Abstract
Fertility is impaired in female patients with rheumatoid arthritis (RA), which is related to disease activity and the use of certain medication. During pregnancy, disease activity usually improves, but less than previously thought. Especially in women with high disease activity, the pregnancy outcome is also impaired. All of this underscores the importance of strict control of disease activity in RA patients who wish to conceive.
Management of RA disease activity during pregnancy might be a challenge as the treatment options are limited. Evidence is accumulating that tumor necrosis factor (TNF) blockers can be safely used during pregnancy, particularly during the first trimester and the beginning of the second trimester.
Far less is known about the problems faced by male RA patients who wish to conceive, in terms of not only fertility and pregnancy outcome but also the safety of medication.
In this paper, the fertility issues in patients with RA, the pregnancy-associated improvement of RA, the pregnancy outcomes, including the long-term effects on the offspring, and treatment options, including those during lactation and for male patients wishing to conceive, will be reviewed.
Case report
Mrs. A, a 41-year-old patient with rheumatoid arthritis (RA), visited the specialized outpatient clinic of the Department of Rheumatology for consultation on her wish to conceive. Her medical history revealed diabetes mellitus type 1 since the age of 25. At the age of 33, she was diagnosed with a rheumatoid factor (RF)- and anti-citrullinated protein antibody (ACPA)-positive, erosive RA. She and her partner had been trying to conceive for >7 years, but this had not resulted in a live birth. The fertility work-up did not reveal any cause for the couple׳s subfertility. In their attempts to conceive, the couple had undergone intrauterine insemination (IUI) 6 times and an in vitro fertilization (IVF) procedure twice. In both IVF procedures, the embryos could be transferred into the uterine cavity. During her attempts to conceive, the patient was treated with sulfasalazine and etanercept, but she reported active RA during this period. At the age of 40, she spontaneously became pregnant when being treated with methotrexate (MTX) and etanercept. Unfortunately, this resulted in a miscarriage at week 7 of gestation.
At the first consultation, the patient still showed active RA (DAS28 3.5; five swollen joints) despite treatment with etanercept and MTX. MTX and etanercept were stopped, and the patient was started on 400 mg of infliximab (5 mg/kg bodyweight) every 6 weeks, 1000 mg of sulfasalazine twice daily, and 400 mg of hydroxychloroquine once daily. Four months thereafter, the RA was in clinical remission (DAS28 2.6; 1 swollen joint); after careful consideration and gynecological consultation, IUI was restarted, as the couple decided not to undergo IVF anymore. The second IUI attempt resulted in a successful pregnancy. At week 20 of gestation, infliximab was stopped and 200 mg of certolizumab every second week was started. The patient gave birth to a healthy baby girl at week 38.0 of gestation. Both infliximab and certolizumab could not be detected in the cord blood. Throughout pregnancy, the RA remained in remission and the diabetes was well controlled. Ten weeks after delivery, the patient experienced a flare of her arthritis for which MTX was added to the combination therapy of certolizumab, sulfasalazine, and hydroxychloroquine. At the following visit, 3 months later, the RA was again in clinical remission.
Fertility in patients with RA
Conceiving a child is a major life event, and most adults try to have a child during their reproductive life span. However, the literature suggests that achieving this goal might be more difficult for women with RA.
Several studies indicate that family size is diminished in women with RA . Some studies , but not all , indicate that this smaller family size is already present before the actual diagnosis of RA. Apart from a smaller family size in women with RA, such patients experience more difficulties in conceiving, as indicated by a longer time to pregnancy (TTP). In a study performed in the Danish National Birth Cohort, it was shown that 25% of patients with RA took more than a year to conceive compared with 15.6% of controls . As only women who gave birth were included in this birth cohort, the actual data might be more concerning, as data on women who do not conceive at all were not incorporated. Clowse et al. reported that 36% of patients with RA had difficulties in conceiving at least once during their reproductive life span. In the Pregnancy-induced Amelioration of Rheumatoid Arthritis (PARA) study, a prospective Dutch cohort study on RA and pregnancy , it was shown that 42% of patients with RA did not conceive within 1 year or at all during the follow-up period of that study. For comparison, in the general population, the median prevalence of subfertility defined as TTP of >12 months is 9%, with a range of 3.5–24.2% depending on the geographic area .
Several factors might be associated with the reported smaller family size and the longer TTP in female patients with RA. As impaired fertility might already be present before the diagnosis of RA and as earlier menopause has been reported in patients with RA , it has been postulated that these patients have a smaller ovarian reserve, which is already present before the onset of the disease and could account both for the impaired fertility and for the earlier menopause . To test this hypothesis, the levels of serum anti-Müllerian hormone (AMH) (the most reliable biomarker for ovarian reserve) were tested in 72 premenopausal women (age range 18–42 years) with early RA. No differences were observed between patients with RA and healthy controls, making it unlikely that the observed impaired fertility, already present at time of diagnosis, is related to a diminished ovarian reserve . Personal choices, due to RA-related concerns, have been shown to be at least partially responsible for the smaller family size , but it cannot account for the observed impaired fertility. Inflammation, as reflected by disease activity, has been shown to be associated with TTP. In women with active disease (DAS28 > 5.1), TTP exceeded 1 year in 67% of women, whereas this was only 30% in women in remission (DAS28 < 2.6) . The use of nonsteroidal anti-inflammatory drugs (NSAIDs) has been associated with increased TTP , most likely through inhibition of the production of prostaglandins. Prostaglandins are involved in ovulation and implantation . In addition, prednisone in a dose exceeding 7.5 mg daily has been associated with prolonged TTP . The effect of prednisone might be related to a transient suppression of the hypothalamic–pituitary–ovarian axis by glucocorticoids or by a direct effect on ovarian function or on the endometrium . Finally, mainly based on studies in oncology and in animal models, it has been suggested that prior treatment with MTX is associated with impaired fertility . However, in a prospective cohort study on RA and fertility, prior MTX treatment did not affect TTP . Furthermore, short-term MTX treatment had no impact on the ovarian reserve in patients with RA . Finally, it has been hypothesized that the impaired fertility in patients with RA is a result of a lower intercourse frequency . Although a high prevalence of sexual dysfunction has been described in mainly postmenopausal RA patients , it is unclear whether any conclusions can be drawn from these studies on the intercourse frequency of young patients with RA who actively wish to conceive.
In conclusion, the doctor must be aware of the impaired fertility and consider various contributing factors, when consulting an RA patient who wishes to conceive.
Fertility in patients with RA
Conceiving a child is a major life event, and most adults try to have a child during their reproductive life span. However, the literature suggests that achieving this goal might be more difficult for women with RA.
Several studies indicate that family size is diminished in women with RA . Some studies , but not all , indicate that this smaller family size is already present before the actual diagnosis of RA. Apart from a smaller family size in women with RA, such patients experience more difficulties in conceiving, as indicated by a longer time to pregnancy (TTP). In a study performed in the Danish National Birth Cohort, it was shown that 25% of patients with RA took more than a year to conceive compared with 15.6% of controls . As only women who gave birth were included in this birth cohort, the actual data might be more concerning, as data on women who do not conceive at all were not incorporated. Clowse et al. reported that 36% of patients with RA had difficulties in conceiving at least once during their reproductive life span. In the Pregnancy-induced Amelioration of Rheumatoid Arthritis (PARA) study, a prospective Dutch cohort study on RA and pregnancy , it was shown that 42% of patients with RA did not conceive within 1 year or at all during the follow-up period of that study. For comparison, in the general population, the median prevalence of subfertility defined as TTP of >12 months is 9%, with a range of 3.5–24.2% depending on the geographic area .
Several factors might be associated with the reported smaller family size and the longer TTP in female patients with RA. As impaired fertility might already be present before the diagnosis of RA and as earlier menopause has been reported in patients with RA , it has been postulated that these patients have a smaller ovarian reserve, which is already present before the onset of the disease and could account both for the impaired fertility and for the earlier menopause . To test this hypothesis, the levels of serum anti-Müllerian hormone (AMH) (the most reliable biomarker for ovarian reserve) were tested in 72 premenopausal women (age range 18–42 years) with early RA. No differences were observed between patients with RA and healthy controls, making it unlikely that the observed impaired fertility, already present at time of diagnosis, is related to a diminished ovarian reserve . Personal choices, due to RA-related concerns, have been shown to be at least partially responsible for the smaller family size , but it cannot account for the observed impaired fertility. Inflammation, as reflected by disease activity, has been shown to be associated with TTP. In women with active disease (DAS28 > 5.1), TTP exceeded 1 year in 67% of women, whereas this was only 30% in women in remission (DAS28 < 2.6) . The use of nonsteroidal anti-inflammatory drugs (NSAIDs) has been associated with increased TTP , most likely through inhibition of the production of prostaglandins. Prostaglandins are involved in ovulation and implantation . In addition, prednisone in a dose exceeding 7.5 mg daily has been associated with prolonged TTP . The effect of prednisone might be related to a transient suppression of the hypothalamic–pituitary–ovarian axis by glucocorticoids or by a direct effect on ovarian function or on the endometrium . Finally, mainly based on studies in oncology and in animal models, it has been suggested that prior treatment with MTX is associated with impaired fertility . However, in a prospective cohort study on RA and fertility, prior MTX treatment did not affect TTP . Furthermore, short-term MTX treatment had no impact on the ovarian reserve in patients with RA . Finally, it has been hypothesized that the impaired fertility in patients with RA is a result of a lower intercourse frequency . Although a high prevalence of sexual dysfunction has been described in mainly postmenopausal RA patients , it is unclear whether any conclusions can be drawn from these studies on the intercourse frequency of young patients with RA who actively wish to conceive.
In conclusion, the doctor must be aware of the impaired fertility and consider various contributing factors, when consulting an RA patient who wishes to conceive.
RA disease course during and after pregnancy
Determination of the RA disease activity during pregnancy
As disease activity is associated with TTP as well as pregnancy outcome in patients with RA, accurately determining RA disease activity in pregnancy is important. However, this is difficult as several measures of determining disease activity or its components are influenced by pregnancy. For example, the erythrocyte sedimentation rate (ESR) is elevated in pregnant women due to increased circulating fibrinogen, plasma expansion, and decreased hemoglobin concentration . The ESR rises during pregnancy from 10 mm/h in the first trimester to 33 mm/h in the third trimester, and it declines post partum to <20 mm/h . Therefore, variants of the disease activity score (DAS) that include the ESR are not preferred as scoring methods in pregnancy . Pregnancy might also influence the visual analogue scale (VAS) of global health, which is incorporated in the DAS . These are some of the reasons for investigating validated alternative DAS scoring methods. Given that C-reactive protein (CRP) is only slightly influenced by pregnancy, RA disease activity during gestation can be most reliably determined with a DAS incorporating a swollen and tender joint count and a CRP, but without considering the VAS scores . The Clinical Disease Activity Index (CDAI) and Simplified Disease Activity Index (SDAI) have not been validated for use in pregnancy and have not been used in pregnancy studies.
RA disease course during pregnancy
Pregnancy is the most studied physiological condition in which RA remits spontaneously . This phenomenon was first described by Hench in 1938 . In this retrospective study, improvement was observed in 90% of the 34 included pregnancies. Following this, multiple retrospective studies between 1950 and 1989 confirmed his initial observation with improvement rates in pregnant RA patients ranging from 54% to 95% (see Table 1 ). Besides retrospective studies, a few small prospective studies within that period of time reported improvement in 71–86% of the patients. In 1993, Nelson et al. showed remission in 39% and improvement in 21% of the 18 prospectively and 39 retrospectively followed pregnancies.
| Study | Study type | Number of patients (pregnancies) | Improvement during pregnancy, % | Deterioration post partum, % |
|---|---|---|---|---|
| Hench | Retrospective | 20 (34) | 90 | 90 |
| Lewis-Faning | Retrospective | 22 | 95 | 81 |
| Torrent | Retrospective | 15 | 80 | – |
| Oka | Retrospective | 93 (114) | 75 | 84 |
| Hargreaves | Retrospective | 10 (11) | 91 | 91 |
| Smith | Prospective | 12 | 75 | |
| Betson | Retrospective | 24 | 54 | – |
| Morris | Retrospective | 17 | 82 | – |
| Neely | Retrospective | 20 | 63 | – |
| Ostensen | Retrospective | 31 (49) | 75 | 62 |
| Ostensen | Prospective | 10 | 90 | 100 |
| Unger | Prospective | 14 | 71 | – |
| Klipple | Retrospective | 93 (114) | 77 | >90 |
| Nelson | Partial retrospective and prospective | 41 (57, retrospective = 18, prospective = 39) | 60 (remission 39%) | – |
| Barrett | Prospective | 140 | 65 (remission 16%) | 62–77 |
| De Man | Prospective | 84 total, 52 with initial DAS28–CRP ≥3.2 | 48 (remission 27%) | 39 |
| De Man | Prospective | 118 | 39 and 75 (depending on the presence of autoantibodies) | 34 and 42 (depending on the presence of autoantibodies) |
After this period, multiple larger prospective studies were conducted. Barrett et al. prospectively studied 140 pregnant patients with RA in the UK to ascertain the influence of pregnancy on RA disease activity. Patients were followed up from the third trimester of their pregnancy until 6 months post partum. In this study, the disease activity was assessed by clinical examination of inflamed joints and a VAS and Health Assessment Questionnaire (HAQ) scoring system. Data were collected at the third trimester of pregnancy and 1 and 6 months post partum. The main disadvantage of this study is that the evaluation started at the third trimester, and therefore disease activity in the first and second trimester was retrospectively assessed by self-report. In this study, in total, 65% of the patients retrospectively reported a decrease in pain and swelling during pregnancy. However, only 16% reached complete remission in the third trimester, defined as no swollen joints and receiving no antirheumatic therapy . From 2002 until 2008, de Man et al. conducted the PARA study, a nationwide prospective cohort study within the Netherlands, to gain insight into the influence of pregnancy on the disease activity and the impact of RA and medication use on the pregnancy outcome. In this study, disease activity was objectively assessed with the validated DAS28–CRP scoring system. The patients were followed up from preconception (if possible) up to 26 weeks post partum. The results of the PARA study showed that 48% of the 52 patients with an initial DAS28–CRP ≥3.2 improved during pregnancy, based on the European League Against Rheumatism (EULAR) response criteria (good and moderate response combined). Improvement occurred although certain medications such as the contraindicated DMARDs were discontinued in all women wishing to conceive. When all patients, including those who were already in remission at the time of conception, were analyzed together, the mean DAS28 decreased from the preconception visit to the third trimester by 0.4 (from 3.8 to 3.4) with a more pronounced decrease in patients with a moderate to high disease activity in the first trimester. Nevertheless, the percentage of patients with moderate to high disease activity (DAS28 ≥ 3.2) in the third trimester was approximately 50%. Furthermore, in the third trimester, in total, 27% of the patients were in remission compared with 11% before conception, according to a DAS28 < 2.6 .
The differences in improvement rates between earlier and more recent studies can be explained by differences in study design (retrospective vs. prospective studies) and in patient selection; in some earlier studies, only patients with active disease were included. Also, various definitions of improvement and remission may contribute to differences in study outcome. In addition, increased numbers of patients with RA enter their pregnancy with low disease activity as treatment regiments have been intensified over the past decades.
RA disease course post partum
After delivery, there is an increased risk of a flare in disease activity. Post partum exacerbations were also described by Hench in 90% of the 34 pregnancies. Other retrospective studies report percentages of patients with a flare between 62% and 90% (see Table 1 ). In the prospective study by Barrett et al. , 66% of the women reported an increase in joint swelling and 77% in pain at 6 months post partum. Furthermore, in this study, 62% of the women had more affected joints post partum than in the third trimester. The median number of affected joints increased from eight during pregnancy to 10 at 6 months post partum .
De Man et al. reported that 39% of patients experienced a flare after delivery, despite restarting medication. Interestingly, also after miscarriage, one-third of patients experience a flare of their disease activity. In a recent study on a sample of 21 patients with RA, Brouwer et al. showed that in 33% the disease flares between the preconception period and 3 months after a miscarriage.
In conclusion, it must be noted that RA improves less and in fewer patients during pregnancy than is generally perceived. Therefore, when patients with RA conceive successfully, medication must be continued and only tapered with caution. In addition, after delivery and even during lactation, continuation of appropriate medication might be necessary to prevent a flare of RA disease activity.
Outcomes
Pregnancy outcomes
In large cohort studies, it has been found that pregnancy outcome in patients with RA is, in general, less favorable than in the normal healthy population , although not to the same extent as that reported for systemic lupus erythematosus (SLE) for example. Differences from the general population are often small; therefore, it is unclear whether the found differences have clinical implications for the individual patient. Notably, none of the studies on pregnancy outcome in female RA patients found increased risks of major congenital malformations or perinatal death .
An overview of studies on RA and pregnancy outcome is shown in Table 2 .
| Study | Study type | Number of patients | Findings on pregnancy outcome |
|---|---|---|---|
| Nelson | Case–control | 144 female RA patients versus 605 unaffected female controls | No increased risk:
|
| Skomsvoll | Retrospective (birth registry) | 4716 female patients with inflammatory arthritis (RA, juvenile RA, ankylosing spondylitis) versus 1,645,029 unaffected female controls | Increased risk:
|
| Wolfberg | Retrospective (birth registry) | 114 female RA patients with rheumatic disease (e.g., RA, systemic lupus erythematosus) versus 18,534 unaffected female controls | Increased risk:
|
| Reed | Retrospective (birth registry) | 243 female RA patients versus 2559 unaffected female controls | Increased risk:
|
| Norgaard | Retrospective (birth registry) | 1199 female RA patients versus 870,380 unaffected female controls | Increased risk:
|
| Lin | Retrospective (birth registry) | 1912 female RA patients versus 9560 matched controls | Increased risk:
|
| de Man | Prospective | 152 female RA patients versus 175,498 unaffected female controls (from the Netherlands Perinatal Registry (PRN)) | Increased risk:
|
| Langen | Retrospective | 46 female RA patients |
|
| Wallenius | Retrospective (birth registry) | 1496 female RA patients versus general Norwegian population | Increased risk:
|
| Rom | Retrospective (birth registry) | 2101 female RA patients versus 1,915,622 unaffected female controls | Increased risk:
|
| Bharti | Prospective | 440 female RA patients | Increased risk:
|
| Brouwer | Prospective | 162 female RA patients versus general Dutch population | No increased risk:
|
Miscarriage
Brouwer et al. recently reported that the risk of a miscarriage in women with RA (17%) is comparable to that in the general population (11–22%). However, they do note that the miscarriage rate in patients with RA might be an underestimation. They explain that, among other reasons, the patients in their cohorts had planned pregnancies and therefore did not use any medication, for example, MTX, which is associated with an increased risk of miscarriages, and that there were fewer smokers and more patients with higher education levels in this cohort than in the general population .
Preeclampsia
Preeclampsia is a relatively uncommon pregnancy complication. Wolfberg et al. showed an increased risk of preeclampsia in 114 women with rheumatic disease (e.g., RA and SLE) compared with unaffected women (8.8% vs. 2.3%). In a large combined Swedish and Danish prevalence study from 1994 to 2006 on 1199 women with RA and a first-time singleton birth registered in population-based health-care databases, Norgaard et al. showed that the risk of preeclampsia increased slightly from 3.4% in unaffected women to 5.0% in women with RA. Some studies did not report an increase in preeclampsia during pregnancy in patients with RA, but this is most likely the result of a lack of power.
Mode of delivery
In a cohort of 243 women from the Washington State birth records, Reed et al. found that women with RA had increased risks of a primary cesarean section (34% vs. 19.5% in the control group). The main reasons for cesarean section in the RA population were cephalopelvic disproportion, dysfunctional labor, and fetal distress in 32%, 31%, and 18%, respectively. These percentages were not significantly different from women without RA . In the previously mentioned study by Norgaard et al. , patients with RA had more cesarean deliveries (26.0% vs. 16.5% in unaffected controls). However, they did not report the indications for cesarean sections. In the study by de Man et al. , cesarean sections were performed more often in the group with DAS28–CRP ≥3.2 (22% vs. 10% in the group with DAS28–CRP <3.2) and an increased number of patients with RA needed an instrumental vaginal delivery (17% vs. 10% in the reference group), but the latter was not associated with disease activity. Cesarean sections were indicated in the majority of the patients due to breech delivery and failure of labor progression . Data on cesarean delivery should be interpreted with caution, as indication to perform a cesarean section may vary for different countries.
Preterm delivery
The definition of preterm delivery in these studies is a birth before 37 weeks of gestation. In the previously mentioned study by Wolfberg et al. and Norgaard et al. , the risk of preterm delivery was increased in women with RA compared with healthy controls (15.2% vs. 7.8% and 9.2% vs. 6.2%, respectively). Other studies report comparable risk percentages. In a recent prospective cohort study on 440 pregnant women with RA within the Organization of Teratology Information Specialists Autoimmune Diseases in Pregnancy Project, Bharti et al. showed that high disease severity (as defined by a HAQ Disability Index (HAQ-DI) >0.5) is associated with preterm delivery (20% vs. 12% in mild disease). In the PARA study by de Man et al. , the use of prednisone was associated with shorter gestational age (<37 weeks) of the offspring at delivery.
Small-for-gestational-age infants
As shown in Table 2 , multiple studies report an increased risk of small-for-gestational-age (SGA) infants ranging from ±3% in unaffected women to ±10% in women with RA. Some studies defined SGA as a birth weight more than two standard deviations below the mean for infants of similar gestational age , and others as a birth weight lower than the 10th percentile adjusted for the gestational age . De Man et al. demonstrated that an increase in RA disease activity during pregnancy is independently associated with lower birth weight in the offspring, although still within the normal range. In this study, the incidence of SGA infants was 3.3% .
Long-term effects on the offspring
Lower birth weight (even within the normal range) has been associated with an increased risk of cardiovascular and metabolic disease in adulthood . This effect is even more prominent if the children display rapid catch-up growth in weight during their first year of life. De Steenwinkel et al. examined growth charts of 167 children born to women with RA. Of these children, 28% expressed rapid catch-up growth in weight, which was associated with maternal disease activity. A selection of 108 of these children was again evaluated at around 7 years of age . At that age, these children did not have a high-risk profile in anthropometric measures, such as an increased blood pressure or altered body composition, compared with children born to healthy mothers.
Maternal use of prednisone during pregnancy has been associated with slightly higher daytime cortisol levels in children of age 7 . The clinical consequence of this is not clear, as this observation was not associated with clinical signs of a higher cortisol level.
In conclusion, pregnancy outcome in patients with RA is only slightly impaired, but it is not of clinical relevance to the individual patient, especially when disease activity is well controlled. This is reassuring for RA patients wishing to conceive in terms of the impact of (well-controlled) RA on pregnancy outcome.
Related posts:
Treating to target in established rheumatoid arthritis: Challenges and opportunities in an era of novel targeted therapies and biosimilars
How does established rheumatoid arthritis develop, and are there possibilities for prevention?
Monitoring in established RA: Role of imaging and soluble biomarkers
Evaluating and mitigating fracture risk in established rheumatoid arthritis
Patient-centred care in established rheumatoid arthritis
Quality in rheumatoid arthritis care
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
