Statin-Induced Cholesterol Lowering and Plaque Regression After 6 Months of Magnetic Resonance Imaging Monitored Therapy

Transcription

1 Statin-Induced Cholesterol Lowering and Plaque Regression After 6 Months of Magnetic Resonance Imaging Monitored Therapy João A.C. Lima, MD; Milind Y. Desai, MD; Henning Steen, MD; William P. Warren, MD; Sandeep Gautam, MD; Shenghan Lai, MD, PhD Background Statin therapy reduces adverse outcomes, with a minimal decrease in vessel stenosis. Magnetic resonance imaging (MRI) noninvasively detects atherosclerotic plaque (AP) reduction. We hypothesized that statin-induced AP regression can be monitored by MRI and detected earlier than previously reported and is significantly associated with its lipid-lowering effect. Methods and Results APs in thoracic aorta were measured by combined surface/transesophageal MRI in 27 patients (treated with simvastatin 20 to 80 mg daily) before and after 6 months of therapy. AP volume and luminal dimensions were measured from 6 cross sections used to construct a 2.4-cm 3D volume of the aorta that included plaque and lumen. Method reproducibility was studied in 10 patients imaged twice, 1 week apart. AP volume was reduced from to cm 3 at 6 months (P 0.02), whereas luminal volume increase was less accentuated (from to cm 3, P 0.06). LDL cholesterol decreased by 23% (from to mg/dl, P 0.05) in 6 months. AP regression (plaque volume/area reduction) was significantly related to LDL cholesterol reduction (P 0.02 and P 0.005, respectively), and luminal volume increase was inversely related to LDL cholesterol reduction (P 0.04). Plaque volume measurement was highly reproducible (intraclass correlation R 0.98 and variability 4.8%). Intraobserver (0.91) and interobserver (0.81) concordances were documented for plaque volume assessment. Conclusions AP regression and reverse remodeling can be detected accurately by MRI 6 months after statin therapy initiation, and it is strongly associated with LDL cholesterol reduction. (Circulation. 2004;110: ) Key Words: lipids atherosclerosis plaque magnetic resonance imaging Atherosclerosis remains the leading cause of death in industrialized societies, and its incidence is projected to increase worldwide in the next 2 decades. 1 The benefits of statin therapy on atherosclerosis have been clearly established in several clinical trials. Typically, sustained statin therapy induces LDL cholesterol (LDL-C) reductions of 30% to 40%, which translate into mortality reductions of 25% to 30%. 2 5 However, the clinical benefits of decreased clinical events are not proportional to parallel reductions in vessel stenoses. 6 This apparent discrepancy can be explained by the results of previous studies documenting plaque regression in the face of minor alterations in luminal stenosis through reverse vessel remodeling. 7 Although plaque regression is considered to be a significant determinant of statin-induced clinical benefits in both primary and secondary prevention, 8 its mechanisms and time course remain incompletely understood. Direct visualization of atherosclerotic plaques by imaging techniques 9 has the potential to provide important insights into the pathophysiology of plaque regression secondary to lipid-lowering therapy. Results from clinical trials using carotid ultrasound 10,11 and intravascular ultrasound 12 and prospective studies using MRI 13 suggest that plaque regression can be documented only after at least 1 year of statin therapy. Conversely, pathological studies have documented statin-induced changes in plaque composition as early as 3 months after therapy initiation. 14 More recently, a small trial using intravenous apolipoprotein A-I Milano to raise HDL cholesterol levels has documented plaque regression by intravascular ultrasound after only 6 weeks of therapy. 15 These observations challenge the current paradigm of atherosclerosis as a chronic process that evolves slowly and requires prolonged lipid-lowering therapy to regress. They have also refocused attention on the mechanisms of statin-induced atherosclerotic plaque regression, which have not been entirely elucidated. Although most previous studies have attributed statin-induced plaque regression to changes in LDL 10,11,16 and/or HDL 6,15 cholesterol, plaque regression has also been documented in response to other types of therapies that do not directly influence lipid levels Moreover, although statin-induced lipid lowering and clinical benefits Received March 29, 2004; de novo received May 27, 2004; revision received June 30, 2004; accepted July 13, From the Cardiology Division of the Department of Medicine (J.A.C.L., M.Y.D., H.S., W.P.W., S.G.), the Department of Radiology (J.A.C.L.) of the School of Medicine, and the Department of Epidemiology of the Bloomberg School of Hygiene and Public Health (J.A.C.L., S.L.), Johns Hopkins University, Baltimore, Md. Correspondence to João A.C. Lima, MD, Cardiology Division, Blalock 524, Johns Hopkins Hospital, 600 N Wolfe St, Baltimore, MD jlima@jhmi.edu 2004 American Heart Association, Inc. Circulation is available at DOI: /01.CIR

2 Lima et al Statin-Induced Plaque Regression by MRI 2337 Figure 1. T2W magnetic resonance images (combined surface and transesophageal) of aortic arch (AA) of same patient obtained at baseline and at 6 months demonstrating excellent reproducibility for atherosclerotic plaque and luminal dimension assessment. Arrow points to bright signal from transesophageal magnetic resonance receiver coil. PA indicates pulmonary artery. may occur in a matter of weeks, 20 statin-mediated plaque regression has been measured in terms of years after the initiation of statin therapy. 6,8,10 13,16 These discrepancies highlight the need for greater insight into the mechanisms and time course of statin-induced plaque regression. MRI, as a noninvasive technique that allows for serial visualization of atherosclerotic plaque morphology, is particularly suited to monitor atherosclerosis in humans and animal models. 9,13,21,22 Paramount to the utilization of MRI to monitor atherosclerosis, however, is the demonstration that the method is both accurate and reproducible, 9 features that are in large part dependent on the ratio of MRI signal to background noise. Transesophageal MRI (TEMRI) is a novel technique that can be combined with standard surface MRI to enhance the ability to image the aortic arch and proximal descending aorta, 23,24 where aortic atherosclerotic plaques are frequently found. We hypothesized that statin-induced plaque regression can be monitored clinically by the combined transesophageal and surface-coil MRI approach, is significantly associated with its lipid-lowering effect, and can be detected noninvasively earlier than previously reported. Methods Study Design In this prospective study, after screening, 59 patients were deemed eligible, and 27 agreed to be enrolled in the study. We included patients with documented atherosclerosis in at least 1 vascular territory: at least moderate ( 3.9 mm) aortic atherosclerosis seen on transesophageal echocardiography, moderate coronary artery disease ( 50% lesion) in at least 1 coronary artery seen at cardiac catheterization, 50% carotid lesion seen on ultrasound, or documented peripheral vascular disease. These patients are enrolled in a randomized trial testing the effects of high- versus low-dose statin (20 versus 80 mg of simvastatin) therapy on aortic plaque regression. We excluded patients with pacemakers, automated implanted cardioverter defibrillators, aneurysm clips, abnormal nasopharyngeal anatomy, active peptic ulcer disease, severe dysphagia, abnormally elevated baseline liver transaminases ( 2 normal), a clinically significant medical event within 3 months before study entry, decompensated congestive heart failure, or inability to give informed consent. Diabetes was defined as fasting blood sugar level 126 mg/dl or use of antiglycemic therapy. Hypertension was defined as a blood pressure 140 mm Hg systolic and/or 90 mm Hg diastolic or use of antihypertensive medications. Smoking history was considered if a subject smoked at any time in the past or did so currently. Positive family history for coronary artery disease was considered when subjects reported physician-diagnosed coronary artery disease or myocardial infarction in a parent, grandparent, or sibling before age 60. A history of myocardial infarction or stroke was self-reported by the patient, on the basis of a physician-based diagnosis. All patients were using statin therapy at baseline, and patients agreed to be switched to simvastatin. Other laboratory studies, including creatine kinase, aspartate aminotransferase, alanine aminotransferase, and serum creatinine levels, were obtained every 3 months; in addition, TEMRIs were obtained at baseline and 6 months. Patients were enrolled in the study after providing written consent according to The Johns Hopkins Committee on Clinical Investigation and General Clinical Research Center guidelines. MRI Protocol MRI of the thoracic aorta was performed in a 1.5-T (40 mt/m) Sigma (General Electric) magnet using a standard flex coil (General Electric) and a specially designed 2-element cardiac phased-array coil positioned on the chest wall (both inches in dimension). We used a TEMRI receiver coil (Intercept Esophageal MR Coil, Surgi-Vision) in the remaining coil position (approximate diameter 8F). It was positioned in the esophagus through a 9F to 10F nasogastric tube by an experienced physician. ECG-gated fast spin-echo and inversion recovery radiofrequency pulses were used to produce a black-blood double oblique sagittal image (candy cane view) of the thoracic aorta. A chemical shift suppression pulse was used to suppress the signal from perivascular fat for proton density weighted (PDW) images. The thickest plaque was identified in the aorta, and through this area, 6 contiguous images (to reduce submillimeter errors in matching of the images at different time points) with a 4-mm slice thickness were obtained (1 slice/breath hold). We imaged perpendicular to the vessel wall (using the double-inversion recovery pulse sequence) with T2-weighted (T2W) and PDW techniques during breath holds (11 to 15 seconds for PDW and 15 to 20 seconds for T2W images). All participants were able to complete the study, and repetitive breath holds were not required. Additional imaging parameters were as follows: 16- to 24-cm field of view, repetition time 2 RR intervals, echo delay time 20 ms (PDW) and 65 ms (T2W), image matrix , echo train length 16 to 24, 1 NEX (number of excitations), no phase wrap, and spatial resolution of 0.63 to 1.41 mm 2. The total duration was 40 minutes (actual scan time was 30 minutes). Image Analysis Follow-up MR images of same patient were reproduced by use of imaging planes with various identical anatomical landmarks (eg, pulmonary arteries; Figure 1). Data from each receiver coil (1 anterior, 2 posterior, and 1 transesophageal) were obtained individually, along with a composite image of all coils (combined). The composite image was used for all plaque and lumen analysis. The MR images were analyzed with Scion image 4.02 software (Scion Corporation) in a blinded manner at a different time than when the patients were imaged. T2W images were used for plaque analysis because this pulse sequence resulted in much better blood suppression, reducing flow artifacts at the plaque/blood interface while maintaining an adequate signal-to-noise ratio and resulting in better

3 2338 Circulation October 19, 2004 plaque delineation. Both PDW and T2W images had a very high correlation for plaque volume measurements (r 0.97). The luminal and outer aortic wall boundaries were traced manually with a region-of-interest tool, and an average plaque area of all 6 slices was calculated by subtracting the inner luminal area from the outer aortic wall area. 3D plaque volume was calculated by integrating the area of plaque in the 2.4-cm (4-mm slices 6 slices) region of aorta imaged with a modification of the Simpson rule: [(image 1 area image 5 area) 4 (image 2 area image 4 area) 2 (image 3 area)]/3 [(image 5 area image 6 area)/2]. 14 This technique ensured increased emphasis on the 4 central slices, thus enabling superior reproducibility because it centered the slices on the same area of maximum plaque. Lumen volume was similarly calculated with the modified Simpson formula by integrating the luminal area from 6 slices in the same 2.4-cm region of the aorta. Reproducibility Studies and Statistical Methods In 10 patients, we obtained a second MRI within 1 week of the initial MRI examination for reproducibility of plaque size measurements. These studies were also analyzed by 2 independent observers using the 2D and 3D methodologies described above. For purposes of assessing method reproducibility, we calculated the coefficient of variation (SD 100/mean) and intraclass correlation coefficient R. The latter coefficient is defined as the proportion of variance due to differences between subjects divided by the total variance. 25,26 Data are expressed as mean SD. A paired Student t test was used to compare MRI and other parameters between baseline and 6 months. The generalized estimating equation approach was used to examine relationships between MRI and lipid parameters, with adjustment for potential confounding factors such as age, gender, and other factors. To examine whether MRI parameters are independently associated with lipid levels, an overall generalized estimating equation model included age, gender, family history, diabetes, smoking, and lipid parameters (including HDL, LDL, total cholesterol, and triglycerides). Those variables that failed to make a significant contribution to the model were eliminated in a stagewise manner, which yielded a final model. Univariate regression analysis was also used to study correlations between changes in lipid parameters and changes in MRI measures of plaque regression and vessel remodeling. All reported probability values are 2 sided, and a probability value of 0.05 was considered to indicate statistical significance. Results Patient Characteristics Twenty-seven patients treated with simvastatin (20 to 80 mg) were followed up for 6 months after therapy initiation. There were no adverse events reported during the course of the study, and changes in liver function tests were clinically insignificant. Baseline demographic data are shown in the Table. Serum Lipids Baseline serum levels of total cholesterol ( mg/dl) and LDL-C ( mg/dl) decreased to and mg/dl, respectively, at 3 months and to and mg/dl, respectively, at 6 months (P 0.05) after the onset of statin therapy. HDL cholesterol changed from to mg/dl (P NS), and triglyceride levels changed from to mg/dl (P NS). Reproducibility The reproducibility of plaque size measurements by the combined surface and transesophageal technique was studied in 10 patients with documented atherosclerosis who returned to a repeat examination within 7 days of the initial study. Demographics of the Study Age, y 69 9 Weight, kg Male gender 77% White race 90% Hypertension 69% Diabetes mellitus 17% Smoking 21% Positive family history 59% Cerebrovascular accident 34% Coronary artery disease 48% Myocardial infarction 31% Peripheral vascular disease 17% Prior statin use 100% Baseline systolic blood pressure, mm Hg Six-month systolic blood pressure, mm Hg Heart rate, bpm Cross-sectional plaque area measurements (2D) were closely correlated between study 1 and 2, with an intraclass correlation coefficient of R 0.91 and a coefficient of variation of 23.9%. However, the most reproducible index of plaque size was 3D plaque volume, with an intraclass correlation coefficient of R 0.97 and a coefficient of variation of 4.8%. Intraobserver and interobserver concordances were 0.91 and 0.81, respectively, for plaque volume. On the basis of these reproducibility studies, we calculated that changes in aortic plaque volume 4.6% can be considered as accurately measured by this MRI method. MRI Plaque Regression During the same period of time, plaque regression measured by MRI was significant for the entire group of patients. Plaque volume decreased from cm 3 at baseline to cm 3 at 6 months (12% reduction, P 0.02), whereas plaque area decreased from cm 2 at baseline to cm 2 (P 0.02), which reflects a 12% reduction (Figure 2). Concomitant changes in vessel luminal volume from cm 3 at baseline to cm 3 at 6 months were of borderline significance (P 0.06; Figure 3), whereas changes in cross-sectional luminal area for the group as a whole were not statistically significant (from to cm 2 at 6 months). Figure 2. Reduction in aortic atherosclerotic plaque volume (A) and area (B) after 6 months of statin therapy (both P 0.02).

4 Lima et al Statin-Induced Plaque Regression by MRI 2339 Figure 3. Change in aortic lumen volume (A) and area (B) after 6 months of statin therapy (both P NS). These results demonstrate that atherosclerotic plaque regression was detectable as early as 6 months after initiation of monitored simvastatin therapy. Multiple regression analysis was performed to examine whether these alterations were related to the magnitude of lipid-lowering changes observed during the same time period. Lipid Lowering and Plaque Regression Relationship Plaque regression and vessel remodeling had a significant correlation with changes in LDL-C after adjustment for age, gender, family history, presence of coronary artery disease, diabetes, and changes in triglycerides and HDL cholesterol. Plaque regression, measured as reductions in plaque volume or plaque area, was directly proportional to LDL-C reduction (regression coefficients 7.07 [95% CI 1.3 to 12.9] and 0.41 [95% CI 0.13 to 0.69], respectively; P 0.02 and P 0.005, respectively). Concomitant increases in vessel lumen volume were inversely related to LDL-C reduction (regression coefficient 70.6 [95% CI 3.3 to 137.9], P 0.04), whereas for lumen area, the relationship was of borderline statistical significance (regression coefficient 2.9 [95% CI 0.5 to 6.38], P 0.10). When changes in LDL-C were correlated by single regression analysis against changes in vessel lumen, LDL-C reduction explained 25% of vessel lumen volume augmentation and 31% of vessel cross-sectional area increase in the first 6 months of simvastatin therapy (P and P 0.003, respectively). Discussion This prospective clinical study demonstrates statin-induced plaque regression 6 months after the onset of MRI-monitored therapy. Simvastatin-induced LDL-C lowering was significantly correlated to plaque volume reduction and inversely correlated with arterial reverse remodeling associated with plaque regression, even after correction for other confounding variables. Finally, we also document the greater reproducibility of plaque volume versus plaque area assessment as a measure of regression of aortic atherosclerosis. Several previous studies have demonstrated the effect of lipid-lowering therapy in preventing plaque progression 11 relative to individuals unexposed to the intervention being tested. Other studies have documented actual regression of the atherosclerotic process expressed as an increase in angiographic vessel lumen 6 or a reduction in arterial wall thickness, 10,21 plaque area, 13 or plaque volume. 15 In the case of statin-induced atherosclerotic plaque modulation, these effects have been consistently reported after at least 1 year of therapy. 6,8,10,11,13,16 Temporal differences in the onset of documented plaque regression relative to the present study can be explained by differences in methods, type or magnitude of statin therapy, or the target study population. Previous studies using luminal changes 6 or changes in plaque thickness measured in the carotid or femoral arteries 10,11 may have had to wait a longer period of time before the biological effect could be detected. Similarly, less intensive statin regimens may have taken longer to produce differences in plaque thickness or vessel lumen. A previous prospective study using MRI found no evidence of plaque regression at 6 months but documented changes in plaque area measured in the aorta and carotid arteries 1 year after the onset of an identical regimen of simvastatin therapy. 13 In the present study, reductions in plaque volume and plaque area could be documented at the 6-month time point from studies performed in the aorta only and with somewhat different MRI methods. In addition, whereas only 1 target per patient was used in the present study, multiple targets were used in the previous study, which could have weighted its results toward nonrespondents or patients taking lower doses of simvastatin. However, despite methodological differences, it is also possible that the different time courses of plaque regression resulted from differences in the lipid-lowering responses among different groups of patients enrolled in the 2 studies. In the present study, plaque regression at 6 months was significantly correlated to LDL-C lowering. Participants in the present study also had significant baseline atherosclerosis, as suggested by the clinical characteristics shown in the Table. It is also possible that the time course and magnitude of plaque regression are directly proportional or easier to detect in patients with greater baseline burden, as suggested by studies in patients with as opposed to those without familial hypercholesterolemia, 10 as well as studies using the femoral wall thickness as opposed to carotid wall thickness measured by ultrasound. 10,16,27 Our understanding of atherosclerotic plaque regression is largely derived from pathological studies performed in experimental animals 21,28 31 and humans. 14,32 There is a large body of evidence indicating that atherosclerotic plaque lipid content is depleted with plasma cholesterol reduction 14 and that such reduction can be monitored experimentally by MRI. 21 Furthermore, concomitant reductions in plaque size and cholesterol levels have been documented in experimental studies without changes in collagen content. Therefore, plaque size reduction is likely to occur in response to reabsorption of lipids from the atherosclerotic plaque. 6 The results from the present study support this concept, given that a significant amount of regression effect seen at the 6-month time point was associated with changes in LDL-C reduction. They are also consistent with results from pathological studies that showed statin-induced atherosclerosis regression as early as 3 months after the initiation of therapy 14 and in part explain the salutary effects of statin therapy in terms of clinical outcomes 16 weeks after an index acute coronary syndrome, as documented in the Myocardial Ischemia Re-

5 2340 Circulation October 19, 2004 duction with Aggressive Cholesterol Lowering (MIRACL) study. 20 Methodological Considerations The data presented in the present study, as in the reports by Corti et al, 13 originate from a clinical trial comparing high 80 mg/d versus 20 mg/d simvastatin therapy and therefore comprise patients randomized to both arms of the trial; however, MRI data analysis was performed by blinded observers relative to all other study variables. Furthermore, in the study by Corti et al, 13 vessel wall and lumen areas were measured in 28 nonatherosclerotic arterial wall segments at baseline and 1 year after therapy without any detectable changes. The TEMRI technique has been validated in autopsy studies, 23,24 and its contribution to the combined approach has been quantified in previous work. 33 The present study also documents the superiority of plaque volume (intraclass correlation coefficient R 0.97 and coefficient of variation 4.8%) relative to plaque area (R 0.91 and coefficient of variation 23.9%). Measurement of plaque area per slice is more dependent on the exact repositioning of the slices, whereas volume measurements tolerate subtler errors caused by misregistration. Our study should be interpreted in light of certain limitations. The sample size is small, although the statistical testing shows that the results are significant results. The TEMRI technique is a semi-invasive procedure and therefore demands some further safety considerations, although none of the patients experienced any adverse effects, and all tolerated the procedure well without any sedation or anesthesia. Because of too few discrete plaques with discernible components, plaque characterization was not attempted during analysis. The field of view, although relatively large, was individually adjusted per patient (16 to 24 cm) according to their body habitus, and the parameters were consistent for the same patients during follow-up imaging. Another limiting factor was that fat suppression was not used for T2W images, which could obscure 1 side of the vessel wall. Finally, even though the reproducibility of plaque assessment is excellent (r 0.95), it is possible that the assessment error might be in the realm of statistically important events in patients with less severe disease, in whom this technique might be very helpful. Conclusions We demonstrate plaque regression 6 months after the onset of simvastatin therapy in patients with documented atherosclerosis. There appears to be a significant correlation between statin-induced lipid lowering and plaque regression, as well as reverse remodeling characterized by cross-sectional luminal increase. The study also demonstrates the feasibility of monitoring alterations of atherosclerotic plaque size by a novel MRI approach that enhances the surface MRI method by the placement of an additional transesophageal antennareceiver coil. Further studies are needed to examine the possibility that statin-induced regression can occur within the first 6 months of therapy, as well as its dependence on dose and magnitude of the disease process, before therapy initiation. Acknowledgments This work was supported by the Donald W. Reynolds Johns Hopkins Cardiovascular Center and by the National Institutes on Aging RO1-AG , NHLBI RO1-HL , and NO1- HC (Johns Hopkins MESA Field Center), the NIH/ NCRR grant MO1 RR00052 (General Clinical Research Center), and Merck Inc. We thank Elisabeth Brady, RN, for her collaboration in and support of this effort. References 1. Michaud CM, Murray CJL, Bloon BR. Burden of disease: implications for future research. JAMA. 2001;285: The Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344: Cholesterol and Recurrent Events (CARE) Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med. 1996;335: Shepherd J, Cobbe SM, Ford I, et al, for the West of Scotland Coronary Prevention Study Group. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med. 1995;333: Downs JR, Clearfield M, Weiss S, et al, for the AFCAPS/TexCAPS. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS. Air Force/Texas Coronary Atherosclerosis Prevention Study. JAMA. 1998;279: Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med. 2001;345: Genest J, Pedersen TR. Prevention of cardiovascular ischemic events: high-risk and secondary prevention. Circulation. 2003;107: Pitt B, Mancini GB, Ellis SG, et al. Pravastatin limitation of atherosclerosis in the coronary arteries (PLAC-I): reduction in atherosclerosis progression and clinical events. J Am Coll Cardiol. 1995;26: Redberg R, Vogel F, Criqui M, et al. Task Force # 3: what is the spectrum of current and emerging techniques for the non-invasive measurement of atherosclerosis? J Am Coll Cardiol. 2003;41: Nolting PRW, de Groot E, Zwinderman AH, et al. Regression of carotid and femoral artery intima-media thickness in familial hypercholesterolemia. Arch Intern Med. 2003;163: Mercuri M, Bond MG, Sirtori CR, et al. Pravastatin reduces carotid intima-media thickness progression in an asymptomatic hypercholesterolemic Mediterranean population: the Carotid Atherosclerosis Italian Ultrasound Study. Am J Med. 1996;191: Tagaki T, Yoshida K, Akasaka T, et al. Intravascular ultrasound analysis of reduction in progression of coronary narrowing by treatment with pravastatin. Am J Cardiol. 1997;79: Corti R, Fuster V, Fayad ZA, et al. Lipid lowering by simvastatin induces regression of human atherosclerotic lesions: two years follow-up by high-resolution non-invasive magnetic resonance imaging. Circulation. 2002;106: Crisby M, Nordin-Fredriksson G, Shah PK, et al. Pravastatin treatment increases collagen content and decreases lipid content, inflammation, metalloproteinases, and cell death in human carotid plaques: implications for plaque stabilization. Circulation. 2001;103: Nissen SE, Tsunoda T, Tuzcu EM, et al. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes. JAMA. 2003;290: de Groot E, Jukema JW, van Swijndregt ADM, et al. B-mode ultrasound assessment of pravastatin treatment effect on carotid and femoral artery walls and its correlations with coronary arteriographic findings: a report of the Regression Growth Evaluation Statin Study (REGRESS). JAm Coll Cardiol. 1998;31: Wiklund O, Hulthe J, Wikstrand J, et al. Effect of controlled release/ extended release metoprolol on carotid intima-media thickness in patients with hypercholesterolemia: a 3-year randomised study. Stroke. 2002;33:

6 Lima et al Statin-Induced Plaque Regression by MRI Simon A, Gariepy J, Chironi G, et al. Intima-media thickness: a new tool for diagnosis and treatment of cardiovascular risk. J Hypertens. 2002;20: Fang JC, Kinlay S, Beltrane J, et al. Effect of vitamins C and E on progression of transplant-associated arteriosclerosis: a randomised trial. Lancet. 2002;359: Schwartz GG, Olsson AG, Ezekowitz MD, et al, for the Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) Study Investigators. Effect of atorvastatin on early recurrent ischemic events in acute coronary syndromes. JAMA. 2001;285: McConnell MV, Aikawa M, Maier SE, et al. MRI of rabbit atherosclerosis in response to dietary cholesterol lowering. Arterioscler Thromb Vasc Biol. 1999;19: Yuan C, Beach KW, Smith LH Jr, et al. Measurement of atherosclerotic carotid plaque size in vivo using high resolution magnetic resonance imaging. Circulation. 1998;98: Shunk KA, Lima JA, Heldman AW, et al. Transesophageal magnetic resonance imaging. Magn Reson Med. 1999;41: Shunk KA, Garot J, Atalar E, et al. Transesophageal magnetic resonance imaging of the aortic arch and descending thoracic aorta in patients with aortic atherosclerosis. J Am Coll Cardiol. 2001;37: Fleiss JL. The Design and Analysis of Clinical Experiments. New York, NY: Wiley; 1986: Rosner B. Fundamentals of Biostatistics. Belmont, Calif: Duxbury Press; 1995: Ubels FL, Muntinga JH, van Doormaal JJ, et al. Effects of initial and long-term lipid-lowering therapy on vascular wall characteristics. Atherosclerosis. 2001;20: Armstrong ML, Megan MB. Arterial fibrous protein in cynomolgus monkeys after atherogenic and regression diets. Circ Res. 1975;36: Clarkson TB, Bond MG, Bullock BC, et al. A study of atherosclerosis regression in Macaca mulatta, IV: changes in coronary arteries from animals with atherosclerosis induced for 19 months and then regressed for 24 or 48 months at plasma cholesterol concentrations of 300 or 200 mg/dl. Exp Mol Pathol. 1981;34: Small DM, Bond MG, Waugh D, et al. Physiochemical and histological changes in the arterial wall of nonhuman primates during progression and regression of atherosclerosis. J Clin Invest. 1984;73: Sukhova GK, Shi GP, Simon DI, et al. Expression of the elastolytic cathepsins S and K in human atheroma and regulation of their production in smooth muscle cells. J Clin Invest. 1998;102: Glagov S, Weisenberg E, Zarins CK, et al. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med. 1987;316: Warren WP, Heath S, Lima JAC. Combined transesophageal and surface MRI provides optimal imaging of aortic atherosclerosis. J Am Coll Cardiol. 2003;41:458A. Abstract.