BACKGROUND: Well-controlled studies have demonstrated that inpatient hyperglycemia is an indicator of poor clinical outcomes, but the use of diabetes-specific enteral formulas in hospitalized patients remains a topic of great debate. METHODS: In two different protocols, postprandial glycemia and insulinemia were measured in 22 subjects with diabetes fed a diabetes-specific or standard formula (protocol 1). Continuous glucose monitoring was used to assess glucose levels in 12 enterally fed patients with diabetes receiving the standard formula followed by the diabetes-specific formula continuously for 5 days each (protocol 2). End points included postprandial glycemia and insulinemia, glycemic variability (mean amplitude of glycemic excursions [MAGE]), mean glucose, and insulin use. RESULTS: In the postprandial response protocol, the diabetes-specific formula resulted in lower positive areas under the postprandial curve (P < 0.001) and peak glucose (P < 0.001) and insulin (P = 0.017) levels. In the protocol using continuous glucose monitoring, glycemic variability (as measured by MAGE) was lower with continuous administration of the diabetes-specific than the standard formula (64.6 +/- 6.8 mg/dL vs. 110.6 +/-15.3 mg/dL, P = 0.003). Also, administration of the diabetes-specific formula resulted in lower mean glucose concentrations during feeding (171.1 +/- 16.1 vs. 202.1 +/- 17.4 mg/dL, P = 0.024) and insulin requirements (7.8 +/- 2.3 vs. 10.9 +/- 3.3 units/day, P = 0.039) than the standard formula. CONCLUSIONS: Relative to the standard formula, the diabetes-specific formula reduced postprandial glycemia, mean glucose, glycemic variability, and short-acting insulin requirements. These results suggest potential clinical usefulness of a diabetes-specific enteral formula for minimizing glycemic excursions in hospitalized patients.