RESUMO
BACKGROUND: Procedures suitable for preoperative localization of insulinomas are widely used but it is unclear whether they determine surgical success. We wanted to clarify whether preoperative localization or intraoperative ultrasound determines the success of surgically identifying and removing insulinomas. METHODS: We performed a systematic literature search (PubMed) yielding 100 publications with patient-level data on preoperative localization procedures, intraoperative ultrasound, success of surgical identification and removal or single, benign insulinomas (published until 31 December 2016). The consequences of successful preoperative localization and intraoperative ultrasound for surgical identification and removal of an insulinoma could be analyzed for 863 and 529 patients with single benign insulinomas, respectively. Sensitivities of commonly employed methods to localize biochemically proven insulinomas were calculated for single benign (1,153 patients), multiple benign, and malignant insulinomas, also in relation to diameter and intra-pancreatic location. RESULTS: The success of surgery (identification and removal of single benign insulinomas) in patients with a biochemically proven insulinoma was high (approximately 98%) even in the absence of preoperative localization. However, successful preoperative localization or intraoperative ultrasonography slightly, but significantly, increased the probability of surgically identifying/removing single benign insulinomas (by 1.4% and 2.0%, P = 0.012 and 0.0032, respectively). Diagnostic sensitivities of localization procedures varied widely, with GLP-1 receptor scintigraphy achieving the highest sensitivity despite its non-invasive nature. CONCLUSION: The probability of surgically identifying and removing a single, benign insulinoma is high even in the absence of successful preoperative localization or intraoperative ultrasound. However, both approaches slightly, but significantly, improve the outcome to near 100%.
