BackgroundPregnancy may cause changes in drug disposition. The clinical consequences may be profound and even counterintuitive; in some cases pregnant women may need more than twice their usual drug dose in order to maintain therapeutic drug levels. For antidepressants, evidence on drug disposition in pregnancy is scarce. The aim of this study was to determine the effects of pregnancy on serum levels of selective serotonin reuptake inhibitors (SSRIs) and venlafaxine in a large and naturalistic patient material, in order to provide tentative dose recommendations for pregnant women.MethodsUsing patient data from two routine therapeutic drug monitoring (TDM) services in Norway with linkage to the national birth registry, dose-adjusted serum drug concentrations of SSRIs and venlafaxine during pregnancy were compared to the women’s own baseline (non-pregnant) values, using a linear mixed model.FindingsOverall, the TDM databases contained 196,726 serum concentration measurements from 54,393 women. After data linkage and drug selection (SSRIs or venlafaxine only), we identified 367 analyses obtained from a total of 290 pregnancies in 281 women, and 420 baseline observations from the same women. Serum concentrations in the third trimester were significantly lower than baseline for paroxetine (–51%; 95% confidence interval [CI], –66%, –30%; p<0.001), fluvoxamine (–56%; CI, –75%, –23%; p = 0.004) and citalopram (–24%; CI, –38%, –7%; p = 0,007), and higher than baseline for sertraline (+68%; CI, +37%, +106%; p<0.001). For escitalopram, fluoxetine and venlafaxine concentrations did not change significantly.ConclusionsFor paroxetine and fluvoxamine the pronounced decline in maternal drug serum concentrations in pregnancy may necessitate a dose increase of about 100% during the third trimester in order to maintain stable concentrations. For fluoxetine, venlafaxine, citalopram, escitalopram and sertraline, the present study indicates that dose adjustments are generally not necessary during pregnancy.