The role of plaque morphology and diameter reduction in the development of new symptoms in asymptomatic carotid arteries

Transcription

1 The role of plaque morphology and diameter reduction in the development of new symptoms in asymptomatic carotid arteries Mark Langsfeld, MD, Anthony C. Gray-Weale, FRCS, and Robert J. Lusby, MD, FRCS, FRACS, New South Wales, Australia To determine the natural history of changes in plaque morphology and luminal diameter of atherosclerotic carotid arteries, we used duplex scanning to follow-up (1) the contralateral artery in 289 patients who had undergone carotid endarterectomy, with a mean follow-up 22 months and a range of 0 to 48 months and (2) the carotid arteries in 130 patients who had no surgical treatment and had been symptom free, with a mean followup period of 15 months and a range of 0 to 48 months. Plaques were graded as to the ratio of echolucency to echogenicity, with type I being most echolucent and type 4 being most echogenic. A normal-appearing artery was classified as type 5. Heterogeneous plaques (types 1 and 2) occurred significantly more (p < 0.001) in symptomatic preoperative arteries than in asympatomatic arteries. Follow-up of the asymptomatic vessels showed that the majority of plaques either remained the same or became more echogenic (fibrous). Approximately one fourth of plaques in each group degenerated (more echolucent). Thirty-one patients (10.7%) developed new symptoms in the contralateral asymptomatic group, with 10 patients (3.5%) having strokes. Fourteen of 130 (10.8%) patients, or 5.4% of vessel territories at risk, in the primary asymptomatic group developed new symptoms, with only two strokes occurring. In the contralateral asymptomatic group those patients who initially had >75% stenoses fared worse than those with primary asymptomatic disease with >75% stenosis. Although the overall development of new symptoms is low in both populations, our data indicate that those patients with heterogeneous plaques or whose plaques have undergone change may be at risk for new symptoms. Longer follow-up studies are needed to define the role of plaque changes in the development of symptoms. For now we advocate a conservative "wait and see" approach to symptom-free patients with >75% stenoses and calcified plaques. We suggest a more aggressive approach, recommending early surgical intervention, to those few patients with heterogeneous plaques. (J VAse SUV.G 1989;9: ) Carotid bifurcation atherosclerosis plays an important role in the development of stroke, yet the natural history of plaque progression and its relationship to symptom onset is not fully understood. Plaque morphology, with particular attention to the presence of intraplaque hemorrhage, has been implicated as an important factor in the development ofneurologic symptoms. 1-3 Duplex scanning gives us the ability to delineate plaque morphology (its shape and composition), allowing the noninvasive monitoring of progression and changes in plaque mor- From the Department of Surgery, University of Sydney, Repartriation General Hospital. Presented at the Non-Invasive Diagnostic Techniques in Vascular Disease Symposium, San Diego, Calif., February 22-26, Reprint requests: Robert J. Lusby, MD, University Department of Surgery, Repatriation General Hospital, Concord, 2139, New South Wales, Australia. 548 phology. The management of patients with asymptomatic carotid plaques, in particular, depends on such information and remains central to the debate on the place of carotid artery surgery. Recent reports have advocated a more conservative approach to the management of diseased asymptomatic vessels or vessels contralateral to an operated carotid artery, s7 and randomized studies are now underway to evaluate the place of surgical therapy. However, these studies do not focus on plaque morphology and its role in the development of new symptoms. During the last 4 years our vascular laboratory has followed up patients with carotid bifurcation disease, paying particular attention to the structure and composition of the plaques seen on duplex scanning. The aim of this study is to determine if there are changes in plaque morphology and luminal diameter reduction in asymptomatic carotid arteries and de-

2 Volume 9 Ntlrnber 4 April 1989 Role of plaque morphology and diameter reduction 549 termine what relationship these changes have to the development of new symptoms. PATIENT POPULATION AND METHODS OF STUDY Two problematic "asympotomatic" population groups that have been followed up in our vascular laboratory were chosen for analysis. The first group consisted of patients who had undergone carotid endarterectomy and had an asymptomatic lesion in the contralateral carotid artery. A total of 289 patients who had undergone unilateral carotid endarterectomy during the previous 4 years constituted this group. The mean age of the group was 66.5 years. There were 239 men and 50 women reflecting the veteran population. Ipsilaterat carotid territory symptoms were the indication for operation in 80% of these patients, the remainder being for nonlocalizing symptoms. Five patients with >75% stenosis elected to have prophylactic carotid endarterectomy. Smoking Was by far the greatest associated risk factor, with 81% of the population having been former heavy smokers or currently smoking. Hypertension was present in 51%, ischemic heart disease in 41%, family history of stroke or heart disease in 34%, and associated peripheral vascular disease in 32% of the patients. Only 8% of this population had diabetes. Of the 289 patients 37 did not undergo preoperative scanning. In this group the unoperated carotid artery was followed from the first postoperative scan. Patients were followed from 0 to 48 months for a mean of 21.6 months. Follow-up duplex scans were performed 3, 6, and 12 months after surgery and then yearly thereafter. The second group of patients with primary asymptomatic disease had not undergone prior carotid artery surgery but were referred to our laboratory because of carotid bruits or other suspected cardiovascular disease. A total of 130 patients were in this group providing 260 carotid arteries for study. The mean age was 67.9 years. There were 109 men and 21 women. All patients had been symptom free for at least I year before referral, but 37 did have previous cerebrovascular symptoms. Mean follow-up time was 15 months. A history of smoking was again the greatest associated risk factor, present in 89% of these patients. Heart disease occurred in 64%, hypertension in 58%, family history for cardiovascular or cerebrovascular disease in 51%, and associated peripheral vascular disease in 49% of the patients. Nearly one third of these patients had diabetes (30%) or hypercholesterolemia (30%). Follow-up duplex scans were performed at 6 and 12 months and then yearly thereafter. At the initial visit each patient was carefully interviewed as to the nature of symptoms, past medical history, and presence of risk factors. On follow-up examinations an interview was again performed with particular attention to interval symptoms. Duplex scanning was then performed with a Hoffrel model 518 with a 7.5 MHz probe (Hoffrel Instruments, Inc., S. Norwalk, Conn.) and a Medasonics model SPA25 spectrum analyzer (Medasonics, Inc., Mountain View, Calif.). The common, internal, and external carotid arteries were imaged in standard longitudinal and cross-sectional views. Degree ofstenosis was estimated by direct visualization on B- mode scanning with electronic caliper estimation of residual lumen and vessel diameter. In addition spectrum analysis criteria 8 were used to confirm the stenotic category, the greater stenosis by either technique being the agreed value. Categories of stenosis were defined as 0% to 20%, 21% to 49%, 50% to 75%, 76% to 99%, and total occlusion. Visualized plaque at the carotid bifurcation was graded as to the ratio ofechogenicity to echolucency. The following classification is used: type 1, predominantly echolucent raised lesion, with thin "egg shell" cap of echogenicity; type 2, echogenic lesions with substantial areas of echolucency; type 3, predominantly echogenic with small area(s) of echolucency deeply localized and occupying less then a quarter of the plaque; type 4, uniformly dense echogenic lesions; and type 5, no plaque seen; increased wall density, or normal. Videotapes were made of each scan for review and polaroid pictures were taken of the bifurcation that best demonstrated the disease. Hard copies of each spectral analysis were stored for each study. First and last scans on asymptomatic unoperated vessels were compared in relation to degree of luminal stenosis and plaque morphology. If a patient developed a new carotid territory symptom the scan closest in time after the new symptom developed was used for the last scan. Patients excluded from the study include those who were lost to followup or those who elected to undergo prophylactic endarterectomy without the development of new symptoms. RESULTS A total of 291 carotid endarterectomies were performed on 289 patients from November 1983 to November 1987 at our hospital. Two patients had early recurrent stenoses and required reoperation. Thirty-one patients (10.7%) developed new carotid territory symptoms on the contralateral side during

3 550 Langsfeld, Gray-Weale, and Lusby Journal of VASCULAR SURGERY Fig. 1. Duplex scan illustrates a type 1 plaque arising from the posterior wall at the origin of the internal carotid artery (ICA). The arrow shows the thin "egg shell" cap ofechogenic material. Fig. 2. Duplex scan illustrates a type 2 plaque arising from the anterior and posterior wall at the origin of the internal carotid artery (ICA). There is more echogenic material contained in this plaque than seen in type 1. follow-up, with mean time to onset of symptoms of 16 months. Of these new symptoms 10 were cerebrovascular accidents (CVAs) (3.5%), 16 were transient ischemic attacks (TIAs) (5.5%), and five patients had amaurosis fugax (1.7%). Nine of the 31 patients who developed new symptoms underwent a contralateral endarterectomy. Indications for these endarterectomies were TIAs in seven patients, CVA with recovery in one patient, and amaurosis fugax in one patient. In the primary asymptomatic group 14 patients (10.8%) developed carotid territory symptoms (5.4% of vessel territories at risk), with a mean time to onset of 15 months. Six of these patients subsequently underwent eight carotid endarterectomies. Symptoms in these 14 patients were 10 TIAs (7.7% of patients, 3.8% of vessel territories at risk), two attacks of amaurosis fugax (1.5% of patients, 0.7% of vessel territories at risk), and two strokes (1.5% of patients, 0.7% of vessel territories at risk). Initial presentation. The plaque types on entry to the study of the vessels associated with symptoms are shown in Fig. 1 and those in the two asymptomatic groups are shown in Fig. 2. The degree of luminal stenosis at initial presentation for the endarterectomy side is shown in Fig. 3 and the distribution in the two different asymptomatic groups is

4 Volume 9 Number 4 April 1989 Role of plaque morphology and diameter reduction 551 Fig. 3. Duplex scan illustrates a type 3 plaque arising mainly from the posterior wall at the origin of the internal carotid artery (/CA). The plaque is predominantly echodense with a small component of echolucent content at the proximal end (arrow). Note the ultrasound shadowing (S) produced beneath the central portion of the plaque. Fig. 4. Duplex scan of a type 4 plaque shows a uniformly echodense lesion at the origin of the internal carotid artery (ICA). shown in Fig. 4. Predominanaly echolucent plaques (types 1 and 2) occurred significantly more often (p < 0.001) in the operated side than in the primary asymptomatic or contralateral vessels. The majority of plaques in the primary asymptomatic or contralateral asymptomatic arteries were predominantly echogenic (fibrous, calcific, or normal), types 3, 4, and 5. In comparing the degree of luminal stenosis, 63% of the operated sides had a >50% stenosis, significantly greater than the 32% of contralateral asympatomatic arteries with >50% stenosis (p < 0.001) or 26% of the primary asymptomatic vessels with >50% stenosis (p < 0.001). Changes in plaque morphology. The changes in plaque morphology in the two different asymptomatic artery groups are shown in Fig. 5, and the changes in luminal stenosis are shown in Fig. 6. Plaques maintained the same morphologic characteristics or increased in echodensity between first and last scans in the majority of patients in both groups. Approximately one fourth of the plaques increased in echolucency or heterogenicity, with ap-

5 552 Langsfeld, Gray-Weale, and Lusby Journal of VASCULAR SURGERY 50-45' 40' 3s- 30-.=.. 2s, 20 L ~_ is, 10 L 5-0 n ~25- ~ ~ 20. ~ ,,,, I PLAQUE TYPES Fig. 5. The distribution of plaque types in 252 vessels of patients with symptoms before endarterectomy and entry into this study. n = i PERCENT DIAMETER REDUCTION Fig. 7. The distribution of degree of diameter reduction in 252 vessels of patients with symptoms before endarterectomy and entry into the study. 50- [] PRIMARY n=260 ASYMPTOMATIC 35 n-2s ASYMPTOMATIC ~ 25 ASYMPTOMATIC,,z, 25 - n=289,z, n IB ~ ~ 10 0 I I I I I I! I I I PLAQUE TYPES PERCENT DIAMETER REDUCTION Fig. 6. The distribution of plaque types in 260 primary asymptomatic and 289 contralateral asymptomatic cases on entry into the study. Fig. 8. The distribution of degree of diameter reduction in primary and contralateral asymptomatic cases on entry into the study. proximately 10% changing to the most heterogeneous plaques, types 1 and 2. Similarly with the degree of stenosis, approximately one quarter of the arteries progressed to a higher stenosis category during follow-up and one quarter were classified in a lower stenotic category. Development of symptoms. A total of 31 patients (10.7%) developed new symptoms in the contralateral, unoperated artery. Five arteries were occluded on initial or follow-up examination, excluding them from evaluation of plaque morphology. Plaque morphologic characteristics changed in 16 of 24 plaques (66.6%) in the new symptom group between the initial scan and the last preoperative scan. It is interesting that only five plaques became more echolucent / heterogeneous; the other 11 changes involved the plaque becoming more echogenic. The majority of contralateral arteries (16 / 30) maintained the same degree of luminal stenosis; eight of 30 were judged to be less stenotic. Fewer than one quarter of the vessels (6/30) increased in diameter reduction between scans. In the primary asymptomatic group, 14 patients (10.8%) developed new carotid territory symptoms (5.4% of vessel territories at risk). Again, most plaques tended to remain the same or increase in echogenicity (9/14), and diameter reduction increased in only four of 14 cases. The total stroke rate on the contralateral side was 10 of 289 patients (3.5%). Six of eight plaque types changed morphologically, but only one became more echolucent. Only one of the 10 arteries increased in diameter reduction, from a 50% to 75% category to total occlusion. Two patients (1.5%) in the primary asymptomatic group developed a stroke, or two of 260 vessel territories at risk (0.7%). Symptom risk. In determining the relative risk of developing new symptoms in the overall study population, both the plaque type on initial presentation and degree of luminal stenosis proved to be important. The more echolucent, heterogeneous plaque types 1 and 2 were associated with significandy more symptoms (p < 0.02) than the denser echogenic plaques (Fig. 7). There was a 43% relative risk of new symptoms developing in patients who initially had a patent vessel of >75% diameter reduction (p < 0.001) (Fig. 8). Tables I, II, and III

6 Volume 9 Number 4 April 1989 Role of plaque morphology and diameter reduction 553 Table I. Primary symptom-free group Plaque type Diameter reduction (%) Remained symptom free New symptoms Total Table II. Contralateral symptom-free group Plaque type Remained symptom free New symptoms Total Diameter reduction (%) Table III. Total population Plaque type Diameter reduction (%) i Remained symptom free New symptoms Total summarize the initial presentation data for those patients who developed new symptoms in the two study groups, as well as the total population. DISCUSSION Our study attempts to determine the influence of plaque morphology on the outcome of a highly select group of patients. The ability of duplex scanning to assess the carotid bifurcation has been well documented. 8-n Most studies have used spectral analysis for predicting the diameter reduction of carotid arteries. However, ultrasonography can also detail the "makeup" of carotid plaque, with denser, fibrous, or calcified material being more echogenic and softer or heterogeneous material, such as blood or lipid, being more echolucent. 4 The importance of plaque composition and intraplaque hemorrhage in the cause of neurologic symptoms has been well documented, 1-s although controversy exists as to the exact relationship of these changes to symptom onsety Duplex scanning now enables us to perform long-term studies of plaque morphology in an effort to identify patients at high risk of stroke. As doubt continues over the value of carotid end- arterectomy in preventing stroke, many physicians have adopted a more conservative approach to treating carotid bifurcation disease, especially in patients without symptoms. We have therefore looked at the natural history of two groups of symptom-free patients referred to our vascular laboratory to determine the influence of plaque morphology and diameter reduction on the development of new symptoms. To study a population group that has had a large number ofneurologic events and appears to carry an increased risk, we chose to look at our population of symptomfree patients who had undergone a previous unilateral carotid endarterectomy. This group also poses a clinical problem as to how to manage the contralateral, asymptomatic side and may account for a substantial portion of carotid endarterectomies being performed yearly. 13 Roederer et al.7 reported small numbers of patients developing postoperative neurologic symptoms in the contralateral artery. Most of these symptoms were TIAs and not unheralded stroke. However, other reports indicate stroke rates ranging up to 36% on the contralateral, unoperated side.1416 Our study confirms the low rate of new symptoms in the contralateral carotid artery. New symptoms in this

7 554 Langsfeld, Gray-Weale, and Lusby Journal of VASCULAR SURGERY [] PRIMARY ASYMPTCe,4ATIC PRIMARY ASYMPTOMATIC i cx~m~'rf.~ 0- Fig. 9. The change in plaque characteristics between the first and last duplex scans in primary asymptomatic and contralateral asymptomatic cases. Fig. 10. The change in degree of diameter reduction between the first and last duplex scans in primary asymptomattic and contralateral asymptomatic cases. group occurred in 31 patients (10.7%), but unheralded stroke occurred in 10 of these patients, for a total stroke rate of 3.5 %. New carotid territory symptoms occurred in 14 of 260 vessel territories at risk (5.4%) in the primary asymptomatic groups, significantly less than that seen in the contralateral asymptomatic group (p < 0.001). The incidence of unheralded stroke was also low, two of 260 vessel territories at risk (0.7%). Although the mean followup time is short in both populations, the group with contralateral asymptomatic arteries appears to have an increased risk of developing new symptoms compared with the primary asymptomatic group, highlighting the need to specify the nature and composition of symptom-free groups in studies of the asymptomatic carotid artery. Several studies have shown the ability of the duplex scan to differentiate between heterogeneous, echolucent plaque and denser, echogenic plaques. 4,~7 In an attempt to define the complexity of the lesions and identify progression in the plaque structural changes, we have divided plaque types into four basic categories depending on the ratio of echolucency to echogenicity. This classification avoids the need to categorically state whether ulceration or hemorrhage is present but rather reflects the echogenic properties of the plaque. Nonetheless, comparison with operating specimens has shown a strong association between echolucent images and plaques containing soft hemorrhage, lipid, or surface ulceration. 4 Because vascular laboratories vary with personnel and types of scanning equipment, it is important for each laboratory to verify its ability to classify plaque types or quantitate diameter reduction. We have continued to confirm our duplex results with ongoing pathologic studies correlating specimens obtained during sur- gery. is Plaque structure, and specifically intraplaque hemorrhage, has been confirmed in this study as being strongly associated with symptomatic carotid artery disease. On initial presentation most symptomatic carotid arteries were classified as types 1 and 2 (predominantly echolucent), whereas the asymptomatic carotid arteries were mainly the more echogenic types 3, 4, and 5. This study has shown an increased incidence in symptom-free patients of symptoms developing in the more echolucent types I and 2 plaques when both primary and contralateral groups are combined. However, the majority of symptom-free patients had the more echodense types 3 and 4 lesions, and the relative risk of symptoms developing in this group was small, certainly much less than that associated with surgery. This suggests a fairly benign course for such lesions and justifies an initial conservative, nonsurgical approach to such lesions. However, little is known about the ongoing process of plaque structural change. Intraplaque hem: orrhage has been postulated to cause an acute luminal diameter reduction, resulting in thrombosis or intimal ulceration with embolization of debris. 13 Repeated plaque hemorrhages are also important in increasing the complexity of the plaque. 12,19 In both of the symptom-free population groups the majority of plaques were predominately echodense, fibrous lesions that remained stable, undergoing no appreciable change in the follow-up period. This observation coincides with the low incidence of symptom onset in the study interval. Nonetheless a change in plaque type was noted in almost half of the vessels studied, with 33% of the contralateral and 18% of primary symptom-free patients' plaques becoming more echodense. This suggests that the fibrous con-

8 Volume 9 Number 4 April 1989 Role of plaque morphology and diameter reduction % 43% 20 p <0.02.< I--,z 10 9% 8% l ~ 6 NNNN 14% PLAQUE TYPE 35- ~: 30-".< e~ m 10. Q % NN % p <0.001 lt% 21% PERCENTAGE DIAMETER REDUCTION Fig. 11. The relative risk of developing new symptoms in the total symptom-free population in relation to the initial plaque type. Fig. 12. The relative risk of developing new symptoms in the total symptom-free population in relation to the initial degree of diameter reduction. tent of the plaque has increased and may reflect a healing and repair process occurring within the plaque. The elements of healing and repair are seen in pathologic specimens of carotid plaques removed at operation. Macrophage infiltrate, new vessel formation, fibrosis, and calcification are recognized within complex plaques and often follow recent intraplaque hemorrhage. 2'2 Twenty-nine percent of the primary and 25% of the contralateral vessel plaques increased in echolucency. This may reflect further intraplaque hemorrhages or the leaking of plasma substances into the plaque from the fragile new vessels associated with the healing and repair response. Tracking of new hemorrhages into a fractured soft plaque may also account for increases in echolucency. The process of plaque healing and repair that has been observed in isolated operative specimens may account for the dynamic changes observed in this study. The majority of patients who developed symptoms exhibited plaques that underwent a change in plaque category. However, this change was not necessarily into a more echolucent category. We suggest that plaque change itself, which reflects a dynamic process occurring within the plaque, may play a role in the development of new symptoms. Changes in plaque structure may reflect softening and degeneration with intimal breakdown with embolization of the fibrous cap or plaque contents. The healing and repair process leads to the formation of complex lesions containing fibrous, cartilaginous, calcific, and bony material that if exposed to the blood stream after ulceration may embolize causing irreversible occlusions of intracerebral vessels. On the other hand the end stage of the healing and repair process may account for the development of the dense fibrous and calcific plaques and explain the apparent reduction in the incidence of strokes occurring 1 year after the onset of cerebral ischemia. 21 We postulate that the more densely fibrous and calcific plaques are the most stable. Only long follow-up studies will help determine the importance of the actual change in plaque structure and the natural progression of the plaque. Degree of diameter reduction has also been proved important in the long-term outcome of patients with carotid artery disease7,22 In the study of patients with asymptomatic carotid bruits, Roederer et al? 2 found that patients who developed a stenosis of the carotid artery >80% were at 47% risk of stroke, TIAs, or occlusion of the carotid artery within 36 months. The incidence of highgrade lesions is low in our own study. Nonetheless, in the combined asymptomatic study population the relative risk of onset of carotid territory symptoms was 43% in patients with >75% diameter reduction. Unfortunately the majority of strokes in this study occurred in patients with <75% stenosis. This observation is consistent with the distribution of stenotic lesions in the patients with symptoms seen in this study, in which most plaques are of <75% stenosis, adding weight to the atheroembolic rather than flow-reduction theory as the leading cause of stroke in carotid artery disease. It has been suggested that patients who are seen de novo with a hemodynamically significant lesion fare worse than patients with a similar lesion opposite to an endarterectomized artery. 23 The reasons for this observation are unclear, but perhaps postoperative antiplatelet therapy and the closer follow-up of patients contribute. Our data indicate that the asymptomatic contralateral artery fares worse than the primary asymptomatic artery, especially if there is a >75% stenosis. An unheralded rate of stroke risk of 3.5% at a mean of 22 months would indicate reasonable susceptibility in this population group com-

9 556 Langsfeld, Gray-Weale, and Lusby Journalof VASCULAR SURGERY pared with a stroke rate of 0.7% for the primary asymptomatic vessels. All of our patients who have undergone endarterectomy have been treated with antiplatelet agents after surgery, perhaps contributing to some instability of the contralateral plaque structure. Conversely, these agents should have offered some protection from platelet emboli. 24"2s We feel that highly stenotic contralateral asymptomatic arteries should be treated more aggressively than those same lesions in the symptom-free population. In their angiographic studies, Javid et al. 26 showed that plaque progression can be paroxysmal. In our study the majority of plaques exhibited the same category of stenosis with the passage of time, but approximately one quarter increased in terms of diameter reduction. Twelve percent of the primary and 23% of the contralateral vessels were classified as being less stenotic on follow-up examination. An increase in luminal stenosis may be explained by further intraplaque hemorrhages, accumulation of plasma exudate from leaky new vessels, plaque fracture with hemorrhage tracking into the plaque, or incorporation of luminal thrombus into the plaque. However, our observation of plaque reduction requires careful consideration. This may represent the end result of healing and repair with scar contracture of the plaque. Alternatively, embolization of the atheromatous cap or evacuation of contents may also lead to a reduction in plaque size. Although we postulate that plaques may reduce in size, we look to further advances in ultrasound imaging systems to confirm these initial observations. The majority of patients who developed symptoms exhibited a change in stenosis, again suggesting that a change in the plaque structure is an important risk factor. Our study also confirms the high incidence of heterogeneous plaques in symptomatic carotid arteries. Overall the incidence of new symptoms in symptom-free patients would appear to be low. However, we have observed a significant increase in symptoms in patients with contralateral asymptomatic vessels compared with primary asymptomatic vessels. Although the degree of stenosis and heterogenicity would appear to be important predictors of stroke risk, the incidence of high-grade heterogeneous lesions in the symptom-free population is low. We have been disappointed in our ability to predict who among the <75% stenotic group are at increased risk. Although those vessels that remain unchanged carry the lowest risk and patients in that category can be reassured, almost half the population studied exhibited a change in their plaque, but only 7.4% of all patients developed symptoms. Our data would suggest that an aggressive approach is justified in those few patients with >75% heterogeneous plaques. We advocate a more conservative "wait and see" approach to the majority of symptom-free patients, recommending repeated scans in those patients with heterogeneous lesions and stressing the importance of seeking medical advice with the onset of suspicious symptoms. We recognize that longer follow-up studies are needed to better define the natural history of plaque progression and the importance of change in plaque structure. REFERENCES 1. Imparato AM, Riles TS, Mintzer R, Baumann FG. The importance of hemorrhage in the relationship between gross morphologic characteristics and cerebral symptoms in 376 carotid artery plaques. Ann Surg 1983;197: Lusby RJ, Ferrell LD, Ehrenfeld WF, Stoney RJ, Wylie EJ. Carotid plaque hemorrhage: its role in the production of cerebral ischemia. Arch Surg 1982; 117: Persson AV, Robichaux WT, Silverman M. The natural history of carotid plaque development. Arch Surg 1983;118: ReiUy LM, Lusby R}', Hughes L, Ferrell LD, Stoney RJ, Ehrenfeld WK. Carotid plaque histology using real-time ultrasonography. Am J Surg 1983;146: Chambers BR, Norris JW. The case against surgery for asymptomatic carotid stenosis. Stroke 1984;15: Johnson N, Burnham SJ, Flanigan P, et al. Carotid endarterectomy: a follow-up study of the contralateral nonoperated carotid artery. Ann Surg 1978;188: Roederer GO, Langlois YE, Lusiani L, et al. Natural history of carotid artery disease on the side contralateral to endarterectomy. J VAse SURG 1984;1: Kohler TR, Zierler RE, Strandness DE. Duplex scanning and spectral analysis. In: Moore WS, ed. Surgery for cerebrovascular disease. New York: Churchill Livingstone, 1987: Comerota AJ, Cranley JJ, Katz ML, et al. Real-time B-mode carotid imaging: a three-year multicenter experience. J VAsc SURG 1984;1: O'DonneU TF, Erdoes L, Mackey WC, et al. Comparison of B-mode ultrasound imaging and arteriography with pathologic findings at carotid endarterectomy. Arch Surg 1985; 120: Goodson SF, Flanigan DP, Bishara RA, et al. Can carotid duplex scanning supplant arteriography in patients with focal carotid territory symptoms? J VAsc SURG 1987;5: Lennihan L, Kupsky WJ, Mohr JP, et al. Lack of a~ssociation between carotid plaque hematoma and ischemic cerebral symptoms. Stroke 1987;18: Dyken ML, Pokras R. The performance of endarterectomy for disease of the extracranial arteries of the head. Stroke 1984;15: Moore WL, Beren G, Malone 1A, Goldstone J. Asymptomatic carotid stenosis: immediate and long-term reuslts after prophylactic endarterectomy. Am J Surg 1979;138: Podore PC, DeWeese JA, May AG, Rob CG. Asymptomatic contralateral carotid stenosis: a five-year follow-up

10 Volume 9 Number 4 April 1989 Role of plaque morphology and diameter reduction 557 study following carotid endarterectomy. Surgery 1980;88: Riles TS, Imparato AM, Mintzer R, Baumann FG. Comparison of results of bilateral and unilateral carotid endarterectomy five years after surgery. Surgery 1982;91: Johnson JM, Kennelly MM, Decesare D, Morgan S, Sparrow A. Natural history of asymptomatic carotid plaque. Arch Surg 1985;120: Gray-Weale AC, Graham J, Byrne K, Lusby RJ. B- Mode ultrasound plaque characteristics: comparison of preoperative appearance with carotid endarterectomy specimen pathology. Aust NZ J Surg (In press). 19. Ammar AD, Ernst RL, Lin JJ, Travers H. The influence of repeated carotid plaque hemorrhages on the production of cerebrovascular symptoms. J VASC SURG 1986;3: Lusby RJ. Lesions, dynamics and pathogenetic mechanisms responsible for ischemic events in the brain. In: Moore WS, ed. Surgery for cerebrovascular disease. New York: Churchill Livingstone, 1987: Kase CS, Wolf PA, Cupples LA. Relationship between transient ischemic attacks and stroke. In: Moore WS, ed. Surgery for cerebrovascular disease. New York: Churchill Livingstone, 1987: Roederer GO, Langlois YE, Jager KA, et al. The natural history of carotid arterial disease in asymptomatic patients with cervical bruits. Stroke 1984;15: Quinones-Baldrich WJ, Moore WS. Asymptomatic carotid stenosis--rationale for management. Arch Neurol 1985;42: Bousser MG, Eschwege E, Haguenau M, et al. "AICLA" controlled trial of aspirin and dipyridamole in the secondary prevention of atherothrombotic cerebral ischmia. Stroke 1983;14:5-14, 25. The Canadian Co-operative Study. A randomized trial of aspirin and sulfinpyrazone in threatened stroke. N Engl J Med 1978;229: Javid H, Ostermiller WE, Hengesh JW, et al. Natural history of carotid bifurcation atheroma. Surgery 1970;67:80-6. BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the JOURNAL OF VASCULAR SURGERY for 1989 are available to subscribers only. They may be purchased from the publisher at a cost of $52.00 ($66.00 international) for Vol. 9 (January to June) and Vol. 10 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, The C.V. Mosby Company, Westline Industrial Drive, St. Louis, MO , USA. In the United States call toll free: (800) , ext In Missouri call collect: (314) , ext Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.