Change in Baseline Response in a PK/PD model

Dear Nmusers, I am currently working on a PK/PD model for the anesthetic effect of nerve blockade and I am now facing issues regarding the PD aspect of my data. I am using a surrogate PD endpoint (current perception threshold; CPT) to quantitatively assess the attenuation of the signal propagation after the injection of the drug. Asssuming complete anesthesia is possible, the attenuation (A) is modeled by an inibitory sigmoid Emax model given by : A = baseline current * [1 + (Ce(t)/EC50)^gama] where EC50 is the effect site concentration causing 50% attenuation (or 100% increase in stimulus intensity for response) Ref: Sarton et al. Anesthesiology 2000:93:1245-54. In this model, the baseline response is very important for the calculation of EC50. We observed that PD baseline is significantly different after surgery compared to time 0 (before the administration of the local anesthetic). I have tried several approaches to handling PD baseline and B1 (typical value and its eta are estimated together with other parameter in the model) (ref: Journ of PKPD 2008 35:269-283) is by far, the best one. However, it assumes that the baseline is constant for a given individual, which is clearly not the case in this setting. Is there any way to take into account changes in PD baseline that occur after surgery ? Any thoughts are welcome... By the way, IOV (interoccasion variability) did not improve the model fit. -- François Gaudreault, Ph.D. Candidate Pharmacométrie / Pharmacometrics Chargé de cours / Lecturer Faculté de pharmacie / Faculty of Pharmacy Université de Montréal