Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Comments

Bellet, 2015

Type of study:

prospective cohort

Setting and country:

pharmacovigilance and teratogenic risk reporting databases, France

Funding and conflicts of interest: non-commercial,

none reported

Inclusion criteria:

­pregnant women who had contacted pharmacovigilance centres about the safety of their medication in pregnancy

Exclusion criteria:

Duplicate database entries, not pregnant at the time of the first contact (preventive and retrospective cases), known developmental anomaly in fetus before the first contact (retrospective cases), ongoing pregnancy at 31 December 2011, follow-up data not available; patients were excluded from the exposed group if they were taking known teratogen(s) (e.g. carbamazepine, carbimazole, cyclophosphamide, lithium, methotrexate, misoprostol, mycophenolate, phenobarbital, retinoids, thalidomide, valproic acid and vitamin K antagonists)

N total at baseline:

(exposed/unexposed)

N=86/N=172

*exposed only in the 1^st trimester: n=56

*exposed in trimester 2 and 3: n= 22

*exposed only in trimester 2: n=8

Mean age ± SD:

(exposed/unexposed)

N=77/N=172

31.8 ± 5.8/31.4 ± 5.4

Smoking (exposed/unexposed

n=30/n=93)

<5 cigarettes/day

10% / 5.4%

≥5 cigarettes/day

36.7% / 5.4%

Alcohol

(exposed/unexposed

n=29/n=90)

<2 drinks/day

17.2% / 8.9%

≥2 drinks/day

3.4% / 0%

Co-medications:

72.1% of exposed patients had co-medications (benzodiazepines, antidepressants, other antipsychotics and anticonvulsants)

Indications:

schizophrenia (n=22; 30.6%),

psychotic disorders not otherwise specified (n= 14; 19.4%), bipolar disorders (n= 12; 16.7%), depression

(n = 11; 15.3%)

(exposed)

use of aripiprazole

(5 to 30 mg/day) during pregnancy

(unexposed)

no use of aripiprazole

or exposed to agents known to be non-teratogenic

2 random unexposed patients were matched for age (± 2 years) and gestational age at the first call (± 2weeks) with each exposed patient

Duration or endpoint of follow-up:

ascertainment of outcomes within 2 months of the expected date of delivery

For how many participants were no complete outcome data available?

(exposed/unexposed)

N: 15/11

Reasons for incomplete outcome data described? Yes

(exposed: 8 miscarriages (9.3%), 8 voluntary abortions (9.3%) and 1 therapeutic abortion)

(unexposed: 10 miscarriages

(5.8%) and 3 voluntary abortions (1.7%))

Outcome measures:

major congenital malformations (resulting in serious medical, surgical or cosmetic consequences), unrelated to a chromosomal defect or a genetic syndrome

Effect estimates for aripiprazole exposure:

OR 2.30 (95% CI 0.32–16.7)

(unadjusted)

Rates of major congenital malformations

(exposed/unexposed):

n=2 (out of 71; 2.8%)

n=2 (out of 161; 1.2%)

malformations in the exposed group:

-oesophageal atresia

-multiple cardiovascular defects (aorta dextroposition, ventricular septal defect, pulmonary stenosis)

authors conclusions:

-no stat. significant association was found between aripiprazole exposure and risk of major congenital malformations

-analyses were not adjusted for potential confounders

Cohen, 2018

Type of study:

prospective cohort

Setting and country:

National Pregnancy Registry for Atypical Antipsychotics, Massachusetts General Hospital, US

Funding and conflicts of interest: non-commercial;

none reported

Inclusion criteria:

pregnant women aged 18-45 yrs with psychiatric illness

Exclusion criteria:

already documented evidence of a birth defect via prenatal testing at the time of enrolment; chromosomal and single-gene abnormalities, and minor anomalies were excluded by a dysmorphologist trained

in teratology and genetics

N total at baseline:

(exposed/unexposed)

N=152/N=205

Mean age ± SD:

(exposed/unexposed)

32.3±4.9 /33.3 ± 4.07 (p=0.04)

mean BMI at baseline

27.1±6.5/25.3±5.1 (p=0.004)

Smoking (exposed/unexposed)

n=32 (21.1%)/n=11 (5.4%) (p=0.012)

Alcohol

(exposed/unexposed)

n=30 (19.7%)/n=64 (31.7%)

Co-medications:

(exposed/unexposed)

FGA 2%/1%, p=0.65

SSRI 31.6%/50.2%, p=0.001

SNRI 7.2%/5.9%, p=0.60

tricyclic ADD 1.35%/1.5%, p=1.00

second-generation ADD 12.5%/15.1%, p=0.48

lithium 2.6%/2.0%, p=0.73

anticonvulsant 46.1%/20%, p=0.001

anxiolytic 29%/19.5%, p=0.044

sedative 9.2%/4.4%, p=0.07

stimulant 4.6%/4.9%, p=0.91

Indications:

(exposed/unexposed)

bipolar disorder 67.8%/28.0%, p=0.001

schizophrenia 1.3%/0%, p=0.10

depression 13.8%/26.8%, p=0.003

anxiety disorder 4%/24.4%, p=0.001

(exposed)

women with psychiatric disorders using quetiapine

during the 1^st trimester of pregnancy

(unexposed)

women with psychiatric disorders not using SGA during pregnancy

Duration or endpoint of follow-up: 12 weeks after delivery

The outcomes were ascertained by telephone interviews at enrollment, 7 month gestation and 12 weeks after delivery

For how many participants were no complete outcome data available?

(exposed)

-dropped out or lost to follow-up (N=69)

-outcome not available yet (N=80)

-no first-trimester exposure (N=35)

(unexposed)

-dropped out or lost to follow-up (N=26)

-outcome not available yet (N=116)

Reasons for incomplete outcome data described? yes, participants not yet given birth

Outcome measures:

major congenital malformations

Effect estimates for quetiapine:

OR 0.90 (95% CI 0.15-5.46)

(unadjusted)

authors conclusions:

-“the safety of fetal exposure to quetiapine in a small sample of well-characterized

participants is reassuring given that the confidence interval does not exceed a fivefold increased risk of major malformations relative to psychiatric control subjects”

did not adjust for confounders

Huybrechts, 2016

Type of study:

prospective cohort

Setting and country:

Medicaid Analytic Extract database, Brigham

and Women’s Hospital, US

Funding and conflicts of interest: non-commercial;

none reported

Inclusion criteria:

all pregnancies that resulted in live births for which Medicaid covered the health care expenses (women and girls aged 12 to 55 yrs had coverage through Medicaid from 3

months before the start of pregnancy to 1 month after delivery; infants had Medicaid for the first 3 months of life unless they died sooner)

Exclusion criteria:

exposure to a known teratogenic medication

during the 1st trimester or a chromosomal abnormality

N total at baseline:

(exposed/unexposed)

N=9,258 (atypical AP)+N=733 (typical AP) /

N=1,331,910 (no AP)

Mean age ± SD:

(exposed atypical AP/unexposed)

25.4±6.4/24±5.8 yrs

(exposed typical AP/unexposed)

26.9±6.1/24 ±5.8 yrs

BMI: not reported

Smoking

(exposed atypical AP/unexposed)

882 (9.5%)/39,418 (2.9%)

(exposed typical AP/unexposed)

68 (9.3%)/ 39,418 (2.9%)

Alcohol

(exposed atypical AP/unexposed)

393 (4.2%)/6669 (0.5%)

(exposed typical AP/unexposed)

38 (5.2%)/6669 (0.5%)

Co-medications (only psychotropic):

(exposed atypical AP/unexposed)

anticonvulsants 2804 (30.3%)/21,055 (1.6%)

antidepressants 6629 (71.6%)/107,131 (8%)

anxiolytics 398 (4.3%)/4,305 (0.3%)

barbiturates 288 (3.1%)/12,992 (1%)

benzodiazepines 2733 (29.5%)/35,889 (2.7%)

other hypnotics 1838 (19.9%)/43,700 (3.3%)

stimulants 937 (10.1%)/7789 (0.6%)

(exposed typical AP/unexposed)

anticonvulsants 184 (25.1%)/21,055 (1.6%)

antidepressants 389 (53.1%)/107,131 (8%)

anxiolytics 25 (3.4%)/4305 (0.3%)

barbiturates 33 (4.5%)/12,992 (0.1%)

benzodiazepines 173 (23.6%)/35,889 (2.7%)

other hypnotics 175 (23.9%)/43,700 (3.3%)

stimulants 29 (3.9%)/7,789 (0.6)

Indications:

(exposed atypical AP/unexposed)

bipolar disorder 3165 (34.2%)/9548 (0.7%)

psychosis 537 (5.8%)/1655 (0.1%)

depression 2,985 (32.2%)/ 62,931 (4.7%)

schizophrenia 761 (8.2%)/873 (0.1%)

anxiety 2373 (25.6%)/41,813 (3.1%)

ADHD 768 (8.3%)/9965 (0.8%)

other 964 (10.4%)/14,930 (1.1%)

(exposed typical AP/unexposed)

bipolar disorder 201 (27.4%)/9548 (0.7%)

psychosis 95 (12.9%)/1655 (0.1%)

depression 192 (26.2%)/62,931 (4.7%)

schizophrenia 160 (21.8%)/873 (0.1%)

anxiety 134 (18.3%)/41,813 (3.1%)

ADHD 24 (3.3%)/9,965 (0.8%)

other 73 (9.9%)/14,930 (1.1%)

(exposed)

use of typical or atypical AP (as class) during the 1^st trimester of pregnancy defined as filling at least 1 prescription during the first 90 days of pregnancy

also analysis separately for aripiprazole, olanzapine, quetiapine, risperidone, ziprasidone

(unexposed)

no use of AP in the 1^st trimester defined as not filling an AP prescription during the 3 months before the start of pregnancy or during the 1st trimester

Duration or endpoint of follow-up:

at least 1 month after delivery, max 90 days

Outcome measures:

major congenital malformations (central nervous system, ear, eye, cardiovascular,

other vascular, respiratory, oral cleft, gastrointestinal tract, genital, urinary, musculoskeletal, limb, or other) identified during the first 90 days after delivery through maternal and infant medical records

*separate analysis for cardiac malformations

effect estimates:

-absolute risk for any congenital malformations:

32.7 (95% CI 32.4–33.0) per 1000 births not exposed to AP were diagnosed with congenital malformations compared with 44.5 (95% CI, 40.5–48.9) per 1000 births exposed to atypical AP and

38.2 (95% CI, 26.6–54.7) per 1000 births exposed to typical AP

-relative risks (RRs) for all congenital malformations:

*unadjusted atypical AP:

RR 1.36 (95% CI 1.24–1.50)

*adjusted atypical AP

RR 1.05 (95% CI 0.96–1.16)

**unadjusted typical AP:

RR 1.17 (95% CI 0.81–1.68)

**adjusted typical AP:

RR 0.90 (95% CI 0.62–1.31)

-absolute risk for cardiac

malformations:

11.6 (95% CI 11.4–11.7) per 1000 births not exposed to AP were diagnosed with congenital cardiac malformations compared with

16.2 (95% CI 13.8–19.0) per 1000 births exposed to atypical and

13.6 (95% CI 7.4–24.9) per 1000 births exposed to typical AP

RRs for cardiac malformations:

*unadjusted atypical AP:

RR 1.40 (95% CI 0.19-1.64)

*adjusted atypical AP

RR 1.06 (95% CI 0.90–1.24)

**unadjusted typical AP:

RR 1.18 (95% CI 0.64–2.18)

**adjusted typical AP:

RR 0.75 (95% CI 0.39-1.43)

Effect estimates for risperidone (adjusted):

-overall malformations

RR 1.26 (95% CI, 1.02–1.56)

-cardiac malformations

RR 1.26 (95% CI 0.88–1.81)

author’s conclusions:

-“use of AP early in pregnancy generally does not meaningfully increase the risk for congenital malformations overall or cardiac malformations in particular”

-“the small increase in the risk for malformations

observed with risperidone requires additional study”

-“in this cohort study of 1.3 million pregnant women, after accounting for psychiatric

conditions and other potential confounding variables, no increased risk for congenital

malformations was found for typical or atypical antipsychotics, with a possible exception

for risperidone”

adjusted for calendar year, age, race, smoking, multiple gestation, indications for AP (schizophrenia,

bipolar disorder, psychosis, depression, anxiety, attention-deficit/hyperactivity disorder, and other psychiatric disorders), other maternal morbidity (ther psychiatric and neurologic conditions (personality disorder, adjustment disorder, delirium, sleep disorders, chronic fatigue syndrome, alcohol and other drug abuse or dependence, epilepsy or convulsions, migraine, and other headaches) and chronic maternal illness (diabetes,hypertension, renal disease, and obesity)), concomitant medication use (anticonvulsants, antidepressants, anxiolytics, benzodiazepines,

other hypnotics, barbiturates, and stimulants), antidiabetic medications, antihypertensives,

and suspected teratogens, as well as methadone, naloxone,

naltrexone, and opioid use as possible proxies for drug abuse or dependence), general markers of the burden of

illness (Obstetric Morbidity Index10 and numbers of distinct prescriptions for medications other than APs, distinct diagnoses, outpatient visits, hospitalizations, and emergency department visits)

Vigod, 2015

Type of study:

population based cohort study (retrospective analysis of prescription and medical records databases)

Setting and country: multiple linked population health databases housed at the Institute for Clinical Evaluative Sciences (ICES) in Toronto, Ontario,

Canada

Funding and conflicts of interest:

grant from the Canadian

Institutes of Health Research (CIHR), and by an

annual grant from the Ontario Ministry of Health and Long-Term Care; author’s funding had no impact on the results and interpretation of the study

Inclusion criteria:

women who delivered a singleton infant in Ontario between 2003 and 2012, who were eligible for provincially funded drug coverage, and who had filled a provincially funded drug prescription within 180 days

before pregnancy and one during pregnancy or within 180 days of delivery

Exclusion criteria:

absence of provincially funded drug coverage

N total at baseline:

(exposed/unexposed)

n=1,209/n=40,314 (unmatched cohort)

n=1,021/n=1,021 (matched cohort)

Mean age ± SD:

(exposed/unexposed)

28.8±6.1/ 26.7±6.3 yrs (unmatched cohort)

28.8±6.2/ 28.8±6.2 yrs (matched cohort)

BMI: not reported

Alcohol and smoking

(substance abuse), n(%)

(exposed/unexposed)

569 (47.1)/ 5523 (13.7) (unmatched cohort)

458 (44.9)/ 415 (40.7) (matched cohort)

Co-medications, n(%):

(exposed/unexposed)

1.unmatched cohort

-SSRI

364 (30.1)/ 1981 (4.9)

-Non-SSRI

322 (26.6)/ 989 (2.5)

-Mood stabilisers

167 (13.8)/ 331 (0.8)

-Benzodiazepines

306 (25.3)/ 630 (1.6)

2. matched cohort:

-SSRI

303 (29.7)/ 204 (20)

-Non-SSRI

264 (25.9)/ 149 (14.6)

-Mood stabilisers

105 (10.3)/ 62 (6.1)

-Benzodiazepines

222 (21.7)/ 135 (13.2)

Indications, n(%):

(exposed/unexposed)

1. unmatched cohort:

-Psychotic disorder

429 (35.5)/ 675 (1.69)

-Bipolar disorder or major depression

937 (77.5)/ 8708 (21.6)

-Personality disorder

393 (32.5)/ 1816 (4.5)

2. matched cohort:

-Psychotic disorder

319 (31.2)/ 160 (15.7)

-Bipolar disorder or major depression

758 (74.2)/ 673 (65.9)

-Personality disorder

295 (28.9)/ 231 (22.6)

(exposed)

 ≥2 consecutive prescriptions for AP between the conception date (estimated using the gestational age at birth) and the delivery date;

at least one of the prescriptions was filled in the 1st or 2nd trimester

(unexposed)*

not exposed to any AP during

pregnancy

Patients in the matched control group had a history of either psychotic disorder, bipolar disorder or major depression, alcohol or substance abuse, or a personality disorder, however it is unclear whether all of the patients in the control group had some sort of psychiatric disorders.

*matched 1:1 to AP patients using propensity scores (based on the HDPS score within 0.2 SD) and on maternal age at delivery within 3 years

Duration of end-point follow-up:

180 days after delivery for exposure, 42 days for the outcome data

Outcome measures:

congenital malformations

Effect estimate (adjusted)*:

matched cohort:

RR 1.19 (95% CI 0.75-1.91)

unmatched cohort (unadjusted):

RR 1.05 (95% CI 0.79 - 1.41)

author’s conclusions:

no significant risk of congenital malformations

*adjusted for the use of SSRI, non-SSRI, mood stabilisers (lithium, valproate, carbamazepine, lamotrigine, oxcarbazepine), or benzodiazepines in the same pregnancy

congenital malformations as a secondary outcome in the sensitivity analysis

for the unmatched cohort unadjusted estimates are shown

Diav-Citrin, 2005

Type of study:

multicentre prospective controlled cohort

Setting and country: multicentre (Israel, Netherlands, Germany, Italy); European network of teratology information services, used data from Teratology Information Services in each country

Funding and conflicts of interest: non-commercial;

none reported

Inclusion criteria:

pregnant women who (or whose physician or midwife) contacted the teratology information service about the safety of gestational exposure to medication in 1989-2001

Exclusion criteria: none reported

N total at baseline:

(exposed/unexposed)

215/631

Median age, yrs:

(exposed/unexposed)

32/30

(stat. sig. different between exposed and unexposed)

BMI: not reported

Smoking:

(exposed/unexposed)

≤5 cigarettes/day (n=157)

6 (3.8%)/20 (3.6%)

>5 cigarettes/day (n=562)

34 (21.7%)/ 33 (5.9%), p<0.001

Alcohol: not reported

Co-medications (%):

any other concomitant psychotropic medication (75.1), concomitant anti-cholinergic drugs (31.6)

Indications (%):

psychosis (33.5), schizophrenia (10.7), depression (9.3), bipolar disorder (4.2), schizoaffective disorder (1.4), anxiety (1.4), panic attacks (0.9), hyperemesis gravidarum (0.5), borderline personality disorder (0.5), suicide attempt (0.5), substance abuse (0.5), Tourette syndrome (0.5); not specified(36.1)

(exposed)

Haloperidol and penfluridol during pregnancy (oral haloperidol, median dose 5 mg (2.25-10 mg) daily, parenteral haloperidol, median dose 100 mg/4 weeks (50-100 mg); oral penfluridol, median dose 20 mg/week (20-40 mg);

in the 1st trimester: haloperidol n=136; penfluridol n=25)

(unexposed)

not exposed to haloperidol or penfluridol, but took drugs which are known to be nonteratogenic

Duration of end-point follow-up:

between the neonatal period and 6 years of age; in most cases within the 1st 2 years of life

For how many participants were no complete outcome data available?

(exposed/unexposed)

N/A

follow-up rate ranged from 63% to 100% depending on the study center

Reasons for incomplete outcome data described?

N/A

Outcome measures:

rates of major congenital malformations;

Effect estimates

(haloperidol and penfluridol together):

-prevalence of congenital malformations

(exposed/unexposed)

3.4% (6/179)/ 3.8% (22/581), p=0.797

--prevalence of congenital malformations

in live births with 1^st trimester exposure:

3.1% (4/128)/ 3.8% (22/581), p=1.000

-2 of the reported malformations were limb defects (absent left 4^th finger, common wrist (carpus) of left 1^st and 2^nd fingers, and upper limb reduction defect and foot deformity)

authors conclusions:

-no major teratogenic risk of penfluridol and haloperidol;

-a possible association of haloperidol and penfluridol with limb defects cannot be ruled out

did not adjust for confounders

Habermann, 2013

Type of study:

prospective cohort

Setting and country: Institute for Clinical Teratology and Drug Risk Assessment in Pregnancy, using data from Teratology Information Service (TIS), Germany

Funding and conflicts of interest: non-commercial;

none reported

Inclusion criteria:

all pregnant women consulted at Teratology information service from January 1997 to March 2009

Exclusion criteria: presence of prenatal pathologic findings; outcome of the pregnancy was known at the time of inclusion

N total at baseline:

(exposed/unexposed)

N=845 (study cohort n=561 + cohort 1 n=284))/ N=1122 (cohort 2)

Median age:

(exposed/unexposed)

32/32 yrs

Median BMI:

(exposed/unexposed)

24.2/23.7 kg/m²

Smoking

(study cohort/cohort 1/cohort2 (unexp.))

(n=529/n=264/n=1111)

≤5 cigarettes/day

27 (5.1%)/ 15 (5.7%)/ 42 (3.8%)

>5 cigarettes/day (n=562)

176 (33.3%)/ 98 (37.1%)/ 58 (5.2%)

Alcohol

(study cohort/cohort 1/cohort2 (unexp.))

(n=526/n=263/n=1112)

≤1 drink/day

23 (4.4%)/ 11 (4.2%)/ 24 (2.2%)

>1 drink/day

14 (2.7%)/ 13 (4.9%)/ 3 (0.3%)

Co-medications (only psychotropic):

(exposed)

total other psychoactive drugs, including antidepressants, anticonvulsants, benzodiazepines, anxiolytics/sedatives, opioids (58%)

Indications:

psychotic disorders, schizophrenia, depression, bipolar affective disorders, anxiety disorders

(exposed)

study cohort: exposed to SGA (concomitant FGA allowed):

olanzapine (n = 187), quetiapine (n = 185), clozapine (n = 73), risperidone (n = 64), aripiprazole (n = 60), ziprasidone (n = 37), amisulpride (n = 16),

zotepine (n = 2)

cohort 1: exposed only to FGA (the most frequent were: haloperidol (n = 64), promethazine (n = 86), flupentixol (n = 44))

(unexposed)

cohort 2:

no antipsychotic use; excluded all women exposed to known teratogenic, fetotoxic or insufficiently studied drugs, allowed drugs known to be nonteratogenic; used a random sample from this cohort in 2:1 ratio

Duration of end-point follow-up:

8 weeks after the estimated date of birth

(via hospital discharge summaries)

For how many participants were no complete outcome data available?

(exposed/unexposed)

18.3%/17.4% lost to follow-up

Reasons for incomplete outcome data described? N/A

Outcome measures:

major congenital malformations (only exposure in the 1^st trimester was used for calculations)

Effect estimates:

major congenital malformations:

-study cohort (SGA) versus. cohort 1 (FGA)

OR 1.22 (95% CI 0.55-2.70)

OR_adj 1.27 (95% CI, 0.57-2.82)

-study cohort (SGA) versus. cohort 2 (unexposed)

OR 2.13 (95% CI 1.19-3.83)

OR_adj 2.17 (95% CI 1.20-3.91)

-cohort 1 (FGA) versus. cohort 2 (unexposed):

OR 1.75 (95% CI 0.80-3.80)

OR_adj 1.71 (95% CI 0.78-3.76)

Rate of major congenital abnormalities:

24/456 (5.2%) study cohort

12/239 (5.0%) cohort 1

25/1014 (2.5%) cohort 2

For the 1^st trimester exposure:

22/430 (5.1%) study cohort

9/213 (4.2%) cohort 1

For individual drugs:

(used in the 1^st trimester):

3.59% (5/139) for quetiapine and 6.81% (3/44) for aripiprazole

amisulpride (n = 13) and zotepine (n = 2): no malformations

cardiovascular malformations:

-study cohort versus. cohort 1

OR 1.50 (95% CI 0.48-4.71)

-study cohort versus. cohort 2

OR 3.21 (95% CI 1.34-7.67)

-cohort 1 versus. cohort 2:

OR 2.13 (95% CI 0.65-7.01)

Rate of cardiovascular malformations (isolated or combined with malformations of other organ systems):

12/430 (2.8%) study cohort

4/213 (1.9%) cohort 1

9/1014 (0.9%) cohort 2

authors conclusions:

-stat. significantly high risk of major malformations in the SGA group;

-an increased risk for atrial and ventricular septal defects after the first trimester use of

SGAs cannot be completely ruled out

-considering first trimester drug use, the malformation rates varied between 3.59% (5/139) for quetiapine and 6.81% (3/44) for

aripiprazole without statistically significant differences; among

amisulpride (n = 13) and zotepine (n = 2) exposed pregnancies,

no malformations were observed

adjusted for alcohol consumption (>1 drink/day)

also assessed maternal age, smoking, BMI, previous spontaneous abortions, previous malformed children in the models

McKenna, 2005

Type of study:

prospective cohort

Setting and country:

multicentre;

Canada, Israel, England (teratogen information centers, including Motherisk

Program at the Hospital for Sick Children in Toronto and an indepen-dent medical charity ‘Drug safety research unit’ in Southampton)

Funding and conflicts of interest: non-commercial;

none reported

Inclusion criteria:

all women who contacted the services regarding the use of antipsychotics and other drugs in the current pregnancy (assuming until 2004)

Exclusion criteria:

women who reported a psychiatric illness or the use of antipsychotics were excluded from the comparison group

N total at baseline:

(exposed/unexposed)

N=151 /N=151

Mean age ± SD: not reported

median BMI:

27.8/ 22.92 kg/m²

sig. different between exposed and unexposed

smoking:

(exposed/unexposed)

 n=38/ n=13

Alcohol

(exposed /unexposed)

heavy/binge n=9/ n=1

casual n=9/ n=11

Co-medications (only psychotropic):

(exposed)

17% anti-epileptics; 12.4% valproate;

6% lithium; 57% anti-depressants; 34% benzodiazepines

Indications (%):

depression (29), schizophrenia (24), bipolar disorder (18), schizoaffective (2), psychotic episode (7), psychotic depression (5), obsessive-compulsive disorder (2), PTSD (1), schizophreniform disorder (1)

(exposed)

Antipsychotics (SGA) during the 1^st trimester of pregnancy (olanzapine n=60, risperidone n=49, quetiapine n=36, clozapine n=6)

(unexposed)

no antipsychotics during pregnancy; taking drugs known to be non-teratogenic, also excluding other psychotropic drugs;

matched with exposed on maternal age ± 2 yrs and gestational age ± 2 weeks

Duration of end-point follow-up:

3-4 months after delivery

For how many participants were no complete outcome data available?

(exposed/unexposed)

N/A

Reasons for incomplete outcome data described?

N/A

Outcome measures:

major congenital malformations

Effect estimates:

rate of congenital malformations (exposed/unexposed)

0.9% (n=1) / 1.5% (n=2)

authors conclusions:

-no indications of a high teratogenic potential of AP, but the study was limited in sample size;

-discontinuation of AP is probably not needed in women with serious psychiatric morbidity

Sadowski, 2013

Type of study:

retrospective analysis of a prospective cohort (prescription database data)

Setting and country:

Motherisk

Program at the Hospital for Sick Children, Toronto,

Canada

Funding and conflicts of interest: non-commercial;

none reported

Inclusion criteria:

all women who contacted the Motherisk services between 2005 and 2009 regarding the use of antipsychotics and other drugs in the current pregnancy

Exclusion criteria:

exposure to teratogenic medications unrelated to the psychiatric disorder treatment; substance abuse (eg, alcohol, marijuana, cocaine, heroin, etc);

fertility-assisted pregnancies, twin or triplet pregnancies, pregnancies with known outcomes at the initial time of contact (eg, contacted Motherisk following the birth, reported abnormal pregnancy screening tests and/or ultrasounds); presence of psychiatric disorders in women of the comparison cohort

N total at baseline:

(exposed/unexposed)

N=133/ N=133

Mean age at conception ± SD (months):

(exposed/unexposed)

378.3±61.8/ 383.6±53.9

BMI: not reported

Pre-pregnancy weight (kg):

(exposed/unexposed)

74.3±20.9/ 64.6±13.0, p<0.001

Smoking N (%)

(exposed/unexposed)

24 (18)/ 0 (0), p<0.001

Alcohol, N (%)

(exposed /unexposed)

8 (6)/12 (9.1)

Co-medications (only psychotropic), N(%):

(exposed)^c

72.2% total; 54.1% combination with antidepressants;

SSRI n=46 (34.6), benzodiazepine n=21 (15.8), anticonvulsants n=19 (14.3), SNRI n=16 (12.0), atypical antidepressant n=12 (9), non-benzo hypnotic n=8 (6), typical antipsychotic n=5 (3.8), serotonin antagonist and reuptake inhibitor n=4 (3), tetracyclic antidepressant n=4 (3), tricyclic antidepressant n=3 (2.3), norepinephrine reuptake inhibitor n=2 (1.5),

synthetic cannabinoid n=2 (1.5), psychostimulant n=1 (0.8), GABA analogue n=1 (0.8), other dopamine antagonist n=1 (0.8)

Indications (%):

bipolar disorder (36.8), depression (27.1),

anxiety and depression (9.8) sleep disorders (9.8), schizophrenia (3), schizoaffective disorders (1.5)

(exposed)

women who initially called the service to inquire about

the safety of an SGA and who confirmed the use of this

medication for a minimum of 4 weeks of pregnancy

(quetiapine (69.9%), olanzapine (16.5%), risperidone (10.5%), aripiprazole (1.5%), paliperidone (0.8%), quetiapine plus olanzapine (0.8%))

(unexposed)

women who contacted Motherisk between 2005 and 2009 and

reported exposure to non-teratogenic agents

matched for age at conception (±3 years) and pregnancy duration at the initial time of contact (±2 weeks)

Duration of end-point follow-up:

between 2009 and 2012

For how many participants were no complete outcome data available?

of 370 women who contacted Motherisk in the study period, 107 women could

not be reached at their last reported contact details, 20 refused participation and 110 did not meet the inclusion criteria

Reasons for incomplete outcome data described?

yes

Outcome measures:

major congenital malformations

Effect estimates:

SGA (n=133) versus no SGA (n=133)

-congenital malformations

6.2% versus 2.6% (not sig.)

SGA monotherapy (n=37) versus. polypharmacy including SGA (n=96)

-congenital malformations

0 versus 7 (8.5%), OR 6.28 (95% CI 0.35-112.86)

authors conclusions:

- 6% of exposed infants

presented with major malformations compared with 2.6% in the comparison group

-although 14.3% of exposed mothers were also taking an antiepileptic (primarily lamotrigine), the reported malformations were not found among those infants exposed to anticonvulsants

- all the malformations were observed in the neonates exposed to polytherapy, but no specific pattern of malformations was found

- In a group receiving multiple psychotropic medications next to antipsychotics, the rate of malformations was (not statistically significantly) higher than in the antipsychotic monotherapy group (8.5% versus. 0%, OR 6.28 (95% CI 0.35-112.86)).

-limitations: small sample size, severity of the baseline disorder that required polytherapy, associated comorbidities, any other unknown environmental factors that were not

assessed

Petersen, 2016

Type of study:

retrospective analysis of prospective primary care data (prescription and electronic health records database)

Setting and country: The Health Improvement Network

(THIN) and the Clinical Practice Research Datalink (CPRD), UK

Funding and conflicts of interest: non-commercial; funded by National Institute for Health Research Health Technology Assessment

program (Grant 11/35/06);

none reported

Inclusion criteria:

all pregnant women and their children Identified in the databases between 1 January 1995 to 31 December 2012

Exclusion criteria: not described

N total at baseline:

(exposed/discontinued/unexposed)

N=290/ N=492/ N=210,966

(used for child outcomes)

Mean age ± SD (yrs): only age categories

Mean BMI ± SD, kg/m²:

(exposed/discontinued/unexposed)

28±6.7/ 27±6.8/ 26±6.4

Obesity, N (%):

(exposed/discontinued/unexposed)

53 (18.3)/ 62 (12.6)/ 15,363 (7.3)

Smoking N (%):

(exposed/discontinued/unexposed)

139 (47.9)/ 183 (37.2)/ 42,502 (20.1)

Alcohol, N (%):

(exposed/discontinued/unexposed)

23 (7.9)/ 28 (5.7)/ 1,124 (0.5)

Co-medications (only psychotropic)

In the 1^st trimester, N(%):

(exposed/discontinued/unexposed)

-anticonvulsant mood stabilisers

27 (9.3)/ 14 (2.8)/ 887 (0.4)

-lithium

11 (3.8)/ 2 (0.4)/ 7 (0)

-antidepressants

169 (58.3)/ 124 (25.2)/ 4,351 (2.1)

-anxiolytics

31 (10.7)/ 24 (4.9)/ 523 (0.2)

-hypnotics

41 (14.1)/ 28 (5.7)/ 423 (0.2)

Pre-existing psychiatric conditions, n (%):

(exposed/discontinued/unexposed)

depression

79 (27.2)/ 152 (30.9)/ 14,626 (6.9)

epilepsy

17 (5.9)/22 (4.5)/ 3,254 (1.5

SMI

180 (62.1)/ 144 (29.3)/ 882 (0.4)

(exposed)

records of prescriptions for antipsychotics during the 1^st trimester of pregnancy (cohort A)

(prescribed AP 2 years before the start of pregnancy and between day 31 and 105 of pregnancy)

(all antipsychotic drugs, except

prochlorperazine)

(unexposed)

no AP use 2 years before pregnancy and during the whole pregnancy (through the delivery date) (Cohort C)

(diseased + discontinued AP):

women with records of antipsychotic treatment in the 2 years before start of pregnancy, but no prescriptions issued after 4 weeks prior to pregnancy start

(Cohort B)

Duration of end-point follow-up:

-days of child follow-up,

median(IQR)

(exposed/discontinued/unexposed):

657 (286, 1351)/

834 (353, 1758)/

738 (349, 1416)

-days to major congenital

malformations record,

median (IQR)

(exposed/discontinued/unexposed):

85 (21, 110) /

136 (33,1167)/

50 (10, 120)

For how many participants were no complete outcome data available?

N/A

Reasons for incomplete outcome data described?

N/A

Outcome measures:

major congenital malformations

Effect estimates:

-cohort A versus cohort B (discontinued)

RR 1.54 (95% CI 0.65-3.63), p=0.32

RR adj. 1.79 (95% CI 0.72, 4.47), p=0.21

-cohort A versus cohort C (unexposed)

RR 1.74 (95% CI 0.93-3.25), p=0.08

RR adj. 1.59 (95% CI 0.84-3.00), p=0.15

authors conclusions:

-no stat. sig. association with major congenital malformations

-antipsychotic treatment in pregnancy carries limited risks of adverse pregnancy and

birth outcomes once adjustments have been made for health and lifestyle factors

adjusted for age, obesity, alcohol problems, smoking, illicit drug use, and antidepressant prescribing and anticonvulsant mood stabilizers

Reis, 2008

Type of study:

cohort (retrospective analysis of prospective data from a drug prescription database)

Setting and country:

Swedish Medical

Birth Register, Swedish Register

of Congenital Malformations, the Hospital Discharge

Register, Sweden

Funding and conflicts of interest:

none reported

Inclusion criteria:

all women giving birth in Sweden

Exclusion criteria: not reported (minor congenital malformations were excluded from the analysis; lithium-exposed pregnancies were evaluated in a separate analysis

N total at baseline:

(exposed/unexposed)

N=2908 /N=955,821

N total children (after excluding lithium):

(exposed/unexposed)

N=2893/N=973,767

N exposed (after excluding lithium, dixyrazine and prochlorperoxine):

n (women)=570

n (children)=576

Mean age ± SD: not reported (only groups; age distribution was different between groups)

Mean BMI: not reported (only groups)

BMI≥26

(exposed/unexposed):

n=225/n=215,796

(OR 2.06 (95%CI 1.72-2.47) versus normal BMI range)

Smoking

(exposed/unexposed)

<10 cigarettes/day

n=77/n=72,882

≥10 cigarettes/day

N=142/n=33,623

Alcohol: not reported

Co-medications (only psychotropic):

(n, exposed/unexposed)

anticonvulsants 23/2467

sedatives, hypnotics 141/3953

antidepressants 172/10,276

TCA 33/1578

SSRI 101/7835

other 31/971

Indications: not reported

(exposed)

women who used AP during the 1st trimester of pregnancy

(unexposed)

all other women in the database

(1994-2005)

Duration of end-point follow-up:

not reported

outcome measures:

any and ‘relatively severe’ congenital malformations, congenital cardiac malformations

effect estimate:

for the exposure to AP excluding dixyrazine and prochlorperoxine:

any congenital malformations

OR 1.31 (95% CI 0.93–1.86)

“relatively severe malformations”

OR 1.52 (95% CI 1.05–2.19)

OR 1.45 (95% CI 0.99–1.41), p=0.055 (after excluding users of anticonvulsants)

congenital cardiac malformations

RR 1.43 (95% CI 0.69–2.63)

author’s conclusions:

“the use of antipsychotics during early pregnancy is not associated with a markedly increased risk for a congenital malformation, but a moderate risk increase may occur”

“Because there seems to be little drug specificity, it is possible that underlying pathology or unidentified confounding factors explain that risk”

adjusted for year of birth, maternal age, parity, smoking, previous miscarriages

when the expected numbers were low, RR was given for the observed over expected numbers from exact Poisson distributions

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

Bellet, 2015

likely

unlikely

likely

likely

Cohen, 2018

likely

unlikely

unlikely

likely

Diav-Citrin, 2005

likely

likely

unlikely

likely

Habermann, 2013

likely

likely

unlikely

likely

Huybrechts, 2016

likely

unlikely

unlikely

unlikely

Reis, 2008

likely

likely

unlikely

likely

McKenna, 2005

likely

likely

unlikely

likely

Petersen, 2016

likely

unlikely

unlikely

unlikely

Sadowski, 2013

likely

unlikely

unlikely

likely

Vigod, 2015

likely

unlikely

unlikely

unlikely