From:Batul Parta
Subject: [NMusers] Renal CL estimations  
Date: Tue, June 15, 2004 7:44 pm  


I think what I am going to ask has been discussed
before. Unfortunately, I cannot find it anywhere
so I ask it here. I am trying to model the PK of a
compound with a fe-value (fraction excreted unchanged
in urine) close to 60%. This value is based on total
amount found in urine within 24 hours after drug administration
(thalf = 3 hrs). All urine was collected and pooled. I was
thinking to estimate the renal CL (CLR) by adding an output compartment:






However, this obviously won't work since the amount measured
in urine is the sum of all that came out in 24 hours. There
is no information over time for the model to be able to pick
up the excretion rate. 

Now, imagine if the urine samples were not pooled and one
could get the individual amounts for each urine sample over
time. Would that enable me to use the equations above to get
an estimation of fe? If so, what is the time-point of the
urine measurement? Half-way from the start to the end of
the urine collection period, as is done for NCA?

I would be very greatful for any published reference.  

Best regards,

B. Parta

From: "Mats Karlsson"
Subject: RE:[NMusers] Renal CL estimations
Date: Wed, June 16, 2004 1:48 am

Hi Batul,

If you have measured amount in urine over the period 0-24 h. Making
an observation in compartment 2 at 24 h will give you a prediction
of just that. You don’t need to think in terms of excretion rate,
neither in observations nor predictions. The coding becomes a bit
tricker if you have multiple urine observations as urine compartment
needs to be “emptied” and collection restarted, which is done in the
data file by using EVID 2 and ON/OFF signaled through CMT = 2 or -2.
I send some coding separately. Whenever you observae amounts in
urine, the time of the observation is when you made it, that
is at the end of the observation interval.

Parametrisation with CLR and CLNR is likely to involve less
covariance between parameters than CL and FE.


Best regards,

Mats Karlsson, PhD
Professor of Pharmacometrics
Div. of Pharmacokinetics and Drug Therapy
Dept. of Pharmaceutical Biosciences
Faculty of Pharmacy
Uppsala University
Box 591
SE-751 24 Uppsala
phone +46 18 471 4105
fax   +46 18 471 4003

From: Leonid Gibiansky
Subject: RE:[NMusers] Renal CL estimations
Date: Wed, June 16, 2004 8:34 am   

A(2) is the drug collected in the urine compartment. So you can include 
total amount found in urine within 24 hours into the data set at time 24 as 
A(2) (you need to use appropriate values of the CMT data item to code it, 
CMT=1 for the central compartment observation and CMT=2 for the A2 
observations). Similarly, if you have cumulative A2 data from time 0 to 
time t, you can include those at time t (at the end of the urine collection 
interval). If you have sample urine data (not cumulative) you need to reset 
the A2 compartment to 0 at each A2 measurement (assuming that this time 
coincides with the start of the new urine sample collection interval)