By Christopher K Zarins
The primary objective of treating an arterial stenosis is to improve blood flow. But how do we determine whether a lesion needs treatment, and how do we know whether that treatment will improve blood flow and relieve the patient’s symptoms? The usual answer is to perform an angiogram to identify obstructing lesions and determine their severity by estimating per cent lumen stenosis. Threshold values of 50% or 70% stenosis are commonly used to select patients for interventional treatment. However, it is well known that visual estimates of the degree of lumen obstruction do not correlate well with the haemodynamic or functional significance of a lesion and that treatment of a stenosis that is not obstructing blood flow will not improve blood flow and relieve symptoms. The prospective, randomised COURAGE trial (NEJM, 2007) showed no survival benefit for patients who undergo angiographically based coronary revascularisation compared to medical therapy alone, leading many to question the benefit of coronary intervention and resulting in a significant reduction in the utilisation of coronary stenting.
Fractional flow reserve (FFR), measured at the time of invasive coronary angiography, has emerged as the reference standard for determining the haemodynamic (functional) significance of coronary artery stenosis. FFR is measured using a pressure-sensing guidewire (St Jude or Volcano) placed across the lesion and determining the ratio of mean coronary pressure distal to the stenosis to the mean aortic pressure during adenosine-induced maximal coronary blood flow. This ratio expresses the coronary blood flow that is still attainable despite the presence of a coronary stenosis. The prospective, randomised FAME trial (NEJM, 2009) showed that FFR-guided coronary intervention resulted in a significant reduction in coronary events and improved survival compared to angiographically-guided intervention. The use of FFR with a combined anatomic-physiologic assessment of coronary artery disease enhances clinical decision-making, improves both short- and long-term event-free survival and results in a significant reduction in healthcare costs. FFR-guided therapy is now considered the standard of care for decisions regarding coronary intervention. However, perhaps because of the invasive nature of FFR determination, it is used in only a minority of patients, with approximately 90% of coronary stents implanted on the basis of visual assessment of angiographic stenosis.
Non-invasive calculation of FFR from coronary CT scans (FFRCT, HeartFlow) is a novel technology that utilises computational fluid dynamics (CFD) to determine the functional significance of coronary artery disease.
Coronary blood flow and pressure are computed throughout the coronary tree under conditions of maximal coronary hyperemia, simulating the conditions of adenosine-induced hyperemia in the cath lab. This provides a three-dimensional map of FFR in all major epicardial vessels without the need for additional imaging, modification of CT acquisition protocols, or administration of medications. Two recently published prospective, multicentre studies (JACC, 2011, and JAMA, 2012) have demonstrated high diagnostic performance of FFRCT and high discriminatory power to identify patients and vessels with haemodynamically significant stenoses which will benefit from revascularisation and, importantly, to identify those lesions that are not functionally significant and are better treated medically.
This computational technology platform introduces a promising new paradigm to help physicians diagnose and plan treatments. The ability to non-invasively identify ischaemia-causing, flow-limiting lesions, will allow physicians to more accurately select patients for interventional treatment or medical treatment. For instance, the FAME studies have shown that medical treatment of visually apparent stenosis that are not functionally significant – ie FFR >0.80 – can be successfully treated with medical therapy with good long-term outcome. At the same time, those with haemodynamically significant lesions do better with intervention than with medical therapy. This technology also allows simulation of the interventional treatment, be it stenting or bypass. The expected blood flow changes can be computed and the haemodynamic benefit after treatment can be predicted.
This would allow pre-treatment comparison of different simulated interventional strategies and may helping guide decisions on stenting vs. bypass.
The strategy of CT image acquisition to define the anatomy and CFD analysis to simulate blood flow and pressure is applicable not only to coronary artery disease, but also to the peripheral circulation. CT scan image datasets of the thoracic and abdominal aorta, carotid arteries and cerebral circulation, renal arteries, mesenteric arteries and lower extremity vessels can be used to build region-specific computational flow models. Flow can be computed under resting and simulated stress/exercise conditions specific to each circulatory bed and potential treatment options can be tested. Development of such clinical applications is underway and can be anticipated in the future.
Christopher K Zarins is the Chidester Professor of Surgery Emeritus, Stanford University, Stanford, USA, and founder, senior vice president for Medical Affairs, HeartFlow.