Back to the Top
Colleagues,
A little less than a yr ago, a member of this list suggested that, in
addition to our usual informal exchange of ideas, we undertake the more
engaging task of preparing conesnsus documents on issues of importance to
clinical practice. Soon afterward we settled on three topics (once-daily
AGs, vanco monitoring, and cont. infusion of beta-lactams). It was decided
that a draft document is first prepared by a "coordinator" and posted on
the U of Oklahoma PK site. Interested professionals would read it and
forward their criticism and suggestions to the coordinator who would
incorporate them in a revised version of the document.
I was given the task of writing the draft document on the ODA. It was
finished and posted at the end of Sept '96. Over the past six months I have
received messages concerning this document from members of this list and
from others. With one or two exceptions, all of the feedback I received was
very positive towards the ODA document and also toward the concept of
consensus documents in general. Some physicians expressed the concern that
using this dosing strategy in pts with marginal renal function may further
compromise their renal status. I have taken account of these concerns in
the revised document
(http://pharm.cpb.uokhsc.edu/pkin/consensus/Odacd.html). I have also
included a caution concerning the use of 24 hrs dosing intervals in pts
with elevated clearance (young adults, cystic fibrosis, etc).
I would like to thank all those who took the time to read the document and
send me their feedback.
I am particularly grateful to Crisanto Ronchera (Colegio Universitario de
Farmacia, University of Valencia, Spain)
for suggesting the idea of consensus documents and for his critical review
of the draft document on aminoglycosides, and to Dr. David Bourne and the
University of Oklahoma PK site for adopting the concept
and forstering the development of consensus documents on topics of critical
importance to the practice of Clinical Pharmacy. I believe this experience
establishes a positve example for other internet groups/lists to emulate.
Dr. Nasr Anaizi
Clin Grp Leader
Associate Prof
Univ of Rochester Med Cntr
nanaizi.aaa.pharmacy.urmc.rochester.edu
Back to the Top
Just finished glancing over the "Updated ODA consensus document." at
http://pharm.cpb.uokhsc.edu/pkin/consensus/Odacd.html
Wasn't on the list when the discussions first occurred, but looking at the
"Dosing" part of things, I wonder if others can clarify it for me.
From my reading of it, I think it suggests that the dose of
gentamicin/tobramycin is 5 mg/kg daily AND depending on the patient's CrCl the
dosing interval should then be increased to 36 and 48 hours. Thus the
recommendation is to alter the interval according to the CrCl.
Therefore, my query is, can we or why don't we change the dose instead?? For
example, if person X has a CrCl of 40 mL/min, is it permissible to use 2-3
mg/kg instead of 5 mg/kg every 24 hours rather than 36 hourly??? For a 70 kg
person, this may translate to 120-180 mg once daily. In most cases, the peaks
will be > 10 x's MIC (think around 14-20 mg/L off the top of my head) and the
aminoglycoside free period will still be attainable. Plus, penicillins are
often used in combination for synergy.
I know from clinical experience that doses of 120 mg/day or even 80 mg/day are
often seen in particularly geriatric patients will low CrCl's.
Wonder if someone can elighten me.
Cheers,
Dave
-------------------------
David Ng, BPharm MClinPharm
PhD Candidate
School of Pharmacy and Medical Sciences
University of South Australia
North Terrace
ADELAIDE SA 5000
Australia
Fax: (61) 08 8302 2373
http://www.merlin.net.au/~psycho
e-mail: psycho.-a-.merlin.net.au
-----------------
Back to the Top
Dr. Nasr Anaizi
I have only joined recently the PK/PD Discussions and was interested to
read your Email, and visit your WWW site regarding a consensus opinion
about once daily dosing of aminoglycosides. I would like to offer some
comments. We have been using a once daily dosing regimen for
gentamicin/tobramycin for 4 years at our institution and have dosed several
thousand patients using this dosing regimen. Over this time we have used a
target Cmax/AUC monitoring technique, which we have found of considerable
use (see refs for this method). The target AUCs that we use are between 70
and 100 mg/L.h. These targets were chosen since it was believed that the
total daily dose of gentamicin/tobramycin should remain the same whether
the dose is given conventionally or as a once-daily dose, and these are
the AUCs that are normally achieved with conventional dosing when peaks of
between 6 and 8 mg/L and troughs of 1 and 1.5 mg/L are targetted. This AUC
method has an advantage in that it gives the clinician a choice of doses to
use, ie a patient with pyelonephritis who is otherwise fit and healthy
should probably receive a smaller 'expsoure' to aminoglycosides than a
patient with septicaemia where the focus of the infection may be in a
distal compartment. Typical starting doses are therefore between 3 and 7
mg/kg/day (depending on renal function as would be used in conventional
dosing). The dose is then adjusted to achieve the appropriate AUC over a
24 period (there is no evidence yet to support extended interval dosing?).
This has resulted in doses between 3 and 11 mg/kg/day.
Because we dose-individualise the aminoglycoside we do not perceive
problems in treating patients with burns, sepsis or other conditions which
may result in increased extravascular volume. In addition we have been
dosing aminoglycosides once daily in neutropenic patients (oncology) for 2
years [the largest single study in once daily dosing was in neutropenic
patients]; and we have recently added haematology neutropenic patients to
our inclusion criteria. Dose individualisation is performed from one or
two (usually two) SDCs. One is taken at approx 30 min post dose and the
second at a time between 6 and 14 hours post dose. The AUC is determined
using Bayesian techniques. The advantage in taking a sample at between 6
and 14 hours is that the AUC may be estimated prior to the next dose,
thereby avoiding the unnecesary continuation of an inappropriate dose. The
graph/nomogram method is likely to produce similar results, although will
not allow the same level of individualisation to occur for a patient. For
example it is possible to follow a patient's PK parameters throughout a
course. It is not uncommon for patients to change the values of their PK
parameters during a course, ie a hyperdynamic septic patient may return to
normal once the sepsis has subsided or alternatively a patient may go off.
Both of these events may be followed as it happens (ie before changes in
serum creatinine have been observed) and appropriate clinical measures
taken where needed.
Measuring a concentration at the trough does not seem to be overly helpful
given the poor limit of detection of the most commonly used assays. It
also does not allow the patient's exposure to the aminoglycoside to be
determined, hence there is a risk of underdosing, or even overdosing in a
patient with very high creatinine clearance. It also does not allow dose
adjustments to be made before the next dose. In our experience once daily
dosing of aminoglycosides continues to have a small therapeutic index, and
depending on the patient a persistent AUC of between 30 to 100% greater
than desirable may result in clinically significant nephrotoxicity.
I realise that my comments are after the fact, but I thought that you might
like to consider them if you intend to update the consensus statement.
Barclay ML, Duffull SB, Begg EJ. Experience with a method of once-daily
dosing of aminoglycosides. Aust NZ J Med 1995;25:230-5
Begg EJ, Barclay ML, Duffull SB. A suggested approach to once-daily
aminoglycoside dosing. Br J clin Pharmac 1995;39:605-9
I hope this is useful.
Steve Duffull
=======================Steve Duffull
Dept Clinical Pharmacology
Christchurch Hospital
Private Bag 4710, CHCH
New Zealand
Ph +64 3 364 0900
Fax +64 3 364 0902
Back to the Top
Dear Steve,
Thank you very much for taking the time to read and comment on the ODA
topic. I am aware of your fine work in the two papers you cited, and
the recent paper (BJCP) in which you compared the two computerized
Bayesian methods (ABBOTTBASE and SeBA-GEN). I do appreciate your point
of view and I believe that the method ("Bayesian
dose-individualization") you are using is probably advantageous and
warranted under certain circumstances (e.g., in hyperdynamic pts).
However, there are a couple of problems with this dosing/monitoring
technique:
1- It negates most of the advantages of the pulse-dosing method,
particularly the ease and simplicity of dosing and monitoring.
2- It follows the standard/traditional methodology without providing a
plausible PD or PK rational as to why we should do so. It is difficult
to explain why should the bacteriological efficacy of AGs correlate
better with Cmax/AUC than with Cmax.
However, I am very much interested in the practical details of how your
method is being carried out at your hospital; from physician order to
the pharmacist, and pharmacist-physician dialog, etc.
I would appreciate any details you can give me.
Thanks
Back to the Top
Dave
I thought that I might comment on your enquiry. We, in our institution,
usually alter the dose rather than the interval. Only when the expected or
observed Cmax becomes less than 10 mg/L (the maximum Cmax achieved on
'conventional' therapy) do we prolong the dosing interval. Typically 3
mg/kg/d is the lowest once daily dose that we use in patients with poor
creatinine clearance (20-30 ml/min) and this usually achieves an AUC(0-24h)
around 100 mg/L.h. If the AUC(0-24h) becomes significantly greater than 100
mg/L.h, in a patient with poor renal function, then we either prolong the
dosing interval to target an average AUC(0-24h) of 100 or (more commonly)
use another antibiotic. As usual we recommend monitoring to ensure that
the dose has been appropriately individualised.
I hope this is helpful.
Steve
=======================Steve Duffull
Clinical Pharmacology Dept
Christchurch Hospital
Christchurch 80201
New Zealand
Ph +64 3 3640900
Fax +64 3 3640902
Back to the Top
Dear Nasr
> Thank you very much for taking the time to read and comment on the ODA
> topic. I am aware of your fine work in the two papers you cited, and
> the recent paper (BJCP) in which you compared the two computerized
> Bayesian methods (ABBOTTBASE and SeBA-GEN). I do appreciate your point
> of view and I believe that the method ("Bayesian
> dose-individualization") you are using is probably advantageous and
> warranted under certain circumstances (e.g., in hyperdynamic pts).
> However, there are a couple of problems with this dosing/monitoring
> technique:
>
> 1- It negates most of the advantages of the pulse-dosing method,
> particularly the ease and simplicity of dosing and monitoring.
>
I guess what we are doing is based on the consideration that there is no
evidence telling us how we should be dosing/monitoring aminoglycosides. We
do not therefore know the ease and simplicity of monitoring. Recently you
will be aware of comments regarding a meta-analysis on ODA published in the
BMJ. In essence it was suggested that in the majority of cases there is no
need to monitor ODA. I think that this was a bit of a paradigm shift -
really we should be saying that until we have evidence to support that we
do not need to monitor we should continue to be vigilant with
aminoglycosides (these drugs afterall remain toxic). The AUC method still
retains some of the perceived advantages since i) we are only giving Ag
once daily hence saving on administration costs and ii) we are only
advocating monitoring every 3 days (unless the clinical picture depicts
otherwise) both of these are less intensive than with conventional dosing.
There is a risk that if practitioners do not monitor in a manner which
allows doses to be adjusted for the individual then preventable toxicity
may occur. This leaves the practitioner open for potential litigation
problems.
>
> 2- It follows the standard/traditional methodology without providing a
> plausible PD or PK rational as to why we should do so. It is difficult
> to explain why should the bacteriological efficacy of AGs correlate
> better with Cmax/AUC than with Cmax.
Bacterial kill has certainly been related to Cmax. In our studies we have
shown that the higher the concentration the greater the bacterial kill. It
might be argued that a Cmax of 50 mg/L would provide better bactericidal
effect than a Cmax of 25 mg/L but this may be at the expense of increased
toxicity. It is likely also that toxicity of aminoglycosides, similar to
other drugs (eg carboplatin), is related to the AUC. Therefore targetting
the highest Cmax that is possible while maintaining the AUC within an
acceptable range would seem an appropriate compromise. Hence we consider
Cmax in terms of effectiveness and AUC in terms of toxicity. The balance
between these two variables is the goal of dosing.
What we do not know yet is what AUC is optimal. We are attempting to
address some of these issues at the moment.
>
> However, I am very much interested in the practical details of how your
> method is being carried out at your hospital; from physician order to
> the pharmacist, and pharmacist-physician dialog, etc.
>
> I would appreciate any details you can give me.
>
For practical purposes the dialogue is relatively simple. The physician
decides whether to use Ags. They then determine a dose that they consider
appropriate for the patient. This is usually in conjunction with the
clinical pharmacist. Considerations for the dose will include: the
patients ability to clear the drug, how sick the patient is, and where
focus of the infection is located. Once the dose is administered two blood
samples are taken off the first dose. The pharmacist then, in conjunction
with the physician, determines the most appropriate next dose, when future
SDCs should be taken, and follows up basic biochemistry. It remains the
responsibility of the physician to chart the doses, although pharmacist
recommendations are accepted almost universally. [All recommendations are
annotated in the patient's notes.] All pharmacists are skilled at TDM using
Bayesian and linear least squares methodology.
In essence therefore the pharmacist drives the Ag dosing once the initial
decision has been made to use one.
I hope this is helpful.
Do not hesitate to contact me if you wish to discuss further any of these
points.
Regards
Steve
=======================Steve Duffull
Clinical Pharmacology Dept
Christchurch Hospital
Private Bag 4710
Christchurch 8020
New Zealand
Ph +64 3 3640900
Fax +64 3 3640902
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)