The train-of-four (TOF) stimulation pattern consists of 4 stimuli (T1, T2, T3, and T4) at 2 Hz, and is used in daily anesthesiological practice to determine the degree of relaxation caused by muscle relaxants. At a surgical levels of relaxation the degree of relaxation can be estimated by counting the number of "measurable" or "visible" muscular reactions to the 4 stimuli in the TOF stimulation pattern (TOF count). During recovery relaxation can be estimated by calculating the TOF ratio (T4/T1). Bartkowski and Epstein described a pharmacokinetic-pharmacodynamic (PK-PD) model to predict TOF ratio by modifying and extending the PK-PD model as described by Sheiner to use a hypothetical distributed effect compartment described by a median equilibration rate constant and a dispersion parameter. We extended the Bartkowski and Epstein PK-PD model to simulate all four TOF twitches by including EC50 terms for T2 and T3. We fit this model to data from the pig and compared the results to fitted models using separate PD models for each TOF twitch (extended Sheiner model). The extended Bartkowski and Epstein model fit the twitch height data from all four TOF twitches better than the extended Sheiner model and has fewer parameters.