The easy misuse of antidepressants during pregnancy is depressing

Bénard-Laribière et al. must be commended for their detailed monitoring of antidepressant prescriptions during pregnancy 1, in a Journal that has previous creditable pieces including warnings about malformations with paroxetine and fluoxetine 2. However, we disagree with their views on potential undertreatment and the over-reassuring conclusion that treatment has been especially adapted in accordance with existing recommendations in a ‘large proportion’ of pregnant women. Firstly, all prescribers should always prioritize pregnancy and drug safety. However, data to ascertain the benefit:harm ratio or guide dosing are scarce. Even when available, such data are often overlooked by prescribers 3. Very few drug registries exist to report postmarketing harms. The European Registry of Antiepileptic Drugs and Pregnancy (EURAP) is among notable exceptions. Secondly, in contrast to the authors' beliefs 1, first-line treatment should always be cognitive behavioural therapy (CBT) – even for severe depression 4. The most serious adverse effects of antidepressants, such as malformations 2 and postpartum haemorrhage 5, cannot be overlooked. Almost all antidepressant prescriptions in this particular observational series may be first-line treatments as psychotherapies are not reimbursed by the mandatory and affluent French health care system. France has overlooked evidence underpinning the Improving Access to Psychological Therapies programme launched in 2008 in England. This beacon, offering largely CBT approaches to people with anxiety or depressive diagnoses, has also been implemented in Quebec and Belgium. Limiting prescriptions to medicines only ties doctors' hands to the patients' detriment. Thirdly, citalopram and escitalopram ranked first for prevalence 1 despite no clinical benefit compared to other antidepressants, and only robust evidence of harms such as deadly Torsades de Pointe. These surely cannot be ‘the drugs of choice in pregnancy’? 1 Indeed, vomiting is common and hypokalaemia, a major risk factor for QT prolongation, is observed in hyperemesis 6. Why is duloxetine ever prescribed 1 when its specific life-threatening serious effects include liver injury and skin reactions including Stevens–Johnson syndrome? Could the authors direct us to any recommendation justifying paroxetine or fluoxetine use and what kind of information women in this series would have received about malformations? 1 It appears no lessons have been learnt from the valproate scandal 6. The first report of congenital malformation due to valproate was made by a French team in 1982. No action was taken until 2015 when a legal case went public thanks to a mother whose two children were consecutively damaged and who created the Association of Parents of Children with Anticonvulsant Syndrome. This association successfully campaigned to restrict prescription of valproate and its by-products in women of childbearing age to specialists who must obtain signed consent 7. They also achieved mandatory pictograms for every medicine with potential harm during planned or unplanned pregnancy and targeted women of childbearing age as restrictions were poorly implemented. Fourthly, we must all be mindful that selective publication, sponsorship bias and ghost-writing have been the norm. GlaxoSmithKline even cynically named their paroxetine programme ‘CASSPER’ after the friendly ghost. The effect size of antidepressants is small or modest (around 0.30), and psychotherapies are at least as effective for severe depression in the real-life setting and long term without adverse effects 8. Could Bénard-Laribière et al. provide further data about other major issues during pregnancy: the prescription of folate and alcohol consumption – itself a frequent cause of depression especially in high consumption France. Getting it right for pregnant women helps everyone. How can pharmacovigilance systems protect sick people when they appear to have so few concerns, monitoring and controls for the most vulnerable unborn future citizens. There are no competing interests to declare.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.