Following a request from EFSA, the Panel on Biological Hazards was asked to (i) identify the serotypes of Yersinia enterocolitica which are pathogenic to humans, (ii) give advice regarding the analytical methods to be used to detect and identify the human pathogenic Yersinia enterocolitica serotypes from food and animals, (iii) consider the need to monitor Yersinia pseudotuberculosis in animals and food and (iv) recommend the monitoring methods for Yersinia spp. in animal populations and foodstuffs that are most optimal from the public health point of view. There are several species of the genus Yersinia. Y. enterocolitica and Y. pseudotuberculosis are widespread in Europe and have the characteristics of typical enteropathogens. Not all Yersinia enterocolitica strains are pathogenic. The BIOHAZ Panel concludes that the best and most reliable indicator of Y. enterocolitica pathogenicity is the biotype as the various biotypes are either pathogenic or non-pathogenic. The serotype is not a reliable marker of Y. enterocolitica pathogenicity because several serotypes are common to both pathogenic and nonpathogenic strains. Strains of biotype 4 (serotype O: 3) and biotype 2 (serotypes O: 9) are commonly associated with human infections in Europe. Biotype 4 predominates in most Member States. However, biotype 2 might predominate in a few other Member States. These biotypes are seldom reported to be isolated from the environment. Animals (pigs and cattle) are the main reservoir and human cases are typically sporadic. Strains of biotype 1A are widely spread in the environment and are often isolated from animal and human stools and from foods, but they are considered non-pathogenic. With regard to Y. pseudotuberculosis, all strains are potentially pathogenic for humans and a wide range of animal species. Serotype I is by far the most common serotype associated with human and animal infections in Europe, followed by serotype III. Wile animals are probably the principal reservoir of Y. pseudotuberculosis in Europe. While several culture methods have been described for the isolation and characterisation of Y. enterocolitica from foods, the environment and animals, no single isolation procedure appears to be optimal for recovery of all human-pathogenic strains of Y. enterocolitica in foods. However, the International Standard Organization method for the detection of pathogenic Y. enterocolitica in foods can also be applied equally well to lymphatic tissues such as tonsils, and includes the parallel use of two isolation procedures. The isolation of Y. pseudotuberculosis in food and environmental samples is relatively difficult and no single selective medium is available that can be used for all strains of this species. Polymerase Chain Reaction (PCR) could be a useful method for preliminary screening for pathogenic Y. enterocolitica in animal, food or environmental samples. An enrichment step prior to PCR is essential to increase the sensitivity and to decrease the risk of false-positive results due to detection of dead cells. PCR could be a useful method to use in parallel with culture methods to screen Y. pseudotuberculosis in animal, food or environmental samples, but its performance first needs to be carefully evaluated. The BIOHAZ Panel also concludes that a reporting system for Y. enterocolitica and Y. pseudotuberculosis cannot rely only on the presence of genetic traits. Isolation of the strains is essential for confirmation and to enable characterisation. Therefore, when PCR methods are used, positive results should be confirmed with culture methods. Monitoring and surveying of human pathogenic Yersinia in animal populations and food should rely on information on human yersiniosis. The BIOHAZ Panel concludes that routine EU-wide monitoring of human-pathogenic Y. enterocolitica and Y. pseudotuberculosis in animals and foods is not recommended. However, more comparable data are needed on the prevalence of pathogenic Y. enterocolitica in the porcine reservoir. These could be obtained, depending on the risk management priorities, by a EU-wide baseline survey on pathogenic Y. enterocolitica in the pig population, or by national surveys on pathogenic Y. enterocolitica in the pig population following a harmonised design. Sampling of pig tonsils at the time of slaughter would provide data on the prevalence at a relevant point in the food chain. If specific biotypes/serotypes represent a serious problem in human yersiniosis, other animal reservoir may be surveyed. When large numbers of animals are to be tested, a pre-screening by serological testing at the time of slaughter could be used to identify infected herds from which pathogenic Y. enterocolitica can be detected thereafter by culture methods. Depending on the human disease situation, consumption patterns and prevalence of pathogenic Y. enterocolitica in pigs, focused surveys in foods of concern in the individual Member State could be considered, e. g. in fresh pork meat. With regard to Y. pseudotuberculosis, the BIOHAZ Panel recommends that in the event of an increased incidence of human Y. pseudotuberculosis infections or if outbreaks occur, focused surveys could be considered in individual Member States guided by the results of epidemiological data. Although the current ISO method for the detection of pathogenic Y. enterocolitica in foods is not optimal for the isolation of all human-pathogenic strains, it is currently recommended as the 'method of choice for monitoring and survey purposes. Efforts should be made at EU level to improve the current isolation methods for Y. enterocolitica and to develop a better and standardised Y. pseudotuberculosis isolation medium. Finally, the BIOHAZ Panel recommends that only pathogenic Yersinia strains should be included when reporting on the occurrence of Yersinia spp. in animals, foods, and human cases of yersiniosis. These strains are (i) all Y. pseudotuberculosis strains (with an indication of their serotype) and (ii) all Y. enterocolitica strains, except those of biotype 1A, with indication of their biotype (and preferably also their serotype).
