1 Introduction [1,2] [3,4] [5] [6] [7] [8] [9,10] [6] [11] [12–14] [15] [16] [17,18] [19,20] [21] 2 Methods 2.1 Study population n 2.2 Vascular MRI protocol [17,18] Figs. 1b and 2b Fig. 2 [22,23] [3,4] [17,18] [24] 2.3 Serum and plasma assays Cholesterol and lipoprotein assays were performed using a Cobas-Mira Analyser (ABX Diagnostics, Shefford, UK). Total cholesterol was assayed using the enzymatic CHOD-PAP method and triglycerides were assayed using the enzymatic GPO-PAP method. HDL-cholesterol was assayed using a homogenous second generation PEGME method (Roche Diagnostics, Burgess Hill, UK). Apolipoprotein AI (apoAI) and apolipoprotein B (apoB) were assayed using immunoturbidimetric methods, using reagents supplied by ABX Diagnostics. C-reactive protein was analysed using ELISA (MP Biomedicals, UK) according to the manufacturer's instructions. 2.4 Statistical analyses χ 2 P 3 Results 3.1 Clinical and biochemical measures Table 1 Table 1 P P Table 2 χ 2 P χ 2 P χ 2 P < χ 2 P Table 3 3.2 Atherosclerosis regression Fig. 1 Fig. 2 P P P P P P χ 2 P Fig. 1 χ 2 P Fig. 2 Figs. 1c and 2c 2 2 P 2 P 2 2 P 2 P 3.3 Physiological measures Table 3 Table 3 3.4 Relationship between variables Within individual patients there was no correlation between MRI quantification of atheroma burden in the aorta and carotid at baseline. Although at a group level, endothelial function and aortic compliance improved and atheroma burden diminished, there was no correlation of these changes within individual patients. Furthermore, there were no significant associations between measures of vascular function or atheroma burden and any of: attained LDL-C; change in LDL-C, HDL-C, apoB, apoA-I, or CRP. 4 Discussion In this study, we have observed that regression of atherosclerosis in response to statin treatment can occur earlier than previously appreciated in both the aorta and carotid arteries. The robustness of this observation is enhanced by the finding that, within individual patients, regression at the early time point of 3 months was closely related to the magnitude and direction of change at 12 months. Patients also showed early and sustained improvement in aortic distensibility and in flow mediated vasodilatation of the brachial artery. [6,7,25] [6] [7] [5] [8] [6] P [26] [27,28] [29] [30,31] [32] [33] [34] [12,14] [13] [35,36] [37] [38] [39,40] [18] [6,22] [41] [42–44] [45] [46] 4.1 Study limitations [5,6] [26] 5 Conclusions This study shows that in a population of statin naïve, clinically stable but otherwise unselected coronary artery disease patients, cholesterol reduction using statins to mean LDL-C of approximately 70 mg/dL was associated with rapid regression of atheroma at 3 months. Early changes were highly correlated with changes after 12 months. These rapid structural changes were accompanied by early improvements in arterial stiffness and endothelial function that were sustained to 12 months. Use of multi-modal vascular MRI to detect early changes in atheroma and vascular function in small numbers of patients could prove to be an efficient strategy to screen novel anti-atherosclerotic agents.