The efficacy of oxime treatment in soman poisoning is limited due to rapid aging of inhibited acetylcholinesterase (AChE). Pre-treatment with carbamates was shown to improve antidotal treatment substantially. Recently, by using a dynamically working in vitro model with real-time determination of membrane-bound AChE activity, we were able to demonstrate that pre-inhibition of human erythrocyte AChE with pyridostigmine or physostigmine resulted in a markedly higher residual AChE activity after inhibition by soman or paraoxon than in the absence of reversible inhibitors. The purpose of the present study was to compare the effect of carbamate pre-treatment and soman challenge with human erythrocyte and muscle homogenate AChE. Both enzyme sources were immobilized on particle filters which were perfused with acetylthiocholine, Ellman's reagent and phosphate buffer. AChE activity was continuously analyzed in a flow-through detector. Pre-inhibition of AChE with pyridostigmine or physostigmine resulted in a concentration-dependent increase in carbamylation, residual activity after soman inhibition and fraction of decarbamylation AChE after discontinuation of the inhibitors without differences between human erythrocyte and muscle AChE. This data support the view that human erythrocyte AChE is an adequate surrogate marker for synaptic AChE in OP poisoning.