Back to the Top
The following message was posted to: PharmPK
Hello,
I'm a pharmacy student starting some research on gentamicin =20
pharmacokinetics on septic patients. We're evaluating if severe sepsis =20=
can change some PK parameters (we're expecting to see some changes in =20=
Vd, due to third spacing phenomenon, for example...). The number of =20
patients in our study will be low (e.g. 20 patients) and it is =20
designed as a cross-over study, because we expect to use patients when =20=
they get better as their own controls.
To start this research i'm looking to see some time-concentration =20
profiles or make some simulations of those, so I can plan when to take =20=
blood samples. I've learned that gentamicin exhibits a two =20
compartmental PK, but i'm having some difficulties in finding macro- =20
or micro-constants to build my simulation. Do you know any references =20=
that have this kind of information? Or at least some time-=20
concentration profiles. I'm only running into papers that use TDM =20
samples to calculate PK parameters, and these are extremely limited, =20
because they use a mono-compartmental approach.
Or, if you have some experience, can you suggest me the sampling times =20=
for gentamicin? Due to the low number of patients in the study I would =20=
prefer to have several time points from one patient (e.g. 8 time =20
points) so I can determine a lot of PK parameters.
Best regards,
Andre Mateus
andrenmateus.at.gmail.com
Faculdade de Farm=E1cia da Universidade de Lisboa
(Faculty of Pharmacy of University of Lisbon)
[I thought I saw pk parameters for gentamicin in Applied =20
Pharmacokinetics and Pharmacodynamics (ISBN 0-7817-4431-8 or earlier =20
edition) but I'm away from the office so can't be sure - db]=
Back to the Top
The following message was posted to: PharmPK
Most papers with parameters calculated for gentamicin were published
back in the late 60's or early 70's. You could use tobramycin or
netilmicin kinetics as proxies as they would be quite similar and
might have been published a few (2-3) years later. I will check my
old materials at home later this evening to see if I still have some
of these references.
Ron Floyd
Back to the Top
Dear Andre,
I'm afraid that your problem is a little more complicated than that.
Gentamicin is not one compound but at least 3 separate molecules (C1,
C1a and C2), the difference among them being 1-2 methyl groups. To the
best of my knowledge you cannot determine pharmacokinetics for a
mixture of components. The only correct way to analyze gentamicin PK
samples is to determine the various components separately. You can do
that by HPLC or LC/MS/MS but the real problem is the reference
material for the analysis.
Consequently, none of the 60's and 70's gentamicin PK studies take the
above into account. In other words they are useless to you. Critically
expressed, there is no pharmacokinetics of gentamicin in humans. You
can get a better idea of the problematics with an example of human PK
from our paper Isoherranen & Soback (2000) Clin. Chem. 46:837-842.
Best regards,
Stefan
Back to the Top
Dear Andre,
Due to the mechanism of action of Gentamicine (Cmax versus MIC), I
believe you should focus your sample points on Cmax.
You are going to have a huge bias if you choose cross over in septic
patients as you are going to compare two different moments in the
course of a disease. There is no way to solve this 100%. I suggest a
matched case design.
Regards,
Andre.
Back to the Top
The following message was posted to: PharmPK
Andre:
The only group that I know characterized the tri-exponential
elimination of the aminoglycosides was that of Dr. Jerome Schentag in
the late 70s. I've requested a copy of JPB 1977 v5 to see if that has
what you might be asking for. However, he was only able to describe
the third elimination phase with extended urine collections, something
that I doubt you are intending or need to do for your purposes.
Experts such as Steve Duffull may wish to weight in on the optimal
sampling times, but those too may depend on what methods you will be
using to characterize the patient's PK parameters. The BMSG ADAPT II
package , PFIMOPT from Dr. Mentre's group, and WinPOPT from Dr.
Duffull are some options for optimizing your sampling times, based
upon the model you choose to use.
You may wish to visit and join the Population Optimal Design
ListServer (PopDesign) at
http://lists.otago.ac.nz/listinfo/popdesign
to learn of more opportunies.
Best regards.
Paul Hutson
--
Paul R. Hutson, Pharm.D.
Associate Professor
UW School of Pharmacy
777 Highland Avenue
Madison WI 53705-2222
Back to the Top
The following message was posted to: PharmPK
Andre-
While, very strictly speaking, Stefan's comments are correct with
regard to the mixture, the three molecules are dealt with by the body
in an reasonably similar manner. For any clinical purpose, the
studies conducted in the 60's - 70's should suffice, particularly if
the assay used was one of the antigen-antibody type. If really
interested, you can probably still find out the relative specificities
for the interaction between the assay antibody and each of the 3
gentamicin molecules. It is an intriguing problem. However, for
getting approximate rate constants for doing simulations for clinical
purposes, the simplifed approach is quite accessible and fairly
accurate.
Ron F
Back to the Top
First of all, thank you for your quick responses.
Ron: Thanks for that information. I'll have a look at those other
aminoglycosides PK and try to get the older papers.
Stefan: I've read your paper and found it interesting that you've
looked at the PK of the different gentamicin fractions. I realized
that you started applying this technique to humans, do you have any
results? Did you try to relate this to PD?
I've read somewhere on the archives of this mailing list that
different gentamicin brands (like generic vs. proprietary brands) had
different gentamicin fractions ratio. Have you tried to quantify the
different fractions in different brands or even in different batches
of the same brand? Perhaps they could be related to different efficacy
due to different PK parameters (e.g. on your paper you showed a time-
concentration profile for one patient and the C2 fraction appeared to
have a greater half-life, this could potentially mean, that a brand
with greater percentage of C2 would have a longer effect).
I also took a look at you paper "Isoherranen et al. Antimicrob Agents
Chemother (2000) 44(6):1443-7". When you show the Total gentamicin
curve and PK parameters, how was this determined? Did you simply sum
the fractions concentrations at each time? Were you able to apply a
compartmental model to this set of points with a good correlation
coefficient? If so do you think it's possible for me to use data from
RIA even with the limitations this could bring, like different
antibody affinity to different fractions? I could reduce this, using
always the same brand of gentamicin administered to the patient, if
the fractions ratio remains the same.
Thanks,
Andre Mateus
Faculdade de Farmacia da Universidade de Lisboa
(Faculty of Pharmacy of University of Lisbon)
Back to the Top
Andre (Daher),
You wrote:
"Due to the mechanism of action of Gentamicine (Cmax versus MIC), I
believe you should focus your sample points on Cmax. "
I believe you are mistaken if you believe that Cmax reflects the
mechanism of action of antibiotics. There have been some useful but
totally empirical connections made between a narrow range of exposure
profiles and clinical outcome (Bill Craig is the father of this family
of empirical rules).
As far as I know, there are no tested theories that connect mechanism
of action (at the site of action in a bacterium) to the exposure
profile in humans which can predict how long treatment should last or
even what dosing interval to use.
My view of the literature is that the switch from q8h to q24h dosing
of aminoglycosides was made on grounds of clinical convenience plus
some minor differences in serum creatinine changes - no clinical
benefit to patient outcome has been conclusively demonstrated. I would
be delighted to hear that my view is wrong but to convince me you must
provide evidence not speculation.
[I have BCCed some researchers who are active in thinking in this area
so see if I can encourage them to contribute]
Nick
--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New
Zealand
n.holford.aaa.auckland.ac.nz
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford
Back to the Top
Dear Andrew,
Thanks for your feedback. In this gentamicin stuff, Nina and myself
represented the analytical and PK part. However, we did have a PK/PD
ambition. At the time we published the method paper, we only had this
one patient. We started a project with a local medical center, but
adequate PK sample sets were not forthcoming.
We did look at different brands of gentamicin and the differences in
component ratios are huge (as also published in the past). Actually,
you can tell from the results, which brand was used. Unfortunately,
again strictly speaking, this means that the PK without component
ratio is determined only for that specific brand of gentamicin. We did
show that the PK of the components also differ among animal species
(note the variable urine pH in them, but I can only guess the clinical
importance of this in humans). And yes, in the AAC article total
gentamicin was just a summation of the components.
As Nick pointed out, I too have my questions concerning the PD
differences versus administration frequencies, but not for the reasons
Nick said. The argument of gentamicin being a 'concentration dependent
killer' was used. I just don't understand from where the 7 mg/kg (once
daily) dose came from. So how relevant is the PD here? The 3 times
daily dose (of 4 mg/kg) is almost twice higher. However, the once
daily dose was also recommended based on toxicological considerations.
This is the second reason why component PK does have a justification.
In short: Do the componets have the same antimicrobial activity and/or
toxicity?
We had some problems in coping with the Schentag paper mentioned by
Paul. The data was analyzed with a bioassay for higher concentrations
and with an immunoassay for the lower concentrations. Now, if the
component ratio does have an effect on the two assays (or one of
them), the two parts of the curve get out of sync. I am not quite
convinced with the generalization by Ron. We observed a tendency of
the immunoassays to produce an additional 'compartment' at low
concentrations, which we did not observe with other methods. Is this a
real thing, a component related issue or a general phenomenon, I don't
know.
Today I would do the gentamicin analysis with LC/MS/MS (instead of
HPLC/UV or FLD) due to much better sensitivity (and avoiding messing
with the derivatization reagents). However, the chromatography is
clearly more complicated in LC/MS/MS due to more limited mobile phase
and column options. Whichever way you choose, it's not easy.
Best regards,
Stefan
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Gentamicin PK" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)