Hello David and all,Back to the Top
I am practically a novice to pharmacokinetics. Now suddenly I am
required to apply pharmacokinetic knowledge to a real life situation
and I am not sure whether i am getting it right. I have the
concentration-time profile of a drug after a 50 mg dose. 50 mg of the
drug is taken every 8 hrs. I have the half life reported and the AUC.I
need to find out the absorption rate constant and the volume of
distribution. For the Volume of distribution as I understand the
bioavalability factor( fraction of drug absorbed) is necessary but I
dont have it so I assumed it is 1. For the absoption rate constant I
tried Wagner nelson method and method of residuals but I am not getting
the correct fit lines. I suspect that the drug may not be behaving in a
one compartment model or it may have slow distribution rate. I
reproduce the concentration -time data. Please try to work out the Ka
and the Vd.
The data is as follows
Dose of drug: 50 mg
Dosing interval: 8 hrs
Half life reported: 6 hrs
AUC after 50 mg dose: 0.75 mcg.hr/ml
Conc-time data
Time (hr) Conc(mcg.ml)
0.25 0.048
0.5 0.269
1 0.216
20.111
4 0.044
6 0.025
8 0.017
12 0.011
24 Not detected
PLEASE HELP ME
Guru
Dear Dr Guru,Back to the Top
I hope that the data are after first test oral dose. I have used
2-compartment model with lag-time and exponential
error model (not multiple oral dose).
Objective function=-103.386 NONMEM (ADVAN8)
KA=9.61 (+/-0.938) 1/hr
ALPHA=0.761 (+/-0.035) 1/hr
BETA=0.106 (+-0.008) 1/hr
K10=0.448 (+-0.013) 1/hr
V1=151 (+-3) L
Tlag=0.234 (+- 0.012) hr
K12/K21=0.239/0.181
MODEL PREDICTION CP(MCG/ML)
fitted 0.048, 0.269, 0.212, 0.114, 0.044, 0.025, 0.018, 0.011
measured 0.048, 0.269, 0.216, 0.111, 0.044, 0.025, 0.017, 0.011
sincerely
Kazimierz H. Kozlowski
The Childrens Memorial Health Institute
Warsaw, Poland
E-mail: khkoz.-at-.czd.waw.pl
Actually it is not a one comp model for the decreasing phase, but twoBack to the Top
comp. Moreover it is surprising to observe a conc of 0.011 at 12h when
the drug is delivered every 8 hours, and taking into account that I
found a lag-time of about 0.25h.
Please find in attachment a fit which looks not so bad, except for the
conc-time point of 12.
Hope this can help
Dr Brigitte Tranchand
Lab de Bact\0xC8riologie
Fac M\0xC8decine Lyon-Sud
BP12
69921 OULLINS Cedex
France
Tel 33 4 78 86 31 53
Fax 33 4 78 86 31 49
e-mail : Brigitte.Tranchand.-at-.adm.univ-lyon1.fr
[Sorry but attachments are not allowed in PharmPK messages - db]
Dear Guru,Back to the Top
From your data I calculated parameters almost similar to that reported
by
Kazimierz Kozlowski. The only relevant difference is in the standard
errors;
my own program (MultiFit) gives about two times higher values. This is
probably a result of the fact that my program uses a 'conservative'
estimate
for the residual error, taking into account the low degrees of freedom
of 2,
i.e. 8 measurements minus 6 parameters. I do not know how this is
handled in
NONMEM. I hope that others may comment upon this difference.
A further comment refers to the interpretation of the three rate
constants ka, alpha, and beta. Actually, the analyis results in two solutions with identical predicted plasma concentration profiles, but with different
parameter sets. In this case, the values of ka and alpha could be
interchanged (with corresponding changes in Vd, k10, k12, and k21; in
contrast, clearance is not affected). Without further knowledge it is
not possible to make a selection between both solutions. By convention, most
programs provide values where ka, alpha, and beta are have a decreasing
magnitude, but there is no obvious reason why this would be always the
case. The argument that absorption is usually fast compared to elimination is
reasonably, but this does not apply to distribution kinetics, so ka is
not necessarily larger than alpha.
In one-compartment kinetics this phenomenon is usually called
'flip-flop', and often regarded as a 'special case'. Please note, however, that
there are always 2 solutions for any concentration profile after extravascular
dosing, analysed with an exponential model with first-order absorption.
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.-a-.farm.rug.nl
At 06:20 AM 6/3/2004, Guru Murthy wrote:Back to the Top
"Please try to work out the Ka and the Vd."
Guru,
You have received several responses to your request. I hope they have
been helpful.
However, I cannot sit by and ignore that an underlying assumption is
being made that is incorrect.
Ka is a coefficient, not a constant. It changes with time, because it
changes in any one location with concentration difference across the
apical membrane and with location in the gut. Constant Ka is a
simplifying assumption that can be useful when a drug is absorbed
rapidly in the upper small intestine. It can also be misleading when a
compound is absorbed over a longer period of time and in the colon.
Our experience is that numerous compounds exhibit strong regional
dependence of permeability, in addition to the natural change in
absorption rate produced by changes in the concentration difference
across the apical membrane of the enterocytes. In GastroPlus, we
provide a calculation of the instantaneous Ka value that would be
needed to use the traditional relationship of absorption rate being
calculated as Ka * Mass in solution. The instantaneous "net Ka" can be
back-calculated from the more correct Fick's First Law absorption model
used in each of the 8 intestinal compartments, allowing for the
changing concentration in the enterocytes as well as the changing
concentration in the various gut compartments as transit, dissolution
and permeability combine to produce such changes.
The result is a net Ka that varies from zero at time zero (no
concentration gradient until the drug is introduced into the first
absorbing compartment), with a rapidly rising value to a peak typically
at 0.5-2 hours, then decreasing over the next 2-10 hours (depending on
the drug).
If all you want to predict is overall Fa, an average value can be
used. But it's like taking AUC/t to get an average concentration - it
can be very misleading when it comes to predicting the actual time
course of absorption, and hence plasma concentration-time.
Walt Woltosz
Chairman & CEO
Simulations Plus, Inc. (AMEX: SLP)
1220 W. Avenue J
Lancaster, CA 93534-2902
U.S.A.
http://www.simulations-plus.com
Phone: (661) 723-7723
FAX: (661) 723-5524
E-mail: walt.at.simulations-plus.com
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)