Author + information
- Francis J. Klocke, MD∗ ()
- Feinberg Cardiovascular Research Institute and Division of Cardiology, Department of Medicine, Bluhm Cardiovascular Institute, Feinberg School of Medicine, Northwestern University, Chicago, Illinois
- ↵∗Reprint requests and correspondence:
Dr. Francis J. Klocke, Northwestern University Feinberg Medical School, Feinberg Cardiovascular Research Institute, 950 North Michigan Avenue, Apartment 5102, Chicago, Illinois 60611-7532.
The RESOLVE study by Jeremias et al. (1) in this issue of the Journal represents a commendable joint undertaking by investigative groups with differing views about the value of resting pressure gradients and the need for pharmacological vasodilation in assessing the functional severity of arteriographic stenoses. Two resting gradient measurements have been evaluated against fractional flow reserve (FFR), the established invasive standard for identifying lesions that will benefit from early revascularization and those in which revascularization may safely be deferred.
FFR: the stenosis pressure gradient during pharmacological coronary vasodilation
The clinical value of FFR, which was developed by Pijls, De Bruyne, Gould, and colleagues approximately 20 years ago, has been confirmed in 3 prospective randomized trials, which now include outcome data (2). The ratio of post-stenotic to pre-stenotic pressure during the full cardiac cycle is taken to reflect regional coronary flow when microvascular resistance has been minimized by pharmacological vasodilation, usually using adenosine. Pijls currently reports that FFRs ≥0.80 exclude inducible ischemia with an accuracy of 95%, whereas values <0.75 indicate its presence in almost 100% of lesions (3). To minimize misclassification of a lesion with impending adverse impact, the recommended clinical cut point is 0.80. The widespread use of FFR continues to be limited by its requirement for pharmacological vasodilation and additional factors covered in recent editorials (4,5).
Instantaneous wave-free ratio and distal coronary artery pressure/aortic pressure: stenosis pressure gradients under resting conditions
As with FFR, resting gradients are reported as the ratio of post-stenotic to pre-stenotic mean pressure (i.e., a value of 0.90 reflects a 10% reduction in incoming arterial pressure). The Imperial College London group led by Davies measures the gradient during the last three-fourths of diastole (excluding the final 5 ms), when post-stenotic impedance to flow is at a low point. Based on wave intensity analysis, this measurement is termed the “instantaneous wave-free ratio” (iFR) (6). Mamas et al. (7) and Johnson et al. (8) have analyzed the ratio of distal coronary artery pressure to aortic pressure (Pd/Pa) during the full cardiac cycle. Because iFR and Pd/Pa evaluate different portions of the cardiac cycle and are measured under resting rather than vasodilated conditions, their absolute values are expected to differ from each other and from FFR.
FFR versus iFR and Pd/Pa
RESOLVE (1) was a core laboratory–based, retrospective, multicenter study including 1,768 patients, two-thirds of whom presented with chronic stable angina. A total of 1,593 lesions proved satisfactory for evaluation. Linear correlations of iFR and Pd/Pa with FFR for the entire data set (Figs. 1A and 1B ) were “moderate” (R2 of 0.66 and 0.69, respectively). Using receiver-operating characteristic–determined cut points of 0.90 and 0.92 for predicting an FFR ≤0.80, overall diagnostic accuracies were 80.4% and 81.5%, respectively.
After concluding that iFR and Pd/Pa are imperfect surrogates of FFR close to the 0.80 cut point, the authors examined whether resting gradients might provide acceptable accuracy at greater or lesser degrees of stenosis severity. To achieve 90% accuracy for predicting an FFR ≤0.80, iFR had to be ≤0.88 and Pd/Pa ≤0.92. For 90% accuracy in predicting an FFR >0.80, iFR had to be ≥0.97 while no upper boundary of Pd/Pa achieved this accuracy. These findings confirm that iFR and Pd/Pa can identify many stenoses that qualify for revascularization by FFR criteria. They also suggest that resting gradients have limited value for identifying lesions in which revascularization can be deferred.
Figure 5 in the paper by Jeremias et al. (1) shows the expected proportions of cases in which pharmacological vasodilation can be avoided for levels of agreement of iFR and Pd/Pa with FFR between 80% and 100%. The 2 relationships “intertwine.” For ≥90% overall concordance with FFR, iFR and Pd/Pa would avoid vasodilation in 65% and 48% of lesions, respectively. For ≥95% overall accuracy, these values fall to 29% and 36%. Importantly, these percentages would all be reduced if only data spanning a narrower range of FFR (e.g., 0.60 to 0.90) were considered.
The RESOLVE study also calls attention to the need for “quality control” in measurements of relatively small gradients. In total, 19.3% of the 1,974 lesions submitted for study had to be excluded because of technical errors/uncertainties. Additional issues include the nonperfect reproducibility of any measurement and potential effects of intracoronary nitroglycerin on the assessment of a stenosis subject to clinically important vasoconstriction.
Fluid dynamic factors influencing pressure gradients
During the 1970s, Young et al. (9), Gould (10), and others showed that the energy loss across a stenosis is influenced by both viscous (frictional) forces and post-stenotic flow separation. Their relative magnitudes vary in individual lesions. Viscous losses depend primarily on stenosis diameter and length and are directly proportional to flow. Separation losses, on the other hand, are proportional to flow squared and are also influenced by post-stenotic geometry (exit angle). A key point is that trans-stenotic pressure gradients increase nonlinearly with flow when separation losses are operative. Thus, stenoses with similar resting gradients but differing propensities for flow separation can generate significantly different gradients during increases in flow caused by pharmacological vasodilation, exercise, or other interventions.
The question of how frequently this latter issue limits the clinical value of a resting gradient measurement merits further study. Experimental studies indicate that flow separation occurs in stenoses of relatively mild severity (11). However, separation effects in arteriographically intermediate lesions may well be less than those in more severe stenoses in which a gradient measurement is deemed unnecessary.
Resting Pd/Pa versus iFR
Rather surprisingly, the 2 measurements of resting gradients in the RESOLVE study correlated closely (R2 = 0.95) and showed generally similar relationships to FFR. A 95% R2 value has also been reported by Johnson et al. (8). iFR is intended to avoid the confounding effects of systolic contraction and thereby focus on downstream microvascular resistance. As noted in the report by Jeremias et al. (1), downstream “resistance” is influenced by capacitive and inertial as well as resistive factors and by the complex effects of contraction on myocardial blood volume and intramyocardial vascular dimensions. Models intended to deal with these issues have been pioneered by several authors of the present study and will no doubt continue to be refined.
Combined flow and pressure gradient measurements
Combined wire-tip pressure and velocity measurements now allow separate analysis of stenosis gradients and the complex pressure-flow behavior of the distal coronary bed. As mentioned in the preceding text, the latter is usually modeled as “microvascular resistance.” Several years ago, Meuwissen et al. in Amsterdam showed that microvascular resistance modulates the relationship between FFR and coronary flow velocity reserve and proposed a hyperemic stenosis resistance index for improving lesion assessment (12). Advocates of velocity as well as gradient measurements point out that an abnormal FFR can result from a large increase in vasodilated flow through a mild stenosis as well as a limited flow increase through a more severe narrowing. The former situation may be relevant to a subgroup of patients with small resting gradients who develop abnormal FFRs despite normal coronary flow velocity reserves and hyperemic stenosis resistance. Combined pressure and velocity measurements are attractive for studies of human coronary physiology as well as individual coronary stenoses. However, the additional time, technical complexities, and operator expertise required for velocity measurements continue to limit their broader implementation.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Klocke has reported that he has no relationships relevant to the content of this paper to disclose.
- American College of Cardiology Foundation
- Jeremias A.,
- Maehara A.,
- Généreux P.,
- et al.
- Kern M.J.
- Samady H.,
- Gogas B.D.
- Sen S.,
- Escaned J.,
- Malik I.S.,
- et al.
- Johnson N.P.,
- Kirkeeide R.L.,
- Asrress K.N.,
- et al.
- Young D.F.,
- Cholvin N.R.,
- Roth A.C.
- Gould K.L.
- Mates R.E.,
- Gupta R.L.,
- Bell A.C.,
- Klocke F.J.
- Meuwissen M.,
- Siebes M.,
- Chamuleau S.A.,
- et al.