Contemplating Campylobacter Conundrums
By Timothy Freier, Ph.D., and Patrick Kennedy
A “conundrum” is defined as “an intricate and difficult problem.” That definition seems perfect for an article discussing Campylobacter. Also, it starts with “c,” and who doesn’t enjoy a good alliteration? Campylobacter spp., particularly thermophilic Campylobacter, have emerged as a leading cause of human foodborne gastroenteritis worldwide, with C. jejuni, C. coli and C. lari reportedly responsible for the majority of human infections (well, maybe; more on this later). Although most cases of campylobacteriosis are self-limiting, this illness represents a significant public health burden. To anyone who has been unlucky enough to experience this illness, the low mortality rate is of little comfort when faced with potentially severe symptoms such as cramping, abdominal pain, fever, nausea and vomiting, and watery, sometimes bloody diarrhea. Perhaps even more important than the gastrointestinal distress, Campylobacter can use a mechanism called molecular mimicry to trigger a very severe autoimmune disease, Guillain-Barré syndrome, which is the most common cause of acute flaccid paralysis. In 2013, U.S. FoodNet data indicated Campylobacter was the second-most common bacterial agent identified in foodborne illness reports (35% of reported infections) following Salmonella (38%).
Campylobacter is also one of the foodborne pathogens that has frustrated public health agencies’ efforts to reduce its contribution to illnesses. The Healthy People 2020 target rate for Campylobacter infections is 8.5 cases per 100,000 people, but the 2014 rate of culture-confirmed infections was reported to be 13.45. There are many interesting conundrums concerning this organism that contribute to its being and continuing to be a leading cause of foodborne illness.
Conundrum #1: Since Campylobacter and Salmonella are both highly associated with poultry, won’t the same interventions being used to control Salmonella also reduce Campylobacter?
The poultry industry has placed the majority of its pathogen-control emphasis on Salmonella. Major outbreaks linked to Salmonella Enteritidis in shell eggs and Salmonella Heidelberg and Hadar in various chicken and turkey products caused the poultry industry and regulatory agencies to declare war on this pathogen in poultry. Campylobacter, although known to be associated with poultry, tends to cause sporadic foodborne illnesses rather than outbreaks; thus, less emphasis has been placed on its control in poultry.
It is well established that Campylobacter is very commonly harbored in the gastrointestinal systems of most domesticated and wild animals. In fact, it has been estimated that greater than 80% of food animals carry Campylobacter. Many of the preharvest interventions used by the egg and poultry industries to control Salmonella have no doubt also had an influence on controlling Campylobacter. For example, protecting poultry from environmental contamination through enhanced biosecurity, litter management and drinking water treatment would be expected to reduce the incidence of both pathogens in live birds. The use of competitive exclusion through the application of probiotics may also reduce the incidence of both pathogens. However, more specific interventions, such as vaccines, bacteriophages and bacteriocins, would need to be targeted specifically to Campylobacter to have an impact. While poultry feed is thought to be a potential source of Salmonella introduction to poultry, it is probably not a source for Campylobacter, as it does not survive well in low-moisture products.
An important question for effective preharvest Campylobacter control is whether transmission of this pathogen is vertical or horizontal. Vertical transmission means that Campylobacter contamination moves from the breeder hen to the egg to the chick. Horizontal transmission is contamination originating from the water, feed or environment. It is well accepted that Salmonella can spread in poultry via vertical transmission, and some serotypes, such as S. Enteritidis, are known to be internalized into the egg. It has been less clear whether this can occur with Campylobacter. Researchers have not been able to demonstrate that internalization of Campylobacter into eggs is a significant factor for transmission. Much work has been done to determine whether viable but nonculturable Campylobacter could be transferred via the egg and then eventually colonize newly hatched chicks, but it has been concluded that this is not a significant occurrence. However, another study showed that when 2,000 chick paper pad tray liners were sampled in commercial chicken hatcheries, 0.75% were positive for Campylobacter, supporting previous findings indicating the potential for Campylobacter to be spread by vertical transmission. While probably not internalized within the egg, Campylobacter can be transmitted via the eggshell. In one study, a total of 2,710 eggs were examined, and viable Campylobacter was found on 4.1% of the eggshell samples while Salmonella was found on 1.1%. In this study, egg yolk samples were negative for both pathogens.
Placing insect screens on poultry houses may be one simple and effective intervention for Campylobacter control. In a study conducted in Denmark, insect screens were placed on some broiler houses and not on others. The houses without the fly screens had a Campylobacter flock-positive rate of 51.4% compared with 15.4% for the houses with the screens. This study was repeated and confirmed with a multiyear study, also conducted in Denmark, which found the prevalence of Campylobacter-positive flocks was significantly reduced, from 41.4% during 2003–2005 (before fly screens) to 10.3% in 2006–2009 (with fly screens). In fly-screen houses, Campylobacter prevalence did not peak during the summer as it did in the houses without screens. Nationally, the prevalence of Campylobacter-positive flocks in Denmark could have been reduced by an estimated 77% during summer had fly screens been part of biosecurity practices.
Postharvest interventions, such as sanitary dressing procedures and the use of antimicrobial solutions during multiple steps of the slaughter and further processing activities, would be expected to reduce the incidence of both Campylobacter and Salmonella. One intervention that has been targeted specifically for Campylobacter reduction is the use of frozen storage. In fact, some countries require the freezing of carcasses originating from more highly contaminated Campylobacter flocks as a public health control intervention. While this practice does appear to have impacted illnesses, other researchers have questioned whether freezing is a truly effective intervention and whether the results may be an artifact of the difficulties in culturing and detecting Campylobacter.
Campylobacter is not typically considered to be a pathogen that can survive or grow in the food manufacturing facility environment. It is rarely, if ever, the focal point of environmental monitoring programs. Investigations of Campylobacter illnesses associated with foods have not indicated that manufacturing facility contamination has been a causative factor. As a difficult and relatively expensive organism to test for, Campylobacter is not an ideal candidate for routine environmental monitoring as a verification of prerequisite programs.
Perhaps the most important consideration for Campylobacter control in the food manufacturing facility environment is the water used for direct product contact or addition to products. Consumption of contaminated drinking water is a significant cause of Campylobacter infections. If it is determined that water testing would be prudent, testing large volumes of water may be required to reach the desired sensitivity for either culture or molecular detection methods, and stressed cells of Campylobacter can be especially difficult to detect. A good practice to reduce this risk is to carefully evaluate all sources of water used in the manufacturing process, being sure to trace and map the entire water system. Water that is used for direct product contact or product addition should have residual chlorine or be otherwise treated to be fit for use.
The U.S. Department of Agriculture (USDA) recently published an updated Draft FSIS [Food Safety Inspection Service] Compliance Guideline for Controlling Salmonella and Campylobacter in Raw Poultry. In this document, the agency offers the following testing recommendations:
“There are no identified index organisms that directly reflect the presence or absence of pathogens in poultry (e.g., Salmonella and Campylobacter). Therefore, FSIS recommends that an establishment test for pathogens at least intermittently and compare its results against the presence or absence of other non-pathogenic organisms (i.e., the indicator organisms the establishment is using) to assess whether it is maintaining process control. The indicator organisms can provide evidence of control, while periodic testing for pathogens may verify that the establishment is reducing pathogens to acceptable levels. Establishments conducting their own ongoing verification sampling of finished product for Salmonella and Campylobacter can use the FSIS performance standards as indicators of process control.”
Due to changes in methodology and sampling programs, it is difficult to use USDA compliance data to prove a parallel reduction in Salmonella and Campylobacter incidence in poultry. However, despite the emphasis on Salmonella control, there does seem to be a generally similar reduction in both pathogens in poultry in the U.S. In the third quarter of 2011, the average prevalence of Salmonella in broiler carcasses was 8.2% and Campylobacter was 8.5%. During the third quarter of 2015, the prevalence of Salmonella and Campylobacter in young chicken carcasses was 1.4% and 2.2%, respectively. These results suggest that while the industry target has mainly been Salmonella, there has also been a concomitant reduction in Campylobacter in U.S. chicken carcasses.
Conundrum #2: Is campylobacteriosis mainly attributed to food, water or other sources?
A review of Campylobacter outbreaks in the U.S. revealed that common vehicles of transmission were food (86%), water (9%) and animal contact (3%). Dairy products were implicated in 29% of the foodborne outbreaks, poultry in 11% and produce in 5%. However, the majority of cases of campylobacteriosis are sporadic, and it is extremely difficult to trace the source of these types of illnesses. It is often generalized that the majority of sporadic cases can be attributed to consuming undercooked poultry or to cross-contamination of ready-to-eat foods with Campylobacter originating from raw poultry, whereas the majority of outbreaks stem from dairy products. However, it seems plausible that nonfood sources may account for an even-greater percentage of the sporadic cases. Besides the extreme epidemiological challenge of attributing the source of sporadic illnesses, there is a challenge in detecting Campylobacter, especially when it is in a stressed condition, as can be expected when trying to isolate the organism from environments such as water or mud. As an example, researchers in Canada found that Campylobacter was frequently detected at low concentrations in a watershed in southern Ontario. When using quantitative PCR, higher prevalence was found compared with a cultural method, probably because of the formation of viable but nonculturable cells. It was concluded that Campylobacter in surface water can be an important vector for human disease transmission and that method selection is important in determining pathogen occurrence in a water environment.
Another interesting phenomenon concerning nonfood sources of campylobacteriosis has been an association with outdoor competitive events. One of the largest reported campylobacteriosis outbreaks in Canada was associated with a mountain bike race that took place in muddy conditions in British Columbia during June 2007. Of the 537 racers included in a follow-up epidemiological study, 225 racers (42%) reported diarrheal illness after the race. C. jejuni clinical isolates (n = 14) were found to be identical by multi-locus sequence typing. Direct accidental ingestion of mud was significantly associated with illness, making mud the most likely source of Campylobacter infection. A similar outbreak occurred during a muddy mountain bike race in the United Kingdom in 2008. An outbreak in Nevada in 2012 was associated with a long-distance obstacle adventure race. These types of outbreaks may represent the tip of the iceberg when extrapolating to sporadic cases and taking into consideration the ubiquitous occurrence of Campylobacter in water, mud and animal feces.
Getting back to the alliterations theme, a very interesting theory explaining a Campylobacter conundrum was put forth in a paper entitled “Flies, Fingers, Fomites, and Food.” The conundrum was that in New Zealand, there is a distinct seasonality of campylobacteriosis, while one of the main food associations is the consumption of cooked chicken at fast-food establishments, a link that did not seem to completely explain the seasonality. The seasonality did, however, correlate quite nicely with the life cycle of flies in the region, with increased illnesses correlating to increased fly foraging activity. The authors postulated that the Campylobacter might be food-associated rather than foodborne, in that the flies might have contaminated fomites such as door handles, and chicken is commonly consumed by handling directly with the fingers that may have contacted the fomites. The contamination may be more a factor of fingers and fomites spreading the contamination from flies directly onto the cooked chicken and into the mouth of the consumer rather than undercooking or cross-contamination in the kitchen, very much changing how this risk can best be mitigated. It should be noted that the New Zealand Food Safety Authority disputed the findings, pointing out yet another conundrum: There is a higher rate of campylobacteriosis in New Zealand among urban dwellers (fewer flies) than rural inhabitants. The agency agreed that there should be a true farm-to-fork approach to controlling Campylobacter.
Conundrum #3: In terms of food safety, the focus has been on C. jejuni, C. coli and sometimes C. lari. What if there are other species that cause a significant number of illnesses?
Despite great advances in detecting, reporting and investigating foodborne illnesses, the U.S. Centers for Disease Control and Prevention estimate that approximately 38.4 million cases of domestically acquired foodborne illness are caused by unspecified agents each year. Many of these illnesses are probably caused by viruses, protozoans and bacteria that are not routinely tested, or for which detection methods do not exist. Because Campylobacter is notoriously difficult to culture in the laboratory, it is reasonable to assume that many of these illnesses due to unspecified agents might be caused by various species of Campylobacter.
More than 20 species are assigned to the genus Campylobacter. Human illnesses attributed to this genus are most commonly associated with C. jejuni and C. coli, but additional species have been identified due to recent advances in methodologies. Less commonly recognized species such as C. lari and C. upsaliensis have been isolated from patients with gastrointestinal diseases.
The use of newer detection technologies that do not necessarily rely on culturing is beginning to indicate that other species of Campylobacter may in fact be causing significant illnesses. In one study, 7,194 fecal samples collected over a 1-year period from patients with diarrhea were screened for Campylobacter spp. using a multiplex-PCR system. Of 349 Campylobacter-positive samples, 23.8% were shown to be C. ureolyticus, using a combination of 16S rRNA gene analysis and highly specific primers targeting the HSP60 gene. The authors of this study suggested that C. ureolyticus may be an emerging enteric pathogen that is capable of causing gastroenteritis. In a review of emerging Campylobacter species causing human illness, C. concisus and C. upsaliensis were also mentioned in addition to C. ureolyticus. C. fetus may be yet another emerging species capable of causing intestinal and systemic illness, although it is probably more associated with immunocompromised individuals. These emerging species are all nutritionally fastidious and would probably not be detected using typical food or clinical cultural screening methods for Campylobacter. As detection methodologies improve, the impact on public health of these emerging species will become clearer.
Conundrum #4: How can a bacterium that is so difficult to grow cause so much illness?
This question has been asked many times, especially by microbiologists who are having trouble keeping the positive lab control culture alive. One of the authors (T.F.) recalls testing for Campylobacter during the early days of routine food testing. Despite the use of complex growth media and gas mixtures, it was very common to “lose” the control culture. Control cultures were expensive and difficult to obtain, so the lab technicians would go to a local butcher shop and buy a freshly butchered chicken carcass. Campylobacter was isolated every time from this source and was used as a control after biochemical identification.
Campylobacter is sensitive to drying, is nutritionally fastidious and cannot tolerate levels of oxygen found in the normal atmosphere. Although difficult to keep alive in the laboratory, some strains of Campylobacter seem able to survive quite well in the natural environment. They probably employ several strategies to survive and may work together with other types of bacteria such as Pseudomonas spp. to form stable mixed biofilms. Another survival strategy is the ability to enter into a viable but nonculturable state, becoming very metabolically inactive, thus better able to survive extreme conditions.
There are probably several more conundrums associated with this complex and vexing bacterial genus. Going forward, many interesting breakthroughs relative to this organism are expected in the areas of epidemiology, detection technologies for food, environmental and clinical samples, and novel risk-reducing interventions as industry, regulatory and public health agencies, and academia collaborate to resolve the many Campylobacter conundrums.
Timothy Freier, Ph.D., serves as division vice president of scientific affairs and microbiology (North America) at Mérieux NutriSciences.
Patrick Kennedy is information services manager at Mérieux NutriSciences.
2. Horrocks, SM et al. 2009. “Incidence and Ecology of Campylobacter jejuni and coli in Animals.” Anaerobe 15(1-2):18–25.
7. Bahrndorff, S et al. 2013. “Foodborne Disease Prevention and Broiler Chickens with Reduced Campylobacter Infection.” Emerg Infect Dis 19(3):425–430.
10. Pitkänen, T. 2013. “Review of Campylobacter spp. in Drinking and Environmental Waters.” J Microbiol Methods 95(1):39–47.
13. Taylor, EV et al. 2013. “Common Source Outbreaks of Campylobacter Infection in the USA, 1997–2008.” Epidemiol Infect 141:987–996.
14. Van Dyke, MI et al. 2010. “The Occurrence of Campylobacter in River Water and Waterfowl within a Watershed in Southern Ontario, Canada.” J Appl Microbiol 109(3):1053–1066.
15. Stuart, TL et al. 2010. “Campylobacteriosis Outbreak Associated with Ingestion of Mud during a Mountain Bike Race.” Epidemiol Infect 138(12):1695–1703.
18. Nelson, W and B Harris. 2006. “Flies, Fingers, Fomites, and Food. Campylobacteriosis in New Zealand–Food-Associated Rather Than Food-Borne.” N Z Med J 119(1240):U2128.
20. Scallan, E et al. 2011. “Foodborne Illness Acquired in the United States–Unspecified Agents.” Emerg Infect Dis 17(1):16–22.
22. Bullman, S et al. 2011. “Campylobacter ureolyticus: An Emerging Gastrointestinal Pathogen?” FEMS Immunol Med Microbiol 61(2):228–230.
23. Man, SM. 2011. “The Clinical Importance of Emerging Campylobacter Species.” Nat Rev Gastroenterol Hepatol 8(12):669–685.
24. Wagenaar, JA et al. 2014. “Campylobacter fetus Infections in Humans: Exposure and Disease.” Clin Infect Dis 58(11):1579–1586.
25. Bronowski, C et al. 2014. “Role of Environmental Survival in Transmission of Campylobacter jejuni.” FEMS Microbiol Lett 356(1):8–19.