Target-controlled infusion (TCI) anesthesia using target effect-site concentration rather than plasma concentration provides less drug consumption, safer anesthesia, less undesired side effects and improved animal welfare. The aim of this study was to calculate the constant that converts propofol plasma into effect-site concentration (k(e0)) in dogs, and to implement it in a TCI system and compare it with the effect on the central nervous system (CNS). All dogs were subjected to general anesthesia using propofol. Fourteen dogs were used as the pilot group to calculate k(e0), using the t(peak) method. Fourteen dogs were used as the test group to test and validate the model. RUGLOOP II software was used to drive the propofol syringe pump and to collect data from S/5 Datex monitor and cerebral state monitor. The calculated k(e0) was incorporated in an existing pharmacokinetic model (Beths Model). The relationship between propofol effect site concentrations and anesthetic planes, and propofol plasma and effect-site concentrations was compared using Pearson's correlation analysis. Average t(peak) was 3.1 min resulting in a k(e0) of 0.7230 min(-1). The test group showed a positive correlation between anesthetic planes and propofol effect-site concentration (R = 0.69; P < 0.0001). This study proposes a k(e0) for propofol with results that demonstrated a good adequacy for the pharmacokinetic model and the measured effect. The use of this k(e0) will allow an easier propofol titration according to the anesthetic depth, which may lead to a reduction in propofol consumption and less undesired side effects usually associated to high propofol concentrations in dogs.