Is Accuracy the Price We’re Paying for Faster Results?
Speed or accuracy? Do we have to choose? Applying the right technologies and paying attention to basic sampling guidelines can reduce overall time-to-result without sacrificing accuracy.
When visiting a grain elevator facility, it is a common sight to see a number of trucks lining up in front of the grain reception point. During harvest season, up to 100 trucks can be lining up over any given day. The reception point has to make a quick decision on whether the trucks are carrying grain that falls below defined mycotoxin contamination limits and can therefore be admitted into the facility’s operations. In harvest years where contamination levels are high,every truck’s payload will be probed and analyzed for mycotoxins before admission into the facility. This has to be done quickly because having results available as soon as possible shortens the waiting time at the back of the queue significantly.
Quick assay turnaround is key but to what extent does it compromise accuracy? These are two important points we shall address next. To provide an answer, we shall first define the right method for such an application.
Every truck is probed and analyzed consecutively; hence a quick, easy-to-use, single sample test method is required. Lateral flow device (LFD) tests, also known as strip tests or dipsticks, are one of the fastest test methods available and a perfect fit for this scenario. Due to their simplicity, these tests can be performed by almost anyone with basic training and typically produce analytical results in just under ten minutes.
This enables quick decisions on whether a truck’s payload can be accepted into the grain elevator’s facility.
Published on:
Mycotoxin
How lateral flow devices work
For a quick LFD test result, a grain sample is ground and extracted with novel water-based extraction solutions, following a subsequent development time of as little as three minutes to form visible test and control lines. In combination with a reader, the lines are then used for quantification. During the three-minute LFD development,the sample extract will flow over the strip and mycotoxin specific antibodies will bind mycotoxins present in the sample.
These blocked antibody sites cannot then be used to bind to antibodies coupled to visible gold colloids to the LFD test line, resulting in a lower test line intensity. Therefore, the higher the mycotoxin concentration in a sample, the lower the test line intensity. The control line will bind any excess of antibody gold conjugates, producing a visible line proving that the LFD development was successful. Such a quick and affordable test result cannot be produced by any other commercial test method. Therefore LFD technology shows its unique strength in reception point analysis.
What else is out there?
Other mycotoxin rapid test methods, such as enzyme-linked immunosorbent assays (ELISA), are also available on the market. ELISA tests, however, require a calibration curve with each analysis which can take around 20 minutes to produce a final result. Hence, such methods are only feasible when multiple samples are analyzed in parallel using the same calibration curve.
Another group of mycotoxin test methods are reference test methods, including high performance liquid chromatography (HPLC) as well as liquid chromatography with tandem mass spectrometry detection (LC-MS/MS). These are typical laboratory methods associated with high accuracy and reproducibility. The drawback of such methods is that they require highly skilled staff and take up to one day to deliver an analytical result. Therefore, they are unsuitable for application in a grain elevator reception point.
The big question of accuracy
The faster an analytical method is, the more often concerns about its accuracy will surface. So how accurate are LFD tests? LFD technology and ELISA tests both use the antibody-antigen reaction detection principle in which a specific antibody detects a unique antigen, in this case a mycotoxin, with a given sensitivity. Both test systems are calibrated against certified reference materials and results are compared to accredited reference methods in thorough validation studies.
Nevertheless, some test methods may not fulfill the operator’s expectations.
What can go wrong?
- Correct sampling (see text box) is the key to producing accurate analysis results. Large lots of grain (e.g.full trucks or barges) must be analyzed for mycotoxins before they are accepted into a facility. It is crucial that representative samples are taken for the analysis.
- One of the most significant factors that can influence an analytical result is the commodity matrix. Ideally, test methods are developed to analyze any kind of food or feed sample for mycotoxin contamination. A limiting factor, however, is that different food or feed samples can have different components which may interfere with the analytical result, depending on their chemical structure and properties. Antibody-based rapid tests are often influenced by certain matrix components which can have major impacts on the antibody-analyte binding reaction, potentially yielding questionable results. Test kit manufacturers face the challenge of reducing or eliminating these interferences through, for example, specific antibody developments and matrix validations.
- Further sources of error can be environmental factors such as the ambient temperature during tests. The antibody-antigen binding reaction — the detection principle of all immunochemical tests — is temperature dependent. For accurate results, manufacturers have developed test incubators to keep temperatures constant, thereby eliminating interferences caused by fluctuating temperatures.
- Operators of analytical tests must be properly trained in order to administer the tests accurately. Different test methods require different operator know-how.The LFD tests are the easiest mycotoxin tests to perform. Basic training is enough to keep operator errors for such methods at a minimum.
Sampling error
The total error of an analytical mycotoxin text result is the sum of sampling, sample preparation
and analytical errors. The sampling error contributes the largest portion of error in the final result. This is due to the fact that mycotoxins are unevenly distributed within bulk grain samples, making it essential to take a representative sample before continuing with sample preparation and the actual measurement.
To get a representative sample from a bulk sample lot, it is necessary to take several incremental samples from multiple locations. These incremental samples are then combined to form an aggregate sample and after homogenization and further splitting by a mechanical divider, an analytical sample is taken for subsequent sample preparation. These procedures are described in official sampling plans by authorities, such as the European Commission (EC 401/2006 & EC 519/2014).
The large influence on the final analytical result makes it essential that the sampling error is kept to a minimum. The simplest way to achieve this is by increasing the number of sampling sites as well as the sample size of incremental samples within a sampling plan.
What is the test kit manufacturer’s responsibility?
What can a test kit manufacturer do to ensure that results produced with their test kits are accurate and reliable? First, a test kit manufacturer needs to educate users on the importance of correct sampling, as this is crucial to result accuracy.
Next, it needs to prove the accuracy of its test kit by conducting a validation study that follows USDA GIPSA or AOAC guidelines. The test kit can be submitted to these official bodies for an external third-party evaluation confirming the claimed performance of the analytical method.
For such third-party validations, a test kit manufacturer must include the validation data for claimed sample matrices to ensure that the test kit performs accurately on these different sample matrices. In validation studies, multiple sample sources, e.g. maize samples from different continents, are typically used to show the reliability of the analytical method.
To further improve the robustness of a test system,environmental factors, such as temperature, must also be eliminated or accounted for. This can be done by using an incubator so that the LFD development always occurs at a constant, controlled temperature.
Is accuracy the price grain elevators pay for fast results?
After observing the operations at grain reception points, it is clear that rapid single sample tests are the methods of choice. Lateral flow devices clearly fulfill all criteria and in addition, reduce many sources of handling errors due to their simplicity.
By following clear and basic rules, speed doesn’t have to come at the price of accuracy. Reliable and accurate results can be obtained in just under ten minutes and that is the key to efficiently analyzing and assessing grain from truck to truck at a reception point.