LUPUS: A GP Guide to Diagnosis
This chapter considers how different tests may be used to make a diagnosis of lupus and how these tests may be useful in the monitoring of the disease. While lupus may have been historically regarded as a difficult disease to diagnose, mimicking as it does many other medical conditions, it is now true to say that the use of specific diagnostic tests is making life easier both for clinicians who suspect the disease and for those who are involved in the day to day care and monitoring of affected patients.
The performance of simple blood and urine tests may often provide very useful clues which make the clinician suspect a diagnosis of lupus. Such a diagnosis is clearly based on the combination of the clinical picture with which a patient presents and the results of much more specific tests, but a full blood count, ESR, CRP, biochemical profile tests (including renal and liver function tests) and a simple dipstick test of the urine can provide an enormous amount of useful information.
A patient with lupus may have a normal blood count when their disease is inactive but the classical features of active, or untreated, disease are: anaemia, leukopenia and thrombocytopenia. There are many reasons why a patient with lupus may be anaemic. The most common picture is of a normochromic normocytic anaemia but, as in any condition affecting the joints, the use of non-steroidal anti-inflammatory drugs, provided by a physician or as a result of self medication, may be associated with an iron deficiency picture due to gastro-intestinal blood loss. A haemolytic anaemia is common in lupus and if the bone marrow is compensating well for continued haemolysis there may be an increased reticulocyte count in the peripheral blood, with a resultant rise in the MCV. A high MCV may also occur in patients receiving drugs such as azathioprine, discussed further below.
Leukopenia is common in lupus and the classical picture is one of a relative lymphopenia. Many patients with lupus are of African or Afro-Caribbean origin and in these patient groups a leukopenia (sometimes called an "ethnic" leukopenia) is quite common. Care should be taken not to over-interpret modest reductions in white blood cell count in the absence of any other haematological abnormalities. The platelet count is often a useful guide in helping reach a diagnosis of lupus. It may be normal, but it is frequently modestly reduced and is rarely high unless there is active bleeding occurring which would be very likely to be clinically manifest in some way. For reasons which are poorly understood, patients with antiphospholipid antibodies are often chronically thrombocytopenic. In some cases, patients with acute disease may present with a picture similar to that of idiopathic thrombocytopenia, as a consequence of the development of anti-platelet antibodies, and destruction of platelets within the reticulo-endothelial system. Platelet counts may be more difficult to interpret in chronic disease (see below). The combination of a mild anaemia, lymphoenia and relative thrombocytopenia in a patient with arthralgia and other symptoms suggestive of lupus would make one strongly suspect that SLE was the diagnosis. While patients with other conditions which enter into the differential diagnosis, particularly rheumatoid arthritis, may be anaemic, leukopenia and thrombocytopenia are not typical features of other acute inflammatory rheumatic disorders. Patients with rheumatoid arthritis, for example, are much more likely to have a leucocytosis and a high platelet count, as do patients who present with systemic vasculitides such as Wegener's granulomatosis or microscopic polyarteritis.
Patients with active lupus characteristically have an elevated ESR. The ESR may be influenced by a number of factors including the level of haemoglobin and the plasma viscosity which is affected by the levels of globulins in the blood. In some patients the ESR is a good guide to disease activity but certain patients have a chronically elevated ESR as a consequence of hyperglobulinaemia. This is particularly true in patients with primary Sjögren's syndrome.
Lupus patients do not typically have a high C-reactive protein. It is undoubtedly true that a high ESR is frequently indicative of active lupus and the CRP may not be elevated at all. Conventional wisdom dictates that if the CRP is abnormally elevated in the disease (in many labs the upper limit of normal is either 5 or lOmg/l), this is suggestive of an infection. A CRP in excess of 50 should definitely make one think of an intercurrent infection. However, in patients with recurrent pyrexias, and in some cases with evidence of serositis (pericarditis or pleurisy), a CRP of between 15 and 40 may occur. All CRP and ESR results should be interpreted very carefully in the appropriate clinical context.
There is a wide range of special tests available which facilitate the diagnosis of lupus. These may broadly be split into two categories:
A number of autoantibodies which help to diagnose lupus, and to distinguish the disease from other causes of vasculitis, can be measured in the serum. These tests are performed in most large hospitals and there are many commercially available kits which enable laboratories which do not have a specialist interest in this area to produce quick and reliable results. Most hospitals perform tests of this type about once a week and all that is required is a serum sample which does not have to be collected in any special way.
These are found in 95 - 99% of untreated patients with lupus and many laboratories will perform an ANA screen before proceeding to more sophisticated tests. Results are presented in the form of a titre and a "pattern" e.g. "ANA positive 1:80 speckled pattern". This test is performed by indirect immunofluorescence, usually on a cell line such as HEp-2 cells, and the pattern relates to the distribution of fluorescent staining which is seen on the test cells under the microscope. Certain patterns of immunofluorescence are associated with the results in the other tests e.g. a speckled pattern is often associated with the presence of autoantibodies to the extractable nuclear antigen, ribonuclear protein. A "weakly" positive ANA (e.g. with a titre of 1:80 or less) may sometimes be a normal finding, particularly in middle aged multiparous females, and a positive result at this level should be interpreted in the light of the clinical picture and the results of other tests which are performed. A positive ANA is sometimes found in patients with other autoimmune disorders, such as myasthenia gravis or immune-mediated thyroid disease and in some patients with rheumatoid arthritis, particularly with extensive extra-articular involvement. In the event of a weakly positive result which does not fit the clinical picture the test should be repeated.
This is now easily measured using an agglutination test and is, of course, found in patients with rheumatoid arthritis and 2 - 10% of healthy adults. However, up to 40% of patients with lupus have a positive rheumatoid factor. A positive rheumatoid factor may also be found in patients with mixed essential cryoglobulinaemia.
These are detected in most laboratories by immuno-electrophoresis or by ELISA. A detailed discussion of the biology of these antigens is clearly outside the scope of this chapter but a number of points are worth remembering. The commonly measured antibodies to ENA directed to Sm (Smith antigen), RNP (ribonuclear protein), Ro and La (once called SS-A and SS-B). Anti-Sm antibodies are common in lupus, and anti-ribonuclear protein antibodies occur in patients who have 'mixed connective tissue disease'. Antibodies to Ro and La are particularly associated with primary Sjögren's syndrome. Many patients with lupus have antibodies which bind to more than one antigen of this type and a patient with a strongly positive ANA who has multiple antibodies to different extractable nuclear antigens is very likely to have lupus. A range of antibodies to other antigens such as 'Jol' or 'PMSCl', are also measured in specialist laboratories. These are associated with various overlap syndromes which may involve the lungs, skin and joints. Antibodies to Ro and La in the context of hypergammaglobulinaemia, a high ESR and sicca syndrome point strongly to a diagnosis of primary Sjögren's syndrome. Ro and La are named after two French-Canadian patients - called Robert and Lavoie.
Antiphospholipid antibodies are found in 30 - 40% of patients with lupus. Both 1gM and IgG antibodies are measured by ELISA. These antibodies are associated with a pro-thrombotic tendency, recurrent abortions and thrombocytopenia when present at high levels and are also found in the 'primary antiphospholipid syndrome' (Hughes syndrome). They may occur in some patients after infections (e.g. with parvovirus, TB, mycoplasma, or streptococci ).These antibodies are very often associated with prolongation of the APTT and a positive 'lupus anti-coagulant' (see below).
A 'lupus anticoagulant test' is performed in many haematology laboratories. This is a functional test which involves mixing the patients' plasma with normal plasma in order to evaluate its effect in in vitro clotting assays. To perform this test the laboratory needs to receive a blood sample taken into ACD.
Antibodies to double-stranded DNA are diagnostic of lupus. Some laboratories perform a "Farr assay" while others use kits of various types. A result should always be quoted with a normal range. As noted below, these antibodies are very useful in patients for monitoring disease. Occasional patients with obvious lupus and other antibodies do not have antibodies to double-stranded DNA.
These are a recently described autoantibody to the collagenous part of the Clq molecule. These antibodies are associated with severe lupus, particularly that affecting the kidneys. They occur in around 30% of patients with the disease and are presently only measured in specialist centres. They are also found in patients with hypocomplementaemic urticarial vasculitis (HUVS).
These are also measured in only a few specialist centres. They are particularly associated with drug-induced lupus.
Hypocomplementaemia is a feature of patients with active lupus. Most laboratories measure antigenic C3 and C4 and some also measure haemolytic complement activity (CH5O). Results should always be quoted with an associated normal range. There are many causes of a low C4 (see table 1). Results should also be carefully interpreted in the clinical context. Complement levels may be very valuable in monitoring disease activity (see below). There are two important practical points which must be made. Firstly, C3 and C4 levels may rise in pregnancy due to enhanced hepatic synthesis and this should be taken into account when interpreting complement tests performed in a pregnant patient. Secondly, it is not uncommon to receive a report from the laboratory in which there is a normal level of C3 and C4 but a CH5O which seems abnormally low. The problem causing this is that there has been in vitro complement activation as a consequence of a delay in the sample reaching the relevant laboratory. In very rare cases it may point to a congenital deficiency of a complement component, usually of the terminal pathway.
| Table 1 - Causes of a low C4 Systemic lupus erythematosus Hereditary angio-oedema (Cl inhibitor deficiency) Cryoglobulinaemia Rheumatoid vasculitis Sub-acute bacterial endocarditis Serum sickness Severe infection Inherited deficiency (very rare) |
As a general rule, most patients with lupus will be under the care of a hospital clinic - either rheumatology, nephrology, or dermatology. Most hospitals which initiate treatment with cytotoxic drugs also undertake to monitor their use and this is particularly the case for patients who are receiving cytotoxic agents which can cause leukopenia or hepatic or renal impairment. Many lupus patients are treated with azathioprine and such patients should have a full blood count and liver function tests measured at least every six to eight weeks and more frequently when the treatment is being initiated. In addition to the monitoring of cytotoxic drug therapy, patients with lupus need regular blood tests to monitor their disease activity.
All lupus patients need to have a full blood count monitored regularly. There are many causes of anaemia in the disease and in all cases a fall in haemoglobin should be carefully investigated. Care should be taken not to miss immune-mediated Vitamin B12 deficiency, iron deficiency due to non-steroidal therapy or another cause of blood loss, or a haemolytic anaemia. Attention, therefore, needs to be paid to the red cell indices and a blood film performed regularly. Measurement of the appropriate haematinics may be necessary.
A biochemical profile, including electrolytes, urea, creatinine, calcium, phosphate and liver function tests should be performed on a regular basis. If a patient is being reviewed in the clinic at 4-6 weekly intervals then both a blood count and biochemical profile would normally be performed at the same time as the special blood tests, indicated below, on each clinic visit.
Anti-nuclear antibodies, rheumatoid factor and antibodies to extractable nuclear antigens are primarily of value in the diagnosis of the disease and are not routinely used in most institutions for its monitoring. In some centres changes in titre of antibodies to extractable nuclear antigens are measured for research purposes. The most useful tests in the monitoring of lupus are anti-DNA antibodies and complement levels. In the majority of patients levels of antibodies to double-stranded DNA will rise markedly when there is a disease flare. This is particularly true for disease flares affecting the kidneys. However, not all patients behave in the same way and it is usually possible to develop a "feel" for how a patient's blood tests vary with disease activity as one sees them on a regular basis over the years. In some patients dsDNA binding is always elevated, in spite of the fact that they may be completely well, and in other patients it remains low even though the disease is active in other ways. In some cases there are fluctuations in the levels of anti-dsDNA antibodies from clinic visit to clinic visit, with no real relationship to disease activity or other serological parameters.
Complement tests are particularly useful in the monitoring of patients with lupus, especially those with renal disease. C4, which is a classical pathway component, is the first to fall in the event of a flare. Many patients, some of whom will have C4 null alleles (genes which result in defective C4 production), have a chronically low C4 level but a relatively normal C3 and in the absence of other signs of disease activity this does not necessarily mean that a change in therapy is required. However, if a patient runs a normal C4 for many months and there is then a precipitate fall in the level this may presage a flare. C3 consumption usually implies even more active disease and a low C3 is frequently found in patients with renal disease. Some centres perform assays of C3 breakdown products - e.g. C3d, and there is a good relationship between the levels of these C3 products and disease activity. As noted above, CH5O levels may need to be interpreted with caution if there is any delay in the specimen reaching the lab but, in general, a low CH5O will reflect complement consumption and often co-exists with a low C3 and C4 level. There is a strong association between hypocomplementaemia and autoantibodies to Clq. At present, however, the latter are only measured in a few specialist centres.
Dr. Sozos Loizou |
Dr. Kevin A. Davies |