Selective Serotonin Reuptake Inhibitors in Human Pregnancy: To Treat or Not to Treat?

Table 1

SSRIs in human pregnancy.

Study	Design	Sample size	SSRI	Results	Comments
Pastuszak et al., [10]	Prospective comparative multicentre cohort study	𝑛=128	Fluoxetine	No increase in the rate of major malformations	Small numbers
Chambers et al., [11]	Prospective comparative cohort study	𝑛=228,   𝑛=101 (with physical examination)	Fluoxetine	No increase in the risk of major anomalies, higher incidence of 3 or more minor anomalies 15.5% versus 6.5%,   𝑃=0.03	Physical examination by a single dysmorphologist
McElhatton et al., [12]	Prospective comparative collaborative ENTIS study	𝑛=689  antidepressants	Fluoxetine 𝑛=96 Fluvoxamine 𝑛=66 Paroxetine 𝑛=3	No increase in the rate of congenital anomalies	Small numbers
Goldstein et al., [13]	Prospective registry, historic controls	𝑛=769	Fluoxetine	No increase in the rate of congenital anomalies	Manufacturer’s data, spontaneous reports
Wilton et al., [14]	Postmarketing survey	SSRIs 187	Paroxetine 𝑛=63 Fluoxetine 𝑛=52 Sertraline 𝑛=51 Fluvoxamine 𝑛=21	No increase in the rate of congenital anomalies	Small numbers
Kulin et al., [15]	Prospective comparative multicentre cohort study	“New” SSRIs 267	Sertraline 𝑛=147 Paroxetine 𝑛=97; Fluvoxamine 𝑛=26	No increase in the risk of major congenital anomalies
Ericson et al.,^a [16]	Swedish Medical Birth Registry, initial report	SSRIs: 𝑛=531	Citalopram 𝑛=375 Paroxetine 𝑛=122 Sertraline 𝑛=34 Fluoxetine 𝑛=16	No increase in the rate of congenital anomalies	Incomplete drug reporting
Unfred et al., [47]	Prospective comparative cohort study	𝑛=101	Paroxetine	Increased risk of congenital anomalies (4/96 (4.2%) 1/195 (0.5%)𝑃=0.04) no pattern	Rate of anomalies in comparison group low, is, unpublished data
Simon et al., [17]	Retrospective cohort	SSRIs: 𝑛=185	Fluoxetine 𝑛=129 Sertraline 𝑛=32 Paroxetine 𝑛=28; >1 SSRI: some	No increase in the rate of congenital anomalies; however, the rate of cardiac malformations was 2.2% in the exposed group versus 0% in unexposed	Prescription study, reliance on routinely collected clinical data, sample of live births rather than pregnancies, large number of comparisons
Hendrick et al., [18]	Review of obstetric and neonatal records	SSRIs: 𝑛=138	Fluoxetine 𝑛=73 Sertraline 𝑛=36 Paroxetine 𝑛=19; Citalopram 𝑛=7; Fluvoxamine 𝑛=3	No increase in the rate of congenital anomalies	Uncontrolled design, small numbers
Malm et al.,^c [19]	Population-based cohort study, Finnish registries	SSRI: 𝑛=1398	Citalopram 𝑛=554 Fluoxetine 𝑛=525 Paroxetine 𝑛=152 Sertraline 𝑛=118 Fluvoxamine 𝑛=65	No increase in the rate of congenital anomalies	Prescription study, data on dose not provided
Sivojelezova et al., [20]	Prospective comparative cohort study	𝑛=125	Citalopram	No increase in the rate of major malformations
Wogelius et al.,^b [29]	Population-based cohort study, Danish registries	SSRIs 𝑛=1051	NA	Increased risk for overall anomalies (Ad RR 1.34 (95% CI 1.00–1.79) early, 1.84 (95% CI 1.25–2.71 2nd-3rd m) cardiovascular 29%	Data on specific SSRIs not provided, prescription study
Wen et al., [21]	Retrospective cohort study	SSRIs: 𝑛=972	Paroxetine by 1/3	No increase in the risk of birth defects	Prescription study
Schloemp et al., [22]	Prospective comparative cohort	𝑛=119	Paroxetine	No increase in the risk of birth defects	Unpublished data
Vial et al., [23]	Prospective comparative cohort	𝑛=683	Paroxetine	No increase in the risk of birth defects	Unpublished data
Källén and Otterblad-Olausson,^a   [30], ^a [31]	Swedish Medical Birth Register, updated	SSRIs: 𝑛=6555	Fluoxetine 𝑛=926; Citalopram 𝑛=2701; Paroxetine 𝑛=959; Sertraline 𝑛=1906; Fluvoxamine 𝑛=38; Escitalopram 𝑛=72	Increased risk of cardiovascular defects with paroxetine OR 1.63 95% CI 1.05–2.53 mostly septal defects 13/20	Incomplete drug reporting, potential detection bias, multiple comparisons
Davis et al., [24]	Retrospective case control study	SSRIs: 𝑛=805	Paroxetine 𝑛=182	No increase in the risk of birth defects	Prescription study
Bérard et al., [48]	Retrospective nested case-control study	𝑛=1403 antidepressants	Paroxetine 𝑛=542; n (>25 mg/d) = 143; other SSRIs 𝑛=443	No increase in the rate of congenital anomalies, increased risk for overall and cardiac malformations in the high-dose (>25 mg/d) group (Ad OR 3.07 (95% CI 1.00–9.42))	Prescription study, calculation of average daily dose affected by duration in the first trimester
Alwan et al., [39]	Retrospective case-control study	Cases:𝑛=9622; Control:s 𝑛=4092	SSRIs 2.4% of cases 2.1% of controls, (sertraline 0.8%, fluoxetine 0.7%, paroxetine 0.5% citalopram 0.2%)	Associations between any SSRI and craniosynostosis, paroxetine/sertraline and anencephaly, paroxetine and right ventricular outflow tract obstruction defects, omphalocele and gastroschisis—small absolute risks	Small numbers of exposed infants for individual anomalies, multiple comparisons, data on dosage unavailable, potential recall and selection biases
GSK/Cole; et al., [49, 50]	Retrospective epidemiologic study	𝑛=4936; antidepressants	Paroxetine 𝑛=1020	Increased risk for overall congenital anomalies for paroxetine (Ad OR 1.89 (95% CI 1.20–2.98))	Manufacturers’ data from a large US insurer using data originally for a study on bupropion in pregnancy
Louik et al., [40]	Retrospective case-control study	9849 infants with defects and 5860 without	SSRIs: 2.7% of infants without malformations, 1.6–4.8% of infants with various malformations	Association between paroxetine and right ventricular outflow tract obstruction defects, clubfoot, undescended testes and NTDs, sertraline and omphalocele and septal defects—small absolute risks	Small number of exposed infants for individual anomalies, multiple comparisons, data on dosage unavailable, potential recall and selection biases
Einarson et al., [25]	Prospective comparative cohort study 8 TISes	𝑛=1174	Paroxetine	No increase in the risk of cardiac defects in the paroxetine group (0.7%) compared to controls	Spontaneously resolved cardiovascular defects not included
Diav-Citrin et al., [32]	Prospective comparative cohort study 3 TISes	SSRI: 𝑛=809; 1467 controls	Paroxetine 𝑛=348 Fluoxetine 𝑛=253	Increased Cr OR for cardiovascular anomalies after paroxetine 3.47 (95% CI 1.13–10.58) Ad OR 2.66 (95% CI 0.80–8.90), Cr OR 4.81 (95% CI 1.56–14.71) Ad OR 4.47 (95% CI 1.31–15.27) after fluoxetine	After adjustment for potential confounders OR significant only for fluoxetine and cigarette smoking of 10 or more/day, septal defects considered major anomalies even when spontaneously closed, large confidence intervals
Oberlander et al., [33]	Population-based cohort study, Canada, BC	SRIs: 𝑛=2625; SRIs+BZ: 𝑛=968; Controls: 𝑛=107,320	SRIs: 𝑛=2625; SRIs+BZ: 𝑛=968; controls: 𝑛=107,320	Increased risk of cardiovascular (CV) defects after combined exposure to SRI and BZ, increased risk for an ASD after SRI monotherapy, major anomalies after fluoxetine and BZ	Clinical significance of the anomaly not verified, many septal defects minor and spontaneously resolve, attempt to control for confounders
Pedersen et al.,^b [34]	Population-based cohort study, Danish registries	SSRIs: 𝑛=1370; 493,113 controls	Fluoxetine 𝑛=348 Citalopram 𝑛=460 Paroxetine 𝑛=299 Sertraline 𝑛=259; >1 SSRI 𝑛=193	Increased prevalence of septal defects with sertraline (OR 3.25 (95% CI 1.21–8.75)) and citalopram (OR 2.52 (95% CI 1.04–6.10))	Prescription study, potential selection bias, underlying condition potential confounder, information on malformation from hospital registry (more sensitive to severe and visible malformations)
Wichman et al., [26]	Retrospective controlled review of medical records at the Mayo Clinic	SSRIs: 𝑛=808; 24,406 controls	Citalopram 𝑛=122 Venlafaxine 𝑛=53 Escitalopram 𝑛=8 Paroxetine 𝑛=134 Fluoxetine 𝑛=184 Sertraline 𝑛=296; >1 SSRI 𝑛=11	3/808 (0.4%) had congenital heart disease after exposure to SSRIs compared with 2.05/24,406 (0.8%) without exposure to SSRIs (𝑃=0.23)	No review of SSRI exposure timing, of demographic or clinical information including use of other drugs, smoking, or alcohol use, data from physician prescription records. Small VSDs may be undetected soon after birth.
Merlob et al., [35]	Prospective comparative hospital-based study	SSRIs: 𝑛=235; 67,636 controls	Paroxetine 𝑛=92 Fluoxetine 𝑛=66 Citalopram 𝑛=43 Escitalopram 𝑛=13 Sertraline 𝑛=8 Fluvoxamine 𝑛=4 Venlafaxine 𝑛=9	Non syndromic heart defects (mild) identified by echocardiography among infants with murmur on 2nd-3rd day of life RR 2.17 (95% CI 1.07–4.39)	Small sample size of exposed group, lack of data on potential confounders, ascertainment of SSRI use based on maternal report, no detection bias; all newborns with murmur examined by a pediatric cardiologist including echocardiography.
Klieger-Grossmann; et al., [27]	Prospective comparative cohort study	𝑛=213	Escitalopram	No increased risk for anomalies	Unpublished data
Kornum et al.,^b [36]	Population-based cohort study, Danish registries, updated	SSRIs: 𝑛=2064; 213,712 controls	Citalopram 𝑛=658 Fluoxetine 𝑛=472 Sertraline 𝑛=352 Paroxetine 𝑛=297 Escitalopram 𝑛=88; >1 SSRI 𝑛=195	SSRI use associated with increased risk of overall malformations (Ad OR 1.3 (95% CI 1.1–1.6)) and cardiac (Ad OR 1.7 (95% CI 1.1–2.5)), for specific SSRIs increased risk for septal defects with sertraline (Ad OR 3.3 (95% CI 1.5–7.5))	Prescription study, potential selection bias, underlying condition potential confounder, information on malformation from hospital registry (more sensitive to severe and visible malformations)
Bakker et al., [37]	Population-based case- control study, the Netherlands	𝑛=678 with heart defects 615 controls	Paroxetine	No significantly increased risk for heart defects overall (AOR 1.5 (95% CI 0.5–4.0)) increased risk for ASD with paroxetine in T1 (AOR 5.7 (95% CI 1.4–23.7))	Small number of exposed infants for individual anomalies
Reis and Källén,^a   [38]	Swedish Medical Birth Register, updated	SSRIs: 𝑛=10,170 1,062,190 controls	Fluoxetine 𝑛=1522 Citalopram 𝑛=3950 Escitalopram 𝑛=153 Paroxetine 𝑛=1208 Sertraline 𝑛=3297 Fluvoxamine 𝑛=42 Unspecified 𝑛=86	Increased risk for cystic kidney (𝑛=9) with SSRIs (OR 2.39 (95% CI 1.09–4.54)), for relatively severe malformation (OR 1.29 (95% CI 1.00–1.67)) with fluoxetine, for any cardiovascular defect (OR 1.66 (95% CI 1.09–2.53) 12/24 septal defects) and for hypospadias (𝑛=9) (OR 2.45 (1.12–4.64)) with paroxetine	Prospective exposure information, largest dataset available. Incomplete drug reporting, potential detection bias, multiple comparisons, possible confounding by the underlying psychiatric condition, smoking, obesity, alcohol, folic acid, association with preexisting diabetes and hypertension
Shechtman et al., [28]	Prospective comparative cohort study	𝑛=241	Citalopram or escitalopram	No increased risk for anomalies	Unpublished data
Malm et al.,^c    [41]	Retrospective cohort, based on Finnish population-based registries	SSRIs: 𝑛=6,976	Citalopram 𝑛=2,799 Fluoxetine 𝑛=1,818 Paroxetine 𝑛=968 Sertraline 𝑛=869 Escitalopram 𝑛=441 Fluvoxamine 𝑛=240	Increased risk for isolated VSDs with fluoxetine (𝑛=19) (Adj OR 2.03 (95% CI 1.28–3.21)) (0.5% absolute risk increase), for right ventricular outflow tract defects with paroxetine (𝑛=3) (Adj OR 4.68 (95% CI 1.48–14.74)) (0.2% absolute risk increase), for NTDs with citalopram (𝑛=8) (Adj OR 2.46 (95% CI 1.20–5.07))	Large dataset, attempt to control for confounders (i.e., maternal age, parity, year of pregnancy, marital status, smoking, purchase of other psychiatric drugs, maternal prepregnancy diabetes). Large number of comparisons, some associations based on small numbers, drug compliance and timing of exposure in pregnancy not confirmed, septal defects considered major anomalies even when spontaneously closed

^aFrom the same database of the Swedish Medical Birth Register, ^b from the same database of the Finnish registries, ^c from the same database of the Danish registries.