DOI: 10.1111/j.1540-8175.2012.01667.xC© 2012, Wiley Periodicals, Inc.

Predictors of Statin-Induced Regression of LeftAnterior Descending Coronary Artery Wall Thicknessas Measured by High-Resolution TransthoracicEchocardiography

Rebecca Perry, B.Sc., D.M.U. (Cardiac), A.M.S., Majo X. Joseph, M.B.B.S., F.R.A.C.P.,Derek P. Chew, M.B.B.S., M.P.H., F.R.A.C.P., Philip E. Aylward, B.M., Ph.D., F.R.A.C.P.,and Carmine G. De Pasquale, B.M.B.S., Ph.D., F.R.A.C.P.

Cardiac Services, Flinders Medical Centre/Flinders University, Bedford Park, South Australia, Australia

Background: Statin therapy has been shown to reduce cardiovascular risk after myocardial infarction(MI). Using a novel technique of high-resolution transthoracic echocardiography (HRTTE), we soughtto assess the statin-induced changes in left anterior descending coronary artery (LAD) wall thicknessin previously statin naive patients over a 12-month period. Method and Results: Thirty subjects un-derwent HRTTE assessment of their LAD wall thickness predischarge post-MI (non-LAD territory) andat 3, 6, and 12 months. The LAD anterior and posterior wall thickness and vessel luminal diameterwere measured. Blood lipid levels were also assessed at each study visit. All subjects were started onmoderate lipid-lowering therapy (40 mg of atorvastatin or simvastatin). There was a sustained decreasein total cholesterol (–23%), triglycerides (–19%), and low-density lipoprotein (–41%) at the 3-monthvisit from the baseline, with no change in high-density lipoprotein level. Overall, there was no changein the LAD wall thickness and external or vessel lumen diameter over the 12-month period. Of thosethat demonstrated regression, the only predictor of percentage change in the LAD wall thickness wasthe baseline LAD wall thickness. Conclusion: Despite a favorable change in blood lipid profile, no overallchange in the LAD wall thickness was detected over a 12-month period in subjects on moderate statintherapy using HRTTE. However, using case-based analysis, regression was able to be predicted by thebaseline LAD wall thickness. HRTTE may be an instructive noninvasive modality to assess response tostatin intervention. (Echocardiography 2012;29:641-646)

Key words: echocardiography, coronary artery disease, statin therapy, plaque regression

Statin therapy reduces future cardiac events inpatients with coronary heart disease and in thoseafter myocardial infarction (MI).1–5 A 30% reduc-tion in mortality risk post-MI has been observedwith lipid lowering over an 8-year follow-up pe-riod.6 However, it has also been shown that thisreduction in events does not necessarily correlateto a reduction in coronary artery disease (CAD) asdemonstrated by angiography.5 In fact, the typ-ical angiographic reduction in stenosis was ap-proximately 1–3% compared with a 25–75% re-duction in acute coronary events.1,5 As plaquesstabilize, it is thought that negative remodelingoccurs causing limited changes in their angio-graphic appearance despite a marked reductionin cardiovascular events.7

Conflicts of Interests: This work (through MXJ) is supportedby a CVL grant from Pfizer Australia.

Address for correspondence and reprint requests: RebeccaPerry, B.Sc., D.M.U. (Cardiac), A.M.S., Cardiac Services,Level 6, Flinders Medical Centre, Flinders Drive, Bed-ford Park, SA 5042, Australia. Fax: +61882044907; E-mail:rebecca.perry@health.sa.gov.au

Lipid lowering has been shown to reduce therate of progression of atherosclerosis and the de-velopment of new lesions compared with a con-trol group in intravascular ultrasound (IVUS) tri-als using high-dose lipid-lowering therapy8 andregression using very high dose lipid-loweringtherapy.9

Accurate baseline measurements of the left an-terior descending artery (LAD) luminal and exter-nal diameters and wall thickness have been shownto be achievable using a recently described tech-nique of high-resolution transthoracic echocar-diography (HRTTE).10–12 We sought to use thistechnique to assess the statin-induced changesin the LAD wall thickness in previously statinnaive patients after non-LAD MI over a 12-monthperiod.

Methods:This study was approved by the Flinders Re-search Ethics Committee, and all subjects gavewritten informed consent to participate in thestudy. Consecutive acute MI inpatients with

641

Perry, et al.

Figure 1. A modified parasternal long-axis view to demonstrate color flow in the left anterior descending coronary artery as it runsalong the interventricular septum. Some color flow in the right ventricular inflow tract can also be detected. IVS = interventricularseptum; LAD = left anterior descending coronary artery; LV = left ventricular cavity; RVIT = right ventricular inflow tract.

angiographically proven CAD (defined as a coro-nary artery stenosis >50% in any coronary arterybranch other than the LAD; n = 30) underwentan ultrasound scan to measure the luminal and

external artery caliber and anterior and posteriorLAD wall thickness.

Examinations were obtained using a commer-cially available ultrasound system (iE33, Philips

Figure 5. Pulsed-wave Doppler image of blood flow in left anterior descending coronary artery (LAD) demonstrating predomi-nately diastolic flow. Velocities are lower than apical coronary artery Doppler due to the angle of incidence parasternally.

642

Echo Detection of Atherosclerotic Regression

Medical Systems, Bothell, WA, USA) with ahigh-frequency transducer (S8-3). The LAD wasrecorded using a parasternal long-axis examina-tion with a slight inferior tilt to obtain long-axisimages of the LAD as it runs along the interven-tricular septum. Readers were blinded to any clin-ical data. The presence of the two linear echoesanterior to the interventricular septum in at leastthree consecutive frames was used to identify theLAD. Segments with the largest luminal diame-ter were selected to ensure that the measuredcross-section of the artery was through the lu-minal center, and thus the results were not con-founded by off-axis images. Where multiple linearechoes were noted anterior to the interventricularseptum, the thicker walled vascular structure wasmeasured to avoid potential confusion with thegreat cardiac vein. Pulsed-wave spectral and colorDoppler (Fig. 1) were used to further delineate theLAD from the great cardiac vein if required. Forall measurements, the distance between the inneredges of the lines representing vascular walls wasused.

A fasting blood test was obtained to mea-sure total cholesterol, triglycerides, low-densitylipoprotein (LDL), and high-density lipoprotein(HDL) levels.

Height and weight were also measured, andthe body mass index (BMI) was calculated.

Statistical Analysis:Data are expressed as the mean ± standard de-viation for continuous variables and as percent-ages for discrete variables. A paired t-test wasused to determine significant changes in theLAD wall thickness or cholesterol profile over the12 months with Bonferroni correction for re-peated measures.

Univariate analysis was performed to evaluatevariables as independent predictors of percent-age change in the LAD wall thickness. The inde-pendent variables examined were age, BMI, to-tal cholesterol, triglycerides, LDL, HDL, C-reactiveprotein (CRP), homocysteine, creatinine, locationof MI, hypertension, diabetes, smoking status,and family history of cardiovascular disease.

All statistical analyses were performed usingSPSS software for Windows (version 17.0, SPSSInc., Chicago, IL, USA). A P-value of <0.05 wasconsidered statistically significant.

Results:Adequate imaging of the LAD was achieved in27 of the 30 (90%) subjects. The data from thethree subjects without LAD imaging data were ex-cluded from the analysis. Five subjects were lost tofollow-up and two subjects died within the studyperiod, one from a subsequent MI and the other

Table I

Baseline Subject Characteristics

Age (years) 53 ± 7Males 15 (75%)BMI (kg/m2) 29 ± 4Total cholesterol (mg/dL) 201.1 ± 50.3Total cholesterol (mmol/L) 5.2 ± 1.3Triglyceride (mg/dL) 81.2 ± 38.7Triglyceride (mmol/L) 2.1 ± 1.0LDL (mg/dL) 123.7 ± 50.3LDL (mmol/L) 3.2 ± 1.3HDL (mg/dL) 46.4 ± 11.6HDL (mmol/L) 1.2 ± 0.3CRP (mg/L) 7.1 ± 10RCA infarct 5 (25%)LCx infarct 11 (55%)Both RCA and LCx infarct 4 (20%)HTN 4 (20%)Diabetes 1 (5%)Smokers 7 (35%)Family Hx CAD 10 (50%)

Continuous data expressed as mean ± standard deviation,discrete variables as number (percentage). BMI = body massindex; CAD = coronary artery disease; LDL = low-densitylipoprotein; HDL = high-density lipoprotein; CRP = C-reactiveprotein; RCA = right coronary artery; LCx = left circumflexartery; HTN = hypertension.

from a noncardiac cause. These data were alsoexcluded. Of the remaining 20 subjects (meanage 53 ± 7 years), 15 were male and 5 were fe-male. Baseline subject characteristics are listed inTable I.

There was a decrease in total cholesterol (23%reduction), triglycerides (19% reduction), andLDL (41% reduction) at the 3-month visit fromthe baseline (P = 0.01), which was sustained overthe study period (Fig. 2). There was no change inthe HDL level.

The intra- and interoperator variability of thismethod was r = 0.86 (P < 0.001), r = 0.86 (P <0.001), respectively. The intraclass correlation co-efficient for intra and interoperator variability was0.89 (P < 0.001) and 0.85 (P < 0.001), respec-tively. Using Bland–Altman analysis, the mean ofthe variance in measurements within the sameoperator for the LAD wall thickness was 0.04 mm,representing an intraoperator variability of 3%.The mean of the variance in measurements be-tween two different operators for the LAD wallthickness was 0.08 mm, representing an interop-erator variability of 6%.

Overall, there was no change in the LADwall thickness, external or vessel luminal diame-ter over the 12-month period (Fig. 3). There were11 (55%) subjects who demonstrated some de-gree of regression despite there being no over-all regression in the total cohort. The percentage

643

Perry, et al.

Figure 2. Graph demonstrating the change inblood cholesterol levels over the study period. HDL= high-density lipoproteins; LDL = low-densitylipoproteins.

Figure 3. Graph demonstrating the change in leftanterior descending coronary artery wall thickness,luminal diameter, and external diameter over thestudy period.

Table II

Univariate Analyses of Predictors of Percentage Change inLeft Anterior Descending Coronary Artery (LAD) Wall

Thickness

Variable r-Value P-Valve

Age –0.53 0.04Sex –0.34 0.16Baseline LAD wall thickness 0.76 0.001Baseline BMI 0.12 0.67Baseline total cholesterol –0.38 0.89Baseline triglyceride –0.16 0.56Baseline LDL –0.11 0.74Baseline HDL –0.29 0.36Baseline CRP 0.18 0.62Homocysteine 0.18 0.70Creatinine –0.36 0.18Location of MI –0.23 0.43HTN 0.36 0.18Diabetes 0.35 0.15Smokers –0.19 0.44Family history CAD 0.13 0.66

BMI = body mass index; CAD = coronary artery disease;LDL = low-density lipoprotein; HDL = high-density lipopro-tein; CRP = C-reactive protein; RCA = right coronary artery;LCx = left circumflex artery; MI = myocardial infarction;HTN = hypertension.

change in the LAD wall thickness was negativelycorrelated with age (r = –0.53, P = 0.04) andpositively correlated with the baseline LAD wallthickness (r = 0.76, P = 0.001) (Table II).

Table III

Univariate Analyses between Baseline Left AnteriorDescending Coronary Artery (LAD) Wall Thickness and the

Changes in Serum Lipids, CRP, Homocysteine, and Creatinine

Variable Univariate Analysis (r) P-Valve

� Total cholesterol 0.24 0.41� Triglycerides 0.34 0.22� LDL 0.16 0.65� HDL –0.22 0.54� CRP –0.04 0.93� Homocysteine 0.69 0.09� Creatinine 0.48 0.08

LDL = low-density lipoprotein; HDL = high-density lipopro-tein; CRP = C-reactive protein.

Changes in the lipid, CRP, homocysteine, andcreatinine levels, which can affect the devel-opment of coronary atherosclerosis, were notrelated to the baseline LAD wall thickness, sug-gesting that the baseline LAD wall thickness canpredict percentage change in the LAD wall thick-ness independently of the aforementioned vari-ables (Table III).

Discussion:This study supports the ability of HRTTE to de-tect subclinical atherosclerosis in the LAD. More-over, these measurements are reproducible over

644

Echo Detection of Atherosclerotic Regression

time and may be used to detect stabilization ofplaque. IVUS studies have demonstrated a reduc-tion of plaque burden with intensive high-dosestatin therapy9 and stabilization of plaque bur-den with high-dose statin therapy,8,9 thus in ourcohort, stabilization or minor progression of theLAD wall thickness was a likely outcome due tothe moderate statin dose given. However, unlikeIVUS, HRTTE is noninvasive, fast, technically sim-ple, and essentially risk free.

As expected, there was a decrease in totalcholesterol, triglycerides, and LDL at the 3-monthvisit which was sustained at subsequent visits.However, the change in serum lipid levels wasnot able to predict percentage change in the LADwall thickness. In fact, the only predictor of per-centage change of the LAD wall thickness was thebaseline LAD wall thickness in that the larger theLAD wall thickness was at the baseline the morelikely it was to demonstrate regression. This is asimilar finding to what was observed in both theREVERSAL8 and ASTEROID9 trials where the areawith the highest plaque burden underwent thegreatest degree of regression, even when the to-tal atheroma volume in the target vessel did notregress. This is intuitive, as a higher plaque bur-den at the baseline results in a larger potential forabsolute reduction.

Those with the baseline LAD wall thickness≥1.8 mm tended toward regression of wall thick-ness. There was no difference in the change in to-tal cholesterol, triglycerides, LDL, and HDL levelsbetween the subjects with the LAD wall thicknessabove and below the cutoff value. It is possiblethat with more aggressive lipid-lowering, regres-sion in the LAD wall thickness may have beenseen, particularly in subjects with a higher plaqueburden.

At the baseline, all subjects demonstrated ab-normal levels of the LAD wall thickness (Fig. 4),and despite a level of regression in some sub-jects, none reduced to a normal level as deter-mined in our previous studies.11 The only othervariable to be associated with wall thickness wasage; intriguingly this was an inverse relationship.The impression that younger patients seemed tohave more potential for LAD wall thickness reduc-tion on HRTTE might be of major future clinical

significance, particularly in the primary preven-tative setting. While older patients clearly havemore coronary atherosclerosis, the regression inyounger patients’ coronaries might reflect a moreactive stage of coronary atherosclerosis and a win-dow where statin therapy might be of greatestclinical value.

Therapeutic intervention and risk analysis ofthe individual patient post-MI may be assistedby the HRTTE method. The simple noninvasivenature of the test makes it easily applicable andcould potentially be used in follow-up and dosetitration of preventative medications; however,this requires further validation.

Limitations:This is a small cohort and therefore it may nothave reached statistical significance; however,given the final sample size of n = 20, the powerof the trial to detect a 30% reduction in the LADwall thickness associated with statin therapy ex-ceeded 90%. This is a proof of the concept studyand paves the way for larger trials to explore thistechnique.

There is a potential risk of confusion of theLAD for the great cardiac vein as in most casesboth vessels run parallel to each other along theanterior surface of the heart. This risk was mini-mized, however, by measurement of the thickerwalled vascular structure visualized and by usingpulsed wave Doppler analysis to define the arterialfrom the venous blood flow in a subset of patients(Fig. 5).

Finally, although it is also possible that varia-tions in coronary anatomy made it difficult to visu-alize the LAD, the proximal LAD is a very rare siteof congenital anatomic abnormality (reported as<0.1%),13 and thus congenital coronary anoma-lies would probably not contribute in this study.

Conclusion:The increased wall thickening seen in our co-hort despite the benign angiographic appearanceof the proximal and mid-LAD region indicatesthat this HRTTE method may be more sensitivethan angiography for the detection of subclini-cal atherosclerosis. Furthermore, this method may

Figure 4. High-resolution transtho-racic echocardiography of the left an-terior descending coronary artery (LAD)in a normal subject A. and in a studysubject B. Arrows demonstrate anteriorLAD wall thickness.

645

Perry, et al.

have potential in the evaluation of response topreventative therapies and allow individual tailor-ing of both primary and secondary preventionstrategies.

References1. Brown BG, Zhao XQ, Sacco DE, et al: Arteriographic view

of treatment to achieve regression of coronary atheroscle-rosis and to prevent plaque disruption and clinical cardio-vascular events. Br Heart J 1993;69:S48–S53.

2. Scandinavian Simvastatin Survival Study Group: Ran-domised trial of cholesterol lowering in 4444 patientswith coronary heart disease: The Scandinavian Simvas-tatin Survival Study (4S). Lancet 1994;344:1383–1389.

3. Cannon CP, Braunwald E, McCabe CH, et al: Inten-sive versus moderate lipid lowering with statins afteracute coronary syndromes. N Engl J Med 2004;350:1495–1504.

4. Sacks FM, Pfeffer MA, Moye LA, et al: The effect ofpravastatin on coronary events after myocardial infarc-tion in patients with average cholesterol levels. Choles-terol and Recurrent Events Trial investigators. N Engl JMed 1996;335:1001–1009.

5. Brown BG, Zhao XQ, Sacco DE, et al: Lipid lowering andplaque regression. New insights into prevention of plaquedisruption and clinical events in coronary disease. Circu-lation 1993;87:1781–1791.

6. Pedersen TR, Wilhelmsen L, Faergeman O, et al: Follow-

up study of patients randomized in the Scandinavian sim-vastatin survival study (4S) of cholesterol lowering. Am JCardiol 2000;86:257–262.

7. Libby P: Molecular bases of the acute coronary syn-dromes. Circulation 1995;91:2844–2850.

8. Nissen SE, Tuzcu EM, Schoenhagen P, et al: Effect of in-tensive compared with moderate lipid-lowering therapyon progression of coronary atherosclerosis: A randomizedcontrolled trial. JAMA 2004;291:1071–1080.

9. Nissen SE, Nicholls SJ, Sipahi I, et al: Effect ofvery high-intensity statin therapy on regression ofcoronary atherosclerosis: The ASTEROID trial. JAMA2006;295:1556–1565.

10. Gradus-Pizlo I, Sawada SG, Wright D, et al: Detec-tion of subclinical coronary atherosclerosis using two-dimensional, high-resolution transthoracic echocardiog-raphy. J Am Coll Cardiol 2001;37:1422–1429.

11. Perry R, De Pasquale CG, Chew DP, et al: Changes inleft anterior descending coronary artery wall thicknessdetected by high-resolution transthoracic echocardiog-raphy. Am J Cardiol 2008;101:937–940.

12. Perry R, Joseph MX, De Pasquale CG, et al: High-resolution transthoracic echocardiography of the left an-terior descending coronary artery: A novel noninvasiveassessment of coronary vasoreactivity. J Am Soc Echocar-diogr 2008;21:134–138.

13. Pelliccia A: Congenital coronary artery anomalies in youngpatients: New perspectives for timely identification. J AmColl Cardiol 2001;37:598–600.

646