In stable coronary stenosis of intermediate tightness on coronary angiography, a 1996 functional approach analysing endocoronary pressure displayed 93% accuracy in detecting myocardial ischaemia . Since 2010, fractional flow reserve (FFR) has been a level IA recommendation from the European Society of Cardiology in the absence of relevant non-invasive evidence of ischaemia, and this indication was extended in 2013 . Twenty years after publication of the seminal study, the French health authorities have given FFR coverage in all stable single-vessel or multivessel clinical situations, in diagnostic coronary angiography and in percutaneous coronary intervention (PCI) when non-invasive assessment is non-contributive or contraindicated. The clinical indications are evidence based, and are unrestrictively suited to the powerful discriminatory potential of FFR in the diagnostic process.

FFR is the ratio of coronary pressure downstream of the stenosis in question (Pd) to aortic pressure (Pa) under hyperaemia (Pd/Pa). A threshold of 0.80 shows optimal specificity (negative predictive value > 95%) in detecting coronary lesions liable to induce myocardial ischaemia . For example, a FFR of 0.70 means that the stenosis is inducing epicardial resistance sufficient to reduce myocardial flow by 30% under effort.

The fundamental prerequisite is to obtain maximal hyperaemia, to avoid the effect of coronary self-regulation, and to be able to perform measurement under minimal myocardial resistance. Incomplete hyperaemia was recently shown to reduce FFR accuracy from 95% to 85% (while angiography alone shows 65% accuracy) . If hyperaemia is incomplete, FFR will be overestimated and lesion severity will be underestimated, reducing diagnostic sensitivity.

The gold-standard technique for obtaining hyperaemia is intravenous adenosine injection, as used in the pivotal clinical studies . However, for reasons of practicality, intracoronary injection is generally preferred, and has been shown to be just as effective. Other methods have been discussed but have yet to be tested clinically. The injection of contrast medium, for example, achieves only 65% hyperaemia. Resting gradient measurement without hyperaemia (instantaneous wave-free ratio [iFR]) has also been described, but threshold values are disputed and clinical studies (DEFINE-FLAIR, NCT02053038 ; iFR-SWEDEHEART, NCT02166736 ) are still underway.

There is no simple criterion to assess maximum hyperaemia, and the cardiologist is the key player in ensuring metric quality (good adenosine dose, effectively reaching the myocardium) and the reliability of the results.

As FFR measurement is liable to have a profound effect on revascularization strategy, the FFR report should be kept separate from the mass of coronary angiographic data, and should specify procedure, hyperaemia and the artery or arteries studied.

Initial studies of FFR in patients with single-vessel disease reported long-term stability for angiographic lesions with negative FFR . These data were probably too good to be generally true, and need to be viewed in the light of coronary progression over time. Following FFR measured at >0.80 in a non-selected population, the risk of major adverse cardiac events was 5% at 1 year and 18% at 4 years . FFR values should be read in the light of their clinical relevance, taking account of risk factors for progression: notably age, smoking, history of acute coronary syndrome and multivessel coronary involvement. FFR makes a major clinical contribution to coronary assessment, but is not in itself an absolute predictive factor in the long term.

FFR ≤ 0.80 indicates depth of ischaemia in the artery being explored. The lower the FFR, the poorer the prognosis , but myocardial ischaemic mass is still a major prognostic factor ; it is not taken directly into account in measuring FFR, and an FFR value that would certainly indicate revascularization in the proximal anterior interventricular artery is not going to be so decisive for the marginal artery . Whereas FFR shows good accuracy and spatial resolution, endocoronary pressure is simply contributive to clinical decision-making and is not by itself the gatekeaper of revascularization.

Clinical studies of FFR have rigorously and meticulously explored the various clinical situations of stable coronary disease managed by angioplasty. Prognosis is improved in single-vessel and multivessel cases, and costs are saved. Extrapolating from the findings of the FAME study in three-vessel disease, the expected cost-savings in various countries of the European Union are considerable, depending on each country’s PCI turnover, and are probably at least several millions of euros for France.

The French multicentre R3F registry followed more than 1000 patients after routine FFR, and showed that results modified treatment in more than 4/10 cases . In almost 500 patients with three-vessel involvement on angiography (stenosis ≥ 50%; mean 60 ± 18%), FFR indicated that 43% of those eligible for surgical revascularization due to a SYNTAX score > 22 could have benefited from a different treatment option . In a prospective series of 200 stable low-risk coronary patients (>80% with ≤1 stenosis on angiography) FFR modified vessel targeting for revascularization in 26% of cases, with only 60% concordance with angiography. Lastly, FFR increased indications for medical management by 24% .

In stable coronary conditions, it is thus well established and agreed that FFR avoids stenting in about 30% of angiographic > 50% stenoses. The European guidelines are founded on these data; extending indications for FFR to multivessel acute coronary syndrome is intuitively attractive, and has been suggested, but at present is not formally accepted .

In non-ST-segment elevation myocardial infarction (NSTEMI), after ruling out periprocedural infarction, Layland et al. reported a trend ( P = 0.19) for an increase in major adverse cardiac events in the FFR study arm, mainly attributed to rapid progression of the culprit lesion despite FFR > 0.80 in the acute phase . In contrast, subgroup analysis of 328 patients with NSTEMI or unstable angina in the FAME study found a favourable trend for the FFR group, but did not exclude periprocedural infarction, which may have masked spontaneous progression in FFR-negative culprit lesions . In the present state of knowledge, it seems reasonable for the culprit lesion in an acute coronary syndrome not to be selected for FFR, given its specific risk of atherothrombotic progression.

Recently, FFR assessment of non-culprit lesions in multivessel acute-phase ST-segment elevation myocardial infarction (STEMI) (with the culprit lesion systematically undergoing PCI) was shown to be feasible, and to reduce rates of subsequent emergency and non-emergency revascularization, compared with culprit-lesion PCI alone . The DANAMI3-PRIMULTI study thus provided no final answer to the question of the role of FFR in this clinical situation, where non-culprit lesions are also routinely revascularized.

The temptation remains to use FFR to stratify treatment and to choose between PCI and bypass in multivessel cases, but without any scientific evidence to support it. If FFR is to cease to be only a tool in selection for angioplasty, and to become a means of constructing revascularization and treatment strategies, this remains to be validated, and is presently the focus of ongoing studies (FUTURE, NCT01881555 ; SAVE-IT NCT02173860 ).

Thus, FFR enables the functional status of an ambiguous coronary lesion to be assessed directly in the catheterization laboratory, with a high level of evidence. As it is now covered by the health budgeting system, the spread of FFR in France offers an opportunity for the medical community in general, and for the cardiology community in particular, to manage patients in a rational and adapted manner; it is also our individual and collective responsibility to make the best use of this powerful – but relatively expensive – tool, requiring rigorous use and global clinical judgement.