TY - JOUR T1 - Diagnostic value and clinical significance of whole body PET/CT and localized PET/MR combined imaging for liver space occupying lesions JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 353 LP - 353 VL - 59 IS - supplement 1 AU - Yuanfan Xu AU - Shifeng Wang AU - Gang Feng AU - Dangfan Yu AU - Chengzhou Li Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/353.abstract N2 - 353Objectives: The purpose of this study is to explore the diagnostic value of combined whole body PET/CT and a hybrid TOF PET/MR localized imaging for liver space occupying lesions and the effect on the tumor staging and clinical decision of the malignant tumor patients. Methods Seventy-one patients suspected of liver space occupying lesions, were performed whole body 18F-FDG PET/CT imaging and 18F-FDG PET/MR (T1,T2 and DWI) localized examinationon the same day. The Pearson correlation analysis, Bland-Altman analysis and t test were applied to analyze the SUVmax and the T/N values in two PET imaging. By using the pathological results as the gold standard,χ2test was used to compare diagnostic efficiency between PET/MR and PET/CT.And then,the percentage of patients with changed tumor staging and clinical decision was calculated. Results In the 71 cases, of which 61 were diagnosed with malignant tumor, 41 cases were liver tumors, including 11 cases of newly diagnosed liver primary tumors, 5 cases of residual and new lesions after hepatocellular carcinoma treatment,25 cases of liver metastases, and 30 cases were diagnosed with benign liver lesions. In the combined imaging, there are 42 cases of lesions have obvious FDG uptake. More lesions with 18F-FDG uptake higher than normal liver background were found in PET/MR comparing with PET/CT in 25 cases.And there was significant statistical difference (χ2= 18.12, p< 0.01) between detection efficiency of PET in PET/MR and PET in PET/CT. The SUVmax (PET/MR) was linearly correlated with SUVmax(PET/CT) (r = 0.901), which was also observed between T/B (PET/MR) and T/B (PET/CT) (r = 0.894). SUVmax (PET/MR). The average of SUVmax (PET/MR) was greater than that in PET/CT and statistic difference (t=2.11, P< 0.05) was found with a mean difference 1.7 (95%CI: -2.3~5.6). The average of T/B (PET/MR) was also greater than that in PET/CT with significant difference (t=3.45, p< 0.01) and mean difference 1.6 (95%CI: -1.4~4.7). The sensitivity, specificity and accuracy rates of PET/CT were 70.7%, 96.7% and 81.7% respectively and they were 100%,93.3%,97.2% for PET/MR. The sensitivity and specificityof the two imaging were statistically different (χ2=14.06,9.02,p< 0.01).Compared with PET/CT examination only, additional lesions were detected in 60.6% (43/71) patients by PET/MR, and the percentage of patients with changed tumor staging is 26.2%(16/61) by the combined imaging,therefore the treatment plans were changed for 31.1%(19/61)patients. Conclusions With optimal system sensitivity and TOF information, there is better signal-to-noise (SNR) for PET images in hybrid PET/MR than in PET/CT and therefore lesions can be shown clearly. Good soft tissue contrast can be observed in MR images which is greatly helpful for the detection and diagnosis. Comparing with PET/CT examination only,improved liver lesion diagnosis and tumor staging was done according to combining with PET/MR imaging and can help to make more reasonable clinical therapy plan. Key Words:liver space occupying lesion;PET/MR;PET/CT; diagnostic value; clinical significance Figure 1. A)PET/CT images;B)PET/MR images.Images from left to right are: Maximum Intensity Projection (MIP) of PET in PET/CT, horizontal axial CT image, horizontal axial PET/CT fusion image, MIP of PET in PET/MR, horizontal axial MR T2image, horizontal axial PET/MR(T2) fusion image.1)Cases 1: 42-year-old man, hepatocellular carcinoma.2) Cases 2: 59-year-old man,hepatocellular carcinoma recurrence.3) Cases 3: 70-year-old man,lung cancer, liver metastases.4) Cases 4: 65-year-old man, esophageal cancer, liver metastases. ER -