A retrospective analysis of the results of all intra-arterial calcium stimulations performed at St. Vincents Hospital, Dublin, in the years 2001-2006. All patients with symptoms suggestive of hypoglycemia had 72 hour fasting test with evidence of inappropriately elevated insulin and c-peptide at the time of hypoglycemia. These patients were investigated further with pancreatic imaging and selective intra-arterial calcium stimulation with hepatic venous sampling (ASVS). Analysis of the results was performed using the Wilcoxon signed rank test. Results were available in 9 patients. The overall catherisation success rate was: minimum four arteries in 7/9, three arteries 1/9 and two arteries in 1/9. CT was positive in 2/7 patients, MRI 0/2, octreotide scan 0/2 and endoscopic ultrasound 0/2. Mean insulin increment was 11.91 fold (95% CI 6.5117.30) in tumour area versus 1.61 fold (95% CI 1.212.01) P=0.002. ASVS was positive in 8 patients. 7 patients were found to have insulinoma and 2 patients were diagnosed with adult nesidioblastosis by means of histological diagnosis. One of nesidioblastosis patient had negative calcium stimulation test but had diffuse hyperinsulinemic picture on ASVS. Our results suggest that selective intra-arterial calcium stimulation with hepatic venous sampling remains a powerful tool for diagnosis of insulinoma. CT pancreas alone combined with ASVS should be the standard of investigation in biochemically proven insulinoma. Three fold insulin levels increment should be used as the cut-off point for positive test after calcium stimulation. We reported a case of failure ASVS. ASVS use should be restricted to units with expertise in this area.