Eurosurveillance - Volume 24, Issue 18, 02 May 2019

Borrelia miyamotoi clusters phylogenetically among relapsing fever borreliae, but is transmitted by hard ticks. Recent recognition as a human pathogen has intensified research into its ecology and pathogenic potential.

We aimed to provide a timely critical integrative evaluation of our knowledge on B. miyamotoi, to assess its public health relevance and guide future research.

This narrative review used peer-reviewed literature in English from January 1994 to December 2018.

Borrelia miyamotoi occurs in the world’s northern hemisphere where it co-circulates with B. burgdorferi sensu lato, which causes Lyme disease. The two borreliae have overlapping vertebrate and tick hosts. While ticks serve as vectors for both species, they are also reservoirs for B. miyamotoi. Three B. miyamotoi genotypes are described, but further diversity is being recognised. The lack of sufficient cultivable isolates and vertebrate models compromise investigation of human infection and its consequences. Our understanding mainly originates from limited case series. In these, human infections mostly present as influenza-like illness, with relapsing fever in sporadic cases and neurological disease reported in immunocompromised patients. Unspecific clinical presentation, also occasionally resulting from Lyme- or other co-infections, complicates diagnosis, likely contributing to under-reporting. Diagnostics mainly employ PCR and serology. Borrelia miyamotoi infections are treated with antimicrobials according to regimes used for Lyme disease.

With co-infection of tick-borne pathogens being commonplace, diagnostic improvements remain important. Developing in vivo models might allow more insight into human pathogenesis. Continued ecological and human case studies are key to better epidemiological understanding, guiding intervention strategies.

Estimating burden of disease (BoD) is an essential first step in the decision-making process on introducing new vaccines into national immunisation programmes (NIPs). For varicella, a common vaccine-preventable disease, BoD in the Netherlands was unknown.

To assess national varicella BoD and compare it to BoD of other vaccine-preventable diseases before their introduction in the NIP.

In this health estimates reporting study, BoD was expressed in disability-adjusted life years (DALYs) using methodology from the Burden of Communicable Diseases in Europe (BCoDE)-project. As no parameters/disease model for varicella (including herpes zoster) were available in the BCoDE toolkit, incidence, disease progression model and parameters were derived from seroprevalence, healthcare registries and published data. For most other diseases, BoD was estimated with existing BCoDE-parameters, adapted to the Netherlands if needed.

In 2017, the estimated BoD of varicella in the Netherlands was 1,800 (95% uncertainty interval (UI): 1,800–1,900) DALYs. Herpes zoster mainly contributed to this BoD (1,600 DALYs; 91%), which was generally lower than the BoD of most current NIP diseases in the year before their introduction into the NIP. However, BoD for varicella was higher than for rotavirus gastroenteritis (1,100; 95%UI: 440–2,200 DALYs) and meningococcal B disease (620; 95%UI: 490–770 DALYs), two other potential NIP candidates.

When considering the introduction of a new vaccine in the NIP, BoD is usually estimated in isolation. The current approach assesses BoD in relation to other vaccine-preventable diseases’ BoD, which may help national advisory committees on immunisation and policymakers to set vaccination priorities.

In 2016, an uncommon outbreak of oropharyngeal tularaemia involving six human cases occurred in Germany, caused by drinking contaminated fresh must after a grape harvest.

We describe the details of laboratory investigations leading to identification of the outbreak strain, its characterisation by next generation sequencing (NGS) and the finding of the possible source of contamination.

We incubated wine samples in different media and on agar plates. NGS was performed on DNA isolated from young wine, sweet reserve and an outbreak case’s lymph node. A draft genome of the outbreak strain was generated. Vertebrate-specific PCRs using primers targeting the mitochondrial cytochrome b gene and product analyses by blast search were used to identify the putative source of must contamination.

No bacterial isolate could be obtained. Analysis of the draft genome sequence obtained from the sweet reserve attributed this sequence to Francisella tularensis subsp. holarctica, belonging to the B.12/B.34 phylogenetic clade (erythromycin-resistant biovar II). In addition, the DNA sequence obtained from the case’s isolate supported our hypothesis that infection was caused by drinking contaminated must. The vertebrate-specific cytochrome b sequence derived from the young wine and the sweet reserve could be assigned to Apodemus sylvaticus (wood mouse), suggesting that a wood mouse infected with F. tularensis may have contaminated the must.

The discovered source of infection and the transmission scenario of F. tularensis in this outbreak have not been observed previously and suggest the need for additional hygienic precautionary measures when processing and consuming freshly pressed must.

In 2017, a food-borne Salmonella Agona outbreak caused by infant milk products from a French supplier occurred in Europe. Simultaneously, S. Agona was detected in animal feed samples in Bavaria.

Using next generation sequencing (NGS) and three data analysis methods, this study’s objectives were to verify clonality of the Bavarian feed strains, rule out their connection to the outbreak, explore the genetic diversity of Bavarian S. Agona isolates from 1993 to 2018 and compare the analysis approaches employed, for practicality and ability to delineate outbreaks caused by the genetically monomorphic Agona serovar.

In this observational retrospective study, three 2017 Bavarian feed isolates were compared to a French outbreak isolate and 48 S. Agona isolates from our strain collections. The later included human, food, feed, veterinary and environmental isolates, of which 28 were epidemiologically outbreak related. All isolates were subjected to NGS and analysed by: (i) a publicly available species-specific core genome multilocus sequence typing (cgMLST) scheme, (ii) single nucleotide polymorphism phylogeny and (iii) an in-house serovar-specific cgMLST scheme. Using additional international S. Agona outbreak NGS data, the cluster resolution capacity of the two cgMLST schemes was assessed.

We could prove clonality of the feed isolates and exclude their relation to the French outbreak. All approaches confirmed former Bavarian epidemiological clusters.

Even for S. Agona, species-level cgMLST can produce reasonable resolution, being standardisable by public health laboratories. For single samples or homogeneous sample sets, higher resolution by serovar-specific cgMLST or SNP genotyping can facilitate outbreak investigations.