Maternal Considerations

Pregnancy in a patient with SLE represents an added risk to an already complex clinical situation. Large-scale epidemiologic studies confirm that despite improvements in overall pregnancy outcome over recent years, maternal morbidity is still high for patients with SLE: hypertensive complications, renal disease, preterm delivery, cesarean section, postpartum hemorrhage and thrombosis, fetal growth restriction, and neonatal death are all more common in SLE pregnancies. One national study found maternal mortality to be 20-fold higher among women with SLE; in addition, roughly one third of patients with SLE have cesarean sections, one third deliver prematurely, and one fourth develop pre-eclampsia. Fortunately, identification of particular risk factors can help predict likelihood of complications and aid in risk counseling. Important risk factors include presence of lupus disease activity, renal involvement, aPLs and anti-SSA/Ro and SSB/La antibodies. Pre-pregnancy evaluation involves assessment of level of disease activity, presence of disease damage, review of current medications, and laboratory tests, especially autoantibody status. The algorithm in Figure 31-1 illustrates a general approach to evaluation of rheumatic disease patients before pregnancy.

Pre-pregnancy evaluation of the rheumatic disease patient.

Medication Assessment

Review of medications before conception allows adjustment with careful follow-up to ensure maintenance of disease control. Detailed recommendations appear below. In brief, however, prednisone, methylprednisolone, hydroxychloroquine, azathioprine, cyclosporine, and intravenous immunoglobulin may be continued through pregnancy; in most circumstances, other immunosuppressive medications should be discontinued or replaced. Continuation of hydroxychloroquine is encouraged, because recent data support fetal safety and a lowered risk of maternal disease flare.

Disease Activity

The importance of quiescent disease at the time of conception cannot be overemphasized. Risk of flare during pregnancy is significantly higher if lupus was active before conception, and pregnancies with high levels of disease activity (but not mild disease) result in increased risk of premature birth and a decrease in live births. Cortes-Hernandez and colleagues found that risk of flare in pregnancy was associated with discontinuation of maintenance therapy before pregnancy and a history of 3 or more severe flares before pregnancy. In general, it is prudent to have quiescent disease for at least 6 months before attempting to conceive.

Evidence of any serious organ damage due to SLE—even in the absence of current disease activity—may represent a contraindication to pregnancy. This includes not only moderate to severe renal insufficiency but also seriously impaired cardiac or pulmonary function, uncontrolled hypertension, or presence of pulmonary hypertension.

Risk of Flare During Pregnancy

Relative risk of flare during SLE pregnancy remains controversial, with estimates ranging from 13% to 60%. Discordant findings likely reflect differences between patient populations and in the definition of lupus flare. For example, in Lockshin’s series of SLE pregnancies, thrombocytopenia, proteinuria, and hypocomplementemia were the most common abnormalities. If interpreted as SLE-related symptoms, the flare rate was 25%; if interpreted as due to pregnancy, pre-eclampsia, or aPLs, rate of flare dropped to 13%. More recently, Ruiz-Irastorza and associates demonstrated higher flare rates in pregnant lupus patients compared with both nonpregnant controls and with the patients themselves after pregnancy, but Georgiou and coworkers found no significant increase in flare during pregnancy, and an overall flare rate of only 13% in their cohort of patients. Existing instruments to measure disease activity have recently been modified and validated to differentiate pregnancy-related physiologic changes from lupus activity, and will allow greater comparability between future studies. Pregnancy-specific disease activity indices include the SLE-Pregnancy Disease Activity Index and the Lupus Activity Index Pregnancy Scale.

Despite the varied estimates of precise risk of flare during lupus pregnancy, it is reasonable to counsel patients regarding a small to moderate increased risk of disease exacerbation, even with long-term quiescent disease, which may occur during any trimester or the postpartum period. Prophylactic steroid therapy is not recommended to prevent flare.

Impact of Lupus Nephritis

Renal insufficiency is an important risk factor for poor fetal outcome and for serious renal deterioration during pregnancy. Precise renal histology is less important than the presence of clinical renal dysfunction. Although exacerbation of renal disease may occur in up to 40% of cases, severe or permanent renal deterioration develops in less than 10%. The physiologic stress of pregnancy, even in patients with quiescent nephritis, may accelerate deterioration, especially in those with initial serum creatinine greater than 1.6 mg/dL. Onset of worsening renal function usually occurs in the latter half of pregnancy. In addition to the risk of moderate renal insufficiency, baseline proteinuria presents an independent risk: Moroni and associates demonstrated that proteinuria greater than 500 mg/24 hours is an independent predictor of poor fetal outcome. Fetal loss occurred in 57% of patients with proteinuria versus 10% of those without. Recent active nephritis also increased risk of renal deterioration during pregnancy: risk of renal flare was 5% in those with quiescent nephritis, and 39% in those with recently active nephritis.

Patients with SLE with renal transplants usually have successful pregnancies. McGrory and colleagues compared pregnancy outcomes in renal transplant recipients with (60 pregnancies) and without (374 pregnancies) SLE. Significant hypertension and cesarean sections were less common in patients with SLE, and other maternal conditions and pregnancy outcomes were comparable in both groups.

Differentiation of Systemic Lupus Erythematosus Flare from Pregnancy-Related Complications

In the SLE population, the differential diagnosis for new-onset proteinuria in the second half of pregnancy includes lupus nephritis, pre-eclampsia, or both. Two thirds of patients with SLE entering pregnancy with pre-existing renal disease develop pre-eclampsia, compared with 14% of patients with lupus without kidney disease. It is often not possible to differentiate between SLE nephritis and pre-eclampsia during pregnancy, and frequently, therapy is directed toward both. Clinical signs of active SLE such as inflammatory arthritis or rash, fever, lymphadenopathy, hematuria, leukopenia, erythrocyte casts, and rising anti-DNA antibody favor the diagnosis of lupus glomerulonephritis. Thrombocytopenia, hypertension, and hyperuricemia occur in both SLE and pre-eclampsia ( Table 31-1 ). ESR, increased in normal pregnancy, becomes an unreliable measure of lupus activity during pregnancy. Normal serum complement favors a diagnosis of pre-eclampsia. Measures of alternative pathway complement activation (Ba and Bb), are reported to be abnormal in active lupus nephritis and normal in lupus pregnancies with proteinuria not due to active SLE. Seizures may occur due to eclampsia or to neuropsychiatric lupus.

Thrombocytopenia during SLE pregnancy varies in presentation, and may be due to flare, aPLs, HELLP syndrome, or even normal pregnancy itself. Active SLE-related moderate thrombocytopenia (50–120 × 10 ⁹ /L) may be difficult to distinguish from benign (mild) thrombocytopenia of late pregnancy, unrelated to SLE. Although rare, severe lupus-related thrombocytopenia may occur, with counts less than 10 × 10 ⁹ /L.

Pregnancy Outcome

Management and Therapy

Treatment of the pregnant patient with SLE does not usually differ from that of the nonpregnant patient. Careful follow-up and monitoring with both the rheumatologist and obstetrician are essential ( Table 31-2 ). Hydroxychloroquine should be continued to prevent disease flare. Clinical flares of skin, joints, blood, kidneys, and nervous system require therapy with (unfluorinated) corticosteroids. Use of immunosuppressive therapies other than corticosteroid is generally limited to azathioprine or cyclosporine (see details later). Corticosteroid should be prescribed for thrombocytopenia if the platelet count drops significantly; life-threatening thrombocytopenia (< 30 × 10 ⁹ /L) or imminent delivery requires platelet transfusion, intravenous IgG, or both.

The risk of pre-eclampsia is significantly increased in lupus pregnancy. Although low-dose aspirin is used to prevent pre-eclampsia in high-risk women with pre-existing hypertension, renal insufficiency, or history of previous pre-eclampsia, there is as yet no systematic study of the use of aspirin to prevent pre-eclampsia in patients with SLE (with or without aPL), although it is frequently prescribed.

There is general agreement on several key observations regarding risk and outcome of lupus pregnancy: first, inactive disease at the time of conception (and for the 6 months preceding pregnancy) is associated with a lower likelihood of disease exacerbation and better fetal outcome. Second, although precise risk of exacerbation is difficult to quantitate, SLE may flare anytime during pregnancy, including the postpartum period. Third, renal insufficiency and hypertension may worsen during pregnancy, and active renal disease is associated with both unfavorable maternal and fetal outcome. Finally, fetal outcome is related to presence of aPL, renal involvement, and high levels of lupus disease activity.

Maternal Considerations

Obstetric manifestations of antiphospholipid syndrome (APS) are not restricted to fetal loss. Current APS criteria include pregnancy loss or early delivery due to pre-eclampsia, IUGR, or fetal distress in the presence of persistent moderate to high titer anticardiolipin antibody (aCL), anti-β2-Glycoprotein I antibody (anti-β2GPI), or lupus anticoagulant. Fetal loss (>10 weeks of gestation) is more strongly associated with aPL than are earlier pregnancy losses. Midtrimester loss is considered characteristic; however, up to one half of losses may be in the first trimester. Since pre-embryonic and embryonic losses (< 10 weeks of gestation) are common in the general population, the diagnosis of APS should be made only with three or more consecutive losses in the absence of other identifiable etiologies. The two greatest risk factors for fetal loss are high titer IgG aCL and a history of previous fetal loss: these patients have up to 80% risk of current pregnancy loss if untreated. Although aCL and LA are closely linked to pregnancy loss, the predictive value of anti-β2GPI for pregnancy loss is less well studied. Concordance between types of aPLs are incomplete, so all should be tested if APS is suspected. Antibodies directed against other phospholipids, such as antiphosphatidylserine or antiphosphatidylethanolamine, do not generally identify additional patients and are not usually tested.

Even in the presence of positive aPL antibody, exclusion of confounding conditions is important in patients with pregnancy morbidity, because more than one factor may be present. Gynecologic conditions may include uterine abnormalities, hormonal imbalance (e.g., luteal phase defect), maternal and paternal karyotype abnormalities, or fetal genetic abnormalities.

Other Maternal Complications

In addition to pregnancy loss, associated maternal obstetric complications include pre-eclampsia, eclampsia and HELLP syndrome. Rates of gestational hypertension and pre-eclampsia range from 30% to 50%. Arterial or venous thrombosis and aPL manifestations such as severe thrombocytopenia may occur. A report of cerebral infarction postpartum with discontinuation of aspirin therapy emphasizes the importance of postpartum thromboprophylaxis. Finally, fifteen cases of catastrophic APS developing during pregnancy have been reported, many associated with concomitant HELLP syndrome.

Pregnancy Outcome

Neonatal complications include prematurity and IUGR ; in one large series, the rate of preterm delivery was 43% and that of IUGR was 31%. Preterm delivery is the strongest risk factor for adverse neonatal outcome and reflects the greatest concern for patients with APS.

IgG aCL can traverse the placenta. Although most large studies of aPL pregnancies do not find fetal or neonatal thrombosis, rare cases of intrauterine thromboembolic stroke, arterial or venous thrombosis, and neonatal seizures have been reported.

Although at least one large study suggests no long-term developmental abnormalities in offspring, the rate of learning disability was increased in a single small series of 17 children of mothers with APS.

Management and Therapy

Low-dose aspirin plus subcutaneous low-dose heparin is the standard prophylactic therapy during pregnancy for patients who have had prior pregnancy events. Untreated patients with high-titer aPL and a history of previous pregnancy loss have a high chance of subsequent fetal loss; with low-dose aspirin and heparin therapy, the likelihood of a full-term delivery rises to 70% to 80%. A recent meta-analysis of randomized controlled trials of therapy for aPL-positive patients with pregnancy loss found that low-dose aspirin and low-dose heparin significantly reduced risk of pregnancy loss compared with aspirin alone (relative risk [RR], 0.46). Aspirin alone did not show a significant reduction in rate of pregnancy loss (RR, 1.05). Prednisone and aspirin treatment increased risk of prematurity (RR, 4.83) but did not significantly decrease the risk of pregnancy loss (RR, 0.85). Although initial studies used unfractionated heparin, most clinicians prefer low-molecular-weight heparin (LMWH) owing to convenience and lower risk of side effects including osteoporosis; recent small studies suggest at least equal efficacy. LMWH dosing is not uniform, but pregnancy-related pharmacokinetic changes suggest LMWH might best be given twice daily, even for pregnancy prophylaxis therapy. LMWH is generally converted to unfractionated heparin several weeks before the expected due date due to the latter’s shorter half-life.

There are no clear guidelines in the management of patients in a first pregnancy, or those who have no, one, or two embryonic or pre-embryonic losses; however, low-dose aspirin treatment is often given to these patients because of its low-risk profile. Low-dose aspirin and heparin in therapeutic doses with midinterval partial thromboplastin time (aPTT) or anti-factor-Xa level monitoring must be used in patients with APS with a history of vascular thrombosis. IVIG has been used in addition to aspirin and heparin for those in whom standard therapy has failed, although a small placebo-controlled study showed no benefit. Suggested guidelines for therapy of aPL pregnancies are shown in Table 31-3 .

All patients require at least 6 weeks of anticoagulation postpartum, whether with aspirin alone, heparin, or warfarin. The importance of fetal monitoring during pregnancy should also be emphasized. Nonstress tests, biophysical profiles, or fetal umbilical artery Doppler velocimetry starting at 26 to 28 weeks may provide early indications of fetal distress requiring intervention.

During long-term follow-up, long-term low-dose aspirin appears to be protective against development of vascular thrombosis in patients with APS with a history of pregnancy morbidity alone and is generally recommended.

Maternal Considerations

The effect of pregnancy on RA is usually positive: up to 73% of patients experience some degree of clinical remission of symptoms during pregnancy, even in the absence of usual medications. In one series of 308 pregnancies, amelioration in symptoms was detected in about 50% of patients in the first trimester, in an additional 14% in the second trimester, and in a further 6% in the third trimester. Improvement persisted throughout the course of the pregnancy. A prospective study of 140 women with RA during pregnancy demonstrated significant variability in degree of response, however, with about one quarter of patients doing worse, two thirds partially improved and only 16% of patients attaining complete remission.

Regardless of the degree of improvement experienced during gestation, most patients relapse following delivery, many of these within the first 6 weeks, and almost all by 4 months. Although this suggests that pregnancy-related hormones might influence remission, breastfeeding does not appear to change the timing of relapse. Recent work suggests that persistence of active disease during pregnancy may relate to a similarity between maternal and fetal HLA (HLA-DQ system).

The effect of pregnancy on the development of RA is controversial. Pregnancy as well as use of oral contraceptives has been suggested to reduce the risk of disease onset, and an increased risk of disease onset is noted in the postpartum period. A recent study, however, suggests no protective effect of parity but does demonstrate a protective effect of breastfeeding against development of RA.

Pregnancy Outcome

Fertility is not clearly decreased in patients with RA, and no increase in fetal loss has been demonstrated. A slightly increased rate of fetal complications of prematurity and low birth weight has been reported, however, and rates of pre-eclampsia and cesarean section are higher than in the general population. A case of marked fetal growth retardation has been described in severe RA with associated vasculitis.

Management and Therapy

Few of the common medications for treatment of RA are considered safe during pregnancy; fortunately, owing to high likelihood of at least partial remission, many patients discontinue medication entirely or can be managed with low doses of corticosteroid.

As with all rheumatic disease patients, the patient with RA should be counseled before attempting pregnancy. The preconception visit should include discussion of the likelihood (but not certainty) of some degree of remission, recommendations for discontinuation of medications as necessary, and the potential adverse effects of any medications that will be continued. Note that testing for presence of anti-SSA/Ro and anti-SSB/La antibodies should be done, and patients with positive results should be counseled and monitored as described earlier in the section on NLE.

Potential fetal effects of medications are discussed in detail at the end of the chapter. In general, NSAIDs are avoided, plaquinil and sulfasalazine may be continued, and methotrexate must be discontinued at least 3 months before conceiving. Leflunomide must be discontinued and “washed out” with a course of cholestyramine. Tumor necrosis factor (TNF) inhibitors have not been proven safe during pregnancy, but reports of immediate pre-pregnancy or early first-trimester use are reassuring, and if required, the fetal risk of continuing these until the diagnosis of pregnancy appears to be quite low. This is especially important for patients with fertility issues: although pregnancy may bring about remission, it may take many months to conceive for some patients. If even young RA patients experience delay in conceiving, aggressive pursuit of fertility therapy may minimize the amount of time their joints are not optimally protected. Low-dose prednisone (with calcium and vitamin D supplementation) is used as necessary throughout pregnancy.

Although uncommon, cervical spine arthritis with atlantoaxial instability requires careful management of the patient under general anesthesia since manipulation of the unstable spine may produce spinal cord compression. Severe cricoarytenoid involvement may also be a relative contraindication to intubation with general anesthesia. Hip range of motion should be assessed for adequacy before anticipated vaginal delivery. Stress dose steroid should be administered if the patient has been on long-term steroid therapy, and prophylactic antibiotics are recommended for patients with prosthetic joints.

A particular concern for the RA patient is the near-certainty of postpartum flare. For patients with severe erosive disease, it may be wise to recommend avoiding breastfeeding and restarting combination therapy immediately postpartum. Patients with mild disease often elect to breastfeed until they flare, then wean and restart medications.

Maternal Considerations

Early reports of maternal morbidity in scleroderma pregnancy were negative: in a combined analysis of early studies, about one third of patients reported pregnancy-related aggravation of their disease. Ten percent died of pregnancy-related complications, primarily hypertension, renal failure, and cardiovascular complications. Importantly, however, these series included a number of cases described before advances in management.

In contrast, a retrospective case-control study found no difference in the rates of accelerated hypertension or renal failure in parous versus nulliparous scleroderma patients, suggesting that pregnancy may not adversely affect disease course. Incidence of hypertension, pre-eclampsia, or proteinuria was no different in pregnancies after diagnosis as compared with pregnancies before diagnosis of disease. In Steen’s prospective studies of 91 pregnancies in 59 patients with systemic sclerosis, maternal outcome was generally good, with worsening of disease during pregnancy in about 20% of patients. Raynaud’s phenomenon improved during pregnancy, whereas esophageal reflux became worse. There were three cases of renal crisis during pregnancy, all in women with early diffuse disease.

Pregnancy Outcome

Scleroderma patients were initially reported to have both decreased fertility and decreased parity when compared with control populations. Early case-control studies showed twice the rate of spontaneous abortion and three times the rate of infertility compared with controls as well as a significant incidence of IUGR and preterm births that occurred with equal frequency before and after the diagnosis of systemic sclerosis. More recent prospective data, however, show no consistent decrease in fertility or increase in frequency of miscarriage except in the subgroup of patients with long-standing diffuse scleroderma, in which miscarriage rates are as high as 45%. Preterm births occurred in 29% of pregnancies, but neonatal survival was good. Of note, four women in this series had five healthy infants while taking angiotensin-converting enzyme (ACE) inhibitors.

Management and Therapy

Preconception evaluation of disease status and organ function is critical in advising patients with systemic sclerosis about pregnancy. Patients with early diffuse disease are advised to delay pregnancy if possible until disease stabilizes, to reduce the risk of renal crisis during pregnancy. ACE inhibitors should be stopped due to risk of fetal toxicity, and other antihypertensives should be substituted. It is prudent to make this change before pregnancy and to observe for a minimum of several months to be confident that blood pressure is adequately controlled. Once pregnant, blood pressure and renal function should be monitored carefully, and include use of home blood pressure monitoring. For true scleroderma renal crisis during pregnancy, a life-threatening complication, ACE inhibitors are the only effective therapy and must be used to protect maternal health.

Skin thickening may be a concern for intravenous access or cesarean section, but wound healing is generally normal. In patients with severe Raynaud’s phenomenon, warming the patient throughout labor and delivery with use of blankets, socks, and increased room temperature may minimize symptoms.

Women with severe cardiac, renal, or pulmonary disease for whom pregnancy is contraindicated may consider in vitro fertilization and surrogate pregnancy with a gestational carrier.