Methodologies

• Most commercial clinical pathology laboratories will offer a range of ELISA-based techniques for a variety of purposes. 
• The general methodology for performing an ELISA test is described above. Each ELISA must be developed and standardized by determining the optimum combination of the various components to be used (eg antigen, primary antibody, secondary antibody, substrate), the optimum buffers to dilute these in, the optimum periods and temperatures of incubation, the optimum period of color development. 
• Between each stage of the test, the wells of the microtiter plate will be washed to remove unbound reactants this washing is relatively stringent and is usually performed with buffer containing a detergent substance. 
• The process of defining the conditions for performing the ELISA is known as chequerboarding where one reactant is varied (titrated) whilst the others remain constant. 
• The assays will generally involve an initial blocking step, where a source of irrelevant protein, eg skim milk powder, is added to the wells to block any free plastic surface to which the specific antigen has not adhered. This prevents the subsequent non-specific binding of other reactants. 
• Each ELISA protocol, including in-practice ELISA kits, must be closely adhered to for optimum results. 
• In a laboratory setting, the color that develops following the enzyme-substrate reaction is quantified by the use of a purpose-designed spectrophotometer (an ELISA plate reader). These are often linked to computer for very precise calculation of test samples versus a standard curve. 
• A standard curve should be run on each ELISA plate, and in the best ELISAs each test sample will be fully titrated to allow comparison of the slope of the standard curve with that obtained from the test sample. 
• Alternatively, the test samples may be tested at a single dilution (usually in triplicate) in a one point assay. 
• For kits designed for in-practice use, such precision is not possible; the company will generally provide some indication of how to score the relative intensity of color development and how to interpret this score.

Control

• Any one ELISA should be carefully validated by the laboratory or company providing the test. 
• This requires validation of the component reagents, and validation of the performance of the test in a field situation, with determination of sensitivity and specificity. 
• Most ELISA tests used in veterinary medicine are subject to this careful validation and are supported by published literature. 
• The reproducibility of an ELISA is generally assessed by establishing the intraplate and interplate coefficients of variation, which optimally would be <5%.

Availability

• ELISA tests are widely performed by commercial laboratories, and in-practice test kits of a variety of types are available.

Validity

Sensitivity

• ELISA is a highly sensitive test capable of detecting very small quantities of the target molecule (at least in the order of µg/ml). 
• A low/negative serology result may be obtained in the early stages of infection and paired serum samples may be required to identify recent exposure/infection.

Specificity

• Like all serological tests, the specificity of the ELISA is in part determined by the specificity of the reagents (antisera) used within it. 
• Many ELISAs are rendered highly specific by the use of monoclonal (rather than polyclonal) antibodies and by the use of highly purified recombinant proteins as target antigens. 
• However, as in all serological tests the discriminatory ability of some ELISAs may be less than optimal, eg if two related microbes share common antigenic epitopes, infection with one may lead to generation of antibody that cross-reacts with both organisms and may give a false-positive result in the ELISA. 
• Seropositivity indicates exposure a particular antigen but does not confirm disease; results must always be interpreted in relation to the clinical presentation and other diagnostic tests.

Predictive value

• This will be calculated for each individual assay.