Abstract
INTRODUCTION: Paradoxical co-existence of insulinoma and diabetes is extremely rare. Although a few case reports addressed this association, a comprehensive study elucidating this relationship has been lacking. We performed a systematic review of published cases of insulinoma in diabetes. METHODS: We conducted a literature search using PubMed and Google Scholar, employing various combinations of the following terms: 'insulinoma', 'diabetes', 'nesidioblastosis', 'endogenous hyperinsulinism', 'hypoglycaemia', and 'hyperglycaemia' (from January 1900 to January 30, 2024). Exclusion criteria included non-English publications, duplicate articles, reports lacking sufficient data, cases of endogenous hyperinsulinemic hypoglycaemia other than insulinoma, and inaccessible articles. Statistical analysis was performed using appropriate methods. RESULTS: Sixty patients were considered for the final analysis. Mean age was 61 ± 15 years (range: 17-96 years) with a slight female preponderance; 88.3% had type-2 diabetes with a median duration of 8 years. The median delay in diagnosis of insulinoma was 6 months. Median blood glucose varied from 30.5 mg/dL to 235 mg/dL, with a mean HbA1c of 5.6 ± 1.3% (range: 2.9%-8.2%). Critical sampling data were available in 75% of cases. The median size of the insulinoma was 2 cm. Furthermore, 5.2% of insulinomas were extra-pancreatic. Among pancreatic insulinomas, 14.5% were multi-focal. One-third of cases were malignant. Surgical resection was done in 70.9% of cases, while 40% received drug therapy and 12.7% received both, with 20.7% overall mortality. Malignant insulinoma (P = 0.007), micro-angiopathic (P = 0.018) and macro-angiopathic complications (P = 0.039), and other co-morbidities (P = 0.009) were associated with unfavourable prognosis, while being overweight and obese (P = 0.020) at presentation was associated with favourable prognosis. CONCLUSION: This first systematic review provides insights into the uniqueness of insulinoma in diabetes.