Relating glucose clamp profiles to reduction of blood glucose after insulin administration.

A model was established allowing prediction of blood glucose response from glucose clamp results performed in healthy volunteers. Data from published studies performed in healthy volunteers were used to establish, test, and validate a model for the evaluation of glucose reductions from glucose clamp results. Studies included those that measured blood glucose and glucodynamic response over time after administration of 0.05 U/kg of regular human insulin (HR) and insulin lispro (LP) with and without the benefit of a glucose clamp procedure. An inhibitory effect E(max) model was used to describe the relationship; the model differed between the HR and LP responses by the intensity of the counterregulatory response as assessed by glucagon measurements. The relationships were used to predict blood glucose responses from a clamp study assessing NPH insulin and HR administrations. Glucose concentrations measured after administration of NPH insulin and HR without a clamp were compared to the model-predicted results to assess the accuracy of the model predictions. The E(max) model successfully correlated the glucose clamp results with the blood glucose depressions in the presence and absence of a counterregulatory response. However, predictions of glucose depression were only accurately modeled in the absence of a counterregulatory glucagon response. The correlations established with a minimal counterregulatory response underscore the value of glucose clamp procedures in defining the time-activity profiles of insulins when the clamp is established at fasting glucose concentrations.