The 11th Campylobacter, Helicobacter and Related Organisms (CHRO) meeting was held in September 2001 in Freiburg (Germany). It was aimed at promoting interdisciplinary contacts and exchange of information on the various topics concerned. The organisers successfully brought together over 700 participants from 54 countries, representing medical microbiologists, clinical gastroenterologists, neurologists, pathologists, epidemiologists, public health experts and veterinarians (1). Campylobacter infections have become a major cause of human gastroenteritis in developed countries. The annual number of reported cases exceeds the even reported Salmonella cases in many European countries. This workshop offered an excellent opportunity to understand the microbiological and epidemiological complexities involved, and develop collaborations between different professionals. This article describes information related to Campylobacter only.

Microbiology and pathogenesis of campylobacteriosis
Campylobacter cells are gram-negative spirally curved rods that are microaerophilic, requiring a low oxygen content for growth. It is, however, sometimes difficult to cultivate the microorganisms. The bacterial cells react to temperature downshift by altering cell morphology and physiology. As the temperature decreases, coccoid cells are formed, resulting in viable but non-cultivable forms. This is considered to be an adaptive response to hostile external environments. The resuscitation of non-cultivable cells has been demonstrated in chickens.
The most common pathogenic species for humans are C. jejuni and C. coli. The pathogenesis of diarrhoea is still not completely understood. Motility by the flagellum is required for efficient colonisation of intestinal cells. Cell invasion is a major pathogenic mechanism of Campylobacter infection. Campylobacters have been shown to produce cytotoxic proteins that may play a role in clinical courses of the disease. However, the studies do not give similar results and therefore the role of toxin production remains unclear. Apart from causing diarrhoea, Campylobacters can cause neurological complications such as Guillain-Barré syndrome (GBS). GBS is an autoimmune disorder of the peripheral nervous system, which causes acute flaccid paralysis. The syndrome is correlated with prior infection by C. jejuni in up to 40% of cases. It has been shown that GBS arises as a result of autoimmune attack due to molecular mimicry that exists between certain lipopolysaccharide (LPS) molecules of C. jejuni strains and human nerve tissue gangliosides. Furthermore, some serotypes of C. jejuni are associated with GBS.
These microbiological characteristics of Campylobacters demonstrate the importance of understanding the microbiological features of a microorganism when interpreting microbiological, clinical and epidemiological data.

Molecular microbiology and serology
A range of subtyping methods has been used to study the epidemiology of campylobacters. Serotyping with commercial antisera is the most frequently used method. Many strains remain untypeable, however, and different molecular sub-typing methods have been developed. As Campylobacters are genetically and phenotypically very diverse, the methods yield results that are difficult to interpret. It has been suggested that the purpose of creating diversity is to enable survival during transmission from host to host. The diversity may reflect differences in virulence. There is evidence that human isolates are more virulent than poultry isolates. There is a strong need for standardisation of a molecular subtyping method. A first attempt was made by Campynet in the shape of a EU network project. Campynet has standardised a pulsed field gel electrophoresis (PFGE) method for definitive typing of C. jejuni and C. coli. This method could form a genotyping scheme of outbreak strains even on an international scale. In the United States, a PFGE method standardised by PulseNet has been successfully used to determine the environmental cause of a Campylobacter outbreak although a food source was initially suspected.

Epidemiology and risk factors
In Europe, different studies have shown associations between sporadic human cases and animal sources, suggesting that these sources would be the most important risk factors for sporadic human infections. The prevalence studies among animals show different percentages for different types of animals and Campylobacter species. In animals, however, many species and serotypes are found at the same time. The most prevalent Campylobacter species in poultry and cattle has been shown to be C. jejuni, in swine, C. coli, and in dogs, C. upsaliensis. Most human infections (90-95%) are due to C. jejuni. Broiler flocks are often contaminated with C. jejuni and C. coli. There is strong evidence for frequent cross contamination during slaughtering and product processing. Some studies have, however, indicated that slaughter processes have a minor influence on the risk for human campylobacteriosis. Rather, reducing the Campylobacter load on chickens seemed to have a significant impact on the number of human cases. Campylobacter carriage in Danish poultry flocks has showed a seasonal variation with a higher contamination rate (50-80%) during the summer (June to October) and a lower rate (13-40%) during the winter season (December to March). Human infections show similar seasonality in many European countries.
Interestingly, a significant association between the infection with Campylobacter jejuni serotype O:6 and eating undercooked poultry has recently been demonstrated in Denmark. The infections of this serotype displayed a distinct peak in summer, suggesting a variable importance of different risk factors during one year. In England and Wales, epidemiological and microbiological data have been integrated within the frame of a sentinel surveillance system (CAMPSITE group). In the first twelve months of the study, over 7000 completed epidemiological datasets were created. A new approach based on case-case comparisons has been applied to generate hypotheses for Campylobacter infection. The cases of C. coli were more likely to have drunk unboiled tap water (OR 2.5, P=0.01) or river/stream/spring water (OR 5.47, P=0.006) than C. jejuni cases. They were also less likely to have eaten pre-packed sandwiches (OR=0.36, P=0.03) than C. jejuni cases. This indicates the variable role of different Campylobacter species depending on risk factors.
Campylobacters have been associated with waterborne outbreaks and travel abroad in several countries. There is evidence that cattle may be an important reservoir for the contamination of drinking water.

Public health impact
Public health impact of campylobacteriosis can be assessed in several ways. In the Netherlands, the health burden of infections with thermophilic Campylobacter species has been evaluated by using DALYs (Disability Adjusted Life Year). DALYs is the sum of Years of Life Lost by premature mortality and Years Lived with Disability, weighed with the factor between 0 and 1 for the severity of the illness. The main determinants of health burden were acute gastroenteritis in the general population, gastroenteritis related mortality and residual symptoms of GBS. The health burden of C. jejuni associated illness in the Dutch population was estimated to range between 1000 and 2000 DALYs per year.
Another interesting example came from Iceland, where an epidemic of domestic human campylobacteriosis occurred in 1998-1999 reaching the peak of 158 cases/100 000 inhabitants. The cause for the epidemic was due mainly to increased consumption of fresh chicken. Intervention measures were actively taken in 1999 and 2000 by testing the broilers at slaughtering, educating farmers and consumers, and by freezing all Campylobacter-positive broiler flocks before sending them for retail. In 2000, there was a reduction of 72% in domestic Campylobacter cases, that the interventions were effective.

Antimicrobial resistance
In the majority of the Campylobacter cases, antibiotic therapy is not necessary. In severe cases, however, erythromycin and ciprofloxacin are the drugs of choice. In the 1990s, antimicrobial resistance increased, especially to fluoroquinolone, ciprofloxacin, in many Campylobacter species. The resistance to erythromycin has not increased significantly. The resistance to ciprofloxacin has reached about 30% in human C. jejuni isolates in many European countries during the last few years. The increase in resistance has been associated with antibiotic use in animal husbandry as well as inappropriate treatment in humans. In Ireland, the resistance to ciprofloxacin among human and poultry isolates did not increase significantly during 1996-1998 despite the licensing of enrofloxacin for use in poultry until 1987. Reports from other countries show similar results. In a study in Denmark, the resistance to fluoroquinolones varied considerably from farm to farm, indicating a possible association between the use of fluoroquinolones and increased resistance. Proper epidemiological studies are still needed to confirm the link between antibiotic use in farm animals and increase in resistance to fluoroquinolones. The resistance to fluoroquinolones has also been associated to travelling abroad, especially in countries outside Europe.
Networks
Two EU network projects were presented in the workshop. The first one was Campynet, which is a network for the standardisation and harmonisation of molecular subtyping of Campylobacter jejuni/coli. The group collected a set of 100 strains from multiple locations throughout Europe and from human, animal and environmental sources. The strains have been speciated and characterised. The list of strain characteristics is available on the Campynet website www.svs.dk/campynet. The second project was a study to evaluate and describe surveillance and laboratory methodologies for Campylobacter in Europe. This project is presented in a separate article in this issue.