Comparative Study of F-18-FDG PET/CT Images Quality in the Diabetic and Nondiabetic Patients
Ziwei Zhu,Zhi Chang,Xiaofen Xie,Xiang Li,Marcus Hacker,Hongzhi Mi,Xiaoli Zhang
2018-01-01
Journal of Nuclear Medicine
Abstract:1568 Objectives: To obtain 18F-FDG PET/CT cardiac viability imaging with good quality in diabetic patients still remains a challenge. The aim of this study was to evaluate the imaging quality of 18F-FDG PET/CT in diabetic and non-diabetic patients with coronary artery disease (CAD) by using standardized fasting + oral glucose loading and intravenous injection of insulin protocol. Furhter, to analyze the factors which may affect the imaging quality and guide our routine clinical practice. Method 257 consecutive patients (226 men, age: 58±10 y) with documented CAD underwent 18F-FDG PET/CT imaging (between June 2016 and Nov. 2017) for viability assessment. Cardiac 18F-FDG PET/CT imaging was performed using standardized fasting + oral glucose loading and intravenously injection of insulin protocol. There were 58 diabetic patients (22.6%) and 199 (77.4%) non-diabetic patients. The quality of 18F-FDG PET/CT images was visually scored by 0-4 (0, excellent; 1, very good; 2, good; 3, fair, and 4, poor). The scoring criteria was based on the noisy signal in blood, liver, spine and other visceral organ nearby the heart related to myocardial uptake. Images with core 0-2 were considered as acceptable, whereas unacceptable images had score of 3-4. Factors may affect image quality were analyzed. These factors included oral glucose loading (g), the unit of insulin injection (u), blood glucose (BG, mmol/L) level at fasting, at the time of FDG injection, and at peak (40-60min after FDG injection). BG increasing rate [(peak BG-fasted BG)/fasted BG×100%], and BG decreasing rate [(peak BG-BG at injection FDG)/peak BG×100%] were also calculated. Considering that impaired glucose tolerance(IGT) may exist in part of non-diabetic patients, we also divided patients into two groups with the threshold of fasted BG>7.0 mmol/L. Finally, factors which may affect the imaging quality were compared with or without diabetes. Result The distribution of images’ score in diabetic and non-diabetic patients was shown in Figure 1. Images with good to excellent scores have been found in 98% cases in non-diabetic patients, whereas 84% of PET images were classified as good quality in diabetic patients (P>0.05). There was significant difference of fasted BG, peak BG, oral glucose loading, BG decreasing rate (%) and insulin i.v. (u) among the four subgroups, all P value 7.0 (n= 77), 17% of patients (n=11) had unacceptable imaging quality, it was significantly higher than that only 1% of patients with fasted BG≤7.0 (n= (χ2=19.49, P 7.0, lower FDG dose was injected in 11 patients with unacceptable imaging quality in comparison with 66 patients with acceptable imaging quality (5.3 ± 0.8 vs. 4.6 ± 0.9, P 7.0mmol/L, it remains a challenge for obtaining good or excellent imaging quality. [Research support]Natural Science Foundation of China(No.81571717, No.81071177).