Serological testing was developed for two major reasons, first to determine if a person had been exposed to an infectious agent in the past. Secondly to provide evidence of recent exposure to an infectious agent where no direct evidence for a specific pathogen could be found or growing the pathogen itself was difficult or dangerous.

The testing is based on the observed host immune response to infection. On first exposure to a pathogen the host produces antibodies of the IgM class. Over a period of days the immunoglobulin production switches from IgM to IgG and these IgG antibodies remain detectable for many years.. Initially only IgG could be measured for routine diagnostic purposes. A single sample of serum assayed for specific IgG against an infectious agent is useful for large-scale sero-epidemiological studies in populations to determine past exposure to disease or confirming the effectiveness of immunization programs. A single determination of specific IgG is also useful in situations where the immune
status needs to be known such as donors and recipients of organ transplants, exposure to an infection during pregnancy when the mother's immunity to the disease is unknown, or post exposure to a disease in which prophylaxis may be offered to those not immune. Many of the original methods for detection were cumbersome and required specific reagents such as one day-old chick red cells for rubella antibodies or guinea pig complement for complement fixation assays. Nowadays most tests of this type are done using enzyme immunoassay techniques (EIA) that lend themselves well to automation. Tests for IgG usually cost about $10 per antibody requested.

Diagnosis of recent infection used to be done by detecting a fourfold rise in levels of IgG antibodies using paired serum samples. This is now rarely done as the availability of specific IgM testing has taken the place of paired sample IgG testing also the EIA technology now used is not routinely designed to measure titres. Testing for specific IgM antibodies used to be available for only a limited number of infectious agents and the methodologies were time-consuming and very expensive such as sucrose density gradient ultracentrifugation methods for rubella IgM detection. Nowadays specific IgM antibody detection is available for a wide variety of infectious agents. The tests are expensive (5 to 10 times the costs of IgG tests) and are technically more difficult to perform. A few are automated but most are manually performed, time-consuming and require considerable technical expertise, such as fluorescent methods. Tests for IgM are sometimes useful but the tests are grossly over-utilized, particularly when multiple antibodies for both IgG and IgM are requested on a
patient. Some tests that are frequently ordered, such as Varicella or Toxoplasma IgM are rarely necessary.

Because of the nature of the testing methods and the time taken to perform the tests serological tests are not "stat" and are usually only performed on week days. (Some tests for transplant donors or other urgent situations such as needle-stick injuries can be provided on week-ends) Tests are done as soon as possible and results are often available on the same day. Most tests, because of low volumes, are "batched" in order to make the best use of the reagents and keep the costs of the tests as low as possible. Batches of tests are usually run once or twice a week.. Batching samples makes sense, a diagnostic kit may contain sufficient reagents to perform 50 tests. Each time the samples are tested there needs to be positive and negative controls run as well. A 50 test kit could run 48 patients samples (plus one positive and one negative control). If tests were done as single patient samples each sample would need a positive and negative control
so only 16 patients samples could be done with the 50 test kit. If the kit costs $1000 then batching 48 samples costs $20.80 per patient. If the tests are done singly it would cost $62.50 per patient sample. This of course does not take into account the additional technologist time required to do single tests. There may be some instances when it is necessary to do single sample determinations but that type of request needs to be kept to a minimum.
In some cases the volumes of test requests are so low that even batching samples does not make sense as the time to accumulate sufficient numbers to run the test may take weeks, often longer than the shelf life of the testing kit. In those cases we purchase the testing from an agency that receives sufficient samples to run the tests within a reasonable time frame. Laboratories have to be licensed to perform diagnostic tests and in some instances, for highly specialized serological testing such as arbovirus serology or HIV testing, only limited laboratories in the province are licensed to perform these tests and all such testing has to be done by these laboratories.

Another issue regarding serological testing that needs to be understood that the current testing methods are approved for use to detect IgG and IgM only in serum and the results from any other body fluid samples are not reliable. Negative and positive results are equally unreliable so it is not worth doing the test to "see if the result is positive". Requests to do serological testing on samples other than serum can not be accepted. There are some very specific tests that have been adapted to detect immunoglobulins in fluids such as CSF but these are research tools and not widely available.

Interpretation.

It needs to be appreciated that no single method is capable of providing absolutely reliable answers in all situations. With any lab test interpretation is much easier if the pre-test probability of the test being positive is substantial. Too often serological tests are ordered as part of a “fishing expedition” and a positive result for some disease that it is most unlikely the patient has merely causes confusion. Serological interpretations are based on the sensitivity and specificity of the test method and as no method is 100% sensitive and specific, certain choices have to be made. All serological tests are essentially screening tests and depending on the test, false negatives or false positives can occur. If the purpose of the test is to screen out all blood transfusions or organ donors who might have HIV then the test needs to have no false negatives. In order to do this a test is used to screen that is highly sensitive but some of the positives will be false positives as the test is not 100% specific. All positives, in this instance, need to be tested by a confirmatory test but using a screen like this no false negatives will occur. Confirmatory tests are usually done off site in large central facilities such as the Public Health Laboratories. It does mean that a few organs or blood donations will not be used that are later found to have been safe to use but at least no one will have been exposed to HIV by receiving tainted blood or organs. If on the other hand the test is being used to determine if a pregnant woman has been exposed to an infection that might cause fetal damage and pregnancy termination is a consideration then a test that has no false positives is desirable, this kind of test would ideally be highly specific but not 100% sensitive so a few false negatives will occur. In general it is "safer" to accept a few false positives and most screening tests licensed for use have "cut off" points that allow for false positives, all the tests available at LHSC/SJHC are of this variety. It does mean that a person may get an initial shock that they are HIV positive and then be told that they are in fact negative but it does make it safer for the recipients. Some tests like HIV are very sensitive and specific and very few false negatives or false positives occur, other tests such as for Lyme Disease are much less accurate. Some laboratories only report positive results on screening tests once they have been confirmed, we report the results of the screening test immediatly and then send a report of the confirmatory test result. Only some positive tests are sent for confirmation, such as HIV and some of the hepatitis tests, positive results from other serological tests are not sent for confirmation. The interpretation of any serological test needs to be made in light of the clinical picture and possible exposure. Doing serological tests in situations where the clinical history and exposure history does not fit with the suspected disease is much more frequently going to confuse, rather than clarify, the issue.

In many instances the test requisitions are incomplete and that means the lab has to make certain assumptions. In order to reduce unnecessary testing when an IgG test is ordered it is automatically checked to see if a previous result is available. If it is found, then the test will not be done and the report will state "previously pos ". It is understood that in some instances it is important to determine that antibodies are still present (individuals who have received multiple transfusions since the previous antibody testing) if this is the case then it must be indicated on the requisition.

Often the lab is asked to perform multiple IgG and IgM assays. When this is requested, IgG will be done first and IgM will be done only in those instances where the IgG is absent. For instance, if a request is received for EBV, CMV, TOXO and VZV IgG and IgM. We will do IgG for all and if IgG is absent for CMV but present in the rest then we will do an IgM on the CMV.
Sometimes testing is repeated on a sample using a different methodology to see if the presence of antibodies can be detected. These tests often measure total immunoglobulin so if the initial request was for IgG and it was negative and a repeat test for total immunoglobulin was positive then it may be that IgG was below the limit of detection with the first method or the positive result is due to the presence of IgM. Clinical judgement is required to interpret this type of test result. If a test that measures total immunoglobulin is used it will be indicated on the report.

Sometimes, despite the best efforts made by the lab, the test result is indeterminate and has to be reported as such. It means that there is no way to determine if the result is positive or negative. This can happen for a number of reasons, most frequently it is because the result obtained falls close to or at the “break-point” or “cut-off point” of the test. For instance, if the manufacturer states that a positive is >6 and a negative is <4 then a result of 5 is indeterminate. Occasionally the patient sample will contain non-specific inhibitors that prevent the test from functioning properly and again this may make the results of the test indeterminate. Whatever the reason an indeterminate test result is just that… indeterminate!…and asking the lab to either repeat the test on the same sample or to judge if the test is positive or negative is fruitless. A repeat sample at a later date can sometimes resolve the issue but again the probability that the patient actually might have the disease in question is critical.

There are a couple of other instances where interpretation of serological results is often problematic. The first is in new borns. It is often much more useful to investigate the mother, she has more blood available than the baby and her immunity (igG antibodies) are circulating in the babe in the early stage of life. Specific IgM antibodies can be measured in the child but this does require substantial quantities of blood. Often there are better methods of diagnosing acute infections in babies that do not rely on serological testing, (culture, rt-PCR etc).
The second instance is in patients who are immunocompromised eg post transplant, or in patients who have received large amounts of replacement fluids or multiple transfusions. In these instances serology is relatively useless. Immune status is much more accurately determined pre-transplant for instance.

It is now a required standard for all labs to indicate on the report what methodology was used to determine the test result. These are usually abbreviated to initials. Here is a short primer:
DFA… this is a direct fluorescent antibody test. (often used for rapid,preliminary results on a sample eg respiratory viruses). It is direct because the labeled antibody “sticks” directly on the target. While the tests are specific they are not very sensitive so a positive result is valuable a negative result is not. Immunofluorescent tests rely on the expertise of the observer and sometimes a weak positive will be interpreted as positive by some and negative by others. Positive results usually mean that a large number of infectious particles are present in the sample.
IFA…indirect fluorescent antibody. In this case it is a two stage test where the antibody has been produced in a particular animal (let’s say a goat). The antibody is allowed to react with antigen in the clinical specimen and sticks there. The slide is washed to remove any non-attached antibody and then another antibody to “goat” that is labeled with flourescein is added to the slide. This will attach to any of the antigen-antibody complexes remaining and the slide will be positive.
Immunofluorescent tests though very good and often cheap do have some observer variation at low levels and also are plagued by “non-specific” fluorescence in the sample.
EIA… This is an enzyme immunoassay. The endpoint is usually machine read and so is not subject to observer variation. These tests are largely automated and it is a very common methodology for many immunologically based tests.
Other tests that are rarely used are such things as CF tests,(Complement Fixation) that use the property of antigen-antibody complexes to “fix” complement. HI tests, (Hemaglutiination Inhibition)…again rarely used now but work on the principle of inhibiting agglutination of red cells.

In order to receive the most value from serological testing:

1. Order only those tests that are needed. Do not order "serology" or "TORCH". Specify which pathogens are of interest but don't order tests that the patient has had no identifiable risk of exposure to. Fill out the requisition completely and accurately.

2. Keep IgM testing to minimum and only request the specific pathogen you are concerned about. If, for instance, there is a concern that a patient in emergency, or your office, might have either EBV or CMV infection, as EBV is more common than CMV ask for EBV and request CMV only if EBV is negative. This can be put on the requisition and will be done by the lab.

3. When ordering, be sure to specify if past infection (immunity) or acute infection is being investigated, do not routinely order IgM and IgG on all samples.

4. Call the Microbiologist or Infectious diseases if in doubt about what tests to order.

5. Don’t order serology for diseases that have almost no clinical overlap. Believe it or not it is not uncommon to have a request for an acute disease investigation that includes Dengue, Chagas disease, Varicella, Lyme disease and Q fever serology in an 80 year old with no travel history or animal contact. It has to be possible to narrow the diagnosis down and if all these tests were to be done it would be an enormous waste of resources !! As it is, it takes a lot of phone calls to determine what tests really need to be ordered.

Reference. Clinical Virology Manual. Steven Specter, Richard L. Hodinka, Stephen A. Young. 3rd Edition. ASM Press. 2000.

The following is the list of infectious diseases serological tests available at LHSC/SJHC,. Hepatitis serology is available but on a different menu. All other infectious disease serology is referred out.

IgG:

Coxiella burnetti (Q fever)
Cytomegalovirus (CMV)
Epstein-Barr Virus (EBV) { Infectious mononucleosis}. Viral Capsid antibody (VCA) only. Early antibody (EA) and Epstein-Barr nuclear antibody (EBNA) can be referred out.
Herpes simplex virus (HSV)
Rubeola (Measles)
Mumps
Mycoplasma pneumoniae
Rubella (German measles)
Toxoplasma gondii
Varicella zoster virus (VZV) {Chicken pox}

IgM:

Coxiella burnetti (Q fever)
Cytomegalovirus (CMV)
Epstein-Bar virus (EBV) {infectious mononucleosis}. Viral Capsid antibody (VCA) only.
Herpes simplex virus (HSV)
Rubeola (Measles)
Mycoplasma pneumoniae
Toxoplasma gondii
Varicella zoster (VZV) {Chicken pox}