We studied the practice of investigations for diagnosing and localising insulinoma in our hospital. We looked at the number of investigations performed before exact localisation of insulinoma.
We identified 5 cases that were confirmed on clinical, biochemical or histopathological bases and were recorded on the histopathology and IT database as insulinoma from 1986 to 2004. We looked retrospectively at the case notes of these cases.
All 5 cases had blood tests for insulin and c-peptide during a 72 h fast. Three out of 5 had abdominal ultrasound and this was normal in 2 and inconclusive in 1. All had abdominal CT scan and was normal in 3. One CT scan was normal 8 years after symptoms onset and a repeat CT, after 15 years of symptoms, localized a mass. One CT localized the lesion quite close to the exact location as found on surgery. Angiogram, portal venous sampling and gut hormone profile were performed in 2 patients. One patient had endoscopic ultrasound which localized insulinoma accurately. Two patients had pancreatic MRI which were inconclusive. Four patients had surgical exploration and all of these had intra operative ultrasound (IOUS). IOUS localized lesions only in 1 of 4 patients. Octreotide scan was done in 2 and both were normal.
These findings suggest that none of the tests performed were able to localize insulinoma accurately all the time. We do not seem to have any consensus on any protocol for specific investigations to localize these disabling tumours and we performed a number of different imaging techniques. There is a need for drawing joint guidelines by physicians and surgeons for localizing these tumours accurately.