Back to the Top
Consider the plasma concentration ( ng/ml ) of a drug which lowers
plasma glucose ( mg/dl ) at a dose of 50 ug/Kg given as a bolus i.v.
Time(h) Conc ( ng/ml)
0.08 197.1 95
I first fitted the data using WinNomlin to a two comp model ( model
8) and I had the following parameters: A, 236.57; B, 50.93; a, 6.98;
Then I used these values to make a pharmacodynamic fit.
Case # 1 PK/PD link and I had the follwing:
AIC - 54.60
Case # 2 Indirect response( stimulation of input, the mechanism of
action of the drug) and I had the following
Clearly judging by AIC and Sc criteria and by looking at the graphs
and residuals I had good fits in both models. Now my question is why
is Emax so different between the two models and so is the EC50? Do
they mean differen things in the two models and if yes what are their
units?Also what would be the units of Kin and Kout? Which PD model is
more physilogical to use.
Dr. Ibrahim wasfi
P O Box 253, Abu Dhabi
United Arab Emirates
Tel + 4092522
Fax + 4463470
Back to the Top
The following message was posted to: PharmPK
> Now my question is why
> is Emax so different between the two models and so is the EC50? Do
> they mean differen things in the two models and if yes what are their
> Case # 1 PK/PD link
> Case # 2 Indirect response( stimulation of input, the mechanism of
>action of the drug)
The two models you use have different assumptions. Would you expect a
horse that was a known winner to be comparable to a winning camel?
The EC50 is not much different but the Emax is describing very
different concepts of drug action.
> Also what would be the units of Kin and Kout? Which PD model is
> more physilogical to use.
Your "PK/PD link" model seems to be a delayed (effect compartment
model) link between conc and response. If you think that the delay in
onset of action is due to distribution to the site of action then
this model would be more physi-logical. But if you think the delay
depends on the kinetics of a mediator then the so called indirect
effect model would be more physi-logical.
The keo=0.011 h-1 means a half-life of distribution of 0.7/0.011 or about 6 h.
This value is unlikely to represent distribution (usually minutes) so
I would favour the indirect effect model as being more physi-logical.
The kout=0.04 h-1 means a half-life of the mediator of about
0.7/0.04 or about 2 h.
The kin physi-logical meaning of 3.83 mg/dL/h reflects the input rate
of glucose (in slightly weird units of conc/time rather than
mass/time). The ratio of kin/kout should be a reasonable
approximation of the pre-drug glucose conc ie. 3.83 h-1*mg/dL/0.04
h-1 or about 96 mg/dL.
> Dr. Ibrahim wasfi
> Camelracing Laboratory
> P O Box 253, Abu Dhabi
> United Arab Emirates
Nick Holford, Divn Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand
email:n.holford.aaa.auckland.ac.nz tel:+64(9)373-7599x6730 fax:373-7556
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (firstname.lastname@example.org)