Koji Tokunaga1, Shigeki Arizono2, Hironori Shimizu3, Koji Fujimoto4, Mariyo Kurata5, Sachiko Minamiguchi6, Hiroyoshi Isoda7, Kaori Togashi8. 1. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: tokunaga.koji.28m@kyoto-u.jp. 2. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: arizono@kuhp.kyoto-u.ac.jp. 3. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: hshimizu@kuhp.kyoto-u.ac.jp. 4. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: kfb@kuhp.kyoto-u.ac.jp. 5. Department of Diagnostic Pathology, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: mariyo@kuhp.kyoto-u.ac.jp. 6. Department of Diagnostic Pathology, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: minami@kuhp.kyoto-u.ac.jp. 7. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: sayuki@kuhp.kyoto-u.ac.jp. 8. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. Electronic address: ktogashi@kuhp.kyoto-u.ac.jp.
Abstract
PURPOSE: To determine the optimal b-value for accurate depiction of pancreatic cancer (PC) in patients with active tumor-associated pancreatitis (TAP), using computed diffusion-weighted imaging (cDWI) with a range of b-values up to 3000 s/mm2. METHODS: The study protocol was approved by the institutional review board. We retrospectively analyzed 34 consecutive PC cases with active TAP who underwent pancreatectomy without preoperative therapy. Four cDWI datasets with b-values of 1500-3000 s/mm2 (cDWI1500-cDWI3000) were generated from the original DWI datasets with b-values of 0 and 1000 s/mm2 obtained using a 3-T scanner. Two board-certified radiologists evaluated images qualitatively (tumor conspicuity and total image quality), and another two board-certified radiologists placed regions of interest for quantitative evaluations (apparent diffusion coefficient [ADC] values of both lesions, contrast ratio [CR] of PC to active TAP, and volume ratio [VR] of PC to surgical specimen). RESULTS: As the b-value increased, tumor conspicuity improved significantly in cDWI2000 and cDWI2500 (P = 0.0121 and 0.0015, respectively), although total image quality decreased in all cDWIs compared with DWI1000 (P < 0.0001). Significantly lower ADC values were seen in PC (P < 0.0001). All cDWI groups showed positive correlation between the tumor conspicuity and ADC difference between PC and TAP. CR increased with the b-value, while VR decreased. Significant equivalence of VR to the surgical specimen was seen on cDWI2000 (P = 0.0031). CONCLUSION: Accurate depiction of PC was optimal with cDWI2000 in the presence of active TAP.
PURPOSE: To determine the optimal b-value for accurate depiction of pancreatic cancer (PC) in patients with active tumor-associated pancreatitis (TAP), using computed diffusion-weighted imaging (cDWI) with a range of b-values up to 3000 s/mm2. METHODS: The study protocol was approved by the institutional review board. We retrospectively analyzed 34 consecutive PC cases with active TAP who underwent pancreatectomy without preoperative therapy. Four cDWI datasets with b-values of 1500-3000 s/mm2 (cDWI1500-cDWI3000) were generated from the original DWI datasets with b-values of 0 and 1000 s/mm2 obtained using a 3-T scanner. Two board-certified radiologists evaluated images qualitatively (tumor conspicuity and total image quality), and another two board-certified radiologists placed regions of interest for quantitative evaluations (apparent diffusion coefficient [ADC] values of both lesions, contrast ratio [CR] of PC to active TAP, and volume ratio [VR] of PC to surgical specimen). RESULTS: As the b-value increased, tumor conspicuity improved significantly in cDWI2000 and cDWI2500 (P = 0.0121 and 0.0015, respectively), although total image quality decreased in all cDWIs compared with DWI1000 (P < 0.0001). Significantly lower ADC values were seen in PC (P < 0.0001). All cDWI groups showed positive correlation between the tumor conspicuity and ADC difference between PC and TAP. CR increased with the b-value, while VR decreased. Significant equivalence of VR to the surgical specimen was seen on cDWI2000 (P = 0.0031). CONCLUSION: Accurate depiction of PC was optimal with cDWI2000 in the presence of active TAP.
Authors: Felix N Harder; Eva Jung; Sean McTavish; Anh Tu Van; Kilian Weiss; Sebastian Ziegelmayer; Joshua Gawlitza; Philip Gouder; Omar Kamal; Marcus R Makowski; Fabian K Lohöfer; Dimitrios C Karampinos; Rickmer F Braren Journal: Cancers (Basel) Date: 2022-01-18 Impact factor: 6.639