a III–IV New York Heart Association (NYHA) class, an angiographic ejection fraction below 35% and rest-redistribution SPECT Thallium-201 scintigraphy. On the basis of presence or absence of angina, patients were divided in two groups: Group A, 97/180 patients experiencing chest pain at rest, on effort or both; and Group B, 83/180 patients with only heart failure symptoms. All patients had coronary angiography in multiple projections and biplane left ventriculograms within 1 month of the scintigraphic studies. The extent of angiographic coronary artery disease was based on the criterion of >75% stenosis in the proximal coronary arteries of the major branches and >50% for left main coronary disease. The patient groups were further subdivided on the basis of their Thallium-201 scan, into those with a majority of viable myocardium and those with a dominance of scar. Patients with recent myocardial infarction (<3 months), valvular heart disease or cardiomyopathies, previous CABG or percutaneous transluminal coronary angioplasty (PTCA) were excluded from the study. 2.1. Thallium-201 scintigraphy All patients underwent a rest-redistribution SPECT Thallium-201 study. Images were obtained in all patients 10 min and 4 h after the injection of 2.5–3.5 mCi of Thallium-201. Images were quantitatively analyzed for maintained Thallium-201 uptake using a bull's eye plot based on the 11-segment model. To define viable or necrotic segments, Thallium-201 uptake was assessed in each of the 11 segments using a cut-off of 54% of the peak [15]. Assignment of the apex to a specific coronary territory was based on coronary anatomy as well as on the spatial distribution of adjacent defects. The attribution of defects to single diseased vessels has been validated elsewhere [16]. In order to quantify the amount of viable myocardium, myocardial viability was assessed according to a continuous parameter defined as a percentage of maintained Thallium-201 uptake in 11 segments such as proximal and distal anterior, septal, inferior, lateral and posterior walls and the apex.2.2. Echocardiography Commercially available wide-angle phased array imaging systems were used. The left ventricle was divided into the same 11 anatomical segments used for Thallium-201, and the bias deriving from single observers in assigning vascular beds was reduced by the Bull's eye plot as previously described for Thallium-201. Segmental wall motion was graded as: normal (score=1), hypokinetic (score=2), akinetic (score=3), dyskinetic (score=4). A wall motion score index (WMSI) was derived by dividing the sum of individual segment scores by the number of interpretable segments [17].Baseline echocardiography was performed at the enrollment and 6 months after revascularization.2.3. Coronary artery by-pass grafting Patients underwent CABG under mild hypothermic (34°C) cardiopulmonary bypass using a membrane oxygenator. In all cases, distal anastomoses between the bypass graft and native coronary arteries were performed under aortic cross-clamp on a heart arrested by blood cardioplegia. In the majority of patients, cardioplegic solution was infused into the aortic root anterogradely. The distal anastomoses were performed during an aortic cross clamp period. The proximal anastomoses for vein graft were completed with side-biting clamp and the cross-clamp period. During cardioplegic arrest, metabolic repletion of the arrested heart was maintained by regular infusion of blood cardioplegia every 15–20 min. Intermediate distal stenoses were not considered as a contraindication to surgical graft. Adverse events occurring in the postoperative phase were always recorded.2.4. Percutaneous transluminal coronary angioplasty Angioplasty was performed through the femoral artery. At the commencement of the angioplasty procedure, 250 mg of acetylsalicylic acid and 100 mg of heparin were administered intravenously, with additional boluses of 50 mg given hourly. The stenoses were dilated with an appropriately sized balloon catheter system. Twenty-two patients received intravascular stents. After completion of the procedure, heparin infusion was commenced to achieve an activated partial thromboplastin time to 2.0–2.5 times control levels for 12–24 h. Procedural clinical success was defined as a reduction in the lumen narrowing to <50% of the reference diameter by visual assessment, with no major complications within 24 h of the procedure.2.5. Follow-up Six months after enrollment, all patients were reassessed by echocardiography for wall motion score index and evaluated for symptoms by interview and examination. To determine their long-term outcome, patients were followed for an additional 36 months. Survival status was determined by contacting all patients or next of kin by telephone. Data have been obtained from one of the following sources: physician interview, review of hospital records, or death certificate. Cardiovascular deaths were defined as death from stroke, acute myocardial infarction, refractory congestive heart failure and any sudden, unexplained death. Cardiac events during late follow-up were defined as cardiac death and cardiac transplantation for refractory heart failure.2.6. Statistics: data analysis Baseline characteristics of medically treated and revascularized patients were expressed as mean±1 standard deviations. A P value of <0.05 was considered statistically significant. Kaplan–Meier survival estimates were used to describe mortality patterns in the overall population of medically and revascularized treated patients [18]. Significant differences ( by two tailed tests) in survival between the medical and surgical groups were tested by the log Rank test.
To assess the association of each variable with survival while controlling for the effects of the other variables, a Cox proportional-hazard regression was performed [19]. This constitutes the multivariate analysis. For the regression model, a forward stepwise procedure was used to select significant variables from among the following: age, sex, diabetes, hypertension, number of vessel disease, ejection fraction, WMSI, presence of viable myocardium. The significant level to enter the model was Follow-up time for medically treated patients was started at the date of Thallium-201 scintigraphy. In revascularized patients follow-up time started at the date of operation. Receiver operator characteristics (ROC) curve was used to determine this ‘optimal’ cut-off value for the prediction of late events, with respect to the percentage of maintained Thallium-201 uptake at rest in the 11 segments.
A total of 180 patients were enrolled in the study. Ninety-seven patients with anginal symptoms were classified in Group A, while the remaining 83 patients with heart failure symptoms only in Group B. Baseline characteristics of Groups A and B are presented in Table 1. Comparison of the two groups of patients demonstrated that many of the baseline clinical descriptors were distributed similarly.
In Group A, 59 patients were revascularized by CABG and 38 by PTCA. The average number of coronary artery by-pass graft was 2.8±1.3, and 54/58 patients received an internal mammary graft on the left anterior descending coronary artery. The average number of PTCA was 2.1±0.9; 32 patients received a PTCA on the left anterior descending coronary artery, 12 on the left circumflex artery and 11 on the right coronary artery. Twenty six patients received an intravascular stent, 20 on the left anterior descending coronary artery and six on the right coronary artery. In Group B, 54/83 patients were revascularized by CABG (2.9±1.1 mean of graft per patient, 51 internal mammary artery on left anterior descending coronary artery), while the remaining 29/83 patients by PTCA (mean 1.9±1.2 procedure per patient, 19 stents on left anterior descending coronary artery). 3.1. Myocardial viability and regional wall motion Since we used a cut-off value determined by the ROC analysis for the detection of myocardial viability, we translated a continuous variable into a dichotomous one: when eight/11 segments had maintained Thallium-201 uptake, about 70% of the myocardium was viable (Fig. 1).
The mean number of viable segments on Thallium-201 was 8.5±1.8 and 8.4±2.3 per patient in Groups A and B, respectively, while the resting WMSI was 2.7±0.4 and 2.8±0.2 in the two groups (P=not significant). According to the cut-off identified by ROC curves, 66/97 and 63/83 patients of Groups A and B showed mostly viable myocardium, while the remaining 31/97 and 20/83 had dominance of scar in dyssynergic segments. 3.2. Early post-revascularization outcome In hospital mortality was 5% in patients of Group A and 7% in Group B (P=not significant). Causes of death were ventricular arrhythmias in two and two patients and cardiogenic shock in three and four patients in Groups A and B, respectively. When considering only patients with mostly viable myocardium, intraoperative mortality was 3% (two/66 patients) in Group A and 3% (2/63) in Group B (P=not significant). In patients with mostly necrotic myocardium the intraoperative mortality was 10% and 20% in patients of Groups A and B, respectively ( versus mostly viable patients). 3.3. Short-term follow up Regional contractile function assessed by echocardiography and improvement of symptoms were evaluated in all patients after 6 months. In patients of Group A with maintained myocardial viability, WMSI decreased from 2.7±0.6 to 1.8±0.5 () and NHYA class changed from 3.4±0.3 to 1.6±0.6 (). All but three patients referred the absence of anginal symptoms. In patients of Group B with maintained myocardial viability, WMSI decreased from 2.8±0.5 to 1.9±0.7 () and NHYA class changed from 3.5±0.5 to 1.6±0.8 (). In patients of Groups A and B with dominance of scar, WMSI after revascularization was not statistically different from the control value before revascularization (2.8±0.8 and 2.9±0.5 in the baseline characterization versus 2.9±1.0 and 2.8±0.7 post-revascularization, P=not significant, respectively, in Groups A and B). 3.4. Long-term follow-up All patients were treated with combinations of drugs occurring to individual clinical conditions and degree of cardiac heart failure.Sixteen deaths occurred during the follow-up for cardiac causes (nine in Group A and seven in Group B) and two patients of Group A and one of Group B had heart transplantation. When considering only patients with mostly viable myocardium, five and five deaths occurred in patients of Groups A and B. 3.5. Univariate and multivariate survival results Univariate analysis indicated that maintained myocardial viability ( ,  ) and diabetes (,  ) were independent predictors of survival. The other variables analyzed were not significant in determining survival at the univariate analysis. In the stepwise analysis, the independent predictor was the presence of myocardial viability, exerting a protective effect on survival (chi square 4.7, 95% C.I.,  ). 3.6. Kaplan–Meier survival curves Fig. 2 shows the Kaplan–Meier survival curves of Groups A and B patients. The 3 years survival was 84 and 78% (log rank=not significant) in patients of Groups A and B, respectively. Fig. 3 shows the Kaplan–Meier survival curves of mostly viable patients of Groups A and B (survival at 3 years: 89 and 87%, respectively in Groups A and B, log rank=not significant).
This study supports the hypothesis that patients with left ventricular dysfunction respond well to revascularization when they have mostly viable myocardium, regardless of whether they experience angina. Revascularization results in reduced symptoms, improved regional wall motion and a better prognosis. The extent of myocardial viability identified by rest injected Thallium-201 myocardial scintigraphy is related to survival. When coronary revascularization was undertaken in patients with a small area of viable myocardium, the procedure was linked to a higher incidence of perioperative cardiac death. 4.1. Comparison with previous studies Patients with ischaemic left ventricular dysfunction represent an important subset of heart failure population in whom myocardial revascularization offers the potential for reducing symptoms and enhancing prognosis [9–14,20]. Although this concept is well accepted, no study compared the short and long term survival of patients with heart failure symptoms with and without angina pectoris treated with revascularization, even if some studies [4,8,9] have underlined the short term benefit of revascularization in patients with heart failure symptoms.The main finding of the present study, i.e. the striking protective effect of revascularization in patients with a dominance of viable segments, is in keeping with the available evidence. This study has several peculiarities:
4.2. In-hospital mortality The severity of left ventricular dysfunction is known to influence intraoperative mortality, particularly if advanced heart failure is present. This study suggests that the extent of viable myocardium is also an important predictor of early death in revascularized patients when heart failure is present, whether or not the patient has angina. In patients with mostly viable myocardium who have coronary anatomy suitable for revascularization and heart failure symptoms without angina, revascularization can be performed with an acceptable mortality. On the other hand, due to the high intraoperative mortality, patients with mostly necrotic myocardium should be carefully evaluated for alternative therapies.4.3. Improvement of heart failure symptoms and left ventricular function In this study, symptoms of heart failure improved in patients with an improvement of regional left ventricular function. Moreover, the presence of mostly viable myocardium was highly predictive of improvement of heart failure symptoms after revascularization, independently of the absence of angina at the enrollment.These results are in agreement with previous findings [8,21]. Bax et al. [8] and Di Carli et al. [21] showed a correlation between the magnitude of improvement in heart failure symptoms, improvement of left ventricular function and the extension of pre-operative myocardial viability. Moreover, Marwick et al. [22] reported a link between improvement of exercise capacity and extent of jeopardized tissue defined by positron emission tomography. In patients with mostly necrotic myocardium there was no wall motion improvement after revascularization, even if the mean WMSI at the enrollment was similar in the two groups of mostly viable and mostly necrotic patients. These findings underscores that a substantial amount of viable myocardium (70% of the left ventricle) needs to be present to result in a significant improvement in regional wall motion abnormalities. Moreover, these results were observed independently from the presence of angina. 4.4. Long term follow-up Angina remains one of the major symptoms in the clinical decision making for revascularization. Moreover, our follow-up data, supported by telephone contact with the patients or next of kin, indicate that the absence of chest pain is not a reason to only consider medical therapy in patients with heart failure. These patients should be evaluated for the extent of viable myocardium (determined on rest redistribution imaging), as a major criteria for the treatment decision. Finally, our follow-up was based on symptoms and survival and the absence of significant differences among the two groups is not supported by relative long-term changes of ejection fraction or similar quantitative data.4.5. Clinical implications Patients with chronic coronary artery disease, severe left ventricular dysfunction and heart failure symptoms represent a common, and increasingly frequent management problem. Documentation of myocardial viability with Thallium-201 scintigraphy provides important diagnostic, prognostic and therapeutic information. From the diagnostic viewpoint, Thallium-201 scintigraphy identifies myocardial viability and a high likelihood of recovery after revascularization. From the prognostic viewpoint, the impact of revascularization in dysfunctioning patients and heart failure symptoms is associated with mostly viable myocardium and is quite similar to that obtained in revascularized dysfunctioning patients with mostly viable myocardium and angina. Finally, dominance of necrosis should be carefully quantified for an increased intraoperative risk.4.6. Limitations of the study For this study, patients were recruited from three medical centers, with a non-randomized non-blinded designed protocol. Although this allowed an expanded database, it also introduced an obvious bias in decision making linked to single institutions but hampered by our inclusion criteria.Only rest-redistribution Thallium-201 SPECT was used to assess myocardial viability and no data on stress ischaemia were available to further characterize the two populations with and without angina. Thus, the question whether angina corresponds to a higher degree of stress ischaemia and whether patients without anginal symptoms have a significant amount of reversible defects cannot be answered in this study. Similarly, the functional status was evaluated subjectively and no accurate assessment of functional class and exercise tolerance was available in this study. WMSI was controlled after revascularization only at 6 months, and thus long term evaluation of these variables was not available. Accordingly, the improvement of outcome observed in patients undergoing CABG and PTCA could not be correlated to long lasting improvements of regional and global left ventricular function. Moreover, similarly to the majority of published papers on viability and outcome, patency data were not available in all our patients. Our study group of revascularized patients included both surgical and PTCA revascularization. Although differences and similarities have been demonstrated above, the two groups do not allow further conclusions from the two groups. Finally, all patients received the best association of drugs available in single centres, and no standardization of medical treatment was defined in the inclusion criteria.
ICVTS on-line discussion Author: Dr. Paul Vogt, Division of Cardiovascular Surgery, Giessen University Justus Liebig, Rudolf-Buchheim-Straôe 8, Giessen, Germany Date: 07-Aug-2002 11:21 Message: The authors compare the outcome of 180 NYHA class III or IV patients with left ventricular dysfunction with and without preoperative anginal symptoms undergoing either coronary artery bypass grafting or PTCA. Left ventricular dysfunction was defined as LV ejection fraction less than 35%. The authors report a similar hospital mortality and three-year survival in patients undergoing revascularisation regardless of the preoperative presence of angina or dyspnea, provided that a distinct amount of viable myocardium was present. This paper deals with a question, which can be intuitively answered, namely that the outcome of coronary artery patients with severe LV dysfunction, undergoing either PTCA or coronary revascularisation, does not depend on the preoperative clinical symptoms but on the amount of viable, hibernating myocardium present at the time of revascularisation. Hence, patients with severe LV dysfunction without preoperative anginal symptoms undergoing coronary artery bypass grafting do have the same outcome as those presenting with dyspnea, but not with angina pectoris. There are several interesting questions in this paper, which remain unanswered. What should we do with patients presenting with severe left ventricular dysfunction and mostly necrotic myocardium? According to the results presented in this study, patients with large myocardial scars had a 3-6 times higher perioperative mortality compared to those where viable myocardium has been demonstrated preoperatively using FDG PET evaluation. Should conventional coronary artery bypass grafting be offered to patients with severe left ventricular function and mostly necrotic myocardium according to a preoperative PET examination? For these patients, would off-pump coronary surgery supported by perioperative intra-aortic balloon pumping be a better option decreasing hospital mortality? Or would PTCA of the culprit lesion be the treatment of choice? And at least, can we achieve a survival benefit in these patients when looking on long-term data, which are not presented in this paper for this subgroup of patients. There are other interesting questions. With regard to cost-effectiveness, does each patient need FDG PET to demonstrate viable myocardium or is there a subgroup of patients with severe left ventricular dysfunction which can safely undergo coronary artery bypass grafting without preoperative FDG PET, probably those patients with preoperative anginal symptoms and three-vessels disease where all three major coronary arteries can be bypassed? Overall the paper confirms the current opinion that patients with severe left ventricular dysfunction should be evaluated for conventional surgical treatment of their end-stage heart failure to reduce the number of patients dying on cardiac transplant lists. Response Author: Dr. Alessia Gimelli, Nuclear Cardiology, Institute of Clinical Physiology, CNR, via Moruzzi, 1, Pisa, Italy Date: 12-Aug-2002 11:36 Message: From a statistic point of view, the number of patients enrolled in our study with mostly necrotic myocardium is too small for providing definite conclusions in terms of short and long term outcome. However, to obtain a better evaluation of the cardiac status of these patients and hence to decide the better treatment for them, probably we need additional information beyond viability, such as the end diastolic diameter, the end diastolic volume (as suggested in a previous study by Chan and colleagues), and the mitral valve condition. Moreover, in these patients with mostly necrotic myocardium well designed randomized clinical studies should be performed in order to obtain indications about the use of off pump coronary surgery plus perioperative intraaortic balloon pumping versus PCI of the culprit lesion for decreasing hospital mortality. Talking about cost-effectiveness, quantitative gated SPECT probably represents the best tool for obtaining information on site and extension of myocardial viability as well as onb regional and global left ventricular ejection fraction. This technique can be applied on a large scale basis in the majority of patients with ischemic heart failure to assist the decision making. Author: Dr. Brigitte Osswald, Department of Cardiac Surgery, University of Heidelberg, Im Neuenheimer Feld 110, Heidelberg, Germany Date: 10-Aug-2002 18:21 Message: The authors present a very interesting work on the outcome of patients with impaired LV function, a growing part of patients in cardiac surgery. These patients are revascularized by either PTCA or CABG. They intended to answer the question, whether the presence or absence of angina prior to the first intervention represents a predictive value for the further outcome. There are some remarks about the endpoints of the study, since in the Material section, a refined definition of cardiovascular death and cardiac events is given, however, in the results, only "survival" is presented. The completeness of follow-up may be 100%, however, it should be stated somewhere. The authors focus on angina, therefore, one could expect to obtain not only information of the average number of PTCA’s in both groups of 2.1 and 1.9 respectively, but to have a look at the time to reintervention (redo PTCA and redo CABG) and, even more important, the angina-free survival. The authors could not find statistically significant differences between PTCA and CABG, however, the number of interventions, as well as long-term results (5 years and more) may, even within the small patient group, have some impact on the outcome, not only in terms of the overall mortality, but concerning life quality and functional status. So, one has to congratulate the authors for their data, which may also serve as a valuable basis for a further evaluation of medical, interventional and surgical concepts in this important and steadily growing patient group.
We are grateful to Prof. William Strauss from Stanford University, California, U.S.A., for his help in reviewing the paper. We are also grateful to Mrs Ilaria Citti and Nadia Sereni for their help in collecting data. PII: S1569929302000026
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
Top
Abstract
1. Introduction
. 
