Researchers at Semmelweis University conducted a study on two coronavirus antibody rapid tests and looked at how accurately they were able to detect infection, government-critical news portal Index.hu reported last week. According to the (unpublished) research, the examined tests returned positive results of only one-third of those actually infected, and showed 7-13% of those actually negative as false-positives. Thus, the results suggest that these coronavirus rapid tests used to detect antibodies in patients are simply not currently suitable for screening for coronavirus infection in themselves.
It is important to note that these rapid antibody tests are not the same that are used in the mass screening campaign led by Semmelweis University and three other Hungarian medical schools. In the initiative, they use a PCR test, which, currently to the best of the scientists’ knowledge, gives the most accurate picture of a person’s virulence and active infection.
Coronavirus: Mass Screening Suggested Long Ago to Only Come with Easing of Restrictions
Regardless, as the coronavirus spreads all around the world, and some people contract the disease without even showing symptoms, many people want to know if they were already infected with coronavirus or not – or maybe they are still infected – so the demand for the rapid antibody tests is huge.
These rapid tests are showing exactly what asymptomatic people want to know: whether there is an antibody in them to the virus which their body has produced during the infection. In addition, these tests give not one, but two results at once: both the IgM (immunoglubulin-M) antibody against the coronavirus, and IgG. These can be used to prove that the body’s immune system has already encountered the virus or that the patient is contagious.
However, the Institute of Laboratory Medicine at Semmelweis University conducted a study in which researchers evaluated two coronavirus antibody rapid tests available in Hungary and concluded that these two rapid tests, detecting specific IgM and IgG antibodies of coronavirus, are inadequate.
During the research, results of the rapid tests were compared with the results of PCR tests, meaning they knew exactly whether the people whose samples were tested were actually infected or not.
The sensitivity of the tests was only 33-35%, so only one in three patients marked positive by the PCR test showed infection with the rapid tests. On the other hand, in uninfected (PCR-negative) individuals, a high proportion of positive results were given by the rapid tests. Of the 100 uninfected individuals, 15–28 showed a positive result (i.e., the specificity of the test was 72–85%).
However, the real results of the tests were even worse. The clinical utility and accuracy of the test results are also affected by the frequency of infection. To put it simply, in a rare disease, even a very well-performing test gives a large number of false-positive results.
Therefore, if the rapid test is positive, a PCR test should be performed in all cases to confirm the infection, and only the latter can be used to make any diagnosis. At Semmelweis University, where patients have been continuously screened since March 11th, uses this method in all cases.
Although Semmelweis University only tested two types of these rapid tests, it cannot be ruled out that most of the antibody tests for coronavirus detection would give similar results.
A similar conclusion was reached by the author of an article published on the website of the Hungarian Academy of Sciences about the usability of tests developed for the detection of coronavirus in general: “Most of the tests to detect the SARS-CoV-2 virus, of which there are already nearly 300 in the world, are unusable. According to the reports of experts, there is also an imported test in Hungary, the predictive value of which does not reach 20%. In Hungary, the expertise needed to select a good test is given, but this expertise should be used at all levels of decision-making.”
Hungarian news portal atlatszo.hu also reported on the unreliability of the tests imported from China, and added that it turned out that the research was about the two tests used by the ambulance services. They claimed that so far, four types of rapid tests that are used in public health have been named, only one of which is truly reliable – which is not yet widely used.
featured photo: MTI/AP/Nirandzsan Sresztha