In a randomized double-blind crossover study in 42 patients with essential hypertension, the metabolic effects of a beta-adrenergic blocker with a pronounced beta 2-agonistic effect, dilevalol 400 mg x 1, were compared with those of metoprolol succinate 200 mg x 1. The effects of glucose metabolism were evaluated by the hyperinsulinemic euglycemic clamp technique and an intravenous glucose tolerance test (IVGTT). Insulin-mediated glucose disposal (M) and the insulin sensitivity index (M/I) increased by 19% (P = .011) and 10% (P = .27), respectively, during dilevalol treatment, but decreased by 10% (P = .15) and 22% (P = .0025), respectively, during metoprolol treatment, giving rise to significant differences between the two treatment regimens. Compared with dilevalol-treated patients, those treated with metoprolol showed increased plasma insulin values at the end of the IVGTT, but there was no difference in plasma glucose concentrations or glucose tolerance. Hemoglobin A1c (HbA1c) levels increased by 5.4% (P = .04) in the metoprolol group. Serum cholesterol, total serum and very-low-density lipoprotein (VLDL) triglycerides, and serum urate levels decreased significantly (by 6%, 22%, 29%, and 14%, respectively) during dilevalol treatment, and there was a significant difference between the effects of the two drugs on total, VLDL, and low-density lipoprotein (LDL) triglyceride and serum urate levels. These data demonstrate that a beta-blocker with a partial beta 2-agonist action differs in its metabolic effect profile from a selective beta 1-blocker and may offer advantages by improving glucose and lipid metabolism.