Intranasal and Inhaled Corticosteroids

Intranasal and inhaled corticosteroids are used for the treatment of allergic rhinitis and asthma. The commonest intranasal and inhaled corticosteroids include beclomethasone, budesonide, flunisolide, fluticasone, mometasone and triamcinolone. They have relatively few systemic effects, due to low bioavailability after administration, which differs depending on the particular corticosteroid and formulation.^2,3 Second-generation agents, including mometasone and fluticasone have been shown to have <0.1–2% oral bioavailability compared to older-generation agents such as triamcinolone, flunisolide and beclomethasone, which have 20–50%.^4,5

Intranasal and inhaled budesonide is the only corticosteroid considered by the FDA to be a Pregnancy Category B medication (i.e. no risk shown in animal studies or sufficient clinical safety data in pregnancy). Epidemiology studies from three Swedish birth registries (Swedish Medical Birth Registry, Registry of Congenital Malfomations and Child Cardiology Registry) showed no increased risk of congenital malformations, premature birth, stillbirth, or decreased birth weight or length, compared to the general population.^6,7 These large population-based studies investigated the safety of inhaled budesonide, but as maternal plasma concentration is lower after intranasal budesonide, the risk is considered comparable. A case-control population study by Kallen showed that the use of intranasal budesonide may have an association with less severe infant cardiovascular defects, however this did not reach statistical significance.⁸

Research on other intranasal corticosteroids is limited, but a randomized placebo-controlled trial of intranasal fluticasone during pregnancy showed no difference in birth weight, fetal growth or APGAR scores.⁹ A Canadian population-based prospective study examined the use of intranasal triamcinolone, other intranasal corticosteroids or no corticosteroid exposure during pregnancy. This showed no associated risk of small for gestational age, miscarriage or congenital malformations with triamcinolone, but did have an increased risk of respiratory defects which was not in keeping with previous population data.¹⁰ Although data are limited, intranasal corticosteroids are considered safe during pregnancy and breastfeeding due to their low bioavailability after administration and previous safety profiles of inhaled corticosteroids at higher systemic doses.¹¹

Inhaled corticosteroids have been examined more extensively than intranasal corticosteroids and have been considered safe to use in pregnancy when required.¹² A 2004 prospective cohort study of 474 women compared fetal outcomes from mothers receiving inhaled beclomethasone, fluticasone, triamcinolone, budesonide and flunisonide.¹³ This study showed no statistical difference in low birth weight, pre-term births and congenital malformations in this cohort compared with the general population. A further systematic meta-analysis of inhaled corticosteroids including fluticasone, beclomethasone, budesonide, triamcinolone and flunisolide showed no significant difference in major malformations, pre-term delivery, low birth weight and pregnancy-induced hypertension.¹⁴

Inhaled corticosteroids have relatively low molecular weight and high lipid solubility and would therefore be transmitted into breast milk. Infants of mothers receiving inhaled budesonide were found to have 1/600 budesonide plasma concentration compared to the mother, and plasma concentrations less than the limit of quantification.¹⁵ Other inhaled corticosteroids have not been studied in this way, but with low systemic bioavailability would be assumed to have similar properties and thus are considered safe to use whilst breastfeeding. Intranasal corticosteroids have lower bioavailability than inhaled and thus would have a similar or reduced transmission of dose during breastfeeding.

Oral and Intravenous Corticosteroids

Oral and intravenous glucocorticoids including hydrocortisone, prednisolone, dexamethasone and betamethasone are pivotal to treatment of inflammatory respiratory disorders. There has been previous concern about congenital malformations associated with corticosteroid use in early pregnancy from several epidemiological studies, which showed an association between corticosteroid use and orofacial cleft.¹⁶–¹⁸ However, a 2011 large population cohort study from Denmark examined 51 973 out of 832 636 live births over a 12-year period exposed to first-trimester corticosteroids and showed no significant increase in orofacial cleft compared with non-exposure.¹⁹ This has also been supported in a recent US case-control study from the National Birth Defect Prevention Study (NBDPS) and a further prevalence study from Denmark, which showed no association between maternal corticosteroid use and orofacial cleft.^20,21 The European League against Rheumatism (EULAR) task force recommended the use of glucocorticoids for rheumatic disease during pregnancy after systematic review of the literature.²² Similarly, The British Thoracic Society and Scottish Intercollegiate Guidelines Network support the use of inhaled corticosteroid and systemic steroids when required for asthma control.²³

Maternal corticosteroid therapy has been associated with increased maternal infection and pre-term birth from premature rupture of membranes,^24,25 however this risk is balanced against the benefit of treating respiratory conditions, as uncontrolled conditions such as asthma can cause similar adverse peri-natal outcomes.²⁶ An additional risk of antenatal corticosteroid therapy is the propensity to increase insulin resistance in an already insulin-resistant state during pregnancy.²⁷ This can lead to gestational diabetes and the complications associated with this condition, including macrosomia, shoulder dystocia and pre-term delivery.²⁸ Prolonged use of high-dose oral steroids (>20 mg/day) increases the risk of premature rupture of the membranes (and therefore pre-term delivery) and has also been associated with fetal growth restriction.²⁹

Both oral and intravenous corticosteroids have a low molecular weight and have been shown to transfer into the breast milk of women taking these medications. One study examined the maternal serum and milk concentration after repeated doses with 10–80 mg/day prednisolone and found milk concentrations between 5% and 25% of the maternal serum level. They reported that at a daily dose of 80 mg prednisolone, the infant would ingest less than 0.1%, which is less than 10% of the infant’s endogenous cortisol production.³⁰ A systematic review of the literature found corticosteroids to be safe to use during breastfeeding, with the advice to avoid breastfeeding during the first 4 hours after a dose of prednisolone >50 mg to minimize the dose of corticosteroid transferred.²²

Tumour Necrosis Factor-alpha Inhibitors

Tumour necrosis factor (TNF)-alpha is a pro-inflammatory cytokine with a central role in inflammatory processes and is a target for prevention of conditions including rheumatic and psoriatic arthritis, ankylosing spondylitis, Crohn’s disease and ulcerative colitis. Tumour necrosis factor inhibitors (TNF-i) are biological agents and include adalimumab, certolizumab, etanercept, golimumab and infliximab. They are monoclonal antibodies, with the exception of certolizumab, which is a monovalent Fab fragment that blocks TNF-alpha, and etanercept, which is a dimeric fusion protein that blocks interaction between TNF-alpha and TNF cell-surface receptors. These medications have been used in pregnancies and to date appear safe for those women that require biologic therapy to control their underlying disease.

Adalimumab, infliximab and golimumab are IgG1 monoclonal antibodies, which have a high molecular weight and therefore do not diffuse freely across the placenta. They have differing structure and placental transfer times, along with a half-life range between 8 and 20 days. Monoclonal antibodies are actively transported across the placenta once placental Fc receptors develop on the trophoblast cells towards the third trimester.³³ Placental transfer is initially limited by cytotrophoblasts in early pregnancy, but increasing active transport of monoclonal antibodies to the fetus towards the end of pregnancy leads to greater fetal concentrations of adalimumab and infliximab in the cord blood at delivery than those in the mother’s serum.³⁴ This process is important to understand as it guides recommendations to stop adalimumab and infliximab either between 6–8 weeks prior to delivery³⁵ or at 20 weeks’ gestation²² to ensure minimal drug levels in the neonate at birth. Human data on golimumab use in pregnancy are limited and thus there is insufficient evidence to inform its use in pregnancy.

Etanercept and certolizumab have a half-life of four and 14 days, respectively. Placental transfer of etanercept is slow due to reduced uptake from placental Fc receptors and certolizumab relies on placental diffusion only, which is limited due to its high molecular weight. The pharmacokinetics of these medications is confirmed by the reduced or negligible fetal cord blood concentrations at delivery, despite continued maternal usage until near delivery, and therefore support the low transfer of these medications across the placental membrane.³⁴ The preferential materno-fetal transfer profile allows certolizumab use throughout pregnancy and has led gastroenterologists to suggest certolizumab as a potential first line TNF-i in women of child-bearing age.³⁶ The EULAR task force suggests that further evidence is required to ensure the fetal and infant safety of certolizumab is confirmed prior to its use in this way.²²

The EULAR task force performed a systematic literature review of all studies involving TNF-i through pregnancy, including adalimumab, certolizumab, etanercept, golimumab and infliximab. The collation of 10 cohort studies, five case-controls, two registry data and 32 case reports, which did not differentiate among TNF-i, found no significant difference in miscarriage or congenital malformation in pregnancy with TNF-i compared to controls.²² This trend was similarly seen for individually studied TNF-i, with most data on infliximab (1161 pregnancies) followed by adalimumab (524 pregnancies). Infliximab, adalimumab and etanercept reached 2b strength of evidence²² with at least one quasi-experimental study.

TNF-i use during pregnancy has been scrutinized after the death of a 4.5-month-old infant of dissemin-ated tuberculosis after a live BCG vaccine, in which the mother had received infliximab two weeks prior to delivery.³⁷ The EULAR task force recommended from this that any neonate who has been exposed to a TNF-i agent in utero should not receive a live vaccine for the first six months of life. All other vaccinations involving killed vaccines can be undertaken.²²

TNF-i have a high molecular weight and it is unlikely that significant amounts of these medications reach therapeutic significance in the infant from breastfeeding. This has been supported by evidence of negligible concentrations of infliximab, adalimumab or certolizumab in breast milk and reducing concentrations of these medications in infants from mothers receiving them, with no adverse outcome seen in the infants.³⁸–⁴¹ There are no data on the transfer of golimumab into breast milk, but it is thought to have similarly low bioavailability due to its large molecular weight, and as such is safe to use in lactation.²² Furthermore, because these agents are degraded by enzymes in the stomach and have low/no oral bioavailability, even if they are present in breast milk, there is unlikely to be systemic absorption by the neonate.

T-cell and B-cell Depletion

Azathioprine, mycophenolate and leflunomide cause both B- and T-cell depletion by interference with DNA and RNA synthesis. They have been used in several diseases as biological therapy including systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), vasculitis, organ transplant, inflammatory bowel disease and sarcoidosis.

Azathioprine is converted to 6-mercaptopurine and then into its metabolites via xanthine oxidase (XO), hypoxanthine phosphoribosyl transferase (HPRT) and thiopurine methyltransferase (TPMT) pathways. 6-Methyl mercaptopurine is then converted to the active substance, thioguanine nucleotides (TGN), which inhibit DNA and RNA synthesis. Genetic variants of the TPMT enzyme can cause increased levels of TGN metabolites and lead to bone marrow suppression or hepatotoxicity. Patients planning to start azathioprine should have their TMPT levels checked prior to initiation. Azathioprine was examined by the EULAR task force and found to have no significant increase in miscarriage or congenital malformation when compared to disease-matched controls in 1327 pregnancies affected by azathioprine.²² They have recommended that this can be used throughout pregnancy and it is compatible with breastfeeding, which has been shown in case-control studies.⁴²–⁴⁴

Mycophenolate is a purine synthesis inhibitor and thus prevents the proliferation of B-cell and T-cells, which is beneficial in controlling inflammatory conditions. This drug has been associated with high rates of miscarriage and congenital malformations with a specific pattern, including external ear anomalies, cleft lip, with or without cleft palate, and ocular anomalies.²² Less frequent features associated with mycophenolate embryopathy include congenital heart defects, distal limb anomalies, oesophageal atresia, vertebral malformations, diaphragmatic hernia, and kidney and central nervous system anomalies.⁴⁵ The data on lactation in this group are minimal, and therefore it should probably be avoided while breastfeeding. Women should stop mycophenolate at least 1.5 months prior to conception.

Leflunomide is a pyrimidine synthesis inhibitor, which is required for DNA and RNA synthesis and hence prevents reproduction of rapidly dividing cells such as B-cells and T-cells. Animal data with leflunomide in rats and rabbits showed a dose-related teratogenicity and embryo-fetal toxicity and it has therefore been contraindicated in pregnancy.⁴⁶ This has been further confirmed by lack of evidence to support safety in a systematic review, despite initial results showing no difference in miscarriage or congenital malformation. There are no human lactation data and therefore it should also be avoided when breastfeeding.

Interleukin Inhibitors (IL-i)

Interleukins are immunomodulatory cytokines involved in immunity and inflammation, including immune cell proliferation, maturation, migration and adhesion. IL-i used in respiratory disease include anakinra (IL-1 inhibitor), basiliximab (IL-2 inhibitor), mepolizumab and reslizumab (IL-5 inhibitors) and tocilizumab (IL-6 inhibitor). These medications have been used for respiratory disease associated with RA, lung transplantation, severe eosinophilic asthma and large cell lung cancer.

Anakinra is a recombinant version of IL-1 and tocilizumab is a monoclonal antibody to the IL-6 receptor antagonist. They have both been trialled in patients with RA-associated interstitial lung disease (ILD). Anakinra showed deleterious effects on ILD but tocilizumab has been shown to be effective in case reports of RA-associated ILD.^47,48 Tocilizumab also has use in large cell lung cancer. Data on these IL-i are limited in pregnancy and although the EULAR task force found no increase in congenital malformations, the recommendation would be to switch to other therapy prior to conception, with use only if there are no other well-studied options available for maternal disease.²² Similarly, breastfeeding with these biologics should be discouraged due to insufficient lactation data. Both are large molecular weight proteins, which suggests limited excretion into breast milk, but physiological active transfer of maternal antibody and IL-1 into breast milk means that these medications are also transferred, sometimes at higher than maternal plasma concentration.⁴⁹

Basiliximab is a chimeric mouse–human monoclonal antibody to the alpha chain of the IL-2 receptor on T-cells. It is used as an induction immunosuppressant and for antibody-mediated rejection in transplanted patients, and is generally well tolerated.^50,51 It is an FDA Pregnancy Category B drug, as no maternal toxicity, embrotoxicity or teratogenicity events were observed in cynomolgus monkeys 100 days post-coitum following basiliximab dosing during organogenesis. Similarly, a 2004 review on teratogenesis and immunosuppressive treatment concluded that its use could be continued in pregnancy as it is only used for induction immunosuppression or treatment of acute rejection.⁵² The manufacturer, Novartis, recommends contraception prior to, during and up to four months after using basiliximab.⁵³ There are no lactation data from mothers taking basiliximab, but it may be assumed that as an immunoglobulin this would be excreted into the breast milk similar to physiological immunoglobulins. The long half-life of basiliximab (seven days) could also mean that basiliximab may be seen in colostrum or mature milk after birth.

Mepolizumab is a humanized monoclonal antibody (IgG1 kappa) and antagonist to IL-5, which selectively inhibits eosinophilic inflammation, reduces eosinophils in sputum and blood, and reduces asthma exacerbations. It is used in the treatment of eosinophilic asthma over the age of 12 in combination with other asthma medications.⁵⁴ Human pregnancy data are insufficient to inform on the fetal-maternal risk of this medication, although animal data from cynomolgus monkeys have shown no evidence for fetal harm through pregnancy at doses 30 times that of the human dose. In addition, there were no differences in fetal or neonatal growth at nine months. Mepolizumab, like previously discussed monoclonal antibodies, is actively transferred at increasing rates across the placenta towards term, and similarly due to the high molecular weight but physiological transfer of IgG, is transferred at minimal dose to breast milk in animals.⁵⁵ The use of mepolizumab may be beneficial in reducing the corticosteroid burden during pregnancy, but further safety data are required to balance the risk and benefits. Reslizumab is also an interleukin-5 antagonist monoclonal antibody used for the treatment of eosinophilic asthma. It prevents the growth, differentiation and activation of eosinophils by binding to the IL-5 receptor and inhibiting its activation. Similarly, animal studies have not shown evidence of embryotoxicity or fetotoxicity, but there is insufficient data in human pregnancy on drug-associated risk.

B-cell Inhibitors

Belimumab is a monoclonal antibody that inhibits B-cell activating factor (BAFF), which is also known as B-lymphocyte stimulator (BLyS), and reduces B-cell survival. This medication is licensed in SLE and further pilot studies have suggested use in diffuse systemic sclerosis, both being conditions that can affect the respiratory system.⁵⁶–⁵⁸

Belimumab, like other immune monoclonal antibodies, has been shown in cynomolgus monkeys to cross the placenta. The initial animal data showed no direct or indirect teratogenicity, but did have some infant and fetal deaths thst were unexplained.⁵⁹

The 2016 EULAR force guidelines collated 153 pregnancies exposed to belimumab, where miscarriage occurred in 26.8% and congenital malformations occurred in 9.9%. They have concluded that further human data are required prior to use of belimumab in pregnancy and alternative therapy should be considered throughout pregnancy, unless essential to control maternal disease.²² Similarly, there are no human lactation data on belimumab; breastfeeding should be avoided as belimumab may be transmitted into the breast milk of animal models and the effect on the infant is unknown.

VEGF Inhibitors

Bevacizumab is a humanized monoclonal IgG1 antibody that binds to vascular endothelial growth factor (VEGF) extracellularly and inactivates VEGF receptors and angiogenesis. It has been licensed for the treatment of advanced non-small cell lung cancer⁶⁰ and has shown benefit in treatment of recurrent respiratory papillomatosis.⁶¹

There are no data on bevacizumab in pregnancy for the indications above, but initial animal studies demonstrated adverse fetal outcomes⁶² and there have been two case reports of non-pregnant women with bevacizumab developing pre-eclampsia mimics.⁶³ A literature review of case reports of intravitreous bevacizumab treatment for choroidal neovascularisation reported cases of miscarriage and normal fetal outcomes.⁶⁴ It is difficult to draw conclusions from this small dataset alone, as the background risk of miscarriage in early pregnancy is high and some of the patients included had several risk factors for miscarriage. Furthermore one would not expect significant systemic drug levels after intravitreous administration. Similar to previously discussed IgG1 monoclonal antibodies, bevacizumab is a large molecule, but may be actively transported into breast milk and therefore breastfeeding should be avoided.

Stay updated, free articles. Join our Telegram channel

Join