Due to the increasing number of COVID-19 cases in Austria, which are caused by the coronavirus SARS-CoV-2, the Austrian Society for Laboratory Medicine and Clinical Chemistry (ÖGLMKC) would like to summarize the cornerstones of laboratory diagnostics for COVID-19. The recommendations are also coordinated with the Austrian Society for Hygiene, Microbiology and Preventive Medicine (ÖGHMP). This summary corresponds to the current state of knowledge; new scientific knowledge of COVID-19 is currently being published according to sometimes greatly shortened peer review procedures and requires ongoing updating and a rational evaluation.
| ||CONTENTS |
|1. ||Diagnosis of a SARS-CoV-2 infection |
|1.1. ||Direct virus detection using PCR |
|1.1.1. ||Sampling and transportation |
|1.1.2. ||Sample handling in the laboratory for virus detection |
|1.1.3. ||PCR test |
|188.8.131.52. ||General |
|184.108.40.206. ||Molecular genetic basis |
|1.1.4. ||Assessment of PCR results |
|220.127.116.11. ||Positivity for only one PCR target |
|18.104.22.168. ||Negative PCR results |
|1.2. ||Direct virus detection using an antigen test |
|1.3. ||Indirect virus detection using an antibody test |
|1.3.1. ||Potential uses of antibody tests against SARS-CoV-2 |
|1.3.2. ||Evaluation of a positive antibody finding |
|1.3.3. ||Examples for the interpretation of the results of an antibody test |
|1.4. ||Definition of terms for the properties of diagnostic tests |
|1.4.1. ||Analytical sensitivity and specificity |
|1.4.2. ||False negative and false positive results |
|1.4.3. ||Clinical sensitivity and specificity |
|1.4.4. ||Positive and negative predictive value (prediction value) |
|1.5. ||Quality assurance |
|1.5.1. ||Legal basis |
|1.5.2. ||Assessment of diagnostic test |
|22.214.171.124. ||Validation and verification of laboratory tests |
|126.96.36.199. ||Practical information |
|188.8.131.52.1. ||Selection of subjects to ascertain clinical sensitivity and specificity |
|184.108.40.206.2. ||Insufficient number of cases and missing information about confidence intervals |
|1.5.3. ||External quality controls |
|2. ||General laboratory diagnostics for COVID-19 |
|2.1. ||Handling of blood samples and other body fluids from COVID-19 patients |
|2.2. ||Value of laboratory parameters with COVID-19 |
|3. ||Further literature |
|4. ||Contact details of laboratories for the detection of SARS-CoV-2 in Austria |
|5. ||Other |
1. Diagnosis of a SARS-CoV-2 infection
1.1. Direct virus detection using PCR 1.2. Direct virus detection using an antigen test
1.3. Indirect virus detection using an antibody test 1.4. Definition of terms for the properties of diagnostic tests
1.1. Direct virus detection using PCR
The laboratory diagnostic gold standard for the diagnosis of an infection with coronavirus SARS-CoV-2 is the direct virus detection from respiratory secretions by means of polymerase chain reaction (PCR) or other nucleic acid amplification techniques (NAT).
1.1.1. Sampling and transportation
Samples are typically taken using a nasopharynx smear or an oropharynx smear from the upper airway. Alternatively, if the clinical situation is appropriate, samples of the deep airways (induced sputum, tracheal secretion or bronchoalveolar lavage) can be used. Coronavirus SARS-CoV-2 was also detected by PCR in stool and in isolated cases in urine, blood / plasma and cerebrospinal fluid, but these materials are currently not recommended for the primary diagnosis of COVID-19 disease.
Coronaviruses are enveloped RNA viruses, the virus particles and their single-stranded RNA genome are therefore sensitive to surfactants and RNAses. For this reason, special requirements must be placed on sampling, storage and transport in preanalytics in order to avoid false-negative results. In the case of smears, it should be noted that suitable swabs and transport media are used for virus detection ("virus swabs" with the appropriate transport medium or, if necessary, dry sterile swabs with a small amount (1-3 ml) of sterile NaCl solution suitable for molecular genetic analysis). The selection of a suitable swab should be checked when establishing the test.
The samples should be sent to the laboratory as soon as possible. If sample storage is necessary, this can be done at 2-8 ° C for a maximum of 3 days. The sample dispatch must be classified as "biological substance, category B" of the UN-No. 3373 according to the specifications of packaging instruction P650.
1.1.2. Sample handling in the laboratory for virus detection
The handling of respiratory samples in the context of laboratory diagnostics of SARS-CoV-2 infections falls into the category "non-targeted activities" and should be restricted to specially trained laboratory personnel. Updated German-language guidelines for obtaining and handling potentially contagious sample material can be found on the websites of AGES (Austrian Agency for Health and Food Safety), the RKI (Robert Koch Institute) and ABAS (Committee for Biological Agents). In particular, we allow ourselves to refer to the recommendations for handling test material from Covid 19 positive / suspect patients in the laboratory of the Austrian Society for Hygiene, Microbiology and Preventive Medicine (ÖGHMP).
In a nutshell, non-targeted activities, such as sample preparation and preparation or inactivation for molecular biological tests (PCR) are carried out under the conditions of biological safety level 2 (BSL-2). All activities that can lead to the release of droplets or aerosols with SARS-CoV-2, e.g. the opening of sample vessels with respiratory material must be carried out in a class 2 safety workbench. In addition to the general safety precautions such as protective gowns and gloves, breathing protection measures (at least FFP-2; filtering face piece, fine dust mask) are recommended to wear safety glasses.
Targeted activities with SARS-CoV-2 (virus isolation, neutralization test or similar) may only be carried out by specially trained personnel in security level 3 (BSL-3) facilities.
1.1.3. PCR test
For the PCR detection of SARS-CoV-2, both validated in-house tests are established in medical laboratories and commercial test systems from different providers are available. Typically, viral nucleic acid (RNA) is extracted from the sample first. After reverse transcription (RT), a PCR (or another NAT such as isothermal amplification) is carried out from this to detect virus-specific nucleic acids.
Depending on the test system, the analysis steps are automated to different degrees. Roughly summarized, different categories of test systems can be distinguished:
Manual or semi-automatic RNA isolation and RT-PCR: different providers and in-house tests available, processing of several samples simultaneously (in batch), medium sample throughput, high molecular genetic expertise of the laboratory staff required.
Fully automated SARS-CoV-2 PCR: Individual providers have launched fully automated extraction / amplification systems for the detection of SARS-CoV-2. These devices can process many samples (e.g. 94 samples) at the same time and allow a high sample throughput. Such systems are particularly suitable for the efficient processing of samples in large laboratories.
Point-of-care systems: Individual suppliers have launched cartridge systems for the SARS-CoV-2 PCR. These complete systems are designed for ease of use and allow automated virus PCR in a point-of-care setting. For the individual sample, the time to the analysis result is short (less than an hour), which is why these systems are also referred to as (molecular-genetic) rapid tests. In addition to the analysis time of the individual sample, the sample throughput of the respective system must also be taken into account. Such systems are particularly suitable for the rapid analysis of individual samples in a point-of-care setting or outside the sampling times of large laboratories.
The availability of reagents for PCR tests and RNA extraction as well as smear systems has improved. Both medical laboratories and manufacturers of commercial test systems continue to work on increasing the analysis capacities. The ÖGLMKC explicitly endorses and supports collaborations to further increase the analysis capacity, but points to the need for adequate quality control of the SARS-CoV-2 PCR analysis to ensure valid and reproducible test results for all patients.
220.127.116.11. Molecular genetic basis
Established in-house and commercial test systems are mostly based on the detection of 2 gene sequences (targets) of SARS-CoV-2. Usually, a gene sequence is selective for the beta corona virus genus and a gene sequence specific for the Cladus SARS-CoV-2.
The test systems used so far usually detect 2 sequences of the following genes: N (nucleocapsid), E (envelope), S (spike) and RdRP (RNA-dependent RNA polymerase).
For example, the test published by the Charité virology in Berlin initially uses the E gene to detect sarbecovirus, a subgenus of the beta corona virus. This is combined with the detection of the RdRP gene, which is carried out in parallel with a reaction specific for SARS-CoV-2 and with a reaction that targets the entire subgenus sarbecovirus (Euro Surveill. 2020 Jan; 25 (3)). doi: 10.2807 / 1560-7917.ES.2020.25.3.2000045.). The test used by the Pasteur Institute in Paris is similar and detects the E gene of sarbecovirus and two sequences of the RdRP gene, which can be combined in a multiplex approach. According to the updated WHO Guidelines for Laboratory Analysis (03/20), if SARS-CoV-2 is more widespread in a region, a simplified workflow with amplification of only one specific region (e.g. PCR detection of the E gene) can also be used (https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/laboratory-guidance).
In addition, the validity of the test results must be ensured by an additional internal control, which checks the nucleic acid extraction and amplification as well as the integrity of the reagents.
The described analytical sensitivity of the PCR assays for SARS-CoV-2 is typically in the range of approx. 10 copies of viral nucleic acids per reaction. (J Clin Microbiol. 2020 Mar 4th pii: JCM.00310-20. Doi: 10.1128 / JCM.00310-20.; Euro Surveill. 2020 Mar; 25 (9). Doi: 10.2807 / 1560-7917.ES.2020.25 .9.2000173). The manufacturer's instructions for the respective assay must be taken into account and verified by comparative measurements before the test is started in the laboratory. If CE-IVD certified PCR assays for SARS-CoV-2 are available, the ÖGLMKC strongly recommends using them.
If the introduction of an in-house PCR test for SARS-CoV-2 is necessary in individual cases because CE-marked tests are not available or are not sufficiently suitable for answering the clinical question, the test performance must be thoroughly validated by the performing laboratory and document it.
1.1.4. Assessment of PCR results
The duration of detection of viral RNA in the nasopharynx secretion seems to be subject to large individual fluctuations and is 12 days (1-24 days) in the median, according to a series of cases. In > 80% of patients, the detection is positive for at least 7 days (JAMA. 2020 Mar 3. doi: 10.1001/jama.2020.3204). In some cases, a PCR positivity for >25 days was also described. A recent work describes a detailed time history of the viral load in different sample materials. At the beginning of the symptom, the concentration of viral RNA was high, decreased rapidly in throat swabs during the course of the disease and was typically detectable there for about two weeks, while in samples of the deep respiratory tract (induced sputum) and in the stool an extended viral excretion was observed (Nature 2020 April 1. doi: 10.1038/s41586-020-2196-x (2020)).
18.104.22.168. Positivity for only one PCR target
Of great practical relevance is that in some confirmed COVID-19 cases the virus RNA is only intermittently detectable (JAMA. 2020 Mar 3. doi: 10.1001/jama.2020.3204). This phenomenon may have pre-analytical (sample handling, experimental use of antivirals) and analytical reasons. Typically, however, it is necessary to be monitored in the late course of the disease, when the viral load in the nasopharynx secretion is low and is at the detection limit of the PCR method. When the load of the virus is low, correct sample collection is also particularly important for virus detection. Therefore, to document that virus excretion is no longer carried out, a twice-negative PCR test is carried out at intervals of recommended for ≥ 24 hours. (https://www.ecdc.europa.eu/en/publications-data/novel-coronavirus-sars-cov-2-discharge-criteria-confirmed-covid-19-cases)
If a PCR test is based on the amplification of two (or more) target sequences of SARS-CoV-2 and only one is positive while the other is negative, there are several causes to consider:
The amount of virus in the sample is at the detection limit of the test
A technical error in the analysis of the sample (both false positive and false negative possible)
A mutation in one of the target sequences of SARS-CoV-2
The presence of a coronavirus other than SARS-CoV-2
In these cases, the PCR result shall be evaluated in combination with the raw data (early or late amplification of the target) and the molecular genetic characteristics of the assay. If a technical error can be ruled out, in the current epidemiological situation it is recommended to consider the amplification of only one PCR target as a positive test result and thus as an indication of the presence of an infection with SARS-CoV-2. This is due to the consideration that only one human pathogenic beta coronavirus is currently circulating, that SARS-CoV-2 has a genetic diversity, that "weak-positive" or "non-evaluable" results may lead to uncertainties in reporting and regulatory procedures, and that any false-negative results could undermine adherence to isolation and quarantine measures.
22.214.171.124. Negative PCR results
To differentiate from this is that some test systems are primarily aimed at the detection of gene sequences of the genus betacoronavirus or the subgenus sarbecovirus, so that the specific detection of SARS-CoV-2 would require subsequent sequencing of the PCR product (https://www.who.int/docs/default-source/coronaviruse/peiris-protocol-16-1-20.pdf?sfvrsn=af1aac73_4) https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/laboratory-guidance.
If only one gene sequence can be detected repeatedly in a patient and the result is not at the detection limit of the test, further confirmation tests should be considered. Where appropriate, the specifics should be forwarded to a reference laboratory for SARS-CoV-2 after prior contact; the Austrian Reference Laboratory is the Center for Virology of the Medical University of Vienna.
A one-time negative PCR result does not 100% rule out SARS-CoV-2 infection. If there are reasonable grounds for suspicion of SARS-CoV-2 infection and initial negative PCR outcome, a re-sampling and examination should be agreed between the clinician and the laboratory physician. This recommendation is supported by series of cases according to which PCR may initially be negative in epidemiologically, clinically and CT morphologically defined suspected cases (Lancet. 2020 Feb 15;395(10223):514-523. doi: 10.1016/S0140-6736(20)30154-9. Radiology. 2020 Feb 12:200343. doi: 10.1148/radiol.2020200343.).
False-negative results may be based in particular on pre-analysis and are based, for example, on poor sample quality, long sample storage, improper sample transport or unfavorable timing of sampling, based on the course of the disease. Inhibition of the PCR or mutation of the virus may play a role in analytics (Clin Infect Dis. 2020 Mar 4. pii: ciaa203. doi: 10.1093/cid/ciaa203.). The former can be detected by using an internal control for nucleic acid extraction and amplification.
1.2. Direct virus detection using an antigen test
Direct detection of viral antigens requires the availability of a specific antibody in the test system. Individual manufacturers have launched antigen tests on the market, but we currently do not recommend diagnostic use of antigen-based rapid tests (see also WHO recommendation https://www.who.int/news-room/commentaries/detail/advice-on-the -use-of-point-of-care-immunodiagnostic-tests-for-covid-19).
1.3. Indirect virus detection using an antibody test
Immunological tests for serological examination detect antibodies against coronavirus SARS-CoV-2 in the blood of patients, which are formed as part of the patient's immune response against SARS-CoV-2. These detect antibodies in the blood of patients, which are formed as part of the patient's immune response to SARS-CoV-2. The time to detecting antibodies in the context of a SARS-CoV-2 infection varies and has not yet been fully investigated for many tests. In general, a seroconversion was described (with great individual variation) after about 10 days.
The available antibody tests differ in essential respects:
Antibody class: Most test systems detect either IgG, IgM or IgA antibodies against SARS-CoV-2. The kinetics of the antibody class in the course of the disease are different and also affect the clinical sensitivity of the antibody test, depending on the time of sampling. Typically, IgM and IgA antibodies occur more quickly, while IgG antibodies persist for longer. For SARS-CoV-2, however, a comparatively early appearance of IgG antibodies has also been described (Lancet. 2020 Mar 23; doi.org/10.1016/S1473-3099(20)30196-1.)
Antibody specificity: Viral proteins are used for antibody testing, to which the patient's antibodies bind. Which proteins of SARS-CoV-2 are used exactly as an antigen for antibody detection differs between the test systems and is not specified by some manufacturers. This can result in a decisive difference in the specificity of the test, since cross reactivity compared to other coronaviruses is generally possible. In addition to the known and rare coronaviruses SARS and MERS, infections with other low pathogenic human coronaviruses (HCoV-HKU1, HCoV-NL63, HCoV-OC43 and HCoV-229E) are common. If an antibody test has cross-reactivity against another coronavirus and a patient has undergone an infection with one, the serological test for SARS-CoV-2 may be incorrectly positive. For many antibody tests, the cross-reactivity of antibodies against other coronaviruses is currently insufficiently characterized, so the specificity cannot be conclusively assessed. False positive results are not only described by cross-reactivities to other corona viruses, but have also been observed in other (viral) diseases, autoimmune diseases or other conditions.
Type of test execution: The type of tests available varies significantly and ranges from classic laboratory tests (such as enzyme-linked immunosorbent assay (ELISA) or types of chemiluminescence immunoassays (CLIA)), which can be carried out with the appropriate equipment in medical laboratories and a simultaneous analysis multiple samples allow up to "rapid tests" which allow analysis of individual blood samples without additional equipment. The latter are often immunochromatographic "lateral flow" procedures, in the form similar to a pregnancy test strip for home use.
Type of result: Depending on the type of test system, the result of the antibody test is purely qualitative (positive / negative) or quantitative. Quantitative tests potentially have the advantage of being able to assess the dynamics of the rise and fall of the antibody.
Currently, many of the antibody tests on the market are insufficiently clinically validated, so that there is often no robust data on the sensitivity and specificity in a specific clinical situation. For diagnostic use, it is crucial that these data have been collected in a collective of patients or healthy individuals relevant to the clinical question. For example, data on the clinical sensitivity and specificity of a test collected in COVID-19 patients in the advanced course of the disease cannot be used to determine a COVID-19 diagnosis at the onset of symptoms.
If the indication for an antibody test is given, then tests should be used which, according to the current state of the art, have the highest sensitivity, specificity and precision and whose performance data have been verified by the respective laboratory. The ÖGLMKC therefore recommends that serological tests for SARS-CoV-2 should only be carried out in medical laboratories and that patient-specific rapid serological tests should only be used in exceptional cases.
In summary, according to current knowledge, serological tests alone (without PCR) are neither suitable for diagnostic detection nor for the exclusion of acute infection by SARS-CoV-2. However, it may make sense to carry out anti-SARS-CoV-2 antibody tests in addition to PCR analyzes, especially if several serum samples are examined with a valid antibody test in the course of the disease in order to assess the seroconversion. The ÖGLMKC therefore recommends preserving serum samples for antibody tests in patients with high clinical suspicion of COVID-19. In addition, antibody tests are essential for epidemiological studies. The ÖGLMKC recommends continuing to refrain from the uncritical use of antibody tests.
1.3.1. Potential uses of antibody tests against SARS-CoV-2
According to current knowledge, serological tests for the detection of antibodies against SARS-CoV-2 are particularly suitable for epidemiological analyses of infections in the population.
o A false positive result of an antibody test (e.g. in cross-reactivity against other coronaviruses) could lead to overestimating the prevalence.
In the diagnosis of a SARS-CoV-2 infection in an individual patient, the antibody detection is inferior to direct virus detection by MEANS of PCR
o This applies in particular to the early stage of the disease. A false negative result of an antibody test, especially in the early stages of the disease, can lead to the possibility of SARS-CoV-2 infection being incorrectly ruled out in a patient. This would also be unavoidable in the context of a possible stage diagnosis (antibody rapid test before PCR), so such a virus is not currently recommended for the diagnosis of sarS-CoV-2 infection. As a result of a falsely negative antibody test in the early stages of infection, a false safety could promote spread (e.g. if no quarantine measures were taken as a result).
o Another problem is immunosuppressed persons, who do not form an antibody for serological detection. In them, a serological test against SARS-CoV-2 will be generally negative and is therefore not suitable for diagnosis.
o In the late phase of the disease, a positive antibody test complementary to PCR may be diagnostically helpful; especially if in the course of the disease several blood samples are present at a time interval with a documented seroconversion.
Antibody testing in addition to PCR is particularly useful if there is a high clinical probability of COVID-19. Serial antibody tests to assess seroconversion are recommended.
It is currently believed that after SARS-CoV-2 infection there is some immunological protection against renewed infection (however, the duration of immunity and the clinical extent is still unclear). In principle, antibody detection with a sufficiently specific test is suitable for the detection of a gone infection. At present, however, there is still too little meaningful data on which antibodies reflect effective immunological protection against a new SARS-CoV-2 infection at what level. Gold Standard for the detection of immunologically active antibodies against SARS-CoV-2 is an neutralization test. Whether and how well the antibodies detected with a particular serological test correlate with a virus-neutralizing effect is still insufficient evidence for most tests.
o A false positive result of an antibody test (e.g. in case of cross-reactivity against other coronaviruses) can result in an incorrectly accepting immunity to SARS-CoV-2 for a person. If no protective measures are in place for such a person, there is a risk of infection with SARS-CoV-2 and as a result of transmission to contact persons.
o In recovered COVID-19 patients (PCR-assured diagnosis), serological tests can be used to assess the extent of antibody formation. This is particularly of direct relevance when plasma preparations from recovered people (as part of studies) are used for the passive immunization of sick people. Classic antibody tests can be used as a pre-screening. The gold standard for the detection of immunologically active antibodies is the neutralization test.
1.3.2. Evaluation of a positive antibody finding
With the test systems currently available, a positive antibody finding should be interpreted with great caution and in the overall clinical context. The main problem is that in most of the currently widespread tests false positive results are repeatedly observed (at least approx. 1 to 3%, in some tests also significantly more; Preprint from medRxiv 2020 March 20. doi: 10.1101 / 2020.03.18.20038059 ). This limitation of specificity is partly due to the cross-reactivity of antibodies against other pathogens. Some manufacturers have launched or announced antibody tests with a false positive rate of significantly <1%. Given the wide availability of new, more specific tests, a timely reassessment of the use of antibody tests in COVID-19 diagnostics can be expected.
The highest significance is a positive antibody test when the seroconversion is documented in the course of the disease.
If the clinical presentation and imaging clearly speak for COVID-19 (i.e. a correspondingly high pre-test probability) and direct virus detection by PCR testing has not been performed or performed too late (and is therefore negative), a positive antibody finding may confirm the suspicion of COVID-19.
In asymtomatic individuals or patients with atypical clinical presentation, the significance of a positive antibody test (without documenting seroconversion) is low. The positively predictive value of a test result depends above all on the specificity of the test and the prevalence of the disease. As of 09.04.2020, the frequency of COVID-19 cases with 148/100,000 inhabitants or 0.15% of the population is documented in Austria. The number of unknown infections is not exactly known, but even assuming a very high number of unreported infections and a 10 times higher frequency (the latest estimates assume a lower dark number), there is a maximum prevalence of 1.5% for COVID-19 cases (total illnesses including recoveries). In a hypothetical antibody test with a sensitivity of 100% and a specificity of 98.5%, taking into account the prevalence for the general population, a positive predictive value of 50% results. In other words, even under ideal conditions, the test is at least as often as often falsely positive as it is really positive. In reality, this rating is much worse for most antibody tests. The above figures apply to IgG-based tests, with IgM or IgA inclusion, the specificity of the test is often significantly lower. The value of a positive antibody finding in asymtomatic individuals or in patients with atypical clinical presentation is therefore critical. In order to be able to assume with the current prevalence of COVID-19 in Austria with the presence of a positive antibody test (also in IgG) with a high probability that a person has undergone an infection with SARS-CoV-2 in the past, the test would have to have a specificity of well over 99% and this would have to be ensured by a sufficiently large study.
In order to also make a statement on immunity, the antibody test should be validated against a gold standard for the detection of immunologically active antibodies against SARS-CoV-2 (neutralization test). For the LIAISON® SARS-CoV-2 S1 / S2 IgG test, the company Diasorin states a match with the results of a plaque reduction neutralization test in patients with confirmed COVID-19 disease. Samples from patients without COVID-19 disease (false positive results) were not considered in this evaluation. The ÖGLMKC therefore expressly recommends that the positive result of a currently available antibody test on its own (without PCR-secured COVID-19 disease) not be interpreted as evidence of immunity to SARS-CoV-2.
Since SARS-CoV-2 was only discovered comparatively recently and reliable data on patients are missing, it is not yet known what the clinical extent of immunity and how long it lasts. In a statement dated April 24, 2020, the WHO stated that there is currently no reliable evidence that people who have recovered from a COVID-19 disease and have developed antibodies are effectively protected against a new infection (https://www.who.int/news-room/commentaries/detail/immunity-passports-in-the-context-of-covid-19). The ÖGLMKC therefore recommends that the detection of an immune reaction - regardless of the test and specific methodological limitations - not yet be interpreted as evidence of complete immunity to SARS-CoV-2.
1.3.3. Examples for the interpretation of the results of an antibody test
In the interpretation of the findings of an antibody test, the limitations according to the current data situation should be explicitly mentioned. An example of the interpretation of a positive or negative anti-SARS-CoV-2 IgG test is given below. Without prejudice to this, an individual interpretation of the findings can and should of course also take place. This can also take into account the respective prevalence of COVID-19.
Anti-SARS-CoV-2 IgG test positive:
The positive test result may indicate contact with the SARS-CoV-2 virus. Cross-reactivity with other viruses cannot be ruled out at this time. With regard to immunity, no statement can be made at this time due to the current data situation.
A positive IgG test does not safely rule out an active infection. In case of appropriate clinical symptoms or other suspicion of an active infection, we recommend conducting a SARS-CoV-2 PCR test of respiratory material (nose-throat smear) to determine the infectivity.
Anti-SARS-CoV-2 IgG test negative:
A negative test result does not safely exclude contact with the SARS-CoV-2 virus. The formation of IgG antibodies takes place with high reliability only 3-4 weeks after pathogen contact. In case of appropriate clinical symptoms or other suspicion of an active infection, we recommend conducting a SARS-CoV-2 PCR test of respiratory material (nose-throat smear) to determine the infectivity.
1.4. Definition of terms for the properties of diagnostic tests
1.4.1. Analytical sensitivity and specificity
The analytical sensitivity and specificity of a test describe the suitability of a test to detect a particular analyte (e.g. SARS-CoV-2 nucleic acids) in a sample. These characteristics of a laboratory test can be determined by a pure technical validation of the test.
The detection limit of a method describes the lowest concentration of the analyte, which can be reliably detected by the test. A test with a low detection limit can detect low concentrations of the analyte and has a high analytical sensitivity.
The analytical specificity of a test describes the ability of the test to detect only the desired analyte and not to be influenced by other substances in the sample (in addition to general interference factors, for example, other coronaviruses).
1.4.2. False negative and false positive results
Differences between the actual presence of a disease and the result of a laboratory test are called false negative or false positive results.
False negative people are sick patients, which the test mistakenly classifies as healthy.
False positives are actually healthy individuals, which the test mistakenly classifies as sick.
Really negative people are healthy people, which the test correctly classifies as healthy.
Really positive people are sick patients, which the test correctly classifies as sick.
1.4.3. Clinical sensitivity and specificity
The clinical sensitivity and specificity of a test describe the suitability of a laboratory test to distinguish the sick and the healthy. These characteristics of a laboratory test can only be collected by clinical validation of the test with patient samples and are only valid for the clinical situation for which the patient group used is representative.
The diagnostic sensitivity describes the proportion of correctly positive test results in patients with a disease and is usually expressed in %. A test has a high diagnostic sensitivity when few false negative results occur.
The diagnostic specificity describes the proportion of correctly negative test results in healthy people and is usually expressed in %. A test has a high diagnostic specificity when few false positives results occur.
1.4.4. Positive and negative predictive value (predictive value)
In addition to its diagnostic sensitivity and specificity, the predictive value of a laboratory test also depends on the frequency or prevalence of the disease in the population. As a result, the predictive value of the same laboratory test changes when the frequency of a disease such as COVID-19 increases significantly in the population.
The positive predictive value is the probability of a disease if the laboratory test is positive (pathological). The higher the clinical specificity of a laboratory test and the more frequent a disease occurs, the higher the positive predictive value of the test.
The negative predictive value is the probability that a disease can be ruled out if the laboratory test is negative (normal). The higher the clinical sensitivity of a laboratory test and the less frequently a disease occurs, the higher the negative predictive value of the test.
1.5. Quality assurance
1.5.1. Legal basis
The testing of sample material for human diagnostic purposes with test kits and devices is subject to the Austrian Medical Devices Act (MPG) and its aim is to ensure the safety and high-quality care of patients and society with medical devices and laboratory diagnostic tests. Compliant, high-quality and therefore legally permissible testing therefore requires
- CE marked test kits or
- test kits correctly validated in-house by the laboratory or
- by means of an exemption from the Ministry of Health in accordance with section 32 (1) at the request of the manufacturer or test kits approved on the basis of section 113a
The requirements of the MPG, including quality assurance, apply to every laboratory that carries out tests for patients, regardless of the legal status or any crisis situations, and are not overridden by the Epidemic Act or related regulations. The Austrian Medical Devices Act is based on the EU Directives 93/42 EG and 98/79 EG, within the framework of which national law must comply. Quality assurance in medical laboratories in the established area is determined by an ordinance of the Medical Association (approved by the Ministry of Health). Quality assurance in medical laboratory analysis in the hospital sector is included in the Cure and Hospital Act. The ÖNORM K 1950 gives a qualified recommendation for the concrete implementation based on the international standard EN ISO 15189: 2014.
The relevant regulations on documentation for the traceability and liability of the medical laboratories are contained in part in the Medical Devices Act, in part in the Physicians Act and in the Health and Medical Institutions Act and are important for patients and authorities in order to be able to make inquiries and, if necessary, assert claims . The structural and quality requirements laid down in the regulations are not only legally required for the correct medical high-quality performance of laboratories, but also one of the most important requirements for the work of clinically active doctors, who rely on medically meaningful findings and accompanying expertise for their work are instructed on the patient.
The Agency for Health and Food Security (AGES) is legally obliged (§68 MPG) to monitor all laboratories that use medical devices (in-vitro diagnostics) to ensure compliance with these legal provisions and to act accordingly in the event of violations to ensure that patients are at risk and can be prevented from the public by faulty laboratory analysis.
1.5.2. Assessment of diagnostic test
In the current pandemic situation, the test kits available on the market (in vitro diagnostics) have either been approved with urgent approvals for the American market (FDA) and then CE-marked by the manufacturers, or are only available as research kits for which there is no comprehensible verification the quality has been achieved by the manufacturer, whereby manufacturers can now only be monitored very incompletely, especially with SARS-CoV-2 kits, and are actually de facto autonomous when applying the CE label. This means increased caution is also required when carrying out CE-marked tests. CE-marked tests for SARS-CoV-2 diagnostics in the laboratory are, according to the current legal situation, not subject to the need for an objective evaluation of the manufacturer's information by an independent notified body.
For research kits, each laboratory must assume full responsibility for the analysis and clinical usability of the findings (in-house test), i.e. carry out the validation / evaluation itself, including acceptance and transport of the samples intended for the test, which are otherwise carried out by the industrial manufacturers in complex test procedures. The laboratory must document that the performance requirements in Annex I of EU Directive 98/79 EC have been met and that the documentation can be submitted to the authority if necessary.
The European Commission has summarized the current status of the literature on performance data from SARS-CoV-2 Test: "Current performance of COVID-19 test methods and devices and proposed performance criteria - Working document of Commission services" (Link: https://ec.europa.eu/docsroom/documents/40805). This document also emphasizes that for many of the laboratory tests there are no independent studies to assess the performance data.
126.96.36.199. Validation and verification of laboratory tests
Every laboratory test that is to be used for diagnostic purposes on human samples must be checked in advance by the laboratory for its suitability. This also applies to CE-marked tests. The EN ISO 15189: 2014 standard differentiates between validation and verification of the test procedure.
Verification: validated testing methods from manufacturers (i.e. CE-marked tests) that are used without modification must be subjected to verification before being introduced into routine use. The performance characteristics of the test specified by the manufacturer must be verified by the laboratory, in particular it must be documented that the performance characteristics for the specific application in the respective laboratory are also achieved.
Validation: Assessment of the performance characteristics of an examination procedure or laboratory test using a - usually more complex - evaluation procedure in the laboratory. Validation is used if there are no valid performance characteristics of an examination method validated by the manufacturer or if the method has been significantly modified in the laboratory. In particular, validation must be carried out for:
- non-standardized procedures (non-CE-marked test, especially all "research use only" tests or research kits);
- procedures designed or developed for the laboratory (in-house tests);
- standard procedures that are used outside their intended scope (CE-marked tests that are used for a purpose for which the manufacturer has not validated them);
- validated and then modified procedures (CE-marked tests that are used in the laboratory with significant modifications to the manufacturer's information).
188.8.131.52. Practical information
In the case of CE-marked laboratory tests for SARS-CoV-2, the manufacturer is obliged to provide the performance data of the test. The quality of this information varies widely among the tests currently available and is currently not being verified by an independent notified body. It is therefore the responsibility of the laboratory to check the quality of the performance data specified by the manufacturer and to assess whether the validation was carried out by the manufacturer according to scientific standards. In addition, the achievement of the performance data must be verified in the respective laboratory (see validation and verification of laboratory test). In practice, there are common problems with manufacturer information, which will be dealt with as examples.
184.108.40.206.1. Selection of subjects to ascertain clinical sensitivity and specificity
Ideally, clinical sensitivity and specificity of an examination procedure should be ascertained on a patient who has been coughed up, which is representative of the clinical question and also allows an assessment of the positive or negative predictive value of the test. In practice, due to the limitations of the test-independent definition of COVID-19 patients, almost all manufacturers use two independent patient cohorts to ascertain the clinical sensitivity and specificity. For example, the clinical sensitivity of an anti-SARS-CoV-2 antibody test in patients with PCR-related COVID-19 disease is assessed. Irrespective of this, sera from test persons who had been archived before the occurrence of SARS-CoV-2 are used to determine the specificity. This procedure makes sense in the current situation, but has certain limitations; in particular, no direct conclusions can be drawn from such studies that include the prevalence of SARS-CoV-2. It is expressly forbidden to use the sum of these two cohorts to calculate a positive or negative predictive value of the test, as is done by individual manufacturers. Such an approach would equate the prevalence of COVID-19 with that resulting from any combination of these two cohorts.
When assessing clinical sensitivity, it is particularly important that the patient cohort be representative of the respective question. If, for example, the clinical sensitivity of an anti-SARS-CoV-2 antibody test was tested exclusively in hospitalized patients in a late stage of the disease, it is inadmissible to apply these results to outpatients in the early phase of the disease at the onset of symptoms without further testing. Several manufacturers have therefore gone over to specifying the clinical sensitivity depending on the stage of the disease.
220.127.116.11.2. Insufficient number of cases and missing information about confidence intervals
"Current performance of COVID-19 test methods and devices and proposed performance criteria - Working document of Commission services" (Link: https://ec.europa.eu/docsroom/documents/40805), emphasizes the need for 95% confidence intervals for the Specify the results of diagnostic sensitivity and specificity of a laboratory test, which was examined in a clinical study using a suitable cohort. In this context, the number of cases of the examined subjects is significant. Figure 1 shows the relationship between the number of cases and the width of the confidence interval in a test with 95% specificity according to Bruderer (https://onlinelibrary.wiley.com/doi/epdf/10.1111/j.1553-2712.1996.tb03538.x).
Figure 1:Example of the relationship between the width of the 95% confidence interval and the number of cases, with the following assumptions: 95% specificity, error probability α=0.05, 1% prevalence
For example, if a manufacturer checks the diagnostic specificity of an Anti-SARS-CoV-2 antibody test using only 50 subjects and no false positive result is shown for these 50 samples, the validity of this test evaluation is relatively low, although the specificity is nominally 100% . Individual manufacturers have evaluated anti-SARS-CoV-2 antibody tests in more than 1,000 subjects, which leads to more statistically meaningful results with a narrow confidence interval.
1.5.3. External quality controls
Round robin tests are an essential means of external quality assurance. For this purpose, samples are sent from an external point to the participating laboratories. The interlaboratory test samples are to be used and processed in the laboratory like patient samples. The results are reported back by the laboratory and assessed by the proficiency testing manager. Both SARS-CoV-2 PCR tests and anti-SARS-CoV-2 antibody tests are now available in round robin tests from different providers. The interlaboratory test samples are to be used and processed like patient samples. The results are assessed by the proficiency testing management. The ÖGLMKC advocates mandatory participation in proficiency testing for SARS-CoV-2 tests. For all laboratories that carry out SARS-CoV-2 analyzes on humans without exception, sanctionable regulations to ensure the required quality in the use of in-vitro diagnostics are essential.
2. General laboratory diagnostics for COVID-19
2.1. Handling of blood samples and other body fluids from COVID-19 patients
In addition to respiratory secretions, coronavirus SARS-CoV-2 was also detected in the stool and in isolated cases in urine, blood/plasma and cerebrospinal fluid by means of PCR. Although no contagion via these bodily fluids (urine, blood/plasma and cerebrospinal fluid) is documented, such a must not be ruled out with the last certainty at present. Based on the current data situation, ÖGLMKC therefore recommends that general hygiene and protective measures be taken into account when working with blood samples from COVID-19 patients and that unnecessary aerosol formation during sample handling should be avoided. For stool samples of COVID-19 patients, the potential risk of infection is currently assessed higher, so the ÖGLMKC recommends reducing the analysis of stool samples from COVID-19 patients or suspected cases to an absolutely necessary minimum and taking additional protective measures to minimize the risk of potential infection. In the course of a laboratory request, the laboratory must in principle be notified by the sender that the submission is samples of a COVID-19 patient or a COVID-19 suspected case.
In detail, we allow ourselves to refer to the recommendations for handling test material from Covid 19 positive / suspect patients in the laboratory of the Austrian Society for Hygiene, Microbiology and Preventive Medicine (ÖGHMP):
2.2. Value of laboratory parameters with COVID-19
Clinical and laboratory chemical data have now been published for the first patients of the COVID-19 epidemic in China. This showed that some laboratory values in COVID-19 are frequently changed (see summary in Lippi G, Plebani M. Clin Chim Acta. 2020 Mar 4. https://doi.org/10.1016/j.cca.2020.03.004). In particular, lymphopenia has been described as typical in most works. However, none of the routine laboratory tests are specific to COVID-19. Routine laboratory procedures cannot therefore replace the specific virus detection by means of PCR.
Within the group of COVID-19 patients, in addition to clinical risk characteristics such as age and pre-existing conditions, an association with severe course and poor prognosis was also described for several laboratory values: lymphopenia, leukocytosis, increases in D-dimer, Troponin, creatine kinase, ferritin, interleukin-6, procalcitonin, lactate dehydrogenase, creatinine and alanine aminotransferase (ALT, GPT) (Zhou F et al. Lancet. 2020 Mar 9. https://doi.org/10.1016/S0140-6736(20)30566-3). In particular, the association of troponin with virus-induced heart muscle damage, the association of procalcitonin with secondary infections and the prognostic relevance of D-dimer in a multivariate analysis should be mentioned. However, the extent to which the data and experience from China can be directly relocated to the situation in Austria is not definitively clarified.
The ÖGLMKC recommends a structured collection and evaluation of laboratory data of Austrian COVID-19 cases in order to pool the experience of the different health institutions and to validate the prognostic significance of biomarkers.
3. Further literature
Federal Ministry of Social Affairs, Health, Care and Consumer Protection
Agency for Health and Food Safety (AGES)
Robert Koch Institute (RKI)
World Health Organization (WHO)
Austrian Society for Hygiene, Microbiology and Preventive Medicine (ÖGHMP)
4. Contact details of laboratories for the detection of SARS-CoV-2 in Austria
The ÖGLMKC has launched a website of specialist medical laboratories for SARS-CoV-2 diagnostics, which lists detailed information on the methodology used and contact data of the laboratories and offers a search and filter function.
You can also find other laboratories. See the homepage of the Agency for Health and Food Safety (AGES)
Note for laboratories:
Interested specialist laboratories who agree to the publication of their contact details on the website mentioned above, please contact the ÖGLMKC office. You can find more information on the website.
All laboratories that carry out SARS-CoV-2 diagnostics are urged to take part in round trials as part of external quality assurance. In Austria, the Austrian Society for Quality Assurance and Standardization of Medical Diagnostic Examinations (ÖQUASTA) offers round-robin tests for SARS-CoV-2 diagnostics.
Authors in alphabetical order:
- Christoph Binder, Clinical Institute for Laboratory Medicine, Medical University of Vienna
- Markus Exner, Labors.at - Mühl-Speiser-Bauer-Spitzauer and partner specialist for med. and chem. Laboratory diagnostics OG, Vienna
- Georg Greiner, Clinical Institute of Laboratory Medicine, Medical University of Vienna
- Andrea Griesmacher, ZCentral Institute for Medical and Chemical Laboratory Diagnostics, University of Innsbruck
- Alexander Haushofer, Institute of Medical and Chemical Laboratory Diagnostics, Klinikum Wels-Grieskirchen
- Gregor Hörmann, Central Institute for Medical and Chemical Laboratory Diagnostics, University of Innsbruck & Clinical Institute of Laboratory Medicine, Medical University of Vienna
- Harald Kessler, Diagnostics & Research Institute for Hygiene, Microbiology and Environmental Medicine, Medical University of Graz
- Georg Mustafa, Medilab Medical - Chemical Laboratory Dr. Mustafa Dr. Richter OG, Salzburg
- Manfred Nairz, Department of Internal Medicine II & Central Institute for Medical and Chemical Laboratory Diagnostics, Innsbruck University
- Matthias Perné-Mayerhofer, Medical Laboratory DDr. Johann Perné, Klagenfurt
- Franz Ratzinger, Ihr Labor ordination community for laboratory diagnostics and microbiology, Vienna
- Christian Schweiger, Clinical Institute for Laboratory Medicine, Medical University of Vienna
- Robert Straßl, Clinical Institute of Laboratory Medicine, Medical University of Vienna
- Thomas Szekeres, Clinical Institute of Laboratory Medicine, Medical University of Vienna
- Andreas Tiran, Laboratory Dr. Tiran, Graz
Central Institute for Medical and Chemical Laboratory Diagnostics, University Hospital Innsbruck & Clinical Institute for Laboratory Medicine, Medical University of Vienna
Postal address: Universitätskliniken Innsbruck, ZIMCL, Anichstraße 35, 6020 Innsbruck
Tel: +43 (0) 512 504 - 83673
E-Mail: email@example.com; firstname.lastname@example.org