Revascularisation for Chronic Heart Failure: A Valid Option?

Coronary artery disease is the commonest cause of heart failure in the western world accounting for up to 60% of cases 1. Although the benefits of coronary artery surgery in patients with angina or evidence of exercise-inducible ischaemia have been defined, the role of revascularisation in patients with post-ischaemic heart failure remains unclear. There is a tendency to equate LV dysfunction with heart failure, and although the two conditions are often associated, there are many patients with markedly depressed LV function without significant heart failure. From a clinical point of view, therefore, we must distinguish between two possible clinical scenarios: (1) patients with severe LV dysfunction and predominant symptoms of angina; and (2) patients with severe LV dysfunction and predominant symptoms of heart failure. The coronary surgery trials enrolled patients with EF greater than 35%, therefore, the results of surgical and medical therapy in patients with coronary artery disease and more severe LV dysfunction can only be extrapolated from non-randomised and/or retrospective reports. Data from 651 patients with left ventricular ejection fraction ≤35% enrolled in the CASS registry 2 indicated a better 5-year survival rate for the surgical group (68%) compared to the medical (54%) group. In addition, a greater benefit from surgery was detected in the subgroup of patients with the most compromised ventricular function (ejection fraction <0.26) with a 5-year survival rate of 63% compared to 43% for the medical group. A study from Duke University (North Carolina) 3, reported the survival rate of 710 patients with significant coronary artery disease and left ventricular dysfunction (ejection fraction ≤40%) treated medically (n=400) or surgically (n=310). The 3-year survival rate in the surgical group (86%) was significantly higher than in the medical (68%) and the advantage of coronary revascularisation over medical therapy was more evident in the patients with the most compromised ventricular function. A number of other reports confirmed that patients with impaired left ventricular function, angina, and coronary artery disease have a survival advantage if treated by surgical revascularisation, and this benefit is more pronounced in the subgroup with lower ejection fractions (Table 1). Despite several limitations, it can be concluded that the advantages of CABG over medical treatment make it reasonable to consider such patients for revascularisation, in the absence of other medical contraindications. Patients with predominant and severe heart failure symptoms constitute the group in whom the clinical decision to recommend coronary revascularisation, heart transplantation or continued medical therapy remains difficult. There are no randomised trials to date that have addressed this issue and most of the studies published include a variable number of patients with these clinical characteristics (Table 2). The CASS registry data comparing the survival of patients with predominant symptoms of heart failure (72 treated medically and 19 surgically), showed a poor 5-year survival of 23% for both groups 2. Despite the small size of the surgical group, it was concluded that in these patients surgery does not convey any long-term survival benefit. These observations and the finding that the presence of heart failure is a significant predictor of operative mortality 4, have largely dictated clinical practice in the 1980s and patients with isolated heart failure symptoms have been mostly considered for either medical therapy or transplantation. In recent years it has become evident that chronic left ventricular dysfunction in patients with coronary artery disease, is not always an irreversible process and left ventricular function may improve in a large subset of patients after successful coronary revascularisation 5,6 due to the presence of hibernating myocardium. Given that hibernating myocardium can be present in patients without angina 7 and that LVEF is a strong prognostic indicator in patients with coronary artery disease 8, it has been proposed that patients with post-ischaemic heart failure and hibernating myocardium may gain functional and prognostic benefit from coronary revascularisation. However, before revascularisation for heart failure is considered a valid treatment option, a number of issues need to be discussed. Confirmation of hibernating myocardium can only be made retrospectively when improvement in contractile function of dysfunctional myocardium is documented following coronary revascularisation. Although the prevalence of recoverable LV dysfunction is largely unknown, data extrapolated from a number of studies indicate that 22–57% of dysfunctional LV segments improve contractile function following CABG 9–11 even in patients without angina and with Q-wave myocardial infarction 7. The improvements in segmental wall motion are associated with clinically meaningful improvements in global LV ejection fraction in up to 50% of the patients enrolled 7,12,13. These data however, come from highly selected patients, and, therefore, may overestimate the real prevalence of hibernating myocardium. The presence of hibernating myocardium can be predicted prior to revascularisation if hallmarks of cellular viability (i.e. inotropic contractile reserve, sarcolemmal integrity and/or preserved metabolic function) can be demonstrated in dysfunctional myocardium subtended by a stenotic coronary artery. A comprehensive review of the three most commonly used techniques used to identify hibernating myocardium, namely dobutamine echocardiography, thallium-201 scintigraphy and positron emission tomography (PET) with fluorodeoxyglucose (FDG), showed a similar predictive accuracy 14. However, these data are extrapolated from studies of patients with moderately impaired LV function and angina, a group in whom decision for revascularisation had been made on conventional clinical criteria. Thus, the ability of these techniques to influence clinical management in patients with post-ischaemic heart failure remains untested. More recently, it has been shown that patients with severe LV dysfunction and advanced heart failure, dobutamine echocardiography may be associated with a significant higher false negative rate compared to PET or Thallium-201 scintigraphy 15,16. This discrepancy seems to be due to the severity of the ultrastructural changes detected in hibernating myocardium 17. A consistent relationship has been reported between the amount of viable myocardium detected preoperatively and the magnitude of LV functional improvement following revascularisation. Clinically meaningful improvements in LV ejection fraction after CABG are more likely to be detected in patients when at least 50% of the dysfunctional myocardium is viable, irrespective of the technique used to assess this 7,13. The surgical risks of performing CABG in these patients have decreased significantly in recent years, from 11 to 16% 18–20 to less than 6% 7,21,22 and this is probably due to improved anaesthetic, surgical and myocardial protection techniques. Nevertheless, the risk of revascularisation in patients with severe heart failure remains consistently higher than in patients with moderate impairment in LV function and the decision to revascularise must balance the risks of peri-operative mortality against all the benefits of revascularising hibernating myocardium. The lack of substantial amount of hibernating myocardium has been identified as an independent risk factor for peri-operative mortality 15,23 and this could help select patients at ‘lower’ peri-operative risk. The potential benefits of performing CABG in patients with substantial amount of hibernating myocardium should be divided into functional and prognostic. Recent observational studies have suggested that CABG in patients with a significant amount of viable myocardium is not only associated with convincing improvements in LV function 13,15,22, but also in a significant increase in exercise capacity 7,24 (Fig. 1) and improved quality of life 7. Although these data are extrapolated from observational data in a small number of highly selected patients, the results compare favourably with medical treatment, the effects of which on some of these end points is disappointing 25. A number of retrospective non-randomised studies have suggested that the presence of myocardial viability in patients with LV dysfunction and coronary artery disease may confer an adverse prognosis. Coronary revascularisation in this group may confer survival advantage compared to medical treatment 26,27. More recently it has also been demonstrated that the extent of preoperative myocardial viability is a determinant of long-term outcome following CABG 23 even in patients with severe heart failure symptoms 7. Five-year survival rates of 85% have been reported 28. These figures are significantly better than those usually reported for patients with class III or IV heart failure treated medically 25,29,30. Enough evidence exists to justify CABG in patients with severe LV dysfunction and predominant symptoms of angina in the opinion of many. The data on patients with predominant and severe heart failure symptoms is not conclusive. CABG could confer substantial benefits in patients with a significant amount of hibernating myocardium. Randomised studies are required.