Valproate: knowledge of the risks associated with taking during pregnancy - Results of the 2018 FAMHP survey

date: 25/06/2019

In the context of pharmacovigilance activities, the FAMHP performed a survey of users of valproate in order to assess their knowledge of the risks associated with taking this medicinal product during pregnancy. The aim was to assess the new prevention measures implemented in Belgium in 2018, following the recommendations of the EMA (European Medicines Agency).

Valproate is a medicinal product indicated for the treatment of epilepsy and bipolar disorders. It is also prescribed for migraine prophylaxis, although this is not an indication that is approved in Belgium.

If taken during pregnancy, valproate significantly increases the risk of birth defects and developmental disorders in the unborn child. The use of valproate should therefore be avoided during pregnancy, unless there is no appropriate alternative treatment.

Valproate is marketed in Belgium under the names CONVULEX, DEPAKINE, VALPROATE EG, VALPROATE MYLAN and VALPROATE SANDOZ.

Results of the 2018 survey                                                                                                       

The survey was carried out at the end of 2018, a little after implementation of the new prevention measures.

Three thousand five hundred randomly drawn patients aged between 18 and 49, who were using valproate in 2017, were invited to participate in the survey. The FAMHP analysed 693 responses.

Finding 1: use of valproate for a condition that is not among the indications approved in Belgium.

Although epilepsy is the main reason for treatment with valproate, a third of respondents cited migraine prophylaxis. Across all of the respondents, only 61 % of the reasons given corresponded to an indication approved in Belgium (epilepsy or bipolar disorders). This proportion did not vary significantly according to age group.

Finding 2: 27 % of the respondents were not aware of the teratogenic and developmental risks to the unborn child.

The youngest women were significantly better informed than the older ones.

The condition justifying taking valproate had little influence on the proportion of women who were informed: 77 % in the case of epilepsy, 69 % for bipolar patients and 73 % for patients suffering from migraines.

Finding 3: doctors (general practitioners and specialists) and pharmacists were the primary source of information.

Three quarters of the respondents who had been informed of the risks of valproate for the unborn child mentioned at least one healthcare professional, namely a doctor or pharmacist, as their source of information. Neurologists (47 %) and general practitioners (33 %) were the healthcare professionals most often mentioned, while pharmacists were mentioned by 16 % of respondents. However, 11.5 % of respondents had only been informed by the media. This proportion was significantly higher in women of 40-49 years of age (15 %) and higher in French-speaking (26 %) than in Flemish-speaking respondents (3 %).

Various educational materials are available for patients and healthcare professionals, in relation to the prescription and dispensing of valproate-based medicinal products. Among the women informed of the teratogenic risks of valproate, 8.4 % had received the patient information leaflet and only 4.7 % had signed the acknowledgement of receipt for the annual risk information bulletin. These documents, downloadable from the FAMHP website, were updated in November 2018, but had already been available since 2015.

Of the 693 people who responded to the survey, 33 % of patients indicated that they had discussed contraception with their doctor or pharmacist, 38 % mentioned a discussion on the subject of risks for the unborn child. The risk associated with abrupt interruption of a course of treatment with valproate had been discussed with 80 % of the patients.


This first survey shows that some valproate users of childbearing age had not been informed of the risks of the treatment in the event of pregnancy. A significant share of these women were using valproate for a condition not corresponding to the indications approved in Belgium for the medicinal product; essentially for migraines, for which safer alternatives are available.

Finally, the educational material provided has not been sufficiently distributed and used. A report with the detailed results of this survey will be available in autumn 2019.

Measures taken and to be taken

In order to prevent exposure to valproate, the EMA published new recommendations in May 2018, consisting among other things of prohibiting the use of valproate during pregnancy for the prophylaxis of migraine and bipolar disorders. This prohibition also applies to the treatment of epilepsy, except in the absence of any effective therapeutic alternative.

Furthermore, valproate should not be used by women and girls of child bearing age unless a strict pregnancy prevention programme is in place. These new measures are set out in detail on the EMA website and in the Summary of Product Characteristics (SPC) and notices, as well as in various educational materials intended for healthcare professionals and patients. A pictogram representing teratogenicity has also been applied to the packaging of valproate-based medicinal products since April 2018, helping to warn of the risks associated with use during pregnancy.

The FAMHP envisages new measures to limit valproate exposure in unborn children and raise awareness in its users of the risks of the medicinal product in the event of pregnancy. In order to measure the progress achieved, a new survey will be carried out in 2022.

The FAMHP thanks all of the patients who responded to this survey, as well as the mutual companies, the National Institute for Health and Disability Insurance (INAMI) and the various partners who have contributed to this survey.

Last updated on 11/07/2019