Dr Buyon suggested that patients with RA should have their disease under as good control as possible prior to conception; careful planning is needed to stabilize disease activity and to modify medication regimens. Because of its risk of teratogenicity, methotrexate (MTX) must be discontinued 1 to 3 months before conception. However, Eck et al found that paternal exposure to MTX within 90 days before pregnancy was not associated with congenital malformations, stillbirths, or preterm birth. Thus, a man with RA need not discontinue MTX therapy when he is trying to conceive a child with his partner. Women with RA who are not currently planning to become pregnant but desire to conceive in the future should be counseled about potential safety issues regarding their current RA medications, including the risks of teratogenicity.

RA disease activity often improves during pregnancy, as described in 1938 by Philip Hench, who first used corticosteroids to treat RA. Dr Buyon indicated that retrospective and prospective studies have found that up to 75% of patients with RA experience improvement in disease activity during pregnancy. However, joint inflammation typically flares 3 to 4 months postpartum. Hydroxychloroquine, sulfasalazine, azathioprine, and corticosteroids are safe to use during pregnancy to treat RA. There is evidence supporting the safety of tumor necrosis factor inhibitors (TNFis) in pregnancy, especially certolizumab pegol, which lacks an Fc region and thus does not undergo transplacental transport. Given the adverse impact of active disease on fertility and pregnancy outcome, a treat-to-target strategy is recommended for patients who are pregnant or wish to conceive. At the 2019 ACR/ARP Annual Meeting, Haase et al presented data in abstract 2279 suggesting that it may be feasible and safe for women to reduce the frequency of TNFi dosing, guided by disease activity. Indeed, for all patients in remission, treatment de-escalation is recommended. I would not use Janus kinase inhibitors or biological agents other than TNFis because of the paucity of data confirming their safety during pregnancy.

In abstract 2283, Mills and Bermas reported the results of a survey that was distributed to 22 rheumatology providers that found considerable knowledge gaps regarding the compatibility of antirheumatic drugs with lactation. At baseline, all providers responded that hydroxychloroquine was compatible with lactation, but there was substantial uncertainty regarding the compatibility of prednisone, MTX, leflunomide, sulfasalazine, mycophenolate, and TNFis with lactation. The authors developed a “Rheumatic Diseases and Lactation” information card to assist providers in determining the compatibility of medications with lactation. The card listed MTX, leflunomide, and tofacitinib among those medications that are contraindicated during lactation. Although tofacitinib has not been studied in lactation, it is likely to pass into breast milk because of its small molecular size. TNFis and the interleukin 6 receptor antagonist tocilizumab were among the agents listed as being compatible with lactation. Prednisone at doses of less than 20 mg/day was considered to be compatible with lactation, but, at doses of greater than 20 mg/day, breastfeeding must be delayed for at least 4 hours after the prednisone dose and breast milk expressed during that time must be discarded. After being provided with the information card, the rheumatology providers demonstrated improvement in their knowledge about antirheumatic drugs and lactation.

The decision to breastfeed or not involves a very personal choice. Some mothers may feel guilty if they do not breastfeed, and I would not want to contribute to that guilt. If a mother indicates that she is interested in breastfeeding, I counsel her about the compatibility of the various antirheumatic drugs with breastfeeding. As more data become available, our recommendations can be more informed.