Application of population modeling to characterize the effect of pregnancy on aspirin pharmacokinetics

Document Type

Conference Proceeding

Publication Date




Low dose aspirin is recommended for preeclampsia prevention in high risk pregnancies. However, there remains uncertainty regarding its optimal dosing due to the lack of pharmacokinetic (PK) based clinical studies in pregnancy. Herein we developed a population PK model for aspirin in preeclampsia using data from a preliminary study to investigate the alterations of aspirin PK through gestation.

Study Design

A population PK model was developed using data from a clinical trial in which aspirin was started in women at risk of preeclampsia (Table 1) at a daily dose of 81 mg in the first trimester (10-14 weeks’) of pregnancy through delivery (N=10). Samples were collected from 0-6h following the first aspirin dose with a similar schedule in the third trimester (28-32 weeks’). Salicylic acid (primary metabolite) was the analyte used in the construction of the PK model. Modeling was done following standard development and qualification procedures.


A one-compartment model with zero-order formation and linear elimination best described the PK of salicylic acid (Table 2). A correlation between volume of distribution (V) and clearance (Cl) was added to adjust for the correlation between random effects. The observed drop in Cmax in the third trimester was explained by the inclusion of the gestational age as a covariate on V as follows: V in pregnancy (L) =5.79 * gestational age (weeks)ˆ0.21, i.e., V increases with the gestational age by a factor beta = 0.21. Cl did not change between trimesters as well as the duration of zero order metabolite formation.


Gestational age increases V of salicylic acid without altering elimination. This model-based approach can help optimizing aspirin dosing for the prevention of preeclampsia understanding that V is the principal parameter altered through gestation. Clinical data at 162mg dosing would assist in further refining the model, which could then be used to model aspirin dose/response for future trials or clinical use.


Additional Files