Aortic Atherosclerotic Disease and Stroke
2006; Lippincott Williams & Wilkins; Volume: 114; Issue: 1 Linguagem: Inglês
10.1161/circulationaha.105.593418
ISSN1524-4539
AutoresItzhak Kronzon, Paul A. Tunick,
Tópico(s)Cardiac Valve Diseases and Treatments
ResumoHomeCirculationVol. 114, No. 1Aortic Atherosclerotic Disease and Stroke Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBAortic Atherosclerotic Disease and Stroke Itzhak Kronzon, MD and Paul A. Tunick, MD Itzhak KronzonItzhak Kronzon From the Charles and Rose Wohlstetter Noninvasive Cardiology Laboratory, New York University School of Medicine, New York. and Paul A. TunickPaul A. Tunick From the Charles and Rose Wohlstetter Noninvasive Cardiology Laboratory, New York University School of Medicine, New York. Originally published4 Jul 2006https://doi.org/10.1161/CIRCULATIONAHA.105.593418Circulation. 2006;114:63–75In the 1940s, most strokes were attributed to cerebral vasospasm, a mechanism that is not given a great deal of credence today. It was not until the early 1950s that Harvard neurologist C. Miller Fisher1 stressed the importance of carotid artery atherosclerosis as a major cause of cerebral infarction. Later that decade, the importance of atrial fibrillation as a cause of cerebral embolism began to be stressed,2 and the presence of a left atrial thrombus was first seen on angiocardiography in 1965.3 Despite the established importance of these 2 causes of stroke, carotid disease and atrial fibrillation, nearly half of strokes were listed as "of undetermined cause" in a large stroke registry as recently as 1989.4 In this series, 40% of 1273 cerebral infarctions in the Stroke Databank of the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) were thought to be cryptogenic (from the Latin crypticus, meaning secret or mysterious). The clinical syndrome in those patients, as well as the angiographic and computed tomographic (CT) findings, could be reclassified as embolic; however, because no source of embolus could be identified, the authors kept these strokes in the undetermined cause category.In 1990, a third leading cause of embolic stroke was identified on transesophageal echocardiography (TEE), namely severe atherosclerotic plaques in the aortic arch.5 The 3 patients described in that initial report were a 68-year-old woman with dysarthria and an embolus to the foot, a 77-year-old woman with a cerebellar infarction after cardiac catheterization, and a 70-year-old man with staggering, diplopia, and a visual field cut. All 3 had severe plaque in the aortic arch on TEE. In addition, freely mobile projections were seen superimposed on the plaques, making it seem likely that these findings were the reason for the patients' embolic events. That atherosclerosis of the aorta and arterial emboli are related is not a new concept. In fact, since 1862, pathologists have suggested that "eroded atherosclerotic plaque" may result in embolic arterial occlusions.5–7 However, it required sophisticated imaging technology such as TEE (Figure 1) (and later CT and magnetic resonance imaging [MRI]) to view these lesions in vivo. Download figureDownload PowerPointFigure 1. TEE showing aortic arch with very severe atherosclerotic plaque.The subject of the risk of aortic arch plaque as seen on TEE has been reviewed several times in the last 15 years.8–10 More recently, the ability to image aortic plaque and its contents has improved significantly through the use of different imaging modalities, and our understanding of the associated clinical syndromes has increased. Although most investigators feel that aortic atherosclerosis is a leading cause of embolic disease, doubts have been raised.11 All of these issues are discussed in this review.DefinitionsAortic atherosclerotic lesions have been referred to in several different ways in the medical literature. These include atheromas, protruding atheromas, atherosclerotic debris, and plaque. For the purposes of this review, we define these lesions as plaques. The mobile components to these plaques have been called mobile debris, mobile plaque, and superimposed thrombi. These mobile lesions are most often thrombi. Clinical data show that the plaques with high risk for embolization are those that are ≥4 mm thick. We refer to these lesions as severe plaques. Finally, the term complex plaque has been used in the literature to refer to those plaques that are ≥4 mm thick (called severe here), contain mobile elements (most often thrombi), or both.Prevalence of Aortic AtherosclerosisMost studies of aortic plaque include patients who were symptomatic and referred for diagnostic studies. However, the Stroke Prevention: Assessment of Risk in a Community (SPARC) study enrolled patients at random and therefore was not subject to referral bias.11 Of 588 patients having TEE as part of the study (average age, 66.9 years), aortic plaque in any location was present in 43.7%, of which complex plaque (defined as ≥4 mm or mobile) was present in 7.6% of patients. Aortic plaques were present in the ascending aorta in 8.4%, but complex plaques were noted in the ascending aorta in only 0.2%. Plaques were present in the aortic arch in 31%, but complex arch plaques were present in only 2.2%. Finally, descending aortic plaques were seen in 44.9%, and complex plaques in the descending aorta were present in 6.0%. Thus, the number with any or complex aortic plaque on TEE increased from the ascending aorta distally.It should be noted that this information is true for a relatively elderly population that is homogeneous (white). Half of the patients were male. The prevalence of aortic plaque in other groups not subject to referral bias is not known.Aortic Atherosclerotic Plaque and the Risk of StrokeIt is important to realize that the prevalence of severe aortic plaque in stroke patients (14% to 21%) is on the same order of magnitude as that of the other 2 important causes of embolic stroke, carotid artery disease (10% to 13%) and atrial fibrillation (18% to 30%), which were documented in 2 large series of consecutive stroke patients.12,13Early investigators looked for a risk of embolization in patients with the largest aortic plaques as measured with TEE.8 In this study, patients with aortic plaque thickness ≥5 mm had a significantly higher risk of stroke and peripheral embolization.The French Aortic Plaque in Stroke (FAPS) group evaluated a large number of patients with varying plaque thickness and found not only that increasing plaque thickness imparted increasing risk but also that the odds ratio (OR) was significantly greater in those with a plaque thickness ≥4 mm.12 The OR for stroke in patients with plaques 1 potential source of embolization to the brain.8,13,27 One retrospective study28 found aortic arch plaques in significantly more stroke and transient ischemic attack patients with carotid stenosis (38%) than in those without carotid stenosis(17%). That study also found that mobile thrombi superimposed on the aortic arch plaques were exclusively found in patients whose carotid stenosis was >80%. Thus, patients with the highest-risk carotid lesions also had the highest-risk aortic plaques. Therefore, when one sees a patient with stroke and severe carotid stenosis, it is important to always consider the possibility that the embolic event may have originated in the aortic arch. When the stroke or transient ischemic attack is contralateral to the carotid stenosis or is associated with peripheral emboli, the aorta should always be evaluated. This evaluation should also take place in patients who have neurological events after recovery from a technically successful carotid operation.Coronary Artery DiseaseThe Framingham study found an association between aortic calcification seen on chest x-ray and the development of coronary artery disease.29 Aortic plaque seen on TEE has been correlated with a higher prevalence of coronary artery disease and the presence of significant angiographic coronary artery stenosis.30 In addition, the lack of aortic plaque on TEE has also been shown to predict the absence of coronary artery disease.31 The sensitivity and specificity of the presence of aortic plaque for the prediction of significant coronary stenosis were both 90%. The positive predictive value of plaque detected by TEE was 95%; the negative predictive value was 82%. In addition, CT has shown that the presence of aortic plaque was predictive of obstructive coronary artery disease independently of coronary artery calcification. The sensitivity of aortic plaque for predicting the presence of obstructive coronary artery disease was 89%, and the specificity was 63%.32 Because a patient without aortic plaque on TEE is less likely to have obstructive coronary artery disease, younger patients without a clinical suspicion of coronary disease who are undergoing open heart surgery to repair congenital defects or acquired valve disease may not need coronary angiography if the TEE does not show aortic plaque. TEE may be very important for an additional reason in patients who are slated to have coronary bypass (or other) heart surgery: Aortic plaque identified by TEE may have a big impact on the outcome of heart surgery. The implications of identifying aortic plaque for heart surgery are discussed below.Renal Artery DiseaseHypertension is common in patients with aortic plaque. One of the sometimes-missed causes of hypertension is renal artery stenosis, which in older patients is yet another complication of the atherosclerotic process. One study using abdominal ultrasound to look for renal artery stenosis in patients with severe aortic plaque on TEE found it in 19% of these patients.33 No renal artery stenosis was detected in any control subjects (without severe aortic plaque). Therefore, renal artery stenosis should be ruled out in patients with severe aortic plaque and hypertension. In addition, TEE detection of aortic plaque may provide a clue to the cause of occult renal dysfunction in the elderly,34 which may occur because of recurrent renal embolization from aortic plaque.Abdominal Aortic AneurysmSevere thoracic aortic plaque was found in twice as many patients with abdominal aortic aneurysm on abdominal ultrasound or angiography (52% in those with Taneurysm versus 25% in those without aneurysm). In addition, researchers performed abdominal ultrasound to look for aneurysms of the aorta in patients with severe thoracic aortic plaque that had been found when TEE was done for various clinical indications. Abdominal aneurysms were found in 10 times more patients with severe thoracic plaque (13.9%) than in those without such plaque (1.4%).35 Because an abdominal aneurysm may have a catastrophic outcome, such an aneurysm should be ruled out if severe plaque is seen in the thoracic aorta.Association of Aortic Plaque With Aortic Stenosis and Mitral Annular CalcificationSeveral studies have reported on the association of valvular aortic stenosis and atherosclerosis. In 1 study,36 92 patients who had severe aortic stenosis and underwent TEE were matched by age and sex with 91 control subjects who also underwent TEE. Severe, complex aortic plaque was noted in 47% of the patients with severe aortic stenosis compared with only 9% of the control subjects. In another study, nonobstructive aortic valve calcification (without stenosis) was also significantly associated with aortic plaque on TEE.37 Aortic plaque was present in 86% of those with aortic valve calcification compared with 30% of control subjects without aortic valve calcification.Mitral annular calcification is most common in the elderly and is associated with known atherosclerosis risk factors. A different study by the same researchers found a significantly higher incidence of aortic plaque, especially complex plaque, in those with annular calcification: 74% of the patients with mitral annular calcification had complex aortic plaque versus only 22% of controls. The thickness of the mitral annular calcification was directly related to the severity of the aortic plaque.38 Therefore, consideration should be given to evaluation of the aorta (especially before valve replacement) in patients with severe aortic stenosis, nonstenotic aortic valve calcification, or mitral annular calcification.Aortic Plaque and Atrial FibrillationThe Stroke Prevention in Atrial Fibrillation (SPAF) study reported 382 patients with "high-risk" nonvalvular atrial fibrillation (age >75 years, hypertension, or previous stroke) who had TEE.39 Complex aortic plaque (mobile, ulcerated, and/or size ≥4 mm) on TEE was present in 35% of the atrial fibrillation patients. Those with plaque had a stroke risk in 1 year of 12% to 20%. The risk was dramatically lower, only 1.2%, in the high-risk nonvalvular atrial fibrillation patients who did not have significant aortic plaque. Therefore, if a stroke occurs in the setting of high-risk nonvalvular atrial fibrillation and an "upstream" severe aortic plaque is present, the atrial fibrillation may not be the only possible cause of the embolus.The SPARC study, also, found an association of aortic plaque and atrial fibrillation using a community-based approach. However, the association was no longer significant after an adjustment was made for age.40Atherosclerosis of the Aorta: Risk Factor or Innocent Bystander?To avoid the referral bias inherent in many of the TEE studies of aortic plaque, the SPARC investigators selected 585 patients randomly from the community for evaluation with TEE and follow-up for 5 years. Simple aortic plaques, seen in 253 persons, were not found to be associated with cardiac or cerebral events. Complex plaques, present in 44 patients, were marginally associated, but this association was no longer present after adjustment for additional clinical risk factors. Why do these results conflict with the data mentioned above in this review, which support the malignant nature of severe aortic plaque seen on TEE? The answer lies in the distribution of complex plaques that were found in these randomly selected, community-based patients. Only a small number of complex plaques were found in an area that could embolize to the brain. Complex aortic plaque was seen in the ascending aorta in only 1 patient (0.2% of their patients). Complex plaque was also seen in the aortic arch in only 2.2%. Most of the complex plaques found in this study were in the descending aorta (6%, 35 patients). Such plaques in the descending aorta are very unlikely to embolize backward in the circulation to the brain (except in the unusual coincident case of severe aortic insufficiency). In fact, in the FAPS study,12 the OR for stroke in patients with severe plaque in the descending aorta was only 1.5 (versus 13.8 for plaques of the same size in the ascending aorta or arch). Thus, although the community-based study described above11 is an important contribution and avoids the referral bias inherent in the other studies of aortic plaque, it does not appear to be powered to detect the association of aortic plaque and stroke because it does not contain enough patients with high-risk plaque in the proximal aorta.In several case reports, there is direct evidence that severe aortic plaque does in fact cause embolization. In a TEE performed for a patient during diffuse embolization (which resulted in multiple organ failure and death), particulate matter was seen detaching from an aorta filled with severe plaque and traveling distally in the circulation.41 In another patient who underwent aortic cannulation during open heart surgery, a mobile component of an aortic arch plaque, presumably thrombus, was actually struck and detached from the plaque during cannulation of the aorta under TEE visualization. The patient awoke from surgery with a stroke.42 Furthermore, patients with aortic arch plaque have far more emboli to the left brain and periphery than they do to the right brain.8 This is true because the innominate artery (which supplies the right brain) is proximal to aortic arch plaque in most cases. Finally, pathological specimens have shown that the mobile components to aortic plaques are most often thrombi, which can detach and embolize (Figure 2). Download figureDownload PowerPointFigure 2. A, TEE showing junction of aortic arch and descending aorta. Atherosclerotic plaque is shown on top with 2 superimposed mobile thrombi (T; arrows). B, Gross pathological specimen of 2 thrombi attached to atherosclerotic plaque (P) surgically removed from aorta (shown on TEE in A). C, Microscopic pathological specimen of the same lesion. D, Gross pathological specimen of thrombus surgically removed from femoral artery of the same patient.Pathological CorrelationsIn 1992, a landmark autopsy study evaluated 500 patients with stroke and other neurological diseases.43 The authors showed that ulcerated plaques were present in the aortic arch in 26% of 239 patients with cerebrovascular disease but in only 5% of 261 patients with other neurological diseases. In addition, the prevalence of ulcerated plaques in the aortic arch was much higher (61%) in 28 patients with "cryptogenic stroke" than in the 155 patients with stroke and carotid disease or another known cause of stroke, in whom the prevalence was only 22%.Since the first reports, investigators have found that a significant number (25% to 50%) of severe aortic plaques seen on TEE have a mobile component. These plaques may be found in patients with both a large and a small atherosclerotic burden. They range in size from 1 mm to several centimeters. Case reports of 2 such patients who underwent surgery documented that these mobile lesions were in fact thrombi superimposed on atherosclerotic plaque.44,45 In one of these patients, red thrombi were seen both superimposed on the aortic plaque and in the specimen removed from the femoral artery where it had embolized (Figure 2).The presence of thrombi on aortic plaques was further documented on autopsy in a more recent study,46 which found aortic thrombi in 17 of 120 cases, as well as a significant association of complex plaque with previous emboli. In addition, mobile lesions in the aorta seen on TEE in 6 patients were seen to be thrombi on surgical pathology.47 A TEE study48 showed changes in morphology when TEE was repeated in the same patients over a period of time. New mobile lesions were seen on plaques that initially had none, and some of the presumed thrombi had disappeared (dissolved or embolized) by the time that the repeat study was performed.In the coronary arteries, thrombosis has been shown to occur when the arterial wall contains a large lipid core, which predisposes to plaque rupture. A similar situation probably exists in the aorta. One group found that aortic plaques that contain a preponderance of lipid are likely to have undergone thrombosis.49 Calcified aortic plaques (which are less lipid laden) have a lower risk of plaque rupture, thrombus formation, and embolization (see Table 2). TABLE 2. Plaque Characteristics and Stroke RiskData derived from Amarenco et al.12Thickness, mm<11–1.92–3.9≥4Risk, OR13.34.113.8Findings associated with high stroke risk Thrombus (mobile or not) Ulceration Large lipid core (hypoechoic)Finding associated with lower stroke risk Plaque calcification (hyperechoic)Cholesterol Crystal Embolization (the Atheroemboli or "Blue Toe" Syndrome)The syndrome of stroke and peripheral embolization described above is most often the result of thromboembolism to medium-sized and larger arteries from unstable aortic plaques. Much less common is the "classic" syndrome of diffuse cholesterol crystal embolization, the atheroemboli or "blue toe" syndrome. In the latter entity, there is a diffuse showering of cholesterol crystals from unstable aortic (or other arterial) plaques. This showering may occur spontaneously or may be precipitated by disruption of plaques during trauma, interventional procedures such as angiography or cardiac catheterization, or vascular surgery. Patients with this syndrome suffer from cerebral dysfunction, renal failure, intestinal infarction, and limb ischemia that is usually diffuse, distal, and bilateral.50,51 There is a very high mortality rate. This syndrome is a much rarer complication of aortic atherosclerosis than is the syndrome of thromboembolism. Thromboembolism occurs in up to 33% of patients with severe aortic plaque on TEE in 1 year.14 In contrast, atheroembolism was reported in 0.7% of patients with severe aortic plaque on TEE.39Imaging ModalitiesContrast AngiographyHistorically, imaging of the aorta was done first (in vivo) with contrast angiography. This technique can define major aortic abnormalities; however, it is invasive and requires the use of contrast and radiation. In addition, angiography may miss important aortic plaques viewable by TEE.52 Therefore, angiography should not be used to evaluate aortic plaque.Transthoracic EchocardiographyTransthoracic echocardiography (TTE) frequently visualizes the aortic root and proximal ascending aorta. In some patients, the aortic arch can be seen from the suprasternal notch and the descending aorta from that window and on apical views. Aortic plaque may be visualized with TTE B-mode imaging.53 Harmonic imaging may add to the accuracy of this technique.54Transesophageal EchocardiographyMost echocardiographers feel that TEE is more accurate than TTE for the critical measurement of plaque thickness (Figure 3 and Table 2) and for the diagnosis of mobile thrombi. The TEE probe is closer to the aorta and can be used at a higher frequency, thus allowing for higher resolution than on TTE. TEE is safe,55 can be brought to the bedside for critically ill patients, and allows the evaluation of other possible reasons for stroke. An accurate and detailed evaluation of the aorta, including the origin of the great vessels, is possible,56 and there is excellent interobserver and intraobserver variability.57Download figureDownload PowerPointFigure 3. TEE showing normal aortic intima (1 mm) and mild (2 mm), moderate (3 mm), and severe (≥4 mm; shown here=7 mm) plaque.Plaque composition may be evaluated with TEE, including plaque calcification. The latter feature is important because the embolic risk is lower if aortic plaques are calcified.58 The reason for this decrease in risk may be that lipid-laden (noncalcified) plaques, which appear hypoechoic on TEE, are more prone to rupture and thrombosis,59 and these superimposed thrombi embolize to cause stroke and other organ damage.The ability to record the amount of plaque at different levels in the aorta on TEE allows estimation of the total plaque burden. This is useful for assessing risk and the response to therapy.60 Newer TEE technology allows 3-dimensional reconstruction of the aorta,61 which may improve the localization and quantification of atherosclerotic plaque burden and risk.The limitations of TEE include occasional discomfort for the patient (gagging), the frequent use of conscious sedation, and the rare risk of oropharyngeal and esophageal damage.62 In addition, the small portion of the ascending aorta that is masked by the tracheal air column near the origin of the innominate artery may not be seen on TEE; thus &2% of plaques may be missed.63Epiaortic Ultrasound ImagingEpiaortic imaging, yet another way to evaluate the aorta, is accomplished by placing a transducer directly on the aorta when the chest is open. The application of ultrasound is discussed later, in the section describing cardiac and aortic surgery in patients with atherosclerosis.Magnetic Resonance ImagingIn 1983, MRI was first used to image aortic atherosclerosis.64 More recently, MRI was compared with TEE for the evaluation of aortic plaque.65 In this study, MR angiography underestimated plaque thickness in the aortic arch, probably because of difficulties in defining the aortic wall on the contrast-enhanced MR angiograms.One important advantage of MRIs is that they reveal contrast between different tissue types and can thus identify morphological components of the atherosclerotic plaque such as calcification, fibrocellular tissue, lipid, and thrombus.66,67 Plaque stability depends on the size of the lipid core (Figure 4), the thickness of the fibrous cap, and inflammation within the cap. All of these parameters can be evaluated with MRI.68 In addition, MRI can be used to monitor the progression and regression of atherosclerotic plaques.69 To improve the signal-to-noise ratio of MRI, a radiofrequency receiver probe may be placed in the esophagus. One study used a transesophageal probe to obtain images that correlated well with those obtained by TEE and provided important information about plaque composition.70Download figureDownload PowerPointFigure 4. MRI showing severe descending aortic plaque. Note the relatively lucent lipid core (arrow). Reproduced from Tunick et al106 with permission from the American Journal of Roentgenology.MR angiography (Figure 5) is a valuable technique because it may supply 3-dimensional images of the entire vascular tree, including the small area masked by the tracheal air column on TEE.63Download figureDownload PowerPointFigure 5. MR angiography showing large protruding lesion, probably a mobile thrombus (black arrow). Open arrow points to the innominate artery. Reproduced from Tunick et al106 with permission from the American Journal of Roentgenology.Although MRI images supply important information about aortic plaque, MRI requires expensive and cumbersome equipment and cannot be done at the bedside or on patients in intensive care. This equipment also cannot be used on patients with ferromagnetic implants, particles, pacemakers, or defibrillators. In addition, claustrophobia in some patients may be a limiting factor. Finally, MRI is difficult to use for assessing mobile thrombi, which are an important factor in the development of embolization from unstable plaques. For these reasons, MRI is not routinely by most cardiologists to evaluate atherosclerosis in patients with stroke or other embolic syndromes.Computerized TomographyCT scanning is yet another modality frequently used to evaluate the aorta and its branches (Figure 6). Unenhanced dual-helical CT with thin sections has been reported to be successful in detecting protruding aortic plaque, especially in areas not visualized by TEE (94% of plaques detected by TEE were seen with CT).71 Additional benefits over the contrast-enhanced CT technique have been reported using spiral computerized dual-helical CT scanning.72 In this study, the plaques were identified in 95% of cases as co
Referência(s)