Bioequivalence and Bioavailability Forum

Dear All!

I want to simulate PK profiles from the publication by Björkman et al (2012) in which a population PK of recombinant factor VIII including body weight and age as covariate is published. It is a two-compartmental model with an additive error, random effects were only assumed for the first compartment:

CL_i = f(BW_i, Age_i, fixed effect_CL,...)*exp(random effect_i,CL)

V1_i = f(BW_i, fixed effect_V1,...)*exp(random effect_i,V1)

What I understand:
Random effects are two-dimensional normal distributed with mean zero (covariance matrix can be found in the paper). Setting the random effects to zero leads to the population parameters as exp(0)=1. It's a multiplicative effect.

What I do not understand:
How do I simulate profiles from this model taking random effects into accout? Exponentiating random draws from the two-dimensional normal distribution and multiplicating it with the population parameters CL and V1 leads to implausible values.

I would be very happy to read your suggestions!

Kind regards and thanks in advance,
Alex

PS:
Björkman S, Oh M, Spotts G, Schroth P, Fritsch S, Ewenstein BM, Casey K, Fischer K, Blanchette VS, Collins PW. Population pharmacokinetics of recombinant factor VIII: the relationships of pharmacokinetics to age and body weight. Blood. 2012 Jan 12;119(2):612-8.