## Predicting Grfghrp from GH Output

Given the knowledge of the state of desensitization of the somatotrophs (indicated by f0) at the beginning of the experiment, and the somatostatin levels (assumed constant for present purposes at s0) throughout the experiment, Eq. (10) can in principle be used to determine the GRF/secretagogue stimulation, which would produce any given pattern of GH output as follows (assuming only one stimulating compound is present). GH released in interval (t, t + At) is approximately given by t+At t+At

Ht,t + A t) = J tf r,s,f)dt = J [k4 + k 5 (1 - 4 [S0])](r + r0)f* +(f - f*) e Kt]dt t t

= [k4 + k5 (1 - 4 [s0])](r + r0)(/b - f*) e-Kt (1 - ) / K

If At is sufficiently small for r to be assumed approximately constant during the interval, and if the other parameters are known for the particular application, then this equation can be solved for r, and Eq. (7) used to obtain the change in f. The process can be repeated (making appropriate adjustments to f0 each time) for each subsequent At. In this way the temporal profile of GRF can be built up sequentially. This simple estimation method is dependent on the parameters determining the somatotroph resensitization rate being known quite precisely. Otherwise, errors would accumulate because of f0 being updated incorrectly at each stage.