Abstract
3008
Objectives: Dopamine transporter single-photon emission computed tomography (DAT SPECT) imaging has been widely used to diagnose Parkinson’s disease (PD) and can improve the diagnostic ability to differentiate PD from non-PD. In DAT SPECT, a visual interpretation and quantitative analysis is performed to evaluate the 123I-ioflupane uptake in the striatum. The specific binding ratio (SBR) and asymmetry index (AI) are quantitative indices used in DAT SPECT. The standardized uptake value (SUV) was first introduced with positron emission tomography (PET). Quantification using the SUV is one advantage of PET over SPECT. A program that can quantitatively analyze bone SPECT results using 99mTc-hydroxymethylene diphosphonate (99mTc-HMDP) has been developed and implemented; this program calculates the SUV, enabling the quantitative evaluation of the uptake degree. We adapted the SUV for DAT SPECT and propose a new quantitative index “functional dopamine transporter volume” (f-DTV).
Methods: Before SUV measurement, the Becquerel calibration factor (BCF) of 123I when using the same program was obtained by a phantom. Thirty patients were enrolled in this study (15 PD and 15 non-PD patients. 17 men, 13 women, 24-81 years old). 123I-ioflupane 167 MBq was injected intravenously, and images were obtained 3 h later using SPECT/CT. CT attenuation correction and scatter correction (TEW) were applied for image reconstruction. The regions of interest were set according to the guideline, and the SBR and AI were calculated using the program (DaT view) specialized for DAT SPECT, while the SUV was calculated using the quantification program for bone SPECT (GI-BONE). We calculated the f-DTV as the active dopamine volume (ADV) × SUVmean. We evaluated the correlation between the SBR and SUV and the SBR and f-DTV. The diagnostic ability of the SBR, SUV and f-DTV to differentiate PD from non-PD was also assessed using a receiver operating characteristic (ROC) analysis, and the area under the curve was calculated. Voxels of interest were set three ways: ADV=11.2 ml, threshold=40%, or SUV≥3.
Results: There was a strong correlation between the SUV and SBR and the f-DTV and SBR. Correlation coefficients: SUVmax vs. SBR, r=0.839; SUVpeak vs. SBR, r=0.852; SUVmean(1) vs. SBR, r=0.763; SUVmean(2) vs. SBR, r=0.767; SUVmean(3) vs. SBR, r=0.733; f-DTV(2) vs. SBR, r=0.874; f-DTV(3) vs. SBR, r=0.82. In the ROC curve, f-DTV had a higher diagnostic ability than SBR and SUV. ROC (AUC): SBR=0.867, SUVmax=0.904, SUVpeak=0.911, SUVmean(1)=0.860, SUVmean(2)=0.859, SUVmean(3)=0.783, f-DTV(1)=0.896, f-DTV(2)=0.984. SBR: Cut-off value=5.65, Sensitivity=90%, Specificity=73.3%. SUVmax: Cut-off value=4.88, Sensitivity=76.7% Specificity=86.7%. SUVpeak: Cut-off value=3.4, Sensitivity=76.7%, Specificity=90.0%. SUVmean(1): Cut-off value=2.45, Sensitivity=86.7%, Specificity=70.0%. SUVmean(2): Cut-off value=2.30, Sensitivity=76.7%, Specificity=83.3%. SUVmean(3): Cut-off value=2.38, Sensitivity=70.0%, Specificity=73.3%. f-DTV(2): Cut-off value=26.6, Sensitivity=86.7%, Specificity=76.7%. f-DTV(3): Cut-off value=28.7, Sensitivity=93.3%, Specificity=96.7%.
Conclusions: We proposed the new quantitative index “f-DTV” on DAT SPECT. It was strongly correlated with SBR and had a high diagnostic ability for differentiating PD from non-PD. F-DTV can evaluate both the volume and degree of active dopamine transporter at once, making it a useful marker in Parkinson syndrome.