? Emptying the Gall Bladder in NONMEM

1 messages 1 people Latest: Oct 13, 1999

? Emptying the Gall Bladder in NONMEM

From: Matt Hutmacher Date: October 13, 1999 technical
From: "HUTMACHER, MATTHEW" <MATTHEW.HUTMACHER@chi.monsanto.com> Subject: ? Emptying the Gall Bladder in NONMEM Date: Wed, 13 Oct 1999 13:03:55 -0500 > Hello All, > > I have an interesting problem, which some of you may have previously > solved. Any insight or help would greatly be appreciated. I am trying to > mimic gall bladder filling and emptying. My problem is: > > I have a three compartment model open model. It is probably easier to > write out the differential equations rather than to use English. Let (.) > represent differentiation with respect to time, k01 is a zero-order input, > k12 the rate of distribution to the gall bladder compartment, k31 is the > rate from CMT 3 to CMT 1, k30 is the rate of elimination. The equations > are: > > i) (A1)= k01-k12*R1+k31*R3 > ii) (A2)= k12*R1 -S > iii) (A3)= -k31*R3+S-k30*R3 > > S is a function to mimic gall bladder emptying and should have the > following properties: CMT 2 (gall bladder) is filled by CMT 1 until a > specified time then all of CMT 2 is emptied as a bolus into CMT 3. S > needs to be defined as the total amount in CMT 2 at the point of this > bolus emptying (note: if it is not the system will go to infinity). At > steady-state S will be the same at each emptying, but when the system is > initiated, S will be dependent on time and so S will be S=S(t). That is, > S(t) will depend on the results of the integration of the equations since > the last emptying. My problem is how to code this in NONMEM. Can one use > dummy bolus doses into CMT 3 where the bioavailability fraction F3 is set > to the amount A2 at the time of dumping and simultaneously reset the about > in CMT 2 equal to 0? I am not sure that it is possible to link the thetas > in the $PK block with a result from the $DES block. > > Any help will be greatly appreciated including other approaches to this > problem. > Thank you. > > Matt