New Diagnostic Criteria for the Localization of Insulinomas with the Selective Arterial Calcium Injection Test: Decision Tree Analysis

Kenji Kajiwara,Takuji Yamagami,Naoyuki Toyota,Hideaki Kakizawa,Masaki Urashima,Masashi Hieda,Yasutaka Baba,Tomoyuki Akita,Junko Tanaka,Kazuo Awai
DOI: https://doi.org/10.1016/j.jvir.2018.05.015
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.
What problem does this paper attempt to address?