My question is that how would you select the most appropriate way for PK analysis (model Selection)?
As a common way to select models among compartmental model, Akaike information criterion(AIC) is commonly used for model selection. The lower the AIC value, the more appropriate model it is. However, when should we consider compartmental models rather than noncompartmental model?
If we got intensive sampling points, is it more appropriate to use noncompartmental analysis instead of compartmental analysis?
Hi Yu-Syuan,
Check this out:
Br J Clin Pharmacol. 1998 Jan; 45(1): 63–70.
The time of maximum effect for model selection in pharmacokinetic– pharmacodynamic analysis applied to frusemide
Monique Wakelkamp, Gunnar Alván,1 and Gilles Paintaud
You should always to perform your analysis firstly using non compartmental analysis. This way you'll have access to parameters such as MRT and Vss WITHOUT anay mathematical hypothesis on the data. Then you can perform different modeling but you should take the physiology and the properties of your API into consederation. In fact, if you use the aikaike criteria most chances are that it will be better for multiple compoartment, it is the same for fitting parameters, but it is not always relevant to perform a multiple compartment analysis. For example, if your API is hydrophilic and you have a distribution volume around 7L it means that it stays in the blood stream and does not go to the tissue, in this case EVEN IF a double compartment models fits better I would not use it. In conclusion, you have to justify chemically and physiologically your modelling choices.
