18F-sodium fluoride PET/CT provides prognostic clarity compared to calcium and Framingham risk scoring when addressing whole-heart arterial calcification

Abstract

To investigate the benefit of utilizing 18F-sodium fluoride (NaF) PET/CT over calcium and Framingham scoring for potential preventative coronary artery disease (CAD) intervention. This retrospective study included 136 participants (ages 21–75, BMI 18–43 kg/m2): 86 healthy controls and 50 patients. CT heart segmentations were superimposed onto PET images and standard uptake values (SUV) were calculated by a semi-auto segmentation method of drawing volumes of interest around the heart. Intergroup comparisons were made matching 37 patient/control pairs based on age, gender, and BMI. ROC curves were generated to determine how well SUV and Framingham methods predicted patient status. Regressions including all 136 participants were performed between SUV, age, and BMI. Patients exhibited higher average SUV (SUVmean; P = 0.006) and Framingham scores (P = 0.02) than controls. However, ROC curves indicated that SUVmean could discriminate patients from controls (AUC = 0.63, P = 0.049), but Framingham scores could not (AUC = 0.44, P = 0.38). Calcium scores and maximum SUV (SUVmax) did not differ between patients and controls. SUVmean correlated with age and BMI among females (age, partial R2 = 0.16, P = 0.001; BMI, partial R2 = 0.12, P = 0.004) and males (age, partial R2 = 0.28, P < 0.0001; BMI, partial R2 = 0.22, P < 0.0001). Unlike calcium scores, NaF PET/CT-derived values differed between patients and controls. Framingham risk score patterns echoed those of SUVmean, but were not sensitive enough to predict patient status. SUVmean values increased with age and BMI. Therefore, incorporation of NaF PET/CT into routine prognostic CAD assessment might prove beneficial for assessing early stage plaque calcification in coronary arteries. ClinicalTrials.gov (NCT01724749).