Problem with multiple dose input

From: Lewis Sheiner <76135.526@compuserve.com> Subject: Problem with multiple dose input Date: 23 Oct 1997 03:31:17 -0400 Previous note: > Dear NONMEM users, > > I would like to alert you to a problem I encountered recently which > cost me some time and nerves. I am not sure what is going on, perhaps > you may have some ideas..... > > I was developing a model with simultaneous input into 2 compartments: > > TIME CON AMT RATE CMT EVID MDV SS > 0 0 5000 20000 1 1 1 0 > 0 0 0 -1 3 1 1 1 > 0 0 9999 -1 3 1 1 0 > > The input to compartment 1 is an infusion at time 0, the input to > compartment 3 is a steady-state infusion followed by a nonsteady-state > infusion both also at time 0 and both of which have their rates > estimated. Surprisingly (to me at least) this dosing code did not > produce ANY input into either compartment. ARE YOU SURE? I CAN SEE WHY THE FIRST RECORD PRODUCED NOTHING (SEE BELOW), BUT IT SEEMS TO ME THAT IF YOU MODELED R3 PROPERLY, YOU SHOULD HAVE SEEN THE RESULTS OF THE SECOND 2 LINES; THAT IS, THE SS INFUSION ENDING AT TIME 0 AND THEN THE SINGLE-DOSE INFUSION BEGINNING AT TIME 0, BOTH INTO CMPT 3. BUT MAYBE THERE'S SOMETHING MORE SUBTLE HERE THAT ESCAPES ME ... > After much trial and error I discovered that the following dosing code > worked. For some reason I had to put the steady-state dose (into CMT= > 3) before the nonsteady-state doses into CMT 1 (or CMT 3). > > TIME CON AMT RATE CMT EVID MDV SS > 0 0 0 -1 3 1 1 1 > 0 0 9999 -1 3 1 1 0 > 0 0 5000 20000 1 1 1 0 > > For me at least it is not intuitive that doses to different > compartments should interfere with each other. I would be interested > in hearing your comments. SS DOSES RESET THE SYSTEM (EXCEPT IF YOU MAKE SS =2, IN WHICH CASE THEY ADD THEIR SOLUTIONS INTO THE SYSTEM). THEREFORE, ANYTHING THAT PRECEEDS A SS RECORD WITH SS=1 DISAPPEARS. THIS IS APPRPRIATE. HERE IS THE APPRPRIATE DISCUSSION FROM THE 'HELP' DOCUMENT: SS labels PREDPP's steady-state (SS) data item. The steady-state data item is optional. It can take one of three values in any event record. 0 indicates that the dose is not a steady state dose. 1 indicates that the dose is a steady state dose, and that the compartment amounts are to be reset to the steady-state amounts resulting from the given dose. Compartment amounts resulting from prior doses into the system are "zeroed out," and infusions in progress or pending additional doses are cancelled. The system is not totally reset: the on/off status of the compartments remains as it was at the time of the prior event record (if any), and the value of time must be greater than or equal to its value on the prior event record (if any). 2 indicates that the dose is a steady state dose and that the compartment amounts are to be set to the sum of the steady-state amounts resulting from the given dose plus whatever those amounts would have been at the event time were the steady-state dose not given. I.e., letting t be the time on the event record, then the amounts in the compartments are updated to amounts valid for time t, and next, these amounts are added to the steady-state amounts. This is meaningful when kinetics are linear and the superposition principle holds.