Impaired coronary flow reserve: a pre-requisite for coronary revascularization

This editorial refers to ‘Effects of coronary revascularization on global coronary flow reserve in stable coronary artery disease’, by T. Aikawa et al., pp. 119–129. The goal of revascularization in stable coronary artery disease (SCAD) is to relieve the myocardium from flow-limiting coronary artery disease. The therapeutic benefit is based on the assumption that removal of the stenosis will improve myocardial perfusion. But the interaction between the artery and its microcirculation is complex,1 and regulated by many factors. These include stenosis severity, lesion morphology, length, eccentricity, surface features, plaque burden, collaterals,2 and the numerous factors influencing microvascular resistance on the tissue level.3 The relationship between coronary stenosis severity and myocardial blood flow is not linear unless resistance is minimized.4 Yet, coronary intervention may improve myocardial flow acutely.5 The prognostic benefit of revascularization in SCAD, however, is only obvious in cases with more severe and proximal disease burden.6 Yet, surprisingly few studies have so far investigated the longitudinal long-term benefit of epicardial coronary revascularization on microvascular perfusion reserve on the tissue level. Positron emission tomography (PET) using tracers of myocardial perfusion such as O-15 water or N-13 ammonia is considered to be the gold standard for absolute quantification of myocardial blood flow and flow reserve.1,3,7 Cardiac PET has been used for decades as a research tool, in order to identify the subtle effects of risk factors and their modification on the coronary microcirculation.7 The PET-derived coronary flow reserve is an integrated, absolutely quantitative marker of all factors influencing perfusion on the tissue level, including the macro- and microvascular components. It can be measured both for myocardial regions as well as for the global myocardium. As such, it differs from fractional flow reserve (FFR), which is measured on the macroscopic level of a single coronary artery,1 and from regional ischaemia on stress imaging, which identifies a relative tissue perfusion deficit in one myocardial region compared with the best perfused region of the same heart.8 Notably, after achieving a more widespread acceptance as a clinical diagnostic tool, PET has been shown to be a powerful tool for the functional assessment of suspected or overt coronary artery disease.1 Global coronary flow reserve has been shown to have incremental value on top of relative regional ischaemia for predicting outcome in large patient cohorts with different risk profiles.9–11 Furthermore, a recent registry analysis not only showed that global coronary flow reserve determined by PET predicts outcome independent of angiographic stenosis severity but also identifies a subgroup of low coronary flow reserve where early global revascularization (as opposed to targeted percutaneous intervention) resulted in reduced event rates.12 Long-term changes of coronary flow reserve after intervention, however, were not determined in this study. Aikawa et al.13 report about their effort to study this open question in a prospective observational study using serial O-15 water PET imaging.13 In 75 patients with SCAD, the clinical decision was taken to undergo either coronary artery bypass grafting (CABG), percutaneous coronary intervention (PCI), or optimal medical therapy (OMT) alone for treatment. Patients were recruited from four centres in Japan, but all travelled to one PET site for quantitative flow measurements before and 6 months after intervention decision. This emphasizes the complexity of the study setup and the considerable effort made by the researchers. The authors made some expected and some unexpected observations. On the one hand, global coronary flow reserve correlated inversely with the severity of angiographic disease burden as measured by SYNTAX and Leaman score. Also, global coronary flow reserve improved significantly after CABG. On the other hand, an improvement after PCI was only observed in subgroups with substantially impaired flow reserve below two at baseline or with high SYNTAX score above 23, but not in the entire PCI group. This was still true after propensity matching for differences in clinical characteristics. Regional analysis of flow reserve in the territory receiving PCI did not show a significant difference to OMT alone, independent of the presence of regional ischaemia or reduced FFR. The results of Aikawa et al. further support the complexity of interaction between epicardial coronary artery circulation and myocardial microcirculation. An improvement in coronary flow reserve following coronary revascularization may be greatest in patients with an impaired coronary flow reserve before revascularization. This is expectedly the case in patients undergoing CABG since typically coronary disease burden is greatest, and the same logic applies to performance of PCI. If coronary revascularization is performed in patients without flow-limiting disease (or impaired coronary flow reserve), beyond a placebo effect, benefits are unlikely. The design of the current study presents some limitations. It was not a randomized trial and treatment effects of one strategy or another on coronary flow reserve cannot be deduced from an observational cohort study. The acute effects of PCI on tissue flow have not been studied, and the small sample size precluded an in-depth analysis of regional differences. The precise mechanisms underlying the lack of observed differences on global microcirculatory flow following PCI remain unclear and require further investigation. Optimal medical therapy alone may improve global coronary flow reserve.14 Such an effect was not noted in this study, but the study was neither designed nor powered to elicit treatment effects and the patients were likely already established on medical therapy. Nevertheless, the study by Aikawa et al. provides a potential explanation for the recent ORBITA trial,15 where PCI combined with medical therapy in SCAD did not lead to a significant between-group difference for the improvement in exercise time on the Bruce protocol at 6 weeks post-randomization compared with treatment with medical therapy only. Approximately 30% of patients had objective evidence of non-flow limiting coronary disease, yet they were still enrolled into the strategy trial with PCI. On this basis, the lack of improvements in exercise duration, time to 1 mm ST-segment depression, peak oxygen uptake (mL/min), and anginal symptoms were to be expected. This issue confounds the interpretation of this trial’s results. In the 5-year follow-up analysis of the FAME-2 trial, the risk of death, myocardial infarction, or urgent revascularization (and notably spontaneous, non-fatal myocardial infarction) was reduced in patients with flow limiting coronary disease (FFR ≤0.80) who were initially treated by PCI, compared with patients treated with medical therapy only. The ISCHEMIA trial (ClinicalTrials.gov Identifier: NCT01471522) is assessing whether coronary revascularization improves symptoms and prognosis in patients with evidence of moderate-severe burden of myocardial ischaemia in patients with stable ischaemic heart disease. The work by Aikawa et al. adds new information on the potential benefits of coronary revascularization when restricted to patients with an impaired coronary flow reserve. The study is timely in relation to ORBITA, FAME-2, and ISCHEMIA. Cardiac PET is increasingly available owing to its utility in clinical oncology, alongside expanding indications in clinical cardiology e.g. infective endocarditis. The results of the ISCHEMIA trial will further inform the utility of revascularization decisions based on measurement of flow reserve and are eagerly awaited. Conflict ofinterest: F.M.B. receives grant support and speaker honoraria from GE Healthcare and Siemens AG. This support is not related to the topic of this manuscript. Based on institutional agreements with the University of Glasgow, Professor Berry has undertaken consultancy and research with Abbott Vascular AstraZeneca, GSK, HeartFlow, Menarini, Novartis, Opsens and Siemens Healthcare. Other author declared no conflict of interest.