Coronary flow velocity reserve adds prognostic information beyond presence or absence of non-obstructive coronary atherosclerosis Free

This commentary refers to the article ‘Coronary flow velocity reserve predicts adverse prognosis in women with angina and no obstructive coronary artery disease: results from the iPOWER study’, by J. Schroder et al., doi: 10.1093/eurheartj/ehaa944 and the discussion piece ‘Is coronary flow velocity reserve just a marker of a higher atherosclerotic burden?’, by F. Radico et al., doi:10.1093/eurheartj/ehab062.

We thank Radico et al.¹ for their interest in our study and welcome this opportunity to discuss the interpretation of its results.²

1. As pointed out, event rates in our study were similar to those reported in the meta-analysis performed by the authors.³ In patients with angina and no obstructive coronary artery disease (ANOCA) event rates are highly dependent on presence or absence of non-obstructive coronary atherosclerosis on invasive coronary angiogram (ICA). A previous study from our group (included in the meta-analysis by Radico et al.) found the lowest event rates in asymptomatic controls, an increased risk in angina patients with normal coronary arteries, and a further increase in risk in angina patients with non-obstructive atherosclerosis.⁴ As a result, the iPOWER study aimed to assess the prognostic significance of coronary microvascular disease presence in ANOCA patients both with and without non-obstructive coronary artery disease.²

2. Radico et al. speculate if low coronary flow velocity reserve (CFVR) may simply be a marker of non-obstructive atherosclerosis presence and enquire whether our analysis was adjusted for non-obstructive atherosclerosis at ICA. In Table 3, we present the association between a low CFVR and the composite outcome: After adjusting for presence of non-obstructive atherosclerosis at ICA in addition to age and hypertension, the hazard ratio was reduced from 1.07 to 1.05 per 0.1 unit decrease in CFVR [confidence interval (CI) 1.01–1.09, P = 0.01], indicating residual confounding by non-obstructive atherosclerosis is not a likely explanation for the association between CFVR and the outcome. Furthermore, in exploratory interaction analysis, we found lower CFVR was associated with the composite outcome both in patients with and without non-obstructive atherosclerosis (interaction P = 0.91), consistent with the above finding. From our study as well as others we find it likely that CFVR impairment is a marker of adverse prognosis in both patients with and without non-obstructive atherosclerosis.

Part of the association between low CFVR and the composite outcome may likely be accounted for by a propensity for developing future epicardial obstructive atherosclerosis, yet our study and others demonstrate this is not the full explanation. We found only 5 out of 35 observed myocardial infarctions were followed by short-term revascularization, indicating myocardial infarction with non-obstructive coronary arteries. CFVR is also impaired in patients with heart failure and perhaps causally associated with heart failure with preserved ejection fraction.⁵

It is of key importance to introduce diagnostic tests with the ability to discern which angina patients are most likely to benefit from both invasive and non-invasive treatment options. A recent study examining the utility of positron emission tomography myocardial perfusion indicated assessment prior to revascularization may help identify patients most likely to gain a prognostic benefit.⁶ Assessment of coronary microvascular function, integrated with anatomical assessment by computed tomography angiography or ICA, could potentially play a similar role in ANOCA patients. Further studies in coronary microvascular function will help us in understanding ischaemia as a driver of events and will hopefully increase our understanding of coronary microvascular dysfunction in other types of heart disease, including heart failure.

Conflict of interest: none declared.

References