Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Rheumatoid arthritis (RA) is an autoimmune, chronic inflammatory disorder of unknown etiology occurring in about 0.5–1 % of the population worldwide. Women are three times more often affected than men with a peak onset of RA between 35 and 55 years of age. The disease is characterized by symmetrical polyarthritis particularly in small and medium size joints, and extra-articular manifestations including serositis, vasculitis, and subcutaneous nodules. The presence of autoimmune features with the production of rheumatoid factor (RF) and cyclic citrullinated peptide antibodies (ACPA) is a hallmark of RA. Susceptibility of RA is associated with certain subtypes of HLA-DR4 (DR1 and DR14) [1].

The spondyloarthropathies (SpA) comprise a group of chronic inflammatory arthropathies affecting joints, entheses, and certain extra-articular sites. Typical sites of joint inflammation are the spine, the sacroiliac joints, and peripheral, mostly large joints. SpA are characterized by absence of serum autoantibodies, and associated with HLA B27 [2, 3]. Disorders encountered among the SpA are reactive arthritis (ReA), psoriatic arthritis (PsA), arthritis associated with inflammatory bowel disease (IBD), juvenile spondyloarthritis and ankylosing spondylitis (AS), the prototype of SpA. The prevalence of SpA varies in different ethnic populations between 0.5 and 4 % and is related to the frequency of HLA B27 in the population. The male:female ratio is 3:1 and the peak incidence is around age 26 years.

RA and SpA share the characteristics of chronicity, joint inflammation, and damage in conjunction with symptomatic pain, stiffness, and reduced functionality. However, the site of joint inflammation differs between these disorders as well as the extent of systemic features [4]. The majority of patients with RA have autoantibodies like RF and ACPA, whereas SpA do not. In both RA and SpA few internal organs are involved which is one reason for the prevailing good fetal outcomes in these diseases (Table 7.1). Because of the relative age of onset of disease in these disorders, women with RA are more likely to have several pregnancies before disease onset whereas women with AS are more likely to become pregnant during the course of their disease because of their younger age at disease onset. Therapy for the symptoms of joint inflammation overlaps in RA and SpA whereas immunosuppressive therapy directed at the underlying disease pathogenesis of these two disorders differs.

Table 7.1 Characteristics of rheumatoid arthritis and spondyloarthropathies
Full size table

In this chapter, issues of reproduction are presented for patients with RA and SpA. The effect of pregnancy on the activity of the disease during and after pregnancy has been studied in detail only for RA and AS; therefore, only data for these two diseases are presented.

Fertility in RA, SpA, and Other Inflammatory Arthritides

Rheumatic diseases can impair fertility (the actual production of children) and, more often, fecundity (time to achieve pregnancy) in several ways as indicated in Fig. 7.1. Conflicting results on fertility relate most often to study design, number and type of patients included, and presence of a control group. A Norwegian population-based study compared fertility in patients with inflammatory rheumatic diseases comprising 338 patients with RA, 75 with juvenile idiopathic arthritis (JIA), and 218 patients with other chronic arthritides (OCA) including PsA, AS, and unspecified arthritis (UA) with age-matched women without rheumatic disease. Significantly more patients than controls (32.6 versus 26.4 %) were nulliparous at entry in the study [5]. When patients childless at the time of diagnosis were compared with childless references of corresponding age at the time of diagnosis, a higher proportion of patients were childless at the end of the fertile period compared with references [6]. The study was not able to examine reasons for childlessness. The probability of continuing to a second pregnancy among women with one child was lower in arthritic patients than in references (Fig. 7.2). Among patients with a first birth at 30 years of age only 40 % continued to a second pregnancy versus 60 % among references [6].

Fig. 7.1
figure 1

Factors contributing to reduced family size in women with rheumatoid arthritis and spondyloarthropathy

Full size image
Fig. 7.2
figure 2

Mean number of children in all patients with chronic inflammatory arthritides (CIA) by age at diagnosis. Red line = normal fertility rate. Lowess fit = regression line. Scatterplot = mean number of children, size weighted for the number of patients diagnosed per year. Adapted from Wallenius et al., Rheumatology (Oxford) 2011;50:1162–67

Full size image

In the Norwegian study [5] the mean number of children in patients was associated with age at the time of diagnosis (Fig. 7.3) as also observed by Katz [7]. Women diagnosed after 30 years of age had a mean number of children comparable to the reference group, most probably because many women diagnosed in the fourth decade of life have completed their family before disease onset [5]. The study did not show any differences in relative fertility rates in RA and OCA patients compared with references before disease onset. In agreement with several previous studies [7–9] RA and OCA women had significantly lower relative fertility rates after diagnosis (Table 7.2) which may be due to the more severe disease of the study population who all were on treatment with disease-modifying antirheumatic drug (DMARDs) [5]. In addition, women with active rheumatic disease may voluntarily limit their family size due to concerns in regard to problems encountered by the disease and in fear of prognosis [8]. Another contributing explanation for lower fertility rates in rheumatic patients compared with references may be a high occurrence of cesarean section (CS) in deliveries of patients [10–14]. A population-based study from the Medical Birth Registry of Norway (MBRN) has reported that women with CS at first delivery will have fewer children than women who start with a vaginal birth [15].

Fig. 7.3
figure 3

Probability of a second child by the mother’s age at first delivery. Total number of fertile patients (RA, OCA, and JIA) versus references. RA rheumatoid arthritis, OCA other chronic arthritides, JIA juvenile idiopathic arthritis. Adapted from Wallenius et al., Scand J Rheum 2012;41:202–7

Full size image
Table 7.2 Relative fertility rates (RFR) in patients from the NOR-DMARD registry versus references, before and after diagnosis
Full size table

The Norwegian study [5] observed an increased inter pregnancy (IP) interval for RA and OCA patients diagnosed between first and second birth, which indicates that women diagnosed after a first birth may postpone a second pregnancy until the disease is controlled (Table 7.3). No difference in IP-intervals was observed for women with births after diagnosis. This contrasts a previous study where an increased interval after diagnosis was reported [9], possibly indicating that improved treatment during the last decade results in better disease control and opportunity to continue to a second pregnancy. Patients with RA and OCA continuing to a second pregnancy were statistically significantly older than references at the time of the first delivery, because they needed to complete their families within a shorter time period given fewer remaining years of potential fertility. This observation is in accordance with a Norwegian population-based cohort study of patients with RA, JIA, and AS [9], but in contrast to observations in a Canadian case–control study of women with recent onset RA [16]. Different study populations and study designs may explain the different results. The Canadian study examined women with new onset RA (within 3 years) and had a small sample size.

Table 7.3 Inter pregnancy intervala in relation to time of diagnosis, crude* and adjusted for maternal age at first delivery**
Full size table

Contraception in RA and SpA

Unplanned pregnancies in RA and SpA carry risks for mother and child; therefore, family planning should be discussed with every patient of fertile age. RA and SpA patients have no restrictions in regard to method of birth control (see also Chap. 11). A population-based study showed equal frequency of use of oral contraceptives in women with RA and AS compared to age-matched healthy women [17]. Estrogens and gestagens have no particular influence on the signs and symptoms of AS, and probably not on other SpA either. By contrast, use of combination oral contraceptives (COC) containing estrogen and gestagen mitigates the severity of RA, though COC do not prevent the onset of RA [18]. A recent study showed that use of COC shortly before symptom onset and during RA or inflammatory polyarthritis had a beneficial effect on functional outcome [19]. As in healthy women, use of COC in RA and SpA women requires absence of antiphospholipid (aPL) antibodies, absence of thrombophilia or a recent thrombosis, and no use of nicotine. Advice on contraception must be given to all patients who start with possible teratogenic drugs, which for RA and SpA mainly is treatment with methotrexate (MTX) either as monotherapy or as combination therapy (see Chap. 11). Safe birth control has also been recommended for leflunomide though controlled, prospective studies in humans have shown neither an increase in the rate of congenital malformations nor a specific pattern of anomalies after first trimester exposure [20, 21]. Information on necessary contraception must be repeated during follow-up of patients to ensure adherence to birth control [22].

Rheumatoid Arthritis

Pregnancy Outcome in RA

Previous studies have not shown any increased risk of spontaneous abortions among RA patients [12, 23–26]. However, only one of these studies is from after year 2000. New and still unpublished population-based data in the period 2000–2009 (Wallenius, Abstract EULAR 13–2595) have shown an increased relative risk of spontaneous abortion among patients with RA both in early pregnancy ≤ week 12 (OR 1.36) and from gestational week 13–22 (OR 1.42).

Women with rheumatic diseases in general have been found to have higher risks for adverse outcomes like preeclampsia, preterm delivery, cesarean section (CS), and small for gestational age (SGA) infants [10–13, 27, 28]. An increased rate of preeclampsia in RA varying between 1.28 [13], 2.2 [11], and 2.9 [29] was found in population-based studies, but patient populations were different, and neither pharmacotherapy or disease activity or comorbidities were known.

A population-based study with a specific focus on first birth examined possible associations between chronic inflammatory arthritides (CIA) and pregnancy outcomes with separate analyses of first and subsequent births before and after diagnosis [14]. Linkage of data from the Norwegian disease-modifying antirheumatic drug (NOR-DMARD) Registry and the MBRN enabled comparison of pregnancy outcomes in CIA and non-CIA-women. Outcomes of first and subsequent births before and after diagnosis were analyzed separately. In the study, 128 first births in previous nulliparous patients and 151 subsequent births were analyzed and compared with reference deliveries from the general population. Also 286 first and 262 subsequent births before the women were diagnosed with CIA were analyzed. Excess risks were related to first birth in women diagnosed with CIA. No increased risk for preeclampsia was detected. A twofold higher risk of vaginal bleeding occurred in patients compared with references in pregnancies related to first birth. Few reports exist on a possible relationship between use of nonsteroidal anti-inflammatory drugs (NSAIDs) and bleeding in pregnancy. Use of NSAIDs is easily underestimated because patients may report only prescribed drugs and not over-the-counter purchases.

Several population-based studies from different geographic areas have found a 1.5–2.0 increased risk for CS in patients with RA [10–12, 29]. The Norwegian study, including patients with RA and other inflammatory arthritides, has examined both acute and elective CS separately [14]. In the study based on the NOR-DMARD registry, acute CS was not observed more often in patients than references, but the rate of elective CS was doubled among patients both in first and subsequent pregnancies after a diagnosis of CIA. Placental dysfunction, cephalopelvic disproportion, and combined causes constituted each one third of the indications. Furthermore, obstetricians may primarily choose CS in women with rheumatic disease because they anticipate problems during delivery.

Prospective studies including RA women with well-defined disease activity found that pregnancy outcome was similar to healthy women in patients entering pregnancy with low disease activity and limited drug therapy [30]. This was also supported by prospective studies including small numbers of RA pregnancies studied in parallel with healthy pregnant women [31, 32].

The Newborn

Several studies have reported a 1.2–1.8 increased risks of preterm delivery in patients with inflammatory arthritic disease and RA, without relating this to birth order [12, 27, 28]. In the NOR-DMARD study the risks of preterm delivery and SGA were higher for first born children in mothers with CIA compared with references [14]. Multiple regression analysis with covariates for maternal age at delivery, gestational age, and gender of child showed small, but statistically significantly lower mean birth weight in first born children of women with rheumatic disease compared to references. The difference was no longer statistically significant in later pregnancies. Birth weight is related to not only placenta function [33, 34] but also other factors, i.e. genetics and drug use. Two prospective studies comprising 285 RA pregnancies found birth weight within normal range, but lower than in healthy women, and even lower in infants of RA mothers with high disease activity [30, 35]. The Dutch prospective study found prematurity associated with the use of prednisone. In RA pregnancies with well-controlled disease activity and with no or minimal use of prednisone, no increase in prematurity or SGA infants was found [30].

Perinatal morbidity and mortality has been found slightly increased in arthritic study populations in which the majority of patients had RA [14, 28]. Three of four children reported in the Norwegian NOR-DMARD study with perinatal deaths were first born children of mothers diagnosed with inflammatory arthritis including RA. The perinatal mortality rate was three times higher in children of patients than in references [14]. In a retrospective study of women with RA, excess of perinatal deaths were reported in infants of mothers who subsequently developed RA after delivery [25].

Few studies have examined the risk of major congenital malformations in offspring of women with RA [14, 27–29]. In a population-based study based on data from the MBRN a higher proportion of congenital malformations was reported among children of patients with arthritis (RA, AS, JIA, and UA combined) compared with references in the period 1967–1998 [28]. No other study has reported a higher risk of birth defects in children born by RA patients or women with CIA.

Effect of Pregnancy on Rheumatoid Arthritis

Since the 1940s it has been known that about 75 % (range 54–86 %) of RA patients experience spontaneous improvement of symptoms and signs of RA during pregnancy [36]. In a prospective study of RA patients’ perception of their general health as measured by the SF36 during the second and third trimester of pregnancy, RA patients experienced an improvement in physical well-being and pain, resulting in scores on the SF36 comparable to healthy controls [37]. Compared with nonpregnant RA patients, pregnant RA patients experienced significantly less pain and had improved physical function. The need for analgesics or NSAIDs decreased throughout pregnancy. The postpartum deterioration of the disease at week 12 and 24 was reflected by significantly worse scores for bodily pain and for impairment of physical function [37].

Two large prospective studies using validated instruments to measure disease activity found reduced or low disease activity in at least 50 % of RA patients in spite of reduced or no drug therapy during pregnancy [38, 39]. Administration of DMARD decreased significantly, and use of NSAIDs was only 4 % compared to 32 % before conception indicating that pregnancy indeed has a beneficial effect on RA [39]. Patients with low disease activity remained stable throughout pregnancy [39]. Patients with high disease activity at conception did not improve during pregnancy, a finding also confirmed by another study [40]. Complete remission indicating absence of arthritis and no need for therapy is limited to a small proportion of pregnant RA patients [38, 39]. In a prospective study of 118 patients, 75 % of patients negative for RF and CCP antibodies improved compared to only 39 % of those positive for RF and CCP antibodies suggesting that antibody negativity is predictive of pregnancy-induced disease amelioration [41]. Another study showed that long-term functional outcome measured by HAQ was worse in ACPA and RF positive patients who had been pregnant compared to autoantibody negative RA patients with pregnancies [42]. Barret et al. found that the disease response to one pregnancy was predictive of the response in a subsequent pregnancy [38].

Maternal Disease Activity Postpartum

Previous studies stated that the majority of RA patients have recurrent disease within 3–4 months after delivery. Prospective studies from the last two decades found a lower relapse rate which partly could be due to altered management of RA patients during and after pregnancy [39, 43]. Barrett et al [43] found a significant increase in the number of affected joints in 62 % of patients within the first 6 months after delivery. At the same time, drug therapy was increased. Deterioration of disease symptoms after delivery was also found in about 50 % of patients in the Dutch prospective study [39]. The simultaneous increase in drug therapy could indicate that the postpartum relapse rate was probably even higher. Another prospective study showed that entering pregnancy with low disease activity was a relevant factor for stable low disease activity during gestation and even for a reduced risk for a postpartum flare [40]. One study reported a correlation between lactation and increased disease activity postpartum in women with RA, who were breastfeeding for the first time [43]. However, earlier studies reported no correlation between a disease flare and breastfeeding [36]. A prospective study evaluating women with a 12-year follow-up found no significant influence of pregnancy on long-term RA outcome, but found a trend for patients with multiple pregnancies to have less radiographic joint damage and a better functional level [44]

Etiology of the Pregnancy-Induced Amelioration of RA

A number of different factors have been considered in the pregnancy-induced amelioration of RA [45]. Research through more than six decades indicates that not a single but a combination of factors are involved in the improvement of RA during pregnancy [46]. The crucial event is the presence of the semiallogeneic fetus in the mother and the resulting hormonal, immunological, and biochemical alterations which support pregnancy. One important factor recently described in association with amelioration of RA during pregnancy is the reversal of abnormalities in the percentage of IgG immunoglobulins lacking the terminal galactose [47]. The percentage of agalactosyl IgG exceeds age-related normal values in patients with RA. In a prospective study of RA pregnancies, high galactosylation levels were associated with low disease activity of RA and vice versa. Interestingly, the increase of galactosylation levels during pregnancy was more pronounced in the RA patients who improved and resembled those of healthy women [47]. Autoantibody levels did not change significantly during pregnancy [40], but the pathogenicity of autoantibodies is modified by changes in glycosylation of IgG. This may have a role in the pregnancy-induced remission of RA, because the glycosylation of IgG determines its ability to bind to complement and to Fc receptors.

HLA-G which is produced by the placenta has anti-inflammatory and immunosuppressive properties and can mitigate inflammation in murine collagen-induced arthritis [48]. Fetal DNA, soluble HLA molecules of fetal origin or peptides derived from these molecules can also contribute to improvement, as could a shift in cytokine production from a TH1 to TH2 profile during pregnancy [45]. The secretion of cytokines and their receptors in peripheral blood lymphocytes and monocytes changes during pregnancy. Proinflammatory cytokines TNF-α, IL-1β, and IL-6 decrease whereas the anti-inflammatory cytokines IL-4 and IL-10 remain stable or increase. Levels of TNF receptors (TNFR), IL-1 receptor antagonist, and IL-6 receptor increase during pregnancy, thereby counteracting effects of TNF-α, IL-1β, and IL-6 [49].

In two studies of mother–child pairs for whom the mother experienced RA improvement during pregnancy compared to those who did not, fetal–maternal disparity in the HLA class II molecules HLA-DR and DQ was observed significantly more often in the former compared to the latter [50, 51]. Regulatory T cells (Treg) increase during pregnancy suppressing activated T effector cells and maternal autoimmune responses [52]. In a prospective study numerical and functional changes of CD4 + CD25high Treg were investigated in RA patients and healthy women during and after pregnancy [53]. Expansion of Treg occurred to a similar degree during pregnancy in patients and healthy women. Numbers of CD4 + CD25high Treg inversely correlated with disease activity of RA in the third trimester and postpartum. In co-culture experiments of Treg and T effector cells, significantly higher amounts of IL-10 and lowered levels of TNFα and IFNγ were found in supernatants of the third trimester compared to postpartum samples. Postpartum TNFα and IFNγ levels were higher in patients with disease flares than in RA patients with inactive disease [53]. The results showed that amelioration of RA in the third trimester was associated with an increased number of Treg that induced a pronounced anti-inflammatory cytokine milieu.

Ankylosing Spondylitis

Fertility in Ankylosing Spondylitis

A retrospective, international study of reproductive performance in 939 female, Caucasian AS patients found on average 2.4 children per woman, not different from the number of offspring in the healthy population [54]. One population study of women aged 40–42 found normal fertility but diminished fecundity in women with AS [17].

The mainstay of therapy in AS are NSAIDs. Prostaglandins are involved in ovulation and implantation. Studies in animals and humans have shown that NSAIDs inhibiting cyclooxygenase (COX) 1 and 2 can inhibit the rupture of the luteinized follicle, and thereby cause transient infertility [55]. One study showed that NSAIDs increase the risk of the luteinized unruptured follicle (LUF) syndrome, particularly in patients with inactive disease [55]. Selective COX-2 inhibitors seem to be more potent inductors of LUF syndrome than nonselective COX-inhibitors [55]. Patients who take NSAIDs especially selective COX-2 inhibitors continuously, in high doses or take agents with a long half-life are at particular risk for the LUF syndrome. Continuous, periovulatory exposure to NSAIDs should be avoided when planning a pregnancy in patients with rheumatic diseases.

Pregnancy and Ankylosing Spondylitis

Pregnancy Outcome in SpA

The retrospective and prospective studies including AS patients with mild, moderate, and severe disease activity found no increase in adverse pregnancy outcomes including miscarriage, premature labor, or preeclampsia compared to healthy women [31, 32, 54, 56–58]. In all studies of pregnancy outcome in AS, at least 90 % of the patients delivered at term [54, 57, 59]. Compared to healthy women, elective CS is more frequently performed in patients with AS. In the Norwegian population-based study, women with AS had the highest occurrence of elective CS (25 %) among the different diagnostic groups [14], a finding also confirmed by a previous study [54]. Inflammation or ankylosis of the sacroiliac joints is not a mechanical hindrance for the progression of parturition nor is spinal disease. Anesthesiologists often fear that neuraxial anesthesia is difficult to establish in patients with AS because of ankylosis in the lumbar spine. However, ankylosis of the lumbar spine is rare in female AS patients and occurs only after 3–4 decades of disease duration [58, 60]. Iatrogenic reasons or the wish of the patient for surgical delivery may be of more impact for elective CS than the severity of AS.

Retrospective and some prospective studies have investigated the effect of pregnancy on the course and severity of AS [31, 32, 56, 61]. In contrast to RA, pregnancy has no profound effect on the symptoms of AS. In the prospective studies clinical and laboratory measurements showed active disease during the first and early second trimester, sometimes accompanied by a flare around week 20 of pregnancy. Pain and stiffness did increase, nocturnal pain intensified, and occasionally arthritis in the knee or the ankle joints occurred. In some patients, pain at the attachment sites of ligaments or tendons and a feeling of tightness in the chest created additional problems. Anterior uveitis could become active during this period. Need for NSAIDs and analgesics was present in about 70 % of the patients [31, 57, 62]. Disease activity decreased again in the third trimester in most AS patients, but complete subsidence of symptoms was not observed. In a prospective study assessing general health perception by the SF 36 throughout pregnancy in patients with AS and RA compared to healthy pregnant women, most patients with AS entered pregnancy with active disease and suffered the greatest impairment of health-related quality of life with persistent active disease throughout pregnancy [37]. In contrast to pregnant RA patients functional scores and bodily pain scores were not altered by pregnancy in AS. Compared with healthy women, patients with AS had worse scores for bodily pain, physical functioning, and general health perception, both during pregnancy and in the postpartum period.

Some 20 % of AS patients improve from spinal and extraspinal symptoms during pregnancy [31, 61]. These patients most often have a history of arthritis in joints other than the spine, or psoriasis or IBD associated with their AS. In patients studied during multiple pregnancies intensity of disease symptoms varied from one pregnancy to the other. No uniform pattern with regard to improvement or aggravation emerged. However, complete remission of symptoms never occurred in any patient with pure spinal disease [31, 56].

Maternal Disease Activity in AS Post Partum

Fifty to eighty percent of the AS patients experience aggravation of symptoms 4–12 weeks after delivery independent from the period of breastfeeding [31]. One study found the flare correlated to active disease at conception, but not related to disease activity during pregnancy [54]. Episodes of acute peripheral arthritis and anterior uveitis occurred 1.5–3 times more often postpartum than during pregnancy [54, 61]. Clinical signs were accompanied by elevations of acute phase reactants [62]. Disease activity returned as a rule to the pre-pregnancy pattern during the year following delivery [31].

Management of Pregnancy in Patients with RA and SpA

The patient with active RA or SpA who contemplates pregnancy or is already pregnant requires therapy compatible with pregnancy (reviewed in detail in Chap. 14). This obviously limits the choice of DMARD considerably and requires more flexibility as to the timing of certain drugs prior to pregnancy. For example, one could decide for a combination of MTX and a TNFα inhibitor as a first line therapy in a patient with early RA or SpA who plans a pregnancy within the near future. The aim would be to achieve disease control within reasonable short time, then to discontinue MTX and to continue with a therapy compatible with pregnancy. The continuation of therapy with DMARD in RA patients who plan to become pregnant may be beneficial. In a study of RA pregnancies, 75 % of patients with inactive disease were on DMARD before conception of whom 38 % continued treatment throughout pregnancy. By contrast, among patients with persistent active disease only 33 % received immunosuppressive therapy before conception, and only 17 % continued therapy during pregnancy [40]. This supports the notion that continuation of compatible DMARD therapy during pregnancy, at least in RA, helps to suppress flares of disease.

There are two reasons for drug therapy during pregnancy and lactation: one is to prevent a flare with possible harmful effects on mother and fetus, the other is treatment of a flare occurring during or after pregnancy. The goal of drug therapy in pregnancy and lactation is to keep the patient at low disease activity or in remission and to prevent progression of inflammation with structural damage. Regular monitoring of disease activity during and after pregnancy is a prerequisite to decide whether drug treatment is necessary. The need for drug therapy in RA and AS differs significantly. The beneficial effect of pregnancy on the symptoms of RA reduces the need for drug therapy and is present in only 15–25 % of pregnant RA patients. The persistent active disease in AS most often requires symptomatic treatment [32]. Patients with SpA, especially with AS, frequently have active arthritis at some stage of pregnancy. Start of DMARD for a disease flare during pregnancy is not ideal since most of the DMARDs compatible with pregnancy all have a delay of 2–3 months in onset of effect. Corticosteroids with the advantage of a very fast action are best suited for treatment of acute arthritis in pregnancy, although they slightly increase the risk for cleft palate formation when used during the first trimester. Intraarticular injections of corticosteroids into one or a limited number of actively inflamed joints can be given to both RA and SpA patients, and systemic corticosteroids can be added for continuous suppression of inflammation if necessary. This goal can as a rule be achieved by 10–20 mg prednisone daily in patients with RA. To prevent corticosteroid-induced osteopenia supplementation with calcium and vitamin D3 is important in patients on prolonged systemic therapy. Frequent controls to detect an increase in blood pressure or the development of gestational diabetes in the mother are necessary. When administrating systemic corticosteroids, the lowest effective dose of corticosteroids should be given for a limited time period to avoid intrauterine growth restriction and premature delivery.

Patients with AS generally do not respond as promptly to prednisone. They often benefit more from NSAIDs although these medications need to be discontinued by week 28–30 of pregnancy to circumvent the risk of premature closure of the ductus arteriosus.

Patients with RA and SpA need more intensive prenatal care than healthy pregnant women. Type and frequency of appointments during pregnancy should be agreed upon with the patient and contacts established between doctors involved in follow-up. Good communication among specialist increases the chance for a successful pregnancy and delivery. The schedule for follow-up visits depends on disease activity, and maternal and fetal risk factors present at conception. Monthly evaluations throughout pregnancy may be sufficient in patients in remission, but active disease demands more frequent visits. Lifestyle adjustments before pregnancy include stopping smoking, alcohol ingestion, and recreational drugs during pregnancy and maintaining a healthy diet containing multiple vitamins [63]. Importance of folic acid supplementation during the early stages of pregnancy needs to be emphasized [64]. The usual dose for healthy pregnant women as well as patients with RA and SpA is 0.4 mg/day. Patients who have received therapy with MTX and sulfasalazine previously should continue 1 mg daily before conception and at least throughout the first trimester.

An aggravation of disease activity occurs in a majority of patients with RA and SpA within the first 6 months after delivery [46]. Frequent visits postpartum are therefore necessary in order to detect a flare and to start effective therapy right away. The time of disease aggravation coincides with the period of lactation. Effective disease control at this stage is of great importance in order to inhibit disease progression and to enable the mother to care for her baby. Unfortunately the available literature on therapy with antirheumatic drugs during lactation is sparse, and many patients will feel that they need to choose between disease control and lactation. Many mothers would rather deprive themselves of therapy rather than deprive the child of milk. Warnings given against breastfeeding when on therapy are often not based on evidence for adverse effects in the child, but on lack of knowledge [65].

Special Problems in RA and AS

Involvement of the cervical spine with cervical instability in RA is rare, but when atlanto-axial subluxation is present, cervical cord damage is a risk [66]. It is important to know of this deformity before considering general anesthesia for a surgical delivery. Cervical radiographs or MR examination before a planned pregnancy is recommended to exclude atlanto-axial subluxation.

In pregnant women with AS sometimes the anesthesiologist will not give epidural anesthesia because he/she fears problems with positioning the epidural catheter in case of ankylosis of the lumbar spine. These concerns can be overcome when doing an X-ray of the lumbar spine before a planned pregnancy. Sometimes calcification of the posterior, longitudinal ligament inhibits spreading of the anesthetic solution by the epidural route, then intrathecal anesthesia may be necessary [67].

Pre-conceptional Counseling for Patients with RA and SpA

At a pre-conceptional visit, the physician must identify possible risks of a future pregnancy in regard to the stage and activity of the disease and to organize adequate monitoring during pregnancy and postpartum. An early stage of disease implies that the pattern of disease severity is not yet apparent and frequent changes of drug treatment may be necessary. A patient with active rheumatic disease should postpone pregnancy until remission or stable disease is achieved and has persisted for at least 6 months. In patients planning a pregnancy a clinical and laboratory work-up should be performed for risk assessment. RA patients positive for rheumatoid factor or ACPA and those with currently active disease have significantly less chance for spontaneous improvement during pregnancy and are at higher risk for adverse outcomes. RA patients without RF or ACPA and those with low disease activity or in remission are the most likely to remain inactive during pregnancy. The rheumatologist should test a woman with RA for the presence of aPL and antibodies against SS-A (anti-Ro) and SS-B (anti-La) since these autoantibodies can occur in 30 % RA patients and these patients may require special monitoring.

An adjustment of therapy with discontinuation of fetotoxic drugs and switch to pregnancy compatible drugs is necessary in patients desiring to become pregnant or in those already pregnant. Therapy must be tailored for the individual patient according to disease activity and secure effective maternal disease control as well as safety of the fetus (see Chap. 14). It is important that the treating physician warns against discontinuation of all therapy when an attempt of conception is made. Withdrawal of drug therapy may result in a disease flare. Other physicians, especially the obstetrician involved in patient treatment, should be informed about necessary antirheumatic therapy in order to avoid contradictory advice. At the end of pregnancy, a discussion of breastfeeding and use of drugs during lactation is recommended.

Before pregnancy, access to familial and social support should be evaluated. Therefore it is advantageous to include the patient’s partner or a family member into pre-conceptual counseling. In order to prepare for the task of child caring after delivery, consulting an occupational therapist is advisable for the patient with impaired function. A meeting with an interdisciplinary team including a physiotherapist, an occupational therapist and a social worker can help to prepare the rheumatic mother for parenting.

Patients are often concerned in regard to heredity of rheumatic diseases. In general genetic factors and environmental factors contribute to the development of rheumatic diseases and only few of them have been recognized or been studied in detail. The gene linked with rheumatoid arthritis is HLA-DR4. The disease rate in first-degree family members of RA patients is only 0.8 % compared to 0.5 % in the general population indicating that genes only modestly increase the risk for RA and that the environment is likely to play a stronger role. The risk to inherit the disease in a child with one RA parent is only 1–3 %. A recent study has indicated that smoking increases the risk for RA in DR4 positive individuals [68]. This implies that children of a RA parent should be warned against smoking.

More than 90 % of Caucasian AS patients have the HLA-B27 gene, compared to the approximately 7 % of the general population who carry this gene. The chance of a child inheriting AS is estimated at about one in six if one parent has the gene HLA-B27, and about one in ten if not. On the other hand, first-degree family members (parents, siblings, and children) of AS patients with the HLA-B27 gene have a 20 % chance of developing the disease. This is probably because they are exposed to the same environmental factors as the patient and have inherited certain other important genes in addition to the HLA-B27 gene like ARTS 1 and IL-23R, two newly identified non-HLA B27 susceptibility genes for AS [69].

Patients with RA and SpA can be assured that there is no increased risk for giving birth to a child with congenital malformations except for patients who have been exposed to teratogenic drugs during the first trimester.

Conclusion

The disease process, therapy, and disease burden contribute to a reduction in the number of children in women with RA and SpA after disease onset. Pregnancy has no major effect on the symptoms of AS, but induces spontaneous improvement in most patients with RA. An aggravation of disease activity within 6 months after delivery is common in both RA and SpA. In RA active disease and therapy with corticosteroids during pregnancy increases the risk for preterm birth and reduced birth weight of neonates. The rate of CS is significantly higher in all women with rheumatic disease; however, acute CS seems to be no more frequent than in healthy pregnant women. The highest rate of elective CS has been found in women with AS, probably related to anticipation of problems during delivery by obstetricians and anesthesiologists.

Management during pregnancy is based on risk assessment for mother and child and depends on disease activity which must be carefully monitored during pregnancy. In order to reduce unplanned and ill-timed pregnancies, pre-conceptional counseling is suggested for all patients of fertile age and includes discussion of contraception and adjustment of therapy. Most women with RA and SpA can have successful pregnancies provided that pregnancy is planned and occurs in a stage of remission or low disease activity and under therapy with drugs compatible with pregnancy.