Kenji Kajiwara1, Takuji Yamagami2, Naoyuki Toyota3, Hideaki Kakizawa4, Masaki Urashima5, Masashi Hieda5, Yasutaka Baba6, Tomoyuki Akita7, Junko Tanaka7, Kazuo Awai6. 1. Department of Radiology, Medical School, Kochi University, 185-1 Kohasu, Oko-cho, Nankoku 783-8505, Japan; Department of Diagnostic Radiology, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan. Electronic address: kaji07kenji@yahoo.co.jp. 2. Department of Radiology, Medical School, Kochi University, 185-1 Kohasu, Oko-cho, Nankoku 783-8505, Japan. 3. Department of Diagnostic Radiology, National Hospital Organization Kure Medical Center and Chugoku Cancer Center, Kure, Japan. 4. Department of Radiology, Hiroshima Red Cross Hospital and Atomic-bomb Survivors Hospital, Hiroshima, Japan. 5. Department of Diagnostic Radiology, Hiroshima City Hospital, Hiroshima, Japan. 6. Department of Diagnostic Radiology, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan. 7. Department of Epidemiology, Infectious Disease Control and Prevention, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan.
Abstract
PURPOSE: To explore optimal diagnostic criteria for localizing insulinomas with the selective arterial calcium injection (SACI) test using decision tree analysis. MATERIALS AND METHODS: A retrospective study included 86 vessels of 18 patients (5 men, 13 women; mean age 67 y; range, 49-73 y) with insulinomas who underwent SACI test between June 2007 and May 2016. Of 27 insulinomas, 7 were found in the head, 13 in the body, and 7 in the tail of the pancreas. Two patients had multiple tumors. To identify optimal diagnostic criteria, decision tree analysis was performed, and sensitivity, specificity, and accuracy of the conventional and the proposed new diagnostic criteria (plasma insulin concentration after calcium injection [ICpost] > 2.0 × plasma insulin concentration before calcium injection [ICpre]) were compared. RESULTS: The proposed new diagnostic criteria for insulinoma obtained by decision tree analysis were (i) ICpost > 2.7 × ICpre and maximum insulin concentration > 60.3 μIU/mL or (ii) ICpost > 2.7 × ICpre and maximum insulin concentration < 60.3 μIU/mL with ICpre being ≥ 7.5 μIU/mL. Sensitivity, specificity, and accuracy of the new criteria for the SACI test were 100%, 91.4%, and 94.2; sensitivity, specificity, and accuracy of conventional criteria were 100%, 69.0%, and 79.1%. CONCLUSIONS: New diagnostic criteria for localization of insulinomas with the SACI test yielded higher diagnostic performance than conventional criteria.
PURPOSE: To explore optimal diagnostic criteria for localizing insulinomas with the selective arterial calcium injection (SACI) test using decision tree analysis. MATERIALS AND METHODS: A retrospective study included 86 vessels of 18 patients (5 men, 13 women; mean age 67 y; range, 49-73 y) with insulinomas who underwent SACI test between June 2007 and May 2016. Of 27 insulinomas, 7 were found in the head, 13 in the body, and 7 in the tail of the pancreas. Two patients had multiple tumors. To identify optimal diagnostic criteria, decision tree analysis was performed, and sensitivity, specificity, and accuracy of the conventional and the proposed new diagnostic criteria (plasma insulin concentration after calcium injection [ICpost] > 2.0 × plasma insulin concentration before calcium injection [ICpre]) were compared. RESULTS: The proposed new diagnostic criteria for insulinoma obtained by decision tree analysis were (i) ICpost > 2.7 × ICpre and maximum insulin concentration > 60.3 μIU/mL or (ii) ICpost > 2.7 × ICpre and maximum insulin concentration < 60.3 μIU/mL with ICpre being ≥ 7.5 μIU/mL. Sensitivity, specificity, and accuracy of the new criteria for the SACI test were 100%, 91.4%, and 94.2; sensitivity, specificity, and accuracy of conventional criteria were 100%, 69.0%, and 79.1%. CONCLUSIONS: New diagnostic criteria for localization of insulinomas with the SACI test yielded higher diagnostic performance than conventional criteria.