Unbiased and flexible iterative computer program to achieve glucose clamping.

Glucose clamping has become a widely used test for assessing insulin sensitivity and beta-cell function. We present a new method of achieving a reliable and unbiased clamp using an iterative computer program. After initial parameter estimation, the program uses no fixed algorithm or physiological preconceptions but makes predictions according to previous glycemic responses to infusion rates. The program can be used for human or animal studies and for euglycemic or hyperglycemic clamps. The program has flexible data entry, graphic display, running statistics of mean infusion data, and mean glucose levels and their standard deviations. These form part of the criteria for the assessment of glucose clamping, which includes the sampling interval, the duration of the clamp, the achieved glucose concentration (together with its coefficient of variation), and the percent change of glucose infused at the end of the clamp. Data for 151 consecutive clamp studies were analyzed. Median coefficient of variation of glucose values at the end of the clamp was 2%.