The elastic properties of arterial walls have been identified as important factors in the regulation of the circulation (1,2). Accordingly, epicardial coronary arterial wall elasticity plays a significant role in modulating coronary blood flow. However, suitable clinical methods have not previously existed to assess the elasticity of deep arteries such as the coronaries. Neither has it been possible to study the specific vasoreactivity of normal and diseased segments of the coronary arterial wall.
Intravascular ultrasound imaging systems can visualize the cross-sectional anatomy of small nonsuperficial arteries (3-7). The ultrahigh-frequency sound waves used in these devices provide greater image resolution than is possible with transthoracic or transesophageal echocardiography. In conjunction with direct determination of pressure, the high-resolution measurements of intraluminal size afforded by intravascular ultrasound are well suited to the assessment of the compliance of the coronary arteries.
At present, few data exist regarding the assessment of the elastic properties of coronary arteries by intravascular ultrasound. Neither are data available concerning the effects of pharmacologic agents on vascular elasticity. Therefore the aims of this study were (a) to apply intravascular ultrasound to the assessment of the elastic properties of regional coronary artery by measuring the change in cross-sectional area and segment perimeter produced by the fluctuation in pressure during the cardiac cycle, and (b) to use this method to evaluate the regional arterial wall function by examining the response to nifedipine.
The study population consisted of 20 patients who were undergoing cardiac catheterization to evaluate chest pain. Based on the results of coronary angiography, there were eight normal patients (Norm) and 12 patients with coronary disease (CAD). The Norm group consisted of four men and four women ranging in age from 56 to 83 years (mean, 67 years), whereas the CAD group comprised eight men and four women ranging in age from 42 to 83 years (mean, 63 years). All patients were free of other evidence of heart disease or congestive heart failure.
No attempt was made to control the medical therapy administrated to these patients, with the exception of omitting calcium blockers. At the time of catheterization, all patients were receiving aspirin, eight were receiving nitrates, six were taking β-blockers, and three were taking angiotensin-converting enzyme (ACE) inhibitors. The same dosages of these drugs were maintained for >6 months before protocol. No patients were receiving other vasodilators.
The protocol for this study was approved by the Institutional Review Committee for Human Subjects, and informed consent was obtained from all subjects enrolled in the protocol.
After diagnostic coronary angiography, an intravascular ultrasound (IVUS) image was recorded at just distal to the right or left coronary arterial ostium by using the IVUS system (Hewlett-Packard Corp.) and a 3.5F ultrasound catheter probe with 30-MHz center frequency (Boston Scientific Corp.). The vessel to be examined by IVUS in any given patient was primarily determined by the presence of lesions that would undergo subsequent angioplasty and therefore require the placement of a guiding catheter. In normal subjects or those who were not candidates for angioplasty, the left anterior descending coronary artery was usually examined. Thus studies were performed in 14 left and six right coronary arteries. The catheter probe was manipulated to obtain a clear image of a proximal portion of the vessel that exhibited a thin intimal-medial layer and dense adventitial layer. The electrocardiogram (ECG) was recorded, and coronary ostial pressure was measured through the side-port of a Y connector on the 8F guiding catheter. The ultrasound catheter probe was then advanced into the vessel until an area containing atherosclerotic plaque (vide infra for criteria) was encountered, from which high-quality images of the entire vessel circumference could be obtained. In normal subjects, a site in the midportion of the vessel was chosen. Baseline ECG, pressure, and IVUS recordings were repeated.
Because the variable progession of atherosclerotic lesions was commonly observed in the coronary artery, the elastic properties of coronary artery are probably different in each coronary site, even in one patient. Then we basically recorded IVUS images at two sites. After baseline recordings, 10 mg nifedipine was sublingually administered while the catheter was left in place. Ten minutes after sublingual drug administration, we again recorded intravascular arterial echograms from the same distal site, ostial pressure, and ECG. In 17 patients in whom recordings were obtained from two sites, the catheter was then withdrawn to the proximal site (recognized by IVUS anatomy, as well as arterial branches and bifurcations, and by plane radiographic image), and recordings were repeated for the final time. In seven of the patients, IVUS images were recorded at plaque and normal sites, and the both were plaque sites in four patients and were normal sites in six patients. The tip of IVUS probe position in coronary artery is confirmed by fluoroscopic monitor and coronary angiography for each recording. All intravascular ultrasound images and ostial pressures were recorded with lead II of the ECG on 1/2-inch super VHS videotape and as a hard-copy freeze frame by a strip-chart recorder at a paper speed of 25 mm/s.
In six patients, the blood concentration of nifedipine was measured at baseline and 10 min after sublingual nifedipine.
IVUS images were replayed on the IVUS system after completing the protocol. For each intravascular echogram, we identified the endothelium-blood border as the brightest echo signal innermost to the transducer artifact and traced this surface by using a calibrated quantitation system and electronic calipers contained within the machine. We measured the cross-sectional area (A) of this traced signal. We then identified the endothelium-blood border of the vessel wall that was normal and that which contained atherosclerotic plaque and measured the segmental perimeter (S) for each. For each recording, measurements were performed at both end diastole and peak systole. End diastole was taken as the peak of the R wave of the ECG. Peak systole was taken as the ECG point coinciding with maximal systolic pressure identified on a simultaneous recording of pressure and ECG. All values represent the mean of five cardiac cycles.
For the purposes of this study, the identification of plaque was based on visual analysis of the IVUS images for increased intimal-medial thickness, high-intensity reflectors within the vessel wall, shadowing, and luminal encroachment. Recognition of plaque was assisted by comparison with normal vessel segments. Although no specific measurements were applied, all vessel segments judged to be normal were found to have an intimal-medial thickness of <0.7 mm by IVUS. We also classified atherosclerotic plaque as hard rather than soft if high-amplitude reflectors with acoustic shadowing were observed within the lesion. On the basis of visual examination, 20 sites were classified as normal and 17 as abnormal. Sixteen abnormal sites had at least some normal wall perimeter, whereas one abnormal site was completely involved with plaque.
From the measurement of arterial size and pressure, we calculated the change in A (in mm2), and S (in mm) per mm Hg pressure change, according to the following formulae: Equations (1)-(2)
To explore the interobserver variability in measuring IVUS images, at 20 randomly selected coronary sites, the circumferential length and at 17 sites the cross-sectional area were measured separately by two independent observers. In addition, one observer measured the same IVUS images twice within 30 min to evaluate intraobserver variability.
All values are expressed as mean ± standard deviation. Values for the elasticity parameters at baseline were compared with those obtained from nifedipine by using paired Student's t test, and those between normal and plaque sites at baseline were compared by using unpaired Student's t test. We performed a multiple regression analysis by taking the cross-sectional variation (ΔA/ΔP or ΔS/ΔP) and peak systolic pressure, end-diastolic pressure, mean ostial pressure, pulse pressure, and heart rate as other variables. A probability (p) value of <0.05 was considered significant.
All patients enrolled in the study completed the protocol without adverse events. Technically adequate IVUS images were obtained in all subjects. On the basis of the criteria previously described (13), patients with atherosclerosis were judged to have soft plaque, whereas four were classified as having hard plaque.
Pulsatile change in the coronary artery at baseline
At baseline, the mean heart rate was 79 ± 17 beats/min for the normal subjects and 79 ± 15 beats/min for the patients with CAD. For normal subjects, peak systolic, enddiastolic, and mean ostial pressure were 137 ± 34, 87 ± 14, and 104 ± 20 mm Hg, whereas those for patients with CAD were 130 ± 25, 74 ± 13, and 93 ± 15 mm Hg, respectively. No significant difference between normal subjects and those with CAD was observed (Table 1).
Baseline values for pulsatile change detected by IVUS were similar for normal subjects and patients with atherosclerosis. A minuscule increase in coronary arterial cross-sectional area was observed from end diastole to peak systole at baseline: from 11.4 ± 3.5 to 11.7 ± 3.7 mm2 at the 20 normal sites and from 9.1 ± 3.8 to 9.5 ± 4.0 mm2 at the 17 lesion sites. The increase in arterial cross-sectional area per change in pressure (ΔA/ΔP) was 8.5 ± 10.2 × 10−3 mm2/mm Hg at normal sites and 6.6 ± 7.0 × 10−3 mm2/mm Hg at lesion sites. The difference in ΔA/ΔP was observed between normal and lesion sites at baseline was not statistically significant.
In those arterial sites with atherosclerosis, the segment length at end diastole was 5.9 ± 3.0 mm for normal segments, 4.7 ± 1.8 mm for soft-plaque segments, and 6.1 ± 3.5 mm for hard-plaque segments; it increased at peak systole to 6.2 ± 3.3 mm for normal and 5.0 ± 1.7 mm for soft-plaque but did not change for hard-plaque segments (6.0 ± 3.5 mm). Changes in segment length per pressure change (ΔS/ΔP) at baseline were similar at normal sites (4.5 ± 8.7 × 10−3 mm/mm Hg) and soft-plaque sites (5.0 ± 3.6 × 10−3 mm/mm Hg), both of which were greater than that at hard-plaque sites (−1.1 ± 0.3 × 10−3 mm/mm Hg; p = 0.002 vs. normal; p = 0.00005 vs. soft).
The ΔA/ΔP and ΔS/ΔP were not significantly related to peak systolic pressure, end-diastolic pressure, mean ostial pressure, pulse pressure, and heart rate.
Effects of nifedipine on pulsatile change
The administration of nifedipine produced substantial reductions in blood pressure with a concomitant small acceleration of heart rate. At 10 min after sublingual administration of nifedipine, heart rate was 83 ± 16 beats/min (p = 0.03 vs. baseline) for normal subjects and was unchanged from baseline for those with CAD (81 ± 16 beats/min). After nifedipine, peak systolic, end-diastolic, and mean ostial pressures for normal subjects were 123 ± 28 (p = 0.01 vs. baseline), 80 ± 14 (p = 0.007 vs. baseline), and 94 ± 18 mm Hg (p = 0.006 vs, baseline), and those for patients with CAD were 118 ± 23 (p = 0.007 vs. baseline), 68 ± 13 (p = 0.03 vs. baseline), and 85 ± 15 mm Hg (p = 0.01 vs. baseline). The hemodynamic response to nifedipine was similar in normal subjects and patients with CAD (Table 1).
After administration of sublingual nifedipine, coronary arterial cross-sectional area at end diastole increased to 12.9 ± 4.6 mm2 at normal sites, and 9.4 ± 3.7 mm2 at lesion sites. Peak systolic cross-sectional area was 13.6 ± 4.8 mm2 at normal (Fig. 1), and 9.7 ± 3.7 mm2 at lesion sites. Compared with baseline, the ΔA/ΔP increased to 16.5 ± 14.4 × 10−3 mm2/mm Hg (p = 0.005) after nifedipine at normal sites. Although ΔA/ΔP increased at some disease sites, it did not change for the group of athero-sclerotic lesions (6.7 ± 7.1 × 10−3 mm2/mm Hg; Fig. 2).
After administration of nifedipine, segment length at end diastole was 5.9 ± 3.1 mm for normal segments, 4.8 ± 1.8 mm for soft-plaque segments, and 5.9 ± 3.4 mm for hard-plaque segments, and increased at peak systole to 6.4 ± 3.5 mm for normal subjects, 5.1 ± 1.7 mm for soft-plaque segments, but was not different for hard-plaque segments (6.0 ± 3.3 mm), respectively. Compared with baseline, the ΔS/ΔP increased to 9.9 ± 10.9 × 10−3 mm/mm Hg (p = 0.02) after nifedipine in normal segments. Although ΔS/ΔP increased at some CAD sites after nifedipine, especially those with soft plaque, this index was not altered for the group of soft-plaque (6.1 ± 4.8 × 10−3 mm/mm Hg) or hard-plaque (1.4 ± 1.6 × 10−3 mm/mm Hg) segments (Fig. 3).
Blood concentration of nifedipine
In six patients, postadministration nifedipine blood concentration was measured at 10.8 ± 10.0 ng/ml. The changes in area and segment length per change in pressure were not significantly related to change in nifedipine blood concentration.
Interobserver and intraobserver variability
Comparison of circumference and cross-sectional area measurements obtained by the two independent observers demonstrated a high correlation, as did the two separate measurements by a single observer (Fig. 4). In measurement of circumference, the mean difference between observer 1 and observer 2 was −0.04 mm with a 95% confidence interval of −0.06 to −0.01, and that between measurement 1 and measurement 2 was −0.02 mm with a 95% confidence interval of −0.05 to 0.01. In measurement of area, the mean difference between observer 1 and observer 2 was −0.02 mm2 with a 95% confidence interval of −0.03 to 0, and that between measurement 1 and measurement 2 was −0.07 mm2 with a 95% confidence interval of −0.08 to −0.06. The limits of agreement are small enough for measuring the circumference and cross-sectional area. Thus interobserver and intraobserver reproducibility in this study was good.
Changes early in the course of atherosclerosis, including alteration in the distribution of fibronectin and various types of collagen (8), may alter arterial elasticity before morphologic changes can be detected in the vessel wall. The results of our study demonstrate that IVUS provides an excellent clinical modality for assessing the elastic properties of coronary arteries. In addition, by using pressure measurements derived from the coronary ostium and ultrasound recordings, we demonstrated that the calcium-entry blocker nifedipine alters indexes of vessel elasticity in normal segments of the vessel wall, but that this change is blunted in those with mild atherosclerosis.
Arterial wall elasticity is regulated by factors inherent in the structural composition of the wall, including collagen fibers, elastic fibers (9), and the vasomotor tone of smooth muscle (10). Therefore it is anticipated that arterial elasticity should be susceptible to alteration by a variety of pharmacologic agents. Prior studies demonstrated that stimulation of smooth muscle by norepinephrine or nitroglycerin altered the elasticity of arteries (11-15). The other commonly used class of drugs that affect smooth-muscle tone is the calcium channel blockers. Although the effect of calcium blockers on coronary arterial lumen size has been reported (16-18), few data exist regarding alterations in regional coronary arterial wall elasticity. Our data show that the parameters derived from the pressure-segment length relation at the normal wall increased after sublingual nifedipine administration, documenting the influence of nifedipine on coronary artery elasticity.
The data in our study clearly establish that atherosclerosis in the mild stage blunts or eliminates the effect of nifedipine on vessel elasticity. It seems clear, however, that nifedipine-induced changes in elasticity are markedly altered in larger lesions and hard plaque. The mechanism for this impaired responsiveness is uncertain. Although it is possible that atherosclerosis inhibits some endothelial cell-mediated action induced by nifedipine, it seems more likely that plaques modulate vascular stiffness by atrophy of the muscular media or by physical constraints on the ability of vascular tissue to expand. The results of these studies suggest that the vasomotor effects of dihydropyridine derivatives in particular, and perhaps calcium entry blockers in general, may be limited in the setting of stenotic lesions involving the entire vessel perimeter.
In light of the size of the available catheter probes and the experimental nature of the study, the lesions examined in this protocol were generally mild in severity. Thus although the cross-sectional areas of the diseased sites were slightly less than those of the normal sites, these differences did not reach statistical significance. The mild degree of disease also was reflected by the fact that some normal wall perimeter existed for nearly all the lesions examined. Accordingly, the change in size during the cardiac cycle and change in area and segmental perimeter per change in pressure also were similar for normal and atherosclerotic vessels. Because the response to nifedipine for segments with soft plaque differed from that of segments with hard plaque, we anticipate that the findings observed in this study would be more marked for lesions of greater severity.
Previously reported measurements of aortic wall viscoelastic properties (19-22) demonstrated that a close relation exists between vessel tension and circumference and was best fitted exponentially. Because exponential relations probably exist between pressure and area or pressure and circumference of coronary arteries, two-point measures may not be sufficient for calculating an index of coronary artery elasticity. Cunha et al. (23) demonstrated the correlation of pulsatile diameter change of the carotid artery with hemodynamic parameters, which may result from the exponential relation between diameter and pressure. In our study, we did not find the significant correlation between distensibility indexes and hemodynamics, because ΔA/ΔP and ΔS/ΔP were corrected by pressure variation. Additionally, Nakatani et al. (24) demonstrated that the distensibility index, ΔA/ΔP, was available for evaluating the arterial distensibility in the clinical setting. Thus although the ΔA/ΔP derived from the cyclic variation of coronary arterial cross-section and pressure in our study were not absolute parameters, an alteration in coronary arterial wall elasticity caused by nifedipine could be readily detected and assessed by these simplified and relative indexes.
We used the ΔS/ΔP as an index indicating regional wall distensibility in the coronary artery, which was negative in some patients. We thought that the cyclic alteration in arterial wall segmental perimeter was regulated by the physical factor of arterial wall and endothelial function. The increases in pressure during systole and in flow volume during diastole may tend to dilate the artery. Therefore although the pressure factor is commonly dominant, the volume factor may be larger than the pressure factor in some cases, which may be the possible mechanism of negative ΔS/ΔP.
Several factors could have influenced the results of our study. Assessment of the relation between arterial size and pressure during a single cardiac cycle is a direct method of evaluating elasticity. In our study, we obtained pressure data at the coronary ostium but not at the ultrasound imaging site within the left anterior descending (LAD), left circumflex (LCX), or right coronary artery (RCA), because a pressure sensor was not present at the tip of catheter probe. Any difference in or loss of pressure at the image site compared with the ostium may have influenced the indexes that were calculated. However, because significant stenosis did not exist between the ostium and the imaging site, the cyclic variation of pressure at the imaging site was likely parallel to ostial pressure. Additionally, the modest lesions studied readily accommodate the 1.2-mm catheter, and half of our intravascular ultrasound data were derived from images close to the ostium and are similar to the overall results. Thus it appears the effects of the site of pressure measurement were minimized in our study.
Some subjects were continuously treated with different cardiovascular drugs, nitrates, β-blockers, or ACE inhibitors during the protocol, which may influence the arterial wall elasticity. This limits considerably the interpretation of data at baseline and even after nifedipine if we consider the possible drug interaction. However, the same dosages of these drugs were maintained for >6 months before the protocol. Therefore if these drugs may affect the arterial wall function and cause a possible drug interaction with nifedipine, the influence on results of our study was within small limits.
Certain constraints exist regarding the IVUS methods applied in our study. Although every attempt was made to maintain the catheter probe in the identical site for imaging, the possibility of slight catheter displacement cannot be entirely eliminated. In addition, distortions of morphology induced by artifacts, eccentricity, or nonuniform rotational velocity may have influenced our results. However, data were obtained before and after an intervention, and imaging distortion was likely constant throughout the procedure.
The data from our study have several implications. It is clear that isolated measures of change in pressure or lumen size cannot be extrapolated to indicate alterations of arterial elasticity. Thus nifedipine produced small changes in cross-sectional area and pressure, but the pressure-area or pressure-segment length relation indicative of elasticity was considerably altered by this intervention. Therefore assessment of coronary artery elasticity will require some measure of luminal size, such as that provided by IVUS. Moreover, IVUS was able to provide accurate measurements of the circumferential expanse per pressure of an atheroma in a vessel with eccentric disease. In this way, we were able to define the influence of various pharmacologic agents on diseased and nondiseased segments of the vessel wall. Such determinations may be of great value in selecting optimal therapy for patients with CAD.
1. Milnor WR. Arterial impedance as ventricular afterload. Circ Res
2. Nichols WW, Conti CR, Walker WE, Milnor WR. Input impedance of the systemic circulation in man. Circ Res
3. Yock PG, Johnson EL, Linker DT. Intravascular ultrasound: development and clinical potential. Am J Cardiac Imaging
4. Tobis JM, Mallery JA, Gessert J, et al. Intravascular ultrasound cross-sectional arterial imaging before and after balloon angioplasty in vitro. Circulation
5. Gussenhoven EJ, Essed CE, Lancee CT, et al. Arterial wall characteristics determined by intravascular ultrasound imaging: an in vitro study. J Am Coll Cardiol
6. Potkin BN, Bartorelli AL, Gessert JM, et al. Coronary artery imaging with intravascular high-frequency ultrasound. Circulation
7. Tobis JM, Mallery J, Mahon D, et al. Intravascular ultrasound imaging of human coronary arteries in vivo: analysis of tissue characterizations with comparison to in vitro histological specimens. Circulation
8. Shekhonin BV, Domogatsky SP, Idelson GL, Koteliansky VE, Rukosuev VS. Relative distribution of fibronectin and type I, III, IV, V collagens in normal and atherosclerotic intima of human arteries. Atherosclerosis
9. Roach MR, Burton AC. The reason for the shape of the distensibility curves of arteries. Can J Biochem Physiol
10. Stone DN, Dujardin JPL. Changes in smooth muscle tone influence characteristic impedance of the aorta. Am J Physiol
11. Dobrin PB, Rovick AA. Influence of vascular smooth muscle on contractile mechanics and elasticity of arteries. Am J Physiol
12. Fitchett DH. Forearm arterial compliance: a new measure of arterial compliance? Cardiovasc Res
13. Smulyan H, Mookherjee S, Warner RA. The effect of nitroglycerin on forearm arterial distensibility. Circulation
14. Latson TW, Hunter WC, Katoh N, Sagawa K. Effect of nitroglycerin on aortic impedance, diameter, and pulse-wave velocity. Circ Res
15. Westling H, Jansson L, Jonson B, Nilsen R. Vasoactive drugs and elastic properties of human arteries in vivo, with special reference to the action of nitroglycerine. Eur Heart J
16. Vatner SF, Hintze TH. Effects of a calcium-channel antagonist on large and small coronary arteries in conscious dogs. Circulation
17. Feldman RL, Hill JA, Conti JB, Conti CR, Pepine CJ. Analysis of coronary responses to nifedipine alone and in combination with intracoronary nitroglycerin in patients with coronary artery disease. Am Heart J
18. Schwartz JS, Bache RJ. Combined effects of calcium antagonists and nitroglycerin on large coronary artery diameter. Am Heart J
19. Hickler RB. Aortic and large artery stiffness: current methodology and clinical correlations. Clin Cardiol
20. Imura T, Yamamoto K, Satoh T, Kanamori K, Mikami T, Yasuda H. In vivo viscoelastic behavior in the human aorta. Circ Res
21. Bourgeois MJ, Gilbert BK, Donald DE, Wood EH. Characteristics of aortic diastolic pressure decay with application to the continuous monitoring of changes in peripheral vascular resistance. Circ Res
22. O'Rourke M. Arterial stiffness, systolic blood pressure, and logical treatment of arterial hypertension. Hypertension
23. Cunha RS, Benetos A, Laurent S, Safar ME, Asmar RG. Distension capacity of the carotid artery and ambulatory blood pressure monitoring: effects of age and hypertension. Am J Hypertens
24. Nakatani S, Yamagishi M, Tamai J, et al. Assessment of coronary artery distensibility by intravascular ultrasound: application of simultaneous measurements of luminal area and pressure. Circulation