Antiepileptic drugs (AEDs) are not indicated for use in pregnancy. However, women with epilepsy, and other approved indications including bipolar disorder, may require or be unintentionally exposed to AEDs during pregnancy. Prior to an AED being marketed there are few data available on drug safety in pregnancy: data from animal models may not translate directly to humans and pregnant women are routinely excluded from clinical trials. The International Lamotrigine Pregnancy Registry was established by GlaxoSmithKline (GSK) in 1992 to monitor the safety of lamotrigine during pregnancy.

The International Lamotrigine Pregnancy Registry aims to assess whether there is a substantial increase in the risk of major congenital malformations (MCMs) following in utero exposure to lamotrigine. Exposure during the first trimester is of primary interest as this represents the period of organogenesis. The Registry is a primarily prospective enrolment and follow-up study. Patients exposed to lamotrigine during pregnancy are enrolled, on a voluntary basis, by their healthcare provider. Enrolment is encouraged early in pregnancy and if possible prior to any prenatal testing. The healthcare provider provides initial information concerning patient demographics; details of the pregnancy including the estimated delivery date and results of any prenatal testing; and the timing, duration and dosage of lamotrigine exposure in pregnancy. The registry accepts exposure reports from anywhere in the world. Within the United States (US) the healthcare provider can contact the registry using a toll free number. Outside of the US enrolments are made through the GlaxoSmithKline local operating company.

Close to the estimated date of delivery the healthcare provider is contacted by the Registry to provide follow up information concerning the pregnancy outcome (live or still birth, spontaneous or induced abortion), the presence or absence of a MCM, and the history of exposure to lamotrigine as well other antiepileptics and antipsychotics during pregnancy. Up to six attempts are made to contact the healthcare provider to obtain pregnancy outcome information. After six attempts, the record is closed as lost to follow up. Pregnancy outcomes are classified as outcomes with MCMs, outcomes without MCMs and spontaneous abortions. The outcomes with and without MCMs are further classified as live births, stillbirths/fetal deaths and induced abortions. Spontaneous abortions are reported separately due to potential for inconsistent identification of malformations in that situation.

It is beyond the scope of the Registry to consistently access pediatric evaluations and medical records. For this reason the main outcome is restricted to major congenital malformations that are external and recognizable in the delivery room or shortly after birth. To provide consistency, reported congenital malformations are classified as major or minor according to criteria used by the Centers for Disease Control and Prevention (CDC)'s Metropolitan Atlanta Congenital Defects Program (MACDP). All malformation reports are reviewed and classified by a paediatrician from the CDC and further information is requested as necessary.

Analyses are restricted to prospectively enrolled pregnancies (enrolment prior to knowledge of the birth outcome). Retrospectively enrolled pregnancies are reviewed for patterns of defect types, but are not included in formal analyses as retrospective reporting can be biased towards more unusual and severe outcomes and are less likely to be representative of the general population experience. The proportion of infants with MCMs among prospectively reported exposures is calculated as: the total number of outcomes with major birth defects (number of outcomes with major birth defects + the number of live births without defects).

Chromosomal malformations are reported descriptively, but are not included in the numerator as it is very unlikely that they are associated with drug exposure during pregnancy. All spontaneous pregnancy losses, as well as induced abortions and fetal deaths without reported defects, are excluded from the denominator due to the potential for inconsistent identification of malformations in those situations. The 95% confidence intervals (CIs) for risk estimates are calculated using exact methods based on the binomial distribution.

Analyses are stratified according to trimester of exposure (with the second trimester starting at week 14 and the third trimester at week 28 of gestation) and by exposure group (lamotrigine monotherapy, lamotrigine polytherapy including valproate and lamotrigine polytherapy without valproate).

The registry does not have an internal comparator group, but descriptive comparisons are made with MCM rates from general population studies in the literature and from other ongoing AED pregnancy registries. Prospective reports are also reviewed to detect any unusual patterns of malformation types that may warrant further investigation. The data from the International Lamotrigine Pregnancy Registry are reviewed, and conclusions developed, by an independent scientific advisory committee. A semi-annual interim report summarizing aggregate data is provided to disseminate information on a regular basis.

Inclusion Criteria:

• Women exposed in utero to lamotrigine (as monotherapy or in a polytherapy combination) during pregnancy. Exposure can occur at any time during pregnancy, though exposure in the first trimester is of primary interest.
• Pregnancies exposed to lamotrigine and reported before the outcome of the pregnancy is known (prospective reporting). Ideally exposed pregnancies are registered prior to prenatal testing, but only those pregnancies enrolled after prenatal testing has diagnosed a birth defect are excluded.
• Retrospectively reported exposures (i.e. exposures registered once the pregnancy outcome is known) are included in the registry, but are considered descriptively and are not included in risk analyses.

Exclusion Criteria:

• Retrospectively reported exposures (i.e. exposures registered once the pregnancy outcome is known) are included in the registry, but are reviewed separately and descriptively. These are not included in risk analyses.
• Patient reported exposures and outcomes that are not verified by a healthcare provider.

Women exposed to lamotrigine during pregnancy anywhere in the world.

This was a prospective enrollment/follow-up study based on exposure/outcome data collected using structured enrollment and birth outcome report forms completed by healthcare providers of pregnant women. This observational study was designed to investigate birth defects, a specific group of serious adverse events (SAEs); no other SAEs were measured.