- Email: [email protected]
Coronary angiography and intravascular ultrasound
Coronary Angiography and Intravascular Ultrasound Steven Nissen,
MD
Clinicians have long used the size of the lumen and the angiogram as a predictor of coronary events. However, cardiovascular disease is not a disease of the lumen but a disease of the vessel wall. In early stages, atherosclerosis outwardly remodels the external elastic membrane; only late in the disease process does luminal narrowing occur, enabling angiographic detection. This has profound implications for drug therapy, because approximately 70% of patients present with acute myocardial infarction (MI) or sudden death, not angina as
the first symptom of coronary disease. Intravascular ultrasound (IVUS) can provide detailed images of the artery and is the only technique currently available that enables physicians to routinely visualize coronary plaques. Due to its sensitivity in measuring plaque volume and content, IVUS may be a useful surrogate marker to evaluate the atherosclerotic process in smaller numbers of patients than required for conventional clinical endpoint trials. 䊚2001 by Excerpta Medica, Inc. Am J Cardiol 2001;87(suppl):15A–20A
ore than 40 years ago, Mason Sones performed the world’s first coronary angiography and M thereby launched the contemporary era of cardiovas-
INTRAVASCULAR ULTRASOUND AND CORONARY REMODELING
cular medicine. However, despite the enormous contributions of coronary angiography to clinical medicine, this imaging modality has misled practitioners about the true nature of the atherosclerotic disease process. By its very nature, angiography focuses our attention on narrowing of the coronary lumen as a means to detect atherosclerotic disease and predict coronary events. However, coronary disease is largely not a disease of the lumen but an abnormality of the vessel wall. Regardless of the size of the lumen, if a plaque ruptures and a thrombus develops, a patient can experience unstable angina, acute myocardial infarction (MI), or sudden cardiac death. In reality, the size of the lumen is not a very good predictor of which patients will have morbid events. Only 14% of all clinical events occur in lesions that are ⱖ70% narrowed, and the majority of coronary events occur at the site of a “hemodynamically insignificant” coronary lesion.1 Intravascular ultrasound (IVUS), developed over the past decade, has many potential advantages in characterizing the atherosclerotic disease process.2 IVUS uses high frequency ultrasound to image not only the coronary lumen, but also the structure of the vessel wall, including the atherosclerotic plaque. The catheter is approximately 1 mm in diameter, operates typically at 30 – 40 MHz, and generates full-motion, cross-sectional images at 30 frames per second. With modern devices, axial resolution is approximately 50 – 80 m, which is approximately 1 order of magnitude greater resolution than available by angiography.
From the The Cleveland Clinic Foundation, Cleveland, Ohio, USA. Address for reprints: Steven Nissen, MD, The Cleveland Clinic Foundation, Department of Cardiology, F15, 9500 Euclid Avenue, Cleveland, Ohio 44195. ©2001 by Excerpta Medica, Inc. All rights reserved.
IVUS has been extraordinarily useful in teaching us about the atherosclerotic disease process. We have learned that the traditional model of the disease, in which the plaque develops in the vessel wall over many years, gradually narrowing the lumen to produce symptoms, is not accurate. A more accurate model of atherosclerosis was originally described by Glagov et al,3 who showed that coronary “remodeling” enables patients to develop large atherosclerotic plaques without reduction in lumen size (Figure 1). In essence, during the early progression of disease, atherosclerotic material grows into the vessel wall, outwardly displacing the external elastic membrane. Accordingly, patients can develop a large atherosclerotic plaque with no luminal narrowing. They hypothesized that this compensatory expansion maintained a constant lumen, which would protect patients against ischemic symptoms. They also proposed that luminal narrowing would occur only late in the disease process. The consequences of Glagov et al’s hypothesis are profound and influence how we diagnose coronary artery disease. Because the lumen is not narrowed during the initial disease process, angiography or other traditional measures would not detect atherosclerosis. In fact, coronary remodeling is no longer a hypothesis, but a fundamental mechanism of early disease development. IVUS has demonstrated that approximately 95–99% of coronary atherosclerosis involves no luminal narrowing. Therefore, an angiogram will remain relatively normal until the disease is far advanced. Once the angiogram becomes abnormal, it will reveal only about 1–5% of the disease.4
INTRAVASCULAR ULTRASOUND, ANGIOGRAPHY, AND THE ASSESSMENT OF ATHEROSCLEROTIC DISEASE: TRIAL RESULTS To illustrate, Figure 2 shows a typical patient with disease of the right coronary. This patient had a 95% obstruction treated successfully with a stent. However, 0002-9149/01/$ – see front matter PII S0002-9149(01)01420-5
15A
FIGURE 1. Glagov et al’s coronary remodeling hypothesis: compensatory expansion maintains the lumen diameter despite increasing plaque size. (Adapted from N Engl J Med.3)
FIGURE 2. Intravascular ultrasound and angiography of the left anterior descending artery in a patient with a 95% stenosis of the right coronary. Although the angiogram is normal (left panel) ultrasound at 4 sites (right panels) (A–D) shows diffuse atherosclerosis.
although the left coronary system appears normal angiographically, using IVUS, multiple large plaques were identified throughout the artery. Although the entire artery was diffusely atherosclerotic, the angiogram remained normal because the lumen size was not altered. Thus, ultrasound can be used to provide very detailed images of the artery and is currently the only technique available enabling physicians to see coronary plaques directly. Figure 3 demonstrates another example of this remodeling phenomenon. In this case, the patient is a presumably healthy 32-year-old woman, who had a lethal head injury after a motor vehicle accident and whose heart was obtained for transplantation. Angiography showed no evidence of luminal narrowing. 16A THE AMERICAN JOURNAL OF CARDIOLOGY姞
However, IVUS shortly after transplantation showed many large atherosclerotic lesions, none of which had narrowed the lumen (Figure 3). At the age of 32, 100% of her coronary disease was extraluminal. My colleague E. Murat Tuzcu and I studied 262 young people from the United States who died in motor vehicle accidents and were heart transplant donors. This study showed that at an average age of 31.5 years, atherosclerotic plaques were large and abundant as identified with IVUS. However, among these subjects, 98% had completely normal angiograms.5 Surprisingly, this study demonstrated that women had nearly as much disease at an early age as did men. This study supports the concept that at least 90% of the coronary arterial tree is atherosclerotic
VOL. 87 (4A)
FEBRUARY 16, 2001
FIGURE 3. Transplant donor: an example of coronary remodeling in a 32-year-old, supposedly healthy woman.
before clinical coronary disease is evident. This study further divided the heart transplant donors to determine the prevalence of atherosclerosis by age. In the group 13–19 years of age (approximately 50 subjects), 1 in 6 had at least 1 atheroma, defined as an intimal thickness ⬎0.5 mm. Among subjects aged 20 –29, 1 in 3 had such an atheroma, and among those aged 30 –39 the occurrence was 60% (Figure 4). This study demonstrates that, as previously reported by necropsy studies, atherosclerosis begins at a surprisingly young age. In another study, we asked practitioners who were about to perform angioplasty on single-vessel coronary disease to place the IVUS probe at the most normal site in the artery. On average, these “normal reference” sites contained an atherosclerotic plaque occupying about 40% of the external elastic membrane. In addition, IVUS offers the potential to identify plaques that are more vulnerable to rupture and produce acute coronary events. Vulnerable plaques are characterized by a thin fibrous cap and a large lipid core.6 IVUS can characterize the plaque because the lipid appears primarily sonolucent, whereas fibrous material is more echogenic. Lipid does not have very much collagen and other materials that reflect ultra-
sound, so it is dark, or sonolucent. IVUS can also roughly measure the thickness of the fibrous cap. By IVUS, some plaques will exhibit a thick, dense, and stable fibrous cap with a very small lipid core. Thus, IVUS offers the opportunity to assess both the quality and the quantity of atherosclerotic plaque. This, we hope, can enable a better understanding of how pharmacotherapy stabilizes coronary disease. Many angiographic studies have shown that most MIs occur at sites with low-grade stenoses.1 This has led to the presumption that MIs are caused by the rupture of small plaques. We examined a group of patients with stable angina and unstable symptoms (MI or unstable angina), examining the culprit lesions by IVUS, not angiography.7 In this study, the unstable lesions exhibited more prominent outward modeling of the external elastic membrane compared with lesions producing stable angina. Although there was minimal stenosis, the plaques were quite large and had so remodeled the artery that they never produced a significant stenosis. In the stable angina patients, exactly the opposite was seen; the external elastic membrane was actually smaller within the culprit lesion than at a nearby reference site. According to Glagov’s theory, the assumption is that the remodeling is adaptive, so that as plaques get bigger, the artery enlarges
A SYMPOSIUM: SURROGATE MARKERS OF ATHEROSCLEROSIS
17A
FIGURE 4. Prevalence of atherosclerosis by donor age.
FIGURE 5. Atheroma regression: “reverse” remodeling. Coronary remodeling in a heart transplant patient.
and protects the individual from stenosis and ischemia. However, our data indicate that although this adaptive change may prevent stenosis, it may actually predispose to an acute coronary event.7 It is widely accepted that lowering low-density lipoprotein cholesterol pharmacologically can reduce the event rate in patients. However, serial angiograms in those patients demonstrate very little improvement in percent stenosis, despite significant benefits in clinical event reduction.8 IVUS is a more sensitive means 18A THE AMERICAN JOURNAL OF CARDIOLOGY姞
of determining plaque volume. Figure 5 shows an atherosclerotic heart that was transplanted into a patient with low coronary atherosclerotic risk. This figure demonstrates that the atheroma that was originally present decreased from about 8 to about 6 mm2, although the lumen size did not change. The external elastic membrane remodeled as the plaque regressed without changing the lumen size. This is likely the best explanation for the discordance between clinical and angiographic benefit.
VOL. 87 (4A)
FEBRUARY 16, 2001
FIGURE 6. Design of the Reversal of Atherosclerosis with Lipitor (REVERSAL) study, which examines the effects on volume of plaque, treating patients with either 80 mg atorvastatin or 40 mg pravastatin. REVERSAL is a prospective, randomized, double-blind, multicenter study of endpoint-percent change in coronary atheroma volume comparing data from the end of the study to data from the beginning of the study.
An IVUS study involving 25 patients randomized half to 10 mg pravastatin per day for 3 years and the others to placebo. Using volumetric plaque assessment by IVUS, after 3 years, a 41% increase in atheroma volume was observed in the placebo patients and a 7% decrease in the pravastatin patients (p ⫽ 0.0005). A major trial using IVUS is underway to compare aggressive to moderate lipid lowering therapy. The Reversal of Atherosclerosis with Lipitor (REVERSAL) study examines the effects of treating patients with either 80 mg atorvastatin or 40 mg pravastatin (Figure 6). The endpoint of this trial is not lumen size but volumetric plaque in the coronary artery. Volume of plaque is obtained by measuring the external elastic membrane, lumen area, and plaque and repeating this measure every millimeter for at least 25–50 mm of artery. This provides the study with extraordinary statistical power to detect regression or progression.
CONCLUSION In summary, IVUS has great potential as a means to accurately assess the atherosclerotic disease process. IVUS has the ability to demonstrate changes in plaque volume and plaque vulnerability over a relatively short period of time, with fewer patients than required for large morbidity and mortality endpoint trials. In the future, for IVUS to become a widely accepted surrogate, there needs to be a clear correlation between plaque volume by IVUS and cardiovascular events. If this correlation can be achieved, IVUS may provide the best tool for evaluating antiatherosclerotic therapies.
DISCUSSION Michael Cressman: How did you figure in the potential for heterogeneity of a response of atherosclerotic lesions into sample size calculations for these studies? Steven Nissen, MD (Cleveland, Ohio): We picked
total volume as the primary endpoint, but prespecified a series of secondary endpoints. For example, we take rolling 10-slice segments, then take the segments with the least atherosclerotic plaque, and that represents a defined endpoint. We take the 10 slices with the greatest atherosclerotic plaque, and that represents another endpoint. Then you can ask, is the primary effect of intensive low-density lipoprotein lowering with a statin to retard the progression of the early lesion or to regress the advanced lesion? Then you can define within the database. We get 30,000 slices (50 slices per patient in 600 patients), and each is potentially an independent variable. It’s not really, but it may be that the very sonolucent and soft plaques are the ones where the greatest changes occur. Because of the methodology and the way we define the secondary endpoints, we’ll be able to figure that out from our data. Michael H. Davidson, MD (Chicago, Illinois): Any evidence that change in IVUS plaque volume correlates with change in cardiovascular events? Dr. Nissen: We don’t have such a trial yet, but we will. There is another trial we’re doing (a 3-arm trial) for conventional endpoints, so we wrapped an endpoint trial around an IVUS volumetric trial. So, if carotid ultrasound can predict coronary events, then certainly coronary ultrasound should be able to predict coronary events. Rem Bolton, PhD (Charlottesville, Virginia): What is fascinating about your data is that it looks almost identical to the Brunich studies that have been done on the carotid artery with transcutaneous ultrasound where they were able to show that the lesions, at first, remodeled and got larger, preserving the lumen, and then above a certain level of plaque they started to shrink and compromise the lumen. The principal appears to be the same in the carotid and the coronary artery. David Herrington, MD (Winston-Salem, North Carolina): That disparity between clinical and angiographic progression slide shows relative change in events compared with absolute change in a metric
A SYMPOSIUM: SURROGATE MARKERS OF ATHEROSCLEROSIS
19A
whose units happen to be percentage. If you look at relative change in events compared with relative change in progression of coronary disease, there is absolutely no discontinuity whatsoever. There is 50 – 70% relative reduction in events and there is 45–75% relative reduction in the rate of progression of disease measured by angiography. So, to promote the notion that there is a big disconnection between progression measured angiographically and progression or regression clinically I think does a disservice to the interpretation of the large body of data that shows that altering progression of coronary disease measured angiographically does indeed predict changes in risk for future events—a body of data for which there is not yet comparable evidence with respect to IVUS. I would not want to be misperceived as being unenthusiastic about IVUS, but to promote IVUS at the expense of the very valuable information that does reside with angiography is a disservice. Dr. Nissen: I disagree but will reserve my comments for further discussion. Karen Friday, MD (New Orleans, Louisiana): Based on your presentation, the great majority of these patients should be dead, which is not the case. We’ve known for a long time that young people do have plaques, but it takes more than a plaque to produce an event. Dr. Nissen: Well, a large number of these people should be dead, and in fact, approximately half the people in this room will die from the disease. That is the point. Yes, it’s true there is a long asymptomatic lead time, but when a man has an MI at age 45, the disease did not start when he was 43, it started when
20A THE AMERICAN JOURNAL OF CARDIOLOGY姞
he was 17. I think we understand why it takes time, but the fact is that you have the disease and I have the disease, and approximately half the people in this room will die of the disease. We need to identify pharmacologic strategies for treating people earlier and more aggressively before they have their first event. The problem is that we don’t get second chances, because with 60 –70% of coronary disease, the very first symptom is either sudden death or acute MI. I think there is a lot of room for the pharmaceutical industry to develop better drugs to be used more aggressively at an earlier age, if we can do so safely. 1. Little WC, Constantinescu M, Applegate RJ, Kutcher MA, Burrows MT, Kahl
FR, Santamore WP. Can coronary angiography predict the site of a subsequent myocardial infarction in patients with mild-to-moderate coronary artery disease? Circulation 1988;78:1157–1166. 2. Nissen SE, Gurley JC, Grines CL, Booth DC, McClure R, Berk M, Fischer C, DeMaria AN. Intravascular ultrasound assessment of lumen size and wall morphology in normal subjects and patients with coronary artery disease. Circulation 1991;84:1087–1199. 3. Glagov S, Weisneberg E, Zarins CK, Stankunavicius R, Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 1987;316:1371–1375. 4. Topol EJ, Nissen SE. Our preoccupation with coronary luminology: the dissociation between clinical and angiographic findings in ischemic heart disease. Circulation 1995;92:2333–2342. 5. Tuzcu EM, Hobbs RE, Rincon G, Bott-Silverman C, De Franco AC, Robinson K, McCarthy PM, Stewart RW, Guyer S, Nissen SE. Occult and frequent transmission of atherosclerotic coronary disease with cardiac transplantation: insights from intravascular ultrasound. Circulation 1995;91:1706 –1713. 6. Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation 1995;92: 657– 671. 7. Schoenhagen P, Ziada KM, Kapadia SR, Crowe TD, Nissen SE, Tuzcu EM. Extent and direction of arterial remodeling in stable versus unstable coronary syndromes: an intravascular ultrasound study. Circulation 2000;101:598 – 603. 8. Brown BG, Zhao XQ, Sacco DE, Albers JJ. Lipid lowering and plaque regression. New insights into prevention of plaque disruption and clinical events in coronary disease. Circulation 1993;87:1781–1791.
VOL. 87 (4A)
FEBRUARY 16, 2001
MD
Clinicians have long used the size of the lumen and the angiogram as a predictor of coronary events. However, cardiovascular disease is not a disease of the lumen but a disease of the vessel wall. In early stages, atherosclerosis outwardly remodels the external elastic membrane; only late in the disease process does luminal narrowing occur, enabling angiographic detection. This has profound implications for drug therapy, because approximately 70% of patients present with acute myocardial infarction (MI) or sudden death, not angina as
the first symptom of coronary disease. Intravascular ultrasound (IVUS) can provide detailed images of the artery and is the only technique currently available that enables physicians to routinely visualize coronary plaques. Due to its sensitivity in measuring plaque volume and content, IVUS may be a useful surrogate marker to evaluate the atherosclerotic process in smaller numbers of patients than required for conventional clinical endpoint trials. 䊚2001 by Excerpta Medica, Inc. Am J Cardiol 2001;87(suppl):15A–20A
ore than 40 years ago, Mason Sones performed the world’s first coronary angiography and M thereby launched the contemporary era of cardiovas-
INTRAVASCULAR ULTRASOUND AND CORONARY REMODELING
cular medicine. However, despite the enormous contributions of coronary angiography to clinical medicine, this imaging modality has misled practitioners about the true nature of the atherosclerotic disease process. By its very nature, angiography focuses our attention on narrowing of the coronary lumen as a means to detect atherosclerotic disease and predict coronary events. However, coronary disease is largely not a disease of the lumen but an abnormality of the vessel wall. Regardless of the size of the lumen, if a plaque ruptures and a thrombus develops, a patient can experience unstable angina, acute myocardial infarction (MI), or sudden cardiac death. In reality, the size of the lumen is not a very good predictor of which patients will have morbid events. Only 14% of all clinical events occur in lesions that are ⱖ70% narrowed, and the majority of coronary events occur at the site of a “hemodynamically insignificant” coronary lesion.1 Intravascular ultrasound (IVUS), developed over the past decade, has many potential advantages in characterizing the atherosclerotic disease process.2 IVUS uses high frequency ultrasound to image not only the coronary lumen, but also the structure of the vessel wall, including the atherosclerotic plaque. The catheter is approximately 1 mm in diameter, operates typically at 30 – 40 MHz, and generates full-motion, cross-sectional images at 30 frames per second. With modern devices, axial resolution is approximately 50 – 80 m, which is approximately 1 order of magnitude greater resolution than available by angiography.
From the The Cleveland Clinic Foundation, Cleveland, Ohio, USA. Address for reprints: Steven Nissen, MD, The Cleveland Clinic Foundation, Department of Cardiology, F15, 9500 Euclid Avenue, Cleveland, Ohio 44195. ©2001 by Excerpta Medica, Inc. All rights reserved.
IVUS has been extraordinarily useful in teaching us about the atherosclerotic disease process. We have learned that the traditional model of the disease, in which the plaque develops in the vessel wall over many years, gradually narrowing the lumen to produce symptoms, is not accurate. A more accurate model of atherosclerosis was originally described by Glagov et al,3 who showed that coronary “remodeling” enables patients to develop large atherosclerotic plaques without reduction in lumen size (Figure 1). In essence, during the early progression of disease, atherosclerotic material grows into the vessel wall, outwardly displacing the external elastic membrane. Accordingly, patients can develop a large atherosclerotic plaque with no luminal narrowing. They hypothesized that this compensatory expansion maintained a constant lumen, which would protect patients against ischemic symptoms. They also proposed that luminal narrowing would occur only late in the disease process. The consequences of Glagov et al’s hypothesis are profound and influence how we diagnose coronary artery disease. Because the lumen is not narrowed during the initial disease process, angiography or other traditional measures would not detect atherosclerosis. In fact, coronary remodeling is no longer a hypothesis, but a fundamental mechanism of early disease development. IVUS has demonstrated that approximately 95–99% of coronary atherosclerosis involves no luminal narrowing. Therefore, an angiogram will remain relatively normal until the disease is far advanced. Once the angiogram becomes abnormal, it will reveal only about 1–5% of the disease.4
INTRAVASCULAR ULTRASOUND, ANGIOGRAPHY, AND THE ASSESSMENT OF ATHEROSCLEROTIC DISEASE: TRIAL RESULTS To illustrate, Figure 2 shows a typical patient with disease of the right coronary. This patient had a 95% obstruction treated successfully with a stent. However, 0002-9149/01/$ – see front matter PII S0002-9149(01)01420-5
15A
FIGURE 1. Glagov et al’s coronary remodeling hypothesis: compensatory expansion maintains the lumen diameter despite increasing plaque size. (Adapted from N Engl J Med.3)
FIGURE 2. Intravascular ultrasound and angiography of the left anterior descending artery in a patient with a 95% stenosis of the right coronary. Although the angiogram is normal (left panel) ultrasound at 4 sites (right panels) (A–D) shows diffuse atherosclerosis.
although the left coronary system appears normal angiographically, using IVUS, multiple large plaques were identified throughout the artery. Although the entire artery was diffusely atherosclerotic, the angiogram remained normal because the lumen size was not altered. Thus, ultrasound can be used to provide very detailed images of the artery and is currently the only technique available enabling physicians to see coronary plaques directly. Figure 3 demonstrates another example of this remodeling phenomenon. In this case, the patient is a presumably healthy 32-year-old woman, who had a lethal head injury after a motor vehicle accident and whose heart was obtained for transplantation. Angiography showed no evidence of luminal narrowing. 16A THE AMERICAN JOURNAL OF CARDIOLOGY姞
However, IVUS shortly after transplantation showed many large atherosclerotic lesions, none of which had narrowed the lumen (Figure 3). At the age of 32, 100% of her coronary disease was extraluminal. My colleague E. Murat Tuzcu and I studied 262 young people from the United States who died in motor vehicle accidents and were heart transplant donors. This study showed that at an average age of 31.5 years, atherosclerotic plaques were large and abundant as identified with IVUS. However, among these subjects, 98% had completely normal angiograms.5 Surprisingly, this study demonstrated that women had nearly as much disease at an early age as did men. This study supports the concept that at least 90% of the coronary arterial tree is atherosclerotic
VOL. 87 (4A)
FEBRUARY 16, 2001
FIGURE 3. Transplant donor: an example of coronary remodeling in a 32-year-old, supposedly healthy woman.
before clinical coronary disease is evident. This study further divided the heart transplant donors to determine the prevalence of atherosclerosis by age. In the group 13–19 years of age (approximately 50 subjects), 1 in 6 had at least 1 atheroma, defined as an intimal thickness ⬎0.5 mm. Among subjects aged 20 –29, 1 in 3 had such an atheroma, and among those aged 30 –39 the occurrence was 60% (Figure 4). This study demonstrates that, as previously reported by necropsy studies, atherosclerosis begins at a surprisingly young age. In another study, we asked practitioners who were about to perform angioplasty on single-vessel coronary disease to place the IVUS probe at the most normal site in the artery. On average, these “normal reference” sites contained an atherosclerotic plaque occupying about 40% of the external elastic membrane. In addition, IVUS offers the potential to identify plaques that are more vulnerable to rupture and produce acute coronary events. Vulnerable plaques are characterized by a thin fibrous cap and a large lipid core.6 IVUS can characterize the plaque because the lipid appears primarily sonolucent, whereas fibrous material is more echogenic. Lipid does not have very much collagen and other materials that reflect ultra-
sound, so it is dark, or sonolucent. IVUS can also roughly measure the thickness of the fibrous cap. By IVUS, some plaques will exhibit a thick, dense, and stable fibrous cap with a very small lipid core. Thus, IVUS offers the opportunity to assess both the quality and the quantity of atherosclerotic plaque. This, we hope, can enable a better understanding of how pharmacotherapy stabilizes coronary disease. Many angiographic studies have shown that most MIs occur at sites with low-grade stenoses.1 This has led to the presumption that MIs are caused by the rupture of small plaques. We examined a group of patients with stable angina and unstable symptoms (MI or unstable angina), examining the culprit lesions by IVUS, not angiography.7 In this study, the unstable lesions exhibited more prominent outward modeling of the external elastic membrane compared with lesions producing stable angina. Although there was minimal stenosis, the plaques were quite large and had so remodeled the artery that they never produced a significant stenosis. In the stable angina patients, exactly the opposite was seen; the external elastic membrane was actually smaller within the culprit lesion than at a nearby reference site. According to Glagov’s theory, the assumption is that the remodeling is adaptive, so that as plaques get bigger, the artery enlarges
A SYMPOSIUM: SURROGATE MARKERS OF ATHEROSCLEROSIS
17A
FIGURE 4. Prevalence of atherosclerosis by donor age.
FIGURE 5. Atheroma regression: “reverse” remodeling. Coronary remodeling in a heart transplant patient.
and protects the individual from stenosis and ischemia. However, our data indicate that although this adaptive change may prevent stenosis, it may actually predispose to an acute coronary event.7 It is widely accepted that lowering low-density lipoprotein cholesterol pharmacologically can reduce the event rate in patients. However, serial angiograms in those patients demonstrate very little improvement in percent stenosis, despite significant benefits in clinical event reduction.8 IVUS is a more sensitive means 18A THE AMERICAN JOURNAL OF CARDIOLOGY姞
of determining plaque volume. Figure 5 shows an atherosclerotic heart that was transplanted into a patient with low coronary atherosclerotic risk. This figure demonstrates that the atheroma that was originally present decreased from about 8 to about 6 mm2, although the lumen size did not change. The external elastic membrane remodeled as the plaque regressed without changing the lumen size. This is likely the best explanation for the discordance between clinical and angiographic benefit.
VOL. 87 (4A)
FEBRUARY 16, 2001
FIGURE 6. Design of the Reversal of Atherosclerosis with Lipitor (REVERSAL) study, which examines the effects on volume of plaque, treating patients with either 80 mg atorvastatin or 40 mg pravastatin. REVERSAL is a prospective, randomized, double-blind, multicenter study of endpoint-percent change in coronary atheroma volume comparing data from the end of the study to data from the beginning of the study.
An IVUS study involving 25 patients randomized half to 10 mg pravastatin per day for 3 years and the others to placebo. Using volumetric plaque assessment by IVUS, after 3 years, a 41% increase in atheroma volume was observed in the placebo patients and a 7% decrease in the pravastatin patients (p ⫽ 0.0005). A major trial using IVUS is underway to compare aggressive to moderate lipid lowering therapy. The Reversal of Atherosclerosis with Lipitor (REVERSAL) study examines the effects of treating patients with either 80 mg atorvastatin or 40 mg pravastatin (Figure 6). The endpoint of this trial is not lumen size but volumetric plaque in the coronary artery. Volume of plaque is obtained by measuring the external elastic membrane, lumen area, and plaque and repeating this measure every millimeter for at least 25–50 mm of artery. This provides the study with extraordinary statistical power to detect regression or progression.
CONCLUSION In summary, IVUS has great potential as a means to accurately assess the atherosclerotic disease process. IVUS has the ability to demonstrate changes in plaque volume and plaque vulnerability over a relatively short period of time, with fewer patients than required for large morbidity and mortality endpoint trials. In the future, for IVUS to become a widely accepted surrogate, there needs to be a clear correlation between plaque volume by IVUS and cardiovascular events. If this correlation can be achieved, IVUS may provide the best tool for evaluating antiatherosclerotic therapies.
DISCUSSION Michael Cressman: How did you figure in the potential for heterogeneity of a response of atherosclerotic lesions into sample size calculations for these studies? Steven Nissen, MD (Cleveland, Ohio): We picked
total volume as the primary endpoint, but prespecified a series of secondary endpoints. For example, we take rolling 10-slice segments, then take the segments with the least atherosclerotic plaque, and that represents a defined endpoint. We take the 10 slices with the greatest atherosclerotic plaque, and that represents another endpoint. Then you can ask, is the primary effect of intensive low-density lipoprotein lowering with a statin to retard the progression of the early lesion or to regress the advanced lesion? Then you can define within the database. We get 30,000 slices (50 slices per patient in 600 patients), and each is potentially an independent variable. It’s not really, but it may be that the very sonolucent and soft plaques are the ones where the greatest changes occur. Because of the methodology and the way we define the secondary endpoints, we’ll be able to figure that out from our data. Michael H. Davidson, MD (Chicago, Illinois): Any evidence that change in IVUS plaque volume correlates with change in cardiovascular events? Dr. Nissen: We don’t have such a trial yet, but we will. There is another trial we’re doing (a 3-arm trial) for conventional endpoints, so we wrapped an endpoint trial around an IVUS volumetric trial. So, if carotid ultrasound can predict coronary events, then certainly coronary ultrasound should be able to predict coronary events. Rem Bolton, PhD (Charlottesville, Virginia): What is fascinating about your data is that it looks almost identical to the Brunich studies that have been done on the carotid artery with transcutaneous ultrasound where they were able to show that the lesions, at first, remodeled and got larger, preserving the lumen, and then above a certain level of plaque they started to shrink and compromise the lumen. The principal appears to be the same in the carotid and the coronary artery. David Herrington, MD (Winston-Salem, North Carolina): That disparity between clinical and angiographic progression slide shows relative change in events compared with absolute change in a metric
A SYMPOSIUM: SURROGATE MARKERS OF ATHEROSCLEROSIS
19A
whose units happen to be percentage. If you look at relative change in events compared with relative change in progression of coronary disease, there is absolutely no discontinuity whatsoever. There is 50 – 70% relative reduction in events and there is 45–75% relative reduction in the rate of progression of disease measured by angiography. So, to promote the notion that there is a big disconnection between progression measured angiographically and progression or regression clinically I think does a disservice to the interpretation of the large body of data that shows that altering progression of coronary disease measured angiographically does indeed predict changes in risk for future events—a body of data for which there is not yet comparable evidence with respect to IVUS. I would not want to be misperceived as being unenthusiastic about IVUS, but to promote IVUS at the expense of the very valuable information that does reside with angiography is a disservice. Dr. Nissen: I disagree but will reserve my comments for further discussion. Karen Friday, MD (New Orleans, Louisiana): Based on your presentation, the great majority of these patients should be dead, which is not the case. We’ve known for a long time that young people do have plaques, but it takes more than a plaque to produce an event. Dr. Nissen: Well, a large number of these people should be dead, and in fact, approximately half the people in this room will die from the disease. That is the point. Yes, it’s true there is a long asymptomatic lead time, but when a man has an MI at age 45, the disease did not start when he was 43, it started when
20A THE AMERICAN JOURNAL OF CARDIOLOGY姞
he was 17. I think we understand why it takes time, but the fact is that you have the disease and I have the disease, and approximately half the people in this room will die of the disease. We need to identify pharmacologic strategies for treating people earlier and more aggressively before they have their first event. The problem is that we don’t get second chances, because with 60 –70% of coronary disease, the very first symptom is either sudden death or acute MI. I think there is a lot of room for the pharmaceutical industry to develop better drugs to be used more aggressively at an earlier age, if we can do so safely. 1. Little WC, Constantinescu M, Applegate RJ, Kutcher MA, Burrows MT, Kahl
FR, Santamore WP. Can coronary angiography predict the site of a subsequent myocardial infarction in patients with mild-to-moderate coronary artery disease? Circulation 1988;78:1157–1166. 2. Nissen SE, Gurley JC, Grines CL, Booth DC, McClure R, Berk M, Fischer C, DeMaria AN. Intravascular ultrasound assessment of lumen size and wall morphology in normal subjects and patients with coronary artery disease. Circulation 1991;84:1087–1199. 3. Glagov S, Weisneberg E, Zarins CK, Stankunavicius R, Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 1987;316:1371–1375. 4. Topol EJ, Nissen SE. Our preoccupation with coronary luminology: the dissociation between clinical and angiographic findings in ischemic heart disease. Circulation 1995;92:2333–2342. 5. Tuzcu EM, Hobbs RE, Rincon G, Bott-Silverman C, De Franco AC, Robinson K, McCarthy PM, Stewart RW, Guyer S, Nissen SE. Occult and frequent transmission of atherosclerotic coronary disease with cardiac transplantation: insights from intravascular ultrasound. Circulation 1995;91:1706 –1713. 6. Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation 1995;92: 657– 671. 7. Schoenhagen P, Ziada KM, Kapadia SR, Crowe TD, Nissen SE, Tuzcu EM. Extent and direction of arterial remodeling in stable versus unstable coronary syndromes: an intravascular ultrasound study. Circulation 2000;101:598 – 603. 8. Brown BG, Zhao XQ, Sacco DE, Albers JJ. Lipid lowering and plaque regression. New insights into prevention of plaque disruption and clinical events in coronary disease. Circulation 1993;87:1781–1791.
VOL. 87 (4A)
FEBRUARY 16, 2001