These studies were performed to evaluate the postprandial blood glucose responses to a variety of differently formulated enteral feeding products in patients with type I diabetes. Eleven subjects with type I diabetes were evaluated in three studies, all using a Biostator (artificial endocrine pancreas) that delivered a small, basal amount of insulin and measured blood glucose levels. Subjects consumed 20 mL of the assigned formula every 15 minutes for the 240 minutes of the study. Study 1 evaluated the response to each of five products: Glucerna, Enrich, Ensure HN, Pulmocare, and Compleat Modified. When the postprandial blood glucose response to Glucerna was greater than when its research formulation (EN-8715) had been tested in 1988, studies 2 and 3 were undertaken to assess why this discrepancy occurred. Study 2 compared stored EN-8715 to Glucerna and study 3 compared frozen and thawed vs nonfrozen EN-8715, because of a concern that the original product had been frozen during shipping. In study 1 the glucose response (assessed as area under the glucose curve) correlated with the grams of carbohydrate present in the enteral feeding formula (r = .58, p = .002). The presence or absence of fiber, in the form of soy polysaccharide, did not affect the glucose response. Glucerna produced a significantly lower blood glucose response than did Enrich, Ensure HN, or Compleat Modified, although this response was greater than the response to EN-8715 in 1988. However, in study 2 no differences were found between stored EN-8715 and Glucerna and in study 3, freezing and thawing was not found to significantly alter the glucose response.(ABSTRACT TRUNCATED AT 250 WORDS)