Posted: May 20th, 2012 - 12:38pm
Source: Leavitt Partners

Does rapid pathogen testing help or hurt?
With foods and ingredients moving across continents, the global nature of our food supply means that foodborne illness associated with a particular product can present at low levels throughout the United States or the world. PulseNet facilitates the identification of potentially related cases of foodborne illness, and the power of that technique—based on methodology that is now more than 20 years old—has been a game changer in the world of outbreak investigations. It looks like the game is about to change again.
In my position as Panel Manager for the USDA NIFA Small Business Innovative Research program, I’ve seen numerous proposals seeking funds to advance the development of rapid methods for the detection of foodborne pathogens. The benefits are obvious: the more quickly you can identify if a food is contaminated, the more quickly actions can be taken to ensure that the food does not reach consumers. However, the theme of the “Culture Independent Diagnostics Forum” meeting held in Atlanta, GA, in late April cast a bleaker vision of a world where rapid methods increase in prevalence—at the expense of culture-dependent methods that feed systems like PulseNet.
The roughly 150 attendees at a recent APHL/CDC/CSTE sponsored meeting on the topic of “culture independent diagnostics” were predominately those from the public health and clinical laboratories. Speakers painted a picture illustrating that the reliance on rapid clinical methods to confirm the diagnosis of illness in a patient meant that the longer, more laborious (and seemingly duplicative) confirmatory tests that yielded a culture would be abandoned. Without this culture—the spot on the plate—tests like Pulsed Field Gel Electrophoresis that yield the molecular fingerprints fed to PulseNet would cease to exist, and several suggested that we would revert to conducting investigations the way they were conducted 50 years ago.
In the past six years, PulseNet has identified outbreaks associated with 15 foods that had never before been associated with outbreaks. Perhaps some members of the food industry reminisce about the days that their products were considered “low risk” and yearn for the inability to definitively link pathogens in the environment, food, animals, and people. I sincerely hope that with the focus on prevention, industry would welcome the knowledge and insight gained by better understanding how different strains and serovars of pathogens move through the environment so they can better defend their food against contamination.
My observation is that these “culture independent” rapid methods are further developed in the clinical community compared to food applications where matrices and inhibitors present many challenges. This is unfortunate, since the food industry could benefit from such advances. While testing cannot assure that all products are safe, testing is an important step in verifying the effectiveness of preventive controls. Given that several foods associated with outbreaks are also highly perishable, it would be advantageous to be able to quickly test foods rather than expending precious shelf life time waiting for laboratory results. Often, a “presumptive” positive is enough to warrant the disposition of the product, and last week we reported that a “presumptive positive” of a regulatory sample is enough for FSIS to initiate a traceback investigation. Some food companies may have the resources to conduct follow up tests in the event of a positive finding, to better understand the issue, but this is the exception, not the norm. On the clinical side, in many cases, upon finding a positive rapid test, clinical labs will not follow up with a “reflex culture” (a secondary, culture-based test) because they will not be reimbursed by insurance companies for their effort.
There is a potential conflict in testing between the food world and the clinical world. While the food industry would benefit from more rapid methods, enabling tests to be conducted by less skilled staff than conventional food microbiology, it was pointed out that if tests become too easy to use in the clinical environment, the results of a test conducted in a doctor’s office may not be submitted to public health agencies. Clearly, outbreaks can’t be identified if data aren’t fed into the system. So what may be an advantage in the food world could have a negative impact on public health in the clinical world, regarding outbreak identification, yet could still provide good individual patient management.
There are concerns about the validity and interpretation of results of culture-independent methods by the food industry. Many Critical Control Points kill pathogens, and there is concern that some test methods may detect viable as well as non-viable organisms. Some could argue that there is value in knowing that the product initially contained pathogens, so that controls could be moved further upstream to prevent an initial contamination, but from the regulatory perspective, only a food containing viable pathogens able to cause illness are of concern.
Unfortunately, none of the companies who manufacture these tests presented their perspectives. However, after one panel discussion, a manufacturer did ask the obvious question: “As clinicians and public health professionals, what is desired in a test kit? Do you want differentiation down to a strain/serotype level, or do you want multiplexing?” There wasn’t a clear answer, as most presenters seemed content to continue the status quo in collecting cultures.
How would the food industry respond to this question? Are the needs different? Should the conversation that occurred within the public health community also take place jointly with the food industry? I would suggest that it should, given that there was but one food industry member present at the meeting. Reinforcement from the food community about the value of being able to differentiate strains of pathogens, which may rule you “in” or “out” of an investigation, should be voiced. If suddenly we only tested for “Salmonella” without a serotype or PFGE pattern, would we even be able to tell if there was an outbreak? If there was a marked uptick in cases of salmonellosis, how would you know who was a part of the outbreak compared to the baseline? Given the high number of cases of salmonellosis annually, how would you know who to collect food histories from? If you went the route of testing food and found some that were positive for Salmonella, how would you know if it was associated with an outbreak? How many products, brands, and industries could be damaged? While it’s great to develop a rapid test so that a doctor rapidly knows “yes, it’sSalmonella” or a food company can make a quick decision about the safety of a product, many of these tests currently provide no big-picture information for the control of what might be an outbreak.
If there is a move toward culture-independent methods, does this mean the end of PulseNet or some next generation of PulseNet? I can’t imagine that will happen (although questions were raised regarding the “next generation” of PulseNet and the potential to rely on newer methods). However, we do need to think through the unintended consequences of advances in this area. Given the great ingenuity often demonstrated in how these methods work, I am confident that by raising awareness of the issue, we’ll find a workable solution. I hope that in “round 2” of this discussion, all stakeholders — public health and clinical communities, food industry members, test kit developers, and others — convene to discuss how we can work together to ensure that we use the best technology possible to prevent foodborne illness.