Utility of Non-invasive Method Using Acoustic Radiation Force Impulse ARFI Imaging for Measurement of Liver Fibrosis Gaku Igarashi, Fumio Tsujimoto, Nobuyuki Matsumoto, Masasumi Miyazaki, Atsuki Koike, Takanori Okamura, Masaru Sakurai, Masaru Okamoto, Yuuki Ikeda, Hideaki Takahashi, Koutarou Matsunaga, Yoshiki Katakura, Chiaki Okuse, Satoshi Koizumi, Takeshi Asakura, Hiroshi Nakano, Masayuki Takagi, Yasuo Nakajima, Sachihiko Nobuoka, Takehito Otsubo, Michihiro Suzuki, and Fumio Ito The accurate diagnosis of fibrosis related to chronic liver diseases is crucial for prognostication and treatment decisions. Although liver biopsy is the gold standard, it has limitations because of its invasive nature. Thus development of new non-invasive methods is desired to diagnose hepatic fibrosis. Recently, acoustic radiation force impulse ARFI imaging has been developed as a non-invasive modality to evaluate sti#ness of tissues. ARFI imaging theoretically measures liver sti#ness of all the segments independently. However, there are no reports for the attempt to diagnose liver fibrosis with ARFI imaging. The aim of this study is to assess the applicability of ARFI imaging for the diagnosis of liver fibrosis. The study enrolled 20 healthy volunteers as a control group 42 patients diagnosed as chronic liver disease. From these patients, we classified 11 patients with F1, 17 with F2, 5 with F3, 9 with F4 accouding to pathological fibrosis staging by the new Inuyama classification. ARFI Vs values were acquired with a Siemens Acuson S2000 Siemens Medical Systems, Germany , which has been modified to obtain ARFI images. We obtained ARFI Vs values from S2 to S8 of Cuinaud[s segmentation system by 3 consecutive measurements. ARFI Vs values of S1 was excluded from the study, because of the location with di$culty for the measurement. We calculated interquartile range median IQR M and standard deviation SD of these 3 data from a total of 434 segments of 62 patients. The dispersion of IQR M 1.1 was smaller than that of SD 2.4 , suggesting the existence of outlier in ARFI Vs values. And to reject those outliers, we chose median to represent ARFI Vs values of each segments. The average of ARFI Vs value of all segments was calculated as the ARFI Vs value of whole liver. The laboratory blood tests, which showed statistically a significant correlation to fibrosis staging by Pearson[s correlation coe$cient, were PT-INR r 0.57, p 0.001 , ICGR15 r 0.497, p 0.001 , hyaluronic acid r 0.416, p 0.01 , and Platelet r 0.436, p 0.01 . ARFI Vs value showed the best correlation r 0.766, p 0.001 compared to blood results. In this study we showed that fibrosis staging is significantly a correlate with whole liver sti#ness which was obtained by ARFI, a newly developed non-invasive imaging technique, which enables us to quantitatively estimate sti#ness of liver. Division of Gastroenterology and Hepatology, Department of Internal Medicine, St. Marianna University School of Medicine. ARFI 211