Surgical removal is the treatment of choice for insulinomas. Definitive biochemical diagnosis of organic hyperinsulinism has to be established before surgery. These tumors are sometimes undetected by preoperative imaging investigations and, in addition, surgical management may also be complicated by the absence of palpable tumors or the presence of multiple tumors. We report the value of the euglycemic clamp technique for diagnosis and surgical treatment in 21 patients with confirmed insulinomas. Data were compared with 12 controls, and nine patients were retested after surgery. During the euglycemic hyperinsulinic clamp, the mean C-peptide value was 3.6+/-2.2 ng/ml and it remained high (3.8+/-2.5 ng/ml), despite exogenous hyperinsulinemia (1762.7+/-233.2 microU/ml for the highest plateau). In contrast, the C-peptide concentration declined in 12 control patients (0.3+/-0.1 ng/ml, P < 0.001) and after successful surgery in nine retested patients (0.3+/-0.2 ng/ml, P < 0.01). During continuous glucose monitoring, successful removal of the insulin-secreting tumor was accompanied by an increase in plasma glucose concentrations and a loss of requirement for endogenous glucose within 36 min (range 28-43 min). The continuing requirement for glucose after the ablation of the tumor revealed the existence of additional and initially undetected tumors in four patients, among whom two had the multiple endocrine neoplasia type I (MEN I) syndrome. We conclude that the euglycemic hyperinsulinic clamp is a reliable and convenient diagnostic test for insulinoma, as it is both safe (no hypoglycemia) and relatively brief (3 x 90 min). Glucose monitoring and glucose clamping provide a reliable indicator of complete removal of insulin-hypersecreting tissue, especially in patients with occult or multiple tumors.