The recent introduction of new diagnostic tests such as the antigen detection test, more reliable now than at the start of the pandemic, has expanded the number of tools available for the detection of SARS-CoV-2 infections. With this, the need for guidelines that help to decide which test should be performed in each particular case and how to interpret its results increases.
Different environments need different tests
There are many situations in which it is necessary to apply diagnostic tests but, in general terms, we can propose the following scenarios:
-Massive screening of the asymptomatic population.
-Screening of the high-risk population (for example, in nursing homes and health personnel).
-The investigation of the close contacts that an infected person has had.
-The clinical diagnosis of a symptomatic person.
-Monitoring the severity of the disease once diagnosed or the duration of the infectivity of the patient.
-The population serological study of past infections.
To date, no diagnostic test meets the requirements to be applied reliably in each and every one of these scenarios. This situation has generated much confusion in the interpretation of the results obtained by the different tests in each of these scenarios. Not only among the population but sometimes among the health workers themselves.
Each test, with its limitations, may be useful in a specific setting and it is necessary to know them in order to make the appropriate clinical decisions based on its results.
The following table shows a summary of the clinical situations where each test can be applied and which, within the recommended applications, are those situations where there is a greater probability of obtaining a false positive (a positive in uninfected people) or a false negative (a negative in people who are infected). All this assuming that there have been no errors in the taking of samples, their transport and the preanalytical processing.
Tests that detect past or late-stage infection
Serological tests consist of the detection of antibodies (IgM, which indicates infection resolving, and IgG, which indicates past infection). They can be useful in epidemiological surveys at the population level in which you want to assess the prevalence of people who have been in contact with the virus.
Specifically, IgM antibodies appear 6-7 days after the onset of infection and greater positivity is detected at 15 days. Around day 20 from the onset of symptoms are no longer detected.
IgG antibodies appear approximately 15 days after onset of infection and confer probable immunity (although at present it is unknown for how long).
This is the case of the seroprevalence surveys carried out in Spain since Carlos III Health Institute and in which it was possible to know the prevalence and characteristics of the population that had been infected during the first wave of the pandemic, through the measurement of antibodies IgG.
However, these tests have a very limited use (if they have any) in the evaluation of active infection, despite the fact that for this purpose they are being applied erroneously in some autonomous communities and in other settings. In addition, they have important limitations, especially those related to the presence of false positive results due to their cross reaction with other viruses.
There are different techniques for the determination of antibodies: ELISA (Enzyme-Immunoassay) and CLIA (Chemo-luminescence) (reference tests for the determination of antibodies) and immunochromatography (or also called rapid test).
The test results of ELISA/CLIA they are quantitative. That is, the titer (or number) of antibodies present is indicated. On the other hand, the results of rapid tests are qualitative (presence or absence of antibodies).
The sensitivity and specificity is higher in the tests of ELISA and CLIA than in rapid tests. However, given the ease of performing rapid tests (capillary blood sample versus serum or plasma and less complexity in performing them), their use has become widespread, especially in private laboratories, despite the greater probability of results. false negatives and positives.
Tests that detect active infection
Among the tests to detect the presence of the virus, the use of PCR – which detects the viral genome – has been established as the gold standard for detecting active infection.
Among its limitations, in addition to the complexity in terms of laboratory equipment, cost and time, we must highlight the false negatives that may depend on the onset of symptoms or viral load, as well as false positives depending on the characteristics of the environment in that are carried out and the temporal dynamics of the infection.
In general, in people with a low probability of being infected (as occurs in general population screenings) the probability of obtaining false positives increases.
On the other hand, a positive result weeks after the onset of symptoms may be due to the detection of nonviable fragments of the virus in people who no longer have infectious capacity.
Within this category of active infection tests are the new antigen detection tests that are considered “fast and cheap”. They detect the presence of SARS-CoV-2 viral proteins and have the advantages of giving results in 15-30 minutes and being able to be performed outside of the clinical laboratory, in the area close to the patient.
Its reception by the population and the political class has been enthusiastic. However, its sensitivity (especially in asymptomatic populations) is lower than that of PCR, with a higher rate of false negatives, so a negative result in someone suspected of being infected requires confirmation with a PCR determination.
The WHO and the ECDC have recommended its use in settings where it is not possible to perform PCR or a rapid result is needed for clinical decision-making (isolation, hospitalization, initiation of specific treatment, etc.), even stating that they must be performed within 5 days from the onset of symptoms.
These tests are not recommended for the detection of infected persons among asymptomatic cases, since their diagnostic yield in this population is low. Studies in these populations in our country place its sensitivity between 45% and 57% (a study that raises it to 79%, but in a population with a very high prevalence of disease). In the case of symptomatic children, with a sensitivity of around 62%, CRP might also be preferable.
Pharmacy and self-administered tests
Some past infection tests (antibodies) have been marketed in community pharmacies for use by patients themselves after prescription. Likewise, some Autonomous Communities and pharmaceutical associations have opened the debate on conducting antigen tests in community pharmacies. A practice that already occurs in countries like France, a country with a health system very different from that of Spain.
At Spanish National Health System (not so much in private insurers) antigen tests are available in primary care centers and the need to refer to community pharmacies a test that can be done immediately at the center must be carefully considered.
However, in the exceptional situation that we are living, and with the need to expand the virus detection capacity, these are initiatives to be evaluated. Despite the discussions in the media about the competencies of each profession and the place where the tests are to be carried out, the important problem must focus on the fact that a diagnostic test requires a rigorous interpretation of its results based on the clinical situation of the patient. patient or the person in which it is performed. This is what must be ensured in each case.
The fact that a test can give false positives and negatives does not mean that it is not useful, but rather that it must be performed in the setting in which it is most useful and be interpreted taking into account the clinical information of the patient and the prevalence of infection in the scope of action.
Therefore, initiatives in which it is the patient himself who picks up the antibody test at the pharmacy to do it at home can lead to multiple confusing situations. These pose a risk to both individual and collective health.
Proper interpretation of imperfect tests
Much of the mess around Covid-19 tests stems from the confusion between asymptomatic and presymptomatic patients, the informative value of each test in the temporal dynamics of the infection and the false discourse of “the more the merrier.”
At the same time, it is forgotten that, as in any other disease, scientific medicine requires the use of the right test, in the right person and at the right time.
Beyond the confusion, the use of diagnostic tests for Covid-19 requires not forgetting some extremely important rules:
People with symptoms or who are in close contact with a Covid-19 case, even if they are asymptomatic, should be isolated and controlled by health care services. All this even if your test results, whatever they are, are negative.
A negative result of an antigen test (or a PCR) does not exclude the development of disease or the possibility of infection (especially in the days immediately following). Nor does it allow relaxing any measure of social distancing (masks, distance, gauges, etc.).
A positive antibody test is not a biological passport. It does not guarantee that a specific person has passed the infection or that they will not be able to contract it again, especially if it has been carried out through rapid tests.
Blanca Lumbreras: University Professor. Preventive Medicine and Public Health Area, Miguel Hernández University.
Salvador Peiró: Researcher, Health Services Research Area, FISABIO SALUD PÚBLICA, Fisabio.
Article published in THECONVERSATION.ES