Author + information
- Francis J. Klocke, MD∗ ( and )
- Daniel C. Lee, MD
- Feinberg Cardiovascular Research Institute, Division of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine and the Bluhm Cardiovascular Institute, Chicago, Illinois
- ↵∗Reprint requests and correspondence:
Dr. Francis J. Klocke, 950 North Michigan Avenue, Apartment 5102, Chicago, Illinois 60611-7532.
- cardiac magnetic resonance
- endocardial-epicardial flow ratios
- subendocardial ischemia
- transmural myocardial perfusion
The cardiac magnetic resonance (CMR) studies of Mordini et al. (1) in this issue and Chiribiri et al. (2) in the May 2013 issue of iJACC probe 2 approaches for evaluating transmural perfusion in humans. The importance of subendocardial reductions in myocardial blood flow in the genesis of myocardial ischemia is unquestioned. Transmural perfusion is most frequently characterized in terms of the ratio of blood flow in the inner and outer layers of the myocardial wall during the entire cardiac cycle. Commonly referred to as the endocardial-to-epicardial flow ratio, this parameter is a potentially useful indicator of a flow limitation capable of producing subendocardial ischemia in patients with known or suspected coronary disease.
Complexities of Endocardial-to-Epicardial Flow Ratios
Experimental studies during the last several decades have identified relevant complexities of transmural perfusion and effects of systolic contraction on coronary blood flow (3). Under normal resting conditions, the “endo-epi” flow ratio for the full cardiac cycle slightly exceeds 1.0, probably because of greater oxygen demand in the inner myocardial layers. As has long been recognized, the restricting effects of cardiac contraction on systolic flow require that the flow needed to maintain the normal O2 demand-supply balance occurs primarily during diastole. Because the magnitude of systolic compressive forces varies transmurally—being greatest in the innermost myocardium and decreasing progressively across the ventricular wall—essentially all endocardial flow must occur in diastole. Additionally, the magnitude of full-cycle coronary flow reserve is usually less in the inner layers of the myocardium and varies with changes in diastolic driving pressure and/or the duration of diastole.
Changes in intravascular blood volume over the course of each cardiac cycle also influence transmural perfusion. Because of contraction's compressive effects, myocardial outflow occurs predominantly during systole, momentarily reducing intramyocardial blood volume and intramural vascular dimensions. Although nearly all outflow exits via the coronary sinus and other venous channels, some blood moves retrograde into the epicardial arteries—and perhaps the subepicardial myocardium as well. Because of the normal transmural variation in compressive forces, systolic reductions and subsequent diastolic increases in intravascular volume are greatest in the subendocardium. In addition, diastolic reopening of the intramyocardial vessels compressed during systole probably occurs more slowly in the subendocardium, and the back pressure opposing diastolic inflow is higher.
Not surprisingly, values of the endo-epi flow ratio vary with hemodynamic and other factors. Under normal conditions, the endo-epi ratio can be estimated by the ratio of diastolic to systolic pressure-time indices: that is, the ratio of the area between aortic and left ventricular pressures during diastole and the area beneath the left ventricular pressure curve during systole (3). Endo-epi ratios measured during pharmacologic vasodilation are reduced globally by factors such as reductions in systemic pressure, tachycardia, increased pre-load, and ventricular hypertrophy. As early as 1974, adenosine-induced increases in heart rate and reductions in systemic arterial pressure were found to reduce the endo-epi flow ratio from 1.06 to 1.16 down to 0.81 to 0.99 in awake dogs (4). An additional consideration is that the full amount of flow reserve measured using pharmacologic vasodilation is not necessarily available to ischemic myocardium (5). This reflects a limited sensitivity of larger resistance vessels to metabolic factors, and perhaps vasoconstrictor responses to alpha-adrenergic stimuli, platelet products, and neurohormonal factors as well.
Current CMR Approaches
Widely used CMR approaches for evaluating myocardial blood flow typically focus on regional differences in myocardial signal intensity during the first passage of an intravenously administered gadolinium contrast agent. Signal intensity increases more slowly and peaks at a lower value in regions in which flow is reduced on a relative basis. Visual analysis of full-thickness images obtained under resting conditions and during pharmacologic coronary vasodilation remains the most common clinical approach. Semiquantitative indices based on parameters such as upslope, peak signal, and area under the signal intensity-time curve have been shown to accurately identify regional limitations in full-thickness flow reserve. Experimentally, the high spatial resolution of CMR has enabled further assessment of vasodilated flow in multiple layers from endocardium to epicardium (6). Using analytic models, myocardial flow can be quantified in absolute terms (7), although further study is ongoing to identify the optimal perfusion pulse sequence, contrast dosing scheme, and modeling algorithm.
In the current study by Mordini et al. (1), a quantitative “pixel-wise” technique was used to analyze transmural perfusion patterns during dipyridamole-induced vasodilation in 67 patients. Single-, double-, and triple-vessel diseases (≥70% stenosis on quantitative coronary arteriography) were currently present in 16, 5, and 2 patients respectively. Seventeen patients had a history of myocardial infarction. Using receiver-operating characteristic analysis, the investigators concluded that obstructive disease was best identified by a <50% ratio of regional endocardial to median epicardial quantitative vasodilated flow. Ratios averaged 0.42 ± 0.05 in patients with arteriographically defined disease versus 1.13 ± 0.19 in others (p < 0.001). Absolute values of flow were also reduced systematically in arteriographically positive areas (1.20 ± 0.53 vs. 3.13 ± 0.61 ml/min/g in the endocardium and 1.73 ± 0.71 vs. 2.99 ± 0.59 ml/min/g in the full-thickness myocardium, both p < 0.001). Coronary flow reserve was not analyzed.
The recent study of Chiribiri et al. (2) followed their development of a semiautomated analysis of transmural perfusion gradients for distinguishing between “ischemic” and normally perfused myocardium during adenosine stress perfusion. This approach uses the instantaneous differences in myocardial signal intensity between the inner and outer thirds of the transmural myocardial wall following bolus contrast injection to construct a 2-dimensional diagram portraying the circumferential extent and temporal persistence of regional reductions in transmural perfusion. The diagram can be segmented at different degrees of inner-outer contrast difference and “thresholded” at a specific degree of reduction. The recent study included 67 patients. Single-, double-, and triple-vessel diseases (≥50% diameter stenosis in 2 orthogonal views) was present in 29, 16, and 6 patients, respectively. Because of its established clinical utility (8), fractional flow reserve was used as the reference standard for identifying hemodynamically significant stenoses. Using a fractional flow reserve cutoff of <0.8, a 20% transmural perfusion gradient provided the best diagnostic threshold in an initial group of 30 patients. When applied to 37 additional patients, this threshold achieved a per-patient sensitivity of 89%, specificity of 83%, and area under a receiver-operating characteristic curve of 0.86.
Taken together, the initial studies of Mordini et al. (1) and Chiribiri et al. (2) suggest that endo-epi flow assessments can add usefully in the noninvasive assessment of coronary disease. Studies including larger patient numbers and a greater prevalence of multivessel disease are especially desirable. In view of the various factors able to influence endo-epi ratios, it seems doubtful that a single cutoff value identifying a propensity for subendocardial ischemia will emerge. Diseases such as left ventricular hypertrophy, aortic stenosis, hypertrophic cardiomyopathy, and syndrome X will no doubt reduce vasodilated endo-epi ratios globally. Independently of any global reduction in the ratio, however, the identification of a relative regional reduction that exceeds measurement variability could indicate coexistent coronary artery disease. Concomitant late gadolinium enhancement imaging can then distinguish between viable myocardium with limited perfusion reserve and infarcted myocardium.
As transmural measurements are made more frequently, an additional consideration is that a modest regional reduction in endo-epi ratio during pharmacologic coronary vasodilation may not always correlate with ischemia in activities of daily life. For example, patients whose activities require only modest increases in myocardial oxygen supply may remain asymptomatic despite a modest reduction in maximum vasodilated flow. Conversely, because epicardial stenoses are subject to dynamic changes in caliber (9)—for example, active constriction during exercise (10)—reductions in the endo-epi ratio may sometimes be greater during vigorous activity than pharmacologic vasodilation.
Continuing improvements in CMR—and probably computed tomography perfusion imaging as well—should allow subendocardial perfusion and transmural flow regulation to be examined in humans more extensively than has heretofore been possible. Further experience should also clarify how frequently quantitative measurements of regionally restricted endo-epi ratios can add to currently available diagnostic imaging capabilities.
↵∗ Editorials published in JACC: Cardiovascular Imaging reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Imaging or the American College of Cardiology.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Mordini F.E.,
- Haddad T.,
- Hsu L-Y.,
- et al.
- Chiribiri A.,
- Hautvast G.L.,
- Lockie T.,
- et al.
- Klocke F.J.
- Lee D.C.,
- Simonetti O.P.,
- Harris K.R.,
- et al.
- Lee D.C.,
- Johnson N.P.
- Klocke F.J.
- Brown B.G.,
- Lee A.B.,
- Bolson E.L.,
- Dodge H.T.