Carotid Plaque Inflammation Imaged by ¹⁸ F-Fluorodeoxyglucose Positron Emission Tomography and Risk of Early Recurrent Stroke

Background and Purpose— Plaque inflammation contributes to stroke and coronary events. 18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) identifies plaque inflammation-related metabolism. Almost no prospective data exist on the relationship of carotid 18 F-FDG uptake and early recurrent stroke. Methods— We did a multicenter prospective cohort study BIOVASC (Biomarkers/Imaging Vulnerable Atherosclerosis in Symptomatic Carotid disease) of patients with carotid stenosis and recent stroke/transient ischemic attack with 90-day follow-up. On coregistered carotid 18 F-FDG PET/computed tomography angiography, 18 F-FDG uptake was expressed as maximum standardized uptake value (SUV max ) in the axial single hottest slice. We then conducted a systematic review of similar studies and pooled unpublished individual-patient data with 2 highly similar independent studies (Dublin and Barcelona). We analyzed the association of SUV max with all recurrent nonprocedural stroke (before and after PET) and with recurrent stroke after PET only. Results— In BIOVASC (n=109, 14 recurrent strokes), after adjustment (for age, sex, stenosis severity, antiplatelets, statins, diabetes mellitus, hypertension, and smoking), the hazard ratio for recurrent stroke per 1 g/mL SUV max was 2.2 (CI, 1.1–4.5; P =0.025). Findings were consistent in the independent Dublin (n=52, hazard ratio, 2.2; CI, 1.1–4.3) and Barcelona studies (n=35, hazard ratio, 2.8; CI, 0.98–5.5). In the pooled cohort (n=196), 37 recurrent strokes occurred (29 before and 8 after PET). Plaque SUV max was higher in patients with all recurrence ( P <0.0001) and post-PET recurrence ( P =0.009). The fully adjusted hazard ratio of any recurrent stroke was 2.19 (CI, 1.41–3.39; P <0.001) and for post-PET recurrent stroke was 4.57 (CI, 1.5–13.96; P =0.008). Recurrent stroke risk increased across SUV max quartiles (log-rank P =0.003). The area under receiver operating curve for all recurrence was 0.70 (CI, 0.59–0.78) and for post-PET recurrence was 0.80 (CI, 0.64–0.96). Conclusions— Plaque inflammation-related 18 F-FDG uptake independently predicted future recurrent stroke post-PET. Although further studies are needed, 18 F-FDG PET may improve patient selection for carotid revascularization and suggest that anti-inflammatory agents may have benefit for poststroke vascular prevention.