Mitchellsmithlaw.com

Selective Serotonin Reuptake Inhibitors and
Risk for Major Congenital Anomalies
Heli Malm, MD, PhD, Miia Artama, MSc, PhD, Mika Gissler, MSocSc, PhD, and Annukka Ritvanen, MD OBJECTIVE: To estimate the risk of major congenital
in the selective serotonin reuptake inhibitor-exposed
anomalies after exposure to selective serotonin reuptake
offspring than in unexposed referent offspring.
inhibitors during pregnancy.
CONCLUSION: Fluoxetine use is associated with an in-
METHODS: A retrospective cohort study based on
creased risk of isolated ventricular septal defects and
national population-based registers (years 1996 –2006)
paroxetine is associated with right ventricular outflow
of births, congenital anomalies, and terminations of
tract defects. The absolute risk for these specific cardiac
pregnancy because of severe fetal anomalies (main-
anomalies is small but should guide clinicians not to
tained by National Institute for Health and Welfare,
consider fluoxetine or paroxetine the first option when
source offspring population n؍635,583) and drug re-
prescribing selective serotonin reuptake inhibitors to
imbursements (Social Insurance Institution) linked by a
women planning pregnancy. Special attention should be
personal identification number. Offspring exposed to
given to alcohol use in pregnant women using selective
selective serotonin reuptake inhibitors during the first
serotonin reuptake inhibitors.
trimester (n؍6,976) were compared with unexposed
(Obstet Gynecol 2011;118:111–20) referent offspring.
RESULTS: Overall major congenital anomalies were not
LEVEL OF EVIDENCE: II
more common in selective serotonin reuptake inhibitor-
exposed offspring compared with unexposed referent
offspring (adjusted odds ratio [OR] 1.08, 95% confidence
As reported in a previous review,1 approximately interval [CI] 0.96 –1.22). Fluoxetine was associated with
10% of pregnant women experience depression, an increased risk of isolated ventricular septal defects
and up to 20% exhibit depressive symptoms.2 More (adjusted OR 2.03, 95% CI 1.28 –3.21) and paroxetine was
women have been using the new antidepressants, associated with an increased risk of right ventricular
including the selective serotonin reuptake inhibitors outflow tract defects (adjusted OR 4.68, 95% CI 1.48 –
(fluoxetine, citalopram, paroxetine, sertraline, fluvox- 14.74). Citalopram use was associated with neural tube
amine, and escitalopram); research shows up to defects (adjusted OR 2.46, 95% CI 1.20 –5.07). Fetal
3%– 6% of pregnant women are using these drugs.3 alcohol spectrum disorders were 10-times more common
Although selective serotonin reuptake inhibitors andtheir effects on pregnancy outcome have been more From the Teratology Information Service, Helsinki University Central Hospital widely investigated than many other drugs, conflict- and HUSLAB, Helsinki, Finland; the Department of Clinical Pharmacology,Helsinki University and Helsinki University Central Hospital, Helsinki, Fin- ing data for possible increased risk of major congen- land; the National Institute for Health and Welfare, Helsinki, Finland; and the ital anomalies have been published.4–26 Fluoxetine, Nordic School of Public Health, Go¨teborg, Sweden. citalopram, paroxetine, and sertraline exposure in Funded by the Social Insurance Institution in Finland, the Finnish Medicines early pregnancy have been associated with an in- Agency, and the National Institute for Health and Welfare. The study materialwas derived from the Drugs and Pregnancy project database. creased risk of cardiovascular anomalies; fluoxetine The authors thank the Drugs and Pregnancy study group for their input in with overall cardiovascular anomalies5; citalopram building and maintaining the combined database for surveillance and study with septal heart defects6; paroxetine with right ven- tricular outflow tract defects,7,8 atrial septal defects,9 Corresponding author: Heli Malm, Teratology Information Service, HUSLAB, and overall major cardiovascular anomalies10–12; and PO Box 790, 00029 HUS, Helsinki, Finland; e-mail heli.malm@hus.fi. sertraline with cardiac septal defects.6,8 An increased Financial Disclosure
risk for several other organ group anomalies or over- The authors did not report any potential conflicts of interest. all major congenital anomalies after exposure to 2011 by The American College of Obstetricians and Gynecologists. Publishedby Lippincott Williams & Wilkins. fluoxetine (overall major congenital anomalies), cita- lopram and paroxetine (pooled birth defects includ- OBSTETRICS & GYNECOLOGY
ing anencephaly, craniosynostosis, and omphalocele), The Register of Congenital Malformations in- paroxetine (overall major congenital anomalies and cludes information on live births and stillbirths and neural tube defects), and sertraline (omphalocele, anal fetuses from pregnancy terminations attributable to atresia, limb reduction defects) has also been ob- severe fetal anomaly, all with at least one detected served in meta-analyses or individual studies.7,8,10,11,13 major congenital anomaly, including major structural However, the results have been inconsistent and anomalies, chromosomal defects, and congenital hy- direct teratogenicity for any of the selective serotonin pothyroidism, classified and coded according to the reuptake inhibitors has not been established. On the ninth version of the World Health Organization basis of all current available information, fluoxetine International Classification of Diseases. Minor anom- and paroxetine are the individual selective serotonin alies are excluded principally according to the exclu- reuptake inhibitors seriously suspected to cause a sion list of the European Surveillance of Congenital marginally increased risk for congenital cardiovascu- Anomalies, EUROCAT.28 In case of severe fetal lar anomalies.1 We began a register-based study to major congenital anomalies or disease, permission for investigate associations between the use of selective termination of pregnancy may be granted on moth- serotonin reuptake inhibitors and major congenital er’s request by a National Board at the National anomalies, including organ group-specific anomalies Supervisory Authority for Welfare and Health (Val- and subgroups of cardiovascular anomalies using vira) until 24 completed gestational weeks, but not Finnish nationwide register data covering 11 years.
later. The register obtains information on all thesecases from Valvira and confirms data on the termina-tion attributable to severe fetal anomalies from hos- MATERIALS AND METHODS
pitals in charge. More than 10% of all cases of major The data for our study were derived from an congenital anomalies in the register are recorded ongoing national joint project, Drugs and Preg- from terminations attributable to severe fetal anoma- nancy, in Finland, established by three governmen- lies.29 The register receives and collects actively data tal organizations: the Finnish Medicines Agency; nationwide from several sources, including hospitals, the Social Insurance Institution in Finland; and the health care professionals, and cytogenetic laborato- National Institute for Health and Welfare for con- ries. It also obtains and confirms data from the tinuous surveillance of safe drug use during preg- Medical Birth Register, the Register of Induced Abor- nancy. The project collects information from three tions, the Hospital Discharge Register, and the Reg- national health registers: The Medical Birth Register ister of Visual Impairments, all nationwide registers and the Register of Congenital Malformations, both maintained by National Institute for Health and Wel- maintained by the National Institute for Health and fare, and from the Cause of Death Register, main- Welfare, and the Drug Reimbursement Register, tained by Statistics Finland. Information on fetal maintained by the Social Insurance Institution in anomalies diagnosed prenatally and confirmed after Finland. The unique personal identification number termination of pregnancy are included in the register.
assigned at birth to all Finnish citizens and permanent The validity of the register is considered good and has residents in Finland makes the linkage of all these been ascertained in several studies.29–31 The Drug Reimbursement Register contains data The Medical Birth Register collects data on ma- on 98% of reimbursed prescription drug purchases, ternal background and medical history, diagnoses including data on several chronic illnesses requiring during pregnancy and delivery, and neonatal out- continuous drug treatment.32 Prescription-only medi- come data up to age 7 days. Data in the register are cines deemed necessary for the treatment of an illness collected from all maternity hospitals and include all are reimbursed under the Social Insurance Scheme, births, including the occasional home births. All which covers all permanent residents in Finland. The neonates are examined at hospital by a pediatrician.
reimbursement system has three levels, including a All live births and stillbirths with gestational age of 22 basic refund category (covering all citizens) and two weeks or more or birth weight of 500 g or more are special reimbursement categories (granted to persons included in the register. The register data are con- with a diagnosed chronic disease such as severe psy- firmed and complemented from the maternity hospi- chotic and other severe mental disorders, diabetes, and tal records in cases of conflicting or missing informa- epilepsy, among others).32 Drug purchases are reim- tion. Data quality studies have shown that the bursed concomitantly on purchase at the pharmacy and majority of the register content corresponds well or drugs are supplied to the patient for 3 months at a time.
satisfactorily with hospital record data.27 The data in the register include the International Ana- Malm et al
SSRIs and Major Congenital Anomalies OBSTETRICS & GYNECOLOGY
tomic-Therapeutic-Chemical classification code and in- flow tract defects (including aortic valve atresia and formation on possible special reimbursement status in- stenosis). In the Register of Congenital Malforma- cluding indication for treatment. Over-the-counter drugs tions, patent foramen ovale and patent ductus arteri- or medications given to institutionalized persons are not osus are not considered as major anomalies, whereas ventricular septal defects are considered as major The study material was derived from the Drugs and congenital anomalies because no extensive follow-up Pregnancy project database and included the births and information is available for possible spontaneous clo- terminations attributable to severe fetal anomalies in the sure of these defects. Neonates exposed to selective years 1996–2006. In that project, beginning of preg- serotonin reuptake inhibitors while in utero during nancy has been defined by subtracting the number of the third trimester are at an increased risk for experi- days corresponding to gestation length from date of encing neonatal adaptation problems, which may birth or termination attributable to severe fetal anomaly.
further lead to more active diagnostic procedures.
The best estimated length of gestation at delivery is When indicated, we performed a subanalysis for based on ultrasound examination or the last menstrual ventricular septal defects by excluding neonates need- period and is recorded in the Medical Birth Register.
ing treatment in the neonatal care unit. Chromosomal Data on drug purchases and special reimbursements defects are registered in the Register of Congenital because of chronic disease were obtained from years Malformations as major congenital anomalies, except 1995 to 2006 originally from the Drug Reimbursement for sex chromosomal numerical defects, and were Register and were collected during preconception, each included in the basic analyses in the whole study pregnancy trimester, and 3 months after pregnancy.
population. We hypothesized that exposure to selec- Pregnancy trimesters were defined as 0–12 weeks (first tive serotonin reuptake inhibitors or any individual trimester), 13–26 weeks (second trimester), and 27 selective serotonin reuptake inhibitor does not in- weeks onward (third trimester). A total of 635,583 crease the risk of congenital anomalies.
mother and child pairs, including all pregnancies ending The use of sensitive health register data in the in live birth, stillbirth, or termination attributable to joint project and for scientific research was approved severe fetal anomaly, were included in the database by the data protection authority. The register linkages have been conducted with the contract between the At least one purchase of one or more selective register-keeping organizations and with permission serotonin reuptake inhibitor drugs (Anatomic-Thera- from the register administrators. The study protocol peutic-Chemical classification code on fourth-level, was approved by the Institutional Ethical Review N06AB: selective serotonin reuptake inhibitor antide- Board at National Institute for Health and Welfare pressants) and individual selective serotonin reuptake and Social Insurance Institution in Finland, as well as inhibitor drugs during the period of 1 month before the Steering Committee of the Drugs and Pregnancy pregnancy and first trimester was considered as an project. The study participants were not contacted indicator for first trimester exposure. The period of 1 and, according to the Finnish legislation, informed month before pregnancy was included to enable inclu- sion of prescriptions received immediately before preg- Statistical analyses were performed using SAS 9.1 nancy and potentially used during early pregnancy.
for Windows. Prevalence of overall major congenital We grouped major congenital anomalies on an anomalies and specific organ group major congenital organ group level, primarily according to the anomalies in offspring exposed to selective serotonin EUROCAT subgroups.28 Cardiovascular anomalies reuptake inhibitors or individual selective serotonin were further grouped as follows: atrial septal defects; reuptake inhibitor drugs in the first trimester was ventricular septal defects (excluding ventricular septal compared with corresponding prevalence in offspring defects in tetralogy of Fallot); right ventricular outflow not exposed to selective serotonin reuptake inhibitors tract defects (including pulmonary valve stenosis, or the individual selective serotonin reuptake inhibi- pulmonary valve atresia, and infundibular pulmonary tor drug (unexposed referent offspring) using ␹2 test or stenosis, and excluding tetralogy of Fallot); transposi- Fisher exact test (both two-tailed). Crude and adjusted tion off great arteries; conotruncal heart defects (in- logistic regression analyses were performed with cluding tetralogy of Fallot, pulmonary artery atresia overall major congenital anomalies and specific organ with ventricular septal defects, double outlet right group major congenital anomalies as dependent out- ventricle, persistent or common truncus arteriosus, come variables. Independent variables considered in aortic septal defect including aortopulmonary win- the adjusted logistic model were maternal age at the dow, and Fallot pentalogy); and left ventricular out- end of pregnancy (younger than 20 years or 35 years Malm et al
SSRIs and Major Congenital Anomalies or older compared with 20 –34 years; 0% missing nancy period. Of the total number of offspring information in the database), parity (no previous (nϭ635,583), 6,976 (1.1%) were exposed to selective deliveries compared with one or more previous de- serotonin reuptake inhibitors during the first trimester, liveries; 0.2% missing information), year of pregnancy including 6,902 (98.9%) live births, 30 (0.4%) stillbirths, ending (0% missing information), marital status (4.9% and 44 (0.6%) fetuses from termination attributable to missing information), smoking during pregnancy severe fetal anomalies compared with 623,402 (99.2%) (3.1% missing information), purchase of other reim- live births, 2,295 (0.4%) still births, and 2,910 (0.5%) bursed psychiatric drugs (including antiepileptics) fetuses from termination attributable to severe fetal during the first trimester, and maternal prepregnancy anomalies in unexposed referent offspring.
diabetes (0.3% missing information both). Missing Overall, major congenital anomalies were more information for categorized variables was categorized common among the offspring exposed to any selec- as the harmless alternative (ie, not exposed to to- tive serotonin reuptake inhibitor, but this difference bacco). All variables were entered simultaneously in did not remain statistically significant after adjusting the model. We did not include body mass index to confounders (Table 2). Variables independently because this information was available starting only associated with major congenital anomalies in the from the year 2004 and for only 23.7% of all patients.
logistic model were maternal age (OR 1.30, 95% CI Crude odds ratios (ORs) and adjusted odds ratios 1.26 –1.34), year of pregnancy ending (2006 com- (ORs) and 95% confidence intervals (CIs) were calcu- pared with 1996, OR 1.31, 95% CI 1.23–1.39), ma- lated. Statistical significance was set at PϽ.05.
ternal diabetes (OR 2.74, 95% CI 2.42–3.10), andpurchases of other psychiatric drugs (OR 1.53, 95% CI 1.39 –1.68). Considering only “other psychiatric The characteristics of pregnant women purchasing drugs” in the logistic model, the association between selective serotonin reuptake inhibitor drugs during selective serotonin reuptake inhibitors and major the first trimester or 1 month before pregnancy congenital anomalies became statistically insignificant (nϭ6,881) are presented in Table 1. Women with (adjusted OR 1.11, 95% CI 0.98 –1.25). When catego- selective serotonin reuptake inhibitor purchases were rizing polytherapy to consist of first trimester use of less likely to be married, twice as likely to smoke or to both selective serotonin reuptake inhibitors and other be entitled to special reimbursement because of psychiatric drugs as the exposure, and comparing to chronic disease, and 20-times more likely to have unexposed (also including selective serotonin re- purchased other psychiatric medications than women uptake inhibitor or other psychiatric drugs used either who had not made any selective serotonin reuptake alone), the exposure was statistically significantly as- inhibitor purchases during the corresponding preg- sociated with overall major congenital anomalies (ad- Table 1. Characteristics of Women With One or More Purchases or No Purchases of Selective
Serotonin Reuptake Inhibitors During the First Trimester or 1 Month Before Pregnancy
Women With Purchases
Women With No Purchases of
of Selective Serotonin
Selective Serotonin
Pregnant Women
Reuptake Inhibitor Drugs
Reuptake Inhibitor Drugs
Entitled to special reimbursement because of chronic disease Severe psychotic and other severe mental disorders Any other psychiatric medication purchases†‡ Assisted reproduction, mainly in vitro fertilization or Data are mean (minimum–maximum) or n (%).
* Mean age at the end of pregnancy.
† Including antiepileptics, antipsychotics, anxiolytics, hypnotics and sedatives, tricyclic antidepressants, and other antidepressants.
‡ Drug purchases during the first trimester or 1 month before pregnancy.
Malm et al
SSRIs and Major Congenital Anomalies OBSTETRICS & GYNECOLOGY
Table 2. Overall Major Congenital Anomalies in Offspring Exposed to Selective Serotonin Reuptake Inhibitors
Overall Major Congenital Anomalies
No. Exposed
No. Unexposed
Adjusted OR
Exposed Offspring (n)
(Prevalence)
(Prevalence)
(95% CI)*
OR, odds ratio; CI, confidence interval.
* Logistic regression. Adjusted to maternal age at the end of pregnancy, parity, year of pregnancy ending, marital status, smoking any time during pregnancy, other reimbursed psychiatric drug purchases, and entitlement for special reimbursement for prepregnancydiabetes. “Exposed” are offspring of pregnant women with one or more selective serotonin reuptake inhibitors drug purchasesduring the period of 1 month before pregnancy until the 12 completed gestational weeks. Comparisons made with unexposedreferent offspring of pregnant women with no purchases of selective serotonin reuptake inhibitors or the individual selectiveserotonin reuptake inhibitor drug analyzed during the same study period. Included are all live births (nϭ630,304) stillbirths(nϭ2,325), and fetuses from pregnancies ending in pregnancy termination because of fetal anomaly (nϭ2,954).
justed OR 1.31, 95% CI 1.04 –1.64). Obesity (body treatment in the neonatal care unit from the analysis mass index 30 or higher) was not independently (PϽ.01; adjusted OR 2.47, 95% CI 1.50 – 4.07; Table associated with a statistically significant risk for major 3). Exposure to paroxetine was associated with an congenital anomalies (OR 1.07, 95% CI 0.99 –1.17) increased risk for right ventricular outflow tract de- when restricting the analyses to patients for whom fects but was based on only three cases (Pϭ.03; information on body mass index was available. After adjusted OR 4.68, 95% CI 1.48 –14.74; Table 3). No excluding all major chromosomal anomalies from statistically significant associations were observed be- the study material, the major congenital anomalies tween the isolated cases of transposition of great rate was 392 of 10,000 in the selective serotonin arteries, conotruncal heart defects, or left ventricular reuptake inhibitor-exposed cohort and 315 of 10,000 outflow tract defects and any of the individual selec- in the unexposed referent offspring and was essen- tive serotonin reuptake inhibitor drugs (Table 3).
tially identical (adjusted OR 1.08, 95% CI 0.96 –1.23) Neural tube defects were more common in the to that observed in the whole study. Of individual selective serotonin reuptake inhibitor-exposed cohort selective serotonin reuptake inhibitors, fluoxetine and (prevalence 22 of 10,000 compared with 9 of 10,000 paroxetine use was associated with an increased risk in unexposed fetuses), and eight fetuses among the for overall major congenital anomalies, but the risk offspring exposed to citalopram had a neural tube did not remain statistically significant after adjusting defect diagnosis (prevalence 29 of 10,000; PϽ.01; adjusted OR 2.46, 95% CI 1.20 –5.07; Table 4). Six of Major cardiovascular anomalies were more com- the fetuses had spina bifida (including five cases of mon in the selective serotonin reuptake inhibitor- lumbal, lumbosacral, or sacral meningomyelocele exposed offspring when compared with the unex- and one case of sacral lipomeningocele), and the posed referent offspring; however, after adjusting to remaining two had frontal encephalocele and total confounders, the risk did not remain statistically craniorachischisis. No difference was observed on significant (adjusted OR 1.09, 95% CI 0.90 –1.32; individual drug level between the exposed and unex- Table 3). Further, considering only other psychiatric posed offspring in the prevalence of other organ drugs in the logistic model, the association turned insignificant (adjusted OR 1.17, 95% CI 0.96 –1.41).
Among the offspring exposed to any selective For individual selective serotonin reuptake inhibitors, serotonin reuptake inhibitor, eight8 had fetal alcohol fluoxetine exposure was associated with an increased spectrum disorders diagnosed, with a prevalence of risk for overall cardiovascular anomalies (adjusted 11.5 of 10,000 compared with 75 among the unex- OR 1.40, 95% CI 1.01–1.95) and isolated ventricular posed referent offspring (prevalence 1.2 of 10,000; septal defects, even when excluding offspring needing Malm et al
SSRIs and Major Congenital Anomalies Table 3. Prevalence of Major Cardiovascular Anomalies in Offspring of Pregnant Women Exposed to
Selective Serotonin Reuptake Inhibitors
Major Cardiovascular Anomalies
No. Exposed
No. Unexposed
Adjusted OR
Exposed Offspring (n)
(Prevalence)
(Prevalence)
(95% CI)*
Left ventricular outflow tract defects࿣ Malm et al
SSRIs and Major Congenital Anomalies OBSTETRICS & GYNECOLOGY
Table 3. Prevalence of Major Cardiovascular Anomalies in Offspring of Pregnant Women Exposed to
Selective Serotonin Reuptake Inhibitors (continued)
Major Cardiovascular Anomalies
No. Exposed
No. Unexposed
Adjusted OR
Exposed Offspring (n)
(Prevalence)
(Prevalence)
(95% CI)*
Escitalopram (441)All major cardiovascular anomalies —, no cases; OR, odds ratio; CI, confidence interval.
The total study population includes 635,583 births.
Isolated ventricular septal defect: including only offspring with isolated ventricular septal defect as the only recorded major congenital “Exposed” are offspring of pregnant women with one or more selective serotonin reuptake inhibitor drug purchases during the period of 1 month before pregnancy until 12 completed gestational weeks. Comparisons made with unexposed referent offspring ofpregnant women with no purchases of selective serotonin reuptake inhibitors or the individual selective serotonin reuptake inhibitordrug analyzed during the same study period.
* Adjusted to maternal age at the end of pregnancy, parity, year of pregnancy ending, marital status, smoking any time during pregnancy, other reimbursed psychiatric medicine purchases, and entitlement for special reimbursement for prepregnancy diabetes.
† Tetralogy of Fallot excluded from analyses.
‡ Including pulmonary valve stenosis, pulmonary valve atresia, and infundibular pulmonary stenosis.
§ Including tetralogy of Fallot, pulmonary artery atresia with ventricular septal defect, double outlet right ventricle, persistent or common truncus arteriosus, aortic septal defect including aortopulmonary window, and Fallot pentalogy.
࿣ Including aortic valve atresia and stenosis.
DISCUSSION
to paroxetine (adjusted OR 4.68, 95% CI 1.48 –14.74).
We found an increased risk of isolated ventricular Citalopram use was associated with neural tube de- septal defects after exposure to fluoxetine, even when fects (adjusted OR 2.46, 95% CI 1.20 –5.07), and we excluding neonates needing neonatal care unit treat- observed a 10-fold increase of fetal alcohol spectrum ment from the analysis (adjusted OR 2.47, 95% CI disorders in offspring of mothers using selective sero- 1.50 – 4.07). We also observed an increased risk for tonin reuptake inhibitor drugs compared with unex- right ventricular outflow tract defects after exposure Malm et al
SSRIs and Major Congenital Anomalies Table 4. Adjusted Odds Ratios and 95% Confidence Intervals for Organ System-Specific and Some
Previously Reported Major Congenital Anomalies in Relation to Selective Serotonin Reuptake
Inhibitor Use
Any Selective
Serotonin Reuptake
Organ System
Inhibitor
Citalopram
Fluoxetine
Paroxetine
Sertraline
Escitalopram
Fluvoxamine
—, no cases.
Data are adjusted odds ratio (95% confidence interval).
Logistic regression analyses adjusted to maternal age at the end of pregnancy, parity, year of pregnancy ending, marital status, smoking any time during pregnancy, other reimbursed psychiatric medicine purchases, and entitlement for special reimbursement forprepregnancy diabetes. “Exposed” are offspring of pregnant women with one or more selective serotonin reuptake inhibitor drugpurchases during the period of 1 month before pregnancy until 12 completed gestational weeks. Comparisons made withunexposed referent offspring of pregnant women with no purchases of selective serotonin reuptake inhibitors or the individualselective serotonin reuptake inhibitor drug analyzed during the same study period.
* Including anencephaly, encephalocele, and spina bifida.
† Also including limb defects.
Our study has several strengths. The register- lies, with the percentage having remained stable based approach enabled us to conduct a population- during the past decade.29 The prevalence of termina- based study with numbers large enough to allow tion attributable to severe fetal anomalies in the study analyses of organ-system specific major congenital material was 47 of 10,000 births and comparable to anomalies on individual selective serotonin reuptake numbers reported from other European countries.33 inhibitor level, focusing on first trimester exposure.
Although experimental animal studies have not The quality of the registers included in the study is demonstrated a well-defined increased teratogenic high, with the coverage of the Medical Birth Register risk for any of the individual selective serotonin and Drug Reimbursement Register being close to reuptake inhibitors, in vitro studies have demon- 100%.27,32 Linkage errors are highly unlikely because strated perturbations in neural crest cell migration of the unique identification number that is assigned to and heart cell differentiation and proliferation from all Finnish citizens and permanent residents in Fin- paroxetine and fluoxetine, possibly referring to car- land. No validity studies have been performed for the diac developmental toxicity.34–36 Two previously pub- Register of Congenital Malformations regarding car- lished studies have suggested an association between diovascular malformations, but because of the nation- fluoxetine and major cardiovascular anomalies,5,14 wide collection of data from multiple health care one of them with partially overlapping material with sources and national registers, the quality of the the present study and covering the years 1996 –2001 register is considered good.29 In Finland, operative but presenting no detailed analyses of organ group- treatment of all severe cardiovascular anomalies is specific malformations.14 Neonatal adaptation prob- centralized in the Helsinki University Central Hospi- lems are frequently encountered in neonates exposed tal Children’s Hospital, which actively notifies all to selective serotonin reuptake inhibitors in utero and these cases to the register. Including all terminations may predispose to information bias related to diag- of pregnancy attributable to severe fetal anomaly is an nostic activity.12,37 We found a risk for isolated ven- additional strength. Of all cases of major congenital tricular septal defects after fluoxetine exposure when anomalies in the register, 10%–12% are recorded neonates needing treatment in the neonatal care unit from termination attributable to severe fetal anoma- were excluded from the analysis, suggesting a true Malm et al
SSRIs and Major Congenital Anomalies OBSTETRICS & GYNECOLOGY
association between fluoxetine and ventricular septal tion. Despite this, ventricular septal defects are mostly defects. The observed higher rate of right ventricular classified as major defects because of frequently short outflow tract defects in paroxetine-exposed offspring follow-up periods in cohort studies, which leave the was based on only three cases with a wide CI long-term outcome unknown, or because there is no (adjusted OR 4.68, 95% CI 1.48 –14.74). However, long-term follow-up in birth register data.1 the parallel association observed in two previous One limitation is that drug compliance and tim- studies7,8 gives additional strength to our findings.
ing of exposure cannot be confirmed in a register- We found a statistically significant association based setting. However, previous studies have shown between citalopram and neural tube defects, with a that compliance with drugs used for treating chronic prevalence more than three-times higher in the ex- illnesses is high during pregnancy.39 If some women posed than in the unexposed cohort (29 of 10,000 did not use their selective serotonin reuptake inhibitor compared with 9 of 10,000; PϽ.01). Two previous drugs, even if obtaining the prescription from the studies with a case-control design have observed pharmacy, then this would bias the estimated OR unconfirmed associations between selective serotonin toward unity. Selective serotonin reuptake inhibitor reuptake inhibitors and neural tube defects or anen- drug treatment is usually discontinued step by step cephaly.7,8 Our findings need to be confirmed in during several weeks with tapering off the dose be- future studies. Because of the large number of com- cause abrupt cessation of therapy may give cause parisons performed in our analyses, it is possible that withdrawal effects. Therefore, even those women who some of the observed associations reflect variation by decided to discontinue the medication when discov- chance. Some of the associations were also based on ering that they were pregnant are likely to have had small numbers and could reflect random associations.
their fetuses exposed during embryogenesis.
The main findings, however, produced highly signif- We conclude that exposure to fluoxetine and icant P values, giving reinforcement to our findings.
paroxetine in early pregnancy is associated with a We observed a nearly 10-fold higher prevalence small but established risk of specific cardiovascular of fetal alcohol spectrum disorders in offspring of anomalies; fluoxetine is associated with isolated ven- women using selective serotonin reuptake inhibitors.
tricular septal defects (0.5% absolute risk increase) In Finland, the diagnosis of fetal alcohol spectrum and paroxetine is associated with right ventricular disorders is made according to the U.S. Institute of outflow tract defects (0.2% absolute risk increase).
Medicine criteria,38 which implies that substantial These findings should guide clinicians to not consider maternal alcohol consumption during pregnancy has fluoxetine or paroxetine as the first options when been ascertained by the clinician before the fetal prescribing these drugs to women planning preg- alcohol spectrum disorder diagnosis can be made and nancy. Special attention should be given to alcohol reported to the Register of Congenital Malformations.
use, smoking, and use of other psychiatric drugs in The diagnosis of fetal alcohol spectrum disorder is pregnant women using selective serotonin reuptake therefore reliable and not dependent on the occa- sional information on alcohol exposure from theMedical Birth Register or the Register of Congenital REFERENCES
Malformations. In addition to the obvious heavy use 1. Ellfolk M, Malm H. Risks associated with in utero and lactation of alcohol, women using selective serotonin reuptake exposure to selective serotonin reuptake inhibitors (SSRIs).
inhibitors during pregnancy smoke more often and use more prescription drugs, including psychiatric 2. Burt VK, Stein K. Epidemiology of depression throughout the female life cycle. J Clin Psychiatry 2002;63:9 –15.
drugs, than selective serotonin reuptake inhibitor 3. Andrade SE, Raebel MA, Brown J, Lane K, Livingston J, nonusers. We found an association between exposure Boudreau D, et al. Use of antidepressant medications during to selective serotonin reuptake inhibitors together pregnancy: a multisite study. Am J Obstet Gynecol 2008;198: with other psychiatric drugs and overall major con- genital anomalies, even after adjusting to confounders 4. Wichman CL, Moore KM, Lang TR, St. Sauver JL, Heise RH Jr, Watson WJ. Congenital heart disease associated with selec- (adjusted OR 1.31, 95% CI 1.04 –1.64).
tive serotonin reuptake inhibitor use during pregnancy. Mayo An important issue capable of reducing the com- parability of results obtained in individual studies is 5. Diav-Citrin O, Shechtman S, Weinbaum D, Wajnberg R, which congenital anomalies should be classified as Avgil M, Di Gianantonio E, et al. Paroxetine and fluoxetine inpregnancy: a prospective, multicentre, controlled, observa- major or minor. The rates of closure for small mus- tional study. Br J Clin Pharmacol 2008;66:695–705.
cular ventricular septal defects are nearly 90% by the 6. Pedersen LH, Henriksen TB, Vestergaard M, Olsen J, Bech first year of life without surgical or medical interven- BH. Selective serotonin reup-take inhibitors in pregnancy and Malm et al
SSRIs and Major Congenital Anomalies congenital malformations: population based cohort study. BMJ 23. Wogelius P, Norgaard M. Gislum M, Pedersen L, Munk E, Mortensen PB, et al. Maternal use of selective serotonin 7. Alwan S, Reefhuis J, Rasmussen SA, Olney RS, Friedman JM, reuptake inhibitors and risk of congenital malformations. Epi- National Birth defects Prevention Study. Use of selective serotonin-reuptake inhibitors in pregnancy and the risk of birth 24. Ericson A, Ka¨lle´n B, Wiholm B. Delivery outcome after the use defects. N Engl J Med 2007;356:2684 –92.
of antidepressants in early pregnancy. Eur J Clin Pharmacol 8. Louik C, Lin AE, Werler MM, Herna´ndez-Díaz S, Mitchell AA. First-trimester use of selective serotonin-reuptake inhibi- 25. Chambers CD, Johnson KA, Dick LM, Felix RJ, Jones KL.
tors and the risk of birth defects. N Engl J Med 2007;356: Birth outcomes in pregnant women taking fluoxetine. N Engl 9. Bakker MK, Kerstjens-Frederikse WS, Buys CH, de Walle HE, 26. Pastuszak A, Schick-Boschetto B, Zuber C, Feldkamp M, de Jong-van den Berg LT. First trimester use of paroxetine and Pinelli M, Sihn S, et al. Pregnancy outcome following first- congenital heart defects: a population based case-control study.
trimester exposure to fluoxetine (Prozac). JAMA 1993;269: Birth Defects Res A Clin Mol Teratol 2010;88:94 –100.
10. Reis M, Ka¨lle´n B. Delivery outcome after maternal use of 27. Gissler M, Teperi J, Hemminki E, Merila¨inen J. Data quality antidepressant drugs in pregnancy: an update using Swedish after restructuring a nationwide medical birth registry. Scand J data. Psychol Med 2010;40:1723–33.
11. Wurst KE, Poole C, Ephross SA, Olshan AF. First trimester 28. EUROCAT Guide 1.3 and reference documents. Instructions paroxetine use and the prevalence of congenital, specifically for the Registration and Surveillance of Congenital Anomalies.
cardiac, defects: A meta-analysis of epidemiological studies.
Birth Defects Res A Clin Mol Teratol 2010;88:159 –70.
12. Bar-Oz B, Einarson T, Einarson A, Boskovic R, O’Brien L, 29. National Institute for Health and Welfare. Congenital anoma- Malm H, et al. Paroxetine and congenital malformations: lies 1993–2008. Statistical Report 7/2011. Available at: meta-analysis and consideration of potential confounding fac- 13. Cole JA, Ephross SA, Cosmatos IS, Walker AM. Paroxetine in 30. Pakkasja¨rvi N, Ritvanen A, Herva R, Peltonen L, Kestila¨ M, the first trimester and the prevalence of congenital malforma- Ignatius J. Lethal congenital contracture syndrome (LCCS) tions. Pharmacoepidemiol Drug Saf 2007;16:1075– 85.
and other lethal arthrogryposes in Finland–an epidemiological 14. Malm H, Klaukka T, Neuvonen PJ. Risks associated with study. Am J Med Genet A 2006;140A:1834 –9.
selective serotonin reuptake inhibitors in pregnancy. Obstet 31. Leoncini E, Botto LD, Cocchi G, Annere´n G, Bower C, Halliday J, et al. How valid are the rates of Down syndrome 15. Oberlander TF, Warburton W, Misri S, Riggs W, Aghajanian J, internationally? Findings from the International Clearinghouse Hertzman C. Major congenital malformations following pre- for Birth Defects Surveillance and Research. Am J Med Genet natal exposure to serotonin reuptake inhibitors and ben- zodiazepines using population-based health data. Birth Defects 32. National Agency for Medicines (Department of safety and Res B Dev Reprod Toxicol 2008;83:68 –76.
drug information) and Social Insurance Institution (Research 16. Be´rard A. Paroxetine exposure during pregnancy and the risk department): Finnish Statistics on Medicines 2006. 1st ed.
of cardiac malformations: what is the evidence? Birth Defects Edita Prima Oy, Helsinki (Finland); 2007.
Res A Clin Mol Teratol 2010;88:171– 4.
33. Garne E, Khoshnood B, Loane M, Boyd P, Dolk H, 17. Einarson A, Choi J, Einarson TR, Koren G. Incidence of major EUROCAT Working Group. Termination of pregnancy for malformations in infants following antidepressant exposure in fetal anomaly after 23 weeks of gestation: a European register- pregnancy: results of a large prospective cohort study. Can based study. BJOG 2010;117:660 – 6.
34. Sloot WN, Bowden HC, Yih TD. In vitro and in vivo 18. Merlob P, Birk E, Sirota L, Linder N, Berant M, Stahl B, et al.
reproduction toxicology of 12 monoaminergic reuptake inhib- Are selective serotonin reuptake inhibitors cardiac teratogens? itors: possible mechanisms of infrequent cardiovascular anom- Echocardiographic screening of newborns with persistent heart alies. Reprod Toxicol 2009;28:270 – 82.
murmur. Birth Defects Res A Clin Mol Teratol 2009;85: 35. Sari Y, Zhou FC. Serotonin and its transporter on proliferation of fetal heart cells. Int J Dev Neurosci 2003;21:417–24.
19. Einarson A, Pistelli A, DeSantis M, Malm H, Paulus WD, 36. Kusakawa S, Yamauchi J, Miyamoto Y, Sanbe A, Tanoue A.
Panchaud A, et al. Evaluation of the risk of congenital cardio- Estimation of embryotoxic effect of fluoxetine using embry- vascular defects associated with use of paroxetine during onic stem cell differentiation system. Life Sci 2008;83:871–7.
pregnancy. Am J Psychiatry 2008;165:749 –52.
37. Bosi G, Garani G, Scorrano M, Calzolari E, IMER Working 20. Be´rard A, Ramos E, Rey E, Blais L, St.-Andre´ M, Oraichi D.
Party. Temporal variability in birth prevalence of congenital First trimester exposure to paroxetine and risk of cardiac heart defects as recorded by a general birth defects registry.
malformations in infants: the importance of dosage. Birth Defects Res B Dev Reprod Toxicol 2007;80:18 –27.
38. Hoyme HE, May PA, Kalberg WO, Kodituwakku P, Gossage 21. Davis RL, Rubanowice D, McPhillips H, Raebel MA, Andrade JP, Trujillo PM, et al. A practical clinical approach to diagnosis SE, Smith D, et al. Risks of congenital malformations and of fetal alcohol spectrum disorders: clarification of the 1996 perinatal events among infants exposed to antidepressant institute of medicine criteria. Pediatrics 2005;115:39 – 47.
medications during pregnancy. Pharmacoepidemiol Drug Saf 39. De Jong van den Berg LT, Feenstra N, Sorensen HT, Cornel MC. Improvement of drug exposure data in a registration of 22. Ka¨lle´n BA, Otterblad Olausson P. Maternal use of selective congenital anomalies. Pilot-study: pharmacist and mother as serotonin reuptake inhibitors in early pregnancy and infant sources for drug exposure data during pregnancy. EuroMAP congenital malformations. Birth Defects Res A Clin Mol Group. European Medicine and Pregnancy Group. Teratology Malm et al
SSRIs and Major Congenital Anomalies OBSTETRICS & GYNECOLOGY

Source: http://www.mitchellsmithlaw.com/celexa/malm.pdf