Good news – it seems that many of the problems initially thought to be related to taking antidepressants during pregnancy (cardiac problems, autism) are, in fact, probably due to risk factors related to the underlying illness. Since psychiatric illness is not only unpleasant and risky for the woman but also somewhat risky to the fetus, keeping it controlled should be the priority during pregnancy, rather than trying to avoid medication.
Ideally, a woman should be stabilised on one of the better-known, lower-risk drugs for at least three months before becoming pregnant. In practice, of course, this is often not going to happen. When a woman who is taking a psychotropic drug becomes pregnant, the temptation is going to be to switch her onto one of the drugs that has the best evidence profile for safety, but in fact this may make matters worse; it means that not only do we risk triggering a relapse in the mother, but we expose the fetus to more medications plus potentially to the effects of psychiatric relapse on the fetus. If a woman who is stable on medication becomes pregnant, therefore, it’s usually best to stick to whatever she’s on.
Use of a single drug at a higher dose is considered preferable to use of multiple drugs at lower doses (Level C evidence – consensus and expert opinion)
Some general risks of antidepressants
- Miscarriage: The article claimed research to have shown AD use to be associated with some increase in the miscarriage rate. I have to say that, from looking up their references, I’m not convinced that this isn’t the same sort of confounded result we previously had with other problems as per the first sentence of this post; at least one analysis of multiple studies found no increase in risk. One study that did find a risk and that broke it down by separate ADs found it to be associated with paroxetine & venlafaxine.
- Premature delivery: Again, there may be some increase in risk or it may be due to confounders. One study found average decrease in gestational age to be only a few days (statistically significant, but uncertain whether clinically so).
- Poor neonatal adaptation syndrome: SSRI and SNRI exposure in the third trimester can be associated with a group of problems including respiratory distress (cyanosis & apnoea), seizures, temperature instability, feeding difficulty and vomiting, hypoglycaemia, hypo- or hypertonia, hyperreflexia, tremor, jitteriness, irritability and constant crying. While all this sounds alarming, symptoms are pretty mild in most cases. It isn’t clear whether or not it’s advisable to taper ADs in the third trimester.
General comments on antipsychotics
- Untreated antipsychotic illness is much more of a risk to a pregnancy than antipsychotic use (this includes risk of suicide or infanticide).
- Placental transfer is lowest for quetiapine, risperidone, haloperidol, and olanzapine.
- CYP1A2 enzymes are downregulated, which means in plain English that olanzapine and clozapine may be metabolised more slowly and therefore the dose may need to be decreased. Similarly, some drugs may need to have their dose increased due to upregulation of the enzymes metabolising them, though no examples were given.
- Many of the antipsychotics – particularly second-generation – are associated with increased risk of gestational diabetes and increased infant birthweight. Ultrasound monitoring of fetal size in late pregnancy may be worthwhile.
- There have been reports of restlessness, dystonia, hypertonia, and tremor in infants who have had in utero exposure to antipsychotics.
- Current research indicates that there may be neurodevelopmental delays in 6-month-olds exposed to second-generation antipsychotics in utero, but these delays are no longer evident at 12 months.
- Some research has been done on in utero exposure to first-generation antipsychotics, and this has found no difference in IQ or behavioural functioning at 5 years.
Specific risks of particular drugs
- Lamotrigine: Conflicting results on whether or not this causes oral cleft defects. Main problem with lamotrigine is that pregnancy may cause the levels to decrease and these should be monitored and adjusted accordingly. Overall, the evidence for lamotrigine’s safety is increasing, and considering that the alternatives for stabilising BPD are worse overall (see below) this is often the safest option.
- Valproate: This is associated with a high risk of malformations, although, as one rapid response pointed out, this risk has been found in studies done on women taking it for epilepsy and the same problem with the illness itself as a confounder is a possibility and hasn’t been allowed for in studies. Two studies have also linked it with autism. Women on valproate should take high doses of folic acid during pregnancy, have their blood levels monitored, and be advised to have a 2nd-trimester ultrasound to screen for major deformity.
- Carbamazepine: Increased risk of malformations; neural tube defects, facial and skeletal abnormalities, hypospadias, and diaphragmatic hernia. It may also increase the risk of neonatal haemorrhage. Best avoided where possible. If used, same advice as for valproate – high-dose folic acid, monitor levels, screen for fetal abnormality.
- Lithium: Increased risk of Ebstein’s abnormality (a cardiac defect), but this occurs in only about one in 1 – 2000 children exposed. There are also reports of perinatal toxicity – hypotonia, cyanosis, goitre and diabetes insipidus. One big issue with lithium in pregnancy, as highlighted by one of the rapid responders, is that the changes in circulating volume during pregnancy and then delivery can play havoc with lithium levels; the increased volume in pregnancy can lead to subtherapeutic levels, and the blood loss at delivery can lead to toxicity. Lithium levels should therefore be closely monitored in pregnancy (it doesn’t say how often) and when labour begins the dose should be stopped or reduced, the woman should remain well hydrated during labour, and after birth the dosage should be reduced to pre-pregnancy levels with, again, close monitoring.
- Paroxetine: Should be avoided if possible (Level B – limited or inconsistent scientific evidence.) If women do take paroxetine in early pregnancy, consider fetal echocardiography.
- Benzodiazepines: Seems to be a 1 in 10 000 risk of oral cleft with benzodiazepine use in pregnancy (Level B evidence again – more recent studies haven’t backed this up) Use shortly before delivery is associated with floppy infant syndrome (this one is Level A). There is also some evidence that a benzodiazepine + SSRI combo may increase the risk of hart defects.
- Gabapentin: One study found this to be associated with increased risk of premature birth (about 6.5% AR), low birth weight (about 6% AR), and SCBU admission (about 35% AR with use in late pregnancy). However, that plus several other studies have found no increased risk of major congenital malformations.
- Zolpidem: If used for >90 days, associated with increased risk low birth weight.