Definition and Epidemiology

Systemic lupus erythematosus (SLE) is a multisystem inflammatory disease of unknown etiology that is characterized by the production of autoantibodies against nuclear, cytoplasmic, and cell membrane antigens. Although its overall prevalence is about 50 per 100,000 population, it is recognized most commonly in women during their childbearing years, with a female-to-male ratio of 9 : 1. Certain groups have a higher prevalence, including African-Americans (2- to 3-fold risk), Asians, and Native Americans. An estimated 1 in 1200 deliveries occur in women with SLE.

Pathophysiology

The etiology of SLE remains unclear. The principal mechanism is thought to be an immune complex disease involving IgG antibodies to double-stranded DNA and other nuclear proteins. Intracellular autoantigens are released by necrotic and apoptotic cells, leading to aberrant sensitization against these antigens. Impaired clearance of apoptotic cells and prolonged exposure to nuclear autoantigens may be involved. Affected individuals have both hyperactivity of the antibody-producing B cells and defects of the helper and suppressor T cells. Genetic defects of immune regulation and possibly environmental triggers including viruses and bacteria lead to a proliferation of B cells capable of producing autoantibodies that may target more than 30 classes of antigens. A variety of antigen-antibody immune complexes are formed, followed by secondary inflammatory responses. Deposits may occur within the skin, choroid plexus, and other endothelial surfaces, with or without an inflammatory response. In certain locations (e.g., the renal glomerulus), deposition of immune complexes and continued inflammation may lead to irreversible injury. However, SLE is not simply an immune complex disorder because some autoantibodies actively bind to erythrocytes, granulocytes, lymphocytes, and macrophages, leading to their removal from the circulation.

Diagnosis

Clinical manifestations of SLE are diverse, owing to the widespread antigenic targets. Box 40.2 outlines objective criteria for the diagnosis of SLE. Although epidemiologic studies require the presence of four or more of these criteria, the clinical diagnosis may be suspected if fewer features are present without another explanation. Typically, the diagnosis of SLE is made before conception, but in 20% of cases the initial diagnosis is made during pregnancy.

Effect of Pregnancy

SLE disease activity may increase during pregnancy, and some pregnancy-related complications may have long-term effects on women with SLE. The PROMISSE study, an observational prospective outcome study of 385 women with SLE with or without antiphospholipid antibodies (aPL), used formal measurements of disease activity to delineate risk. Renal disease, SLE disease activity, and the presence of aPL are considered significant risk factors for maternal deterioration during pregnancy. Assessments using the SLE Disease Activity Index found that 50% to 65% of women with active disease had deterioration during pregnancy in both retrospective and prospective studies. Such flares occur most commonly in the second and third trimesters and the puerperium, are not more severe than those in nonpregnant patients, and mostly respond to conservative management. The risk for significant disease activity during pregnancy is increased sevenfold if active disease is present in the 6 months before conception, and a preconception history of nephritis predicts adverse maternal outcome.

Women with SLE have a 2- to 4-fold increase in the overall rate of pregnancy complications, with increased rates of hypertension, renal disease, preeclampsia, preterm delivery, cesarean delivery, and maternal and neonatal death. Nevertheless, maternal and fetal outcomes have improved with advances in disease management.

Effect on the Mother

Most women with SLE do not have renal impairment at conception, possibly because renal insufficiency impairs fertility. If lupus nephritis pre-exists, deterioration in renal function may occur during pregnancy. Although mild and reversible in about 8% of pregnant patients with SLE, 3% of women may suffer irreversible progression of renal dysfunction. Long-term glomerular filtration rate may be preserved.

Renal involvement is also associated with maternal hypertension. The presence of hypertension, edema, and proteinuria in both lupus nephritis and preeclampsia makes distinguishing between the two difficult. It is not clear whether preeclampsia is more common in patients with SLE, but a large meta-analysis suggested an association between lupus nephritis and preeclampsia. The distinction is critical because treatments are different (immunosuppressive therapy for lupus nephritis versus delivery for preeclampsia). Increased serum uric acid concentration, proteinuria without active urinary sediment, and liver enzyme abnormalities suggest preeclampsia rather than SLE. The two conditions may also co-exist.

SLE may cause thrombocytopenia. When thrombocytopenia occurs in a pregnant woman, preeclampsia, HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome, and disseminated intravascular coagulation must also be considered. Anemia, a common manifestation of SLE, must be differentiated from nutritional anemia and the physiologic anemia of late pregnancy.

Ligamentous relaxation often occurs during late pregnancy and may worsen the pain of lupus arthritis. Patients with SLE occasionally require joint replacement, most commonly of the femoral head. These prostheses may become painful, dislocated, or infected during pregnancy. Neurologic complications of SLE are rare during pregnancy but include seizures, chorea gravidarum, and stroke.

Effect on the Fetus

Maternal SLE impairs fetal survival and increases the risk for preterm delivery. The primary predictor for adverse pregnancy outcomes is the presence of lupus anticoagulant (LAC). A systematic review of papers published between 1980 and 2009 showed that preterm delivery occurred in 39.4% of 2751 pregnancies in 1842 patients. In the Hopkins Lupus Pregnancy Cohort, preterm birth occurred in 38 of 57 (67%) pregnancies in women with moderate to severe active SLE, compared with 68 of 210 (32%) pregnancies in women with inactive or mild active SLE. Improved perinatal management and control of disease activity have reduced the rate of fetal loss from 43% (between 1960 and 1965) to 17% (between 2000 and 2003). Data from California showed a preterm delivery rate in SLE that was six times higher than that found in the general population.

Neonatal lupus erythematosus (NLE) is a syndrome that results from maternal autoantibodies against Ro (SS-A) or La (SS-B) crossing the placenta and binding to fetal tissue. These autoantibodies are found in up to 87% of patients with SLE, but NLE occurs in only a small proportion of patients. The condition is generally benign and self-limiting, and reversible manifestations such as cutaneous lupus, elevation in aminotransferase levels, and thrombocytopenia resolve as maternal antibodies disappear from the neonatal circulation within 8 months of birth. Anti-Ro/anti-La antibodies may bind to fetal cardiac conduction cells in utero, leading to cell death and irreversible fetal heart block. Neonatal congenital heart block occurs in 2% of neonates when anti-Ro antibody is detected in the mother, and mortality occurs in about 18% of affected babies. Fetal echocardiography reveals atrioventricular dissociation, cardiac dilation, and pericardial effusion. Treatment includes prompt delivery, newborn cardiac pacing, antepartum administration of dexamethasone, and consideration of apheresis to remove maternal antibodies.

Medical Management

Optimally, women with SLE should delay pregnancy until their disease has been quiescent for at least 6 months, and they should be taking “acceptably safe” medications at the time of conception. Medications with acceptable safety are used to minimize disease activity during gestation.

Disease-modifying antirheumatic drugs (DMARDs) and immunosuppressive agents form the mainstay of treatment. Antimalarial drugs, most notably hydroxychloroquinine (HCQ), are frequently used to reduce SLE activity. Discontinuation of HCQ just before conception or in early pregnancy leads to a significant increase in disease activity. A systematic review of English literature (1982 to 2007) found that antimalarial drugs, particularly HCQ, prevent lupus flares; increase long-term survival; contribute to protection against irreversible organ damage, thrombosis, and bone loss; and have low toxicity. HCQ should be continued in all women who were taking it before conception, and it may be used to treat flares during gestation. In contrast, mycophenolate mofetil should be discontinued before conception owing to the risk for teratogenicity.

Immunosuppressive agents such as azathioprine, cyclosporine, and tacrolimus are considered safe for use in pregnancy and may be considered if corticosteroid avoidance is desirable. Azathioprine should be continued if used before conception. The fetal liver does not express the enzyme necessary to convert azathioprine to its active form, but maternal use of azathioprine has been associated with reversible neonatal lymphopenia, depressed serum immunoglobulin levels, and decreased thymic size in the newborn. Transplant registries do not report increased risk for congenital abnormalities after its use.

Corticosteroids may be used to treat flares of SLE disease activity. Antenatal exposure to low-dose prednisone (less than 20 mg daily) appears to be safe, and most children born to mothers who have used antenatal corticosteroids develop normally. However, fluorinated glucocorticoids such as dexamethasone or betamethasone readily cross the placenta and may cause fetal growth restriction and abnormal neuronal development, although orofacial cleft abnormalities are not associated with their use. Corticosteroid therapy may precipitate gestational diabetes, and patients should be monitored for evidence of glucose intolerance. Striae, gastrointestinal ulceration, and bone demineralization may complicate long-term corticosteroid therapy. Regular use of antacids may be required. The pediatrician should be alerted to the possibility of neonatal adrenal suppression.

Aspirin, nonsteroidal anti-inflammatory drugs (NSAIDs), and stronger analgesics may be used to manage lupus arthritis. Although there is no evidence of teratogenicity with these agents, concern exists that NSAIDs may cause premature closure of the fetal ductus arteriosus, and high-dose aspirin and NSAIDs ideally should be discontinued by 30 weeks’ gestation.

Biologic agents such as rituximab and belimumab are being used more frequently, but assessment of their efficacy has been limited by the heterogenous nature of SLE and the connective tissue disorders. There is little consensus on their use at present.

Obstetric Management

Patients with SLE are at increased risk for intrauterine fetal death and preterm delivery. Estimation of the gestational age is obtained with ultrasonography at the first prenatal visit and again at 20 weeks’ gestation. Continued surveillance consists of nonstress testing, biophysical profile measurement, and/or umbilical artery Doppler velocimetry beginning at 26 to 28 weeks’ gestation and performed weekly until delivery.

The coexistence of antiphospholipid antibodies (either lupus anticoagulant or triple-positive antiphospholipid antibody; see later discussion) predicts a much higher maternal and fetal risk. Maternal serologic markers are checked regularly, together with platelet count, creatinine clearance, 24-hour urine protein level, uric acid, and complement levels. In normal pregnancy, serial complement levels gradually increase. However, declining levels of C3 and C4 suggest active disease and lupus nephritis. Thromboprophylaxis is important in patients with antiphospholipid syndrome (see later discussion). Aspirin resistance may predict adverse maternal and neonatal outcomes. Regular assessment of blood pressure, weight gain, and proteinuria is performed to detect the development of preeclampsia. Prediction of cardiovascular risk in women with SLE may be possible using the Pulse-Mass Index.

The timing and route of delivery are individualized. Although vaginal delivery is preferred, a cesarean delivery rate of 40% has been reported in parturients with SLE.

Anesthetic Management

The obstetrician, rheumatologist, and anesthesia provider should formulate a joint plan for delivery. Maternal organ system involvement, current disease severity, and particularly the presence of flares must be assessed.

Pericarditis is common in patients with SLE and is typically asymptomatic. A history of dyspnea on exertion or unexplained tachycardia may suggest pericarditis or myocarditis. Cardiac tamponade has been reported. Prolongation of the PR interval or nonspecific T-wave changes may be seen on the electrocardiogram. Coronary artery vasculitis, accelerated atherosclerosis leading to myocardial ischemia, and even myocardial infarction in young women have been reported.

An echocardiographic study in 69 patients with SLE showed a high incidence of valvular abnormalities, including valvular thickening in 51%, vegetation in 43%, regurgitation in 25%, and stenosis in 4%. Current American Heart Association guidelines recommend antibiotic prophylaxis only for patients at highest risk for infective endocarditis in whom there is both significant risk and consequence of infection. Prophylactic antibiotics are not recommended for women with common valvular lesions undergoing genitourinary procedures, including vaginal delivery, but are specifically indicated for those with previous infective endocarditis, unrepaired cyanotic congenital heart disease, implanted prosthetic material or devices, or a history of cardiac transplantation with cardiac valvulopathy.

The prevalence and progression of pulmonary hypertension in 28 patients with SLE has been studied. The prevalence increased from 14% at initial evaluation to 43% 5 years later. Epidural anesthesia for cesarean delivery in parturients with pulmonary hypertension has been reported (see Chapter 41 ). The abrupt onset of sympathetic blockade and decreased venous return may cause precipitous systemic hypotension and hypoxemia. One report described the administration of general anesthesia in a parturient with SLE and pulmonary hypertension, with coexisting SLE-related restrictive lung disease, pulmonary edema, and orthopnea. In one report of three parturients with pulmonary hypertension secondary to SLE and antiphospholipid syndrome, two died of right-sided heart failure within 48 hours of delivery.

Subclinical pleuritis is common, but significant pleural effusions are rare. Patients may suffer from infectious pneumonia or lupus pneumonitis. The latter condition is characterized by fleeting hemorrhagic infiltrates that may become consolidated. Pulmonary embolism and diaphragmatic dysfunction have been reported.

Central and peripheral sensorimotor and autonomic neuropathies are observed in as many as 25% of patients with SLE, and vocal cord palsy has been reported. These deficits should be documented before the administration of either neuraxial or general anesthesia. Migraine headache and cerebral vasculitis resulting from SLE must be considered in the differential diagnosis of a postpartum headache. Psychological disorders and frank psychosis can occur during disease flares. Seizures can occur, especially if chronic anticonvulsant medications are discontinued inadvertently.

Hematologic abnormalities, including anemia, thrombocytopenia, and coagulopathy, should be documented. An abnormality of the activated partial thromboplastin time (aPTT), which is not corrected with a 1 : 1 control plasma mix, suggests the presence of either lupus anticoagulant (a coexistent but separate disease entity) or, more rarely, true autoantibodies against specific coagulation factors (e.g., VIII, IX, XII). Lupus anticoagulant is a laboratory artifact that does not cause clinical coagulopathy. True coagulation factor autoantibodies (or inhibitors) may result in a significant bleeding diathesis, which contraindicates the administration of neuraxial anesthesia.

Long-term use of NSAIDs leads to qualitative platelet abnormalities, but their role in causing spinal epidural hematoma remains conjectural. In prospective studies of patients undergoing orthopedic procedures (924 patients receiving spinal or epidural anesthesia, and with preoperative antiplatelet medications taken by 39% of patients) and epidural steroid injection for chronic pain management (1035 patients, with NSAID use reported by 32% of patients), no cases of spinal epidural hematoma were observed. In the Collaborative Low-dose Aspirin Study in Pregnancy (CLASP), a large, multicenter randomized trial, 9364 pregnant women received either low-dose aspirin (60 mg daily) or placebo for prevention and treatment of preeclampsia. Of 5000 enrollees, at least 1069 patients received epidural analgesia, and no cases of epidural hematoma were observed. Measurement of the bleeding time before neuraxial injection in patients taking aspirin or NSAIDs is not indicated. Thromboelastography has been suggested as an alternative but is not widely available. One case report described administration of general anesthesia for cesarean delivery in a patient with SLE and antiphopholipid syndrome who had suffered a thrombotic cerebrovascular event; thromboelastography was used as an adjunct to coagulation studies. Atypical blood antibodies may complicate cross-matching of blood for patients with SLE, and additional time may be required.

Prosthetic orthopedic joints should be positioned carefully during vaginal or cesarean delivery. Lupus arthritis rarely involves the cervical spine. Women who have undergone long-term corticosteroid therapy should receive a peripartum stress dose of a corticosteroid.