Electrophoresis and immunofixation, pp. 342–343).Evaluation of specific antibody production (antibacterial and antiviral antibodies)
Electrophoresis and immunofixation
‘Bence–Jones proteins’; urine electrophoresis and immunofixation
Acute phase proteins (CRP, ESR, SAA)
Measurement of serum complement components
Haemolytic complement (lytic complement function tests)
Double-stranded DNA antibodies
Antibodies to extractable nuclear antigens (ENA)
Other patterns of autoantibodies identified on ‘autoantibody screen’
Autoantibodies in autoimmune hepatitis
Antibodies to gastric parietal cells and intrinsic factor
Thyroid disease (thyroid peroxidase antibodies)
Islet cell antibodies (ICA), anti-GAD antibodies, and insulin receptor and insulin antibodies
Adrenal antibodies and other endocrine autoantibodies
Classification of autoimmune polyglandular syndromes
Endomysial (EMA) and tissue transglutaminase (tTG) antibodies
Autoantibodies and neurological disease
Normal range (adults only):
The three main classes of Igs are measured by either rate nephelometry or turbidimetry on automated analysers. The principles are similar, dependent on immune complex formation, using antisera specific for the class of antibody. Rarely radial immunodiffusion (RID) may be used; this is slow and less accurate. For automated analysers, coefficients of variation should be in the range of 5–10%. Results are standardized against international standards. In the UK, an external quality assurance (EQA) scheme operates. Laboratories should provide normal ranges, which vary according to age and sex. Unfortunately many laboratories do not adjust ranges for age and sex, which may lead to confusion.
•Suspected immunodeficiency (1° or 2°); diagnosis and monitoring.
•Suspected myeloma, plasmacytoma; diagnosis and monitoring.
•Liver disease (PBC, hepatitis, cirrhosis).
•Post-BM/stem cell transplantation; monitoring.
Measurement of serum Igs does not provide categorical diagnosis in any disease. Normal serum Igs do NOT exclude immunodeficiency. In all cases, measurement of Igs MUST be accompanied by serum electrophoresis and immunofixation to look for paraproteins ( Electrophoresis and immunofixation, pp. 342–343).
•X-linked agammaglobulinaemia (XLA) (absent B cells; all Igs low/absent).
•Common variable immunodeficiency (CVID) (reduced T/B cells; low Igs).
•Hyper-IgM syndrome (normal/raised IgM; low/absent IgG, IgA).
•Selective IgA deficiency (absent IgA; normal IgG, IgM).
•Severe combined immunodeficiency (SCID) (mainly children; all Igs low; absent T cells).
•Lymphoma (reduced IgM; IgA normal; IgG normal or low; disease, chemotherapy, or radiotherapy).
•Herpesviruses (rare, EBV in X-linked lymphoproliferative disease).
•Drugs (immunosuppressants, e.g. cyclophosphamide, azathioprine, chemotherapy).
•Renal loss (IgM normal; IgG and IgA reduced).
•GI loss (IgM normal; IgG and IgA reduced).
•Chronic infection—all Igs raised:
•Sjögren’s syndrome—raised IgG (all IgG1).
•Sarcoidosis—raised IgG and IgA; IgM usually normal.
•PBC—IgM, may be very high (>30g/L) with small monoclonal bands on a polyclonally raised background.
•Alcohol-related—↑ IgA, polyclonal, β–γ bridging on electrophoresis.
•Autoimmune hepatitis (↑ IgG, IgA; normal IgM).
•Hodgkin’s disease—IgE raised (also eosinophilia).
•Acute common viral infections—raised IgM, normal IgG and IgA
•HIV—all Igs raised (IgG very high, but polyclonal).
ALL patients with recurrent infections* should be reviewed by an immunologist or a paediatric immunologist, according to age, irrespective of age. Any patient with recurrent infections and low serum Igs has an immunological problem until proven otherwise.
Note: (*) recurrent infections can be pragmatically defined as two or more major microbiologically/virologically proven infections, requiring hospitalization, within 1 year. One major infection and recurrent minor infections should also be referred where minor infections are documented infections requiring treatment in the community.
Patients with unusual infections or with illness caused by opportunistic or normally non-pathogenic organisms and patients with infections in unusual sites (without good reason) should all be referred for further investigation.
Normal range (adults):
As for serum Igs, IgG subclasses are normally measured by nephelometry or turbidimetry. RID is still occasionally used.
There are no absolute indications for testing, as significant immunodeficiency can occur in the presence of normal subclasses, and conversely complete genetic absence of a subclass may be completely asymptomatic. Measurement is usually performed as part of the work-up of patients with recurrent infections. IgG4 disease has been described recently; this is associated with a wide range of organ-based diseases. IgG4 levels are significantly raised.
Low levels may be significant in the context of presentation with recurrent infections. Deficiency of IgG3, which is involved in immunity against viruses and associated with asthma and intractable epilepsy. IgG2 deficiency may be seen in patients with IgA deficiency and may be associated with poor responses to polysaccharide antigens such as the capsular polysaccharides of bacteria.
Raised IgG1 with normal or reduced IgG2, IgG3, and IgG4 is seen in Sjögren’s syndrome and is a specific pattern, which may occasionally be helpful in diagnosis.
Raised IgG4 suggests IgG4 disease.
Units: variable: u/L, IU/mL, μg/mL.
Ranges: variable, check with reporting laboratory.
•Pneumococcal antibodies: >20u/L (asplenics >35).
•Tetanus antibodies: >0.1IU/mL (minimum protective level).
•Haemophilus influenzae type B: >1.0μg/mL (full protection; asplenics >1.5).
Measurement of antibody production against defined pathogens or antigens purified from pathogens plays an important role in the investigation of suspected immunodeficiency. Most assays are carried out by enzyme-linked immunoassay, but some viral antibodies are still measured by haemagglutination or complement fixation. Pre- and post-immunization samples should be run on the same run for direct comparison, as coefficients of variation for the assays tend to be high—15–25%! An EQA scheme exists. Assays have tended to focus on agents for which there are safe and effective vaccines. Live vaccines should NEVER be given to any patient in whom immunodeficiency is suspected.
Antibodies normally run in immunology laboratories include pneumococcal polysaccharides, which may be further differentiated as IgG1 and IgG2, H. influenzae type B (Hib), and tetanus. Diphtheria antibodies are not run by many laboratories, as the assay performance has been so poor. Meningococcal C polysaccharide antibodies are run by a few specialized laboratories, but correspondence with known clinical status has been poor. Anti-streptolysin O titre (ASOT) may be helpful.
Antibodies to pneumococcal serotypes are available from reference laboratories and may be valuable in assessing response to the conjugated pneumococcal serotype vaccine. There is debate about the protetctive levels (0.2–0.35).
Viral antibodies may be valuable to natural exposure and immunization antigens such as polio, measles, mumps, rubella, chickenpox, EBV, and hepatitis B (if immunized).
These assays should be used in the work-up of patients with suspected immunodeficiency or in monitoring change in such patients. Responsiveness to immunization is a helpful marker of immunological recovery post-bone marrow transplant (BMT). Annual monitoring of levels may be valuable in asplenic patients, as such patients lose immunity more rapidly than a eusplenic population.
The interpretation is entirely dependent on the context. Assays for pneumococcal polysaccharides measure a composite of responses to the 23 strains in the Pneumovax® vaccine. This can be misleading as not all strains represented in the vaccine are equipotent as immunostimulators. This means that a ‘normal’ response may actually mean a good response to the immunogenic strains, masking failure of response to the less immunogenic strains. For this reason, evaluation of such patients should be carried out by an immunologist with an interest in immunodeficiency. More weight should be placed on change in response to immunization than to actual values.
A ‘normal’ response to immunization has never been standardized, with publications frequently using different criteria, rendering comparison impossible. The following is a useful working definition: ‘a 4-fold rise in titre, which rises to well within the normal range’.
This usually presents with upper and lower respiratory tract infections (Streptococcus pneumoniae, Haemophilus, Staphylococcus, Klebsiella), leading to chronic bronchiectasis and sinusitis; GI infections (Salmonella, Giardia, Campylobacter); skin infections (recurrent boils); autoimmune features (ITP, haemolytic anaemia, diabetes, thyroid disease) only in CVID, not XLA. ↑ incidence of lymphoma.
Normal serum Igs DO NOT EXCLUDE antibody deficiency (IgG subclass deficiency, specific failure of antibody production against polysaccharides). Use test immunization with killed or purified component vaccines to test humoral responses.
Patients are at ↑ risk of overwhelming sepsis: S. pneumoniae, H. infleunzae, Staphylococcus aureus, Meningococcus, Klebsiella species, Capnocytophaga canimorsus (from dog bites), fulminant malaria, and babesiosis. Risk is lifelong. Asplenia may result from trauma, involvement in malignancy, removal for diagnosis (rare these days), coeliac disease, and sickle disease and rarely due to congenital absence. Serum Igs and IgG subclasses are normal, but responses to polysaccharide antigens are often poor, especially in patients with lymphoma. Blood film will show Howell–Jolly bodies. Absence (if not known from records) will be shown by ultrasound (US).
IgD is rarely measured in clinical practice, as its main function is as a membrane receptor. Elevated levels may be seen in periodic fever syndrome, in hyper-IgD syndrome, due to deficiency of mevalonate kinase, and in IgD-secreting myeloma. Measurement is usually by RID.
Units: not applicable to electrophoresis (qualitative). Paraprotein quantitated by scanning densitometry reported in g/L.
Normal range: N/A
In serum or urinary electrophoresis, the relevant body fluid is applied to an electrolyte-containing agarose gel. A current is applied across the gel and causes the proteins to migrate through the gel on the basis of their charge and, to a lesser extent, their size until they reach a neutral point in the electric field. The proteins are then visualized with a protein-binding stain. If the total protein is known, then the electrophoretic strip can be scanned and the absorption by the stain measured, which will be proportional to the amount of protein in the particular region in the gel. Thus, any monoclonal bands can be directly measured. This is useful for patients with myeloma, as immunochemical methods for measurement of Igs may be inaccurate in patients with myeloma.
Immunofixation is the technique by which monoclonal Igs are identified by overlaying the electrophoresed strips with antisera against heavy and light chains. These precipitate with the monoclonal proteins in the gel, and unbound antisera can be washed free prior to staining.
The same techniques can be carried out with urine, although this may require concentration to provide the clearest results.
Electrophoresis and, if necessary, immunofixation of serum is an integral part of measurement of serum Igs. ALL requests for serum Igs must have electrophoresis carried out; failure to do so will lead to important abnormalities being missed. There is no place for carrying out electrophoresis as a stand-alone test.
See Table 4.1 for interpretations of results.
| Report | Interpretation |
| Reduced albumin | Chronic inflammation, nephritic syndrome |
| Absent α1 band | Absent/reduced α1-antitrypsin |
| ↑α2 band | Chronic inflammation/infection; also seen in nephrotic syndrome, due to selective retention of α2-macroglobulin |
| ↑β | Seen in pregnancy (raised β-lipoprotein) and iron deficiency (transferrin) |
| β–γ bridging | Caused by raised polyclonal IgA, e.g. cirrhosis |
| ↑γ | Caused by polyclonal ↑ in IgG: infection/inflammation |
| Faint band(s) on polyclonal background | Caused by monoclonal escape during polyclonal response to infection/inflammation (does NOT indicate myeloma) |
| Monoclonal band in γ | Due to myeloma, lymphoma, and MGUS* |
| Absent/reduced γ | Due to inherited or acquired Ig deficiency |
* MGUS, monoclonal gammopathy of uncertain significance—most evolve to myeloma, given time (years).
Monoclonal proteins may polymerize to give >1 band. Some patients will have >1 clone present producing different Igs.
Densitometry cannot be used where the monoclonal protein overlies the β-region, as the figures include non-Ig proteins.
Modifications in the techniques for measurement of free, as opposed to bound, light chains are valuable in monitoring light chain-only myelomas and other myelomas that produce excess free light chains, in addition to a whole paraprotein, which would previously have been monitored by measurement of 24h urinary light chain excretion. Urinary measurement can be problematic where there is renal impairment, and as light chains are nephrotoxic, as the disease advances, urinary measurements become less accurate.
Bence–Jones proteins are urinary free light chains, i.e. unbound to heavy chains. During normal antibody synthesis, a small excess of light chains are produced which are excreted. Hypergammaglobulinaemic states, such as RhA and chronic infection, may therefore be accompanied by excretion of polyclonal free light chains. Monoclonal free light chains are seen in myeloma and may be the only marker in light chain-only myelomas, which do not produce any heavy chains at all.
Testing is carried out as for serum.
Units: usually reported qualitatively, but a ‘cryocrit’ can be measured in a similar way to a manual Hct using capillary tubes.
Normal range: tiny amounts of cryoglobulins may be found in normal individuals.
Cryoglobulins are Igs that precipitate when serum is cooled. The temperature at which this occurs determines whether disease will result. If the blood circulates through a part of the body where the temperature is below the critical temperature, then the protein will precipitate in the capillaries, causing obstruction, vascular damage, and eventually necrosis. The temperature of the hand is ~28°C at ambient room temperature. To check for the presence of cryoglobulins, take blood using a warmed syringe into a warmed bottle and transport to the laboratory at 37°C, using a Thermos™ flask with either pre-warmed sand or water at 37°C. The laboratory will allow the blood to clot at 37°C and then cool the serum. Cryoglobulins will form a precipitate as the temperature drops. The precipitate is then washed and re-dissolved for analysis by electrophoresis and immunofixation.
Cryoglobulins are not the same as cold agglutinins (a feature of Mycoplasma pneumoniae infection) ( Chapter 3).
All patients with Raynaud’s phenomenon of new onset or with winter onset of purpuric or vasculitic lesions on the extremities should be tested. Chronic hepatitis C infection is often accompanied by type II cryoglobulinaemia and a characteristic syndrome—‘mixed essential cryoglobulinaemia’ = autoimmune phenomena, arthritis, ulceration, glomerulonephritis, neuropathy. C3 normal; C4 reduced. Patients with myeloma, SLE, Sjögren’s syndrome, and RhA are also at risk.
Interpretation of the results of testing cryoglobulins is provided in Table 4.2.
| Type | Nature of cryoprecipitate |
| Type I | All monoclonal Igs; myeloma, lymphoma |
| Type II | Monoclonal Ig with RF activity; myeloma, lymphoma, connective tissue diseases, infections (especially HCV; SBE) |
| Type III | Polyclonal RF: connective tissue diseases, infections |
This is found less commonly than cryoglobulins. It will not be detected unless both EDTA and heparinized blood samples are sent warm to the laboratory. The main association is with occult malignancy (and thrombophlebitis migrans). Also associated with connective tissue disease, pregnancy, OCP use, DM, and cold urticaria.
Normal range: 1–3mg/L.
The test measures free β2-microglobulin, which normally forms the light chain of HLA class I molecules but is shed when there is ↑ lymphocyte turnover. It is usually rapidly cleared by the kidneys. Measurement is usually by an automated analyser, nephelometry, or turbidimetry. RID is still used.
The main indication is as part of routine monitoring of patients with myeloma and HIV. Other biomarkers are now felt to be more useful. Very high levels are seen in renal failure and patients on dialysis, which can cause β2-microglobulin amyloid.
Levels elevated in:
•HIV infection (surrogate marker of progression).
•Myeloma (marker of tumour mass).
•CVID (correlation with severity).
•Renal dialysis (depending on type of membrane).
•C-reactive protein (CRP): mg/L.
•Erythrocyte sedimentation rate (ESR): mm/h.
Normal ranges:
•ESR ( Erythrocyte sedimentation rate, p. 252).
•SAA (not measured routinely).
Serum proteins CRP and SAA are amenable to measurement by nephelometry or turbidimetry. Measurement of the ESR is covered in Chapter 3.
Acute and chronic infections, vasculitis, connective tissue disease, arthritis.
Clinicians are usually confused by ESR and CRP—they do not give the same information and should be used together. CRP is like blood glucose, whilst ESR is like HbA1c. CRP rises within hours of onset of inflammation/infection and falls quickly once treatment is instituted. It is therefore useful for rapid diagnosis and monitoring response. The ESR rises slowly, being dependent, in part, on fibrinogen, a long-lived protein, and falls equally slowly (see Fig. 4.1). In active SLE, the ESR is high, but CRP is not elevated. CRP is driven by interleukin (IL)-6 and may be elevated in myeloma. See Table 4.3 for interpretation of results.
Fig. 4.1 Time course of acute phase response proteins. ESR in acute phase response parallels fibrinogen level.
Table 4.3 Causes of elevated CRP
| Level of CRP | Common associations |
| Little or no change (<4–100mg/L) | |
| Moderate elevation (100–200mg/L) | |
| Large elevation (>200mg/L) | |
| Huge elevation (>400mg/L) |
CREST, calcinosis, Raynaud’s syndrome, oesophageal motility dysfunction, sclerodactyly, and telangiectasia.
Levels in very young children may be much lower for a given stimulus. A very small number of patients do not make inflammatory responses that exceed the normal range but seem to run on a lower ‘normal’ range (10-fold less); ultra-sensitive assays for low-level CRP are available.
Amyloid refers to the deposition of altered proteins in tissues in an insoluble form. The precursor protein varies according to the cause and can often be measured specifically. Amyloid is usually confirmed by special stains on histological examination of biopsies. Measurement of serum Igs and electrophoresis, β2-microglobulin, and CRP is essential if amyloid is suspected (see Table 4.4).
| Amyloid protein | Protein precursor | Clinical syndrome |
| AL, AH | Light or heavy chain of Ig | Idiopathic, multiple myeloma, γ-heavy chain disease |
| AA | Serum amyloid A | 2°, reactive: inflammatory arthritis, familial Mediterranean fever, hyper-IgD syndrome, TRAPS (periodic fever), Behçet’s, Crohn’s disease |
| Aβ2M | β2-microglobulin | Dialysis amyloid |
| ACys | Cystatin C | Hereditary cerebral angiopathy with bleeding (Iceland) |
| ALys, AFibA | Lysozyme, fibrinogen Aa | Non-neuropathic hereditary amyloid with renal disease |
| AIAPP | Islet amyloid polypeptide | DM type 2; insulinoma |
| AANF | Atrial natriuretic peptide | Senile cardiac amyloid |
| ACal | Procalcitonin | Medullary carcinoma of the thyroid |
| AIns | Porcine insulin | Iatrogenic |
| ATTR | Transthyretin | Familial amyloid polyneuropathy, senile cardiac amyloid |
| Aβ | Aβ-protein precursor | Alzheimer’s disease |
| AprP | Prion protein | Spongiform encephalopathies |
TRAPS, tumour necrosis factor receptor-associated periodic syndrome.
Normal ranges:
•Factor B (rarely measured routinely).
•Other components usually reported as percentage of normal human plasma.
C3, C4, and factor B are usually measured by rate nephelometry or turbidimetry. Other components are measured by RID or simply by double diffusion where presence or absence is the only result of interest. Complement breakdown products are measured by RID or enzyme-linked assay (EIA).
Valuable in:
•Suspected complement deficiency (C3, C4, haemolytic complement).
•Suspected anaphylaxis (anaphylotoxins C4a, C5a).
•Suspected hereditary angioedema (C3, C4, C1q, C1 esterase inhibitor, immunochemical AND functional).
Complement deficiency is common (especially C4 and C2 deficiencies); predisposes to recurrent neisserial disease, bacterial infections (C3 deficiency), and immune complex disease (lupus-like). Anyone with >1 episode of systemic neisserial disease has a complement deficiency until proven otherwise.
•Functional: reported as percentage activity, compared to normal fresh plasma.
Normal range:
•Immunochemical: 0.18–0.54g/L (paediatric ranges not well defined, but lower than adults).
•Functional: 80–120% of normal plasma.
Immunochemical measurement carried out by RID; functional assay is usually a colorimetric assay.
Key indication is angioedema occurring WITHOUT urticaria at any age. If urticaria is present, diagnosis is virtually never C1 esterase inhibitor deficiency. C4 is a useful screen; normal C4 during an attack excludes C1 esterase inhibitor deficiency.
C1 esterase inhibitor deficiency causes hereditary angioedema.
Two types:
•Type I (common, 80%); absence of immunochemical C1 esterase inhibitor (C1-inh).
•Type II (rare, 20%); presence of non-functional C1-inh; immunochemical levels normal or high.
Both are inherited as autosomal dominant. Present with angioedema, NO urticaria; may involve the larynx and gut, usual onset at puberty. C4 absent during acute attacks. Treat with purified C1-inh (FFP may be a substitute but can make attacks worse) or icatibant; maintenance therapy with danazol, stanozolol, or tranexamic acid to ↓ frequency of attacks. Pregnancy/oral contraceptive exacerbate.
A rare type (type III) of hereditary angioedema is also described, thought to be due to gain-of-function mutations in clotting factor XII. Angioedema can also be caused by deficiency of ACE and C4-binding protein.
Rare acquired form due to autoantibody to C1-inh (SLE, lymphoma); C1q levels are reduced, and paraproteins may be present.
Units: can be reported in arbitrary units, or as a percentage of normal plasma, but better reported as normal reduced or absent.
Normal range: present (80–120% of reference plasma).
Haemolytic complement assays screen for the integrity of the classical and alternate pathways and the terminal lytic sequence, and use either antibody-coated sheep cells (CH100, classical pathway) or guinea pig red cells (APCH100, alternate pathway). Either a gel or liquid assay can be used, but the gel is easier! Both tests must be performed in parallel.
Any patient in whom deficiency of a complement component is suspected. Testing is also used to monitor the effectiveness of the MoAb eculizumab, used to treat atypical HUS.
Reduced levels of haemolytic activity will be seen during infections and during immune complex diseases such as serum sickness and SLE. Testing for absence of a component needs to be undertaken a minimum of 4–6 weeks after recovery from infection. Absence in both CH100 and APCH100 indicates a deficiency in the terminal lytic sequence C5–C9 (C9 deficiency will give slow lysis). Absence of CH100 indicates a missing component in the classical pathway C1–C4. Absence of APCH100 indicates deficiency in the alternate pathway (factor D, factor B, C3).
Follow-up testing to identify the missing component will be performed by the laboratory automatically (if they are doing their job!).
Anyone who has a single episode of neisserial meningitis with an unusual strain or a second episode with a common strain MUST be assumed to have a complement deficiency until proven otherwise. REFER after recovery to the immunologist for investigation.
Deficiency of either of these factors may lead to HUS. Measurement is possible in specialized centres, with follow-up genetic testing. Autoantibodies against these proteins have been described.
This is an autoantibody which reacts with a neo-antigen in the alternate pathway convertase C3bBb to stabilize the convertase and ↑ complement breakdown. Typically seen in membranoproliferative glomerulonephritis and in lipodystrophy. C3 will typically be extremely low.
Autoantibodies are usually divided into organ-specific and organ-non-specific, but clinical testing rarely follows this pattern. They are therefore covered in convenient groups, associated with types of testing.
•Titre (particle agglutination assay).
Normal range:
Tests detect autoantibodies binding to human Ig; these can be of any class, but assays commonly recognize IgG and IgM autoantibodies. Suggestions that IgA RF may be helpful have not been widely accepted. Assays use either agglutination of Ig-coated particles (visual assay) or latex particle-enhanced nephelometry.
The test is only relevant in patients already diagnosed with RhA.
NOT a diagnostic test for RhA! Only +ve in 70–80%. High titre in a patient with known RhA is a risk factor for extra-articular manifestations and poorer prognosis.
RF is also found in:
•Healthy elderly (asymptomatic).
•Chronic bacterial (SBE) and viral infections (HIV, HCV).
•Acute viral infections (transient; especially adenovirus).
•Myeloma (often type II cryoglobulins).
•Connective tissue diseases (SLE, Sjögren’s, systemic sclerosis, polymyositis, undifferentiated connective tissue disease (UCTD)).
Antibodies to cyclic citrullinated peptide (anti-CCP) may be more valuable in the diagnosis of RhA, as they are more specific.
Another abused test! Usually used as an immunological fishing expedition. Multiple tissues (rodent), often with human Hep-2 cell line; gives rapid and semi-quantitative results for the following autoantibodies:
•ANAs (see Table 4.5 for patterns).
•Antimitochondrial antibodies (AMA).
•Anti-smooth muscle antibodies (ASMA).
•Anti-liver–kidney microsomal (LKM) antibodies.
•Anti-gastric parietal cell (GPC) antibody.
Where Hep-2 cells are used, other patterns of nuclear and cytoplasmic fluorescence may be seen. See Fig. 4.2 for examples.
Increasingly, laboratories are using multiplex analysers (modified enzyme-linked or flow cytometry-based testing) which allow high-throughput screening for nuclear and related antibodies.
The traditional method is to overlay suitably diluted serum into frozen sections of rodent liver, kidney, and stomach, and human Hep-2 cells. Bound antibody in the serum is then identified using a fluoresceinated anti-human IgG (or IgM, IgA) as the second stage. Slides are then read manually. Enzyme-based immunoassays are being introduced for screening, including multiplex bead-laser array systems. These can be very specific when purified or recombinant antigens are used but lose out because of their inability to pick up unexpected patterns.
Enzyme-linked assays are used for confirming antigens such histone antibodies and double-stranded (ds-)DNA antibodies.
The correct use of testing is to identify which specific autoantibody is being sought as part of the differential diagnosis.
Because of the multiple patterns detected in this system, the interpretation for each is covered separately (see Table 4.5).
Table 4.5 Patterns of antinuclear antibodies (ANAs)
| Homogeneous | SLE, drug-induced SLE (ds-DNA or histones) |
| Coarse-speckled | UCTD*, SLE (U1-RNP) |
| Fine-speckled | SLE, Sjögren’s (Ro and La) |
| Nucleolar | Systemic sclerosis, polymyositis, SLE |
| Centriole | Commonest in Mycoplasma pneumonia, also scleroderma |
| Proliferating cell nuclear antigen (PCNA) | SLE—highly specific, but rare |
| Centromere | ‘CREST’ syndrome, limited scleroderma, Raynaud’s, never diffuse scleroderma; may be confused with multinuclear dot pattern, which is seen in mitochondrial antibody-negative PBC |
| Histones | SLE, drug-induced (>90% of patients); other connective tissue diseases (low frequency) |
* UCTD, undifferentiated connective tissue disease (previously mixed connective tissue disease).
Note: ribosomal antibodies associated with SLE, especially neuro-lupus (cytoplasmic pattern, not nuclear).
Titre of antibodies does NOT correlate with disease activity.
ANAs may be seen transiently after viral infections, especially in children. Therefore, observe a child where low-titre antibodies occur in the absence of clinical symptoms compatible with juvenile arthritis.
Normal range: varies according to assay:
The original and still best assay is the Farr assay, a radioisotope-based assay. EIA tends to be widely used but is less specific, due to frequent contamination with single-stranded DNA. Crithidia assay used only rarely, as quick fluorescent screen (kinetoplast is pure ds-DNA).
Suspected SLE or autoimmune hepatitis; used to monitor SLE, in conjunction with complement studies.
Sensitive and specific for SLE and autoimmune hepatitis (AIH); +ve result is significant (in correct clinical context).
Units: reported qualitatively.
Normal range: dependent on antibody, normally −ve.
Usually carried out by enzyme-linked immunoassay. However, countercurrent immunoelectrophoresis and western blotting are still widely used, especially for rare antibodies. EIA is much more sensitive than other methods and has led to clinical confusion.
Should always be carried out in patients with suspected connective tissue disease. Monitoring at yearly intervals should be carried out in diagnosed patients, as the antibody pattern may change with time and this may correlate with changes in the clinical profile.
Reported qualitatively; normally laboratories will carry out a 6-antigen screen: Ro, La, ribonucleoprotein (RNP), Sm, Jo-1, and Scl-70 (see Table 4.6). A wide range of other antibodies are described, some of which may be available through reference laboratories.
Table 4.6 Antibodies to extractable nuclear antigens
| Ro/SS-A | SLE, Sjögren’s, neonatal lupus, neonatal congenital complete heart block (cause of ANA-negative lupus, as not picked up by standard ANA screen, which does not include Hep-2 cells). Newer assays will distinguish Ro52 and Ro60 |
| La/SS-B | SLE, Sjögren’s, neonatal lupus, neonatal congenital complete heart block (rare) |
| (U1)-RNP | Mixed (undifferentiated) connective tissue disease (if present alone); SLE (if present with ds-DNA). Other RNPs may be reported |
| Sm | Highly specific marker for SLE (mainly West Indians; rare in Caucasians) |
| Jo-1 | Polymyositis, dermatomyositis; transferase syndrome (fibrosing lung disease; 65% +ve); many other specificities are known (all recognizing transfer RNA (tRNA)-transferases) |
| Scl-70 | Systemic sclerosis (diffuse scleroderma); only 30% of patients are +ve |
| Pm-Scl (PM1) | Scleroderma–myositis overlap |
| Ku, Ki | Rare: SLE, UCTD, Sjögren’s syndrome, polymyositis |
| Mi-2 | Rare: steroid-responsive polymyositis |
| Histones | Found in connective tissue diseases, especially drug-induced lupus |
ANAs and ds-DNA antibodies, both +ve; low C3 and C4; raised complement breakdown products in active disease (C3d); normal CRP. ANA-negative lupus often anti-Ro positive. West Indian lupus anti-Sm+. Always check for cardiolipin antibodies and lupus anticoagulant (dRVVT) ( Antiphospholipid antibodies, p. 362). Ribosomal P antibodies may be a marker for neuropsychiatric lupus.
Monitor with CRP (differential between infection and flare of disease), C3, C4, and ds-DNA antibodies (no value from serial ANAs); rising titre of DNA antibodies often heralds a relapse (actual titre not related to disease activity).
Many types of antibodies to phospholipids are recognized. However, routinely available tests are IgG and IgM anticardiolipin antibodies detected by EIA, and the presence of a ‘lupus anticoagulant’ detected using the dilute Russell viper venom test (dRVVT). BOTH tests must be carried out together, as either may be +ve without the other, but the clinical significance is the same. Antibodies to β2-glycoprotein I are also important, as the protein is a key co-factor for pathogenic antibodies. Other specificities are available only as research tools.
Patients with unexplained venous or arterial thrombosis; recurrent miscarriage (>3); connective tissue disease; early TIAs or stroke (<60 years). Livedo reticularis is a cutaneous marker.
•Prolonged APTT and reduced platelet count (80–120 × 109/L) are typical features. May get false +ve VDRL (Venereal Disease Research Laboratory).
•Persistent +ve IgM anticardiolipin antibodies as the only marker of syndrome unusual, but clinically significant.
•Hughes’ syndrome = antiphospholipid antibodies without evidence of other connective tissue disease or vasculitis. Presents with recurrent miscarriage, thrombosis, or strokes. Fulminant disease causes multi-organ failure.
•Antibodies also found in SLE, but NOT cerebral lupus, Behçet’s, and Sneddon’s syndrome. Also seen after EBV infection (asymptomatic and disappear).
•Treatment is lifelong warfarinization if symptomatic (heparin in pregnancy); aspirin may be acceptable if asymptomatic (but no controlled studies). No indication for intensive immunosuppression.
Units: reported as titre (AMA) Anti-M2 reported as +ve or −ve.
Normal range: not usually detectable.
AMA are identified on fluorescent screen, and previously unknown +ves should be followed up with EIA or blot-based test for anti-M2 antibodies.
Suspected liver disease, especially with raised ALP (and in ♀); investigation of unexplained pruritus (early feature of PBC).
Ninety-five per cent of PBC patients +ve for AMA (anti-M2, against dihydrolipoamide acyltransferase, E2). The remainder may have antibodies against S100 antigen of the nuclear membrane (Nsp-II pattern; multinuclear dots) or gp210, another nuclear antigen. The type of antibody present has no influence on prognosis or response to therapy. Marked elevation in IgM.
Units: reported as titres (ASMA, anti-LKM antibodies). Anti-liver cytosol (LC-1) and soluble liver antigen (SLA) reported as +ve or −ve.
Normal range: not usually detectable.
ASMA and LKM are identified on fluorescent screen, and previously unknown +ves should be followed up with EIA or blot-based test for anti-LC and anti-SLA antibodies. Also do ANCA ( Vasculitic syndromes, pp. 372–373).
Suspected autoimmune liver disease (abnormal LFTs, alcohol and viral infections excluded).
Antibodies to HCV or HCV PCR+ = exclusion criteria for AIH.
•Type 1 (AIH-1) is ANA +ve, smooth muscle antibody (SMA) +ve, p-ANCA +ve, and SLA antibody +ve. Typically occurs in adults and has a better prognosis and responds well to therapy.
•Type 2 (AIH-2) is typically LKM-1 and LKM-3 antibody +ve and LC-1 antibody +ve. AIH-2 is seen in children and has a worse prognosis with poor response to therapy.
•In serological studies, 50% of AIH-1 are ANA +ve/SMA +ve, 15% ANA +ve only, and 35% SMA +ve only. However, there are biopsy-proven serologically −ve hepatitis; 8% of AIH-1 are SLA +ve only. 43% of AIH-2 are LC-1 +ve only. Therefore, necessary to do SLA and LC-1 tests. Prognosis is dependent on type and early diagnosis.
•LKM antibodies are also associated with drug-induced hepatitis (especially halothane) and chronic hepatitis C or D.
•Non-actin SMA may be seen in SLE and after viral infections (especially adenovirus).
•Sclerosing cholangitis may be associated with atypical p-ANCA ( Vasculitic syndromes, pp. 372–373).
Units: GPCs may be reported as titre or simply +ve/−ve (as titre of no clinical value); IF antibodies reported as +ve/−ve.
Normal range: GPC antibodies found in healthy normals without evidence of B12 deficiency or gastritis—may be at risk of later PA. IF antibodies only found in PA.
GPC antibodies are detected as part of the ‘autoantibody screen’; IF antibodies usually detected by either RIA or EIA, but assays are inconsistent. There is a robust EQA scheme for GPC antibodies, but none currently for IF antibodies.
GPC antibodies should be checked in all patients with thyroid disease, as there is a close association (‘thyrogastric disease’). Also check in patients with unexplained macrocytosis ± low B12. Positive antibodies identified incidentally should be followed up with a blood count and, if the MCV is high, a B12 level. IF antibodies may help in patients with a high MCV and low B12 to confirm PA. Diagnosis of PA more difficult now Schilling test withdrawn.
IF antibodies are NOT suitable for screening, as present in only 50% of patients with PA and pre-administration of B12 will interfere with detection.
GPC antibodies found in 90% of patients with PA and in 40% of patients with other organ-specific autoimmune diseases; the antigen is the β subunit of gastric H+/K+ ATPase. Anti-IF antibodies are of two types: those that block B12 binding to IF (70% of patients with PA) and those that block uptake of the B12–IF complex (35% of patients with PA).
Units: variable—check with the laboratory.
Normal range: low levels of antibodies may be found in asymptomatic individuals, although higher levels may indicate a predisposition to later development of thyroid disease.
TPO antibodies are the test of choice. Tg antibodies not monitored now, except as part of monitoring for thyroid carcinoma (interfere with assays for Tg, which is used as a tumour marker). Antibodies to TSH receptor may be stimulating or blocking, but these can only be identified by bioassay or RIA in specialized laboratories.
Suspected thyroid disease and as reflex testing when thyroid function is abnormal; often now combined with thyroid testing on same analyser. Also check in patients with PA. Use TPO antibodies. Other antibodies are for specialist use only.
TPO (thyroid microsomal) antibodies in 95–100% of Hashimoto’s thyroiditis, 70% of Graves’ disease; highest titres seen in Hashimoto’s. Tg antibodies add little to diagnosis (also found in other endocrinopathies and thyroid carcinoma). Antibodies to TSH receptor (stimulating or blocking) found in 95% of Graves’ patients.
Units: qualitative (previously measured in JDF units) for ICA; numeric values for other tests.
Normal range: 0.4% of normal population +ve for ICA.
ICA can be detected by immunofluorescence on pancreatic sections; other antibodies detected by EIA.
All newly diagnosed early-onset diabetics should be tested; they should also be tested for coeliac disease using endomysial or tTG antibodies. May be used to screen normoglycaemic first-degree relatives for likelihood of developing diabetes. Also advisable to screen children with coeliac disease.
ICA found in 75–86% of type 1 diabetics, but only 10% of type 2 diabetics and 2–5% of first-degree relatives. Levels decline with time and may disappear completely in long-standing type 1 patients. Antigen is glutamic acid decarboxylase (GAD) (similar antibodies cause stiff man (person!) syndrome, although a different epitope on the molecule is recognized); GAD65 and GAD67. Other target antigens for autoantibodies, with high specificity for diabetes, are:
•Insulin/proinsulin (seen at diagnosis in 40% of type 1 diabetics).
•Insulin receptor (associated with acanthosis nigricans and insulin resistance).
•IA-2 (present in 60% of newly diagnosed type 1 DM).
•ZnT8 (zinc transporter) found in 60–80% of type 1 DM.
Monitoring is of no value.
Note: IgE antibodies to porcine or bovine insulin may occur in insulin allergy, due to exogenous ‘foreign’ insulin.
Normal range: not detectable.
Usually detected by immunofluorescence on appropriate tissue (adrenal, ovary, testis, parathyroid, pituitary). EIA for 21-hydroxylase, 17-hydroxylase, and P450 side chain cleavage enzymes (targets of adrenal antibodies) in research centres.
Suspected autoimmune endocrinopathy (adrenal insufficiency, premature ovarian failure, hypoparathyroidism).
Positives indicate autoimmune disease of relevant organ.
Refer to Table 4.7.
Table 4.7 Classification of autoimmune polyglandular syndromes
| Syndrome | Major criteria | Minor criteria |
| Type I | ||
| Type II (Schmidt’s syndrome) | ||
| Type III | Thyroid disease |
IDDM, insulin-dependent (type 1) diabetes mellitus.
Units: qualitative for EMA; numeric for tTG.
Normal range: not usually detected in healthy individuals. +ve IgA EMA or tTG with normal biopsy probably indicates risk of later development of coeliac disease.
Originally identified as anti-reticulin R1 antibodies on ‘autoimmune screen’. Later identified as antigliadin antibodies by modified immunofluorescence or by EIA. EMA detected by immunofluorescence on monkey oesophagus (also umbilical vein, but this is difficult to read); purified recombinant human tTG used for EIA (assays with guinea pig tTG are less sensitive).
Patients with malabsorption, wheat intolerance, ‘irritable bowel’; children with type 1 diabetes (and suggested that they should be monitored every year!). Suspected dermatitis herpetiformis. Unexplained hyposplenism, small bowel lymphoma.
IgA EMA has nearly 100% sensitivity and specificity for coeliac disease. Antigen is tTG. In IgA deficiency (commoner in coeliac), IgG EMA is as good a diagnostic test. Gliadin antibodies (IgG or IgA) are not sensitive and specific and should no longer be used; they are found in a range of bowel diseases and in healthy individuals. Reticulin R1 antibodies may be picked up on routine autoantibody screen; these may also be found in IBD, especially with liver involvement, and non-specific bowel disease (post-infectious, allergic, etc.). Also found in dermatitis herpetiformis (typical rash). Strong association with juvenile type 1 DM.
Antibodies disappear with strict gluten-free diet, over 6–12 months. Persistent positivity of IgA EMA or IgA tTG antibodies in a patient on a gluten-free diet is an indication of incomplete/non-compliance with the diet. Monitoring tTG antibodies is valuable.
Gluten sensitivity enteropathy is also associated with neurological disease, typically cerebellar ataxia. Reports have suggested that gliadin antibodies are a marker for this syndrome; the lack of specificity of gliadin antibodies means that it is unlikely that these antibodies can be used as an accurate diagnostic test, and specific tests (EMA and tTG) should be used instead.
Units: normally reported qualitatively.
Normal range: antibodies when detected are highly disease-specific.
Autoantibodies to components of the skin can be detected either by direct immunofluorescence (DIF) on a skin biopsy of affected tissue or indirectly in the patient’s serum using monkey oesophagus as a substrate. Saline splitting of the skin may be used to identify the precise location of the antigenic target to differentiate between epidermolysis bullosa acquisita (antigen on the dermal side) and pemphigoid (antigen on the epidermal side).
Bullous (blistering) skin diseases.
Antibodies bind to the dermal–epidermal junction and recognize hemidesmosome antigens (BP 230 (BPAG1) and BP 180 (BPAG2)); detect by DIF on skin biopsies or in serum (only 70% of patients have detectable circulating antibodies in serum). Linear basement membrane deposition of IgG and C3 on DIF.
Antibodies to BP 180 (BPAG2) binding to the dermal–epidermal junction on DIF of biopsies, often −ve for serum antibodies. Linear basement membrane deposition of IgG and C3 on DIF.
Antibody binds to the cell surface of stratified squamous epithelium and recognizes the desmosomal proteins desmoglein I and III. Eighty to 90% of patients have detectable antibody in serum. Chicken-wire pattern of immunofluorescence in the epidermis. May occur as a paraneoplastic phenomenon in patients with lymphoma, but antigenic specificity is for desmoplakin I and II.
Granular deposits of IgA at the dermal–epidermal junction in dermal papillae on DIF; EMA and tTG antibodies will be +ve.
Most neurological autoantibodies of interest are rare and are available from reference laboratories. A variety of methods are used; some assays are reported with numeric values.
These are restricted to very specific neurological syndromes.
•AChRAb associated with MG (90%); reported numerically as high and low +ves to distinguish different clinical phenotypes. May also be found in asymptomatic relatives. Striated muscle antibodies (immunofluorescence on striated muscle section) strongly associated with underlying thymoma.
•Lambert–Eaton myasthenic syndrome (LEMS), occurring with small-cell carcinoma of the lung, associated with autoantibodies to voltage-gated Ca2+ channels, α and β subunits. Same tumour also associated with retinal autoantibodies. Antibodies to voltage-gated K+ channels are associated with neuromyotonia.
•Anti-ganglioside antibodies associated with GBS (GM-1, GD1a) and variants (Miller–Fisher—GQ1b, GT1a), chronic variants (chronic inflammatory demyelinating polyneuropathy), and neuropathy associated with paraproteins.
•Myelin-associated glycoprotein (MAG) antibodies associated with paraproteinaemic neuropathy, especially with Waldenström’s macroglobulinaemia.
•Antibodies to myelin basic protein found in MS, but not useful diagnostically.
•Anti-Yo (Purkinje cell antibodies) found in paraneoplastic cerebellar degeneration (gynaecological or breast tumours associated); anti-Hu (anti-neuronal nuclear antibodies (ANNA)) associated with paraneoplastic neuropathies and myelopathies (small-cell carcinoma). Anti-Ri (anti-neuronal nuclei) associated with cerebellar ataxia and opsiclonus (small-cell carcinoma, gynaecological or breast tumours associated). Other specificities have been defined.
•Anti-neuronal antibodies in the CSF of 74% of patients with cerebral lupus; also associated with anti-ribosomal P antibodies.
•Anti-GAD antibodies found in stiff person syndrome; same antigen to that found in pancreatic islets; diabetes usually occurs in stiff person syndrome. Antibodies appear to inhibit the production of γ-aminobutyric acid (GABA), an inhibitory neurotransmitter. Rasmussen’s encephalitis associated with autoantibodies to GluR3 receptor, which cause hyperexcitability of neurones.
•Anti-aquaporin-4 antibodies are associated with neuromyelitis optica (Devic’s disease).
Units: usually expressed as titre (qualitative EIAs are used in some centres).
Normal range: in adults, normal = undetectable.
Screening is usually carried out by immunofluorescence on ethanol-fixed human neutrophils; rapid EIA screening tests exist for +ve/−ve testing in emergencies. EIA to specific antigens is an essential follow-up. Distinction between antinuclear and perinuclear staining may require the use of Hep-2 cells.
Suspected vasculitis; acute glomerulonephritis.
•Two main patterns recognized: c-ANCA (mostly Wegener’s; 90% of Wegener’s ANCA +ve) and p-ANCA (some Wegener’s, microscopic polyarteritis, Churg–Strauss syndrome, glomerulonephritis, sclerosing cholangitis, AIH, ulcerative colitis).
•c-ANCA pattern due to antibodies against proteinase-3; p-ANCA pattern due to antibodies against MPO, lactoferrin, cathepsin, and elastase. Follow-up EIA required to identify antigenic specificity: minimum PR3 and MPO ELISA; other antigens available through supra-regional referral laboratories.
•In Wegener’s, monitoring titre of ANCA is useful; a rising titre in a patient in clinical remission heralds relapse.
•ANCA may be seen as epiphenomenon in states of chronic neutrophil activation and turnover, e.g. cystic fibrosis.
Units: qualitative (quantitation available through reference centres—used only to follow patients post-plasmapheresis or transplant).
Normal range: not detectable.
Usually carried out by EIA; screening by immunofluorescence is not sensitive. Biopsies will show linear IgG deposition in glomeruli (and alveoli). Antibodies recognize the NC1 region in the α3 chain of type IV collagen.
Acute glomerulonephritis, particularly if associated with pulmonary haemorrhage.
Positive antibodies confirm Goodpasture’s syndrome. Urgent plasmapheresis is required, and monitoring reduction of antibody with treatment is advisable. Disease may recur in transplanted kidneys, so monitoring of antibody is advised. Some patients may be both ANCA and anti-GBM antibody +ve.
Units: quantitative, through reference laboratories only.
Normal range: not detectable.
Usually carried out by EIA or western blot. Indirect immunofluorescence also available using a cell line expressing PLA-2.
Marker antibody for idiopathic membranous nephritis (IMN).
Very high specificity for IMN but may also be found in lupus.
Unit: kU/L.
Normal range: <100kU/L (>14 years old).
Previously carried out by RIA, now by EIA.
There are FEW indications for testing. Screening for atopic disease; investigation of suspected hyper-IgE syndrome (Job’s syndrome, a rare immunodeficiency), Churg–Strauss vasculitis. Gating requests for radioallergosorbent tests (RASTs) on the basis of IgE is scientifically unsound (see Interpretation below).
Significant allergic disease is possible with low levels of total IgE (including anaphylaxis). Only patients with undetectable IgE (<7) are unlikely to have allergic disease. Conversely, levels above the normal range are compatible with no clinical allergic disease. IgE >1000 associated with atopic eczema; IgE >50,000 confirms hyper-IgE syndrome (although patients may have lower levels—diagnosis is clinical). Raised levels are also seen in parasitic infections of the bowel, filariasis, lymphoma (especially Hodgkin’s disease), and Churg–Strauss vasculitis.
Unit: mm wheal size, compared to histamine and saline controls.
Normal range: no wheal.
This remains the gold standard for allergy diagnosis. It identifies IgE-mediated reactions (type I) such as inhalant allergy, anaphylaxis, and food allergy. It is dependent on triggering the release of histamine from cutaneous mast cells. Solution of allergen or controls (histamine or saline) placed on the skin and pierced through by a lancet. After 15min, wheal will be visible. Positive is at least 2mm greater than −ve control. Histamine control must be +ve. Not interpretable if the patient is dermographic (−ve control gives wheal). Can use factory-prepared allergens; also use double-prick technique with fresh foods (prick food, then the patient)—useful where allergens are labile, e.g. fruits.
Mainstay for diagnosis of all types of allergic disease. Contraindicated when there is significant skin disease, previous severe allergic reactions (use RASTs first), patients on antihistamines (need to be off drug for a week). Other drugs will interfere, e.g. Ca2+ channel blockers, tricyclic antidepressants.
Results can only be interpreted in the context of the clinical history. Testing should be tailored to individual patients to answer specific questions. May need to be followed up by open or blinded challenges where −ve results are obtained in patients with good histories.
Good specificity for inhalant allergens and some foods (nuts, fish); results comparable with RAST testing.
•Latex allergy associated with food reactions: banana, avocado, kiwi fruit, chestnut, potato, tomato, cannabis, lettuce. Also birch pollen allergy commoner.
•Birch pollen allergy (asthma, rhinitis) with food-related reactions to nuts, apples, plums, cherries, carrots, and potatoes (‘oral allergy’ or ‘pollen–fruit’ syndrome).
•Ragweed pollen with melon and banana.
Units: continuous numeric scale in IU/mL. Also (less frequently now) reported as grades 0–6.
Normal range: grades 0 and 1 indicate insignificant specific IgE.
For an overview of grades, see Table 4.8.
Previously tested by RIA (hence the acronym RAST = radioallergosorbent test); now identified by enzyme-linked or fluorimetric assays.
‘RAST’ tests are expensive and should be reserved for cases where skin prick testing is not possible: extensive skin disease, patient on antihistamines, severe reactions, small children, dermographic patients. Testing MUST be guided by the history—do not request ‘allergen screen’.
Presence of specific IgE does NOT equate to allergic disease—indicates sensitization only. Must be interpreted in the context of the clinical history. Numerical value does NOT correlate with severity of clinical reactions.
RAST tests are of little value for identifying allergy to fruits and vegetables, as the allergens are labile, and to drugs (unreliable). False +ves possible when total IgE is very high due to non-specific binding (less of a problem with newer assays).
Normal range: 2–14μg/L.
Measured by EIA. Analyte is stable in clotted blood.
Valuable test for the investigation of acute? allergic reactions. Released when mast cells degranulate, and stable in serum for up to 24h. Also useful for monitoring patients with mastocytosis.
Raised levels indicate mast cell degranulation and will help distinguish anaphylactic and anaphylactoid reactions from other causes of reactions (vasovagal, hyperventilation, carcinoid, phaeochromocytoma, etc.). Persistent elevated levels may indicate mastocytosis.
Investigation of severe drug allergy is a specialized field and all patients should be referred to an appropriate expert for an opinion, usually a consultant in allergy or clinical immunology in a regional centre. Testing will usually involve skin prick testing, followed by intradermal testing and patch testing and, if necessary, blind challenge.
Unit: scored qualitatively.
Normal range: −ve.
This test identifies cell-mediated reactions = delayed-type hypersensitivity (type IV reactions). It should not be confused with skin prick testing. Allergens in petrolatum jelly are placed in contact with the skin for 48h under occlusion with aluminium cups. The test result is read at 96h, looking for eczematous change and blistering. Usually carried out by dermatology departments.
Investigation of contact reactions, e.g. eczema.
Positive results are invariably significant. Common allergens include metals such as nickel and chromium, dyes and chemical in leather, rubber chemicals (accelerators), and cosmetic chemicals. Panels of allergens used, depending on the clinical history.
Investigation of cellular function of lymphocytes, neutrophils, macrophages, and natural killer (NK) cells are restricted to specialized regional immunology laboratories. The tests are labour-intensive and difficult to standardize, with the exception of basic lymphocyte markers. EQA schemes are available only for basic lymphocyte markers.
All tests, other than basic lymphocyte markers, should only be requested after discussion with a consultant immunologist. Their role is in the investigation of suspected cellular immunodeficiency, particularly SCID and 1° disorders of neutrophils. In such cases, urgent referral of the patient to an appropriate paediatric or adult immunologist is more appropriate than fiddling around trying to get tests done, as the immunologist will have direct and immediate access to the appropriate tests. Lives have been lost due to delay in transfer, whilst inexperienced clinicians have tried to make diagnoses.
Unit: cells/μL.
Normal range: see Table 4.9.
Table 4.9 Normal ranges for lymphocyte surface markers
| CD3+ (total T cells) | 690–2540 |
| CD19+ (total B cells; CD20 is equivalent) | 90–660 |
| CD3+CD4+ (T helper cells) | 410–1590 |
| CD3+CD8+ (cytotoxic T cells) | 190–1140 |
| CD16+CD56+ (NK cells) | 90–590 |
Lymphocyte surface markers should be carried out on a single-platform flow cytometer, which will give a direct absolute count, not requiring a total lymphocyte count from a haematology analyser. Absolute counts are the preferred value; percentages are not useful. Fresh samples are required for optimum results. Many other surface markers are available to answer more specific immunological questions, but these will usually be of interest only to clinical immunologists. An EQA scheme operates.
There are no absolute indications. Investigation of lymphocyte subsets is an important part of the work-up of any patients with suspected 1° or 2° immunodeficiency and of patients with unexpected lymphopenia. Serial measurements are valuable in patients undergoing BMT or stem cell transplantation, those with 1° immunodeficiencies, those on any immunosuppressive therapy, and those with HIV on therapy with highly active antiretroviral therapy (HAART).
Results can only be interpreted in the context of the clinical question. Lymphocyte surface marker analysis CANNOT be used as a surrogate for HIV testing, as many acute viral and bacterial infections, as well as other medical problems, will give rise to a reduction in CD4+ T cells.
Baby <6 months with lymphocyte count <2 × 109/L = SCID until proven otherwise: IMMEDIATE referral to a SCID BMT unit (in the UK, Newcastle General Hospital, Newcastle upon Tyne, and Great Ormond Street Hospital for Sick Children, London). Look at the differential white count, not just the total white count!
These are highly specialized and should only be carried out on the recommendation of a clinical immunologist.
There are no absolute indications for testing, but it is usually carried out as part of the specialized work-up of patients with known or suspected SCID and in the monitoring of immunological reconstitution post-BMT.
Is complex and dependent on the precise clinical circumstances.
This is a specialized test. Samples do not transport well, and results are often abnormal if the patient has an active infection or is on antibiotics. It is preferable to refer patients to a clinical immunologist who will organize testing if appropriate.
Patients with deep-seated abscesses, recurrent major abscesses (exclude diabetes, staphylococcal carriage, and hidradenitis suppurativa first), major oral ulceration, and unusual fungal or bacterial infections (Pseudomonas, Serratia, staphylococci, Aspergillus). Atypical granulomatous disease, including atypical Crohn’s disease. Genetic defects of neutrophil function may present at any age.
Interpretation is complex; defects of oxidative metabolism may indicate chronic granulomatous disease; defects of phagocytosis are recognized. Also MPO deficiency. Neutropenia may be chronic or cyclical. Genetic testing to follow up may be required.
Is rarely required. The main indication is in recurrent severe infection with herpesviruses. Assays are complex and need specialist interpretation. Always discuss cases with a clinical immunologist.
Shoedfeld Y, Meroni PL, Gershwin ME (eds.). Autoantibodies, 3rd edn. Sand Amsterdam: Elsevier, 2014.
Spickett G.Oxford Handbook of Clinical Immunology and Allergy, 3rd edn. Oxford: Oxford University Press, 2013.
