Value of visual analysis and SUVR during 18F-AV45 PET/CT imaging in the diagnosis of mild cognitive impairment and Alzheimer′s disease / 中华核医学与分子影像杂志

ABSTRACT

Objective:To evaluate the value of visual analysis and standardized uptake value ratio (SUVR) during 18F-florbetapir (AV45) PET/CT brain imaging in diagnosis of β-amyloid (Aβ) deposition in patients with mild cognitive impairment (MCI) and Alzheimer′s disease (AD), and to explore the clinical ancillary value of the two indexes. Methods:From December 2018 to July 2019, a total of 47 subjects, including 5 (3 males, 2 females, age (58±13) years) normal controls (NC), 8 (2 males, 6 females, age (66±10) years) patients with AD and 34 (16 males, 18 females, age (70±7) years) patients with MCI were enrolled. All subjects underwent 18F-AV45 PET/CT scan. All images were evaluated by visual analysis and SUVR were calculated. The diagnostic efficiencies of visual analysis and SUVR were compared by McNemar test and Kappa test. One-way analysis of variance and Welch test were used to compare data differences. The best threshold value of SUVR was obtained by receiver operating characteristic (ROC) curve analysis. Results:The positive rate of Aβ deposition for all subjects was 46.81%(22/47) by SUVR analysis, and 38.30%(18/47) by visual analysis. There was no significant difference between the two methods ( χ2=33.15, P>0.05), and the consistency was good ( Kappa=0.83). Considering the clinical diagnosis as the"gold standard", the Aβ deposition obtained by visual analysis and SUVR analysis can effectively distinguish AD from NC, and the sensitivities were 7/8 vs 8/8, respectively, both specificities were 5/5( χ2=9.48, P>0.05), with good consistency ( Kappa=0.84). SUVR quantitative analysis could distinguish AD from NC, AD from MCI ( F values 3.99-8.79, all P<0.01), but could not distinguish NC from MCI (all P>0.05). ROC curve analysis showed that the best threshold value of precuneus′ SUVR was 1.08 for the differential diagnosis of AD and NC; for the differential diagnosis of AD and MCI, the best threshold value of lateral temporal′s SUVR was 1.06. Conclusion:Visual analysis was consistent with SUVR′s qualitative determination during 18F-AV45 PET/CT imaging for brain Aβ deposition, while SUVR quantitative analysis could assist in the differential diagnosis of AD and NC, AD and MCI.