Systemic lupus erythematosus is an autoimmune disorder that causes inflammation of connective tissue. The most common type of this disorder is discoid lupus erythematosus (DLE), which affects only exposed areas of skin. The more serious form, systemic lupus erythematosus (SLE), affects many of the body systems, including the skin.
In both varieties of lupus erythematosus, the symptoms periodically subside and recur with varying severity.
In DLE, the rash starts as one or more red, circular, and thickened areas of skin. These areas of skin subsequently become scarred. The patches may occur on the face, behind the ears, and on the scalp.
SLE causes a variety of symptoms. A characteristic red, blotchy, butterfly shaped rash may appear on the cheeks and the bridge of the nose; other symptoms include fatigue, fever, nausea, loss of appetite, joint pain, and weight loss. There may also be associated problems including anaemia, neurological or psychiatric problems, kidney failure, pleurisy (inflammation of the membrane that lines the lungs), arthritis, and pericarditis (inflammation of the membrane surrounding the heart).
Diagnosis and treatment
Diagnosis is made by blood tests and sometimes a skin biopsy. DLE is usually treated with topical corticosteroid drugs. Both DLE and SLE may be treated with the antimalarial drug hydroxychloroquine. Sufferers of mild forms of SLE may have near normal health for many years, and treatment with corticosteroid drugs and immunosuppressant drugs can improve life expectancy. Other treatments are available to treat specific features of the disease; however, SLE is still a potentially fatal disorder.
Systemic lupus erythematosus and related disorders in detail - technical
- Historical perspective
- Aetiology and pathology
- Clinical features
- Differential diagnosis
- Clinical investigation
- Criteria for diagnosis
- SLE in pregnancy
- Occupational and psychological aspects of SLE
- Controversial areas and future prospects
- Further reading
Systemic lupus erythematosus (SLE) is an autoimmune rheumatic disorder that can present with symptoms in almost any organ or system of the body. It is 10 to 20 times commoner in women than men, and commoner in Afro-Caribbeans than Asians than whites.
Aetiology is multifactorial, incorporating genetic, hormonal, and environmental elements. No single abnormality of the immune system can be considered responsible, pathogenesis depending on the interplay of a number of different factors, including autoantibodies, T lymphocytes, cytokines, the complement system, and apoptosis.
Common symptoms are constitutional (fatigue, anorexia), musculoskeletal (arthralgia/arthritis, myalgia), dermatological (alopecia, butterfly rash, vasculitic skin lesions, purpura), cardiopulmonary (breathlessness, pleurisy), and neurological (migraine, seizures, depression, psychosis).
Examination may show evidence of weight loss, low-grade fever, lymphadenopathy, arthritis (but rarely synovitis or deformity), skin rash, oral ulcers, dry eyes/mouth, pleural rub, and peripheral neuropathy (usually sensory).
Investigation and diagnosis
Investigation commonly reveals abnormalities in the following systems: (1) renal—proteinuria, microscopic haematuria, impaired glomerular filtration rate; (2) haematological—anaemia, leucopenia, lymphopenia, thrombocytopenia; and (3) cardiopulmonary—pulmonary function abnormalities.
The American College of Rheumatology classification criteria require four or more of the following to be present at some time: (1) malar rash; (2) discoid rash; (3) photosensitivity; (4) oral ulcers; (5) arthritis; (6) serositis; particular types of (7) renal, (8) neurological, and (9) haematological disorders; (10) immunological disorders (particular autoantibodies); and (11) raised titres of antinuclear antibody. In everyday practice, however, these requirements may be too stringent, and systemic lupus erythematosus should be suspected on the basis of typical clinical findings in one organ or tissue combined with the presence of appropriate autoantibodies. The antinuclear antibody assay is a sensitive (>95%) but not specific test for SLE, hence the absence of antinuclear antibody in a patient with suspected lupus raises serious doubt about the diagnosis. The presence of anti-dsDNA (and anti-Sm) antibodies is virtually specific for lupus. The most reliable measures of highly active disease are depletion of complement, and high anti-dsDNA levels.
Prognosis and management
SLE can kill (mortality c.10% at 10 years from diagnosis), but it may run a fairly indolent course in which an initial flare is followed by many years of low-grade activity. General treatment measures include (1) rest—as appropriate; (2) avoidance of overexposure to sunlight; (3) attention to modifiable cardiovascular risk factors—women with lupus between 35 and 45 years have a 50× increased risk of coronary disease; and (4) prophylaxis / treatment of osteoporosis—usually induced by steroid therapy.
Mild disease—patients whose disease activity is confined to arthralgia, tiredness, and/or mild rash can often be treated symptomatically, e.g. with simple analgesics and/or nonsteroidal anti-inflammatory agents (NSAIDs), with hydroxychloroquine added if these are not sufficient.
Treatment of flares of disease—corticosteroids and cytotoxic agents are used. A mild flare of arthralgia, myalgia, and general fatigue may be alleviated by a single intramuscular dose of corticosteroid. More severe flares of arthritis, pleuritis or pericarditis require oral prednisolone (20–40 mg daily). Renal flares require the most aggressive treatment, generally involving both corticosteroids (high-dose oral and/or intravenous pulse) and cyclophosphamide/mycophenolate mofetil. Biological therapies will be increasingly used in the future: B-cell depletion with the anti-CD20 chimeric reagent rituximab looks very promising, and trials are under way with many other agents.
Antiphospholipid antibody syndrome—immunosuppression is rarely useful and aspirin (150–300 mg daily) is recommended, with lifelong anticoagulation advised for those who have suffered recurrent thromboses or cerebral infarcts.
Pregnancy—SLE may be exacerbated during the pregnancy. Babies born to mothers with lupus are prone to the transient condition of neonatal lupus, also to heart block (particularly if the mother has anti-Ro and anti-La antibodies).
Read more: SLE: diagnosis and management
Systemic lupus erythematosus (SLE) is an autoimmune rheumatic disorder that can present with symptoms in almost any organ or system of the body. Classification criteria that have been published by the American College of Rheumatology should be used to make the diagnosis. These are shown in Table 1 and demonstrate the wide variety of clinical and serological features that are associated with this condition. They provide a useful guide to the clinical features that should place the suspicion of SLE in the mind of a clinician. It is important not to be too dogmatic in searching for ‘pathognomonic’ features of the disease, e.g. although the characteristic butterfly rash over the face is perhaps the best-known sign of SLE, many patients will never develop such a rash.
Although the term ‘lupus’ has been used for several hundred years, its meaning was vague until Cazerave and Clausit coined the term ‘lupus erythematosus’ in 1852, confining the condition to a skin rash affecting the face. The photosensitive nature of the rash (noted by Hutchinson in 1879) and the accompanying internal organ involvement (Kaposi in 1872) helped to frame the current usage of the term ‘systemic lupus erythematosus’ to define a multisystem disorder rather than a purely cutaneous condition. The development of the LE cell test (Hargreaves in 1948) and more importantly (and more specifically) the identification of anti-double-stranded (ds) DNA antibodies by four different laboratories (in 1957) facilitated the classification criteria by which SLE is now widely recognized.
|Table 1 Criteria of the American College of Rheumatology for the classification of systemic lupus erytematosus*|
|1 Malar rash|
|2 Discoid rash|
|4 Oral ulcers|
|7 Renal disorder|
|8 Neurological disorder|
|9 Haematological disorder|
|10 Immunological disorders|
|11 Antinuclear antibody in raised titre (in the absence of drugs known to be associated with drug-induced lupus)|
* …a person shall be said to have SLE if four or more of the 11 criteria are present, serially or simultaneously, during any interval of observation.’ (Tan EM et al. (1982). The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis and Rheumatism 25, 1271–7).
Aetiology and pathology
The aetiology is multifactorial, incorporating genetic, hormonal, and environmental elements. The best-established genetic link is with the presence of null alleles of genes encoding early components of the complement cascade (C1q, C2, and C4). Over 90% of patients homozygous for C1q deficiency and 75% of those with C4 deficiency develop a lupus-like disease (similar clinical features but a relative paucity of antibodies). Major histocompatibility complex (MHC) genes, particularly HLA A1, B8, and DR3, have also been associated with the presence of lupus in family studies, although part of this association may be due to linkage disequilibrium with the C4 and C2 genes also present in that region of chromosome 6.
Hormones are likely to play a role in pathogenesis, because SLE is far more common in women than in men (see below). There is a relatively high incidence of the condition in Klinefelter’s syndrome (males with the XXY karyotype), which is associated with abnormalities in oestrogen metabolism.
Viruses may be important in triggering the autoimmune dysfunction that leads to the production of pathogenic autoantibodies in SLE. Reactivation of BK polyomavirus infection, in particular, has been associated with the presence of antibodies to dsDNA in Norwegian studies. This association has not yet been confirmed in large populations.
Certain drugs induce a form of SLE that is generally characterized by the presence of antihistone rather than anti-dsDNA antibodies, a milder course of disease, and total remission when the causative drug is withdrawn. The most common drugs involved are isoniazid, procainamide, hydralazine, minocycline, penicillamine, and anticonvulsants.
No single abnormality of the immune system can be considered to be the sole cause of SLE. The pathogenesis of the disease depends on the interplay of a number of different factors, the relative importance of which may differ from one patient to another. These include autoantibodies, T lymphocytes, cytokines, the complement system, and apoptosis. Research to unravel this complex system of interrelated factors has been carried out by studying properties of cells and tissue components derived from patients with SLE and by studying mouse models of the condition.
Abnormalities of the immune system
B lymphocytes and autoantibodies
Autoantibodies are those that bind to antigens present within the tissues of the body itself. A wide variety of different autoantibodies has been described in SLE. Those most frequently reported are listed in Table 2.
Anti-dsDNA antibodies have been cited widely as possible causative agents in SLE, particularly in lupus glomerulonephritis. Raised titres of anti-dsDNA antibodies are found in 50 to 70% of patients with SLE, but hardly ever in healthy people or those with other diseases. Levels of these antibodies often rise and fall with disease activity, and deposits of anti-dsDNA antibodies occur in the glomeruli of patients with lupus nephritis. In experimental murine models of SLE, monoclonal anti-dsDNA antibodies can be shown to deposit in the glomeruli and to be associated with proteinuria.
|Table 2 Major autoantibodies associated with systemic lupus erythematosus and their approximate prevalence in patients with the disease|
|Autoantibodies||Antigen/epitope||Approximate prevalence (%)|
|Histone||H1, 2A, 2B, 3, 4||30–80|
|Sm||B/B’, D, E, F, G||30 (Afro-Caribbean), ∼ 10 (Caucasian)|
|U1RNP||A, C, 70 kDa ribonucleoprotein||20–35|
|rRNP||Three subunits: 38, 19, 17 kDa||5–15|
|Ro/SS-A||60, 52 kDa protein bound to cytoplasmic RNA (hY1–hY5)||10–15|
|La/SS-B||48 kDa protein bound to variety of RNA, U1RNA, hY RNA||10–15|
|Heat shock protein (hsp)||hsp 90||30|
|hnRNP||A2 protein (also known as RA-33)||30|
|Neuronal antigen||Expressed on neuronal cell lines grown in vitro||70–90 (+CNS), ∼ 10 (–CNS)|
|Lymphocyte||HLA component||∼ 75 (IgM), ∼ 45 (IgG)|
|Red cell||Non-Rh related||< 10|
|Rheumatoid factor||Fc region of IgG||∼ 25|
The titre of anti-dsDNA antibodies present in the bloodstream of patients with SLE can be a useful indicator of disease activity. It is increasingly clear, however, that not all anti-dsDNA antibodies are equally likely to be associated with tissue damage. Antibodies of the IgG isotype, which show specific high-affinity binding to dsDNA, generally show the closest association with disease activity in patients and the greatest ability to cause renal damage in experimental models.
Why are anti-dsDNA antibodies produced in patients with SLE?
Studies of monoclonal anti-dsDNA antibodies derived from patients or mice indicate that those that show the isotype and binding properties described above often show sequence characteristics suggestive of antigen-driven somatic mutation. This is the process whereby mutations accumulate in the expressed immunoglobulin gene sequences of a B lymphocyte under the influence of a particular antigen. The mutations are accumulated nonrandomly, such that the end-result is an increase in specificity and affinity of binding. This process is dependent on help from T lymphocytes and on the presence of an appropriate antigen. Naked mammalian DNA is, however, a poor immunogen in experimental animals, and the concentration of free DNA in the bloodstream is low even in patients with SLE. It is therefore believed that the antigen that stimulates production of high-affinity anti-dsDNA antibodies is probably a complex of DNA and protein. Nucleosomes derived from cell apoptosis (see below) may be the most important antigens involved in stimulating both T cells and B cells in SLE, although a role for viral DNA-binding proteins has also been suggested.
How do anti-dsDNA autoantibodies exert their pathogenic effects?
Deposition of IgG and complement in inflamed tissues such as kidney and skin is a consistent feature of active SLE. The pathogenic potential of autoantibodies in SLE (particularly IgG anti-dsDNA) may therefore rest on their ability to deposit in these tissues and to activate complement. Why are anti-dsDNA antibodies deposited in target tissues? Much of the work designed to answer this question has concentrated on autoantibodies in lupus nephritis. Originally, it was felt that DNA–anti-DNA immune complexes would form in the bloodstream and accumulate in glomeruli as the blood was filtered there. However, it has not been possible to demonstrate large quantities of such complexes in the blood of patients with SLE, although their clearance may well be abnormal due to complement deficiency. Anti-dsDNA antibodies may be targeted to the kidney due to cross-reaction with cell surface proteins there, or may deposit due to an interaction with histones and heparan sulphate. According to this latter model, anti-dsDNA antibodies bind to DNA in nucleosomes, and the positively charged histones in these nucleosomes bind to negatively charged heparan sulphate in the renal basement membrane.
Between 20 and 30% of patients with SLE possess serum antiphospholipid antibodies. The origin of these antibodies may be similar to that of anti-dsDNA antibodies because monoclonal antiphospholipid antibodies from patients with SLE also show antigen-driven accumulations of somatic mutations. The antigen in this case may be phosphatidylserine on the outer surfaces of blebs derived from apoptotic cells.
Antiphospholipid (APL) antibodies may be present in healthy people or in those with infectious diseases such as syphilis, in which case they have no adverse effects. In patients with SLE, however, APL antibodies may cause arterial or venous thromboses or miscarriages. The combination of these clinical problems with the presence of APL antibodies defines the antiphospholipid syndrome (APS). APS may occur either in patients with other autoimmune diseases (particularly SLE), or alone in the absence of other disease (primary APS). Although it was previously thought that APL antibodies exerted their effects almost wholly through promotion of thrombus formation, it is now clear that they may have many other direct effects on platelets, monocytes, endothelial cells, and the trophoblast. The mechanism by which thrombosis is altered is not fully understood, but it has become clear that APL antibodies found in APS often bind to protein antigens associated with phospholipids rather than the phospholipids themselves. The most important of these proteins is β2-glycoprotein 1, and a direct test for anti-β2-glycoprotein 1 is an alternative diagnostic test for APS.
As the process of antigen-driven selection of mutations in B lymphocytes is dependent on help from helper T lymphocytes, it would be reasonable to suppose that antigen-specific T cells might also contribute to the pathogenesis of the disease. The isolation of T-cell clones reactive with DNA and/or DNA-binding proteins such as histones has been demonstrated from both patients with SLE and murine models of the disease. The clones frequently show specificity for histone epitopes that are cryptic (i.e. not exposed) in normal chromatin. These results reinforce the idea that the antigenic stimulus for production of pathogenic T cells and autoantibodies in SLE may be a DNA/histone complex rather than DNA alone.
Patients with SLE have decreased levels of the subset of T cells carrying the CD4 and CD45 Ro surface markers. This population may be involved in stimulation of suppressor T lymphocytes such that suppression in these patients is insufficient to prevent the production and survival of autoreactive B-lymphocyte and helper T-lymphocyte clones.
Apoptosis and complement
The links between apoptosis, complement, and lupus are complicated. MRL lpr/lpr mice are deficient in apoptosis because they lack the Fas protein that plays a major role in promoting this process. These mice develop a disease very similar to SLE, with death resulting from glomerulonephritis. One possible reason for this might be the failure of the immune system to delete by apoptosis autoreactive clones of T or B lymphocytes, which are then able to cause autoimmune disease. By contrast, humans with the equivalent genetic lesion to MRL lpr/lpr mice do not develop SLE, and other strains of mice show an accumulation of apoptotic debris within nephritic kidneys, which resemble those of SLE. A simple deficiency in apoptosis is therefore unlikely to be the underlying mechanism in SLE.
Apoptosis leads to the production of surface blebs of cellular material. These blebs include a number of the antigens to which autoantibodies develop in SLE, notably DNA and associated nuclear proteins and negatively charged phospholipids. A deficiency in the clearance of products of apoptosis has been demonstrated, which might allow the production of as wide a spectrum of autoantibodies as found in SLE. Removal of immune complexes containing such potentially antigenic material may be compromised in patients with SLE. Monocytes derived from such patients show reduced phagocytosis of cell debris in vitro. This process may be complement dependent, as people with homozygous C2 deficiency process immune complexes very differently from normal controls. Administration of fresh frozen plasma to such patients as a source of complement is successful in ameliorating the symptoms of SLE and in normalizing (albeit transiently) their processing of immune complexes.
C1q knockout mice develop a form of glomerulonephritis similar to that seen in SLE, and their kidneys are characterized by accumulations of apoptotic debris. In fact knockout mice deficient in both the classic and alternative pathways of complement develop this form of glomerulonephritis, showing that in this model the protective effects of complement outweigh its role as an effector of inflammation. Similarly, as noted earlier, humans homozygous for C1q deficiency develop a form of SLE with the frequent occurrence of nephritis.
Cytokines enhance the ability of cells to interact and are therefore critically important in abnormalities in both T- and B-cell functions seen in patients with lupus. Table 3 summarizes the major differences between the different subsets of T-helper (Th) cells in terms of their cytokine profiles and functions. The balance between cytokines from Th1 and Th2 cells is essential in determining the outcome of the immune response. Lupus might be expected to be a disease in which Th2 cells predominate, resulting in excessive help for B cells and overproduction of antibodies. In support of this notion, increased levels of interleukin 10 (IL-10) have been found in patients with lupus. This cytokine promotes secretion of antibodies by B lymphocytes but suppresses Th1 cells and thus impairs cell-mediated immunity, a characteristic feature of the disease. Both macrophage and natural killer cell-mediated cytotoxicity are frequently impaired in patients with lupus. Interferon-γ-induced enhancement of both types of cytotoxicity is also impaired, despite normal levels of interferon-γ production by lupus Th1 cells.
Accessory cells in lupus seem to produce insufficient amounts of IL-1 to provide the necessary activation signals for T cells. Both CD4+ and CD8+ T cells have been described as producing either normal or decreased amounts of IL-2 in response to exogenous antigens. Such a reduction is likely to have a profound effect on T-cell responses.
Interferon-α is increasingly recognized as playing a major role in the development of SLE. Levels of this cytokine are raised in patients with SLE, genetic studies show that interferon-related genes are risk factors for SLE, and gene expression studies show that such genes are activated in patients with the disease. Lupus in the NZB/NZW mouse model of lupus can be accelerated by the presence of interferon-α. Cytokines such as interferon-α and IL-10 are increasingly being considered as possible targets for future therapeutic agents in lupus.
|Table 3 (a) Subsets of CD4+ T cells|
|T-helper 1 cell||Cell mediated immunity||IFN-γ, IL-10 (humans only), IL-12, TNF-α|
|T-helper 2 cell||B cell help||IL-4, IL-10|
|(b) Cytokine profiles in patients with active systemic lupus erythematosus|
|Cytokine||Serum level*||Spontaneous||Stimulation in vitro|
|TNF-α||↑ (or normal)||↓ (DR2, DQw1; ↑ nephritis), ↑ (DR3, 4; ↓ nephritis)||↓|
|IL-1||n.d.||↑ PBM production||↓ Monocyte production|
|IL-10||↑ (or normal)||↑ (or normal)||Normal|
* Serum levels of cytokines are difficult to interpret since these may be affected by soluble cytokine receptors which are shed from cells. Among the known shed receptors are those for IL-1, IL-2, IL-6, TNF-α, and IFN-γ. Soluble TNF-αR and IL-2R levels are increased in systemic lupus and correlate with disease activity and lupus nephritis.
Abbreviations: IFN-γ, interferon-γ; n.d. = not detected; PBM = peripheral blood mononuclear cell, TNF-α, tumour necrosis factor-α.
The two tissues most often subjected to biopsy in SLE are the skin and kidneys.
Skin biopsies are chiefly carried out to facilitate the diagnosis of an atypical rash. If SLE is suspected, it is important to take a sample of apparently normal skin as well as skin from the rash. Both should show deposition of IgG and complement at the dermoepidermal junction (Fig. 1).
Figure 1: Immunofluorescence microscopy showing deposition of IgG at the dermoepidermal junction in the skin of a patient with systemic lupus erythematosus (sometimes called the lupus band test).
Renal biopsy may be performed to establish the diagnosis of SLE when this is not certain, e.g. in a patient with a poorly characterized multisystem disease with renal involvement. In a patient with known SLE it may be employed to help determine prognosis and decide on treatment when renal function is deteriorating, e.g. with the development of nephrotic syndrome and/or declining glomerular filtration rate (GFR). The glomerular pathology can be graded on a scale of I to V according to the World Health Organization criteria, and scores for activity and chronicity can be used to determine appropriate treatment and the risk of a progressive decline in renal function.
The incidence of SLE in the United Kingdom is about 4 cases/100 000 people per year. It occurs between 10 and 20 times more frequently in women than in men, and is more common in some ethnic groups. A study in Birmingham, United Kingdom gave the prevalence of SLE in women as 206 per 100 000 in people of African–Caribbean origin, 91 per 100 000 in those of Asian origin, and 36 per 100 000 in white people. These gender and racial differences are broadly consistent with those reported from studies in the United States of America and the Caribbean, although the reported prevalence of SLE in Africa is much lower.
SLE is a chronic condition punctuated by flares of acute activity. The overall severity of the disease in a particular patient depends on the nature and frequency of these flares and the long-term permanent damage that they cause.
The diverse clinical features of SLE mean that the disease may present to any of a number of different specialists, including rheumatologists, dermatologists, nephrologists, and general physicians. It is important to be aware of SLE as a possible diagnosis in any patient, especially a woman aged between 15 and 50, in whom a number of different organs are inflamed either simultaneously or sequentially. The frequency of occurrence of symptoms in various organs is shown in Table 4.
According to the diagnostic guidelines published by the American College of Rheumatology (see Table 1), SLE may be diagnosed where a patient meets at least 4 of the 11 criteria specified (though not necessarily at a single time). In everyday practice, however, these requirements may be too stringent, and SLE is often suspected on the basis of typical clinical findings in one organ or tissue combined with the presence of appropriate autoantibodies.
Patients with SLE find fatigue to be the most troublesome feature of the disease, excessive tiredness being both very common and difficult to treat. Hypothyroidism coexists in 5 to 10% of patients with SLE and so thyroid function tests should be performed in the fatigued patient. There is an ongoing debate as to whether fibromyalgia is a significant comorbid condition.
Weight loss and low-grade fever may both be indicative of disease activity. Lymphadenopathy is also recognized. The nodes may be markedly enlarged but show no diagnostic features on biopsy, which may nevertheless be necessary to exclude other conditions such as lymphoma.
Arthralgia/arthritis is the most common symptom in SLE, occurring in 90% of patients. This may be severe but is rarely associated with frank synovitis. Effusions may occur but the fluid shows no diagnostic features.
Erosive arthritis is uncommon, but up to 5% of patients may have an overlap syndrome with features of rheumatoid arthritis as well as SLE. These patients tend to have both serum rheumatoid factor and erosions.
When progressive deformity of the hands does occur in SLE, it is usually due to an aggressive tenosynovitis and tendon dysfunction rather than to joint damage. This leads to reversible subluxation of the joints, often known as Jaccoud’s arthropathy (Fig. 2).
Figure 2: Deforming Jaccoud's arthropathy.
Development of hip pain in patients who have been treated with corticosteroids should raise the suspicion of avascular necrosis of the femoral head, which may be diagnosed on a plain radiograph or, in earlier stages, by MRI. Corticosteroids also promote osteoporosis, which can be diagnosed in the presymptomatic phase by bone density scanning, but may present with the acute pain of a vertebral fracture.
Myalgia is common and a true myositis may occur in 5% of cases. Corticosteroid-induced proximal myopathy may also be a problem where these drugs have been used for long periods.
|Table 4 Cumulative prevalence of clinical features in patients with systemic lupus erythematosus|
|Clinical feature||Approximate cumulative prevalence (%)|
|Shortness of breath||40|
|Pulmonary function abnormalities||85|
|Serum albumin < 35 g/l||30|
|Serum creatinine > 125 μmol/l||30|
|Reduced 24-h creatinine clearance||35|
|Cranial nerve lesions||10|
|Anaemia (iron deficiency)||30|
|Anaemia (of chronic disease)||75|
|Autoimmune haemolytic anaemia||15|
|Erythematous maculopapular eruption||35|
|Relapsing nodular non-suppurative panniculitis||< 5|
|Vasculitic skin lesions||40|
Cutaneous and mucosal involvement
Photosensitivity is very common, particularly in white female patients. Patients should be advised to avoid strong sunlight and to wear protective clothing and/or a high-factor sunblock.
The butterfly rash over the malar area of the face occurs in up to one-third of patients. A number of other forms of cutaneous involvement can occur, although these are less specific for SLE. These include maculopapular rash, discoid lesions, alopecia, and nailfold infarcts. Scarring alopecia may be particularly distressing and difficult to treat.
A variant of SLE in which cutaneous manifestations dominate is known as subacute cutaneous lupus. This condition is often associated with anti-Ro antibodies and may be exacerbated by smoking cigarettes.
APL antibodies are associated with a nonraised lattice-like rash concentrated particularly over the thighs and arms: livedo reticularis.
Recurrent crops of oral ulcers are common enough to be recognized as one of the diagnostic criteria for SLE. About 20% of patients develop secondary Sjögren’s syndrome; in this condition the dry eyes and mouth may respond to artificial tears and saliva.
Glomerulonephritis is the most serious and potentially lethal manifestation of SLE. Its presence may be detected by the finding of haematuria and/or proteinuria on routine stick testing of the urine. It may present as the nephrotic syndrome or, less commonly, as a florid nephritis with haematuria, proteinuria, hypertension, and acute renal failure with red cell casts in the urine.
It is important to be aware of the possibility of glomerulonephritis in any patient with SLE. Measurement of blood pressure and analysis of urine should be carried out at each consultation. Early diagnosis and treatment are invaluable in avoiding deterioration of renal function to the extent that dialysis or renal transplantation become necessary.
Patients with APS may develop a different type of renal lesion characterized by thrombi in small renal vessels rather than by glomerulonephritis. These patients develop hypertension and impairment of renal excretory function, detected as a fall in estimated GFR (eGFR), rather than proteinuria, and are best managed by anticoagulation rather than immunosuppression.
The most common form of respiratory involvement in SLE is pleuritis, manifesting either as pleuritic chest pain or as breathlessness caused by pleural effusion. The lung parenchyma is more rarely involved, but fibrosis can occur.
A patient with SLE may present with shortness of breath or chest pain for a number of reasons. Pulmonary emboli must be suspected in those with APL antibodies. Infections are common in immunosuppressed patients and rib fractures may occur, particularly in those rendered osteoporotic by treatment with corticosteroids.
The shrinking lung syndrome is characterized by reduced lung volumes and poor respiratory reserve in the face of a normal appearance of the lung parenchyma on CT. It is believed to arise from basal atelectasis in association with diaphragmatic dysfunction.
The most common cardiac manifestation of SLE is pericarditis, which occurs in about 15% of patients. This generally presents with chest pain or an asymptomatic friction rub. Pericardial effusions may occur, but are rarely large enough to cause haemodynamic compromise.
Myocarditis and endocarditis are less common, though postmortem and echocardiographic studies suggest that both may occur without symptoms in a significant proportion of patients with SLE, e.g. the classic endocarditis described by Libman and Sacks is characterized by small vegetations that often do not cause murmurs or cardiac compromise, but which have been identified in up to 50% of patients with SLE post mortem.
It is increasingly recognized that atherosclerosis and its sequelae in the cerebral and cardiac circulations are more common in patients with SLE than in the general population. Women with lupus aged between 35 and 45 have a 50 times increased risk of coronary disease. This may be partially due to the use of corticosteroids, which raise serum cholesterol and can promote hypertension. Patients possessing APL antibodies are also at a higher risk of stroke or arterial thrombosis.
Raynaud’s phenomenon occurs in about a third of patients with SLE, although it is not usually as severe as that seen in systemic sclerosis. Vasculitis presents with a skin rash or ulcers that may be very difficult to heal, but rarely affects the internal organs.
Similar to pleuritis, peritonitis may occur in patients with SLE and must be considered in the event of abdominal pain, while remembering that patients with lupus are not protected from more common causes of this condition, e.g. appendicitis.
Involvement of the liver and pancreas is recognized but uncommon. The term ‘lupoid hepatitis’ was previously used for a form of autoimmune hepatitis characterized by the presence of autoantibodies, but these patients do not generally have any form of SLE and the term is misleading. Minor enlargements of the liver and/or spleen occur in 10 to 25% of cases, but these are usually asymptomatic and require no treatment.
SLE can affect the nervous system in many ways, so that the true incidence of neuropsychiatric involvement is difficult to quantify. Symptoms such as poor memory, change of personality, and depression or anxiety occur in many patients. It is difficult, however, to be sure whether these are caused by cerebral SLE or represent a reaction to the diagnosis and treatment of the disease.
More florid presentations such as psychotic episodes and convulsions are well recognized but rare. By contrast to the milder symptoms noted above, these manifestations generally call for immunosuppression.
Migraine occurs in up to 40% of patients with SLE, particularly in the presence of APL antibodies. Peripheral neuropathy can occur, and is usually sensory rather than motor. Cranial nerve palsies are less common, as is transverse myelitis (another feature linked to APL antibodies).
Ocular involvement can include episcleritis, conjunctivitis, and the presence of cytoid bodies (white patches on the retina). Patients treated with high-dose steroids may develop cataracts.
A normochromic normocytic anaemia is frequently seen in SLE, particularly during periods of high disease activity. Microcytic iron-deficiency anaemia may result from blood loss from gastritis and ulcers in patients treated with nonsteroidal anti-inflammatory drugs. Anaemia may also result from chronic renal failure in lupus nephritis.
A positive Coombs’ test, signifying the presence of antibodies to red blood cells, is present in up to 20% of patients with SLE but does not always indicate haemolytic anaemia.
The presence of lymphopenia (less than 1.5 × 109/l) is common, occurring in up to 80% of patients. Neutropenia may occur secondary to the use of cytotoxic drugs such as azathioprine or cyclophosphamide.
Three different types of thrombocytopenia occur in SLE. The mildest form is characterized by stable platelet levels of between 50 and 100 × 109/l, is rarely symptomatic, and usually requires no treatment. Other patients develop an acute autoimmune thrombocytopenia with levels dropping rapidly below 10 × 109/l, but usually rising when treated with oral steroids. A third group of patients present with thrombocytopenia alone, are treated with steroids, intravenous immunoglobulins, rituximab, or (rarely these days) splenectomy, and some years later develop full-blown SLE. Thrombocytopenia is also one of the cardinal features of the APS and may be severe enough to necessitate splenectomy in that condition.
Other complicating disorders
About 30% of lupus patients have another autoimmune condition, including Sjögren’s syndrome (the most common, affecting some 15 to 20% of patients with lupus), APS (10 to 15%), autoimmune thyroid disease (hyper- or hypothyroidism, 5 to 10%), and (less frequently) rheumatoid arthritis, myasthenia gravis, coeliac disease, diabetes, and pernicious anaemia.
SLE is truly a chameleon of a disease. Its protean clinical manifestations mean that, theoretically, it should enter the differential diagnosis of virtually any unexplained symptom/feature. However, its predilection for women during the child-bearing years does restrict this potential diagnostic confusion significantly.
Among its general features, fever, weight loss, anorexia, and lymphadenopathy may easily be confused with a lymphoproliferative cancer, hence biopsy of a swollen lymph gland may often be required to be sure of the underlying problem. Polyarthralgia or polyarthritis in a young woman could be due to rheumatoid arthritis or lupus. As both rheumatoid factor and antinuclear antibodies can be found in both conditions, there is obvious room for confusion, but anti-CCP antibodies (rheumatoid) and anti-dsDNA antibodies (lupus) are more specific tests that can usually resolve the matter.
About 15% of patients presenting with ‘idiopathic’ thrombocytopenic purpura will eventually develop lupus. Unfortunately there are usually no clues at the onset to distinguish this 15% and very long-term follow-up may be needed.
Antibodies to Ro, La, Sm, RNP (ribonucleoprotein), and dsDNA may be detectable in the serum of patients who will develop lupus up to 10 years before the disease becomes clinically manifest, and rarely these antibodies are found in the healthy relatives of patients.
Occasionally lupus may present with isolated central nervous system (CNS) disease including convulsions, schizophrenic-like conditions, and neuropathies. As with many other clinical features, the most important consideration is to try to include systemic lupus in the differential diagnosis. The serological tests for it are widely available and will help to establish, quickly, the correct diagnosis.
The most commonly requested test to screen for SLE is the antinuclear antibody assay. A positive antinuclear antibody simply indicates that the patient’s blood contains antibodies that will bind to the nuclei of a sample of cells used in the test. The test is a sensitive one because over 95% of patients with SLE are antinuclear antibody positive. Although a small group of patients do seem to have persistently antinuclear antibody-negative SLE, the absence of antinuclear antibody in a patient with suspected lupus raises serious doubt about the diagnosis.
The specificity of the antinuclear antibody test for SLE is not high. The titre of antibody represents the highest dilution of the patient’s serum at which the test is still positive. Low-titre antinuclear antibody (1 in 10) is of little significance and may occur in healthy people. Higher titres (1 in 160 or more) are more worrying and are found in most patients with SLE and in patients with other autoimmune conditions including rheumatoid arthritis, systemic sclerosis, and Sjögren’s syndrome. However, some people with high-titre antinuclear antibody may be followed in rheumatology clinics for years without developing a frank autoimmune disease.
The finding of a positive antinuclear antibody in a patient with symptoms suggestive of SLE should lead to a series of other autoantibody tests. These are listed in Table 2 together with the identity of the target antigen and the approximate prevalence of the antibodies.
Anti-dsDNA antibody levels are particularly useful. This test is virtually specific for SLE (as is the anti-Sm antibody), especially if the immunoglobulins are of the IgG isotype. The anti-dsDNA result is usually quantified and this value is often a measure of the activity of the disease, but there is a group of patients who have persistently high anti-dsDNA antibody levels but no clinically active disease (serologically active, clinically quiescent). Long-term follow-up suggests that many but not all of these patients will eventually flare. In one study, trial patients were treated with high-dose corticosteroids on the basis of anti-dsDNA levels alone. In comparison with a control group treated only when symptoms or signs also suggested disease activity, the trial group had less disease activity overall and fewer flares. However, frequent large doses of corticosteroids resulted in significant side effects and a number of participants dropped out of this arm of the trial. Anti-dsDNA should be used only as an adjunct to the clinical impression of disease activity when deciding on a treatment regimen.
Anti-Ro and anti-La antibodies are linked to concurrent Sjögren’s syndrome. Mothers who have these antibodies have a higher incidence of neonatal lupus (see below) and should be advised about this before embarking on a pregnancy. Anti-Ro antibodies are also associated with photosensitivity.
There are no good antibody markers for the presence of disease of the CNS. Antibodies to ribosomal protein P were previously thought to have some value in the diagnosis of CNS lupus, but this has not been borne out by later results and the test is not available routinely in most laboratories. More recently, murine and clinical studies have suggested that antibodies to the N-Methyl-D-aspartate receptor (anti-NMDAR) may act on cerebral tissue to cause clinical features of CNS lupus, but this test is not used routinely in clinical practice.
APL antibodies can be recognized by one of two assays. The enzyme-linked immunosorbent assay for binding to cardiolipin distinguishes IgM and IgG isotypes. This is helpful because the level of IgG APL antibodies is a better predictor of clinical sequelae than that of IgM. APL antibodies can also be diagnosed by testing the clotting properties of the blood in vitro in the Russell’s viper venom test. An abnormal result in this assay is reported as showing the presence of a lupus anticoagulant. It is quite possible for the anticardiolipin test to be positive while the lupus anticoagulant assay is negative or vice versa. If either is positive, the patient may be at risk of manifestations of the APS. Anti-β2-glycoprotein 1 tests are now becoming available commercially, offering a third way of detecting APL antibodies.
Coombs’ test and assays for antithyroid antibodies are often requested in patients with SLE, particularly those with coexisting anaemia or hypothyroidism.
Measures of disease activity and end-organ damage
Blood and urine tests
The most reliable measures of highly active disease are depletion of complement and high anti-dsDNA levels. The erythrocyte sedimentation rate (ESR) also tends to be increased in active disease, unlike the level of C-reactive protein (CRP). The combination of high ESR and normal CRP in a patient with a multisystem disorder should raise the suspicion of SLE, leading to appropriate autoantibody tests as described above. The CRP may, however, be raised in the presence of infection, serositis, or arthritis.
Complement components C3 and C4 are the most commonly measured, and both tend to fall in active SLE. A persistently very low level of either C3 or C4 (or a high level of their degradation products C3d or C4d), regardless of immunosuppressive therapy, may signify the presence of a homozygous complement deficiency disorder. Although such disorders are very rare, it is important to diagnose them because they respond better to infusions of fresh frozen plasma than to immunosuppression.
It is important to measure creatinine and electrolyte values regularly and to check the urine for proteinuria and/or haematuria. These measures ensure that renal involvement is diagnosed early. It must be remembered that substantial deterioration in renal function may occur before serum creatinine rises beyond the normal range, an issue emphasized by the now routine reporting of eGFR. If the patient can reliably perform a 24-h urinary collection, then creatinine clearance can provide a more precise estimate of GFR, but this, or radio-isotopic methods of measuring GFR, are rarely required in routine clinical practice. Persistent proteinuria on dipstick testing should be quantified by measuring the albumin/protein:creatinine ratio in a spot urinary sample (or with a 24-h urinary collection).
Liver function tests are usually normal in SLE (abnormal in less than 10% of patients), but a baseline value should be measured, particularly in cases where potentially hepatotoxic drugs such as azathioprine may be used. Thyroid function abnormalities, particularly hypothyroidism, are well recognized to coexist with SLE.
A full blood count should be measured regularly. Falling haemoglobin, white cell count, and platelet counts may all occur (see under Haematological involvement above). Anaemia in the presence of a positive Coombs’ test may indicate haemolysis, which can be confirmed by requesting a blood film and serum haptoglobins.
Infections occur commonly in patients with SLE, particularly in those on high-dose immunosuppressants. Infection may not always be accompanied by high fever or leucocytosis, although CRP is usually raised. It is wise to carry out blood and urine cultures whenever even mild pyrexia is accompanied by a deterioration in health.
Plain radiographs are rarely useful in SLE. There is no characteristic appearance in the joints, and chest radiographs are unlikely to show abnormalities except in the presence of infection or effusion.
Requests for more specialized imaging studies should be directed by the clinical findings, e.g. the presence of dyspnoea and abnormal respiratory function tests often necessitates a CT scan of the thorax, which is the investigation of choice for diagnosis of pulmonary fibrosis. Echocardiography is useful if pericardial effusion, myocarditis, or endocarditis is suspected clinically. Bone density scanning is becoming increasingly important, since patients with SLE are often at risk of osteoporosis due to the use of corticosteroids and reduced capacity for physical exercise during young adult life.
Criteria for diagnosis
Although strictly speaking classification and diagnostic criteria are not synonymous, the classification criteria discussed earlier, and shown in Table 1, are widely used for diagnostic purposes. Some unease with these criteria has been expressed, notably why 4 of the 11 features are confined to one organ system (mucocutaneous), and whether anti-nucleosome antibodies should be allowed as an alternative to anti-dsDNA antibodies. The Systemic Lupus International Collaborating Clinics group is reassessing the currently used classification criteria.
SLE is a disease that still has the potential to kill. In many cases, however, the condition runs a fairly indolent course in which an initial flare is followed by many years of low-grade activity. General measures of value in the treatment of SLE are shown in Table 5.
In the pharmacological management of a patient with SLE, the clinician will typically seek to answer four ‘classic’ questions:
- Can the patient be managed without immunosuppression?
- If immunosuppression is needed, how should it best be started?
- If immunosuppression is being used, is the current level of immunosuppression inadequate or excessive? How should it be increased or reduced?
- Does the patient have any side effects from the drugs?
A fifth is increasingly being posed: would any of the new biologic treatments be of value?
|Table 5 Treatment of lupus—general measures|
|1 Rest as appropriate; try to avoid stress|
|3 Try to adhere to a low-fat diet and consider adding fish oil derivatives|
|4 Vaccination, for foreign travel etc., apart from ‘live’ vaccines in patients on immunosuppressives, is not contraindicated though the precise nature of the immune response differs from that in healthy individuals|
|5 Medium- or high-oestrogen contraceptive pills should be avoided— progesterone only or the lowest possible oestrogen pill (or other methods of contraception) are advised|
|6 The use of hormone replacement in the menopause remains controversial. Many patients do tolerate it without flaring, but not all|
Is immunosuppression required?
Patients whose disease activity is confined to arthralgia, tiredness, and/or mild rash do not usually have greatly raised ESR or anti-dsDNA antibodies or reduced complement. These patients can often be treated symptomatically, e.g. with agents such as paracetamol and diclofenac to control joint pain.
Where such symptoms are more severe, the antimalarial agent hydroxychloroquine at a starting dose of 400 mg/day may be useful. This drug has less potential for retinal toxicity than the closely related chloroquine and is therefore preferred in SLE. It is often possible to reduce the dose to 200 mg/day after 3 months and gradually withdraw it thereafter. Regular blood tests are not required to monitor the effects of hydroxychloroquine, but there is a very small risk of retinopathy such that review by an ophthalmologist every 6 to 12 months is considered advisable in many units. The drug has the useful ‘side effect’ of lowering lipid levels.
Where the main symptoms in a patient with SLE are those of the APS immunosuppression is rarely useful. Aspirin at a dose of 150 to 300 mg daily is recommended for those with mild symptoms of the disease or who have other risk factors for thrombosis. Patients who have had recurrent thromboses or cerebral infarcts and who have serum APL antibodies should usually be treated with lifelong anticoagulation. This is a major commitment for a young patient and raises particular problems in pregnancy (discussed below).
Some patients require a low maintenance dose of oral steroids to control their symptoms even though laboratory indices do not indicate high activity of disease. A dose of 5 to 7.5 mg daily is typically used in such cases. Topical steroids may be useful where lupus activity is confined to the skin.
Judging the dose of immunosuppression
Corticosteroids and cytotoxic agents are used to treat flares of disease. A mild flare of arthralgia, myalgia, and general fatigue may be alleviated by a single intramuscular dose of a corticosteroid preparation such as prednisolone acetate (usually 50 to 125 mg are given).
More severe flares of arthritis, pleuritis, or pericarditis require oral prednisolone at a dose of 20 to 40 mg daily. This usually leads to a rapid improvement in symptoms, and the dose of prednisolone can then be reduced by 5 mg every 1 to 2 weeks until it reaches 5 mg/day. It may not be possible to withdraw the drug completely for several months.
Alternatively, a shorter course of corticosteroids can be given intravenously. A typical course would consist of 500 to 750 mg methylprednisolone given over 3 to 4 h on each of 3 successive days. This requires admission to hospital, making it less convenient than oral therapy, and it is generally reserved for those patients who are not responding to oral prednisolone or cannot tolerate that drug in high doses.
Autoimmune haemolytic anaemia requires higher doses (60–80 mg/day) of oral prednisolone, with the dose reduced in 5- to 10-mg increments according to the clinical response. Azathioprine may be required as a steroid-sparing agent and is used at a dose of 2.5 to 3 mg/kg per day.
Renal flares of SLE require the most aggressive treatment, generally involving both corticosteroids and cyclophosphamide. A number of regimens have been used, with debate continuing as to which is optimal. An ‘older’ regimen of high-dose oral prednisolone and 750 mg intravenous cyclophosphamide monthly for 6 months, then 3 monthly for 2 years, has fallen into disrepute. Although reasonably effective, its side effects (especially infection and infertility) have led to the use of alternative regimens, including lower doses of intravenous cyclophosphamide (around 500 mg) for shorter periods of time and the use of oral mycophenolate mofetil (generally 2 to 3 g/day) instead of cyclophosphamide. Trial data suggest that mycophenolate is about as effective and has fewer serious side effects. In renal SLE it is critically important to control the patient’s blood pressure. Angiotensin converting enzyme (ACE) inhibitors, α-adrenergic antagonists such as doxazosin, and calcium channel blockers such as nifedipine are the agents most commonly used.
The treatment of CNS lupus varies depending on the manifestation of cerebral dysfunction. Mild cases may respond to relatively small doses of oral steroids (up to 30 mg/day). More florid manifestations such as convulsions or major psychosis require treatment with appropriate anticonvulsants or antipsychotic drugs, higher-dose oral steroids (60 to 80 mg/day), and sometimes azathioprine or intravenous pulses of cyclophosphamide in similar doses to those used in renal SLE.
Does the patient have drug side effects?
The side effects of corticosteroids are well known. The most common early problems are weight gain, hirsutism, easy bruising, and insomnia. It is difficult to prevent them, except by using the lowest dose of steroid that is effective and reducing it as rapidly as possible while maintaining control of the disease.
Longer-term sequelae of corticosteroid use include increased susceptibility to infection, osteoporosis, avascular necrosis, and diabetes mellitus. The most rapid loss of bone in steroid-induced osteoporosis occurs within the first year of treatment, although doses of 7.5 mg/day or less of prednisolone are thought to have little effect on bone. At higher doses, it may be advisable to carry out a bone density scan and to give either calcium and vitamin D tablets or a bisphosphonate (either etidronate or alendronate is commonly used) as prophylaxis.
Cyclophosphamide causes alopecia, nausea, bladder toxicity, and gonadal dysfunction that may lead to infertility. The problem of infertility becomes more likely with increasing age, women over 30 given cyclophosphamide being at particular risk. Again, the best way to prevent such problems is to use as small a cumulative dose of the drug as is feasible. Bone marrow suppression may occur. During a programme of cyclophosphamide pulses the white blood cell count falls to a nadir 10 days after each pulse and should be measured at that time to decide whether the next pulse can be given safely. Nausea and vomiting during pulses may be so severe that antiemetics such as metoclopramide or granisetron are necessary.
Azathioprine also causes bone marrow suppression and can cause abnormalities of liver enzymes, which resolve once the drug is withdrawn.
We are entering an exciting era in which improved understanding of the causes of lupus is leading to newer therapeutic approaches. In particular B-cell depletion achieved by using 1 g on two occasions 2 weeks apart of the anti-CD20 chimaeric reagent rituximab (or in more severe cases using this with intravenous cyclophosphamide 750 mg twice and intravenous methylprednisolone 100 to 250 mg twice) looks very promising. The use of fully humanized anti-CD20 monoclonal antibodies is also being explored.
Recently double-blind controlled trials using an anti-CD22 reagent (epratuzumab) and a monoclonal antibody (belimumab) directed against the B cell activating factor BLyS have been reported to meet their end points. Other major trials of antibodies blocking other B cell activating factors (e.g. atacicept) and interferon alpha are ongoing but studies of a B cell toleragen (LJP-394) and CTLA4-Ig (which interferes with the stimulation of T cells by antigen-presenting cells) have failed to prove benefit so far.
SLE in pregnancy
Mothers often ask whether their children are likely to inherit SLE. Inheritance of the adult form of the disease is very rare (approximately 1% of all cases), although a transient illness termed ‘neonatal lupus’ can occur. The characteristics of this condition are rash, hepatitis, anaemia, and thrombocytopenia, which usually resolve by 8 months after birth, and inflammation of the cardiac conducting tissues that may lead to heart block in the fetus. The cardiac problem may be diagnosed by ultrasound scans of the fetal heart between 16 and 24 weeks’ gestation. Treatment of the mother with 4 mg oral dexamethasone/day may prevent progression from incomplete to complete fetal heart block. If complete heart block occurs, the neonate may require a cardiac pacemaker. Interestingly, children born with neonatal lupus sometimes develop heart block later in life, with one reported case of this problem occurring at the age of 35.
SLE itself does not usually reduce the ability to conceive, although as described the drugs used to treat it, notably cyclophosphamide, may induce infertility due to gonadal failure. There is an increased risk of spontaneous abortion, particularly in the presence of high-titre APL antibodies. Pregnant mothers with a high APL antibody level and a history of previous miscarriage should be considered for anticoagulation until the birth of the baby. As warfarin is potentially teratogenic, heparin may be used from the second trimester until parturition.
The presence of maternal anti-Ro and anti-La antibodies predicts a higher risk of neonatal lupus. Where both are present the risk is approximately 2 to 5%. It is believed that the antibodies cross the placenta and bind to some component of the fetal cardiac tissue. Ro itself is expressed on the surface of cardiac myocytes at some point in fetal development and may be an important target for the antibodies. This may be why the fetal heart, but never the mother’s heart, is affected by anti-Ro antibodies.
Although overall the risk of a flare during pregnancy is probably no greater than at other times, SLE may be exacerbated during the pregnancy. Corticosteroids may be used in moderate doses without affecting the fetus, but higher doses (over 30 mg) given for long periods can potentially cause fetal adrenal suppression. If lupus activity is such that these doses are required, the risk to the fetus of not treating the disease adequately should outweigh any risk from the drug.
Cyclophosphamide and methotrexate are contraindicated in pregnancy, although there have been many successful pregnancies in transplant recipients taking azathioprine without obvious increased risk of adverse effect. Use of hydroxychloroquine is not recommended by the manufacturers, but there is little evidence that it has adverse effects.
It may be difficult to distinguish pre-eclampsia from a flare of renal lupus. Both can cause hypertension and proteinuria, but in pre-eclampsia—unlike SLE—there are rarely urinary casts and levels of anti-dsDNA antibodies and complement are normal.
Occupational and psychological aspects of SLE
In making the diagnosis of SLE, therefore, the doctor must consider the effect of this condition on the overall life of the patient as well as his or her individual organs. A sympathetic understanding of the anxieties associated with the diagnosis is vital. SLE typically presents in young people, especially women. The onset of a chronic, essentially incurable condition at a time of life when the patient is otherwise healthy and has many plans and responsibilities is an unexpected and unwelcome burden. Many concerns arise; in particular the outlook for fertility and the ability to care for children are substantial worries. In those cases where the use of high-dose corticosteroids and immunosuppressive agents is essential, detailed explanations of the benefits and risks of these treatments in both the short and the long term are necessary. Although a 10-year survival rate of 90% may appear reassuring, it is probably less so to a 25 year old who recognizes a 10% chance of dying by the age of 35.
Morbidity from systemic lupus can be considerable. From the less serious, but troubling, severe fatigue to the necessity for renal dialysis, many aspects of lupus result in it having a big effect on quality of life. Analyses using the medical outcome survey, short form 36 (SF-36) quality-of-life index have shown that patients with lupus have impaired scores in every aspect of this index. Mortality from SLE has fallen significantly over the last half century. Whereas SLE was reported to have a 50% 5-year survival rate in the 1950s, 10-year survival rates rose to between 80 and 90% by the 1970s. Since then, survival rates have improved a little, but deaths from renal failure have become less common, while those from infection and cardiovascular disease have increased. Infection is generally associated with immunosuppressive therapy, highlighting the need for better and more accurately targeted methods of treating the underlying immunological abnormalities in this disease.
A damage index for lupus, derived by the Systemic Lupus International Collaborating Clinics (SLICC) group, has shown that within 10 years about two-thirds of patients have acquired permanent problems. Furthermore, early acquisition of damage—within 1 year of diagnosis—substantially increases the risk of mortality (fourfold) within 10 years.
Controversial areas and future prospects
Plasma exchange and intravenous immunoglobulin therapy have been tried in SLE, particularly in renal crises. Overall, the results do not suggest that either form of treatment should be used routinely. New drugs such as rituximab and anti-IL-10 have been administered to relatively small numbers of patients. Some encouraging results have been reported, but it is too early to decide on the place of these agents in the management of the disease. It will be particularly interesting to determine whether biologic agents can be used sequentially, or even simultaneously.We do not yet have a cure for SLE, or even a method of controlling the disease without the risk of significant side effects. The main sources of controversy concern attempts to develop new forms of treatment and to establish indices of disease activity that can be used to measure the effects of these treatments.
There are now many different murine models of SLE. These differ in their clinical and serological characteristics and each represents at best a partial approximation to the human disease. This is important, because it is now possible to administer agents such as monoclonal anticytokine antibodies to these mice and to assess the effect on the disease process, but how far such studies can be used to predict which of these agents might be effective in humans remains unknown.
If new drugs or monoclonal antibodies are to be used in human SLE, it is necessary, given that mortality is now (thankfully) an uncommon end-point, to have recognized ‘tools by which to judge the response to treatment’. A combination of a disease activity index, a damage index, a patient health perception index, a record of toxicity, and cost is required. Several global score disease activity indices, e.g. the systemic lupus activity measure (SLAM) and the European Community lupus activity measure (ECLAM) have been developed and provide a ‘rough and ready’ guide to activity. A more sophisticated approach based on the ‘physician’s intention to treat principle’ has been derived by the British Isles Lupus Activity Group, providing an ‘at a glance’ review of activity in eight different organs or systems. A single damage index (the SLICC/ACR damage index) has been developed, which records a wide variety of potential permanent (present for at least 6 months) changes (e.g. avascular necrosis, myocardial infarction) that can occur in patients with lupus as part of the development of the disease. The medical outcome survey SF-36 provides a useful health perception index for patients with lupus. Although not designed specifically for this condition, it has been widely used in a number of ongoing drug trials, but it is likely that a more specific lupus quality-of-life index will be more widely used in future (three are being validated at present).
It is likely that the treatment of SLE in 10 years’ time will be different from that given now. Basic science research is starting to identify the various strands of immune dysfunction at the core of this disease. At the same time, drug development is providing agents that are capable of selectively targeting single cell types or cytokines within the immune system. At least some of these agents are likely to be relevant to the dysfunctional mechanisms in SLE. In addition, clinicians are becoming more aware that conditions such as atherosclerosis and osteoporosis are common in patients with SLE. By increasing efforts to detect and control these associated conditions, as well as seeking to attack the underlying autoimmune disease, it should be possible to improve the lives of patients with SLE, even if a cure for the disease remains a distant prospect.
Read more: SLE: diagnosis and management
Arbuckle MR, et al. (2003). Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med, 349, 1526–33.
Boumpas DT, et al. (1992). Controlled trial of methyl prednisolone versus two regimens of pulse cyclophosphamide in severe lupus nephritis. Lancet, 340, 741–5.
Casciola-Rosen LA, Anhalt G, Rosen A (1994). Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. J Exp Med, 179, 1317–30.
Cervera R, et al. (1993). Systemic lupus erythematosus—clinical and immunologic patterns of disease expression in a cohort of 1000 patients. Medicine (Baltimore), 72, 113–21.
Chan TM, et al. (2005). Long-term study of mycophenolate mofetil as continuous induction and maintenance treatment for diffuse proliferative nephritis. J Am Soc Nephrol, 16, 1076–82.
Contreras G, et al. (2005). Sequential therapies for proliferative lupus nephritis. N Engl J Med, 350, 971–8.[CrossRef] [Web of Science]
Datta SK, et al. (2005). T-helper cell intrinsic defects in lupus that break peripheral tolerance to nuclear antigens. J Mol Med, 83, 267–78.
Goldblatt F, et al. (2005). New therapies for systemic lupus erythematosus. Clin Exp Immunol, 140, 205–12.
Hahn BH (2003). Systemic lupus erythematosus and accelerated atherosclerosis. N Engl J Med, 349, 2379–80.
Isenberg DA, et al. (2006). Systemic lupus erythematosus—2005 annus mirabilis. Nature Clin Pract Rheum, 2, 145–52.[CrossRef]
Isenberg DA, et al. (2007). 50 years of anti-dsDNA antibodies – are we approaching journey’s end? Rheumatology, 46, 1052–6.
Johnson AE, et al. (1995). The prevalence and incidence of systemic lupus erythematosus in Birmingham, England: relationship to ethnicity and country of birth. Arthritis Rheum, 38, 551–8. [Web of Science]
Kaliyaperumal A, et al. (2002) Naturally processed chromatin peptides reveal a major autoepitope that primes pathogenic T and B cells of lupus. J Immunol, 168, 2530–7.
Khamashta MA, et al. (1995). The management of thrombosis in the antiphospholipid antibody syndrome. N Engl J Med, 332, 993–7.
Koffler D, Schur PH, Kunkel HG (1967). Immunological studies concerning the nephritis of systemic lupus erythematosus. J Exp Med, 126: 607–24.
Lu TY, Ng KP (2009). Cambridge G, Leandro MJ, Edwards JC, Ehrenstein MR, Isenberg DA. A retrospective seven-year analysis of the use of B cell depletion therapy in systemic lupus erythematosus at University College London Hospital: the first fifty patients. Arthritis Rheum, 61, 482–7.
Manzi S, et al. (1997) Age-specific incidence rates of myocardial infarction and angina in women with systemic lupus erythematosus: comparison with the Framingham study. Am J Epidemiol, 145, 408–15.
Okamura M, et al. (1993). Significance of enzyme linked immunosorbent assay (ELISA) for antibodies to double stranded and single stranded DNA in patients with lupus nephritis: correlation with severity of renal histology. Ann Rheum Dis, 52, 14–20.
Rahman A (2004) Autoantibodies, lupus and the science of sabotage. Rheumatology, 43, 1326–36.
Rahman A and Isenberg DA. (2008). Systemic Lupus Erythematosus. New England Journal of Medicine, 358, 929–39.
Tan EM, et al. (1982). The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum, 25, 1271–7. [Web of Science]
Walport MJ (1993). Inherited complement deficiency—clues to the physiological activity of complement in vivo. Q J Med, 86, 355–8.