Recommendations for the Use of Disease‐Modifying Antirheumatic Drugs in Pregnancy and Reproductive Health for Patients With Rheumatic Disease: A Scoping Review

Abstract

Objective

Autoimmune rheumatic diseases commonly affect individuals of childbearing age, with historically increased adverse pregnancy outcomes in this group. The advent of disease‐modifying antirheumatic drugs (DMARDs) has fostered more suitable conditions for pregnancy; however, this is accompanied by challenges in ensuring safe use in reproductive health. The aim of this review is to compare existing guideline recommendations for the use of DMARDs in pregnancy and reproductive health for patients with rheumatic disease.

Methods

A scoping review was performed with Medline and Embase, in addition to a hand search, to identify guidelines published since 2014 by academic societies in rheumatology that addressed the management of DMARDs in pregnancy in any of rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, and systemic lupus erythematosus. Conventional synthetic DMARDs (csDMARDs) (methotrexate, sulfasalazine, leflunomide, and hydroxychloroquine), biologic DMARDs (bDMARDs) (adalimumab, etanercept, infliximab, golimumab, certolizumab, abatacept, tocilizumab, rituximab, and anakinra), and targeted synthetic DMARDs (tsDMARDs) (tofacitinib, baricitinib, and upadacitinib) were targeted. Two authors performed data extraction in duplicate (AC, AT).

Results

A total of 18 guidelines were included. Recommendations for DMARD use in preconception were present in 10 guidelines (56%), lactation in 12 guidelines (67%), and male fertility in 6 guidelines (33%). A total of 13 guidelines (72%) included recommendations for csDMARDs, 13 guidelines (72%) included recommendations for bDMARDs, and 5 guidelines (28%) included recommendations for tsDMARDs. There was moderate evidence supporting relatively uniform csDMARD recommendations, compared to minimal evidence for bDMARD and tsDMARD use with variable recommendations.

Conclusion

There is heterogeneity in the formulation of guidelines on the use of DMARDs in pregnancy. Recommendations for csDMARDs were similar between guidelines. There was significant variability in recommendations for bDMARD and tsDMARD use, reflecting current minimal literature in this area.

INTRODUCTION

Autoimmune rheumatic diseases (AIRDs) such as rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), and systemic lupus erythematosus (SLE) commonly affect individuals of childbearing age and have historically had significant effects on reproductive health. ¹ Patients with AIRDs traditionally have been observed to be at greater risk of adverse perinatal outcomes, including risk of preeclampsia, emergency Cesarean section, preterm delivery, and low birth weights. ¹ , ² , ³

SIGNIFICANCE & INNOVATIONS.

• There is relative uniformity in guidelines regarding the use of conventional synthetic disease‐modifying antirheumatic drugs (DMARDs) in pregnancy in patients with rheumatic diseases.

• There has been a gradual shift toward recommendations of certain biologic DMARD use in reproductive health in the last 10 years.

• There is a lack of guidelines regarding DMARD use in male fertility and newborn health.

• Greater consumer representation is important in future formulation of guidelines in this area.

The advent of novel disease‐modifying antirheumatic drugs (DMARDs) for treatment of these rheumatic conditions has significantly improved disease control and hence fostered more suitable conditions for pregnancy. ⁴ As such, there has been a rise over time in number of births among women with rheumatic disease, and, notably, risks of adverse pregnancy and low fetal birth weight have reduced. ⁴ , ⁵

However, this is accompanied by new challenges in ensuring safe use in reproductive health, with concerns regarding teratogenicity, safety in breastfeeding, and the preconception state. Certain DMARDs have clear evidence of potential harm in pregnancy; leflunomide has been observed to be embryotoxic in preclinical animal studies, ⁶ and methotrexate is a current standard medical treatment for ectopic pregnancy. ⁷ There are limited data regarding the use of newer agents such as biologic DMARDs (bDMARDs) and targeted synthetic DMARDs (tsDMARDs) in reproductive health, although there is emerging literature to support the use of bDMARDs for disease control in women with inflammatory arthritis during pregnancy. ⁸ Similarly, given DMARDs are immunosuppressive by nature of mechanism of disease control, implications for newborn immunity requires consideration. ⁹

The potential for unknown teratogenic or developmental implications in children is always challenging in the use of novel drugs during pregnancy. However, the use of DMARDs in reproductive health must be balanced with the benefits in achievement of parental disease control, which appears to cultivate positive peripartum outcomes. ¹⁰ , ¹¹ This review aims to compare existing clinical practice guideline recommendations for the use of DMARDs in pregnancy and reproductive health for patients with rheumatic disease.

MATERIALS AND METHODS

Type of review and search strategy

We sought to identify gaps in literature in this area by reviewing a broad range of sources, and therefore a scoping review was performed following established guidelines from the Joanna Briggs Institute Scoping Review Methodology Group. ¹² Two main searches for relevant publications from Medline and Embase databases were performed, with support from research librarians. A single reviewer (MK) performed the initial title and abstract prescreening, with subsequent further screening by two reviewers (AC and AT). Additional guidelines were identified from hand search of reference lists of guidelines included. Full‐text reviews were then performed by two reviewers (AC and AT), and articles were assessed against predetermined inclusion and exclusion criteria. Inclusion criteria were the following: (1) recommendations, guidelines type, and published within the last 10 years (2014–2024); (2) author or group of national or international academic societies in rheumatology; (3) target disease and patients (adult patients with RA, AS, PsA, juvenile idiopathic arthritis, or SLE); and (4) target drugs including conventional synthetic DMARDs (csDMARDs) (methotrexate, sulfasalazine, leflunomide, hydroxychloroquine, and minocycline), bDMARDs (adalimumab, etanercept, infliximab, golimumab, certolizumab, abatacept, tocilizumab, rituximab, sarilumab, anakinra, and secukinumab), and tsDMARDs (tofacitinib, baricitinib, upadacitinib, and filgotinib). The exclusion criteria were the following: (1) systematic reviews and scoping reviews, meta‐analyses, observational studies, and randomized controlled trials; (2) author or group of individual hospitals or research groups; (3) target drugs (publications exclusively focused on immunosuppressants); and (4) no English translation available. Any discrepancies were resolved by discussion.

We searched Medline and Embase, in addition to a hand search of references, to identify the most recent versions of guidelines by national or international academic societies in rheumatology that addressed management of DMARDs in pregnancy in any of adult patients with RA, AS, PsA, and SLE. csDMARDs (methotrexate, sulfasalazine, leflunomide, and hydroxychloroquine), bDMARDs (adalimumab, etanercept, infliximab, golimumab, certolizumab, abatacept, tocilizumab, rituximab, anakinra, and secukinumab), and tsDMARDs (tofacitinib, baricitinib, and upadacitinib) were targeted. Guidelines published since 2014 were included.

Data extraction

Two authors (AC and AT) performed data extraction in duplicate. Data were extracted based on pre‐established characteristics of interest, including year of publication, country or region of publication, included rheumatic conditions, included DMARDs, inclusion of guidelines on specific areas of reproductive health (breastfeeding, preconception, male fertility, and newborn health), conflict of interest declaration and management, panel structure, method of formulation of recommendation, and evidence system used. Guideline recommendations regarding DMARD use in reproductive health were extracted, with exclusion of in‐text recommendations. Review and discussion between the two authors following initial extraction was performed to further refine data.

RESULTS

Literature search

A total of 743 unique articles were identified from the search strategy. A hand search identified a further two articles, and full‐text review was performed on a total of 125 articles following title and abstract screening. A total of 107 articles were excluded after full‐text review for the following reasons: no pregnancy recommendations (n = 49); review, editorial, or summary article (n = 21); not available in English (n = 9); not an international or national society guideline (n = 6); target population not included (n = 6); target drug not included (n = 5); not the most recent version of society guideline (n = 2); and other (including article subsequently retracted, comparison of guidelines, and defining scope for future guidelines; n = 9). The final review included 18 articles (Figure 1).

Figure 1.

Flowchart of included studies.

Characteristics of included guidelines

The 18 guidelines were published by national and international societies from a range of regions, including 7 guidelines from Europe (38.9%), ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ 5 guidelines from Asia (27.8%), ²⁰ , ²¹ , ²² , ²³ , ²⁴ 4 guidelines from North America (22.2%), ²⁵ , ²⁶ , ²⁷ , ²⁸ 1 guideline from South America (5.6%), ²⁹ and 1 guidelines from Africa (5.6%). ³⁰ In addition to pregnancy, recommendations on DMARD use in breastfeeding were included in 13 guidelines (72.2%), ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ²⁰ , ²¹ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ preconception in 10 guidelines (55.6%), ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ²⁰ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ male fertility in 6 guidelines (33.3%), ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ²⁵ , ²⁸ and postpartum newborn health in 2 guidelines (11.1%). ¹⁴ , ²⁷ Four guidelines (22.2%) were a standalone guideline for the use of DMARDs in pregnancy. ¹³ , ¹⁴ , ²⁵ , ²⁸ More than half of the guidelines (n = 10; 55.6%) were RA‐specific guidelines, ¹⁵ , ¹⁶ , ¹⁷ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁶ , ²⁷ and two guidelines (11.1%) were specific to SLE. ¹⁹ , ²⁹ Most guidelines addressed csDMARDs ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ¹⁸ , ²⁰ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ (n = 13; 72.2%) and/or bDMARDs ¹³ , ¹⁴ , ¹⁶ , ¹⁹ , ²⁰ , ²¹ , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ³⁰ (72.2%), and only six guidelines (33.3%) ¹³ , ¹⁴ , ¹⁶ , ²⁰ , ²¹ , ²⁷ had recommendations for tsDMARDs.

Formulation of recommendations

The identified guidelines had a broadly homogenous approach to guideline formulation, with most performing a literature review before using panel voting to reach consensus for recommendations. All guidelines that specified a predefined level of agreement ¹³ , ¹⁴ , ¹⁷ , ¹⁹ , ²¹ , ²³ , ²⁴ , ³⁰ (n = 8; 44.4%) required panel consensus >75% for a recommendation to be published. The Delphi or modified Delphi method ³¹ was used in five guidelines (27.8%). ¹³ , ¹⁵ , ²⁴ , ²⁷ , ³⁰ All voting panels included rheumatologists. A range of other professions were also included, including medical specialists (obstetricians, geneticists, internal medicine physicians, ophthalmologists, epidemiologists, pediatric neonatologists, infectious diseases specialists, nephrologists, dermatologists, and cardiologists), allied health professionals, and research fellows. Consumers were involved in the development of five guidelines ¹³ , ¹⁴ , ²⁵ , ²⁶ , ³⁰ (27.8%). Funding was disclosed by eight ¹³ , ¹⁵ , ²⁰ , ²¹ , ²³ , ²⁵ , ²⁸ , ²⁹ societies (44.4%), and three ²⁰ , ²⁸ , ²⁹ societies (16.7%) received pharmaceutical funding. The Appraisal of Guidelines for Research & Evaluation II ³² tool was used by three societies ²¹ , ²³ , ²⁸ (16.7%) during the formulation of recommendations.

Rating of evidence

There was a varied approach to rating of evidence within these guidelines. The most used evidence rating system was Grades of Recommendations, Assessment, Development and Evaluation (GRADE) ³³ (n = 7; 38.9%), then the Oxford Centre for Evidence‐Based Medicine Levels of Evidence ³⁴ (n = 2; 11.1%). Other systems used were the classification system by Shekelle et al, ³⁵ Harbour and Miller, ³⁶ and Scottish Intercollegiate Guidelines Network Grading System ³⁷ (n = 1 each; 5.6%). Two societies (11.1%) based their recommendations on previously published guidelines, and four societies (22.2%) did not specify rating of evidence.

Recommendations in pregnancy

Recommendations for csDMARDs

A total of 13 guidelines ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ¹⁸ , ²⁰ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ included recommendations for csDMARD use in pregnancy (Table 1). There was 100% agreement among these recommendations between guidelines, with a moderate level of evidence referenced. All guidelines recommended that methotrexate and leflunomide were contraindicated in pregnancy. Hydroxychloroquine and sulfasalazine were considered safe in pregnancy, with recommendation to continue. Two guidelines ²⁰ , ²⁸ (11.1%) specifically recommended supplemental folate supplementation with sulfasalazine use in pregnancy.

Table 1.

Recommendations for disease‐modifying antirheumatic drug use in pregnancy*.

Recommendations for bDMARDs

Thirteen guidelines ¹³ , ¹⁴ , ¹⁶ , ¹⁹ , ²⁰ , ²¹ , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ³⁰ included recommendations for bDMARD use in pregnancy. There was a low level of evidence for these recommendations. Recommendations differed between guidelines, however, there was a high level of agreement within individual panels. Three guidelines ¹⁹ , ²¹ , ²⁸ published before 2018 recommended avoiding all bDMARDs in pregnancy due to lack of evidence, without recommendations for specific medications. The other 10 guidelines all recommended continuing tumor necrosis factor (TNF) inhibitors (adalimumab, golimumab, etanercept, infliximab, and certolizumab); 7 guidelines were conditional recommendations. ¹³ , ¹⁶ , ²³ , ²⁵ , ²⁶ , ²⁷ , ³⁰ Certolizumab was strongly recommended to be continued throughout the entire pregnancy by five guidelines, ¹⁴ , ¹⁶ , ²⁰ , ²⁴ , ²⁵ and three guidelines specified continuing certain TNF inhibitors (adalimumab, golimumab, and etanercept) for the first two trimesters only. ²⁰ , ²⁴ , ²⁶

Most guidelines recommended cessation in pregnancy due to lack of evidence when specifically discussing other biologic agents, including abatacept (n = 4), ¹³ , ¹⁶ , ²⁰ , ²⁵ tocilizumab (n = 5), ¹³ , ¹⁶ , ²⁰ , ²⁵ , ²⁷ and anakinra (n = 3) ¹³ , ²⁰ , ²⁵ ; however, one guideline had a conditional recommendation for continuing these medications in pregnancy “if no other drug [is] available,“ ¹⁴ and another recommended they “should be used during pregnancy only when no other pregnancy‐compatible drug can effectively control maternal disease.” ¹³ Two guidelines included recommendations on secukinumab use in pregnancy ¹⁴ , ²⁵ ; one guideline recommended discontinuation, ²⁵ and one recommended conditional use if no other medication was suitable. ¹⁴ Rituximab was specifically addressed in six guidelines, with half (n = 3) ¹³ , ¹⁶ , ²⁰ recommending cessation and the others (n = 3) ¹⁴ , ²⁵ , ²⁷ conditionally recommending continuing if no alternate option and required for disease control in pregnancy.

Recommendations for tsDMARDs

Six guidelines ¹³ , ¹⁴ , ¹⁶ , ²⁰ , ²¹ , ²⁷ included recommendations for tsDMARD use in pregnancy. There was low‐level to no evidence in this area, and all guidelines recommended avoiding tsDMARDs due to insufficient data on use in pregnancy. These recommendations had a high level of agreement within individual panels.

Recommendations for the preconception state

Recommendations for csDMARDs

Ten guidelines ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ²⁰ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ included preconception recommendations for csDMARD use. Evidence was limited to cohort studies, with a high level of agreement within individual panels. These guidelines all had strong recommendations regarding methotrexate, with eight guidelines (80%) recommending cessation three months before conception ¹³ , ¹⁵ , ¹⁷ , ²⁵ , ²⁶ , ²⁷ , ²⁸ and two guidelines (20%) recommending cessation at least one month before. ¹⁴ , ²²

Six guidelines included recommendations for leflunomide, ¹³ , ¹⁴ , ²⁰ , ²⁵ , ²⁷ , ²⁸ and all recommended either a washout with cholestyramine or cessation two years before conception, or both. A blood test to check drug levels before conception was recommended by three of these guidelines (50%). ²⁵ , ²⁷ , ²⁸ Two guidelines included recommendations for preconception use of sulfasalazine and hydroxychloroquine ¹⁴ , ²⁵ ; continuation of these drugs was recommended by both societies.

Recommendations for bDMARDs

Four guidelines included preconception recommendations for bDMARD use ¹³ , ²⁰ , ²⁵ , ²⁶ ; notably, one based their recommendations on another included guidelines’ recommendations. ²⁰ There was low‐level evidence to support these recommendations. Regarding TNF inhibitors, three guidelines had conditional recommendations to continue these drugs preconception, ¹³ , ²⁵ , ²⁶ with two guidelines sharing a strong recommendation to continue certolizumab specifically. All four guidelines did not recommend continuing rituximab before conception, one guideline specified cessation 12 months before, ²⁶ and one guideline specified 6 months before. ²⁰ Two guidelines recommended discontinuing abatacept before conception, ¹³ , ²⁶ and one guideline specified 10 weeks before conception. ²⁶ The same two guidelines recommended ceasing tocilizumab; one guideline specified three months (5.5 half‐lives) before conception. ²⁶ One recommended cessation of anakinra before conception. ¹³ Notably, one guideline conditionally recommended continuing abatacept, tocilizumab, anakinra, and secukinumab during the preconception stage and ceasing once pregnant. ²⁵

Recommendations for tsDMARDs

Two guidelines included preconception recommendations for tsDMARD use, specifically tofacitinib. Both recommended cessation in the preconception stage, with differing time frame (two months before ¹³ vs two weeks before ¹⁴ ).

Recommendations for breastfeeding

Recommendations for csDMARDs

Eleven guidelines included recommendations for csDMARD use in breastfeeding, ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ¹⁸ , ²⁰ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ with low‐to‐moderate levels of evidence available (Table 2). There were uniform recommendations for methotrexate (n = 10) ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ²⁰ , ²² , ²⁵ , ²⁶ , ²⁷ , ²⁸ and leflunomide (n = 7) ¹³ , ¹⁴ , ²⁰ , ²⁵ , ²⁶ , ²⁷ , ²⁸ that these two medications were not safe for use in lactation. Guidelines also had shared recommendations to continue hydroxychloroquine (n = 7). ¹³ , ¹⁴ , ¹⁸ , ²⁰ , ²⁵ , ²⁶ , ²⁷ Although there was also consensus that sulfasalazine was compatible in breastfeeding (n = 5) ¹³ , ¹⁴ , ²⁰ , ²⁵ , ²⁶ for healthy, full‐term infants, one of these guidelines specifically advised precaution in use with infants who had glucose‐6‐phosphate dehydrogenase deficiency, hyperbilirubinemia, or prematurity. ²⁶

Table 2.

Recommendations for disease‐modifying antirheumatic drug use in breastfeeding*.

Recommendations for bDMARDs

Eight guidelines included recommendations for bDMARD use in breastfeeding. ¹³ , ¹⁴ , ¹⁶ , ²⁰ , ²¹ , ²⁵ , ²⁷ , ²⁸ Three guidelines, all published before 2017, recommended globally avoiding bDMARDs due to insufficient evidence. ²¹ , ²⁷ , ²⁸

The remaining five guidelines uniformly recommended that TNF inhibitor use was appropriate during breastfeeding, however, there were different recommendations for use of other bDMARDs during lactation. Notably, one article ¹³ stated that there was high disagreement between experts (25–31%) on clinical use of bDMARDs without data on transfer into breastmilk (abatacept and tocilizumab). Use of abatacept and tocilizumab in breastfeeding was conditionally recommended in two guidelines, ¹⁴ , ²⁵ and secukinumab was felt compatible with breastfeeding in one guideline. ¹⁴ Anakinra was also conditionally recommended in two guidelines ¹⁴ , ²⁵ ; one guideline recommended avoidance due to no data. ¹³ Rituximab had varying recommendations among guidelines; one guideline strongly recommended this was compatible, ²⁵ one guideline provided a conditional recommendation to continue, ¹⁴ and three guidelines specifically advised avoidance in lactation. ¹³ , ¹⁶ , ²⁰

Recommendations for tsDMARDs

Five guidelines included recommendations for tsDMARD use in breastfeeding. ¹³ , ¹⁴ , ²⁰ , ²¹ , ²⁷ There are no current data available, and all guidelines recommended avoidance during lactation, with a high level of agreement among panels.

Recommendations for male fertility

Recommendations for csDMARDs

Five guidelines included recommendations for csDMARD use in male fertility. ¹⁴ , ¹⁵ , ¹⁷ , ²⁵ , ²⁸ Four guidelines included conditional recommendations for methotrexate; three guidelines (75%) recommended continuation of methotrexate with a moderate level of evidence, ¹⁴ , ¹⁷ , ²⁵ with one guideline specifying “paternal exposure to low‐dose [at <25 mg/wk] [methotrexate] is compatible with pregnancy.” ¹⁴ One guideline recommended cessation of methotrexate three months before conception, with a low‐grade level of evidence (based on expert opinion). ¹⁵ Leflunomide also had differing recommendations, with two guidelines recommending conditional continuation in men (2 of 3, 66.7%), ¹⁴ , ²⁵ and one recommending discontinuation two years before or washout with cholestyramine and measurement of drug levels before conception in “both parents (men and women)…planning pregnancy.” ²⁸ In two guidelines addressing hydroxychloroquine and sulfasalazine use in male fertility, ¹⁴ , ²⁵ there was shared strong recommendation for continuing hydroxychloroquine and conditional continuation of sulfasalazine. However, one guideline suggested performing semen analysis and ceasing sulfasalazine if conception is delayed by 12 months. ²⁵

Recommendations for bDMARDs

Three guidelines included recommendations for bDMARD use in male fertility. ¹⁴ , ¹⁶ , ²⁵ There was a low level of evidence available for these recommendations. All three guidelines recommended conditionally continuing rituximab in men. When included, there was also 100% agreement of recommendations to continue TNF inhibitors (2 of 2), anakinra (2 of 2), abatacept (2 of 2), and tocilizumab (2 of 2) in men trying to conceive.

Recommendations for tsDMARDs

Two guidelines included recommendations for tsDMARD, specifically tofacitinib, use in male fertility. ¹⁴ , ¹⁶ One guideline recommended continuation, although with limited evidence, ¹⁴ and one guideline recommended avoidance due to insufficient data. ¹⁶

Recommendations for postpartum newborn health

Two guidelines ¹⁴ , ²⁷ included recommendations for newborn health, specifically pertaining to maternal bDMARD use. Both had shared recommendations to discontinue TNF inhibitors (except certolizumab) in the third trimester to avoid risk of infection in the newborn; one guideline suggested ceasing all at 30 weeks, ²⁷ and the other had more specific recommendations ¹⁴ (adalimumab cessation at 28 weeks gestation, etanercept cessation at 32 weeks gestation, golimumab cessation at 28 weeks gestation, and infliximab cessation at 20 weeks gestation). One guideline had recommendations regarding newborn vaccinations ¹⁴ ; if abatacept, tocilizumab, rituximab, anakinra, or TNF inhibitors (except certolizumab) are used in the third trimester, all live vaccinations should be avoided until the infant is six months old. The same guideline recommended that certolizumab may be safely used throughout pregnancy without alteration to infant vaccination schedule. There were no specific recommendations pertaining to maternal use of csDMARDs or tsDMARDs and newborn health.

Other recommendations for drugs outside of inclusion criteria

Glucocorticoids

Seven guidelines ¹³ , ¹⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ had recommendations regarding the use of glucocorticoids in pregnancy. There was a moderate level of evidence, with six guidelines providing conditional recommendations for use of glucocorticoids such as prednisolone in pregnancy. All of these guidelines recommended using the lowest dose possible, and two guidelines specifically recommended daily dose equivalent of <20 mg of prednisolone. ¹⁴ , ²⁵ One guideline specific to SLE and antiphospholipid syndrome recommended avoiding routine use of glucocorticoids in pregnancy. ²⁹

Limitations

This scoping review was limited by the available low‐quality evidence and minimal existing literature of use of DMARDs in reproductive health, in particular, for newer medications. The lack of evidence restricted recommendation formulations for bDMARDs and tsDMARDs, with generation of mostly conditional recommendations only. There was also distinct paucity of recommendations for DMARD use in male fertility and newborn health. Our review also only included guidelines from rheumatologic societies and hence broadening the scope to include other specialties that manage autoimmune conditions (such as renal and gastroenterology) would strengthen a future review of existing data and recommendations for DMARD use in reproductive health.

DISCUSSION

Guidelines for the use of DMARDs in reproductive health are readily available, however, they vary regarding specific recommendations. Evidence for use of drugs in pregnancy is often scant, given significant ethical implications for controlled trials in this cohort. Consideration is required for both maternal and fetal outcomes, and many guidelines rely on preclinical research, cohort studies, and retrospective review of drug exposure during pregnancy to formulate recommendations.

This review identified that there were more uniform recommendations for older medications (csDMARDs) between guidelines, likely due to greater existing literature of use in reproductive health. Recommendations regarding bDMARDs varied significantly between guidelines, with a clear temporal trend observed to favor more conditional recommendations for use in newer guidelines. This is observed over a 10‐year period (2014–2024) and reflects growing literature available from retrospective cohort studies of use and exposure in peripregnancy settings, in addition to longevity in market of medications. The earliest class of bDMARDs approved by the US Food and Drug Administration were TNF inhibitors (etanercept; 1998), and guidelines from 2019 appear to favor use in pregnancy, compared to more recently approved bDMARDs such as abatacept (2005), which guidelines up to 2023 still recommended avoiding in the peripartum setting. Given biologic drugs are recombinant proteins typically similar to maternal IgG antibodies, there is varying ability for these drugs to physically cross the placenta and have direct effect on the fetus, ⁸ and emerging research on placental transfer may also be contributing to the gradually increasing confidence in use of DMARDs when necessary for disease control. A preliminary update of EULAR Points to Consider for Use of Antirheumatic Drugs in Reproduction, Pregnancy, and Lactation has been recently presented ³⁸ and includes recommendations in favor of the use of all TNF inhibitors in pregnancy and the use of other bDMARDs such as abatacept, rituximab, tocilizumab, anakinra, and secukinumab if needed to control maternal disease. These significant changes since the previous points to consider published in 2016 further support the observation that relatively rapid changes in recommendations are occurring in this field.

Currently, the newest agents, tsDMARDs, had very limited recommendations due to a lack of evidence. Notably, tsDMARDs are used in other autoimmune conditions such as inflammatory bowel disease (IBD), and there is emerging registry data on tsDMARD exposure during pregnancy in this group of patients, ³⁹ with recent Pregnancy in IBD and Neonatal Outcomes consensus statements favoring the use of JAK inhibitors if “no viable alternatives to maintain maternal health.” ⁴⁰

Although there was overall uniformity among drug recommendations in the preconception state, guidelines differed in duration that a medication needed to be held before pregnancy. Whereas timing of drug cessation before conception was specified, only one guideline clearly stated this was in relation to drug half‐life. ²⁶ There were scant recommendations pertaining to DMARD use in male fertility, and, interestingly, there was general agreement among guidelines on bDMARD use but different recommendations for csDMARD use in this setting, with overall very limited data available. The preliminary updated EULAR points to consider provides a broader list of DMARDs that have not demonstrated a clinically relevant impact on fertility or pregnancy outcome, ³⁸ including sulfasalazine and methotrexate, highlighting the value of ongoing research on the impact of these drugs on sperm. ⁴¹

Only two guidelines discussed newborn health guidelines with maternal bDMARD use, ¹⁴ , ²⁷ and only one guideline provided recommendations on alteration of vaccination schedules in newborns. ¹⁴ Although it is necessary to exercise caution in use of any new medication in the context of pregnancy and potential implications on fetal developmental outcomes, initial studies suggest that there is no interference of bDMARD use up to infancy ⁴² ; however, more literature is required to strengthen recommendations in this area.

The process of formulating recommendations was similar among guidelines; however, only five guidelines included consumers on their voting panel. Notably, women with autoimmune disease are particularly vulnerable during pregnancy, with high rates of ambivalence and depression, ⁴³ and a retrospective study in women with AIRD revealed several unmet needs in knowledge about reproductive issues, with concerns that both active disease and medications required to control disease may harm the newborn. ⁴⁴ This is supported by a study by Birru Talabi et al ⁴⁵ that found up to 80% of women chose to discontinue DMARDs, including nonfetotoxic agents, and risked enduring inflammatory arthritis due to concerns of safety during pregnancy, despite emerging literature that there is minimal safety concerns with use of bDMARDs in pregnancy but higher risk of adverse pregnancy outcomes with uncontrolled inflammatory disease. ⁸ This highlights the need for consumer representation and integration of the patient perspective in formulation of these guidelines to support personalized discussions between clinicians and patients, as counseling and a multidisciplinary approach appears to herald a positive effect on family planning and confidence in decision‐making. ⁴⁴ , ⁴⁶

AUTHOR CONTRIBUTIONS

All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding author, Dr Chin confirms that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.

Supporting information

Disclosure Form: