Our Preoccupation With Coronary Luminology
The Dissociation Between Clinical and Angiographic Findings in Ischemic Heart Disease
Abstract Nearly 40 years after its invention, the angiogram is still considered by most physicians to be the “gold standard” for defining coronary anatomy. Careful investigations have revealed many deficiencies inherent in this approach. The purpose of this article is to outline the evidence that our current preoccupation with coronary “luminology” may be misguided and to propose a rational paradigm for future practice and investigation. Angiography depicts coronary anatomy from a planar two-dimensional silhouette of the lumen. Angiography is limited in resolution to four or five line pairs per millimeter. Confounding factors include vessel tortuosity, overlap of structures, and the effects of lumen shape. After intervention, a hazy, broadened silhouette may overestimate the actual gain in lumen size. Studies show marked disparity between the apparent severity of lesions and their physiological effects. After myocardial infarction, cardiologists too often do not make an attempt to demonstrate the physiological significance of the stenosis before performing percutaneous coronary revascularization. Similarly, the allure of a better, more gratifying angiogram with new interventional devices appears to be a dominant factor in their popularity. Interventional cardiologists should be aware that techniques yielding marked angiographic benefit may also generate important but unrecognized hazards. The dissociation between the angiogram and clinical outcome should influence future research efforts. Our review of the literature indicates that we may benefit from shifting the current focus and preoccupation with coronary luminology to achieving the desired clinical end point: promoting survival and long-term freedom from myocardial infarction and the disabling symptoms of coronary heart disease.
In 1957, Sones performed the first selective coronary angiogram, offering the first opportunity to visualize atherosclerotic coronary disease in vivo and heralding a new subspecialty of cardiovascular medicine.1 Today, nearly 40 years later, the angiogram is still considered the “gold standard” for definition of coronary anatomy. Cardiologists have increasingly relied on coronary angiography to guide both clinical practice and research, with more than a million procedures now performed annually in the United States.2 In recent years, many research studies, particularly in thrombolysis and intervention, have used computerized measurement techniques, a process known as quantitative coronary angiography. This practice has created an entirely new jargon with terms such as minimal luminal diameter, acute gain, late loss, and relative gain.3 4 5 Such indexes are now universally used to appraise new revascularization devices and evaluate treatments aimed at limiting restenosis.
Coronary angiography has had a profound impact on the diagnosis and management of ischemic heart disease, setting up the potential for both surgical and percutaneous coronary revascularization and the foundation for contemporary myocardial reperfusion therapy. The enormity of the significance of coronary angiography and how it has completely revamped cardiovascular medicine in the past four decades cannot be adequately emphasized. Although the value of coronary angiography remains unquestioned, radiographic imaging depicts the coronary artery as a simple two-dimensional projection of the lumen. Unfortunately, the silhouette or “luminogram” is a relatively poor representation of coronary anatomy and a limited standard on which to base therapeutic decisions. The purpose of this article is to outline the evidence that our current preoccupation with coronary luminology may be significantly misguided and to propose a rational paradigm for future clinical practice and investigation.
Limitations of Angiography
As early as the 1960s, investigators began questioning the accuracy and reproducibility of coronary angiography.1 6 7 8 9 10 11 12 13 Initial studies established that visual interpretation exhibited clinically significant intraobserver and interobserver variability, with differences in the estimation of stenosis severity approaching 50%.6 7 Comparative studies reported major discrepancies between the apparent angiographic severity of lesions and postmortem histology,8 9 10 11 12 13 with angiography significantly underestimating the extent of atherosclerosis. More recently, using functional testing, investigators have focused on the discordance between angiographic lesion severity and the physiological effects of the stenosis.14 15
Observer Variability and Comparisons With Histology
Angiography depicts intricate coronary cross-sectional anatomy from a planar two-dimensional silhouette of the contrast-filled vessel lumen. However, both necropsy studies and intravascular ultrasonography demonstrate that coronary lesions are often complex, with markedly distorted or eccentric luminal shapes. For a complicated coronary lesion, any arbitrary angle of view may significantly misrepresent the extent of narrowing (Fig 1A⇓). Theoretically, two orthogonal angiograms should accurately reflect the severity of most lesions. However, adequate orthogonal views are frequently unobtainable because of foreshortening, overlapping side branches, or disease at bifurcation sites. Even when unlimited projections are available, the angiographic silhouette cannot accurately depict certain complex luminal shapes (Fig 1B⇓).
After intervention, the limitations of projection imaging are particularly problematic. Necropsy and intravascular ultrasound studies demonstrate that mechanical interventions exaggerate the extent of luminal eccentricity by fracturing or dissecting the atheroma.16 The angiographic appearance of the complex postinterventional vessel often consists of an enlarged, although frequently “hazy,” lumen. After extensive plaque fracture, the hazy, broadened angiographic silhouette may overestimate the true gain in lumen size (Fig 2⇓). It is more difficult for a computer program to accurately measure the angiographic dimensions of the hazy, dissected luminogram to calculate the residual diameter. Yet, this is precisely how most clinical studies in the 1990s assess the efficacy of new interventional devices.
The traditional method for characterizing angiographic lesion severity relies on measurement of the percent stenosis. This process requires comparison of dimensions within both the lesion and an adjacent, uninvolved “normal” reference segment. However, necropsy studies demonstrate that coronary disease is frequently diffuse and contains no truly normal segment.9 10 12 In the presence of diffuse disease, calculation of the angiographic percent stenosis will predictably underestimate disease severity (Fig 3⇓). In extreme circumstances, diffuse, concentric, and symmetrical disease involving the entire vessel will result in the angiographic appearance of a small but normal artery.
Angiography is often confounded by the phenomenon of coronary “remodeling,” first described in 1987 by Glagov et al17 (Fig 4⇓). The remodeling process is observed histologically as the outward displacement of the external vessel wall overlying the atheroma. The adventitial enlargement opposes luminal encroachment, thereby concealing the presence of disease. Although remodeled lesions do not restrict blood flow, clinical studies have demonstrated that these low-grade lesions represent the most important source for acute coronary syndromes.18 Recently, we reported that such atheromas are virtually always present in ergonovine-positive patients with a “normal” angiogram.19
Physical Limits of Angiography
The resolution of modern angiographic equipment is surprisingly modest.20 The best image intensifiers can resolve only about four or five line pairs per millimeter, often somewhat less with aging equipment. Prudence limits radiation doses to about 25 μR per frame, exacerbating an image flaw known as “quantum statistical noise.” Quantum noise is an unyielding source of image degradation that can be attenuated only by increasing radiation doses to unacceptable levels. Because of these limits, structural features smaller than about 0.2 mm are invisible to the angiographer. Notably, important features fall within this size limit, including intracoronary thrombi and small focal calcifications.
Angiographic resolution is further degraded by rapid coronary artery motion.20 The velocity of the right coronary artery can reach 50 mm/s. Although pulse-mode radiation “freezes” coronary motion, the shortness of each pulse is limited by thermal loading of the x-ray tube and the need for a small “focal spot.” Adequate exposure in a large patient typically yields pulse durations of 4 to 7 ms, which may rise to 10 ms for compound angles. A coronary artery moving at 50 mm/s will produce a motion blur of 0.50 mm during a 10-ms pulse width. This indistinct border confounds the usual measures of stenosis severity.
Complexity of Coronary Anatomy
Coronary arteries characteristically follow a serpiginous path with complex curvatures, sharp bends, and unpredictable twists and turns. Angiography depicts lesions in tortuous vessels correctly only when the x-ray beam is orthogonal to the vessel. Optimal angiography requires two “triple orthogonal views,” defined as a pair of projections orthogonal both to each other and to the vessel.21 However, the angiographer possesses no certain means to achieve orthogonality. Accordingly, current practice typically consists of imaging at a variety of angles to find the optimal projection: one that yields the “minimum” diameter. Unfortunately, this empirical process is limited by factors including toxicity of contrast agents, radiation exposure, and operator patience. Experienced angiographers are aware that, despite meticulous imaging in multiple views, some lesions may be evident only in a few frames.
The left main coronary artery (LMCA) represents a particularly troublesome site for angiographic imaging.11 22 23 The LMCA is often short and overlaps other structures, providing a limited opportunity to identify a normal reference segment (Fig 5⇓). Detection of ostial LMCA disease is confounded by the tendency for catheters to “seat” beyond the lesion, requiring reflux of contrast into the aorta for adequate visualization. Unfortunately, refluxing contrast fills the aortic cusp, which may obscure the lesion. The distal LMCA is often obscured by the confluence of shadows from the overlapping left anterior descending, left circumflex, and ramus medianus arteries. Other coronary bifurcations present comparable difficulties, because superimposition of the parent and daughter often partially obscures the bifurcation (Fig 6⇓). Almost diabolically, atherosclerosis exhibits a predilection for bifurcation sites. Thus, the segments most difficult to image are often the sites with the most significant disease.
Problem of Coronary Flow Reserve
In chronic ischemic coronary disease, symptoms result principally from the ability of stenoses to blunt increases in blood flow in response to metabolic demands. This phenomenon, originally described by Gould et al in the 1970s,24 is commonly called coronary flow reserve (CFR).24 25 Determination of CFR requires measurement of blood flow at rest and after induction of reactive hyperemia, usually by administration of a coronary vasodilator. Animal and human studies have documented that a normal CFR, in the absence of epicardial stenoses, should exceed ≈5:1 (ratio of hyperemic to basal flow). Several methods for measurement of CFR in patients were developed during the 1980s and 1990s, including intracoronary Doppler flow probes, digital angiography, and quantitative positron emission tomography.14 15 25 26 Each of these approaches has documented major discrepancies between the apparent angiographic severity of coronary lesions and their physiological effects. Animal studies demonstrate that flow reserve remains normal (typically a 5:1 to 7:1 ratio) until the stenosis severity approaches 75%. Between 75% and 95%, CFR falls progressively to reach values approaching a 1:1 ratio. Accordingly, the angiographic differences between moderate and severe lesions may be only a few tenths of a millimeter. Such differences are difficult to discern, given the limitations in angiographic resolution, the confounding effects of projection angles, the irregularity of luminal shape, and the impact of diffuse disease. Other factors further weaken the correlation between angiography and flow reserve, including the presence of ventricular hypertrophy, the metabolic state of the myocardium, and microvascular disease. Thus, the epicardial stenosis represents only one factor responsible for a reduction in flow reserve in patients with clinical symptoms. Accordingly, a stenosis incapable of producing angina in one patient may result in severe functional limitation in another.
Quantitative Angiography: The Emperor Without Clothes
Originally described by Brown et al27 in the late 1970s, quantitative coronary angiography (QCA) was intended to replace the “eyeball” interpretation of the angiogram with an objective, computer-based method. Proponents believed that the failings of the angiogram originated principally from arbitrary, visual inspection. Initial studies demonstrated that computerized determination of vessel borders was highly reproducible, enabling serial measurements of angiograms without significant intraobserver or interobserver variability. Accordingly, QCA gradually became the standard for evaluating the effects of thrombolysis, new interventional devices, and regression or progression of atherosclerosis. A whole new industry developed, fueled by clinical trials, in which self-described “core laboratories” measured luminograms at $100 to $400 per patient and reported the luminal size in hundredths of a millimeter.
Eventually, the proponents of QCA confused reproducibility with accuracy. In reality, QCA possesses virtually all of the limitations of conventional interpretation methods. No matter how precisely measured, the luminogram of a complex, cloverleaf–shaped lesion or the silhouette of dissected, “hazy” atheroma poorly represents the actual lumen size (Figs 1⇑ and 2⇑). Intraluminal filling defects from thrombus (Fig 7⇓, left) or the absence of a disease-free reference segment (Fig 7⇓, right) confound the analysis. QCA cannot distinguish the relative narrowing of a diffusely diseased vessel or accurately quantify LMCA or bifurcation lesions with overlapping contrast-containing structures (Figs 3 through 6⇑⇑⇑⇑).
Published studies showing the excellent reproducibility of QCA bear little resemblance to the clinical situation.28 Generally, these validation studies presented the computer operator with carefully chosen, optimally opacified frames from preselected projections. The performance of QCA in clinical practice is confounded by a staggering array of variables. A recent analysis by Keane et al29 of 10 core QCA laboratories showed marked variability between laboratories for the straightforward quantification of phantom stenoses. Previously, we reported that when observers were permitted to select from available projections and frames, the variability of computer-analyzed angiography approached that of visual inspection.28 Many experienced angiographers believe that visual assessment of stenoses should always be performed on moving images to integrate the interpretation over many frames. No quantitative angiographic program can perform this function.
The described theoretical and practical limitations of angiography have been confirmed through clinical trials of thrombolysis, angioplasty devices, and restenosis. Studies demonstrate that angiography cannot predict which patients will require angioplasty after thrombolysis,30 31 32 33 34 35 36 cannot reliably identify the patients who benefit from new interventional devices, and cannot identify those patients likely to undergo plaque rupture. These studies must remind us of the hazards of formulating clinical decisions principally or solely on the basis of the angiographic appearance of lesions.
Acute Myocardial Infarction: The Oculostenotic Reflex
The first randomized trials in interventional cardiology were performed in the setting of acute myocardial infarction. The Thrombolysis and Angioplasty in Myocardial Infarction (TAMI-1) trial, published in 1987, demonstrated that angioplasty was unnecessary and potentially deleterious after successful thrombolysis.30 Subsequent trials corroborated this finding, confirming that the severity of the residual stenosis after thrombolysis poorly predicts recurrent ischemia or reocclusion.31 32 33 34 35 36 A vivid demonstration of this dissociation occurs with the “no-reflow” phenomenon, in which the epicardial lumen appears adequate but contrast echocardiography shows no tissue perfusion.37 Accordingly, before the residual stenosis in an infarct vessel is addressed, there should be demonstration of either spontaneous or provocable signs of ischemia.30 31 32 33 34 35 36 However, nearly a decade later, this finding has not been fully integrated into clinical practice.
The term “oculostenotic” reflex was coined to denote what appears to be an irresistible temptation among some invasive cardiologists to perform angioplasty on any significant residual stenosis after thrombolysis.38 Although this approach is not supported by the American College of Cardiology and American Heart Association guidelines,39 the ritual of reflex angioplasty is exercised thousands of times each year. Systematic functional assessment in patients after successful reperfusion suggests that fewer than half will have objective evidence of provocable ischemia.40 Importantly, a randomized trial of patients with a negative functional test but a significant infarct vessel stenosis demonstrated worse outcome in the group assigned to angioplasty.41
However, despite data to contrary, it remains the overriding belief of some invasive cardiologists that the residual stenosis must be alleviated to favorably affect the prognosis. Recent data from the GUSTO trial (United States patients) and a National Registry of Myocardial Infarction indicate that more than 40% of patients undergo angioplasty after receiving thrombolytic therapy.42 43 Unfortunately, in a large number of these patients, functional assessment is not performed. A large insurance claims database revealed that, in the late 1980s, only 9% of patients with recent myocardial infarction underwent exercise testing before coronary intervention.44
Directional Atherectomy: Angiographic Gratification
Introduced in the United States in late 1990, directional coronary atherectomy (DCA) generated considerable excitement for its potential to remove (debulk) coronary atheromata.45 Registry data, which led to commercial approval, supported a high procedural success rate, with a complication rate similar to that of historical balloon angioplasty controls.46 DCA quickly became popular, constituting nearly 15% of the procedures performed in the United States in 1992. However, in 1993, the results of the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT) yielded surprising results.47 Compared with balloon angioplasty, atherectomy induced more significant complications, particularly non–Q-wave infarction (twofold risk) and abrupt vessel closure. At 6 months, there were higher rates of death and nonfatal myocardial infarction, and at 1 year, a persistently higher mortality rate in patients assigned to atherectomy.47 48 There was no reduction in repeat revascularization and minimal evidence of reduced angiographic restenosis.
Taken as a whole, these findings provide little support for widespread application of directional atherectomy. However, in the United States, the overall use of directional atherectomy was 15% of percutaneous coronary revascularization procedures before this trial and remained 15% in 1994. Thus, negative data from a large, multicenter effort, conducted by experienced operators, had little to no effect on the frequency of use of atherectomy. Although other explanations are possible, the dominant factor sustaining the use of this procedure appears to be “angiographic gratification,” the allure of a better, more gratifying angiographic image after atherectomy. Early in the development of directional atherectomy, studies showed that debulking of plaque resulted in improved luminal caliber compared with balloon dilatation.49 This difference in lumen size between atherectomy and angioplasty was validated in the CAVEAT and Canadian Coronary Atherectomy Trial (CCAT).47 50
On the basis of lumen size data in atherectomy patients and related findings from nonrandomized balloon angioplasty registries, the “bigger is better” paradigm was introduced. This concept refers to the theory that a larger luminogram attained by intervention will invariably result in better angiographic and clinical outcomes.4 5 However, for DCA, the bigger is better hypothesis remains unproven, and a randomized trial of balloon versus optimal atherectomy (BOAT) is currently in progress. Nevertheless, in CAVEAT, the angiographic outcome at 6 months was marginally improved, but paradoxically, death and non–Q-wave myocardial infarction were increased in patients with a larger lumen achieved via atherectomy.47 Recently, proponents of a bigger luminogram have focused so narrowly on lumen size that clinical end points may be disregarded. This phenomenon is exemplified by a recent matched-pair study of patients undergoing atherectomy or angioplasty in which the authors failed to report on any clinical outcomes.51 Interestingly, the inattention to atherectomy-related complications appears confined to the United States; outside this country, fewer than 3% of patients having percutaneous interventions undergo directional atherectomy.
Stents: A Critical Reappraisal
In contrast to atherectomy, two randomized comparisons with balloon angioplasty demonstrated that the Palmaz-Schatz coronary stent reduced angiographic restenosis.52 53 Like DCA, the cardinal feature stimulating the growth of stenting is greater improvement in the coronary luminogram. Compared with balloon angioplasty, patients randomized to stenting had a 30% greater improvement in initial lumen caliber, although there was attrition of the benefit over time. Unlike atherectomy, angiographic gratification, in the case of stenting, appears to be associated with enhanced durability and reduced recurrence rates. However, the margin of benefit appeared to narrow after a protocol-mandated angiogram after 6 months.52 53 54 Stents were commercially approved for elective coronary revascularization in August 1994, and their use has catapulted rapidly, with estimates that by 1996, nearly 50% of interventions will involve stenting rather than balloon angioplasty.53A
Superficially, these trials support the concept that bigger is better, but disquieting concerns remain. At 6 months, luminal diameter showed a relatively small, and in one trial statistically insignificant, advantage for stenting.52 The stent trials reported increased complications and costs, thus exchanging a slightly more gratifying luminogram for potentially serious immediate risks of hemorrhage and prolonged hospitalization.52 53 54 To avoid intensive anticoagulation, research is presently directed toward optimization of deployment via high-pressure inflations, ultrasound guidance, or both.55 56 57 However, the long-term risks of a permanent endovascular prosthesis remain unknown in patients with de novo lesions. Of potentially greater concern, the desire to optimize the lumen size has increased application of stenting in patient groups not yet fully studied or approved by the Food and Drug Administration. These include patients with restenosis or saphenous vein graft disease. While coronary stenting for a variety of patient and lesion subsets may indeed prove beneficial, it remains unwarranted to extrapolate excessively from short-term angiographic studies in patients with new native-vessel lesions.
Critical review of pharmacological and device trials to reduce restenosis reveals marked differences between the angiographic and clinical results (Table 1⇓).47 52 53 58 59 60 61 In CAVEAT,47 compared with balloon angioplasty, atherectomy patients showed the same extent of diameter improvement as produced by stenting in the BENESTENT trial. However, only the latter trial demonstrated a significant difference in clinical outcome. In a study of nitric oxide donors,58 angiographic measurements showed less restenosis at 6-month follow-up but no change in the need for repeat revascularization. A trial of angiopeptin,59 a somatostatin-like growth factor inhibitor, showed no measurable angiographic benefit but significant improvement in clinical results. Directional atherectomy in saphenous vein graft disease61 produced angiographic benefits similar to those of CAVEAT but a significant reduction of repeat revascularization at 6 months.47 61 These trials dramatically exemplify the lack of correlation between the luminogram and clinical results.
Atherosclerosis Regression and Plaque Rupture
Atherosclerosis regression studies provide a salient example of the dissociation between angiography and clinical outcomes. Nine multicenter, randomized lipid-lowering trials using both angiographic and clinical assessment showed a negligible improvement of luminal caliber, typically an absolute difference of only 1% to 3%. Yet these same studies yielded a staggering 25% to 75% reduction in acute events, including myocardial infarction.62 63 64 Thus, the differences between angiographic and clinical end points exceed an order of magnitude. Although in the 1960s and 1970s most authorities believed that the tightest stenoses would most likely progress to occlusion, subsequent clinical studies have established that acute syndromes are most commonly produced by rupture of minor plaques.65 66 67 Thus, it appears that the benefits of lipid-lowering therapy are derived by stabilization of these low-grade, lipid-rich plaques, not changes in angiographic lumen size.
Mechanisms and Implications
What are the explanations for the apparent dissociation between angiographic and clinical outcomes in thrombolytic and interventional trials? A full understanding remains uncertain, but several factors appear important, including the requirement for repeat coronary angiography in asymptomatic patients, potentially triggering an oculostenotic reflex. Other factors include injury to the microcirculation or myocardial damage, which may reduce the clinical significance of a flow-limiting stenosis supplying an infarcted segment. Alternatively, partial or complete denervation may suppress symptoms.68
Insights From Intravascular Ultrasound
Coronary intravascular ultrasound provides valuable insights into the mechanisms underlying the dissociation between angiographic and clinical outcomes. Studies have compared ultrasound and angiographic dimensions in vivo, examining truly normal vessels, arteries with angiographically occult disease, and coronary lesions before and after intervention.69 These comparisons show that the two imaging modalities correlate closely in undiseased vessels with a nearly circular lumen shape. However, as the lumen becomes progressively more irregular, the correlation between a silhouette imaging method (angiography) and a tomographic modality (ultrasound) diverges significantly.70 71 These differences are most profound in imaging of postinterventional arteries with a dissected atheroma or otherwise distorted luminal shape. In such vessels, the reported correlation between angiography and ultrasound lumen size is r=.30.72 73 Fundamentally, angiography cannot accurately depict the true size of the complex luminal shapes commonly encountered after interventions.
The disparity between angiographic and ultrasound dimensions after interventions creates important paradoxes. A “suboptimal” angiographic silhouette may result from an intervention that yields a relatively round and smooth but moderate-sized lumen (Fig 8⇓, left). However, this modest lumen may have a cross-sectional area that is actually larger than a vessel with a “better” angiographic result in which the lumen gain was derived from complex cracks or splits in the atheroma (Fig 8⇓, right). A recent analysis of patients undergoing balloon angioplasty, directional atherectomy, or rotational ablation confirmed this phenomenon.74 The disparity between angiographic and ultrasound lumen sizes varied considerably for different interventional devices and paralleled the degree of irregularity of lumen shape. Importantly, a complex or irregular lumen will produce greater flow resistance than a circular lumen, which may explain why some patients have continued symptoms despite an optimal angiographic result (Fig 8⇓).
Other phenomena observed by intravascular ultrasound may help explain the results of several of the cited clinical trials. The lumen shape by ultrasound after directional atherectomy is surprisingly irregular, with deep cuts or grooves in the plaque or fronds of material extending into the lumen (Fig 9⇓).75 76 These surface irregularities are invisible to the angiographer but may facilitate thrombus formation or embolization, which may explain the high incidence of acute events after DCA. Ultrasound also shows many casualties of the dogmatic “bigger is better” philosophy. Fig 10⇓ shows two examples of deep and extensive injury arising from overly aggressive attempts to enlarge the lumen with directional atherectomy. Although unproven, it is possible that such severely injured arteries will be maximally stimulated to develop severe neointimal hyperplasia.
Intravascular ultrasound provides a useful perspective on the inability of angiography to predict untoward clinical events. Ultrasound commonly reveals atherosclerosis at coronary sites without any disease apparent by angiography.71 Because low-grade plaques are clearly implicated in acute coronary events, the extent of unrecognized atherosclerosis may determine the prognosis, not the degree of enlargement on the luminogram at the interventional site.
Laboratory Surrogates in Clinical Cardiology
On the basis of the findings of clinical trials and intravascular ultrasound, we can now add angiography to the list of diagnostic procedures in which surrogate end points have failed to predict clinical outcome (Table 2⇓). Examples include suppression of premature ventricular contractions with antiarrhythmic agents,77 improving the hemodynamics in heart failure to promote longevity,78 and using the activated partial thromboplastin time to ensure optimal anticoagulation in thrombolysis.79 In each of these cases, use of a laboratory surrogate seemed intuitively reasonable but was subsequently proved by randomized trials to be incorrect and dangerous. In each case, the false impressions arising from overinterpretation of a laboratory test or measurement generated a therapeutic backfire, in which mortality was paradoxically increased. These observations emphasize the importance of avoiding the precocious reliance on any untested laboratory parameter as a true surrogate for clinical outcome. For example, if a larger coronary lumen after stenting correlates with improved long-term clinical outcomes, this finding does not necessarily translate to rotational ablation or directional atherectomy.
Effects on Clinical Investigation
The dissociation between the angiogram and clinical outcome should influence future research efforts. Clinical trials must be large enough to retain the statistical power to differentiate clinical outcomes, not only a surrogate end point such as angiographic lumen size. Ideally, important clinical trials should incorporate a mechanistic substudy to determine the linkage between clinical and laboratory end points. This approach aided the Global Utilization of Streptokinase and Tissue Plasminogen Activator in Occluded Coronary Arteries (GUSTO-I) trial, in which an angiographic substudy in 2431 of the 41 021 enrolled patients demonstrated the pivotal relationship between the 90-minute patency and 30-day mortality.80 81 82 Similarly, restenosis trials currently evaluating platelet glycoprotein IIb/IIIa inhibitors will include a 6-month angiogram in 900 of the 4000 to 5000 patients.83 The design of these substudies preempts the possibility that follow-up angiography per se will drive repeat revascularization procedures but permits objective validation of angiographic indexes of the clinical outcomes.
In addition to closely tracking clinical end points, future trials of restenosis and new interventional devices should not rely exclusively on measurements obtained from an angiographic silhouette.84 85 Intravascular ultrasound dimensions are frequently divergent from quantitative angiographic measurements, whereas the tomographic perspective of ultrasound provides unique insights into the mechanism, shape, and extent of lumen enlargement. Ultrasound has important limitations, too, including its inability to characterize thrombus, artifactual distension of the lumen via the catheter, lack of accessibility for distal lesions and tortuous vessels, nonuniform rotational distortion, nonorthogonal positioning, and cost. Nevertheless, future trials of restenosis and new revascularization devices should consider the use of ultrasound or a suitable alternative that provides full interrogation of the diseased vessel wall. Similarly, trials of atherosclerosis regression should include assessment of the vessel wall to identify not just the magnitude but also the mechanism by which the tested agent reduces the frequency of clinical events.84 85
Implications for Clinical Practice
Examples demonstrate that miscues can arise when clinicians and investigators rely excessively on angiography for clinical decision-making. How should we incorporate this knowledge into clinical practice? Procedures should not be performed solely to improve the luminal appearance—so-called coronary “cosmetology.” Operators should exercise restraint in selecting specific devices because they provide a more gratifying angiogram, particularly if there is no clear documentation of a clinical advantage. Interventional cardiologists need to be aware that techniques yielding marked angiographic benefit may also generate important but initially unrecognized hazards. This appreciation of the limitations and boundaries of angiographic data may help switch our focus from the current preoccupation with coronary luminology to that of achieving the desired clinical end point promoting survival and long-term freedom from myocardial infarction and the disabling symptoms of ischemic heart disease.
- Received February 13, 1995.
- Revision received April 19, 1995.
- Accepted May 10, 1995.
- Copyright © 1995 by American Heart Association
Graves EJ. Detailed diagnoses and procedures: National Hospital Survey 1990. Washington, DC: National Center for Health Statistics Vital Health Stat 13(113); 1992.
Rensing BJ, Hermans WRM, Deckers JW, de Feyter PJ, Tijssen JGP, Serruys PW. Lumen narrowing after percutaneous transluminal coronary balloon angioplasty follows a near gaussian distribution: a quantitative angiographic study in 1,445 successfully dilated lesions. J Am Coll Cardiol. 1992;19:939-945.
Kuntz RE, Safian RD, Carrozza JP, Fishman RF, Mansour M, Baim DS. The importance of acute luminal diameter in determining restenosis after coronary atherectomy or stenting. Circulation. 1992;86:1827-1835.
Zir LM, Miller SW, Dinsmore RE, Gilbert JP, Harthorne JW. Interobserver variability in coronary angiography. Circulation. 1976;53:627-632.
Galbraith JE, Murphy ML, Desoyza N. Coronary angiogram interpretation: interobserver variability. JAMA. 1981;240:2053-2059.
Arnett EN, Isner JM, Redwood CR, Kent KM, Baker WP, Ackerstein H, Roberts WC. Coronary artery narrowing in coronary heart disease: comparison of cineangiographic and necropsy findings. Ann Intern Med. 1979;91:350-356.
Grodin CM, Dyrda I, Pasternac A, Campeu L, Bourassa MG. Discrepancies between cineangiographic and post-mortem findings in patients with coronary artery disease and recent myocardial revascularization. Circulation. 1974;49:703-709.
Isner JM, Kishel J, Kent KM. Accuracy of angiographic determination of left main coronary arterial narrowing. Circulation. 1981;63:1056-1061.
Vlaodaver Z, French R, van Tassel RA, Edwards JE. Correlation of the antemortem coronary angiogram and the postmortem specimen. Circulation. 1973;47:162-168.
Little WC, Constantinescu M, Applegate RJ, Kutcher MA, Burrows MT, Kahl FR, Santamore WP. Can arteriography predict the site of a subsequent myocardial infarction in patients with mild-to-moderate coronary artery disease? Circulation. 1988;78:1157-1166.
Nissen SE. Radiographic principles in cardiac catheterization. In: Roubin GS, Califf RM, O’Neill W, Phillips H, Stack R, eds. Interventional Cardiac Catheterization: Principles and Practice. New York, NY: Churchill Livingstone Inc; 1993:409-425.
Nissen SE, Tuzcu EM, DeFranco AC, Moliterno D. Evaluation of stenosis severity: quantitative angiography, coronary flow reserve, and intravascular ultrasound. In: Ellis SG, Holmes D, eds. Strategic Approaches to Coronary Intervention. Baltimore, MD: Williams & Wilkins; 1995. In press.
Elliot J, Tuzcu EM, Guyer S, Lefkovits J, DeFranco AC, Nissen SE. Relationship of left main coronary dimension to branch vessel size: a new index for detection of left main coronary artery disease. J Am Coll Cardiol. 1994;23:210A. Abstract.
DeFranco AC, Tuzcu M, Eaton G, Raymond RE, Franco I, Guyer A, Nissen SE. Detection of unrecognized LMCA disease by intravascular ultrasound in patients undergoing interventions: prevalence and severity. Circulation. 1993;88(pt 2):I-411. Abstract.
Vogel RA, LeFree MT, Bates ER, O’Neil WW, Foster R, Kirklin P, Smith D, Pitt B. Application of digital techniques to selective coronary arteriography: use of myocardial appearance time to measure coronary flow reserve. Am Heart J. 1983;107:153-164.
Nissen SE, Elion JL, Booth DC. Value and limitations of computer analysis of digital subtraction angiography in the assessment of coronary flow reserve. Circulation. 1986;73:562-571.
Brown BG, Bolson E, Frimer M, Dodge HT. Quantitative coronary arteriography: estimation of dimensions, hemodynamic resistance, and atheroma mass of coronary artery lesions using the arteriograms and digital computation. Circulation. 1977;55:329-337.
Keane D, Haase J, Slager CJ, van Swijndregt E, Lehmann KG, Ozaki Y, Di Mario C, Kirkeeide R, Serruys PW. Comparative validation of quantitative coronary angiography systems. Circulation. 1995;91:2174-2183.
Topol EJ, Califf RM, George BS, Kereiakes DJ, Abbottsmith CW, Candela RJ, Lee KL, Pitt B, Stack RS, O’Neill WW, and the Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) Study Group. A randomized trial of immediate versus delayed elective angioplasty after intravenous tissue plasminogen activator in acute myocardial infarction. N Engl J Med. 1987;317:581-588.
Simoons ML, Arnold AER, Betriu A, deBono DP, Col J, Dougherty FC, von Essen R, Lambertz H, Lubsen J, Meier B, Michel PL, Raynaud P, Rutsch W, Sanz GA, Schmidt W, Serruys PW, Thery C, Uebis R, Vahanian A, Van de Werf F, Willems GM, Wood D, Verstraete M. Thrombolysis with tissue plasminogen activator in acute myocardial infarction: no additional benefit from immediate percutaneous transluminal coronary angioplasty. Lancet. 1988;1:197-203.
SWIFT Trial Study Group. SWIFT trial of delayed elective intervention versus conservative treatment after thrombolysis with anistreplase in acute myocardial infarction. Br Med J. 1991;302:555-560.
Ellis SG, Topol EJ, George BS, Kereiakes DJ, Debowey D, Sigmon KN, Pickel A, Lee KL, Califf RM. Recurrent ischemia without warning: analysis of risk factors for in-hospital ischemic events following successful thrombolysis with intravenous tissue plasminogen activator. Circulation. 1989;80:1159-1165.
Reiner JS, Lundergan CF, van den Brand M, Boland J, Thompson MA, Machecourt J, Py A, Pilcher GS, Fink CA, Burton JR, Simoons ML, Califf RM, Topol EJ, Ross AM, for the GUSTO Angiographic Investigators. Early angiography cannot predict post-thrombotic coronary reocclusion: observations from the GUSTO Angiographic Study. J Am Coll Cardiol. 1994;24:1439-1444.
Ito H, Tomooka T, Sakai N, Yu H, Higashino Y, Fujii K, Masuyama T, Kitabatake A, Minamino T. Lack of myocardial perfusion immediately after successful thrombolysis: a predictor of poor recovery of left ventricular function in anterior myocardial infarction. Circulation. 1992;85:1699-1705.
Topol EJ. Coronary angioplasty for acute myocardial infarction. Ann Intern Med. 1988;109:970-980.
ACC/AHA Task Force Report. Guidelines for percutaneous transluminal coronary angioplasty: a report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (Subcommittee on Percutaneous Transluminal Coronary Angioplasty). J Am Coll Cardiol. 1988;12:529-545.
Sutton JM, Topol EJ. Significance of a negative exercise thallium test in the presence of a critical residual stenosis after thrombolysis for acute myocardial infarction. Circulation. 1991;83:1278-1286.
Ellis SG, Mooney MR, George BS, Ribeiro da Silva EE, Talley JD, Flanagan WH, Topol EJ. Randomized trial of late elective angioplasty versus conservative management for patients with residual stenoses after thrombolytic treatment of myocardial infarction. Circulation. 1992;86:1400-1406.
Mark DB, Naylor CD, Hlatky MA, Califf RM, Topol EJ, Granger CB, Knight JD, Nelson CL, Lee KL, Clapp-Channing N, Sutherland W, Pilote L, Armstrong P. Use of medical resources and quality of life after acute myocardial infarction in Canada and the United States. N Engl J Med. 1994;331:1130-1135.
Rogers WJ, Bowlby LJ, Chandra NC, French WJ, Gore JM, Lambrew CT, Rubison RM, Tiefenbrunn AJ, Weaver WD. Treatment of myocardial infarction in the United States (1990 to 1993): observations from the National Registry of Myocardial Infarction. Circulation. 1994;90:2103-2114.
Topol EJ, Ellis SG, Cosgrove DM, Bates ER, Muller DWM, Schork N, Schork MA, Loop FD. Analysis of coronary angioplasty practice in the United States using a private insurance database. Circulation. 1993;87:1489-1497.
Topol EJ, Leya F, Pinkerton CA, Whitlow PL, Hofling B, Simonton CA, Masden RR, Serruys PW, Leon MB, Williams DO, King B, Mark DB, Isner JM, Holmes DR, Ellis SG, Lee KL, Keeler G, Berdan LG, Hinohara T, Califf RM, on behalf of the CAVEAT Study Group. A comparison of coronary angioplasty with directional atherectomy in patients with coronary artery disease. N Engl J Med. 1993;329:221-227.
Elliott JM, Berdan LG, Holmes DR, Isner JM, King SB, Keeler GP, Kearney M, Califf RM, Topol EJ, for the CAVEAT Study Investigators. One-year follow-up in the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT-I). Circulation. 1995;91:2158-2166.
Adelman AG, Cohen EA, Kimball BP, Bonan R, Ricci DR, Webb JG, Laramee L, Barbeau G, Traboulsi M, Corbett BN, Schwartz L, Logan AG. A comparison of directional atherectomy with balloon angioplasty for lesions of the left anterior descending coronary artery. N Engl J Med. 1993;329:228-233.
Serruys PW, De Jaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, Emanuelsson H, Marco J, Legrand V, Materne P, Belardi J, Sigwart U, Colombo A, Goy JJ, van den Heuvel P, Delcan J, Morel MA, for the BENESTENT Study Group. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med. 1994;331:489-495.
Fischman DL, Leon MB, Baim DS, Schatz RA, Savage MP, Penn I, Detre K, Veltri L, Ricci D, Nobuyoshi M, Cleman M, Heuser R, Almond D, Teirstein PS, Fish D, Colombo A, Brinker J, Moses J, Shaknovich A, Hirshfeld J, Bailey S, Ellis S, Rake R, Goldberg S, for the Stent Restenosis Study Investigators. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med. 1994;331:496-501.
Lemaitre DT, Barber AP, Mullen MG, Luft H. Interventional cardiology. In: Industry Strategies. Boston, Mass: Cowen & Co; February 22, 1995.
Colombo A, Hall P, Martini G. Stenting with echo guidance but without anticoagulation. In: Current Review of Interventional Cardiology. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1995:115-133.
The ACCORD Study Investigators. Nitric oxide donors reduce restenosis after coronary angioplasty: the ACCORD Study. J Am Coll Cardiol. 1994;23:59A. Abstract.
Emanuelsson H, Beatt KJ, Bagger JP, Balcon R, Heikkila J, Piessens J, Schaeffer M, Suryapranata H, Feogh M, for the European Angiopeptin Study Group. Long-term effects of angiopeptin treatment in coronary angioplasty. Circulation. 1995;91:1689-1696.
Folland ED, Vogel RA, Hartigan P, Bates ER, Beauman GJ, Fortin T, Boucher C, Parisi AF, Veterans Affairs ACME Investigators. Relation between coronary artery stenosis assessed by visual, caliper, and computer methods and exercise capacity in patients with single-vessel coronary artery disease. Circulation. 1994;89:2005-2014.
Holmes DR, Topol EJ, Califf RM, Leya F, Berberg PB, Talley JD, Kellett MA, Shani J, Gottlieb RS, Whitlow PL, Adelman AG, Pinkerton CA, Lee KL, Pieper K, Keeler GP, Ellis SG, for the CAVEAT-II Investigators. A multicenter, randomized trial of coronary angioplasty versus directional atherectomy for patients with saphenous vein bypass graft lesions. Circulation. 1995;91:1966-1974.
Brown BG, Zhao XQ, Sacco DE, Albers JJ. Arteriographic view of treatment to achieve regression of coronary atherosclerosis and to prevent plaque disruption and clinical cardiovascular events. Br Heart J. 1993;69:S48-S53.
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;87:1781-1790.
Fuster V. Lewis A. Conner Memorial Lecture: mechanisms leading to myocardial infarction: insights from studies of vascular biology. Circulation. 1994;90:2126-2146.
Davies MJ, Krikler DM, Katz D. Atherosclerosis: inhibition or regression as therapeutic possibilities. Br Heart J. 1991;65:302-310.
Peterson MD, Abe T, Morgan DA, Gutterman DD. Role of adenosine in postischemic dysfunction of coronary innervation. Circ Res. 1995;76:76-95.
Nissen SE, Grines CL, Gurley JC, Sublett K, Haynie D, Diaz C, Booth DC, DeMaria AN. Application of a new phased-array ultrasound catheter in the assessment of vascular dimensions: in-vivo comparison to cineangiography. Circulation. 1990;82:660-666.
Nissen SE, Gurley CL. Assessment of coronary angioplasty results by intravascular ultrasound. In: Serruys PW, Strauss BH, King SB, eds. Restenosis After Intervention With New Mechanical Devices. Dordrecht, Netherlands: Kluwer Academic Publishers; 1992:73-96.
DeFranco AC, Tuzcu E, Abdelmeguid A, Lincoff AM, Brown D, Bernardi M, Ellis SG, Whitlow PL, Sutton JM, Casale PM, Raymond RE, Topol EJ, Nissen SE. Intravascular ultrasound assessment of PTCA results: insights into the mechanisms of balloon angioplasty. J Am Coll Cardiol. 1993;21:485A. Abstract.
Tobis JM, Mallery J, Mahon D, Lehmann K, Zalesky P, Griffith J, Gessert J, Moriuchi M, McRae M, Dwyer ML, Greep N, Henry WL. Intravascular ultrasound imaging of human coronary arteries in vivo. Circulation. 1991;83:913-926.
DeFranco AC, Tuzcu EM, Moliterno DJ, Elliott J, Berkalp B, Franco I, Raymond RE, Whitlow PL, Guyer S, Nissen SE. Overestimation of lumen size after coronary interventions: implications for randomized trials of new devices. Circulation. 1994;90(pt 2):I-550. Abstract.
Berkalp B, Nissen SE, DeFranco AC, Moliterno DJ, Franco I, Raymond RE, Sutton J, Guyer S, Tuzcu EM. Intravascular ultrasound demonstrates marked differences in lumen shape following interventional devices. Circulation. 1994;90(pt 2):I-58. Abstract.
Tuzcu EM, Hobbs H, Rincon G, Bott-Silverman C, DeFranco AC, Robinson K, McCarthy P, Stewart R, Guyer S, Nissen SE. Occult and frequent transmission of atherosclerosis coronary disease with cardiac transplantation. Circulation. 1995;91:1706-1713.
Pratt CM, Moyé LA. The Cardiac Arrhythmia Suppression Trial: casting suppression in a different light. Circulation. 1995;91:245-247.
Packer M, Carver JR, Rodeheffer RJ, Ivanhoe RJ, DiBianco R, Zeldis SM, Hendrix GH, Bommer WJ, Elkayam U, Kukin ML, Mallis GI, Sollano JA, Shannon J, Tandon PK, De Mats DL, for the PROMISE Study Research Group. Effect of oral milrinone on mortality in severe chronic heart failure. N Engl J Med. 1991;325:1468-1475.
The GUSTO IIa Investigators. Randomized trial of intravenous heparin versus recombinant hirudin for acute coronary syndromes. Circulation. 1994;90:1631-1637.
Simes RJ, Topol EJ, Holmes DR, White HD, Rutsch WR, Vahanian A, Simoons ML, Morris D, Betriu A, Califf RM, Ross AM, for the GUSTO Investigators. Link between the angiographic substudy and mortality outcomes in a large randomized trial of myocardial reperfusion: the importance of early and complete infarct artery reperfusion. Circulation. 1995;91:1923-1928.