Microvascular (Dys)Function and Clinical Outcome in Stable Coronary Disease

• coronary atherosclerosis
• coronary flow
• fractional flow reserve
• microvascular function
• microvascular resistance

The term “coronary artery disease” is often restricted to the presence of ≥1 stenosis in the epicardial arteries exceeding a diameter of 50% of the adjacent reference vessel on angiographic images (1) but does not acknowledge that flow beyond these lesions is part of a black box called the microcirculation. One of the reasons for this scenario is that epicardial arteries are easily accessible for diagnosis and treatment while the microcirculation is not. Consequently, our knowledge about so-called significant coronary artery disease is largely based on this definition of 50% diameter stenosis that has been used to risk-stratify patients, validate noninvasive techniques, generate prediction models, justify revascularization, and serve as an entry criterion or endpoint in multiple clinical trials. A recent position paper by Pries et al. (2) for the European Society of Cardiology’s Working Group on Coronary Pathophysiology and Microcirculation provides a timely reminder of the complexity of the coronary circulation, its remarkable plasticity, and the myriad of factors controlling myocardial blood flow. Although the clinician must be aware that the coronary microcirculation is not a single homogeneous compartment, most of its features are presently beyond the domain of imaging and catheter investigation. Therefore, a simplified framework of the main features and metrics of the coronary circulation (Figure 1) might be helpful for clinical decision making.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Macrocirculation and Microcirculation Across Segments and Sizes of the Arteries

Fractional flow reserve (FFR) versus the index of microvascular resistance (IMR) versus coronary flow reserve (CFR).

In this issue of the Journal, Lee et al. (3) report the clinical outcome of 313 patients with at least 1 angiographically intermediate epicardial stenosis and in whom fractional flow reserve (FFR), the index of microcirculatory resistance (IMR), and coronary flow reserve (CFR) had been measured. This study is important because the authors focus their analysis on patients with hemodynamically nonsignificant stenoses (FFR >0.80, n = 230) that were not immediately revascularized. This report is the first of the natural history of (mainly) stable patients with documented coronary atherosclerosis and specific characterization of the macrocirculation by FFR and the microcirculation by IMR. Patients with an FFR value >0.80 were subdivided into 4 groups according to the values of CFR and IMR. The main finding of the study is that clinical outcome is worst in those with both an elevated IMR and a low CFR, clearly targeting microcirculatory dysfunction as a possible reason (or marker) for this poor outcome.

As the authors acknowledge (3), the study has a number of limitations, some of which are worth mentioning. First, neither the patients nor the operators were blinded to the results of CFR and IMR. Second, the division of a study population of 230 patients into 4 unequal subgroups produced small numbers of patients per group, thus increasing the probability that the observed differences might be a result of chance. Third, there was no systematic attempt to document noninvasive evidence of ischemia. Finally, during follow-up, the differences in patient-oriented combined outcome were mainly driven by subsequent revascularization in the patients with low CFR and high IMR, and this outcome was reported to be due to “progression” of the epicardial stenosis. Stenting an epicardial stenosis is of course unlikely to influence the underlying microvascular problems, and the need to perform percutaneous coronary intervention in these patients suggests that the microvascular dysfunction may have been a marker rather than the cause of the poorer outcome. Nevertheless, the differences in patient-oriented combined outcomes suggest that patients with hemodynamically nonsignificant epicardial stenoses can be further triaged into those with a greater or lesser risk of adverse events based on an assessment of their microcirculation.

Previous research by Fearon et al. (4) highlighted the prognostic value of IMR as a specific index of microvascular dysfunction in patients experiencing an acute myocardial infarction. An IMR value >40 immediately after primary percutaneous coronary intervention was associated with a 4-fold increased risk of death. In patients with stable coronary artery disease, the data are scarce. Van de Hoef et al. (5) showed that in vessels with angiographically nonsignificant disease, a CFR value <2.7 was associated with a >3-fold higher event rate than when CFR was above this threshold. However, this difference in CFR was due to a higher resting flow velocity value rather than a decrease in minimal microvascular resistance, a pattern suggestive of a problem of autoregulation. In several studies of positron emission tomography scanning, reduced CFR was associated with adverse events (6,7). However, in most of these studies, the coronary anatomy was not known or incompletely characterized; thus, an abnormal CFR may have reflected disease of the macrocirculation, the microcirculation, or both. The data presented by Lee et al. (3) are novel because, in addition to a full anatomic characterization, the stenosed arteries also underwent a thorough physiological assessment (FFR and IMR), thus allowing the investigators to distinguish and quantify the dysfunction related to each of these respective compartments.

Lee et al. (3), as with other researchers (8), report “discordance” between FFR and CFR in approximately one-third of cases. This outcome is not surprising. The same holds for discordance between CFR and IMR. As illustrated in Figure 1, CFR takes into account both compartments of the circulation, FFR allows decision making specifically about the macrocirculation, and IMR is specific for the microcirculation. Second, although the FFR cutoff value used to guide revascularization decisions has been defined and confirmed based on clinical outcome data, normal and ischemic thresholds for CFR have not been similarly defined, and reported cutoff values vary widely from 1.44 (9) to >3.0 (10), depending on both the method used for assessing CFR and the comparator. There is therefore little obvious reason to use a cutoff value of 2.0 to define discordance, and a cutoff value of 3.0 would have drastically changed the comparison. It is interesting to note that when both FFR and CFR are known, the minimal true normal value of CFR can be calculated by dividing the CFR by the FFR; for example, in a vessel with a CFR of 2.2 and an FFR of 0.80, the minimal “true” normal CFR value is 2.75. Third, in patients with discordant CFR and FFR, high resting indices of flow were the main drivers of low CFR values. Lee et al. report a resting mean transit time of approximately 0.30 s, which is unusually short and suggests some degree of hyperemia at rest. The same phenomenon has been reported with Doppler flow velocity measurements (8), in which patients with discordant CFR and FFR had resting flow velocity values markedly higher than patients with concordant CFR and FFR; the hyperemic flow velocity values were similar. This finding may reflect abnormal regulation of resting flow, but it also highlights the difficulty of establishing true resting conditions in the catheterization laboratory and is in marked contrast to the available data from the positron emission tomography scan in which resting flow values are remarkably constant (approximately 1.2 ml/min/g) across different tertiles of CFR. Taken together, these limitations suggest that CFR is not the index of choice for the evaluation of the microvascular compartment.

For the time being, let us use the specific tools that are available. If we need to decide about revascularization, use FFR. If we are concerned about the contribution of the microvasculature to myocardial ischemia, use IMR. The main question that Lee et al. (3) leave unanswered is whether IMR alone has an incremental prognostic value in stable patients with FFR >0.80 stenoses.

• 2016 American College of Cardiology Foundation