Phenytoin (All indications)

Study	Type of data	Exposure measurement	Outcome assessment	Adjustment
Adab (Phenytoin), 2004	retrospective cohort	A clinician conducted semi-structured interviews of mothers and clinical records were used to confirm information.	A semi-structured interview was used to collect data on additional educational needs. Dysmorphic features and major malformations were recorded by an examiner. A neuropsychologist used the Wechsler Intelligence Test for Children (6 to 16 yo) and the Schedule of Growing Skills II (birth to 5 yo).	None.
Adams (Phenytoin), 2022	retrospective cohort	Women were interviewed during the peri- and post- partum periods regarding multiple pregnancy characteristics and risks.	A blinded psychometrist who performed The Wechsler Intelligence Scale for Children, 3rd Edition (WISC-III) to measure general mental abilities in the context of a comprehensive neuropsychological testing battery.	Each ASM-exposed group was matched to an unexposed group of mothers according to educational and socioeconomic characteristics, and their children were then selected to match according to gender and age at testing.
Al Bunyan (Phenytoin), 1999	retrospective cohort	The antenatal and perinatal records of the pregnant epileptic patients were examined.	Records and the post-partum examinations were carried out by a paediatrician who documented any congenital anomalies present.	Not specified.
Alsfouk (Phenytoin) (Controls exposed to Lamotrigine, sick), 2021	retrospective cohort	Patients’ electronic and paper-based medical records.	Patients’ electronic and paper-based medical records.	None.
Alsfouk (Phenytoin) (Controls unexposed, sick), 2021	retrospective cohort	Patients’ electronic and paper-based medical records.	Patients’ electronic and paper-based medical records.	None.
Arkilo (Phenytoin), 2015	retrospective cohort	Questionnaires were sent to women.	Questionnaires were sent to women. The developmental outcomes were assessed at 2 years of age by developmental specialists assigned by the school districts.	None.
Artama (Phenytoin), 2005	retrospective cohort (registry)	Information on the antiepileptic drugs received during pregnancy was abstracted from medical records of the mothers with epilepsy.	Information on pregnancy outcomes was abstracted from medical records of the mothers with epilepsy. Anomalies were classified according to the International Classification of Diseases, ninth revision (ICD-9).	Adjustment for maternal age at delivery or number of previous births did not affect the risk of congenital malformations in offspring related to different antiepileptic medications (results not shown).
Artama (Phenytoin) (Controls exposed to Lamotrigine, sick), 2013	population based cohort retrospective	The Special Refund Entitlement Register and the Register on Reimbursement Drugs contained informations on maternal drug treatment under ATC category N03.	The Medical Birth Register data are collected from all maternity hospitals. Information on all deaths is revised and supplemented from the Cause-of-Death Register. Information on malformations are obtained from the Finnish Malformation Register.	No adjustment for this group of comparison.
Artama (Phenytoin) (Controls unexposed, disease free), 2013	population based cohort retrospective	The Special Refund Entitlement Register and the Register on Reimbursement Drugs contained informations on maternal drug treatment under ATC category N03.	The Medical Birth Register data are collected from all maternity hospitals. Information on all deaths is revised and supplemented from the Cause-of-Death Register. Information on malformations are obtained from the Finnish Malformation Register.	Adjusted for maternal age at delivery, parity, university hospital district, socioeconomic status and major congenital anomalies.
Artama (Phenytoin) (Controls unexposed, sick), 2013	population based cohort retrospective	The Special Refund Entitlement Register and the Register on Reimbursement Drugs contained informations on maternal drug treatment under ATC category N03.	The Medical Birth Register data are collected from all maternity hospitals. Information on all deaths is revised and supplemented from the Cause-of-Death Register. Information on malformations are obtained from the Finnish Malformation Register.	Adjustment for maternal age at delivery, parity, university hospital district, socioeconomic status and major congenital anomalies.
Arteaga-Vázquez (Phenytoin), 2012	case control	The intake of anticonvulsants was obtained by direct interview of the mother during the first 24 hours after delivery or by consulting the obstetric record.	Clinical examination systematized of all consecutive births that occur in the hospitals that participate in the program. Uniform information is obtained through the filling of a specifically designed clinical form for the registry.	None.
Aydin (Phenytoin) (Controls exposed to Lamotrigine, sick), 2020	retrospective cohort	Data were obtained from Perinatology Division’s computerized system. Antiepileptic drug use were reported.	Data were obtained from Perinatology Division’s computerized system.	None.
Aydin (Phenytoin) (Controls unexposed, sick), 2020	retrospective cohort	Data were obtained from Perinatology Division’s computerized system. Antiepileptic drug use were reported.	Data were obtained from Perinatology Division’s computerized system.	None.
Bànhidy (Phenytoin), 2011	case control	Mothers were mailed a questionnaire (after the selection of cases and controls) requested information on medicinal products taken during pregnancy and to send their prenatal maternity logbook and other medical records. Regional nurses were asked to visit and question the non-respondent.	Notification of cases with congenital abnormality is mandatory for physicians to the HCAR. Pathologists sent a copy of autopsy report for stillbirths and infant deaths cases and defect diagnosed in prenatal diagnostic centers with or without termination of pregnancy were also included.	Matched according to sex, birth week in the year when cases were born, and district of parents’ residence.
Bank (Phenytoin) (Mixed indications), 2017	prospective cohort	Medical records were obtained for review from the treating obstetrician, the delivery hospital, and the treating pediatrician. Umbilical cord and maternal venous blood were collected to measure total anti-epileptic levels by liquid chromatography/mass spectrometry.	Medical records were obtained for review from the treating obstetrician, the delivery hospital, and the treating pediatrician.	None.
Battino (Phenytoin), 1992	prospective cohort	At least every 4 weeks throughout the pregnancy blood samples were taken for measurement of antiepileptic drugs concentrations.	Vital data of newborns at birth were recorded according to a standardized protocol; a more detailed clinical examination was performed on Day 5.	None.
Battino (Phenytoin), 1999	prospective cohort	Mothers were seen by an obstetrician and neurologist at least every 4 weeks throughout pregnancy, and blood samples were taken for the measurement of antiepileptic drugs concentrations.	The vital data of newborns at birth were recorded according to a standardized protocol; a more detailed clinical examination was performed on the fifth day. The infants were weighed within 1 hour of life, and their head circumference measured.	No adjustment for this group of exposure.
Battino (Phenytoin) (Epilepsy), 2024	prospective cohort	Reporting physicians collected information on drug therapy after each trimester.	Abnormalities in the offspring were recorded descriptively by reporting physicians. A committee blinded to type of exposure assessed and categorized these abnormalities. When necessary, the committee solicited additional information from the reporting physicians.	Exclusion of pregnancies exposed to known teratogenic drugs, and those with comorbidities associated with teratogenic risks. No adjustment for this group of comparison.
Bromley (Phenytoin) (Controls exposed to Lamotrigine, sick), 2008	prospective cohort	Information is collected during the pregnancy regarding the mother's epilepsy and medical history. Maternal epilepsy information and pregnancy were verified using medical case notes.	The diagnosis of autism spectrum disorder (ASD) was made as part of a routine clinical referral independent of the study. Children were reported meeting DSM-IV criteria for ASD or features of ASD including language impairment, a lack of attention, social difficulties and restricted interests.	None.
Bromley (Phenytoin) (Controls unexposed, disease free), 2008	prospective cohort	Information is collected during the pregnancy regarding the mother's epilepsy and medical history. Maternal epilepsy information and pregnancy were verified using medical case notes.	The diagnosis of autism spectrum disorder (ASD) was made as part of a routine clinical referral independent of the study. Children were reported meeting DSM-IV criteria for ASD or features of ASD including language impairment, a lack of attention, social difficulties and restricted interests.	None.
Bromley (Phenytoin) (Controls unexposed, sick), 2008	prospective cohort	Information is collected during the pregnancy regarding the mother's epilepsy and medical history. Maternal epilepsy information and pregnancy were verified using medical case notes.	The diagnosis of autism spectrum disorder (ASD) was made as part of a routine clinical referral independent of the study. Children were reported meeting DSM-IV criteria for ASD or features of ASD including language impairment, a lack of attention, social difficulties and restricted interests.	None.
Canger (Phenytoin), 1999	prospective cohort	The patients received monthly obstetric and neurologic examinations, and antiepileptic drug blood levels were tested monthly.	At the time of delivery the infants underwent a standardized examination by a pediatrician, and a more detailed clinical examination on day 5 in San Paolo Hospital only and during the first months in other hospitals (if so medical records were also acquired).	None.
Christensen (Phenytoin) (All indications) (Controls exposed to LTG), 2024	population based cohort retrospective	Prenatal exposure was identified via national prescription registers. These registers contain the Anatomical Therapeutic Chemical (ATC) classification code (www.whocc.no) and the date of dispensing.	Information on birth weight, gestational age and head circumference at birth was obtained from the medical birth registers for all births occurring at 22 weeks’ gestation or later.	Singleton only. No adjustment for this group of comparison.
Christensen (Phenytoin) (All indications) (Controls unexposed, general population), 2024	population based cohort retrospective	Prenatal exposure was identified via national prescription registers. These registers contain the Anatomical Therapeutic Chemical (ATC) classification code (www.whocc.no) and the date of dispensing.	Information on birth weight, gestational age and head circumference at birth was obtained from the medical birth registers for all births occurring at 22 weeks’ gestation or later.	Singleton only. All models were adjusted for country of birth, year of birth, sex of child, maternal age, maternal parity, maternal marital or cohabitation status, pre-pregnancy hospital admittances, maternal education, smoking in early pregnancy, maternal psychiatric history (F00-F99), and use of psychotropic drugs in pregnancy (ATC N06A, N05A, N05B, excl. N05BA09) and maternal epilepsy.
Cohen (Phenytoin), 2011	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records.	Blinded assessors used the Motor Scale from the Bayley Scales of Infant Development - II. The child’s parents and preschool teacher/daycare worker complete the Adaptive Behavior Assessment System - II and the Behavior Assessment System for Children. Parents complete the Parent Stress Index - III.	Adjusted for maternal IQ, standardized dose, gestational age at birth, and site location (US/UK) for the Bayley Scales of Infant Development - II and adjusted for standardized dose and maternal education beyond high school (yes/no) for the Adaptive Behavior Assessment System - II.
Cohen (Phenytoin), 2013	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records.	The child’s parents and teacher complete the Adaptive Behavior Assessment System Second Edition (ABAS-II), to complete the Behavior Assessment System for Children (BASC) and the Parent Stress Index Third Edition (PSI-III). Each rating scale was scored by assessors blinded to the AED used.	Mean adjusted for maternal IQ, standardized AED dose, birth gestational age, and preconception folate use.
Czeizel (Phenytoin), 1992	case control	Mothers were mailed a questionnaire requested information on medicinal products taken during pregnancy and to send their prenatal maternity logbook which related drug prescriptions. Regional nurses were asked to visit and question the non-respondent.	Notification of cases with congenital abnormality is mandatory for physicians to the HCAR. Pathologists sent autopsy report for stillbirths and infant deaths cases. Prenatally diagnosed and electively terminated malformed fetuses have also been registered.	None.
D'Souza (Phenytoin) (Controls unexposed, disease free), 1991	prospective cohort	Every two weeks, clinical assessment of seizure control, and adjustment of anticonvulsant dosage in accordance with serum drug concentration or clinical response.	Each infant was examined by a paediatrician and a geneticist, neither of whom was aware of whether the mother was in the study group or the control group.	Matched for maternal age, parity, and social class.
D'Souza (Phenytoin) (Controls unexposed, sick), 1991	prospective cohort	Every two weeks, clinical assessment of seizure control, and adjustment of anticonvulsant dosage in accordance with serum drug concentration or clinical response.	Each infant was examined by a paediatrician and a geneticist, neither of whom was aware of whether the mother was in the study group or the control group.	No matching for this group of exposure.
Dean (Phenytoin), 2002	retrospective cohort	A structured interview was carried out by a trained research nurse using questionnaires.	Standardised assessment was carried out by a trained research nurse using examination schedules. Clinical photographs were assessed for facial features. Behavior disorders, developmental delay and later childhood medical problems were recorded and/or diagnosed by specialists.	None.
Dean (Phenytoin) (Controls exposed to Lamotrigine, sick), 2007	retrospective cohort	A trained research nurse undertook interviews recording antiepileptic drugs use in each pregnancy. Maternal antiepileptic drugs dose was verified from case records where possible.	A trained research nurse undertook interviews recording children’s medical and developmental histories. Malformations and medical histories for the children were confirmed by hospital record review and clinical examination. Facial photographs were reviewed for features of FACS by blinded physicians	None.
Dean (Phenytoin) (Controls unexposed, sick), 2007	retrospective cohort	A trained research nurse undertook interviews recording antiepileptic drugs use in each pregnancy. Maternal antiepileptic drugs dose was verified from case records where possible.	A trained research nurse undertook interviews recording children’s medical and developmental histories. Malformations and medical histories for the children were confirmed by hospital record review and clinical examination. Facial photographs were reviewed for features of FACS by blinded physicians	None.
Díaz-Romero (Phenytoin), 1999	cohort	Not specified	The neonates were evaluated by the principal author who used a metallic calliper calibrated in millimeters and a glazed fiberglass tape for the head circumference. The measurements were performed twice and collected by a second observer; the final value was the average of both measurements.	None.
Fedrick (Phenytoin), 1973	retrospective cohort (registry)	In order to ascertain the drugs taken during the relevant pregnancies, the general practitioner of each epileptic mother was asked for details of all anticonvulsant drugs taken.	For hospital births, informations are abstracted by clerks from hospital notes and in domiciliary deliveries the midwife who delivered the infant sends her notes and clerks then abstract and code the information. General files were also searched for any subsequent hospital admissions or deaths.	None.
Forsberg (Phenytoin) (Controls unexposed NOS), 2011	retrospective cohort (registry)	The Medical Birth Register is built from data in standardized medical records. To identify the use of anticonvulsants, the original medical documents were retrieved and scrutinized and the information was added to the records of deliveries from women with epilepsy.	The Swedish School Mark Registry was used to obtain information about school grades from the last year of compulsory school, at age 16. School grades in four subjects were further analyzed: sports, and the three core subjects, mathematics, English, and Swedish.	Not specified.
Forsberg (Phenytoin) (Controls unexposed, sick), 2011	retrospective cohort (registry)	The Medical Birth Register is built from data in standardized medical records. To identify the use of anticonvulsants, the original medical documents were retrieved and scrutinized and the information was added to the records of deliveries from women with epilepsy.	The Swedish School Mark Registry was used to obtain information about school grades from the last year of compulsory school, at age 16. School grades in four subjects were further analyzed: sports, and the three core subjects, mathematics, English, and Swedish.	Not specified.
Guveli (Phenytoin) (Epilepsy), 2017	retrospective cohort	The clinical data of patients (dosages of AEDs used during pregnancy) were recorded retrospectively from patient files.	Echocardiography was checked by a pediatric cardiologist. Parents and children were examined and photographed by a medical geneticist. Children older than six months were examined by a pediatric dentist for developmental dental anomalies (blinded to exposure).	None.
He (Phenytoin) (Controls exposed to Lamotrigine, sick), 2017	prospective cohort	Data collected included antiepileptic drug use.	Clinical records were scanned to verify pregnancy complication and delivery complication. Neonatologist ascertained the health of offspring (neonatal asphyxia, congenital malformation or Apgar score less than or equal to 7).	None.
He (Phenytoin) (Controls unexposed, sick), 2017	prospective cohort	Data collected included antiepileptic drug use.	Clinical records were scanned to verify pregnancy complication and delivery complication. Neonatologist ascertained the health of offspring (neonatal asphyxia, congenital malformation or Apgar score less than or equal to 7).	None.
Hernández-Díaz (Phenytoin) (Epilepsy), 2012	prospective cohort	Women were questioned with a computer-assisted telephone interview 3 times.	Women were questioned with a computer-assisted telephone interview 3 times and medical records were obtained. The written descriptions in the pediatricians’ examinations were reviewed separately by the clinical teratologist.	No adjustment for this group of comparison.
Holmes (Phenytoin) (Controls unexposed, disease free), 2001	retrospective cohort	Women are interviewed and completed questionnaires administered by a research assistant.	The infants were examined by a study physician or the results of the pediatrician’s examination were reviewed if a mother chose not to enroll. For major malformations: features found on prenatal ultrasonography but not on physical examination were excluded.	For cigarette and alcohol use, substance abuse, severity of seizures, and head circumference and height of the mother.
Holmes (Phenytoin) (Controls unexposed, sick), 2001	retrospective cohort	Women are interviewed and completed questionnaires administered by a research assistant.	The infants were examined by a study physician or the results of the pediatrician’s examination were reviewed if a mother chose not to enroll. For major malformations: features found on prenatal ultrasonography but not on physical examination were excluded.	No adjustment for this group of comparison.
Hvas (Phenytoin) (Controls unexposed, disease free), 2000	retrospective cohort	Women completed a questionnaire. They were categorised as users or nonusers of anticonvulsant drugs, by self-reported daily intake of any anticonvulsant drug during the first trimester.	Women completed a questionnaire. Information about major congenital malformations was also collected.	No adjustment for this group of exposure.
Hvas (Phenytoin) (Controls unexposed, sick), 2000	retrospective cohort	Women completed a questionnaire. They were categorised as users or nonusers of anticonvulsant drugs, by self-reported daily intake of any anticonvulsant drug during the first trimester.	Women completed a questionnaire. Information about major congenital malformations was also collected.	No adjustment for this group of exposure.
Kaaja (Phenytoin), 2003	prospective cohort	Follow-up of the pregnant women with epilepsy included assessment of serum concentration of antiepileptic drugs at the end of the first trimester.	Infants were examined by a neonatologist at birth and at discharge from the hospital and charts for the infants admitted to the pediatric clinic or in case of termination of pregnancy were reviewed. Autopsy was performed on stillbirth.	No adjustment for this group of exposure.
Källén (Phenytoin) (Controls exposed to Lamotrigine, sick) (Indications NOS), 2013	population based cohort retrospective	At the midwife interview at the first antenatal care visit, the woman was asked if she had used any drugs since she became pregnant. Or determined by the use of the Swedish Register of Prescribed Drugs (since 2006).	The Swedish Medical Birth Registry contain information based on standardized medical records from the first and further antenatal visit, the delivery and the paediatric examination. Supplemented with data from the Register of Birth Defects and Hospital Discharge Register.	No adjustment for this group of comparison.
Källén (Phenytoin) (Controls unexposed, NOS) (Indications NOS), 2013	population based cohort retrospective	At the midwife interview at the first antenatal care visit, the woman was asked if she had used any drugs since she became pregnant. Or determined by the use of the Swedish Register of Prescribed Drugs (since 2006).	The Swedish Medical Birth Registry contain information based on standardized medical records from the first and further antenatal visit, the delivery and the paediatric examination. Supplemented with data from the Register of Birth Defects and Hospital Discharge Register.	Adjustment was made for year of birth, maternal age (5-year class), parity, smoking in early pregnancy and BMI.
Kaneko (Phenytoin), 1999	prospective cohort	Mothers were seen by an obstetrician and neurologist at least every 4 weeks throughout pregnancy, and blood samples were taken for the measurement of antiepileptic drugs concentrations.	Congenital malformations were examined at each center at birth, at 5 days, and at the 1-month visit by a team of obstetricians and neurologists, according to a standardized check-list based on the report of the Japanese Association of Obstetricians for Maternal Welfare.	None.
Katz (Phenytoin), 2001	retrospective cohort	Telephone interviews were given to the patient as a supplement to the chart analysis. Spouses were interviewed only when additional information could not be provided by the patient.	Data collected in chart review about the newborn included the presence of developmental delay. Including pervasive developmental delay (based on the DSM-IV-TR criteria), combined or isolated speech and gross motor delay and attention deficit disorder with speech and auditory processing delay.	None.
Kelly (Phenytoin), 1984	prospective cohort	Efforts were made to ensure that periodic determinations of blood levels of anticonvulsants were obtained, especially during pregnancy.	Each child born at the University Hospital was examined neonatally by one of the investigators. When birth occurred elsewhere, newborn records were requested. Then further evaluation was conducted later on at clinic visits or home visits. Data were augmented by local health records.	None.
Kilic (Phenytoin) (Controls exposed to Lamotrigine, sick) (Mixed indications), 2014	population based cohort retrospective	The Danish Register of Medicinal Product Statistics holds information on all redeemed prescriptions. Monotherapy exposure was defined as redemption of prescription for one type of AED with the Anatomical Therapeutic Codes.	The Danish Medical Birth Registry contains data on newborn babies including information on gestational age at birth, birth weight, and head circumference.	No adjustment for this group of comparison.
Kilic (Phenytoin) (Controls unexposed NOS) (Mixed indications), 2014	population based cohort retrospective	The Danish Register of Medicinal Product Statistics holds information on all redeemed prescriptions. Monotherapy exposure was defined as redemption of prescription for one type of AED with the Anatomical Therapeutic Codes.	The Danish Medical Birth Registry contains data on newborn babies including information on gestational age at birth, birth weight, and head circumference.	No adjustment for this group of comparison.
Kilic (Phenytoin) (Controls unexposed, sick) (Mixed indications), 2014	population based cohort retrospective	The Danish Register of Medicinal Product Statistics holds information on all redeemed prescriptions. Monotherapy exposure was defined as redemption of prescription for one type of AED with the Anatomical Therapeutic Codes.	The Danish Medical Birth Registry contains data on newborn babies including information on gestational age at birth, birth weight, and head circumference.	No adjustment for this group of comparison.
Kini (Phenytoin) (Controls exposed to Lamotrigine, sick), 2006	retrospective cohort	Structured interviews were conducted with the mothers and medical records were available to confirm details in about 80% of participants.	A clinical examination was carried out to check the growth parameters and detect dysmorphic features and malformations. Clinical photographs were reviewed based on whether the gestalt was suggestive of fetal anticonvulsant syndrome by a panel of blind dysmorphologists.	None.
Kini (Phenytoin) (Controls unexposed, sick), 2006	retrospective cohort	Structured interviews were conducted with the mothers and medical records were available to confirm details in about 80% of participants.	A clinical examination was carried out to check the growth parameters and detect dysmorphic features and malformations. Clinical photographs were reviewed based on whether the gestalt was suggestive of fetal anticonvulsant syndrome by a panel of blind dysmorphologists.	None.
Koch (Phenytoin), 1996	prospective cohort	The pregnant women were asked to continuously monitor their seizures, treatment and additional illnesses.	Authors developed their own neonatal score consisted of symptoms of apathy and hyperexcitability. Then six-year-old children were examined for major and minor neurological dysfunction with the examination of Touwen. The examiners were not blinded for this part of the study.	Matched for five variables, namely socioeconomic status (lower and middle class), age of the mother at delivery, parity, amount of smoking during pregnancy and number of abortions previous to the subject’s birth.
Lowe (Phenytoin) (Controls unexposed, disease free), 1973	population based cohort retrospective	Detailed information has been collected, coded, and computer stored. The hospital and general practitioner records of women with a history of epilepsy were traced and inspected.	Detailed information has been collected, coded, and computer stored whether the infant is observed to be congenitally malformed.	None.
Lowe (Phenytoin) (Controls unexposed, sick), 1973	population based cohort retrospective	Detailed information has been collected, coded, and computer stored. The hospital and general practitioner records of women with a history of epilepsy were traced and inspected.	Detailed information has been collected, coded, and computer stored whether the infant is observed to be congenitally malformed.	None.
Madley-Dowd_SE (Phenytoin) (Controls exposed to LTG) (Mixed indications), 2024	population based cohort retrospective	Self-reported maternal consumption of antiseizure medications were prospectively recorded at antenatal care visits (from a structured interview at the first visit at 8–10 weeks of gestation, supplemented from 2005 with drug all prescription dispensations in Sweden (tetrachoric correlation 0.98).	Inpatient and specialized outpatient diagnoses were obtained using ICD-8/9/10 codes from the National Patient Registry, and to further identify cases of ADHD, prescription information on licensed ADHD medications was also incorporated.	Adjusted for ASM presumed indication, any maternal neurodevelopmental diagnoses, maternal age, residential region, hazardous drinking or illicit drug use during pregnancy, gravidity, maternal health care use and seizure events (year before pregnancy), use of antipsychotics/ antidepressants (year before pregnancy), hyperemesis or antiemetic prescriptions, year of birth and socioeconomic position.
Madley-Dowd_SE (Phenytoin) (Controls unexposed, general pop) (Mixed indications), 2024	population based cohort retrospective	Self-reported maternal consumption of antiseizure medications were prospectively recorded at antenatal care visits (from a structured interview at the first visit at 8–10 weeks of gestation, supplemented from 2005 with drug all prescription dispensations in Sweden (tetrachoric correlation 0.98).	Inpatient and specialized outpatient diagnoses were obtained using ICD-8/9/10 codes from the National Patient Registry, and to further identify cases of ADHD, prescription information on licensed ADHD medications was also incorporated.	Adjusted for ASM presumed indication, any maternal neurodevelopmental diagnoses, maternal age, residential region, hazardous drinking or illicit drug use during pregnancy, gravidity, maternal health care use and seizure events (year before pregnancy), use of antipsychotics/ antidepressants (year before pregnancy), hyperemesis or antiemetic prescriptions, year of birth and socioeconomic position.
Madley-Dowd_SE (Phenytoin) (Controls unexposed, sibling) (Mixed indications), 2024	population based cohort retrospective	Self-reported maternal consumption of antiseizure medications were prospectively recorded at antenatal care visits (from a structured interview at the first visit at 8–10 weeks of gestation, supplemented from 2005 with drug all prescription dispensations in Sweden (tetrachoric correlation 0.98).	Inpatient and specialized outpatient diagnoses were obtained using ICD-8/9/10 codes from the National Patient Registry, and to further identify cases of ADHD, prescription information on licensed ADHD medications was also incorporated.	Adjusted for ASM presumed indication, any maternal neurodevelopmental diagnoses, maternal age, residential region, hazardous drinking or illicit drug use during pregnancy, gravidity, maternal health care use and seizure events (year before pregnancy), use of antipsychotics/ antidepressants (year before pregnancy), hyperemesis or antiemetic prescriptions, year of birth and socioeconomic position.
Madley-Dowd_UK (Phenytoin) (Controls exposed to LTG) (Mixed indications), 2024	retrospective cohort (claims database)	Prescriptions were identified from primary care records.	Diagnoses were obtained from primary care records using Read codes for all patients and from inpatient and outpatient Hospital Episode Statistics (HES) using ICD-10 codes. For ADHD, prescription on ADHD medications (methylphenidate, dexamfetamine, lisdexamfetamine and atomoxetine) was incorporated.	Adjusted for ASM presumed indication, any maternal neurodevelopmental diagnoses, maternal age, residential region, hazardous drinking or illicit drug use during pregnancy, gravidity, maternal health care use and seizure events (year before pregnancy), use of antipsychotics/ antidepressants (year before pregnancy), hyperemesis or antiemetics, year of birth and socioeconomic position.
Madley-Dowd_UK (Phenytoin) (Controls unexposed, general pop) (Mixed indications), 2024	retrospective cohort (claims database)	Prescriptions were identified from primary care records.	Diagnoses were obtained from primary care records using Read codes for all patients and from inpatient and outpatient Hospital Episode Statistics (HES) using ICD-10 codes. For ADHD, prescription on ADHD medications (methylphenidate, dexamfetamine, lisdexamfetamine and atomoxetine) was incorporated.	Adjusted for ASM presumed indication, any maternal neurodevelopmental diagnoses, maternal age, residential region, hazardous drinking or illicit drug use during pregnancy, gravidity, maternal health care use and seizure events (year before pregnancy), use of antipsychotics/ antidepressants (year before pregnancy), hyperemesis or antiemetic prescriptions, year of birth and socioeconomic position.
Madley-Dowd_UK (Phenytoin) (Controls unexposed, sibling) (Mixed indications), 2024	retrospective cohort (claims database)	Prescriptions were identified from primary care records.	Diagnoses were obtained from primary care records using Read codes for all patients and from inpatient and outpatient Hospital Episode Statistics (HES) using ICD-10 codes. For ADHD, prescription on ADHD medications (methylphenidate, dexamfetamine, lisdexamfetamine and atomoxetine) was incorporated.	Sibling. Adjusted for ASM presumed indication, any maternal neurodevelopmental diagnoses, maternal age, residential region, hazardous drinking or illicit drug use during pregnancy, gravidity, maternal health care use and seizure events (year before pregnancy), use of antipsychotics/ antidepressants (year before pregnancy), hyperemesis or antiemetics, year of birth and socioeconomic position.
Mawer (Phenytoin) (Controls exposed to Lamotrigine, sick), 2010	prospective cohort	The research nurse recorded personal and medical history with forms. An epilepsy specialist confirmed antiepileptic medication. Information from the patient was supplemented from clinical records.	Congenital abnormalities were identified from reports of fetal scans and paediatric medical examination of the baby before discharge. Major malformations were distinguished from minor anomalies using Eurocat instructions and those with chromosome abnormalities were not included.	No matching for this group of comparison. No adjustement.
Mawer (Phenytoin) (Controls unexposed, disease free), 2010	prospective cohort	The research nurse recorded personal and medical history with forms. An epilepsy specialist confirmed antiepileptic medication. Information from the patient was supplemented from clinical records.	Congenital abnormalities were identified from reports of fetal scans and paediatric medical examination of the baby before discharge. Major malformations were distinguished from minor anomalies using Eurocat instructions and those with chromosome abnormalities were not included.	Matched for age (within a five-year band), for parity, for residential district (postal code) and employment. No adjustement.
Mawer (Phenytoin) (Controls unexposed, sick), 2010	prospective cohort	The research nurse recorded personal and medical history with forms. An epilepsy specialist confirmed antiepileptic medication. Information from the patient was supplemented from clinical records.	Congenital abnormalities were identified from reports of fetal scans and paediatric medical examination of the baby before discharge. Major malformations were distinguished from minor anomalies using Eurocat instructions and those with chromosome abnormalities were not included.	No matching for this group of comparison. No adjustement.
Meador (Phenytoin), 2006	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records.	All reports of adverse events were reviewed and classified by the principal investigator, and written reports of physical abnormalities were assessed by a board-certified medical geneticist.	No adjustment for this group of exposure.
Meador (Phenytoin), 2009	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records.	Assessors who were unaware of drug exposure evaluated cognitive outcomes with the use of the Mental Developmental Index of the Bayley Scales of Infant Development, second edition (21 to 34 months of age) and Differential Ability Scales (33 to 45 months of age).	Adjustment for maternal IQ and age, antiepileptic-drug dose, infant’s gestational age at birth, and maternal preconception use of folate.
Meador (Phenytoin), 2011	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records.	Cognitive outcomes were evaluated by blinded assessors using the Differential Ability Scales, Preschool Language Scale (4th edition), Peabody Picture Vocabulary Test (fourth edition) and Developmental Test of Visual-Motor Integration (fifth edition).	Mean age 3 index scores adjusted for maternal IQ, maternal age, dose, race, alcohol use during pregnancy and folate.
Meador (Phenytoin), 2013	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records and assessed antiepileptic drug levels for 229 (75%) mothers.	Assessors, masked to treatment, evaluated cognitive outcomes with differential ability scales, Bayley scales (BSID), children’s memory scale, behavior rating inventory of executive function, NEPSY, expressive one-word picture vocabulary test and the developmental test of visual motor integration.	Mean IQ scores are adjusted for maternal IQ and sometimes age, dose, periconceptional folate, and gestational age at delivery. For verbal and non-verbal index: maternal IQ, maternal age, gestational age, and ethnic group. For the general memory index: for maternal IQ, dose, maternal age, and alcohol use. For NEPSY: maternal IQ, gestational age, and maternal age. Only for dose for BRIEF.
Miškov (Phenytoin) (Controls exposed to Lamotrigine, sick), 2016	prospective cohort	Data collected on pregnant women during outpatient visits to the neurologist, gynecologist and pediatrician neurologist.	Newborns were examined by pediatrician neurologist/geneticist at Clinical Department of Pediatrics and followed up yearly. Data on the control group were obtained from Clinical Department of Gynecology and Obstetrics and Clinical Department of Pediatrics .	Not specified.
Miškov (Phenytoin) (Controls unexposed, disease free), 2016	prospective cohort	Data collected on pregnant women during outpatient visits to the neurologist, gynecologist and pediatrician neurologist.	Newborns were examined by pediatrician neurologist/geneticist at Clinical Department of Pediatrics and followed up yearly. Data on the control group were obtained from Clinical Department of Gynecology and Obstetrics and Clinical Department of Pediatrics .	Not specified.
Miškov (Phenytoin) (Controls unexposed, sick), 2016	prospective cohort	Data collected on pregnant women during outpatient visits to the neurologist, gynecologist and pediatrician neurologist.	Newborns were examined by pediatrician neurologist/geneticist at Clinical Department of Pediatrics and followed up yearly. Data on the control group were obtained from Clinical Department of Gynecology and Obstetrics and Clinical Department of Pediatrics .	Not specified.
Morrow (Phenytoin) (Controls exposed to Lamotrigine, sick), 2006	prospective cohort	Information was collected at registration and changes of antiepileptic drugs during pregnancy were detected during the follow-up duration by sending a standardised questionnaire to the patient's general practitioner. Other health care practitioners were contacted if identified.	Outcome data were collected by sending the patient’s general practitioner a standardised questionnaire for completion during the follow-up duration. Other health care practitioners were contacted if identified. Pregnancy losses with no major malformation are excluded.	No adjustment for this group of comparison.
Morrow (Phenytoin) (Controls unexposed, sick), 2006	prospective cohort	Information was collected at registration and changes of antiepileptic drugs during pregnancy were detected during the follow-up duration by sending a standardised questionnaire to the patient's general practitioner. Other health care practitioners were contacted if identified.	Outcome data were collected by sending the patient’s general practitioner a standardised questionnaire for completion during the follow-up duration. Other health care practitioners were contacted if identified. Pregnancy losses with no major malformation are excluded.	No adjustment for this group of exposure.
Nulman (Phenytoin) (Controls unexposed, NOS) (Mixed indications), 1997	prospective cohort	At the first visit during the first trimester of pregnancy was recorded time and dose of the antiepileptic drugs before and during pregnancy.	Details were retrieved from the medical records. More informations were obtained from the mothers and by a report from the children’s pediatricians. Morphological testing were studied by a team of blinded pediatricians, neurologists and geneticists.	Matched on age, gravidity, parity, and socioeconomic class.
Nulman (Phenytoin) (Controls unexposed, sick) (Mixed indications), 1997	prospective cohort	At the first visit during the first trimester of pregnancy was recorded time and dose of the antiepileptic drugs before and during pregnancy.	Details were retrieved from the medical records. More informations were obtained from the mothers and by a report from the children’s pediatricians. Morphological testing were studied by a team of blinded pediatricians, neurologists and geneticists.	No matching for this group of comparison.
Pennell (Phenytoin), 2012	prospective cohort	They obtained information from the patients, their seizure or medication diary, or review of their medical records.	Outcomes measured at birth and later ages by the labor and delivery medical team and the pediatrician, as per local standard clinical practice. Outside records were obtained for head circumference and weight. Standardized Apgar scores are measured at one and five minutes.	Controlling for the covariates tobacco use, gestational diabetes, and gestationnal age.
Samrén (Phenytoin), 1999	retrospective cohort	Data were collected from medical records and include medication. The prescribed dose of thedrugs was also retrieved from obstetric files.	Data were collected from medical records and include information on pregnancy and child. Information on major congenital abnormalities was completed with information from the pediatrician whenever necessary.	Matched for age (±2 years) and parity of the mother, and sex, birth year, and hospital of delivery of the child.
Sonneveld (Phenytoin) (Controls unexposed NOS), 1990	prospective cohort	Since 1981, drugs ingested during pregnancy have been included in the information collected.	Propspective information was available for assessment.	None.
Sonneveld (Phenytoin) (Controls unexposed, sick), 1990	prospective cohort	Since 1981, drugs ingested during pregnancy have been included in the information collected.	Propspective information was available for assessment.	None.
Steegers-Theunissen (Phenytoin), 1994	prospective cohort	The use of antiepileptic drugs was checked in each woman by measurement of the blood concentration. Data such as antiepileptic drug treatment were collected. The consulting doctors were finally contacted to obtain additional information on treatment regimen.	The infants were examined systematically for major malformations according to the ICD9 British Paediatric Association System and for minor malformations according to Mehes and Stalder by a trained research fellow.	None.
The NAAED (Phenytoin) (Controls exposed to LTG) (Indications NOS), 2023	prospective cohort	Women are interviewed at enrollment, at 7 months’ gestation and at 8 –12 weeks after the expected date of delivery. The computer-assisted interviews include questions on start and stop dates of each antiepileptic drugs taken, dose, frequency and changes in medication.	Women were questioned with a computer-assisted telephone interview 3 times and medical records were obtained. The written descriptions in the pediatricians’ examinations were reviewed separately by the clinical teratologist.	No adjustment for this group of comparison.
The NAAED (Phenytoin) (Controls unexposed, disease free) (Indications NOS), 2023	prospective cohort	Women are interviewed at enrollment, at 7 months’ gestation and at 8 –12 weeks after the expected date of delivery. The computer-assisted interviews include questions on start and stop dates of each antiepileptic drugs taken, dose, frequency and changes in medication.	Women were questioned with a computer-assisted telephone interview 3 times and medical records were obtained. The written descriptions in the pediatricians’ examinations were reviewed separately by the clinical teratologist.	Not specified.
Thomas (Phenytoin) (Controls exposed to Lamotrigine, sick), 2021	prospective cohort	Women were instructed to record the use of the antiepileptic drugs on a daily basis in the pregnancy diary that was given to them.	Screening for major congenital malformation with antenatal screening by serum alpha fetoprotein estimation and detailed anomaly scan between 12 and 18 weeks of pregnancy; physical examination at birth, at 3 months of age (at least) and at 1 year of age if possible.	No adjustment for this group of comparison.
Thomas (Phenytoin) (Controls unexposed, disease free), 2021	prospective cohort	Women were instructed to record the use of the antiepileptic drugs on a daily basis in the pregnancy diary that was given to them.	Screening for major congenital malformation with antenatal screening by serum alpha fetoprotein estimation and detailed anomaly scan between 12 and 18 weeks of pregnancy; physical examination at birth, at 3 months of age (at least) and at 1 year of age if possible.	None.
Thomas (Phenytoin) (Controls unexposed, sick), 2021	prospective cohort	Women were instructed to record the use of the antiepileptic drugs on a daily basis in the pregnancy diary that was given to them.	Screening for major congenital malformation with antenatal screening by serum alpha fetoprotein estimation and detailed anomaly scan between 12 and 18 weeks of pregnancy; physical examination at birth, at 3 months of age (at least) and at 1 year of age if possible.	Adjusted for age and educational status of mother, and epilepsy class.
Thomas a (Phenytoin), 2008	nested case control	The maternal use of antiepileptic drugs is recorded in the protocol forms from the month prior to the last menstrual period through the entire pregnancy and postpartum period on a monthly basis.	A cardiologist blinded to antiepileptic drugs exposure of the infants carries out a clinical examination and echocardiogram on all infants. A second cardiologist confirmed malformation whenever required.	None.
Thomas a (Phenytoin), 2022	prospective cohort	The details of Antiseizure medications exposure were extracted from the clinical records in the registry.	Assessment of intelligence and language was performed with the Wechsler Intelligence Scale for Children (WISC 4) or Adult (if > 18 years) and clinical examination of language fundamentals (CELF 4). The assessors were blinded to the details of the antenatal antiseizure medications exposure.	None.
Thomas b (Phenytoin) (Epilepsy) (Controls exposed to LTG), 2022	prospective cohort	Each pregnant woman in the Kerala Registry of Epilepsy and Pregnancy (KREP) maintained a pregnancy diary to record the daily use of antiseizure medications (ASMs). Data were then transferred to the clinical records of the registry during their clinic visits.	Developmental paediatricians and developmental therapists, who were blinded to antenatal ASM exposure evaluated motor and mental development of infants (>12 months) using the Developmental Assessment Scale for Indian Infants (an Indian adaptation of the Bayley Scale of Infant Development).	None (the adjusted motor and mental development quotients (aMODQ and aMEDQ) not provided, only their difference from the reference group (unexposed infants)).
Thomas b (Phenytoin) (Epilepsy) (Controls unexposed, sick), 2022	prospective cohort	Each pregnant woman in the Kerala Registry of Epilepsy and Pregnancy (KREP) maintained a pregnancy diary to record the daily use of antiseizure medications (ASMs). Data were then transferred to the clinical records of the registry during their clinic visits.	Developmental paediatricians and developmental therapists, who were blinded to antenatal ASM exposure evaluated motor and mental development of infants (>12 months) using the Developmental Assessment Scale for Indian Infants (an Indian adaptation of the Bayley Scale of Infant Development).	None (the adjusted motor and mental development quotients (aMODQ and aMEDQ) not provided, only their difference from the reference group (unexposed infants)).
Titze (Phenytoin) (Controls unexposed, disease free), 2008	prospective cohort	The pregnant women were asked to continuously monitor their treatment.	In adolescents, the German versions of the revised Wechsler Intelligence Scales for children and for adults were applied.	Matched for socioeconomic status (SES), nicotine consumption in the last trimester, maternal age at birth, parity, and number of previous abortions. No adjustment for this group of exposure.
Titze (Phenytoin) (Controls unexposed, sick), 2008	prospective cohort	The pregnant women were asked to continuously monitor their treatment.	In adolescents, the German versions of the revised Wechsler Intelligence Scales for children and for adults were applied.	No matching for this group of comparison. No adjustment for this group of exposure.
Tomson (Phenytoin), 2018	prospective cohort	Information obtained during early pregnancy and follow-up by the treating physician who transferred it online to a EURAP national coordinator. He then, reviewed the reports for completeness and accuracy before transmission to the EURAP central database.	Abnormalities in the offspring were reported descriptively by enrolling physicians These reports were reviewed and classified (2005 EUROCAT criteria) by an outcome committee unaware of the type of exposure. Supplementary information from the reporting physician can be request.	None.
Trivedi (Phenytoin) (Controls exposed to Lamotrigine, sick), 2018	prospective cohort	Women were instructed to record the use of the antiepileptic drugs on a daily basis in the pregnancy diary that was given to them.	Women were under regular antenatal follow up by the obstetricians.	No adjustment for this group of comparison.
Trivedi (Phenytoin) (Controls unexposed, sick), 2018	prospective cohort	Women were instructed to record the use of the antiepileptic drugs on a daily basis in the pregnancy diary that was given to them.	Women were under regular antenatal follow up by the obstetricians.	Adjusted for enrollment status, age at conception, type of epilepsy, folic acid use in pregnancy, seizures in pregnancy, Generalised Tonic Clonic Seizures in pregnancy, prior spontaneous fetal loss, parental major congenital malformation.
Vajda (Phenytoin) (Controls exposed to Lamotrigine, sick), 2013	prospective cohort	Details are obtained by four in-depth telephone interviews with the women. Details provided by the participants are confirmed, as far as possible, by the treating medical practitioners.	Details are obtained by four in-depth telephone interviews with the women. Details provided by the participants are confirmed, as far as possible, by the treating medical practitioners.	None.
Vajda (Phenytoin) (Controls unexposed, sick), 2013	prospective cohort	Details are obtained by four in-depth telephone interviews with the women. Details provided by the participants are confirmed, as far as possible, by the treating medical practitioners.	Details are obtained by four in-depth telephone interviews with the women. Details provided by the participants are confirmed, as far as possible, by the treating medical practitioners.	None.
Vajda a (Phenytoin) (Controls exposed to LTG) (Epilepsy), 2024	prospective cohort	Not specified. All contact between the pregnant women and the Melbourne-based register has been by telephone.	The register has collected data concerning each woman’s medical details at enrolment, at about 28 WG and around the end of the first post-partum month. The accuracy of the information provided by the pregnant women was checked with their treating medical practitioners, as far as feasible.	Exclusion of previous ‘lost’ pregnancies and pregnancies where either parent had a foetal malformation.
Vajda a (Phenytoin) (Controls unexposed sick) (Epilepsy), 2024	prospective cohort	Not specified. All contact between the pregnant women and the Melbourne-based register has been by telephone.	The register has collected data concerning each woman’s medical details at enrolment, at about 28 WG and around the end of the first post-partum month. The accuracy of the information provided by the pregnant women was checked with their treating medical practitioners, as far as feasible.	Exclusion of previous ‘lost’ pregnancies and pregnancies where either parent had a foetal malformation.
Vajda b (Phenytoin) (Epilepsy) (Controls exposed to Lamotrigine, sick), 2024	prospective cohort	Details are obtained by four in-depth telephone interviews with the women (at time of enrolment; at approximately 28 weeks of pregnancy; after the 1st post-partum (pp) month, and at one year pp). The accuracy of the information provided by the pregnant women is checked with their treating doctors.	Details are obtained by four in-depth telephone interviews with the women (at time of enrolment; at approximately 28 weeks of pregnancy; after the 1st post-partum (pp) month, and at one year pp). The accuracy of the information provided by the pregnant women is checked with their treating doctors.	None.
Vajda b (Phenytoin) (Epilepsy) (Controls unexposed, sick), 2024	prospective cohort	Details are obtained by four in-depth telephone interviews with the women (at time of enrolment; at approximately 28 weeks of pregnancy; after the 1st post-partum (pp) month, and at one year pp). The accuracy of the information provided by the pregnant women is checked with their treating doctors.	Details are obtained by four in-depth telephone interviews with the women (at time of enrolment; at approximately 28 weeks of pregnancy; after the 1st post-partum (pp) month, and at one year pp). The accuracy of the information provided by the pregnant women is checked with their treating doctors.	None.
Viinikainen (Phenytoin) (Controls unexposed, disease free) a, 2006	prospective cohort	Obstetric data derived from the pregnancy registry and supplemented with medical records. And plasma levels of the antiepileptic drugs were monitoring during pregnancy at each control visit.	Reviewed through pregnancy charts, birth reports, medical records, including discharge notes of the pediatrician and control visits.	No adjustment for this group of exposure.
Viinikainen (Phenytoin) (Controls unexposed, sick) a, 2006	prospective cohort	Obstetric data derived from the pregnancy registry and supplemented with medical records. And plasma levels of the antiepileptic drugs were monitoring during pregnancy at each control visit.	Reviewed through pregnancy charts, birth reports, medical records, including discharge notes of the pediatrician and control visits.	No adjustment for this group of comparison.
Vinten (Phenytoin), 2005	retrospective cohort	The women recruited were initially interviewed to ascertain information including the antiepileptic drug dose throughout the pregnancy. Clinical records were used to confirm the medical information.	Each child received a physical and neurologic examination. The Wechsler Intelligence Scale for Children III was administered.	None.
Waters (Phenytoin) (Controls unexposed, disease free), 1994	prospective cohort	An attending neurologist took a history, performed an examination, and monitored the administration of antiepileptic drugs throughout the pregnancy. Documented at each visit on a standard form.	Infants underwent a standardized examination by a physician and/or nurse trained to recognize major and minor anomalies. Data from the obstetric clinic and the infants' delivery charts.	None.
Waters (Phenytoin) (Controls unexposed, sick), 1994	prospective cohort	An attending neurologist took a history, performed an examination, and monitored the administration of antiepileptic drugs throughout the pregnancy. Documented at each visit on a standard form.	Infants underwent a standardized examination by a physician and/or nurse trained to recognize major and minor anomalies. Data from the obstetric clinic and the infants' delivery charts.	None.
Wide (Phenytoin), 2000	prospective cohort	Mothers attended the outpatient clinic for pregnant women with epilepsy at the department of neurology where drug plasma levels were determined at monthly intervals.	One of the participating paediatricians organize the examination of the newborn infant within 4 days of birth. A blind child psychologist performed the Griffiths’ test, in the child’s home at 9 months after birth. Data on birthweight were obtained from medical records.	Matched for gestational age and mode of delivery. Sex-matched control infants were found for a majority of subjects.
Yerby (Phenytoin), 1992	prospective cohort	Women with epilepsy had a determination of antiepileptic drug concentrations. They were scheduled for monthly visits and with their babies two times post partum	The infants of both case and control mothers were examined by a blinded developmental pediatrician. A checklist for major and minor malformations was used, and facial measurements were taken.	None.

master protocol