Is the process of stabilization of carotid plaque more dynamic than expected? a high-resolution 3D-CMR statin-naive human study

Introduction

Atherosclerosis is a dynamic process thought to stabilize with statin therapy. However, the uniformity of plaque stabilization and subsequent regression when examined under high-resolution 3D CMR is unknown. We hypothesize that plaque characteristics illustrate marked heterogeneity with both plaque regression and progression as lipid lowering therapy is instituted.

Purpose

Methods

Via CMR (1.5T GE), 707-two mm contiguous in vivo slices of advanced carotid disease (>50%; mean 64±21 ) representing 42 complete bilateral human plaques (age 68±15yrs) were analyzed for 2D/3D extent of vascular wall: lipid pool, fibrous cap, outer wall area (OWA), vessel wall area (VWA), lumen area (LA) and lipid pool (LP). All were related to fasting lipids relative to %stenosis via QPlaque (Medis). Plaque morphology was determined by CMR (T1/T2/PD) at baseline and one year following lipid lowering agent (simvastatin or simva/ezetimibe). Plaque progression was defined as LP pre/LP post < 1 while plaque regression was defined as LP pre/LP post >1.

Results

39/42 in vivo plaques in statin naïve pts were successfully imaged. Resolution: 1x1x2mm. Pre therapy, mg/dL range of LDLC was 60-189 (mean 142), HDLC: 23-71 and TG: 80-214. LP represented 30±4% and fibrous laque 9±22% of total vessel wall. Post therapy, LDL was 66±31mg/dL. In 707 slices, 378 (53.5%) demonstrated progression while 329 (46.5%) showed regression. In those plaques that regressed there was significant decrease in both OWA, VWA and fibrous plaque as well as a paradoxical decrease in LA (p<0.0001 for all) while he converse was true for progression (p <0.0001 for all but LA). Specifically, LA decreased from 27.0 to 21.8mm² to (19%; p<0.0001) while LA increased from 24.4 to 26.0mm² (9%; p=NS). Segmenting for quartiles of LDL favorably trended with ΔLA; r=0.35, p=0.08 while lumen size was related to ΔLDL, p<0.02).

Conclusions

In statin naïve pts, administration of lipid modulating agents appear to have initial paradoxical effects on lumen size as assessed by high-resolution 3D CMR: as VWA and LP decreases so does the LA. Similarly, as VWA and LP increases so does LA. All appears driven by the effectiveness of the ΔLDL achieved. Thus, marked heterogeneity in plaque compositional changes exist that are only resolved once ΔLDL is known.