SLE features various combinations of facial skin rash, arthritis, nephritis, systemic symptoms such as fever and malaise, and inflammation of serous membranes such as the pericardium. The disease is by far more frequent in women, predominantly young or middle-aged adults. In one large series the mean age at diagnosis was 30 years, with a skewed distribution: the majority between ages 14 and 40 years, a lesser but significant number between ages 40 and 55 years, and a few scattered from infancy to age 14 years and over age 55 years. Hepatomegaly is found in 30% of patients (literature range, 23%-44%), splenomegaly in 20% (9%-41%), adenopathy in 50% (37%-68%), and arthritis in 60% (52%-78%).

Laboratory findings. Anemia is present in 60%-75% of SLE patients, most often mild or moderate and of the normocytic-normochromic type. About 30% (literature range, 18%-65%) of patients have a positive direct Coombs’ test result, although only 10% or less develop autoimmune hemolytic anemia. Leukopenia is found in about 50% of patients (literature range, 43%-65%) and thrombocytopenia in about 15% (literature range, 5%-26%). There are often one or more manifestations of abnormal plasma proteins; these may include cold-precipitable cryoglobulins, circulating anticoagulants, autoantibodies, elevated gamma-globulin levels, circulating immune complexes, false positive RA and syphilis serologic reactions, and certain even rarer phenomena.

In SLE with active nephritis, total serum complement (C, Cў, or CH50, Chapter 22), C3, and C4 levels are usually all decreased. This complement pattern is not absolutely specific for lupus nephritis and may be found in serum sickness, SBE, and immune complex disease.

Lupus erythematosus cell preparation. The first reasonably good test for lupus, and one that could still be used, is the LE preparation (LE prep). This test detects antibody against nuclear deoxyribonucleoprotein (DNP; soluble nucleoprotein[sNP], DNA-histone complex; see the box) that is produced in SLE. The LE prep technique first provides a source of nuclei from laboratory-damaged cells, either tissue cells or WBCs. The nuclei are then incubated with patient serum, during which time the nuclear material is converted by the ANA against sNP into a homogeneous amorphous mass that stains basophilic with Wright’s stain. This mass is then phagocytized by nondamaged polymorphonuclear neutrophils; these neutrophils containing the hematoxylin (blue-staining) bodies in their cytoplasm are the so-called LE cells.

Some Autoantibodies Found in Systemic Lupus Erythematosus and Other Autoimmune Disorders
I. Anti-DNA
A. Double-stranded (“native”) DNA (dsDNA)
B. Single-stranded DNA (ssDNA)
II. Antinucleoprotein (soluble nucleoprotein [sNP]; DNA-histone)
III. Antibasic nuclear proteins (histones)
IV. Antiacidic nuclear proteins (extractable nuclear antigens[ENAs]: includes Smith[Sm] and ribonucleoprotein [RNP])
A. Nuclear glycoprotein (Sm)
B. Ribonucleoprotein (RNP)
C. Sjцgren’s syndrome A and B (SS-A[Ro] and SS-B)
V. Antinucleolar (nuclear RNA)
VI. Anticytoplasmic (ribosomal RNA and others)

Artifacts may be confused with true LE cells. To be definitive, the basophilic hematoxylin body must be completely amorphous, without any remaining nuclear structure whatever. In many LE preps one finds neutrophils or monocytes with phagocytized nuclear material that still retains some identity as a nucleus, such as a residual chromatin pattern. These are not true LE cells; they are called “Tart cells.” Increased numbers of these may be seen in SLE, but they do not have any diagnostic significance. Neutrophils with ingested RBCs can also be misdiagnosed as LE cells by inexperienced persons. Adrenocorticosteroid therapy often suppresses LE cell production.

In SLE, positive LE prep results were reported in 70%-80% of cases, with a range in the literature of 40%-100%. The great differences in the literature were due to several factors. First, many of the studies report data based on LE preps obtained at various times during the entire illness rather than the result obtained on the first test at the first admission. Cumulative or repeated studies would result in a higher percentage of patients with positive results. Various modifications of the basic LE prep technique were widely used, and the modifications differ in sensitivity. Finally, patients with active disease were more likely to have a positive LE prep than those with inactive disease. A realistic figure for likelihood of a positive LE prep result in a person with SLE on initial workup is probably 60%-70%.

Because LE prep methodology is not standardized, preparing and examining the slides is time-consuming, and many laboratory workers lack expertise in interpreting the slides, the LE prep has been mostly replaced by Crithidia anti-DNA assay.

Antinuclear antibody (ANA) test. The antinuclear factor that causes the LE cell phenomenon is not the only autoantibody or “factor” demonstrable in SLE. A wide variety of such factors has been demonstrated, reactive against either nuclear or cytoplasmic constituents with varying degrees of tissue and cellular constituent specificity. An ANA test usually involves incubating patient serum with nucleated cells. Afterward, a tagged antibody against gamma globulin is added to detect patient ANA coating the cell nuclei. The antibody tag can be fluorescent (FANA test) or can be an enzyme (EANA test). Various types of tissue cells have been used as sources of cell nuclei. Most of these tests yield initial-visit positive results in 95% or more of patients with SLE, compared with positive results on the LE prep of 60%-70%. However, the ANA test is more frequently abnormal than the LE prep in conditions other than SLE. Therefore, current practice is to employ the ANA procedure as a screening test. If the ANA test result is negative, chances that the patient has SLE are very low. If the ANA test result is positive, other studies must be done to obtain a definitive diagnosis.

Like all laboratory tests, the ANA test is not perfect. Both tissue culture cells (tumor-derived Helen Lake [HeLa] cells, human epithelial cell-derived cells [HEp-2 cells, originally obtained from a laryngeal tumor], and human amniotic cells) and thin tissue sections derived from various sources (most commonly rat liver or mouse kidney) have been used to provide cell nuclei for the procedure; the different nuclear sources do not all have equal sensitivity. Most of the early work was done with rat liver or kidney tissue sections. Tissue culture cells generally are reported to be more sensitive than rat tissue. On the other hand, the more sensitive tissue culture cells detect more low-titer FANA reactions, of which some are produced by disorders other than SLE (if screening for SLE is the only consideration, these would be considered false positive reactions. However, they may be a clue to the other diseases). Also, tissue culture cells demonstrate chromosome centromere staining in patients with scleroderma (progressive systemic sclerosis, or PSS), whereas discrete centromere structures are very difficult to see on rat liver sections and instead tend to have a nonspecific speckled appearance. Similarly, rat liver sections will usually not detect ANA against the Sjцgren’s syndrome. A, (SS-A or Ro) antigen (found most frequently in Sjцgren’s syndrome but also in some patients with SLE and other connective tissue disorders). The anti-human globulin used to detect ANAs can be either a polyclonal type or specific for IgG. Specific anti-IgG eliminates a certain number of false positive reactions due to RF or other IgM antibodies. However, it would fail to detect IgM antibodies against histones due to some causes of drug-induced SLE and some other collagen diseases. In addition, about 4% of SLE patients are reported to produce IgM-type ANA. Results of procedures that use acetone-fixed cell substrates are more likely to be positive with ANA against saline-extractable antigens (extractable nuclear antigens, or ENAs) such as Smith (Sm) and ribonucleoprotein (RNP) than those that omit substrate fixation.

About 5% of patients with SLE have negative FANA test results using rat liver or mouse kidney tissue sections. About two thirds of these give a speckled nuclear reaction using HEp-2 tissue culture cells; the majority of these were found to be SS-A (Ro) antibodies, which are most often found in Sjцgren’s syndrome but also can be found in SLE. When they are associated with SLE, the predominant clinical symptom seems to be a photosensitive skin rash. In addition, there appears to be a very small subset of SLE patients with anticytoplasmic antibodies rather than ANAs.

Antinuclear antibody test patterns. The method of reporting ANA results deserves some comment. Both the titer of positive results and the pattern (distribution) of nuclear fluorescence are important. In general, the higher the titer of certain ANA patterns known to be associated with SLE, the more likelihood that the patient has SLE. In addition, the ANA staining pattern can sometimes provide useful information. There are several recognized patterns that appear to be produced by ANAs against certain nuclear proteins. It must be emphasized that none of the fluorescent patterns is specific for only one autoantibody or exclusive for the diseases with which it is traditionally associated. Also, some investigators disagree with the findings of others regarding which pattern is found more often in various diseases.

Rim (peripheral) pattern. Fluorescence is much more intense on the nuclear border. This was originally thought to be specific for antibody against dsDNA. However, it is now known to be produced by ANA against several additional nucleoproteins, including sNP, single-stranded DNA (ssDNA), and histones. However, a high ANA titer (1:160 or greater) with a rim pattern strongly suggests SLE. Drug-induced SLE usually does not have a rim pattern, nor is a rim pattern common in other rheumatoid-collagen diseases (present in <10% of those patients). However, when the ANA titer is borderline or only mildly elevated, the rim pattern is less helpful because of overlap with the other diseases.

Solid (homogeneous) pattern. Fluorescence is uniform throughout the nucleus. This is the second most common ANA pattern and, like the rim pattern, can be produced by ANA against dsDNA, ssDNA, sNP, and histones. This is the most frequently seen pattern in SLE, but it is less suggestive of SLE, since it is found more frequently than the rim pattern in other rheumatoid-collagen diseases and is also seen in some patients with drug-induced ANA. The ANA titer is usually less than 1:160 in diseases other than SLE, so that the possibility of SLE is increased in patients with higher titers.

Speckled pattern. There are many small fluorescent dots throughout the nucleus that do not involve the nucleoli. This is the most commonly encountered ANA pattern. ANAs responsible include those against acidic nuclear proteins, such as ENA (Sm and RNP), Sjцgren’s syndrome A (SS-A; Ro), SS-B (La), Jo-1, histones, and scleroderma 70 (Scl-70). This pattern is the one most frequently associated with the mixed connective tissue (MCT) diseases (discussed later), although the MCT disease does not produce the majority of speckled ANA reactions. It has also been reported in about 25% of patients with SLE (due to anti-Sm) and in some patients with RA, progressive systemic sclerosis, Sjцgren’s syndrome, drug-induced ANAs, and some aged persons. However, if the ANA titer is very high with a speckled pattern, this suggests MCT disease. At least one report indicates that many fluorescence patterns reported as speckled actually consist of small rods or filaments with dots, which is not a true speckled configuration. This pseudospeckled pattern is said to represent most of the conditions other than MCT disease that others include in the speckled category. If the speckled pattern is present in moderately elevated or high titer, specific antibody testing is useful since specific ANA against Sm is strongly suggestive of SLE, ANA against RNP suggests MCT disease, ANA against SS-B suggests Sjцgren’s syndrome, and ANA against Scl-70 suggests scleroderma (PSS).

Nucleolar pattern. Fluorescence only in the nucleolar areas of the nucleus is seen as several small areas of irregular shape and different sizes. This is due to ANA against nucleolar ribonucleic acid (RNA; 4-6s RNA). It suggests PSS (55%-90% of PSS patients), especially if the titer is high. Lower titers may be found in SLE and occasionally in other collagen diseases.

Centromere pattern. The centromere pattern is made up of moderate-sized speckles that vary in size and are fewer in number than expected with a speckled pattern. Chromosomes of mitotic cells show the same speckles along the chromosomal spindles. This pattern is due to ANA against centromere antigen and is highly suggestive of the CREST (calcinosis, Raynaud’s phenomenon, esophageal motility dysfunction, sclerodactyly, and telangiectasia) variant of PSS.

ANA tests can be reported in several ways. The most common are by titer or some descriptive semiquantitation of the reaction strength. In general a test result positive only at the screening titer is considered a weak positive (for HEp-2 cells, this is usually 1:40; for kidney tissue, 1:20). When positive at one dilution above screening titer, it is called moderately positive; when positive two dilutions above screening titer, it is considered a strong positive reaction (3+).

Assays for specific antinuclear antibodies. At present, with the possible exception of the anti-DNA assay using the Crithidia method, the various assays for specific ANAs are obtained mostly in large reference laboratories or laboratories specializing in autoimmune disease. It should be emphasized that correlation of specific autoantibodies with specific diseases usually depends on the presence of substantial titer elevations of the antibody; low titers are more likely to be less specific.

Anti-DNA assay. Assays for various specific autoantibodies have been developed. Currently the most popular of these assays detects antibody against dsDNA. This is usually referred to as anti-DNA, although there is a different antibody against ssDNA. There are currently three well-recognized techniques for measurement of anti-DNA antibody. The Farr method is the oldest; it is a radioimmunoassay (RIA) technique using radioactive tritium (3H) or carbon 14 that depends on ammonium sulfate precipitation or Millipore filter filtration to separate the antigen-antibody complex from nonbound antigen. At least two commercial companies have marketed kits using more traditional radioactive iodine reagents. Assays have also been developed that use a protozoan organism named Crithidia luciliae; this has a large intracytoplasmic structure called a “kinetoplast” composed mostly of dsDNA. Various commercial companies have kits with the dead organisms on prepared slides. Patient serum is added, and the anti-DNA antibodies attach to the kinetoplasts of the organisms. A fluorescent anti-human globulin antibody is added that demonstrates fluorescence of the kinetoplast if anti-DNA antibody is attached (the technique is basically similar to that used for the ANA test). In general, reports indicate that the Farr technique is a small but significant percent more sensitive than commercial RIA kits. There is some disagreement in the literature as to whether the sensitivity of the Crithidia procedure is equal to or less than the RIA procedures. Some of the disagreement is attributable to comparison of the Crithidia kits to the Farr technique versus comparison to a commercial RIA method.

There have been claims in the literature that the anti-DNA test is highly sensitive and specific for SLE. However, review of the published evaluations discloses that overall sensitivity and specificity (as a composite of all anti-DNA methods) is about the same as those of an optimally performed LE prep. Data from the Crithidia assay, which is now the most widely used procedure, suggest that overall sensitivity of Crithidia is about the same as the LE prep, but that specificity for SLE is better than the LE prep, with somewhat fewer positive results in diseases other than SLE. However, positive results do occur in diseases other than SLE. Some data suggest that anti-DNA assay results by any method are more likely (by 20%-40% of cases) to be positive in patients with active SLE than in patients with inactive SLE. Also, SLE is more likely than other conditions to produce a high antibody titer. Test results of drug-induced ANA are usually negative, and if positive are usually not greatly elevated.

Anti-sNP antibody reacts against a DNP complex with histone. Antibody to the sNP complex is reported in about 50% of SLE patients, fewer than 10% of patients with RA, Sjцgren’s syndrome, or MCT disease, and uncommonly in patients with scleroderma, dermatomyositis, polyarteritis nodosa, or discoid lupus. When anti-sNP antibody is present, the ANA test shows a solid (homogeneous) ANA pattern. Anti-sNP antibody is thought to produce the LE cell phenomenon. Since anti-sNP antibody is not the only autoantibody to evoke a solid ANA pattern, data on the incidence of sNP in various diseases and on the incidence of the solid ANA pattern in the same diseases may not correlate exactly.

Anti-ENA is not really a specific antibody but instead is a group of antibodies to certain antigens that consist of RNA and protein; of these, the most important are Sm and RNP (discussed separately in the next paragraphs). The ENA antigen is extracted from thymus using phosphate-buffered saline and therefore, as previously mentioned, is sometimes referred to as saline-extractable antigen. An order for anti-ENA antibodies would generate assays for anti-Sm and anti-RNP. These two ANAs both produce a speckled ANA pattern.

Anti-nuclear glycoprotein Smith (Sm) antibody is present in 30% of patients with SLE (range, 20%-40%) and 8% of patients with MCT disease but not in patients with most of the other rheumatoid-collagen diseases. Therefore, anti-Sm antibody is fairly specific for SLE, but the sensitivity of the test is poor. Anti-Sm antibody is associated with the speckled ANA pattern.

Anti-RNP antibody is reported in nearly 100% of patients with the mixed connective tissue syndrome and about 25% of patients with SLE, discoid lupus, and progressive systemic sclerosis (scleroderma). Anti-RNP antibody (like anti-Sm antibody) is associated with a speckled ANA pattern. In high titer, anti-RNP antibody is suggestive of MCT disease.

Antinucleolar antibody is reported in about 55% of patients with PSS, 25% of patients with SLE, and 10% of patients with RA. In high titer it is very suggestive of PSS.

Anticentromere antibody is directed against the centromere area of chromosomes and is reported to be suggestive of the CREST syndrome. This antigen is seen when FANA is performed on tissue culture cells but not with rat liver or kidney.

Anti-SS-A (Ro) and Anti-SS-B (La) antibodies react against nuclear antigens extracted from human B lymphocytes (Wi12 cell line) grown in tissue culture. SS-A (Ro) is found in the cytoplasm of some tissue cells (e.g., human spleen) and in the nucleus of others (HEp-2 cells). It is difficult to detect in FANA using rat liver or kidney cells, but is visible using HEp-2 cells. The SS-A is found in about 70% of patients with Sjцgren’s syndrome without RA symptoms, 10% of Sjцgren’s patients with RA, and 25% of patients with SLE (including about two thirds of SLE with rim-reactive ANA tests and over 90% of patients with neonatal SLE) but usually not in patients with RA. The SS-B is found in about 50% of patients with Sjцgren’s syndrome without RA, less than 5% of patients with Sjцgren’s syndrome with RA, but not in patients with SLE or RA. Therefore, high titers of the SS-A or SS-B anti-Sjцgren antibodies are fairly good markers for Sjцgren’s syndrome, although sensitivity is not good.

Rheumatoid arthritis precipitin (RAP) is an autoantibody against RA nuclear antigen (RANA) derived from reaction of Epstein-Barr virus with human lymphocytes in tissue culture. RA precipitin is present in about 65% of patients with RA, 70% of patients with Sjцgren’s syndrome and RA, but in less than 10% of patients with SLE or other rheumatoid-collagen diseases. High titers of the RAP antibody are useful markers for RA when RA symptoms are present in other collagen diseases. The SS and RAP antibodies may be present in patients who have a negative ANA test result.

Anti-scleroderma-associated antigen (Scl-70) antibody. Scl-70 autoantibody is found in about 45% of patients with PSS.

Anti-Jo-1 antibody. This autoantibody is fairly specific for myositis and is found in about 25% of myositis syndromes (polymyositis, dermatomyositis) as a group and 35%-45% of polymyositis cases.

Anticytoplasmic autoantibodies. These include antimitochondrial antibodies (found in primary biliary cirrhosis), neutrophil anticytoplasmic antibodies (found in certain types of vasculitis, such as Wegener’s), antimicrosomal antibodies (not detected by HEp-2 cells; associated with chronic active hepatitis), and antiribosomal antibodies (may be present in SLE).

Skin biopsy with immunofluorescence staining. It is possible to obtain a full-thickness biopsy specimen from an uninvolved area of epidermis next to a skin lesion, prepare frozen section from the biopsy tissue, and apply immunofluorescent stains containing antisera against IgG, IgM, and IgA. Several characteristic immunofluorescent patterns have been described in certain diseases. In SLE, there is either a solid or stippled band involving the epidermis basement membrane area using IgG antiserum. This pattern is said to be present in up to 90% of patients with SLE or discoid LE if the disease is active. The incidence is considerably less if the disease is not active. Other diseases associated with characteristic skin immunofluorescent patterns are pemphigus vulgaris (intraepidermal fluorescence of spaces between squamous epithelial cells using IgG), bullous pemphigoid (“tubular” or linear solid staining of the basement membrane using IgG), and dermatitis herpetiformis (short irregular speckled band at the basement membrane confined to the tips of the dermal papillae, using IgA antiserum).

Slide latex tests for systemic lupus erythematosus. Commercial companies have marketed 2-minute slide latex agglutination tests for SLE based on latex particles coated with DNA (most often, thymus-derived DNA). The tests available so far have substantial differences in sensitivity but in general yield fewer positive results in SLE than the LE preparation or anti-DNA assay by the Crithidia method. Some of the tests do not have adequate clinical evaluations; in some cases, I was not able to obtain any published evaluations.

Serum complement in systemic lupus erythematosus. Total serum complement (C, Cў, CH50; Chapter 22), as well as the complement fractions C3 and C4, are often reduced in SLE patients with lupus nephritis. In one study, 38% of SLE patients had decreased C3 levels on initial workup and 66% had decreased C3 levels at some time during that illness. One report suggests that the combination of increased anti-dsDNA levels plus decreased C3 levels is highly specific for SLE; this combination was found in 32% of SLE patients on initial workup and 61% at some time during their illness.

Drug-induced systemic lupus erythematosus. A syndrome very similar to SLE can be produced by certain medications (drug-induced SLE). The three drugs considered proven SLE-inducing agents are procainamide, hydralazine (Apresoline), and isoniazid. Patients on procainamide develop clinical SLE in 20%-35% of cases. 50%-85% of patients on more than 1.5 gm of procainamide/day have positive LE preps; 1%, anti-dsDNA; and 75%, detectable ANAs. Of patients taking hydralazine, 2%-21% develop SLE and about 24%-50% have positive ANA test results. A small number (<1%) of patients taking isoniazid develop the SLE syndrome, and about 20% have detectable ANAs. Certain other drugs (e.g., methyldopa, phenytoin, quinidine, and chlorpromazine) have, on occasion, been reported to induce positive LE prep results or ANA test results, but are usually not associated with the SLE syndrome. Drug-induced SLE produces ANAs that are most often directed against histones, whereas spontaneous SLE produces ANAs most often directed against dsDNA. Although both spontaneous SLE and drug-induced SLE may produce a solid (homogeneous) ANA test pattern, spontaneous SLE often produces a rim (peripheral) pattern not seen with the drug-induced syndrome, whereas drug-induced ANA results frequently show a speckled pattern, which is uncommon in spontaneous SLE. Assay for anti-dsDNA (usually ordered as “anti-DNA”) is helpful to differentiate SLE and drug-induced ANA or SLE symptoms. High titers of anti-DNA are suggestive of SLE, whereas anti-dsDNA in drug-induced positive ANA is either negative or only minimally elevated.