Letter: A

Therapeutic amfetamines, including lisdexamfetamine and dexamfetamine, are licensed to treat attention deficit hyperactivity disorder (ADHD).

Untreated ADHD in the general population can increase the likelihood of risk-taking behaviour, including smoking and alcohol or recreational drug use. In addition, ADHD often manifests as anxiety-type behaviours in females; therefore, the continued treatment of ADHD in pregnancy may be beneficial for some patients. An individualised discussion of the risks and benefits of continued treatment is highly recommended.

Five studies describe approximately 6,200 unique therapeutic amfetamine-exposed pregnancies. Where details were provided, most of the exposures were documented to have occurred in early pregnancy only; as such, data regarding exposure to therapeutic amfetamines in the later stages of pregnancy are lacking.

The available safety data, although somewhat limited, have not associated use of therapeutic amfetamines in early pregnancy with an increased risk of congenital major malformation overall, cardiac malformation overall or any specific cardiac malformation. Use in the first half of pregnancy has not been associated with increased risks of perinatal death, impacts on birth weight outcomes (such as low birth weight, high birth weight or small for gestational age). However, the risk of growth impairment with use of therapeutic amfetamines in the later stages of pregnancy is unknown. Given that amfetamines can cause vasoconstriction, there is a theoretical risk of impaired placental perfusion, with possible impacts on fetal growth.

A large population-based and methodologically robust study has suggested a small increased risk of preterm delivery following maternal therapeutic amfetamine use in the first half of pregnancy. The findings from this study suggest an absolute risk of approximately 10%, relative to a background risk of 7.5%. There are methodologically limited data which suggest that the absolute risk may be further raised with continued exposure in the later stages of pregnancy (up to approximately 12%).

As therapeutic amfetamines are centrally acting stimulants, a possibility of neonatal withdrawal and associated complications may exist. Possible associations with placental abruption, pre-eclampsia, low 1-minute Apgar scores, and the requirement for neonatal resuscitation and neonatal admission have been described following maternal use in pregnancy. Pregnancies with exposure to CNS-acting medication in the later stages of gestation should ideally be delivered in a unit with adequate neonatal facilities to monitor for neonatal complications.

No studies investigating the risk of miscarriage, neurodevelopmental impairment or childhood cancer have been located in the literature. Animal studies of intrauterine amfetamine exposure have described learning and memory deficits and altered locomotor activity in the offspring. However, it is unclear how accurately these studies predict risks for human pregnancy exposures.

Exposure to therapeutic amfetamines at any stage in pregnancy would not usually be regarded as medical grounds for termination of pregnancy. Owing to a lack of sufficient safety data, additional fetal monitoring, particularly of fetal growth, may be warranted on a case-by-case basis. Other risk factors may be present in individual cases which could independently increase the risk of adverse pregnancy outcome. Clinicians are reminded of the importance of considering such factors when performing case-specific risk assessments.

Anthrax is a bacterial infection caused by Bacillus anthracis. Morbidity and mortality following infection are high, and rapid treatment following exposure is therefore required. An inactivated anthrax vaccine is available and, although not routinely administered, may be offered to individuals at risk of occupational exposure.

When clinically indicated, pregnant women should be treated as for the non-pregnant patient. Discussion with UKTIS is recommended in all cases. Ciprofloxacin is the first-line treatment for anthrax. Amoxicillin may also be used for treatment or prophylaxis but only if the specific strain is known to be penicillin-sensitive.

Published human data regarding maternal anthrax infection are limited to 18 case reports of infection in the latter half of pregnancy. Preterm delivery was reported in three of the cases even after successful treatment of the maternal infection. Intrauterine death preceded maternal death in three severe cases. There are no published reports of fetal outcome following anthrax infection in the first trimester of pregnancy, therefore an evidence-based assessment of malformation risk following first trimester exposure is not possible. Exposure to anthrax at any stage in pregnancy may warrant enhanced fetal monitoring which should be considered on a case-by-case basis. Discussion with UKTIS is recommended in all such cases.

The available published data regarding exposure to the anthrax vaccine during pregnancy describe more than 7,000 first trimester-exposed pregnancies. There is no compelling evidence that exposure to anthrax vaccine in pregnancy is teratogenic. However, data relating to neonatal complication and neurodevelopmental outcomes are lacking.

It is recommended that pregnant women should ideally defer anthrax vaccination, unless they are at risk of exposure to Bacillus anthracis. Advice from the UK Health Security Agency (UKHSA) is that anthrax vaccine may be given to pregnant women when clinically indicated.

Inadvertent exposure to the anthrax vaccine at any stage in pregnancy would not usually be regarded as medical grounds for termination of pregnancy or any additional fetal monitoring. However, other risk factors may be present in individual cases which may independently increase the risk of adverse pregnancy outcome. Clinicians are reminded of the importance of consideration of such factors when performing case-specific risk assessments.

Artemisinin and its derivatives (artesunate, artemether, dihydroartemisinin) are used in the treatment of malaria, usually in combination with other antimalarials, which is termed artemisinin-based combination therapy (ACT). In the UK the only licensed artemisinin derivative is artemether (in combination with lumefantrine).

Artemisinins have been associated with embryolethality in animal studies, therefore theoretical concerns regarding use in humans exist. However, human data do not provide evidence of an increased risk of miscarriage following exposure to artemisinin and its derivatives during pregnancy. 

The available human data do not currently provide any reliable evidence indicating that first trimester artemisinin use is associated with an increased risk of malformation. However, the data are highly limited which precludes the ability to provide an informed assessment of the risk. A large number of second and third trimester pregnancy exposures have been documented which do not signal that gestational exposure to artemisinin or its derivatives is associated with increased rates of stillbirth, low birth weight or preterm delivery above those observed in women treated with other antimalarials.

Untreated or inadequately treated maternal malaria infection poses a serious risk to both the mother and the fetus, therefore treatment with artemisinin or artemisinin derivatives should not be withheld at any stage of pregnancy if considered essential for maternal treatment. The UK malaria treatment guidelines (2016) and WHO guidelines for the treatment of malaria (2015) state that uncomplicated falciparum malaria should be treated with artemether-lumefantrine in the second and third trimester, and with quinine and clindamycin in the first trimester. Women with severe malaria in any trimester of pregnancy should be treated the same as non-pregnant patients, with artesunate preferred over quinine. Specialist advice should be sought regarding treatment of malaria during pregnancy.

Exposure to artemisinin and derivatives of the compound at any stage in pregnancy in the absence of maternal malaria infection would not usually be regarded as medical grounds for additional fetal monitoring. However, other risk factors may be present in individual cases which may independently increase the risk of adverse pregnancy outcome. Clinicians are reminded of the importance of consideration of such factors when performing case-specific risk assessments. Discussion with UKTIS is recommended in all cases.

Aspirin is an acetylated salicylate with analgesic, antipyretic, anti-inflammatory and antiplatelet properties. Low-dose aspirin (75-300mg/day) is used as an antiplatelet medication in the prophylaxis of thromboembolic cerebrovascular disease and myocardial infarction. Higher doses of aspirin (up to 4g/day) are used in the control of mild-to-moderate pain and pyrexia. 

The National Institute for Health and Care Excellence (NICE) guidelines state that pregnant women at high risk of pre-eclampsia should be offered low-dose aspirin from 12 gestational weeks. Low-dose aspirin is sometimes used from conception for women undergoing fertility treatment and those with a history of recurrent miscarriage or conditions such as antiphospholipid syndrome, as some studies have suggested an improvement in live birth rates.

There is no specific information on malformation rates following use of low-dose aspirin in pregnancy but in most cases this treatment is initiated after 12 weeks of pregnancy when fetal organogenesis is complete and there is little risk of medication-induced structural malformation. Increased risks of gastroschisis, cleft lip and palate, and neural tube defects have been reported following maternal use of aspirin during pregnancy (mainly at analgesic doses), however data are conflicting and may also be confounded, and a causal association with aspirin has not been proven. Data relating to risk of cardiac malformations are reassuring.

The majority of evidence suggests that there is no association between the use of low-dose aspirin and miscarriage, fetal growth restriction, or preterm delivery. Stillbirth data are limited to one small randomised-controlled trial that found no increased risk following exposure to low-dose aspirin. No data are available on pregnancy outcomes following aspirin use at analgesic doses, and data on neurodevelopmental outcomes following in utero aspirin exposure at any dose are too limited to facilitate an evidence-based assessment of risk.

Chronic exposure to analgesic doses of aspirin >300mg/day from 30 weeks of pregnancy may be associated with neonatal bleeding complications and premature closure of the ductus arteriosus (DA), resulting in pulmonary vasculature abnormalities and persistent pulmonary hypertension of the newborn (PPHN). These effects have not been reported with low-dose aspirin use.

Regular use of analgesic doses of aspirin in the third trimester should be avoided if possible. In circumstances where the maternal clinical condition requires treatment with analgesic doses of aspirin during the third trimester, fetal monitoring for oligohydramnios and ductus arteriosus patency is recommended. Exposure to analgesic doses of aspirin prior to the third trimester would not be considered an indication for any additional fetal monitoring. 

Aspirin overdose can result in severe maternal toxicity and consequently adverse fetal effects.  Pregnant women who have overdosed on aspirin should be managed as for the non-pregnant patient and may require additional interventions and/or fetal monitoring. Discussion with UKTIS and NPIS is recommended in all such cases.

Other risk factors may also be present in individual cases which may independently increase the risk of adverse pregnancy outcome. Clinicians are reminded of the importance of consideration of such factors when performing case-specific risk assessments.

Atenolol is a cardioselective beta-blocker licensed for the treatment of hypertension, angina pectoris, cardiac arrhythmias, for secondary prevention after acute myocardial infarction, and also off-license for migraine prophylaxis.

Data on overall rates of fetal structural malformations, or of specific malformations following first trimester use of atenolol, are too limited to permit an evidence-based risk assessment. There are also insufficient data to assess the risk of miscarriage, stillbirth, and adverse neurodevelopmental outcomes following gestational atenolol exposure.

It is currently unclear whether gestational use of atenolol increases the risk of preterm delivery. Although the currently available limited data do not suggest that this is the case, further research is required to confirm this finding.

There is some evidence that atenolol exposure in pregnancy may adversely affect fetal growth, although data are conflicting. Some studies suggest that the risk of adverse effects on fetal growth may be higher when exposure commences prior to 20 gestational weeks but the possibility of a maternal disease effect cannot be excluded.

Use of beta-blockers near term may result in neonatal beta-adrenoceptor blockade, leading to neonatal bradycardia, hypotension and hypoglycaemia. Accordingly, some (but not all) studies have reported increased rates of these neonatal complications in atenolol-exposed infants, although the absolute risk remains unquantified. Assessment of the neonate for these effects is thus advised.

Exposure to atenolol at any stage in pregnancy would not usually be regarded as medical grounds for termination of pregnancy. In pregnancies complicated by maternal hypertension and/or where atenolol has been administered, careful monitoring of fetal growth is advised. Other risk factors may also be present in individual cases which may independently increase the risk of adverse pregnancy outcome. Clinicians are reminded of the importance of consideration of such factors when performing case-specific risk assessments.

Atomoxetine is a selective noradrenaline reuptake inhibitor (NRI) used in the treatment of attention deficit hyperactivity disorder (ADHD).

The published data regarding atomoxetine use in human pregnancy consist of two case reports and four cohort studies which incorporate a total of 1,621 atomoxetine-exposed pregnancies, including at least 1,465 first trimester exposures. In addition, two population-based cohort studies which utilised an overlapping dataset have investigated ADHD medications as a group.

There is no robust evidence of increased risks of CM overall, cardiac malformations or limb malformations, but the analyses are based on a small number of outcomes which produced imprecise risk estimates.

The controlled data for ADHD medicines as a group describe a possible association with miscarriage, preterm delivery, and decreased five-minute Apgar scores. However, data confounding by the underlying maternal condition may have influenced the observations and a separate analysis of the atomoxetine exposures was not undertaken.

The currently available data for atomoxetine do not raise concern of increased risks of low birth weight and preterm delivery but are too limited to exclude these risks. Data regarding miscarriage and intrauterine death are limited or absent and therefore it is not currently possible to provide an evidence-based assessment of the risks of these outcomes.

It is important that maternal ADHD is adequately controlled during pregnancy. The risks of destabilisation and maternal relapse must be taken into account when considering dose reduction or switching a patient from atomoxetine to another medication(s).

As with other centrally acting drugs, there is a potential risk of poor neonatal adaptation syndrome (PNAS)/neonatal withdrawal effects and/or persistent pulmonary hypertension of the newborn (PPHN) in the neonate. Infants exposed to atomoxetine in utero should ideally be delivered in a unit with neonatal support and monitored for symptoms of PNAS.

Exposure to atomoxetine at any stage in pregnancy would not usually be regarded as medical grounds for termination of pregnancy. However, other risk factors may be present in individual cases which may independently increase the risk of adverse pregnancy outcome. Clinicians are reminded of the importance of consideration of such factors when performing case-specific risk assessments. Given the lack of sufficient human pregnancy exposure data, the need for additional antenatal anomaly scans and/or fetal monitoring should be considered on a case-by-case basis following maternal atomoxetine use in pregnancy.

Atropine is a tropane alkaloid with antimuscarinic effects. It is used in the treatment of symptomatic bradycardia, as an antisialagogue, as an antidote, and topically as a mydriatic and cycloplegic.

Data on the use of atropine in pregnancy are extremely limited, but do not currently indicate an increased risk of fetal malformation. Information on other adverse pregnancy outcomes is too limited to permit an evidence-based assessment of any risk.

Atropine should not be withheld in a pregnant patient if there is a compelling clinical indication for use. Enhanced antenatal surveillance may be warranted following maternal atropine exposure and should be decided on a case-by-case basis. Discussion with UKTIS is recommended in all cases of exposure to atropine at any stage of pregnancy.

Azathioprine is a prodrug and is rapidly metabolised to its active metabolite mercaptopurine (6-MP). Mercaptopurine is a purine analogue which interferes with the synthesis of DNA and RNA and has immunosuppressant properties. Azathioprine is licensed to treat various auto-immune disorders such as inflammatory bowel disease, rheumatoid arthritis and psoriasis, and for the prevention of allograft rejection. It is also used off-licence to treat severe refractory eczema. Mercaptopurine is licensed to treat acute and chronic leukaemia, as maintenance therapy in acute lymphoblastic and myelogenous leukaemia, and is used off-license to treat severe ulcerative colitis and Crohn’s disease.

The available data do not demonstrate that azathioprine/mercaptopurine exposure during early pregnancy increases the risk of congenital malformation, preterm delivery, or adversely affects fetal growth. There is also no robust evidence of increased risks of intrauterine death, miscarriage, or altered neurodevelopment following gestational azathioprine exposure. However, as the available data for many of these outcomes are limited in quantity and sometimes also methodologically, more robust epidemiological data are ideally required.

Neonatal leucopoenia and thrombocytopenia have been reported in a number of case reports following azathioprine exposure in utero but no epidemiological studies quantifying these risks are available. There is therefore a theoretical concern that the use of azathioprine/mercaptopurine during pregnancy could result in immunosuppression in the neonate, leading to an increased risk of infection and adverse effects following administration of live vaccines. Advice from the MHRA states that infants exposed in utero to certain immunosuppressive medications should not receive live vaccines until they are least 6 months old, although no specific advice relating to azathioprine exposure was issued.