Food Protection Trends - July/August 2024 - 303

and not the same as a negative or not-detected result from a
qualitative method using a larger test portion.
Qualitative methods are used to screen for the presence of
specified microorganisms, such as Salmonella or Listeria. These
methods generally use enrichment, capture, or other amplification
approaches to detect as few as one cell per sample test
portion. Because it is common for test portions to be 25-375
g, the qualitative method is nominally sensitive to between
1 CFU per 25 g and 1 CFU per 375 g of the target pathogen.
Compared with the 10 CFU/g sensitivity of the quantitative
indicator method, this represents a 250- to 3,750-fold increase
in sensitivity. Hence, a sample with an unquantified indicator
organism count that is below the LOD could test positive for a
pathogen with no errors in either method.
Comparing results from these two classes of methods
may indicate a misunderstanding of their capabilities and
purpose. The indicator test is intended to provide insight as
to the likelihood of a pathogen being present, but only the
qualitative pathogen test is designed to specifically detect it.
Different test portions
Even with sensitive and specific methods, pathogens can
be hard to detect. The low prevalence, low concentration, or
nonuniform distribution of pathogens creates reduced odds of
detection within the environment or foodstuff lot. Increasing
the number of test portions increases the likelihood that the
target organism will be detected if present. However, people
generally assume uniform distribution of microorganisms
across a decision unit (i.e., the product lot from which sample
portions are taken), whereas organisms are often randomly
distributed and may be found in hot spots within a product.
Therefore, retesting a product lot following a positive pathogen
test result is never permissible if the intention is to obtain a
different result and any decision from such testing will be used
to release products into commerce. The 2008-2009 Salmonella
outbreak associated with Peanut Corporation of America
products (28) and the egregious decisions of executives
associated with analytical testing results (4) illustrate the
potential dire consequences of such decisions.
Nonconfirming presumptive positives
For historical reasons, pathogen detections from rapid
screening platforms are formally called presumptive positive
results, or presumptives for short, because they are not based
upon the isolation of the target as a colony on a plate. These
presumptive results are often subsequently confirmed using
a cultural isolation method, such as those detailed in the
Microbiology Laboratory Guidebook from the U.S. Department
of Agriculture's Food Safety Inspection Service (30) or
the U.S. Food and Drug Administration's Bacteriological
Analytical Manual (33). Sometimes, the cultural method
fails to produce an isolated, morphologically typical colony
of the target pathogen that can be confirmed by biochemical
or molecular means. This cultural " not detected " result gives
a " nonconfirming presumptive result " by the screening
method, often considered a " false positive. " This discrepancy
can occur for numerous reasons, which are detailed in
subsequent sections, but both results could be correct and
nonconfirming presumptives are not always false positives.
Rapid screening tests usually use a cellular component (a
DNA sequence or a cell-surface antigen) as the test target,
serving as a proxy for the viable, infectious pathogen that
is the primary concern. When that test target is present but
associated with dead cells (e.g., after pasteurization), the test
may correctly detect its target, but cultural confirmation will
not yield an isolate. It is also possible that the LOD of a PCR
test is lower than the LOD of the cultural method, so analysts
should take care if calling a PCR result a false-positive result
when the cultural result is negative.
Ultimately, many strategies can be employed to investigate
the sample; however, all must be interpreted within
the context of what decision or decisions will be made. One
investigational approach for resolving such discrepancies is
to use a molecular method to " confirm " an isolate, even if it
is atypical in appearance. This involves testing an isolated
colony with a molecular method and is a validated procedure
for multiple molecular platforms. Ideally, the confirmation
platform would not be the one that gave the initial presumptive
result so that each platform would detect a different
DNA sequence associated with the target organism. Another
approach is to use an alternative method that employs a
detection chemistry different from that of the initial screening
method. For example, a result from a nucleic acid-based
method " confirmed " by an immunoassay could be considered
the strongest indication of a true positive without a cultural
confirmation, because in general, an immunoassay would
be less sensitive, giving a greater likelihood that it detected
an organism that grew during the enrichment culture. A
not-detected result using any rapid confirmation method
cannot be used to overturn the first presumptive result. This
is further discussed in Case Study 1 and stands in contrast to
cultural results, which can be used to overturn initial rapid
screening result presumptives. The inability to overturn
initial presumptive results may greatly limit the appeal of
molecular confirmation for many pathogen test users. Finally,
in cases with a presumptive positive from a rapid screen but
no growth on culture plates, we can consider a metagenomic
analysis of the enrichment to determine whether the sample
contains DNA from the target organism. If this is a rare
occurrence, cost would be outweighed by ensuring the safety
of the product before it is released into commerce. Confirmation
is not necessary to take corrective action but can support
effective corrective actions and, if necessary, modifications to
the food safety plan.
Matrix interference
As discussed in previous articles (5, 17), third-party
method validation studies, such as from AOAC International
July/August Food Protection Trends 303
Food Protection Trends - July/August 2024

Table of Contents for the Digital Edition of Food Protection Trends - July/August 2024
