Intravascular Ultrasound Study of Patterns of Calcium in Ruptured Coronary Plaques

Pmfifilfttslelutcw

Cr

0d

Intravascular Ultrasound Study of Patterns of Calciumin Ruptured Coronary Plaques

Kenichi Fujii, MD, Stéphane G. Carlier, MD, PhD*, Gary S. Mintz, MD,Hideo Takebayashi, MD, Takenori Yasuda, MD, Ricardo A. Costa, MD,Issam Moussa, MD, George Dangas, MD, PhD, Roxana Mehran, MD,

Alexandra J. Lansky, MD, Edward M. Kreps, MD, Michael Collins, MD,Gregg W. Stone, MD, Jeffrey W. Moses, MD, and Martin B. Leon, MD

Coronary calcium is intimately associated with coronary atherosclerotic plaque de-velopment, although it is controversial as to whether coronary calcium is associatedwith plaque instability. We analyzed 101 IVUS-detected ruptured plaques and com-pared them with 101 computer-matched control plaques without evidence of plaquerupture. The arc of calcium was measured every 0.5 mm within 10-mm-long segmentsthat spanned the minimum lumen cross-sectional area, and the number and length ofcalcium deposits were assessed. Ruptured plaques had a significantly larger numberof individual calcium deposits than control plaques (3.5 � 1.7 vs 1.8 � 1.1, p <0.001).However, the arc of the largest calcium deposit was smaller and the length of thelargest calcium deposit in each plaque was shorter in ruptured plaques comparedwith control plaques (67.3° � 41.4° vs 114.9° � 77.4°, p <0.001, and 1.6 � 1.3 vs 4.0 �2.7 mm, p <0.001, respectively). There was no difference in the number of superficialcalcium deposits between the 2 groups, although ruptured plaques had significantlysmaller arcs of superficial calcium compared with control plaques (56.2° � 35.5° vs95.8° � 65.2°, p <0.001). Conversely, the number of deep calcium deposits wassignificantly larger in ruptured plaques than in control plaques (1.8 � 1.4 vs 0.3 � 0.6,p <0.001), although the arc of deep calcium was similar in the 2 groups. Rupturedplaques had quantitatively less calcium, especially superficial calcium, but a largernumber of small calcium deposits, especially deep calcium deposits. In conclusion,ruptured plaques are associated with a larger number of calcium deposits within anarc of <90°, a larger number of deep calcium deposits, and a remodeling

sp

M

DiNIoniwgccwitd

laque rupture with subsequent thrombus formation is theost frequent cause of acute coronary syndrome.1–3 Thinbrous caps and large lipid pools increase atheroscleroticbrous cap stresses and are common features of ruptured

esions. Although coronary calcification is a marker foruture acute coronary syndrome events,4,5 the relation be-ween coronary calcification and plaque instability is con-roversial. Previous histopathologic studies have demon-trated that lesions with fatal acute coronary syndrome areess likely to be calcified.6,7 However, Burke et al5 and Farbt al8 reported that calcium is common in plaque rupture andess common in plaque erosion. Conversely, intravascularltrasound (IVUS) studies have been associated with quan-itatively less calcium,9,10 although the patterns of lesionalcium and the effect of calcium location have not beenell studied. For this reason, we undertook the present

Cardiovascular Research Foundation and Columbia University Medicalenter, New York, New York. Manuscript received January 3, 2005;

evised manuscript received and accepted March 22, 2005.* Corresponding author. Tel.: 212-851-9371; fax: 212-851-9330.

sE-mail address: [email protected] (S.G. Carlier).

002-9149/05/$ – see front matter © 2005 Elsevier Inc. All rights reserved.oi:10.1016/j.amjcard.2005.03.074

tudy to evaluate the pattern of lesion calcium in rupturedlaques.

ethodsPatient and lesion populations: All IVUS images from

ecember 1999 to December 2003 were reviewed by 1nvestigator who was unaware of patients’ clinical findings.inety-nine patients (106 lesions) who had high-quality

VUS images that were obtained before percutaneous cor-nary intervention and showed plaque rupture in a de novoative coronary lesion were identified. Clinical demograph-cs were obtained by hospital chart review. Unstable anginaas classified as new-onset severe angina, accelerated an-ina, or angina at rest. Recent myocardial infarction oc-urred within 4 weeks. Stable angina was defined as nohange in frequency, duration, or intensity of symptomsithin 4 weeks. Asymptomatic patients were typically stud-

ed because of positive stress tests in the absence of symp-oms. Patients who had unstable angina or recent myocar-ial infarction were categorized as having acute coronary

www.AJConline.org

dttpHa

dtt

msssio((Ta

rwvn

citmaLq

wwcutcaatdf

sIwu

Fto

353Coronary Artery Disease/Ruptured Plaque and Coronary Calcium

rome were categorized as having stable angina or asymp-omatic ischemia. Hypercholesterolemia was defined as aotal cholesterol level �240 mg/dl or medication use. Hy-ertension was defined as systolic blood pressure �140 mmg, diastolic blood pressure �90 mm Hg, or use of an

ntihypertensive drug.Ruptured plaques required the agreement of 2 indepen-

ent experienced observers. The rate of agreement for rup-ured plaque was 98% (104 of 106). The 2 cases of ques-ionable ruptured plaque were excluded from our study.

We compared patients who had ruptured plaque with aatched group of patients from the Cardiovascular Re-

earch Foundation database of 2,214 de novo native athero-clerotic plaques from patients who had nonacute coronaryyndrome and no IVUS evidence of plaque rupture. Match-ng criteria included (1) identical gender, (2) identical cor-nary artery segment, (3) similar minimum lumen area, and4) similar reference vessel size. One hundred one lesions97 patients) met the matching criteria for the control group.hus, 101 ruptured plaques and 101 control plaques weressessed in this study.

Angiographic analysis: Preprocedural angiograms wereeviewed by a core angiographic laboratory. Calcificationas identified as readily apparent radiopacities within theascular wall at the site of the stenosis and was classified as

igure 1. Typical IVUS images of a ruptured plaque. Top, longitudinal IVhe largest intraplaque cavity (middle image slice) indicates 2 superficial cf a cavity (asterisk).

one, mild, moderate (radiopacities noted only during the t

ardiac cycle before contrast injection), or severe (radiopac-ties noted without cardiac motion before contrast injectionhat generally compromised the 2 sides of the arterial lu-en). Quantitative angiographic analysis used a computer-

ssisted, automated edge detection algorithm (CMS, MEDIS,eiden, The Netherlands) using standard qualitative anduantitative definitions and measurements.

IVUS imaging protocol: Written informed consentas obtained before all procedures. All IVUS studiesere performed before any intervention and after intra-

oronary administration of 100 to 200 �g of nitroglycerinsing a commercially available system (Boston Scien-ific, San Jose, California). The 30- or 40-MHz IVUSatheter was advanced �10 mm beyond the lesion, andn imaging run (using an automated transducer pullbackt 0.5 mm/s) was performed to a point �10 mm proximalo the lesion. IVUS imaging was recorded only during trans-ucer pullback onto 0.5-in high-resolution s-VHS videotapeor off-line analysis.

IVUS analysis: Qualitative and quantitative IVUS mea-urements were made according to definitions of previousVUS studies, especially for ruptured plaque.11–13 Each caseas analyzed by 2 experienced IVUS observers who werenfamiliar with the clinical data. A ruptured plaque con-

age shows scattered calcification (arrowheads). Bottom, image slice withdeposits (arrowheads) and 1 deep calcium deposit (arrow) at the bottom

US imalcium

ained a cavity that communicated with the lumen with an

ofiwps

Cdt

esmamdww

das4a(

cwdwWp

fwo

srslb(nCwa(merC

wCpdMac

TB

AMAHHDSPPPAT

RMDLC

TI

V

P

R

M

C

354 The American Journal of Cardiology (www.AJConline.org)

verlying residual fibrous cap fragment (Figure 1).12,13 Assure without a cavity communicating with the true lumenas not included in the analysis.11–13 Lesion plaque com-osition was assessed visually at the minimum lumen cross-

able 1aseline patient characteristics

RupturedPlaques(n � 101)

ControlPlaques(n � 101)

p Value

ge (yrs) 61.4 � 11.0 62.3 � 9.9 0.6en 92 (91%) 92 (91%) 1.0cute coronary syndrome 66 (65%) 0 (0%) �0.001ypertension 62 (61%) 65 (64%) 0.8ypercholesterolemia 76 (75%) 62 (61%) 0.03iabetes mellitus 36 (36%) 34 (34%) 0.8moker 66 (65%) 58 (57%) 0.3revious coronary angioplasty 27 (27%) 36 (36%) 0.2revious coronary bypass 5 (5%) 10 (10%) 0.2revious myocardial infarction 16 (16%) 28 (28%) 0.1ngiographic anaysisarget coronary artery 0.9Left anterior descending 48 (47%) 46 (45%)Right 40 (40%) 41 (41%)Left circumflex 11 (11%) 13 (13%)Left main 2 (2%) 1 (1%)eference diameter (mm) 3.45 � 0.56 3.43 � 0.67 0.8inimal lumen diameter (mm) 1.22 � 0.52 1.35 � 0.60 0.2iameter stenosis (%) 65 � 12 61 � 14 0.1esion length (mm) 12.9 � 5.2 13.6 � 6.7 0.6oronary calcium 0.1None 67 (66%) 59 (58%)Mild 20 (20%) 17 (17%)Moderate 13 (13%) 19 (19%)Severe 1 (1%) 6 (6%)

Data are presented as means � 1 SD or numbers (percentages).

able 2ntravascular ultrasound analysis

ariable RupturedPlaques(n � 101)

ControlPlaques(n � 101)

p Value

laque morphology �0.001Soft 54 (53%) 14 (14%)Fibrous 8 (8%) 47 (46%)Mixed 35 (35%) 25 (25%)Calcified 4 (4%) 15 (15%)eference segmentEEM CSA (mm2) 16.8 � 4.7 16.9 � 4.4 0.8Lumen CSA (mm2) 9.6 � 2.7 10.1 � 3.3 0.2inimal lumen site

EEM CSA (mm2) 18.4 � 5.3 17.0 � 4.4 0.04Lumen CSA (mm2) 4.5 � 2.4 4.5 � 2.2 0.9Plaque & media CSA (mm2) 13.9 � 4.9 12.5 � 4.1 0.03Plaque burden (%) 76 � 14 74 � 11 0.1Area stenosis (%) 58 � 16 52 � 21 0.2Remodeling index 1.10 � 0.18 1.01 � 0.13 0.001avity area (mm2) 2.5 � 1.5

Data are presented as means � 1 SD or numbers (percentages).

ectional area (CSA) according to the American College of t

ardiology Clinical Expert Consensus Document on Stan-ards for Acquisition, Measurement, and Reporting of In-ravascular Ultrasound Studies.14

Calcium produced bright echoes (brighter than the refer-nce adventitia) with acoustic shadowing of deeper arterialtructures.14 Ten-millimeter-long segments that included theinimum lumen CSA were analyzed. Cross-sectional im-

ges were digitized every 0.5 mm,15 the arc of calcium waseasured in each image slice, and the number of calcium

eposits was counted. The location of each calcium depositas assessed by 2 independent experienced observers whoere unaware of the clinical data.16

First, the arc of each calcium deposit was measured inegrees with a protractor that was centered on the lumen,nd the largest arc of calcium was identified. The intraob-erver variability of the arc of calcium measurement was.8° � 3.8°. The largest arc of calcium was then classifieds 1 quadrant (�90°), 2 quadrants (91° to 180°), 3 quadrants181° to 270°), or 4 quadrants (271° to 360°).

Second, the location of calcium was defined as superfi-ial (the leading edge of the acoustic shadowing appearedithin the most shallow 50% of the plaque plus media) oreep (the leading edge of the acoustic shadowing appearedithin the deepest 50% of the plaque plus media thickness).hen present, the plaque rupture cavity was included in the

laque plus media thickness.Third, calcium length (in millimeters) was measured

rom the number of seconds of videotape in which calciumas identified (millimeters equal to the product of secondsf videotape and 0.5 mm/s).

Image slices with the largest intraplaque cavity, imagelices with minimum lumen CSA, and proximal and distaleference sites were identified and measured. Referenceites were defined as sections with the largest lumen and theeast plaque within 5 mm proximal and distal to the lesionut before any side branch.12 Using planimetry softwareTapeMeasure, INDEC Systems, Mountain View, Califor-ia), external elastic membrane (EEM) CSA and lumenSA were measured. Further, the largest intraplaque cavityas measured and extrapolated to the ruptured capsule

rea.11–13 Plaque plus media CSA including cavity areaEEM CSA minus lumen CSA), plaque burden (plaque plusedia CSA divided by EEM CSA), area stenosis (mean ref-

rence lumen CSA minus lesion lumen CSA divided by meaneference lumen CSA), and remodeling index (lesion EEMSA divided by mean reference EEM CSA) were calculated.

Statistical analysis: Statistical analysis was performedith StatView 5.0 (SAS Institute, Cary, North Carolina).ontinuous variables were reported as mean � 1 SD. Un-aired Student’s t test tested differences between 2 sets ofata with normal distributions. If normality tests failed, theann-Whitney U statistic test was used. Categorical vari-

bles were reported as frequencies and compared usinghi-square statistics. A p value �0.05 was considered sta-

ap

R

fialII

c(0(mcpipwep1T3p

apa

ai(1p5cmmoF

rNr�r

Fabcei

Fci(dr

355Coronary Artery Disease/Ruptured Plaque and Coronary Calcium

nd stay criteria of 0.10 was used to find independentredictors of ruptured plaques.

esultsBaseline patient characteristics and angiographic

ndings: Baseline patient and angiographic characteristicsre presented in Table 1. Coronary angiography detectedesion calcium in 34% of ruptured plaques (p �0.001 vsVUS), and 42% of control plaques (p �0.001 vs IVUS).VUS findings are presented in Table 2.

IVUS calcium: Fifty-nine percent of plaques had �1alcium deposit at the section with minimum lumen CSA53% of ruptured plaques vs 65% of control plaques, p �.1). In addition, 64% of plaques had �1 calcium depositregardless of size) somewhere else than at the section withinimum lumen CSA (84% of ruptured plaques vs 45% of

ontrol plaques, p �0.001). Lesions that contained rupturedlaques had a significantly larger number of calcium depos-ts than did control plaques (3.5 � 1.7 vs 1.8 � 1.1,�0.001). However, the arc of the largest calcium depositas smaller and the length of the largest calcium deposit in

ach plaque was shorter in ruptured plaques than in controllaques (67.3° � 41.4° vs 114.9° � 77.4°, p �0.001, and.6 � 1.3 vs 4.0 � 2.7 mm, p �0.001, respectively; Figure 2).he frequency distribution of these arcs is shown in Figure. In addition, among 202 plaques, 7 plaques (7 rupturedlaques and 0 control plaques, p � 0.01) had a maximum

igure 2. Four examples of IVUS lesion calcium. (A) Superficial calciumrc of 152°. (B) Deep calcium arc of 34° with an extrapolated lumen/intimaorder (dotted line) and a cavity (asterisk). (C) Arc of calcium (arrow) isoncordant to maximum plaque thickness (double-headed arrow), with anxtrapolated media/adventitia border (dotted line). (D) Arc of calcium (arrow)

�s perpendicular to maximum plaque thickness (double-headed arrow).

rc of calcium of �10°, and 29 plaques (22 rupturedlaques and 7 control plaques, p � 0.003) had a maximumrc of calcium of �30°.

At least 1 superficial calcium deposit somewhere within10-mm-long segment centered on the lesion was observed

n 82% of ruptured plaques and 93% of control plaquesp � 0.02). Numbers of superficial calcium deposits were.7 � 1.2 in ruptured plaques and 1.6 � 1.0 in controllaques (p � 0.6). Arcs of superficial calcium measured6.2° � 35.5° in ruptured plaques and 95.8° � 65.2° inontrol plaques (p �0.001). Lengths of superficial calciumeasured 1.7 � 1.4 mm in ruptured plaques and 4.3 � 2.8m in control plaques (p �0.001). Frequency distributions

f the arc and length of superficial calcium are shown inigure 4.

At least 1 deep calcium deposit was seen in 82% ofuptured plaques and 22% of control plaques (p �0.001).umbers of deep calcium deposits were 1.8 � 1.4 in

uptured plaques and 0.3 � 0.6 in control plaques (p0.001). Arcs of deep calcium measured 38.9 � 19.4° in

uptured plaques and 48.6 � 26.3° in control plaques (p

igure 3. Frequency distribution of the maximum arc and number of lesionalcium deposits. Top, 1-quadrant calcium was significantly more frequentn ruptured plaques (white bars) than in control plaques (black bars)p �0.001). Bottom, most calcium deposits counted in control plaques wereistributed between 1 and 3, whereas 50% of calcium deposits counted inuptured plaques were distributed between 4 and 7 (p �0.001).

0.08). Lengths of deep calcium measured 1.5 � 1.1

mpa

ftpwdc�

D

Ttsco

mveppmw

Fcmmm

Fsrbwcdt

356 The American Journal of Cardiology (www.AJConline.org)

m in ruptured plaques and 2.3 � 1.2 mm in controllaques (p � 0.001). Frequency distributions of the arcnd length of deep calcium are shown in Figure 5.

Multivariate logistic regression analysis was per-ormed to identify independent predictors of plaque rup-ures. The following variables were identified as inde-endent predictors of culprit plaque ruptures in patientsho had acute coronary syndrome: number of calciumeposits within an arc �90° (p � 0.01), number of deepalcium deposits (p � 0.02), and the remodeling index (p

igure 4. Frequency distribution of the maximum arc and length ofuperficial calcium. Top, 1-quadrant superficial calcium was common inuptured plaques (white bars) compared with control plaques (blackars) (p �0.001). Middle, most superficial calcium in ruptured plaquesas �2.5 mm in length, whereas the length of superficial calcium in

ontrol plaque was distributed evenly (p �0.001). Bottom, frequencyistribution of the number of superficial calcium deposits was similar inhe 2 groups (p � 0.6).

0.048). (

iscussion

he present study demonstrates significantly different pat-erns of lesion calcium in ruptured plaques compared withtable plaques. Ruptured plaques had quantitatively lessalcium, especially superficial calcium, but a larger numberf small calcium deposits, especially deep calcium deposits.

Currently, electron-beam computed tomography isost commonly used to detect coronary calcium nonin-

asively, whereas cinefluorography and IVUS are used tovaluate lesion calcium before interventions. Raggi et al4

erformed electron-beam computed tomography in 172atients who had acute myocardial infarction as the firstanifestation of coronary artery disease. Most patientsith at least moderate angiographic disease (105 of 110,

igure 5. Frequency distribution of the maximum arc and length of deepalcium. Top, deep calcium of �180° was not seen in either group. Middle,ost deep calcium in ruptured plaques was �2.5 mm (p � 0.03). Bottom,ost ruptured plaques (white bars) had deep calcium deposits, whereasost control plaques (black bars) did not have calcium deposits

p �0.001).

9cpstcgdtmoa

ucciwswwIrmwh

pwmacwoac

1

1

1

1

1

1

1

1

1

357Coronary Artery Disease/Ruptured Plaque and Coronary Calcium

6%) had measurable calcium deposits on electron-beamomputed tomograms. The angiographic status of theseatients was not reported. Overall, 165 patients (96%)howed coronary calcium by electron-beam computedomography. In 150 of these patients (87%), the extent ofalcium was greater than expected for their age andender. These data suggest that, even in patients whoevelop acute coronary syndrome as the first manifesta-ion of coronary artery disease, coronary calcium is al-ost always present and usually exceeds the amount

bserved in asymptomatic patients or those who havetypical symptoms.

In contrast, several IVUS studies have shown thatnstable clinical symptoms are associated with less cal-ium.9,10 However, in previous IVUS studies, coronaryalcium was assessed only at the minimum lumen CSAmage slice.3,8,17 In the present study, 53% of patientsho had ruptured plaque had �1 calcium deposit at the

ection with minimum lumen CSA, and 84% of patientsho had ruptured plaque had �1 calcium deposit some-here else than at the section with minimum lumen CSA.

n accordance with these findings, a previous IVUS studyeported that small calcium deposits are significantlyore frequent in the culprit lesion segments in patientsho have acute coronary syndrome than in patients whoave stable angina pectoris.18

Although the data were collected prospectively by inde-endent monitors and entered into a dedicated database, thisas a retrospective analysis. However, we used a case-atch analysis to minimize bias between ruptured plaques

nd nonruptured plaques. Some ruptured plaques with smallavities may have been missed, especially when the cavityas filled by thrombus. The interval between symptomnset and IVUS imaging may have influenced the calciumppearance. Superficial calcium could obscure a deep cal-ium deposit.

1. Davies M, Thomas A. Thrombosis and acute coronary artery lesions insudden cardiac ischemic death. N Engl J Med 1984;310:1137–1140.

2. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis ofcoronary artery disease and the acute coronary syndromes. N EnglJ Med 1992;326:242–250.

3. Virmani R, Kolodgie FD, Burke AP, Farb A, Schwartz SM. Lessonsfrom sudden coronary death: a comprehensive morphological classi-fication scheme for atherosclerotic lesions. Arterioscler Thromb VascBiol 2000;20:1262–1275.

4. Raggi P, Callister TQ, Cooil B, He ZX, Lippolis NJ, Russo DJ,Zelinger A, Mahmarian JJ. Identification of patients at increased riskof first unheralded acute myocardial infarction by electron-beam com-

5. Burke AP, Taylor A, Farb A, Malcom GT, Virmani R. Coronarycalcification: insights from sudden coronary death victims. Z Kardiol2000;89(suppl 2):49–53.

6. Gertz SD, Roberts WC. Hemodynamic shear force in rupture of coronaryarterial atherosclerotic plaques. Am J Cardiol 1990;66:1368–1372.

7. Cheng GC, Loree HM, Kamm RD, Fishbein MC, Lee RT. Distributionof circumferential stress in ruptured and stable atherosclerotic lesions.A structural analysis with histopathological correlation. Circulation1993;87:1179–1187.

8. Farb A, Burke AP, Tang AL, Liang TY, Mannan P, Smialek J,Virmani R. Coronary plaque erosion without rupture into a lipid core.A frequent cause of coronary thrombosis in sudden coronary death.Circulation 1996;93:1354–1363.

9. Rasheed Q, Nair R, Sheehan H, Hodgson JM. Correlation of intra-coronary ultrasound plaque characteristics in atherosclerotic coronaryartery disease patients with clinical variables. Am J Cardiol 1994;73:753–758.

0. Nakamura M, Nishikawa H, Mukai S, Setsuda M, Nakajima K,Tamada H, Suzuki H, Ohnishi T, Kakuta Y, Nakano T, Yeung AC.Impact of coronary artery remodeling on clinical presentation of cor-onary artery disease: an intravascular ultrasound study. J Am CollCardiol 2001;37:63–69.

1. von Birgelen C, Klinkhart W, Mintz GS, Papatheodorou A, HerrmannJ, Baumgart D, Haude M, Wieneke H, Ge J, Erbel R. Plaque distri-bution and vascular remodeling of ruptured and nonruptured coronaryplaques in the same vessel: an intravascular ultrasound study in vivo.J Am Coll Cardiol 2001;37:1864–1870.

2. Maehara A, Mintz GS, Bui AB, Walter OR, Castagna MT, Canos D,Pichard AD, Satler LF, Waksman R, Suddath WO, et al. Morphologicand angiographic features of coronary plaque rupture detected byintravascular ultrasound. J Am Coll Cardiol 2002;40:904–910.

3. Fujii K, Kobayashi Y, Mintz GS, Takebayashi H, Dangas G, MoussaI, Mehran R, Lansky AJ, Kreps E, Collins M, et al. Intravascularultrasound assessment of ulcerated ruptured plaques: a comparison ofculprit and nonculprit lesions of patients with acute coronary syn-dromes and lesions in patients without acute coronary syndromes.Circulation 2003;108:2473–2478.

4. Mintz GS, Nissen SE, Anderson WD, Bailey SR, Erbel R, FitzgeraldPJ, Pinto FJ, Rosenfield K, Siegel RJ, Tuzcu EM, Yock PG. AmericanCollege of Cardiology clinical expert consensus document on stan-dards for acquisition, measurement and reporting of intravascularultrasound studies (IVUS). A report of the American College ofCardiology Task Force on Clinical Expert Consensus Documents.J Am Coll Cardiol 2001;37:1478–1492.

5. Scott DS, Arora UK, Farb A, Virmani R, Weissman NJ. Pathologicvalidation of a new method to quantify coronary calcific deposits invivo using intravascular ultrasound. Am J Cardiol 2000;85:37–40.

6. Mintz GS, Popma JJ, Pichard AD, Kent KM, Satler LF, Chuang YC,Ditrano CJ, Leon MB. Patterns of calcification in coronary arterydisease. A statistical analysis of intravascular ultrasound and coronaryangiography in 1155 lesions. Circulation 1995;91:1959–1965.

7. Falk E. Coronary thrombosis: pathogenesis and clinical manifesta-tions. Am J Cardiol 1991;68(suppl):28B–35B.

8. Ehara S, Kobayashi Y, Yoshiyama M, Shimada K, Shimada Y, FukudaD, Nakamura Y, Yamashita H, Yamagishi H, Takeuchi K, et al. Spottycalcification typifies the culprit plaque in patients with acute myocar-dial infarction: an intravascular ultrasound study. Circulation 2004;