Comparative analysis of the accuracy of visual qualitative assessment and semi-quantitative analysis in 18F-florbetaben β-amyloid imaging / 中华核医学与分子影像杂志

ABSTRACT

Objective:To compare the accuracy of visual qualitative assessment and semi-quantitative analysis for 18F-florbetaben ( 18F-FBB) β-amyloid (Aβ) imaging in the diagnosis of Alzheimer′s disease (AD) and to explore their clinical application value. Methods:From January 2019 to October 2019, 17 patients (8 males, 9 females, age (74.1±8.5) years) with mild/moderate-stage clinically probable AD and 17 cognitive normal control (NC; 9 males, 8 females, age (64.5±6.3) years) were prospectively enrolled in this study. All patients underwent dynamic 18F-FBB PET/CT brain imaging in the Chinese PLA General Hospital. Visual qualitative assessment and semi-quantitative analysis methods were used to analyze PET brain imaging results. The difference of standardized uptake value ratio (SUVR) between the two methods was analyzed by using independent sample t test. The consistency of the two methods and clinical results was analyzed by Kappa test. The cut-off value of SUVR for the diagnosis of AD was determined by receiver operating characteristic (ROC) curve. Results:The sensitivity, specificity and accuracy of visual qualitative assessment to diagnose AD were 14/17, 16/17 and 88.2% (30/34). The global SUVR of NC and AD group were 1.09±0.85 and 1.75±0.25 ( t=-10.263, P<0.001), and the composite SUVR were 1.16±0.57 and 1.89±0.15 ( t=-10.789, P<0.001), respectively. The cut-off value of SUVR for the diagnosis of AD was 1.47, with the sensitivity of 15/17, the specificity of 16/17 and the accuracy of 91.2%(31/34). The visual qualitative assessment and semi-quantitative analysis had good consistency with clinical diagnosis results with Kappa value of 0.765 and 0.824 respectively (both P<0.001). Conclusion:The visual qualitative assessment and semi-quantitative analysis methods used in 18F-FBB Aβ imaging to diagnose AD patients show high accuracy and can provide effective value for clinical diagnosis, but the visual qualitative assessment method is concise and easy to grasp, which is worth further promotion and use in clinical.