Executive Summary: British Society for Rheumatology guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice

NICE has accredited the process used by BSR to create its clinical guidelines. The term began on 27 February 2012 and the current renewed accreditation is valid until 31 December 2023. More information on accreditation can be viewed at www.nice.org.uk/accreditation. The rationale behind this update on the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding [1, 2] was described in detail in the guideline scope [3]. In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy [4–7] the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths [8]. Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities [8]. In all decisions regarding medication choices and changes, it is also important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). Therefore, the exposure of the foetus to different drugs when switches are made must be balanced against possible foetal gains and understanding the potential impact of reduced control of the medical disorder on a pregnancy is vital [9]. Patients with inflammatory rheumatic disease (IRD) should be counselled to achieve and then maintain remission or low disease activity before/during pregnancy to reduce the risk of adverse pregnancy outcomes [10]. This goal is primarily achieved through adjustment of therapy to ensure disease control with disease modifying anti-rheumatic drugs (DMARDs) and/or immunosuppressive drugs that are compatible with pregnancy. These medications are reviewed in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids [11]. Many patients with IRD, however, have an additional burden of pain and comorbid illness [12] that require treatment with other medications. The compatibility of various comorbidity medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients to ensure more confident prescribing in these scenarios and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and Faculty of Sexual and Reproductive Healthcare [13, 14]. To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease for the following drug categories: pain management; non-steroidal anti-inflammatory drugs (NSAIDs); low-dose aspirin (LDA); anticoagulants; colchicine; dapsone; bisphosphonates; anti-hypertensives; and pulmonary vasodilators. This revised guideline was produced by consensus review of current evidence to answer specific questions in relation to each drug as follows. Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure. The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men planning to conceive, and patients who have accidentally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, renal physicians, dermatologists and general practitioners who may prescribe these medications to patients in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth [15]. This guideline does not cover the management of infertility or acute pain relief during labour, hence morphine was excluded. Other drug categories: antimalarials; corticosteroids; disease modifying anti-rheumatic and immunosuppressive therapies; and biologic drugs are considered in another guideline [11]. This guideline was commissioned by the BSR Standards, Audit and Guidelines Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders (Table 1). In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website. Composition of guideline working group; list of group members and relevant stakeholders All members were involved in data review, formulation of recommendations and editing of the manuscript. Composition of guideline working group; list of group members and relevant stakeholders All members were involved in data review, formulation of recommendations and editing of the manuscript. The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Tetralogy Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement. Most medications covered in this guideline have been comprehensively and systematically reviewed in multiple other documents, since the first BSR guideline on this topic. Therefore, a consensus-based approach was taken to compile and assess most significant evidence published since 2013 to December 2020 through a comprehensive search of MEDLINE, PubMed and EMBASE databases with specific search terms (Supplementary Table S1, available at Rheumatology online). Filters were applied to capture National Institute for Health and Care Excellence (NICE) guidance, international guidelines, systematic reviews, cohort studies or case-series. Information was preferentially selected from NICE guidance and/or largest/most recent systematic reviews and where lacking was extracted from largest cohort, case-series or abstract. Findings were cross-referenced with the previous BSR guideline [2], as well as the Cochrane, Lactmed (a National Library of Medicine database on drugs and lactation) and UKTIS databases. Two independent reviewers screened the title and abstract of 2997 articles, identified 130 and selected the most recent/largest systematic reviews or largest cohort study or case-series as well any NICE guidance and international guidelines. Thirty-six studies (Fig. 1) met the inclusion criteria and relevant information was extracted into data-extraction tables. Flow diagram of study selection The working group met regularly to formalise search strategy, review evidence, resolve disagreements and finally to determine recommendations. This guideline was developed in line with BSR’s Guideline Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA). Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk or more uncertain. Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C) reflecting the confidence in the estimates of benefits or harm. The wording of each recommendation was discussed until all members were satisfied they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement) and then 20/23 members with full voting rights scored each recommendation on the same scale and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and a data-analyst expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations, so while they fully agreed with each, they did not wish to score each one and did not contribute to the final SOA score. The search was conducted in December 2020. Limits were placed for English language and filters as described above. The guideline will be updated in five years. Drugs are considered in the following categories: pain management; NSAIDs and low-dose aspirin in the management of multisystem rheumatic disease; anticoagulants; bisphosphonates; anti-hypertensive medication in the management of multisystem rheumatic disease; and pulmonary vasodilators. The overall findings for maternal and foetal breastmilk exposures to each drug, including information and key references from the previous BSR guideline [2] are summarised in the full-length guideline and updated recommendations listed below. Paternal exposures and recommendations are described separately after maternal data. An overall summary of compatibility of each drug pre-conception, during pregnancy, breastmilk exposure and paternal exposure is shown in Table 2. Generic recommendations were developed based on evidence as shown in Supplementary Table S2, available at Rheumatology online. Summary of drug compatibility in pregnancy and breastfeeding For further information and caveats, see relevant recommendations and main text in the Executive Summary and full Guideline. Intermittent use advised—see main text for details. Based on limited data and no association with adverse foetal development or pregnancy outcome, therefore unlikely to be harmful. Limited evidence, but unlikely to be harmful. Limited evidence regarding use for treatment of chronic pain in pregnancy. High dose folic acid (5 mg/day) recommended. Cessation of anti-depressant therapy in post-natal period is not recommended. Possible association with miscarriage and malformation. Drugs not included in systematic search, but added because of their relevance. Summary of drug compatibility in pregnancy and breastfeeding For further information and caveats, see relevant recommendations and main text in the Executive Summary and full Guideline. Intermittent use advised—see main text for details. Based on limited data and no association with adverse foetal development or pregnancy outcome, therefore unlikely to be harmful. Limited evidence, but unlikely to be harmful. Limited evidence regarding use for treatment of chronic pain in pregnancy. High dose folic acid (5 mg/day) recommended. Cessation of anti-depressant therapy in post-natal period is not recommended. Possible association with miscarriage and malformation. Drugs not included in systematic search, but added because of their relevance. Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant experience as appropriate to optimise all therapy, including non-pharmacological options for chronic pain management during pregnancy (GRADE 1A, SOA 99.5). The risks and benefits of drug treatment to mother and foetus should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99). The cause of pain and other symptoms should be assessed and managed appropriately (GRADE 1B, SOA 98.5). The requirement for analgesia should be assessed and minimum effective dose should be prescribed and titrated according to response (GRADE 1B, SOA 100). Tricyclic antidepressants are preferred over other antidepressant medications to manage chronic pain (GRADE 1B, SOA 98.1). Cessation of anti-depressant therapy that is being used as chronic pain medication in the post-natal period is not recommended, due to the risk of adverse impact on mood (GRADE 1C, SOA 96). LDA (≤150 mg/day) is recommended in all patients at high risk for pre-eclampsia (GRADE 1A, SOA 99.5). Low molecular weight heparin is the preferred anticoagulant (GRADE 1A, SOA 100). Nifedipine is the preferred vasodilator (GRADE 1B, SOA 98.5). Paternal drug exposure may reduce male fertility but has not been associated with adverse foetal development or pregnancy outcome. Although evidence is weak, we recommend that men are reassured about the safety of fathering a pregnancy whilst taking medicines to manage comorbidities as described in this guideline (GRADE 1C, SOA 98). Recommendations were based on two systematic reviews [16, 17], UKTIS [18], Royal College of Obstetricians (RCOG) guidance [19] and LactMed [20], as shown in Supplementary Table S3, available at Rheumatology online. Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5). Recommendations were based on a systematic review [16] (including a large Norwegian population-based cohort study [21], a large case-control study [22] and a case-control study [23]), UKTIS [18], breastfeeding exposures from references [24–26] and LactMed [20], as shown in Supplementary Table S3, available at Rheumatology online. Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98). Recommendations were based on a systematic review [16] (including a large prospective cohort study of over 1.6 million women [27]), UKTIS [18] (including case reports [28]), LactMed [20] (including [29]) and RCOG guidance [19], as shown in Supplementary Table S3, available at Rheumatology online. Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8). Recommendations were based on NICE guidance [30], a systematic review [16], UKTIS [18] and LactMed [20], as shown in Supplementary Table S4, available at Rheumatology online. Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100). Recommendations were based on [16, 18, 19, 31–35] and [20], as shown in Supplementary Table S4, available at Rheumatology online. Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3). Recommendations were based on NICE guidance [30], two systematic reviews [16, 33], UKTIS [18] and LactMed [20], as shown in Supplementary Table S4, available at Rheumatology online. Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8). Recommendations were based on [16, 18, 30, 36–39] and [20], as shown in Supplementary Table S4, available at Rheumatology online. Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3). Recommendations were based on NICE guidance [40], four systematic reviews [16, 41–43], a case report/review [44], UKTIS [18], LactMed [20] and United States Food and guidance as shown in Supplementary and available at Rheumatology online. findings from large studies with on the use of NSAIDs in the first trimester of pregnancy the of a low risk of miscarriage and malformation. Therefore, these drugs should only be used in the first trimester of pregnancy (GRADE 1B, SOA Intermittent than use of all NSAIDs LDA is recommended throughout pregnancy and from end of trimester to by to of the (GRADE 1B, SOA 98). there are limited data on they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). NSAIDs are compatible with breastfeeding (GRADE 1C, SOA 98.8). LDA of may be throughout pregnancy and NICE guidelines for in pregnancy treatment with LDA of until (GRADE 1B, SOA LDA is compatible with breastfeeding (GRADE 2C, SOA There are limited data on but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA Recommendations were based on and [20], as shown in Supplementary Table available at Rheumatology online. therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). may be used in pregnancy (GRADE 2C, SOA may be used in breastfeeding (GRADE 2C, SOA 98.3). may be used in breastfeeding and due to the risk of it is advised to the infant for of in or breastfed infants (GRADE 2C, SOA Recommendations were based on and [20], as shown in Supplementary Table available at Rheumatology online. Low molecular weight heparin is compatible throughout pregnancy (GRADE 1A, SOA 100). is compatible with breastfeeding (GRADE 1C, SOA 100). The use of in pregnancy is associated with increased foetal risk throughout pregnancy and has limited therefore should only be considered in (GRADE 1B, SOA 98.8). is compatible with breastfeeding (GRADE 1A, SOA 100). be recommended in pregnancy (GRADE 1C, SOA may be considered in breastfeeding (GRADE 2C, SOA Other are not recommended in breastfeeding due to to data and from studies (GRADE 1C, SOA may be considered in pregnancy and breastfeeding if there is an or adverse response to (GRADE 2C, SOA Recommendations were based on and [20], as shown in Supplementary Table available at Rheumatology online. There is data to recommend in pregnancy or to a specific for them to be stopped their in of up to and no evidence of from limited reports of their use in pregnancy, a recommendation is that they should be stopped in of pregnancy (GRADE 2C, SOA There are no data on to a recommendation for the use of during breastfeeding (GRADE 2C, SOA 98.5). Recommendations were based on and [40], as shown in Supplementary Table available at Rheumatology online. and should be stopped as as possible when pregnancy is in the first trimester and if an alternative compatible with pregnancy given (GRADE 1A, SOA 100). should be avoided in the and trimester but may be considered advice in (GRADE 1C, SOA 98.5). Based on limited evidence, is compatible with breastfeeding (GRADE 2C, SOA 98.5). Recommendations were based on as shown in Supplementary Table available at Rheumatology online. Nifedipine is compatible with pregnancy with no evidence of at doses up to (GRADE 1A, SOA Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). can be considered in pregnancy and breastfeeding as there is no evidence of (GRADE 1C, SOA Recommendations were based on 18, and [20], as shown in Supplementary Table available at Rheumatology online. pulmonary a to pregnancy. pregnancy the use of these pulmonary vasodilator drugs in pregnancy should be considered only as of a (GRADE 1C, SOA 99.5). Limited evidence the use of to during pregnancy (GRADE 2C, SOA Limited evidence the use of to during pregnancy (GRADE 2C, SOA is in and although there is no evidence of from pregnancy, the evidence is to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data to breastfeeding exposure to pulmonary on to a recommendation (GRADE 2C, SOA 98.8). Recommendations were based on and as shown in Supplementary Table available at Rheumatology online. Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). NSAIDs are compatible with paternal exposure (GRADE 1C, SOA Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline is unlikely to be (GRADE 2C, SOA of these guidelines will clinical practitioners and patients in and will be through at and national to of these guidelines are Patients with rheumatic disease should counselling and then be reviewed during pregnancy and the period by clinical practitioners with in the management of rheumatic disease in pregnancy, in to their obstetric should have to information on relevant medications in pregnancy and breastfeeding that is and them to decisions regarding compatibility of drugs in pregnancy. The of current evidence the for a national pregnancy for patients with rheumatic as for women with All women with rheumatic disease who pregnant would be to or not they are on anti-rheumatic The prospective pregnancy data would then be published to information on outcomes as miscarriage and in patients with anti-rheumatic and other drug These data would also be used to answer specific questions where data is Data to the impact of paternal exposure to these drugs fertility and as well as breastfeeding exposure is and further in these is An to assess with these guidelines is shown in Supplementary Data S1, available at Rheumatology online. The full guideline is available at Rheumatology online. Supplementary data are available at Rheumatology online. All relevant data produced during the guideline development process are in the guideline or in the specific was from any in the or to the work described in this manuscript. All have made declarations of in line with BSR Supplementary available at Rheumatology online). has from and from and has from and has from has from and from and and has from has from and and from and has from has and from has and from has from and is an of and independent advice to the working The have no of A full list of Standards, Audit and Guidelines Working members can be found in Supplementary Data S2, available at Rheumatology online. All members are in the This is the summary of Society for Rheumatology guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology For the full guideline, see