OBJECTIVE To test the hypothesis that, in contrast to administration of glucose or glucagon, administration of the amino acid Ala or of the beta 2-adrenergic agonist terbutaline produces sustained glucose recovery from hypoglycemia. METHODS We developed a model of clinical hypoglycemia using subcutaneous injection of insulin (0.15 U/kg) in patients with IDDM. In comparison with nondiabetic subjects, patients with IDDM exhibited reduced glucagon (P = 0.0001), epinephrine (P = 0.0060), and pancreatic polypeptide (P = 0.0001) responses to hypoglycemia. In addition to placebos, the following were administered during hypoglycemia (2 h after insulin injection) in IDDM patients: oral glucose, 10 and 20 g; subcutaneous glucagon, 1.0 mg; oral Ala, 40 g; oral terbutaline, 5.0 mg; and subcutaneous terbutaline, 0.25 mg. RESULTS Glucose (10 and 20 g) and glucagon raised plasma glucose (P = 0.0163, 0.0060, and 0.0001, respectively) from 3.0-3.3 mM to peaks of 5.4 +/- 0.4, 6.8 +/- 0.7, and 11.8 +/- 0.8 mM within 30, 45, and 60 min, respectively, but the responses were transient. Oral Ala raised glucose levels (P = 0.0401) to 4.0 +/- 0.4 mM within 30 min; glucose levels then rose gradually to a 6-h value of only 7.1 +/- 0.9 mM. Oral terbutaline raised glucose levels (P = 0.0294) to 4.3 +/- 0.3 mM within 30 min; glucose levels then rose substantially. In contrast, subcutaneous terbutaline raised glucose levels (P = 0.0249) to 3.7 +/- 0.1 mM within 15 min; the levels plateaued at 5.0 mM from approximately 60-150 min and then paralleled the placebo curve. CONCLUSIONS Ala and terbutaline produce sustained glucose recovery from hypoglycemia in IDDM and are therefore potentially useful agents for the treatment of mild or moderate iatrogenic hypoglycemia, or the prevention of iatrogenic hypoglycemia, when food intake is not anticipated over the following several hours.