Paroxetine

Pregnancy

MotherToBaby reports mixed evidence about first-trimester paroxetine and birth defects. Bérard et al. 2016 combined 23 studies and reported higher odds of any major congenital malformation (pooled OR 1.23, 95% CI 1.10-1.38) and major cardiac malformation (OR 1.28, 95% CI 1.11-1.47). Huybrechts et al. 2014 studied 949,504 Medicaid pregnancies; after adjustment among people with depression, the SSRI cardiac-defect estimate was RR 1.06 (95% CI 0.93-1.22), and paroxetine was not significantly linked with right ventricular outflow tract obstruction (RR 1.07, 95% CI 0.59-1.93). The current PAXIL label retains a less-than-twofold heart-malformation caution and says other options should be considered before starting for someone planning pregnancy or in trimester one. These observational results do not establish that paroxetine caused the outcomes. For miscarriage, MotherToBaby says some studies suggest a small increase but that untreated depression or other factors may also matter. Nakhai-Pour et al. 2010 compared 5,124 clinically detected miscarriage cases with 51,240 matched controls and reported an association for paroxetine (OR 1.75, 95% CI 1.31-2.34). This single prescription-based case-control study does not prove a cause. For preterm birth, MotherToBaby says some studies suggest a link with paroxetine throughout pregnancy. Martin et al. 2026 analyzed 2,528,916 deliveries across the UK, Norway, and Sweden. Antidepressant use overall was associated with preterm birth, while the article describes the paroxetine estimate as imprecise. The exact paroxetine-specific estimate was not verified from the abstract and full-text main article, so it is omitted. For PPHN, Huybrechts et al. 2015 studied 3,789,330 pregnancies. The adjusted estimate for all PPHN after restricting to people with depression was OR 1.10 (95% CI 0.94-1.29); a restricted primary-PPHN analysis was OR 1.28 (95% CI 1.01-1.64). These are SSRI-class estimates, not paroxetine-specific results. The PAXIL label and MotherToBaby describe short-term newborn adaptation effects after later exposure. Cornet et al. 2024 studied 280,090 term infants and reported delayed neonatal adaptation in 11.2% of late-SSRI-exposed infants and 4.4% of unexposed infants, with adjusted OR 2.14 (95% CI 1.96-2.32); the abstract does not give a paroxetine-only estimate. Aref et al. 2026 used target-trial methods in one academic center. In its periconception and early-pregnancy cohorts, continued SSRI use was not significantly associated with congenital anomalies or NICU admission, but the study was not paroxetine-specific and had limited sample size. Untreated depression and other factors can also affect pregnancy outcomes, as MotherToBaby, ACOG, SMFM, and the 2026 Martin study note.

Breastfeeding

LactMed reports low milk levels and no detectable paroxetine in the blood of most tested infants. Den Besten-Bertholee et al. 2024 studied 37 lactating women and infants; 8 of 9 infants exposed to paroxetine through milk had no detectable plasma level. This was a small pharmacokinetic study. LactMed lists occasional reports of sedation, agitation, irritability, poor feeding, and poor weight gain, and advises monitoring for these signs. Uguz and Arpaci 2016 studied paroxetine and sertraline together and reported a 12.5% adverse-event rate across the combined group; it does not provide a paroxetine-only rate. LactMed says no developmental effects have been found in infants followed to age five. Heinonen et al. 2025 examined IQ in 97 mother-child dyads, including 22 children with SSRI exposure through breastfeeding. The study found no lower adjusted IQ scores in the postnatal-SSRI group, but it was class-wide, did not measure all areas of development, and does not establish paroxetine-specific long-term effects. LactMed's preferred-antidepressant statement is attributed to its review of several sources, not presented as a universal recommendation.

What was tested

Pregnancy evidence is mainly from observational cohorts, case-control studies, and meta-analyses. The reviewed sources did not identify a randomized trial testing paroxetine pregnancy outcomes. Important studies include a 23-study first-trimester meta-analysis, a 949,504-pregnancy cardiac-defect cohort, a 3,789,330-pregnancy PPHN cohort, and the 2026 three-country birth-outcome cohort. Breastfeeding evidence includes small milk and infant-plasma studies, infant follow-up cohorts, and a 2025 SSRI IQ cohort of 97 dyads. Limits include prescription records that cannot confirm doses taken, confounding by the reason for treatment, mixed-drug groups, and class-wide results that do not give a paroxetine-specific estimate.