Development of method of predicting prognosis for sepsis-associated encephalopathy using pattern recognition analysis of nuclear magnetic resonance data

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K15663
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Emergency medicine
• Research Institution: Kyoto University
• Principal Investigator: Suzuki Takao 京都大学, 医学研究科, 客員研究員 (40328810)
• Co-Investigator(Kenkyū-buntansha): 平川 慶子 日本医科大学, 医学部, 助手 (30165162) ; 小池 薫 京都大学, 医学研究科, 教授 (10267164) ; 佐藤 格夫 愛媛大学, 医学系研究科, 寄附講座教授 (30409205) ; 金涌 佳雅 日本医科大学, 医学(系)研究科(研究院), その他 (80465343)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 敗血症 / 敗血症関連脳症 / 脳脊髄液 / ＮＭＲ

Outline of Final Research Achievements

We collected cerebrospinal fluid from rat septic models, following sepsis induction by intraperitoneal administration of LPS (lipopolysaccharide). Then, the cerebrospinal fluid sample was subjected to proton (1 H) NMR measurement using a 7 tesla (300 MHz) FT-NMR apparatus (JEOL). We pattern-recognized this NMR spectrum and spectral data of a single pulse was analyzed by partial least squares regression (PLS-DA). The NS group (control group; administered physiological saline), LPS 2 group (administered 2 mg/kg LPS), and LPS 10 group (administered 10 mg/kg LPS) could not be clustered on the score plot. However, results from the analysis of the T2 spectrum data by PLS-DA suggested that there was a possibility of distinguishing between the LPS 2 and 10 groups. We confirmed that these three groups were clustered on the score plot to a certain extent.

Free Research Field

救急医学