Isdoib�ctarscc

M

TCs5ge

f2

A

0d

Changes in Left Anterior Descending Coronary Artery Wall Thickness Detectedby High Resolution Transthoracic Echocardiography

Rebecca Perry, BSc, DMU (Cardiac)*, Carmine G. De Pasquale, BMBS, PhD,Derek P. Chew, MBBS, MPH, Lynn Brown, RN, Philip E. Aylward, BM, PhD,

and Majo X. Joseph, MBBS

Recently, it has been demonstrated that high-resolution transthoracic echocardiography(HRTTE) is able to detect differences in the wall thickness of the left anterior descendingcoronary artery (LAD) between patients with coronary artery disease (CAD) and normalvolunteers. The aim of this study was to further validate this technique. One hundred tenvolunteers, 58 patients with angiographically proved CAD and 52 control subjects, under-went assessments of their LADs using HRTTE. Anterior and posterior wall thicknessesdiffered between subjects in the CAD group and controls (1.9 � 0.6 vs 1.2 � 0.3 mm,p <0.001, and 1.8 � 0.5 vs 1.2 � 0.3 mm, p <0.001, respectively). External LAD diameterwas also greater in subjects in the CAD group compared with controls (5.2 � 1.9 vs4.4 � 0.9 mm, respectively, p � 0.01). However, there was no difference in luminaldiameter between subjects in the CAD group and the controls (1.9 � 0.9 vs 2.1 � 0.8mm, respectively, p � 0.3). In conclusion, HRTTE demonstrated that LAD wallthicknesses and external diameters in patients with CAD were significantly larger thanin normal volunteers. Luminal diameter, however, was maintained in the 2 groups,indicating that subjects in the CAD group had undergone positive remodeling at thesite measured. This objectively visualized evidence of coronary atherosclerosis withHRTTE would likely be undetected during coronary angiography. © 2008 Elsevier

Inc. All rights reserved. (Am J Cardiol 2008;101:937–940)

etaaucTwguPhcsatr

estwtwlssvw

ntravascular ultrasound and epicardial echocardiographictudies have demonstrated that coronary atherosclerosis is aiffuse disease process and rarely spares the proximal cor-nary arteries, especially the proximal left anterior descend-ng coronary artery (LAD).1–7 In fact, it has been shown thatefore clinical coronary artery disease (CAD) is evident,90% of the coronary artery tree is atherosclerotic.5,8 Ac-

urate baseline measurements of the LAD luminal and ex-ernal diameters and wall thickness have been shown to bechievable using a recently described technique of high-esolution transthoracic echocardiography (HRTTE).9,10 Weought to use this technique to demonstrate differences inoronary atherosclerosis between patients with CAD andontrols.

ethods and Results

his study was approved by the Flinders Research Ethicsommittee, and all subjects gave written informed con-

ent to participate in the study. Healthy volunteers (n �2) without histories of or risk factors for CAD (controlroup; no clinical hypertension, no hypercholesterol-mia, nonsmoking, nondiabetic, and no history of periph-

Cardiac Services, Flinders Medical Centre/Flinders University, Bed-ord Park, South Australia, Australia. Manuscript received September 7,007; revised manuscript received and accepted November 13, 2007

Dr. Joseph is supported by a Cardiovascular Lipid Grant from Pfizerustralia, Sydney, Australia.

*Corresponding author: Tel: 61-882046075; fax: 61-882044907.

cE-mail address: rebecca.perry@fmc.sa.gov.au (R. Perry).

002-9149/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved.oi:10.1016/j.amjcard.2007.11.053

ral or cerebral vascular disease) and consecutive hospi-al inpatients with angiographically proved CAD (defineds a coronary artery stenosis �50% in any coronaryrtery branch other than the LAD; n �58) underwentltrasound scans to measure luminal and external arteryaliber and anterior and posterior LAD wall thicknesses.he patients with CAD were prospectively identifiedhile in the hospital. Subject characteristics for the CADroup are listed in Table 1. Examinations were obtainedsing a commercially available ultrasound system (iE33;hilips Medical Systems, Bothell, Washington) with aigh-frequency transducer (S8-3). The LAD was re-orded using a parasternal long-axis examination, with alight inferior tilt to obtain long-axis images of the LADs it runs along the interventricular septum. In our hands,he inter- and intraoperator variability of this method is� 0.86 (p �0.001) and r � 0.86 (p �0.001), respectively.

Readers were blinded to any clinical data. The pres-nce of 2 linear echoes anterior to the interventriculareptum in �3 consecutive frames was used to identifyhe LAD. Segments with the largest luminal diametersere selected to ensure that the measured cross section of

he artery was through the center and thus the resultsere not confounded by off-axis images. Where multiple

inear echoes were noted anterior to the interventriculareptum, the thicker walled vascular structure was mea-ured to avoid potential confusion with the great cardiacein. Pulsed-wave spectral and color Doppler (Figure 1)as used to further delineate the LAD from the great

ardiac vein if required. For all measurements, the dis-

www.AJConline.org

tv

patc

ps

s5w

Fi

TS

V

AMTTLHCRLPP

PHDS

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

ance between the inner edges of the lines representingascular walls was used.

Continuous variables are expressed as mean � SD. Aaired Student’s t test was used to determine if there wasny significant difference in each measurement betweenhe 2 groups. Discrete variables were compared using

igure 1. A modified parasternal long-axis view to demonstrate color flown the right ventricular inflow tract (RVIT) can also be detected. LV � le

able 1ubject characteristics of coronary artery disease group in total cohort an

ariable C

ge (yrs)en

otal cholesterol 4.9 � 1riglyceride 2.0 � 1ow-density lipoprotein 3.0 � 1igh-density lipoprotein 1.4 � 1-reactive protein (mg/L)ight coronary diseaseeft circumflex diseaserevious statin therapyrevious angiotensin-converting enzyme inhibitor/angiotensin

receptor blocker therapyrevious aspirin therapyypertensioniabetes mellitusmokers

Continuous data are expressed as mean � SD and discrete variables as

hi-square analysis and are expressed as numbers and T

ercentages. A p value �0.05 was considered statisticallyignificant.

Adequate imaging of the LAD was possible in 50 of the 52ubjects (96%) in the control group (42 men) and in 50 of the8 subjects (86%) in the CAD group (32 men). The 10 subjectsithout adequate imaging were excluded from the analysis.

LAD as it runs along the interventricular septum (IVS). Some color flowicular cavity.

oup analysis

up (Total CADrt, n � 50)

CAD Subgroup (CAD SubjectsAged �55 Years, n � 22)

pValue

1 � 7 44 � 2 0.012 (84%) 17 (77%) 0.22l/L (190 � 65 mg/dl) 4.9 � 1.6 mmol/L (190 � 62 mg/dl) 0.57l/L (77 � 50 mg/dl) 1.9 � 0.9 mmol/L (78 � 35 mg/dl) 0.44l/L (117 � 62 mg/dl) 3.1 � 1.5 mmol/L (52 � 59 mg/dl) 0.92l/L (54 � 46 mg/dl) 1.2 � 0.3 mmol/L (46 � 12 mg/dl) 0.15.0 � 9.4 8.4 � 10.7 0.457 (74%) 16 (73%) 0.550 (60%) 14 (63%) 0.431 (42%) 9 (41%) 0.29 (18%) 4 (18%) 0.54

8 (36%) 7 (31%) 0.433 (26%) 6 (27%) 0.553 (6%) 1 (5%) 0.537 (34%) 11 (50%) 0.07

r (percentage).

in theft ventr

d subgr

AD GroCoho

54

.7 mmo

.3 mmo

.6 mmo

.2 mmo8332

11

1

numbe

he subjects in the CAD group were significantly older than

tap

nt0LgmdCr

eg�tmsitd4

d(1dg

rct2

D

Tdawnnsdacrgtt

iiaTowstn

ComrThpt

tCr0cvn

trhmatp

Lmp

Flac

Flm

939Coronary Artery Disease/Echocardiographic Detection of Subclinical Atherosclerosis

hose in the control group (51 � 6 vs 35 � 9 years, p �0.001)nd had higher body mass indexes (29 � 4 vs 25 � 4 kg/m2,� 0.02).Anterior and posterior wall thicknesses differed sig-

ificantly between subjects in the CAD group and con-rols (1.9 � 0.6 vs 1.2 � 0.3 mm, p �0.001, and 1.8 �.5 vs 1.2 � 0.3 mm, p �0.001, respectively). ExternalAD diameter was also increased in subjects in the CADroup compared with controls (5.2 � 1.9 vs 4.4 � 0.9m, respectively, p � 0.01). However, there was no

ifference in luminal diameter between subjects in theAD group and controls (1.9 � 0.9 vs 2.1 � 0.8 mm,

espectively, p � 0.3) (Figure 2).To determine if this difference in wall thickness and

xternal diameter was due to the effects of age, a sub-roup analysis was performed using control subjects aged35 years and subjects with CAD aged �55 years. In

his subgroup analysis, there were 23 control subjects (13en) and 22 patients with CAD (17 men). The subgroup

ubject characteristics in the CAD group did not signif-cantly differ from those of the total cohort, except thathey were significantly younger (Table 1). There was noifference in age between the 2 subgroups (control group2 � 5 years vs CAD group 44 � 2 years, p � 0.3).

The anterior and posterior wall thickness significantlyiffered between the age-matched CAD group and controls1.7 � 0.5 vs 1.2 � 0.3 mm, p � 0.001, and 1.8 � 0.5 vs.3 � 0.4 mm, p � 0.001, respectively). The external LADiameter was again also increased in the age-matched CAD

igure 2. Graph demonstrating the differences in LAD wall thickness,uminal diameter, and external artery diameter between normal subjectsnd patients with angiographically proved CAD �50% in any part of theoronary tree other than the proximal or mid LAD.

igure 3. Graph demonstrating the differences in LAD wall thickness,uminal diameter, and external artery diameter between age-matched nor-al subjects and patients with angiographically proved CAD.

roup compared with controls (5.2 � 1.9 vs 4.5 � 0.8 mm, s

espectively, p � 0.03). However, as seen in the totalohort, there was no difference in luminal diameter betweenhe age-matched CAD group and the controls (2.0 � 1.1 vs.4 � 0.7 mm, respectively, p � 0.5) (Figure 3).

iscussion

his study confirms that HRTTE can detect structuralifferences in proximal LAD morphology suggestive oftherosclerosis-induced positive remodeling in patientsith confirmed significant luminal CAD in other coro-ary vascular territories. We found the LAD wall thick-esses and external diameters of subjects with CAD to beignificantly larger than those of normal volunteers, in-icating atherosclerotic buildup (Figure 4). Luminal di-meters, however, were the same in the 2 groups, indi-ating the well-recognized phenomenon of positiveemodeling at the measured site in subjects in the CADroup. Because of the preservation of luminal diameter inhis group, this atherosclerotic thickening was not de-ected during angiography.

Subjects in the CAD group were older, thereby introduc-ng a bias that could potentially account for the differencesn wall thickness and external artery diameter, a limitationlso observed by the pioneering work of Gradus-Pizlo et al.9o counter this, we performed a subgroup analysis using thelder control subjects (�35 years) and the younger subjectsith CAD (�55 years). The comparative difference in vi-

ualized proximal LAD existed between the groups afterhis analysis, thereby confirming that the differences wereot solely age related.

The increased wall thickening seen in subjects in theAD group despite the benign angiographic appearancef the proximal and mid LAD region indicates that thisethod of HRTTE may be more sensitive than angiog-

aphy for the detection of subclinical atherosclerosis.his may have future clinical relevance in an era whenigh-risk subgroup-targeted primary prevention thera-eutic intervention exists, such as statin and aspirinherapy.

In the study by Gradus-Pizlo et al,9 the average wallhickness, luminal diameter, and external diameter in theAD group were 1.9 � 0.4, 2.2 � 0.5, and 6.0 � 1.1 mm,

espectively, and in the control group, these values were.9 � 0.1, 2.1 � 0.6, and 3.9 � 0.7 mm, respectively. Theorresponding values in this study closely agree with thesealues, supporting robust interobserver variability for thisovel technique.

There is a potential risk for confusion of the LAD forhe great cardiac vein because in most cases, the 2 vesselsun parallel to each other along the anterior surface of theeart. This risk was minimized, however, by measure-ent of the thicker walled vascular structure visualized

nd by using pulsed-wave Doppler analysis to definehe arterial from the venous blood flow in a subset ofatients.

There were a number of subjects (10%) in whom theAD could not be imaged well enough to make accurateeasurements of the wall thickness. Echocardiography,

articularly at a high frequency, is limited in imaging

ubjects with poor intercostal spaces, significant lung

dtvar

1

d in th

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

isease, and obesity. Finally, although it is also possiblehat variations in coronary anatomy made it difficult toisualize the LAD in some subjects who may have had andequate parasternal window, the proximal LAD is a veryare site of congenital anatomic abnormality.

1. McPherson DD, Sirna SJ, Hiratzka LF, Thorpe L, Armstrong ML,Marcus ML, Kerber RE. Coronary arterial remodelling studied byhigh-frequency epicardial echocardiography: an early compensatorymechanism in patients with obstructive coronary atherosclerosis. J AmColl Cardiol 1991;17:79–86.

2. Tuzcu EM, Hobbs RE, Rincon G, Bott-Silvermann C, De Franco AC,Robinson K, McCarthy PM, Stewart RW, Guyer S, Nissen SE. Occultand frequent transmission of atherosclerotic coronary disease withcardiac transplantation: insights from intravascular ultrasound. Circu-lation 1995;91:1706–1713.

3. Abizaid A, Mintz GS, Pichard AD, Kent KM, Satler LF, Popma JJ,Leon MB. Is intravascular ultrasound clinically useful or is it just aresearch tool? Heart 1997;78:27–30.

4. Topol EJ, Nissen SE. Our preoccupation with coronary luminology:the dissocation between clinical and angiographic findings in ischemicheart disease. Circulation 1995;92:2333–2342.

Figure 4. HRTTE of the LAD of subjects in the control group (A) an

5. McPherson DD, Hiratzka LF, Lamberth WC, Brandt B, Hunt M, KiesoRA, Marcus ML, Kerber RE. Delineation of the extent of coronary

atherosclerosis by high-frequency epicardial echocardiography. N EnglJ Med 1987;316:304–309.

6. Hausmann D, Johnson JA, Sudhir K, Mullen WL, Friedrich G, Fitzger-ald PJ, Chou TM, Ports TA, Kane JP, Malloy MJ, Yock PG. Angio-graphically silent atherosclerosis detected by intravascular ultrasoundin patients with familial hypercholesterolemia and familial combinedhyperlipidemia: correlation with high density lipoproteins. J Am CollCardiol 1996;27:1562–1570.

7. Hermiller JB, Tenaglia AN, Kisslo KB, Phillips HR, Bashore TM,Stack RS, Davidson CJ. In vivo validation of compensatory enlarge-ment of atherosclerotic coronary arteries. Am J Cardiol 1993;71:665–668.

8. Nakamura Y, Takemori H, Shiraishi K, Inoki I, Sakagami M, Shi-makura A, Usuda K, Kubota K, Takata S, Kobayashi K. Compensatoryenlargement of angiographically normal coronary segments in patientswith coronary artery disease. In vivo documentation using intravascu-lar ultrasound. Angiology 1996;47:775–781.

9. Gradus-Pizlo I, Sawada SG, Wright D, Segar DS, Feigenbaum H.Detection of subclinical coronary atherosclerosis using two- dimen-sional, high-resolution transthoracic echocardiography. J Am CollCardiol 2001;37:1422–1429.

0. Perry R, Joseph MX, De Pasquale CG, Chew DP, Yiu D, Aylward PE,Mangoni AA. High resolution transthoracic echocardiography of theleft anterior descending coronary artery: a novel non-invasive assess-

e CAD group (B). Arrows demonstrate anterior LAD wall thickness.

ment of coronary vasoreactivity. J Am Soc Echocardiography. Inpress.