Back to the Top
The following message was posted to: PharmPK
1. What is (are) the most sophisticated out-patient drug dosage
calculator(s) available over the internet.
2. Does anyone know of specific calculators aimed at special
c. pediatric (I know some of these exist)
Back to the Top
Any answer posted so far regarding the message below? I am interested too.
>Subject: PharmPK Dosage Calculators
>Date: Thu, 17 Aug 2000 11:38:22 -0500
>1. What is (are) the most sophisticated out-patient drug dosage
>calculator(s) available over the internet.
>2. Does anyone know of specific calculators aimed at special
> a. obese
> b. elderly
> c. pediatric (I know some of these exist)
> d. anesthetics
> d. etc.
Back to the Top
I thought that my question had been culled. I'm pleased to see that it
did get distributed.
On the other hand the general lack of response could be an indication that
little attention is being paid to the process of clinical dosage adjustment.
There are many known factors that suggest doses should be adjusted, but
unfortunately there are also many unknowable factors that make this a tricky
My sense is that some physicians do not consider issues of weight (up and
down), age, gender, disease and knowable factors in establishing initial
doses. Is this justified? Is the variability due to unknown factors so great
that it is not worth the effort to consider known factors? I would anticipate
that the current approach might be to give a dose and fix it if the side
effects are extreme. However, the elderly frequently report more dose
related side effects, suggesting that obvious dosing adjustments may not be
I am interested in the concept of prescribing assisted by linked clinical
data bases. With these systems, information about drugs, drug interactions,
and PATIENT CHARACTERISTICS can automatically be considered. However it is
not obvious how such patient characteristics might be included in drug
selection and dosing. Is dosage adjustment based on creatinine clearance
When a patient has multiple characteristics that suggest dosage
adjustments could be warranted how should it be handled? (Example: female,
congestive heart, obese) I am reluctant to concluded that you can simply
sum up factors to come up with the correct dose adjustment. Although this
would be logical, it potentially exposes a practitioner to legal challenged if
the projected dosage results in a prescription where the doses "appear" to be
In addition, while many drugs are prepared in a variety of strengths,
these strengths do not seem to reflect common dosing groups. For example, an
average woman is said to be about .85 of an average man. There are no .85 *
male strength dosage-forms generally available. Thus, female patients are, to
some extent, overdosed and also to some extent more likely to exhibit high
dose side effects.
I am unaware of general dosing guidelines for athletes (consider the
Olympics). Clearly this group would be expected to clear many drugs more
quickly than an average patient. To complicate the thought, lets compare the
dosing challenges between a 7' African American professional basketball player
to a 80 lb elderly vietnamese grandmother. We have both of these types in
our patient populations in Houston.
There are some rules for children, however children may metabolize many
drugs at or above adult rates and this is may not be considered by simple
There are some rules for the obese but they seem to be empirical and I am
uncertain as to what the theoretical basis is for handling such patients.
Another characteristic of dosing for the obese is that the regimens reported
in the literature are frequently non-linear, i.e. use rate x up to such a
weight and then use rate y. There would seem to be a pressing need for dosage
adjustment information for the obese in light of the increasing number of the
obese people in this country.
So the question is: has it been possible to develop dosage adjustments
that take patient characteristics into consideration for the outpatient drug
use? Are they clinically good enough?
(Good enough might be defined as: the rule correctly doses 90% of the patients
that it us used for).
I am aware of the book Applied Pharmacokinetics: Principles of Therapeutic
Drug Monitoring, ed Evans et al. which considers some of these issues.
Back to the Top
I think you have summed up a range of typical problems
facing the pharmacist/clinician on a day to day basis. I am
not going to attempt to address any of these issues
specifically - other than to say most of us consider many of
these factors when arriving at a somewhat empirical dose.
The computer program "Practical Pharmacokinetics"
(http://www.clinpharmint.com/) attempts to address some of
these problems. I am unaware if it meets your criteria of
"good enough" though.
> (Good enough might be defined as: the rule
> correctly doses 90% of the patients
> that it us used for).
> dosing groups. For example, an
> average woman is said to be about .85 of an
> average man. There are no .85 *
sounds rather perilous...
School of Pharmacy
University of Queensland
Brisbane, QLD 4072
Ph +61 7 3365 8808
Fax +61 7 3365 1688
Back to the Top
The issue of individualizing drug dosage regimens is most
important, but most physicians do not want to do it, as they think it
is too complicated, and too mathematical. Curriculum committees avoid
it. This is why all the pharmacology that gets taught is about how
drugs act, but nothing about how to use them optimally.
Our laboratory has been working in this area since out first
software for digitalis in 1967, on a time-shared mainframe, as a
consulting service for physicians. It permitted significant reduction
in toxicity without loss of efficacy. The reaction has generally been
that if ignoring it, except for a nomogram we did, which was liked
for a while. Use of more up-to date software has made digoxin dosage
much easier and more effective for me. When I had an elective class
for 3rd and 4th year students, even those who came to med school with
good math backgrounds, was "more work than it's worth", and "more
appropriate for pharmacists". Since when is a physician not a
therapist?! If he is not that, what is he?
Anyway, there is actually much work being done in this area,
but most of it is geared to drug development and the drug industry.
Application to patient care by monitoring drug levels is the kiss of
death for a company's drug as far as marketing goes, but they would
do it if physicians asked for it. We spend all this money for these
very expensive drugs for AIDS, for example, and then never protect
our investment to make it go optimally - and at what human cost? The
same for cancer - little optimization to true individualization. They
will treat to an AUC now, but it is usually a population AUC, and no
one really asks if the AUC can be made more optimal for each
So your surprise at the lack of response is well taken.
However, there is software available for dosage adjustment, such as
the MW/Pharm from the Netherlands, the Abbottbase program from
Abbott, the APIS from Marseilles, and our USC*PACK software, which as
far as I know is the oldest package for patient care. Take a look at
our web site (see below), and you might be interested in what you
see. In addition, we are having a workshop on the topic in Singapore
Oct 16-18, and in LA on Nov 3-4. Info is on the web site.
Please let me know if you have further questions or comments.
It is good to hear your voice.
Roger W. Jelliffe, M.D. Professor of Medicine, USC
USC Laboratory of Applied Pharmacokinetics
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.aaa.hsc.usc.edu
Our web site= http://www.usc.edu/hsc/lab_apk
Back to the Top
Dear Dr Lewis,
This is a very big question and I am going to give an uncharacteristically
The covariates typically available in the clinic are at best poorly
correlated with pharmacokinetic and pharmacodynamic factors, and have only
vague mechanistic links for specific drugs. Dose adjustments which ignore
this, for example, dosing linearly according to BSA and assuming a zero
intercept actually increase variability in pharmacological exposure.
If someone worked out how to accurately measure enzyme function (specific
for each drug) in the liver, this may change. We can measure renal
function to some degree and should exploit this.
TDM is a lot harder than most people think because of IOV and systematic
evolution of pharmacokinetic processes, not to mention pharmacodynamic
Okay, so it wasn't that short. I await the backlash. Specific agents may
be exceptions to the above general comments.
Back to the Top
I mentioned physiologically-based pharmacokinetic (PBPK) models to this
group a few weeks ago and got a lot of negative feedback about their lack of
utility in clinical pharmacology - but I would like to point out that they
could be used to address this issue of dosage calculators in a
straightforward manner. For example, if someone is obese, adjust the tissue
blood flow rate and tissue volume parameters appropriately and the model
will provide PK predictions customized for that individual. These will be
some uncertainty associated with these predictions, but arguably less
uncertainty than if the predictions are obtained with a nonphysiological
The gist of the objections to PBPK modeling in response to my earlier
posting were that it was "useless" and that the models were too difficult to
develop. Model complexity is an issue, but the model development process
tends to be much more challenging for the first member of a class of
chemicals than for subsequent members. Modifying an existing model for a
structurally related chemical can often proceed relatively quickly. In
addition, algorithms are being developed to predict the values of key
parameters, such as partition coefficients. Some work has also been done on
algorithmic prediction of clearance (for polychlorinated biphenyls). I
suspect that if some of the resources of the pharmaceutical industry used to
examine how PBPK models might be exploited in support of drug development
that rapid progress would be made.
PBPK modeling shouldn't be dismissed as useless or too demanding without
thinking carefully about the cost-benefit issues. It may sometimes be
useful to develop an "expensive" PBPK model, whatever that means, if the
model can then be used to help optimize the design of clinical trials,
provide individualized dosage calculation, help interpret toxicology data,
and so on.
- Rory Conolly
Rory B. Conolly, Sc.D., D.A.B.T.
Six Davis Drive
Research Triangle Park, NC 27709
[Interested members might want to look at MacDope
http://www.chime.ucl.ac.uk/Models/macdope.htm - db]
Back to the Top
One of the problems is the evaluation tools availability. We have
few data such as lean body weight, general hepatic and renal
function, ethnic origin etc. But tools for individual's body
functions are generally not available, expensive or at research
state. Using dose adjustment during therapy can only be done in
centers with lots of analytical supports or experience(emergency
We know patients defficient in dihydropyrimdine dehydrogenase can
have mucositis when treated with normal dose of 5FU, What about
cytochrome 2D6(genetic) and 3A4(variable)? Simple tools for
evaluation are not available.
To complicate matter, high dose chemotherapy can reach the limit
of normal functions. An example is a trial of high dose carboplatin
with cytoxan and etoposide. We did not reach our target AUC in dose
escaulation. Is it due to measurement error, hydration(urine output~4
liters/day) which changes plasma concentration, or saturated renal
reabsorption which may not occure at lower doses.
The 3A4 test is now commercially available and hopefully more
will follow, than individualized dosing will be reality.
Memorial Sloan Kettering Cancer Center
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (firstname.lastname@example.org)