270 STROKE V O L 17, No 2, MARCH-APRIL 1986

angioplasty: experience with 9 cadaver hearts. Am J Cardiol 50:1209-1211, 1982

4. Eldar M, Battler A, Neufeld HN, et al: Transluminal carbon diox-ide-laser catheter angioplasty for dissolution of atheroscleroticplaques. J Am Coll Cardiol 3: 135-137, 1984

5. Hass WK, Fields WS, North RR, Kricheff II, Chase NE, BauerRB: Joint study of extracranial occlusion. JAMA 203: 159—166,1968

6. Saunders MD, Young HF, Becker DP, et al: The use of the laser inneurological surgery. Surg Neurol 14: 1-10, 1980

7. Mehta J, Roberts A: Human vascular tissue produces thromboxaneas well as prostacyclin. Am J Physiol 244: R839-R844, 1983

8. Weichert W, Pauliks V, Breddin HK: Laser-induced thrombi in ratmesenteric vessels and antithrombotic drugs. Haemostasis 13:61-71, 1983

9. Treat MR, Weld FM, White JV, et al: Effect of CO2 laser on the

luminal surface of blood vessels in vivo. Lasers Surg Med 3:247-254, 1983

10. Day AL: Effects of the CO2 laser on peripheral nerve regeneration.Congress on Laser Neurosurgery, 1982

11. Silberbaurer K, Sinzinger M, Winter M: Prostacyclin productionby vascular smooth muscle cells. Lancet 1: 1356-1357, 1978

12. LeeG.IkedaRM, TheisJH.etal: Acute and chronic complicationsof laser angioplasty: vascular wall damage and formation of aneu-rysms in the atherosclerotic rabbit. Am J Cardiol S3: 290-293,1984

13. Grundfest WS, Litvack F, Forrester JS, et al: Laser ablation ofhuman atherosclerotic plaque without adjacent tissue injury. J AmColl Cardiol 5: 929-933, 1985

14. Linsker R, Srinivasan R, Wynne JJ, Alonso DR: Far-ultravioletlaser ablation of atherosclerotic lesions. Lasers Surg Med 4:201-206, 1984

Evaluation of a Scoring System for Extracranial CarotidAtherosclerosis Extent With B-Mode Ultrasound

JOHN R. CROUSE, M . D . , GARY H. HARPOLD, M . D . , * FREDERIC R. KAHL, M.D. , JAMES F. TOOLE, M.D. ,* AND

WILLIAM M. MCKINNEY, M.D.*

SUMMARY We have developed a scoring system to quantify extent of extracranial carotid artery athero-sclerosis using real-time ultrasound (B-mode). To evaluate repeatability of this scoring system we correlatedrepeat scores obtained within a short interval of one another (6 months) in 52 individuals. We comparedrepeatability of extent measurements with repeatability of a measure of severity (single most severe lesion).Correlations between first and second studies for severity were weak (r2 = 0.20) but significant (p < 0.001).Extent scores correlated much better (r2 = 0.77, p < 0.001).In another group of 22 patients we found thatthe extent of atherosclerosis decreased following endarterectomy. We used this method to determinechanges in extent of carotid atherosclerosis with age in two sets of individuals. One consisted of a cohort of22 patients who underwent repeat B-mode studies separated by V/t-3 years. This cohort demonstrated anincrease in carotid score with age (p < 0.05). In a second group of volunteers undergoing cardiac catheter-ization and B-mode evaluation of the carotid system, carotid scores could be compared in individuals withage differences that averaged 15 years. Extent of carotid atherosclerosis was significantly greater in olderindividuals (p < 0.01) and differences in extent with age were exaggerated in patients with coronary diseasecompared to coronary disease free controls.

Stroke Vol 17, No 2, 1986

METHODS FOR QUANTITATION OF SEVERITY*

of individual atherosclerotic plaques in the carotid ar-

terial system are of great clinical importance for fol-

lowing patients longitudinally. A single strategically

placed lesion may be the source of cerebral emboli or

the cause of a hemodynamic reduction of flow and

pressure.M0 However, little is known about the signifi-

cance of the extent of carotid artery disease. Because

•In this discussion severity refers to the axial thickness (by B-mode)of a single luminal obstruction while extern refers to the sum (or aver-age) of axial thicknesses for a number of such obstructions.

From the Departments of Medicine, and Neurology,* Wake ForestUniversity Medical Center, The Bowman Gray School of Medicine,Winston-Salem, North Carolina 27103.

Supported by Specialized Center of Research in ArteriosclerosisGrant HL-14164 from the NHLBI and Cerebrovascular Research CenterGrant NS06655 from the NINCDS.

Address correspondence to: John R. Crouse M.D., Department ofMedicine, Bowman Gray School of Medicine, 300 S. Hawthorne Rd ,Winston-Salem, North Carolina 27103.

Received February 26, 1985; revision #1 accepted June 12, 1985.

extensive carotid atherosclerosis that results in modest

lumenal stenosis may have no recognized clinical cor-

relate, and because evaluation of extent is more diffi-

cult than evaluation of severity of individual lesions,

extent is only rarely studied in the clinical setting."'l2

However, by analogy with coronary atherosclerosis,

statistical strength may be gained by considering

atherosclerosis extent as well as severity as a response

variable.13 Coronary atherosclerosis extent may be

evaluated by counting the number of coronary arteries

with stenosis of 50% or more14"18 or by more sophisti-

cated scoring systems such as that suggested by the

American Heart Association Grading Committee" or

others20"23 that include lesions at multiple sites in the

coronary system.

Using evaluation systems for coronary artery

atherosclerosis as a model we have developed a scor-

ing system for extent of extracranial carotid artery dis-

ease. We have employed this system to evaluate

change in extent with age and in patients with and

without coronary artery atherosclerosis and change in

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

EXTENT OF CAROTID ATHEROSCLEROSIS/Crourc et al 271

extent following carotid endarterectomy. The datashow the method to be repeatable and of biologicalrelevance.

MethodsB-Mode Ultrascan

Studies began in March, 1981. Two B-mode ultra-scan instruments were utilized, an HRL (Horizon'sResearch Laboratories, Inc.) high resolution 5 MHZ

small part scanner, and a Biosound compact 7 MHZ

real time imager. Both machines were equipped withelectronic calipers for measurement of axial thickness(mm) of all echogenic lesions and of minimal residuallumen. Scan images, sonographer comments, andcali-per measurements were recorded using video tape andinterpreted by two readers (G.H. and W.M.McK.).

Patients were examined in the seated position andeach carotid system was imaged in anterior oblique,lateral, and posterior oblique planes. Each scan of thecommon carotid artery began just above the clavicleand was moved cephalad through the bifurcation andalong both the internal and external branches as fardistally as possible. The sonographer consistentlyaimed for the clearest image of the far wall of thecarotid artery so as to be certain that the chord ofgreatest length was achieved, and therefore that theimage represented the true diameter of the vessel.Where echogenic lesions were identified, measure-ments were made using the view that displayed thegreatest axial thickness of the lesion (usually the poste-rior oblique view). An image representative of thequality of the studies is reproduced in figure 1.

Follow-up scans were performed according to thesame procedures. Sonographers performing the secondexam were aware of the location of the most severelesion found in the previous study but not of its axialthickness nor of the extent of disease. Readers were"blinded" as to results of the first exam.

For this study readings made at four sites are consid-ered: near and far wall of the common carotid >5 mm

FIGURE 1. Representative B-mode ultrasound image of thecarotid bifurcation showing a plaque measuring 3.3 mm on thefar wall of the right internal carotid. EC = External Carotid;IC = Internal Carotid; Bif = Bifurcation.

from the flow divider; near and far wall < 5 mm belowthe flow divider; near and far wall of the internal carot-id <5 mm above the flow divider; and, near and farwall of the internal carotid artery > 5 mm above theflow divider. These sites are identified as "low com-mon carotid", "below bifurcation", "above bifurca-tion", and "high internal carotid" respectively. To ob-tain a summary score for the left and right carotidarteries (separately), the maximal plaque thickness onthe near and far wall at each of these four sites issummed for each side. The analysis does not take intoconsideration the length of individual plaques; it israther based on evaluations of multiple plaques. Thecarotid summary score is derived by summing thescores of the left and right carotid systems.

Ultrasound exams showing 100% occlusion wereexcluded from study as were those of patients who hada high carotid bifurcation which could not be com-pletely imaged.

PatientsFour groups of patients were evaluated. In three we

evaluated test-retest B-mode studies. Internal carotidstatus with regard to endarterectomy or no endarterec-tomy was established by chart review. The first groupof 52 patients was used to test the repeatability of themethod through comparison of repeat clinical studiesseparated by no more than 6 months. None of thesepatients underwent endarterectomy between the firstand the second study.

The second group of 22 patients was used to test theability of the method to distinguish changes in carotidatherosclerosis following endarterectomy. Repeatscores in this group were separated by varying inter-vals from 2 to 30 months (mean = 12 months), and allsubjects underwent endarterectomy between the firstand second study.

The third group of 22 individuals was used to testwhether an age effect could be demonstrated in pairedtests separated by more than one year (average 2'/3years; range 1 '/4—3 years). None of these patients hadundergone endarterectomy.

In the fourth group of volunteers the possible effectof age was considered by comparing scores in olderindividuals with those in younger individuals. Each ofthe 323 individuals in this group had a single ultra-sound evaluation. They volunteered for an ongoingultrasound study of carotid atherosclerosis in patientswith and without angiographically demonstrable coro-nary artery disease. Volunteers underwent randomsampling stratified according to age (older than oryounger than 50), gender, and coronary status in orderto access equal numbers of volunteers in each of theseeight categories at equal rates.

Statistical AnalysisData sets were analyzed by paired or unpaired "t"

testing and by Pearson product-moment correlation co-efficients.24 Estimates of confidence intervals of differ-ences were carried out as described.15

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

272 STROKE VOL 17, No 2, MARCH-APRIL 1986

First Study

= 0.4 x+1 .5 (p< .001; rz=0.20)FIGURE 2. Scatter diagram of maximum axial thickness ofindividual most severe lesion; two studies performed on thesame individuals separated by s6 months and with no interven-ing endarterectomy. O = Studies in which the site of the lesionof greatest thickness in the first study = that in the secondstudy. A = Studies in which the site of the lesion of greatestthickness in the first study was not the same as in the secondstudy, and the axial thickness of the lesion in the first study was> that in the second study. V = Studies in which the site of thelesion of greatest thickness in the first study was not the same asin the second study, and the axial thickness of the lesion in thefirst study was < that in the second study.

ResultsFifty-two patients had technically adequate repeat

studies separated by 6 months or less. Three of thesepatients had undergone an endarterectomy prior totheir first B-mode evaluation. Twenty-three pairs ofrepeat studies were performed wholely with the HRLinstrument, 23 were performed with the Biosound in-strument, and 6 pairs were "split": the first study (earli-

er) was performed with the HRL and the second (later)with the Biosound instrument.

We were able to develop correlation coefficients fora number of measurements in these repeat studies.First, we correlated the axial thickness of the lesionwith the greatest axial thickness in each study with thatof its "pair" (same site) in the study performed withinthe subsequent (or preceding) six months. For com-parisons we chose the lesion of greatest axial thick-ness, whether it appeared in the first or the secondstudy, and related it to the lesion of the same site in thecorresponding (earlier or later) study of the same pa-tient . Figure 2 is a scatter diagram of these repeats. Thesite of the lesion with the greatest axial thickness in thefirst study was the same as that in the second study in31/52 repeats. The lesion with greatest axial thicknesswas found at a different site in the second study for21/52 repeats. The most significant lesion in the firststudy was of greater axial thickness than that in thesecond study for 12 of these 21 repeats, while in theremaining 9 the situation was reversed. Although therewas a statistically significant correlation between theaxial thickness found at the first and second readingsfor the most significant lesion (p < .001), the correla-tion was weak (r2 = 0.20).

The correlation between the first and second readingimproved when the sum of the axial thicknesses oflesions at the near and far wall found at the first andsecond readings were compared for each of the 4 spe-cific sites described under methods (with one excep-tion — readings in the high internal carotid on the leftside) (table 1). An additional marked improvementwas achieved by combining (summing) readings of thenear and far wall immediately above and below thebifurcation, and yet slightly higher correlations wereobtained when all observations on either the right orthe left side were combined (left summary score, rightsummary score). The best correlation of repeat mea-surements was obtained when a summary score wasdeveloped that included (summed) lesions at all 4 sitesin the common and internal carotid on both the left andthe right side (r2 = 0.74). For clinical use a score of 24was arbitrarily set as a maximum, and all studies thatexceeded this were given a score of 24. This wasnecessary in 4/52 repeat studies. With this modifica-

TABLE 1 Correlation Coefficients for First vs Second Reading

Lowcommon

.60

Right side

Below Abovebifurcation bifurcation

.51 .59

.81*

.82t

Highinternal

.53

.86*

Lowcommon

.67

Left side

Below Abovebifurcation bifurcation

.50 51

.72*

.75t

Highinternal

-0.06

•Correlation coefficient (first vs second study) for sum of lesions immediately above and below the bifurcation (flowdivider).

tCorrelation coefficient for sum of all lesions on left and (separately) right sides.^Correlation coefficient for "summary score" — sum of all lesions on left plus right side.

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

EXTENT OF CAROTID ATHEROSCLEROSIS/Ooiw et al 273

tion the correlation coefficient for the first and secondcarotid summary scores was 0.88 (r2 = 0.78) (fig. 3).

This approach was used to evaluate the relation be-tween atherosclerosis in the left and right carotid arter-ies. For this the left summary scores and (separately)the right summary scores for the first and second stud-ies were averaged for 49 of the 52 patients describedabove. Eliminated from this analysis were three pa-tients who had undergone unilateral endarterectomiesprior to the first of the 2 B-mode studies. The left andright scores correlated with one another (r2 = .54; p <0.01). (1-/3) testing of differences in paired valuesshowed that we could be 95% confident that the meansof the left and right scores differed by less than ±0.96mm (mean ± S.D. left = 6.14 ± 3.74; right = 6.04± 3.74). Similar calculations applied to the sum of thearea above and below the bifurcation showed that leftcorrelated with right bifurcation (r2 = .51; p < 0.01)and we could be 95% confident that the means of theleft and right scores at that location differed by lessthan ±.79 mm [mean ± S.D. left (above + below)bifurcation = 5.11 ± 3.25; right (above + below)bifurcation = 4.88 ± 2.96].

We reasoned that, if it were of biologic relevance,the summary score for the left or right side shoulddiminish following endarterectomy while the score forthe control artery not subjected to endarterectomyshould be unaffected. We could identify 22 patientswho had an endarterectomy (or bilateral endarterecto-mies) in the interval between repeat B-mode studiesseparated by from 2 to 30 months (mean = 12months). Of these 22, 8 had a left endarterectomy, 11had a right endarterectomy, and 3 had bilateral endar-terectomies. One patient had occlusion of the arterycontralateral to that which was operated on, and this

25 i o o

10 15

First Study

20 25

=0.9 x + 1.0 (p< .0001; r2 = 0.78)FIGURE 3. Scatter diagram of the summary carotid score forB-mode studies separated by ^6 months. Each (O) representsthe summary score for a single individual. Maximum score setat 24.

one side of this individual was not considered in dataanalysis. All together we were able to compare 24arteries that had been subjected to endarterectomy with17 that had not. For this comparison we comparedcontrol with operated arteries with regard to the scorederived by summing of the area above and below thebifurcation (since this was the area that had been sub-jected to endarterectomy). Twenty of the 24 repeatstudies of arteries that had been subjected to endarter-ectomy had a lower score than the score obtained at theoriginal scan. For control arteries, 8 of 17 had a lowerscore. In order to quantitate this we compared themean decrement in score (score before — score aftersurgery) of arteries that had been operated on (3.6 ±3.52) with the mean decrement in score for contrala-teral control arteries (.39 ± 2.27). This difference wasstatistically significant (p < 0.005). Since the controlarteries were less likely than those that had been oper-ated on to have severe and extensive atherosclerosis atthe time of the initial scan, and thus less likely to showa decrement in the carotid score, we also compareddecrement in score for control vs endarterectomy arter-ies matched for extent of disease on the first study. Wewere able to compare the decrement in 10 such match-es with an initial score ^5 .0 . The mean decrement forthe side that had been operated on was 5.08 ± 3.85;for the control side the decrement was 1.23 ± 2.22(p < 0.01).

We also reasoned that, with age, the extent of carot-id atherosclerosis should increase. We tested the abili-ty of B-mode to reflect this in two ways. First, wecompared repeat B-mode examinations separated by1 Vi to 3 years (mean = 2'/4 years) in 22 volunteers whohad not undergone an endarterectomy in the interven-ing years. In this test-retest comparison, we found thatthe B-mode score of individuals increased with age.Mean score at the first exam was 10.35 ± 6.4 and atthe second exam it was 12.8 ± 8.2 (p < 0.05). Itappeared that the statistical significance of this differ-ence depended in part on the coronary status of thepatients. When patients were divided into those withovert history of coronary disease (post coronary bypassgraft, history of myocardial infarction, positive tread-mill, or angina) (n = 5) and those without such history(n = 17), the difference with age only appeared inthose with a history of coronary disease (first exam =13.74 ± 5.1; second exam = 21.28 ± 5.9, p < 0.05)and not in those with no coronary disease history (firstexam = 9.35 ± 6.5, scondexam = 10.3 ± 7.1, p =0.35). The mean age of these patients was 65 ± 8years (for coronary disease patients mean age was 68.6years; for patients with no overt history of coronarydisease mean age was 63.9 years).

In addition to this longitudinal comparison of B-mode scores, we also took advantage of data beinggenerated in an ongoing study of B-mode examina-tions in patients undergoing cardiac catheterization.When we evaluated the mean summary carotid arteryscores in our coronary artery disease free patients, wefound that the mean summary score for older volun-teers was greater than that for younger. Carotid artery

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

274 STROKE VOL 17, No 2, MARCH-APRIL 1986

TABLE 2A Volunteers Free of Coronary Disease

Gender/age

males =s 50

males > 50

females s 50

females > 50

n

31

36

35

48

TABLE 2B Volunteers

Gender/age

males £ 50

males > 50

females s 50

females > 50

n

49

48

33

49

Meanage

42

59

42

60

Meancarotidartery

summaryscore

2.5 ±(2.4)

4.8 ±(4.9)

1.8±(1.9)

5.1 ±(5.3)

with Coronary Disease

Meanage

44

60

44

63

Meancarotidartery

summaryscore

3.5 ±(3.6)

9.4 ±(6.2)

4.2 ±(4.0)

11.0 ±(6.5)

p = 0.010

p < 0.001

p < 0.0001

p < 0.0001

summary scores in all categories for individuals withcoronary disease were 1.4-2.3 times greater than thosefor individuals free of coronary disease and scoreswere significantly higher in older than in younger vol-unteers (table 2).

DiscussionAlthough a number of authors have reported the

correlation between severity of carotid atherosclerosisand clinical events,1"10 little has been written about thecorrelation of clinical phenomena with extent of carot-id atherosclerosis. This is in contrast to the situationwith coronary artery disease where, in studies that usecoronary angiography to describe disease, coronarydisease extent has frequently been used as a responsevariable.14"23 In general, studies that take into accountthe extent of coronary atherosclerosis are more suc-cessful at uncovering a relation between risk factorsand extent of disease than studies that use disease pres-ence or absence as a disease descriptor.13 For prognosisthe number of coronary vessels involved with athero-sclerosis is one of the most important factors.26 Use ofextent of atherosclerosis as the response variable pro-vides the opportunity to relate risk factors to disease ina dose-response fashion — the higher the level of therisk factor, the greater the extent of disease. In addi-tion, there is an inherent methodologic advantage inconsidering potential for disease at several sites. Asdemonstrated in this paper, there appear to be severemethodologic limitations in utilizing ordinary B-modescanning of any single lesion to develop a single scorefor any individual. The correlation between repeatmeasurements, while statistically significant is weak(r2 = 0.20). With increasing number of lesions used todevelop the score, the reproducibility increases. Thislikely occurs because carotid atherosclerosis tends notto be limited to one single site, and the average (orsum) is a better representation of the overall process.

Lost in this analysis is the potential clinical importanceof severe atherosclerosis at any site. We acknowledgethat more careful measurements of individual lesionsmight result in better reproducibility for these analysesas well.

Previous studies have shown correlations betweenindividual atherosclerotic lesions in the extracranialcarotid arteries identified by B-mode and by carotidangiography27 and further comparisons may be antici-pated to result from the recently completed B-scanassessment study.28 No attempts were made to corre-late B-scan readings with carotid angiography in thepresent studies since not all patients who underwentnon-invasive testing had angiograms. Furthermore, noeffort was made in the present study to evaluate lumenstenosis. The B-mode partly evaluates a different enti-ty from that evaluated by angiography, since the focusof the angiographic assessment is the arterial lumen,while that of B-mode is the arterial wall. It is thereforetheoretically possible to have atherosclerosis that in-volves the wall, but, because of arterial dilatation,does not narrow the lumen. This would appear as alesion to the ultrasonographer, but not to the angiog-rapher; similarly a vasospastic lesion would appear as alesion to the angiographer but not the ultrasonog-rapher. These differences in measurement modalitiesmay explain in part the less than perfect correlationsbetween ultrasound and carotid angiography that aredocumented in the literature. The present effort dem-onstrates that an extent score can be developed com-bining measurements of lesions at several sites that isreproducible and that has biologic relevance. The re-producibility of the summary score for two sites aboveand below the bifurcation of the carotid artery is almostas good as that for the entire carotid system, and, foreconomy of effort in investigational studies of carotidatherosclerosis, these two sites might be sufficient;little additional information is gained by evaluating theentire internal and common carotid bed. It is likely thatmeasurements undertaken as part of a prospectivestudy would yield yet more reproducible results.

The comparisons of atherosclerosis between the leftand the right carotid system demonstrate a reasonablyhigh correlation between the two sides (r2 = 0.54). Weobserved this correlation despite the fact that many ofthese patients were referred for focal neurologic signsor unilateral bruits, and thus represent a select popula-tion. It is reasonable to believe that correlations be-tween the left and right side for a random sample ofvolunteers would be greater. Not surprisingly wefound a decrease in the carotid score following endar-terectomy. Although it was thus possible to demon-strate clinical "regression" of atherosclerosis resultingfrom endarterectomy, an increase in extent of diseasewas more difficult to demonstrate. There was a sugges-tion of an increase in atherosclerosis extent in the re-peat studies performed after an interval of severalyears, however, the differences for patients free ofcoronary disease symptoms were not statistically sig-nificant. Our studies of differences in extent of athero-sclerosis in patients undergoing coronary angiography

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

EXTENT OF CAROTID ATHEROSCLEROSIS/Ooujf et al 275

may bear on this since it is evident that an increase inextent of atherosclerosis with age occurs at a slow rate,at least for patients free of coronary disease. Thesevolunteers differed from the patients studied with thetest-retest protocol in that their coronary status wascarefully denned; in addition, their extent of carotidatherosclerosis was much less because they were notstudied in order to evaluate cerebrovascular symp-toms. Comparisons of measurements in older withyounger individuals are not valid indices of progres-sion since they do not represent sequential measure-ments of individual volunteers. However, in a generalsense the observations likely approximate those thatwould be obtained from sequential sampling of asymp-tomatic individuals.

The reasons for individual differences in rate ofchange of carotid atherosclerosis and for the greaterage-related rate of increase in carotid artery score forpatients with coronary disease is unclear, but mostlikely relate to differences in risk factor profiles thatdifferentiate individuals with from those without coro-nary disease. These issues will be clarified in subse-quent publications.

AcknowledgmentsWe wish to acknowledge the expert assistance of Mrs. Bobbie Lind-

say who aided in preparation of this manuscript, and of Cathy Nunn whowas responsible for performance of B-mode examinations on patients.

References1. Bamett HJM: Progress towards stroke prevention: Robert Warten-

berg Lecture. Neural 30: 1212-1225, 19802. Harrison MJG: Pathogenesis. In Transient Ischaemic Attacks. Ed

Warlow C and Moms PJ. Marcel Dekker, N.Y. 19823. Ruff RL, Talman WT, Petito F: Transient ischaemic attacks associ-

ated with hypotension in hypertensive patients with carotid arterystenosis. Stroke 12: 353-355, 1981

4 Julian OC, Dye WS, JavidH, Hunter JA: Ulcerative lesions of thecarotid artery bifurcation. Arch Surg 86: 803-809, 1963

5. Moore WS, Hall AD: Ulcerated atheroma of the carotid artery. Acause of transient cerebral ischaemia. Amer J Surg 116: 237-241,1968

6 Harrison MJG, Marshall J: Prognostic significance of severity ofcarotid atheroma in early manifestations of cerebrovascular dis-ease. Stroke 13: 567-469, 1982

7. Ziegler DK, Hassanein RS: Prognosis in patients with transientischaemic attacks Stroke 4: 666-673, 1973

8. Busuml RW, Baker JD, Davidson RK, Machleder HI: Carotidartery stenosis — haemodynamic significance and clinical course JAmer Med Assoc 245: 1438-1441, 1981

9.Moore WS, Boren C, MaJone JM et al: Natural history of nonsteno-tic asymptomatic ulcerative lesions of the carotid artery. Arch Surg113: 1352-1358, 1978

10. Podore PC, De Weese JA, May AG, Rob CG: Asymptomatic

contralateral carotid artery stenosis: a five year follow-up studyfollowing carotid endarterectomy. Surgery 88: 748-750, 1980

11. Hennerici M, Aulich A, SandmannW, Freund HJ: Incidence ofasymptomatic extracranial arterial disease. Stroke 12: 750-758,1981

12. Candelise L, Bianchi F, Galligoni F et al: Italian multicenter studyon reversible cerebral ischemic attacks: III — Influence of age andrisk factors on cerebrovascular atherosclerosis. Stroke 15:379-382, 1984

13. Pearson TA: Coronary artenography in the study of the epidemiol-ogy of coronary artery disease. Epidemiol Rev 6: 140-166, 1984

14. Pearson TA, Bulkley BH, Achuff SC, Kwiterovich PO, Gordis L:The association of low levels of HDL cholesterol and arteriographi-cally defined coronary artery disease. Amer J Epidemiol 109:285-295, 1979

15. Murray RG, Tweddel A, Third JLHC, Hutton I, Hillis WS, Lor-imer AR, Lawrie TDV: Relation between extent of coronary arterydisease and severity of hyperlipoproteinaemia. Brit Heart J 37:1205-1209, 1975

16. Barbonak JJ, Rimm AA, Anderson AJ, Tristani FE, Walker JA,Flemma RJ: Coronary artery occlusion and blood lipids. AmerHeart J 87: 716-721, 1974

17. Bloch A, Dinsmore RE, Lees RS: Coronary arteriographic findingsin type-II and type-lV hyperlipoproteinaemia. Lancet 1: 928—930,1976

18. Scott DW, Gotto AM, Cole JS, Gorry GA: Plasma lipid as collater-al risk factors in coronary artery disease — A study of 371 maleswith chest pain. J Chron Dis 31: 337-345, 1978

19 Austen WG, Edwards JE, Frye RL et al: AHA Committee Report.A reporting system on patients evaluated for coronary artery dis-ease. Report of the Ad Hoc Committee for Grading of CoronaryArtery Disease, Council on Cardiovascular Surgery, AmericanHeart Association. Circulation 51: 7^0 , 1975

20 Nitter-Hauge S, Enge I: Relation between blood lipid levels andangiographically evaluated obstructions in coronary arteries. BritHeart J 35: 791-795, 1973

21. Balcon R, Cattell MR, Stone DL, Fuerlicht JA: A computer gener-ated index for the assessment of coronary angiography. Acta MedScand 615 (Suppl): 25-30, 1978

22 Miller NE, Hammett F, Saltissi S, Rao S, van Zeller H, Coltart J,Lewis B: Relation of angiographically defined coronary artery dis-ease to plasma lipoprotein subfractions and apolipoproteins. BritMed J 282: 1741-1744, 1981

23. Jenkins PJ, Harper TW, Nestel PJ: Severity of coronary atheroscle-rosis related to lipoprotein concentration Bnt Med J 2: 388-391,1978

24. Snedecor GW, Cochran WG Statistical Methods, 6th ed. IowaState University Press, Ames, IA.

25 Lapin L: Statistics, Meaning and Method. Harcourt Brace Jovano-vich, New York, 267-268, 1975

26. European Coronary Surgery Study Group: Long-term results ofprospective randomised study of coronary artery bypass surgery instable angina pectoris. Lancet 2: 1173-1180, 1982

27. Jones AM, Biller J, Cowley AR, Howard G, McKinney WM,Toole JP: Extracranial carotid artery arteriosclerosis. Diagnosiswith continuous-wave doppler and real-time ultrasound studies.Arch Neural 39: 393-394, 1982

28. Toole JF, Berson A: Multicenter trial for assessment of B-modeultrasound imaging. In Clinical Diagnosis of Atherosclerosis, Ed.Bond MG, Insull W Jr, Glagov S, Chandler AB, Comhill JF.Springer-Verlag, New York, 389-397, 1983

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

J R Crouse, G H Harpold, F R Kahl, J F Toole and W M McKinneyultrasound.

Evaluation of a scoring system for extracranial carotid atherosclerosis extent with B-mode

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 1986 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.17.2.270

1986;17:270-275Stroke. 

http://stroke.ahajournals.org/content/17/2/270World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Stroke Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from