Pharmacokinetics and Drug Interactions in Pregnancy

Pregnancy rewires nearly every system involved in processing medications, from the liver enzymes that break drugs down to the kidneys that flush them out, to the expanded blood volume that dilutes them. These shifts can make a dose that worked perfectly before conception either too weak or unexpectedly potent, and they evolve across trimesters rather than staying fixed. When drug-drug interactions are layered on top, the picture gets even less predictable, because pregnancy itself acts like an invisible co-medication altering the same pathways that other drugs target.

How Liver Enzymes Change During Pregnancy

The liver is the body’s main drug-processing plant, and pregnancy remodels it in selective, sometimes contradictory ways. The key players are a family of enzymes called cytochrome P450s, or CYPs. Pregnancy does not simply speed them all up or slow them all down. Some ramp up dramatically while others drop off, and the direction of change depends on the specific enzyme.

Two of the most clinically important enzymes, CYP2D6 and CYP3A4, become substantially more active as pregnancy progresses. CYP2D6 activity climbs steadily across trimesters, rising by roughly a quarter in the first trimester and nearly half by the third, compared to postpartum levels. CYP3A4 activity increases by about 35 to 38 percent throughout pregnancy and stays elevated the whole time.¹ These two enzymes together handle a huge share of commonly prescribed medications, including many antidepressants, pain relievers, and anti-nausea drugs.

Meanwhile, CYP1A2 heads in the opposite direction. Its activity drops by about a third in the first trimester and falls further as pregnancy advances, reaching roughly a 65 percent reduction by the third trimester.¹ CYP1A2 metabolizes caffeine and several psychiatric medications, so a drug cleared through this pathway can linger longer in the body during pregnancy. The practical consequence is that a standard dose may produce higher blood levels than expected.

This push-pull pattern is well documented but still underappreciated clinically. Pregnancy alters enzyme function in ways that are both gestational-stage-specific and tissue-specific, meaning the magnitude of change can differ depending on how far along the pregnancy is and which organ is doing the work.² This matters because a dose adjustment that works in the second trimester may be wrong by the third.

What Happens to Kidney Clearance

The kidneys handle a different slice of the drug-elimination workload, and pregnancy accelerates them too. Blood flow to the kidneys increases significantly during gestation, and the filtration rate climbs along with it. By late pregnancy, the rate at which the kidneys filter blood can nearly double compared to non-pregnant levels.³

For drugs that are eliminated primarily through the kidneys rather than the liver, this acceleration means the medication is cleared out of the body much faster. Blood levels drop, and the drug may not maintain effective concentrations between doses. This effect is especially relevant for certain antibiotics, antivirals, and blood thinners that depend heavily on renal excretion. The increase in kidney clearance is one reason why some drugs need more frequent dosing or higher doses during pregnancy, and why the dosing often needs to be pulled back quickly after delivery when the kidneys return to their baseline pace.

Where the Clinical Consequences Show Up

The enzyme and kidney changes described above are not abstract physiology. They create real, measurable problems for specific drug classes that pregnant people commonly need.

Anti-Seizure Medications

Lamotrigine is one of the most-studied examples. It is widely prescribed for epilepsy and bipolar disorder, and its clearance accelerates sharply during pregnancy. In a study tracking over 300 blood samples across 53 pregnancies, total lamotrigine clearance peaked at roughly 94 percent above non-pregnant baseline by the third trimester.⁴ That is nearly double the normal rate. The clinical fallout was clear: women whose blood concentrations dropped below a certain threshold relative to their pre-pregnancy baseline had significantly more seizures in the second trimester.

What makes lamotrigine tricky is that the clearance change varies widely between individuals. Some women experience a moderate increase while others see their levels plummet. Research into dosing strategies has found that early trimester blood-level monitoring can help identify whether a given patient falls into a low-clearance or high-clearance subgroup, but a blanket dose-increase strategy does not work well for everyone.⁵ Regular blood-level checks throughout pregnancy are the standard recommendation for this drug.

Antidepressants

Selective serotonin reuptake inhibitors, the most commonly prescribed class of antidepressants, are also affected, though the pattern is less uniform. Some women metabolize SSRIs faster during late pregnancy and need higher doses, while others show little change. For sertraline, dose increases are often required early in the third trimester to prevent depressive symptoms from breaking through. For fluoxetine, evidence suggests clearance rises between about 20 and 36 weeks of gestation. Across the class, some women end up needing nearly twice their pre-pregnancy dose to maintain the same clinical benefit.⁶

The interindividual variability here is a recurring theme. Not every pregnant person on an SSRI will need a dose adjustment, but many will, and the timing and magnitude are hard to predict without monitoring. Equally important is what happens after delivery: the pregnancy-driven enzyme acceleration reverses, and a dose that was appropriate in the third trimester can suddenly become excessive postpartum. Clinicians usually plan a taper back toward the pre-pregnancy dose around delivery.

Blood Thinners

Low-molecular-weight heparins like enoxaparin are commonly used during pregnancy to prevent or treat blood clots. Their pharmacokinetics shift significantly during gestation, driven largely by the increase in kidney clearance. A study comparing the same women during pregnancy and when non-pregnant found that enoxaparin behaved measurably differently in the pregnant state, with faster elimination and lower peak drug activity.⁷ Dose adjustments guided by blood-level monitoring were common throughout pregnancy, and the need for increased doses was especially apparent in women receiving preventive (rather than treatment) levels of anticoagulation.⁸

Antibiotics

Beta-lactam antibiotics, the family that includes penicillins and cephalosporins, are among the most frequently prescribed drugs in pregnancy. They are eliminated faster during gestation, resulting in lower blood concentrations than the same dose would produce outside of pregnancy.⁹ A more recent pharmacokinetic analysis confirmed that pregnant women in their second and third trimesters showed higher clearance and larger distribution volumes than non-pregnant or postpartum women, leading to lower drug concentrations and overall exposure.¹⁰ Whether the standard dose still hits the concentration needed to kill the infecting bacteria is a real clinical question, and for some infections it may not.

Drug-Drug Interactions on Top of Pregnancy Changes

Pregnancy-related changes to enzyme activity do not exist in a vacuum. When a pregnant person takes multiple medications, the drug-drug interactions that would occur in anyone can be amplified, diminished, or redirected by the underlying pregnancy physiology. This creates a layered problem that is genuinely difficult to predict from first principles.

A concrete example comes from modeling work on UGT1A1, an enzyme involved in processing several drugs. Pregnancy naturally induces UGT1A1 activity. When simulations added rifampicin, a powerful enzyme inducer used to treat tuberculosis, the pregnancy-related induction and the drug-related induction compounded each other, accelerating drug clearance even further. Conversely, when the HIV drug atazanavir, which inhibits UGT1A1, was modeled in a pregnancy context, the pregnancy-driven induction partially cancelled out atazanavir’s inhibitory effect, potentially leaving the patient with less enzyme inhibition than a non-pregnant person taking the same combination.¹¹ The direction of the interaction depends on whether pregnancy and the co-medication are pushing the same enzyme in the same direction or opposite directions.

This kind of interplay is one reason why simply looking up a drug interaction in a standard reference and applying it to a pregnant patient can be misleading. The magnitude of the interaction may be larger or smaller than what was measured in non-pregnant adults, and in some cases the direction can effectively reverse.

The Placenta Is Not a Simple Wall

A common misconception is that the placenta acts as a straightforward barrier keeping drugs away from the fetus. In reality, it is more like a selectively permeable checkpoint. Most small-molecule drugs do cross the placenta to some degree, and the extent of crossing depends on the drug’s physical and chemical properties as well as on active transport systems built into the placental tissue.

The placenta expresses a range of transporters, including efflux pumps that actively push certain drugs back toward the maternal circulation and uptake transporters that pull others toward the fetal side.¹² It also contains its own set of drug-metabolizing enzymes, including CYP enzymes, whose levels fluctuate throughout gestation.¹³ The placenta can therefore partially metabolize a drug before it ever reaches the fetus, adding yet another variable to predicting fetal exposure.

Once a drug does reach the fetal side, there is a further wrinkle: it can accumulate in the amniotic fluid through fetal kidney excretion, then be swallowed by the fetus and reabsorbed. This recycling loop means that some drugs persist in the fetal compartment longer than their half-life in the mother would suggest.¹⁴ The fetus’s own liver enzymes are immature and process drugs more slowly, so clearance on the fetal side is limited.

Prenatal Vitamins and Supplement Interactions

Drug interactions in pregnancy are not limited to prescription-to-prescription combinations. Prenatal vitamins, which nearly all pregnant people take, can themselves interfere with drug absorption. A well-documented example involves iron supplements: when iron is combined in a prenatal formula with calcium carbonate and magnesium oxide, iron absorption drops because calcium and magnesium inhibit it.¹⁵ The same mineral interactions can affect the absorption of thyroid medications, certain antibiotics, and other drugs taken around the same time as a prenatal vitamin. Spacing doses apart by a couple of hours is a simple but often-overlooked solution.

Why Dosing Data Is So Limited

One of the most frustrating aspects of pharmacokinetics in pregnancy is how little high-quality evidence exists. Pregnant individuals have been systematically excluded from clinical drug trials for decades, largely out of ethical concern for fetal safety. The result is that for most medications, the dosing recommendations used in pregnancy are extrapolated from studies in non-pregnant adults, sometimes with small pharmacokinetic studies in pregnant volunteers tacked on after the drug has been on the market for years.¹⁶

To fill this gap, researchers have increasingly turned to computer-based modeling. Physiologically based pharmacokinetic (PBPK) models simulate what happens to a drug in the body by incorporating known physiological changes of pregnancy, such as blood volume expansion, altered enzyme activity, and increased kidney filtration, into a mathematical framework. These models can predict how drug concentrations will change across trimesters and can be validated against the limited real-world data that does exist. Recent work has extended this approach to predict not just maternal drug levels but also fetal exposure, including for biologic therapies like therapeutic antibodies.¹⁷ PBPK models have also been used to simulate what happens when drug-drug interactions and pregnancy-related changes overlap, providing mechanistic insight that would be unethical or impractical to study in a traditional trial.¹¹

These models are promising but imperfect. They depend on the accuracy of the physiological parameters fed into them, and for some pregnancy-specific changes the data are sparse. They are best understood as tools that help researchers design smarter, smaller clinical studies rather than as replacements for real human data.

Genetic Variation Adds Another Layer

Even outside of pregnancy, two people taking the same drug at the same dose can end up with very different blood levels depending on their genetic makeup. Variants in the genes encoding drug-metabolizing enzymes can make someone an ultra-rapid metabolizer who burns through a medication quickly, or a poor metabolizer who clears it very slowly. Pregnancy amplifies whatever baseline metabolic profile a person has. If you are already a fast metabolizer of a drug cleared by CYP2D6, and pregnancy further boosts CYP2D6 activity, the combined effect on drug clearance can be dramatic.

The interplay between genetic variation and pregnancy-related enzyme changes is an area of active research. Investigators have shown that pharmacogenomic variability contributes to the wide range of drug responses seen in pregnancy, and that standard dosing adjustments cannot account for the full spectrum of individual variation. More targeted approaches that incorporate both genetic testing and therapeutic drug monitoring are under investigation, but for now this kind of personalized dosing in pregnancy remains more of a research aspiration than routine clinical practice.

Drug Transfer into Breast Milk

The pharmacokinetic story does not end at delivery. For people who breastfeed, the question shifts from placental transfer to milk transfer. Most drugs enter breast milk primarily through passive diffusion from maternal blood, driven by the drug’s size, fat solubility, protein binding, and electrical charge.¹⁸ Some drugs also enter milk through active transport mechanisms. The physicochemical properties of the drug largely determine how much crosses over, and measuring drug levels in milk directly is considered the gold standard for assessing infant exposure.¹⁹

For most medications, the dose an infant receives through breast milk is quite small relative to the mother’s dose. The general benchmark used in lactation pharmacology is whether the infant’s exposure exceeds 10 percent of the weight-adjusted maternal dose. Most drugs fall well below that threshold. However, drugs with low clearance in the infant are a particular concern, because even a small amount delivered through milk can accumulate over time in a newborn whose liver and kidneys are still maturing. Opioids are the most frequently cited example: pharmacokinetic principles predict that drugs with very low clearance are likely to exceed the 10 percent threshold, and the potential for severe effects in the infant means extra caution is warranted. Individual variation in both the mother’s milk production and the infant’s capacity to metabolize the drug makes the risk hard to predict from general tables alone.²⁰

The postpartum period also brings its own transitional pharmacokinetics for the mother. The pregnancy-driven increases in enzyme activity and kidney filtration do not snap back to baseline immediately after delivery. For drugs like lamotrigine, where the dose was escalated during pregnancy, failure to taper promptly after birth can lead to toxicity as clearance slows. Coordinating the mother’s dose adjustments with the shift from fetal to neonatal exposure via milk is one of the more underappreciated challenges in perinatal medicine.