An updated meta-analysis of the distribution and prevalence of Borrelia burgdorferi s.l. in ticks in Europe

Abstract

Background: The bacteria of the group Borrelia burgdorferi s.l. are the etiological agents of Lyme borreliosis in humans, transmitted by bites of ticks. Improvement of control measures requires a solid framework of the environmental traits driving its prevalence in ticks.

Methods: We updated a previous meta-analysis of the reported prevalence of Borrelia burgdorferi s.l. in questing nymphs of Ixodes ricinus with a literature search from January 2010-June 2017. This resulted in 195 new papers providing the prevalence of Bb for 926 geo-referenced records. Previously obtained data (878 records, years 2000-2010) were appended for modelling. The complete dataset contains data from 82,004 questing nymphs, resulting in 558 records of B. afzelii, 404 of B. burgdorferi s.s. (only 80 after the year 2010), 552 of B. garinii, 78 of B. lusitaniae, 61 of B. spielmanii, and 373 of B. valaisiana. We associated the records with explicit coordinates to environmental conditions and to a categorical definition of European landscapes (LANMAP2) looking for a precise definition of the environmental niche of the most reported species of the pathogen, using models based on different classification methods.

Results: The most commonly reported species are B. afzelii, B. garinii and B. valaisiana largely overlapping across Europe. Prevalence in ticks is associated with portions of the environmental niche. Highest prevalence occurs in areas of 280°-290° (Kelvin) of mean annual temperature experiencing a small amplitude, steady spring slope, together with high mean values and a moderate spring rise of vegetation vigor. Low prevalence occurs in sites with low and a noteworthy annual amplitude of temperature and the Normalized Difference Vegetation Index (colder areas with abrupt annual changes of vegetation). Models based on support vector machines provided a correct classification rate of the habitat and prevalence of 89.5%. These results confirm the association of prevalence of the three most commonly reported species of B. burgdorferi s.l. in Europe to parts of the environmental niche and provide a statistically tractable framework for analyzing trends under scenarios of climate change.

Keywords: Borrelia burgdorferi s.l.; Distribution; Europe; Meta-analysis; Modelling of prevalence.

Fig. 1

The workflow of the study. We collected literature data (a) in the period 2010–2017 about the prevalence of Borrelia burgdorferi s.l. (Bb) and B. miyamotoi in questing ticks (but only data from questing nymphs were finally used). This allowed to plot raw maps of the reported distribution of each species of Bb. We built scripts in R to mass download 8-days 1 km. satellite images from MODIS repository (b) that were transformed into Fourier components using in-house developed scripts (c). All the scripts and the transformed satellite imagery are available in http://datadryad.org/resource/doi:10.5061/dryad.2h3f2.2. We tessellated the complete territory with a grid of hexagonal cells of 0.25° of radius, and environmental data as well as prevalence values (converted to categories) were transferred (d). The categories of prevalence for each species were also transferred (e) to a standard definition of the ecological classes in Europe (LANMAP2) to obtain wide area estimates of Bb species composition and prevalence. Data on prevalence and environmental traits were used to produce a map in the environmental niche about the restriction of Bb species and prevalence to different portions of the niche outlined by temperature and vegetation (f). These grid-derived values were also used to model the influence of the environmental traits on the observed values of prevalence by training a Neural Network

Fig. 2

The reported distribution and prevalence of B. afzelii in questing nymphs of I. ricinus (data for years 2010–2017). The size and the levels of grey of the plots in the legend define the prevalence at the coordinates of the points

Fig. 3

The reported distribution and prevalence of B. garinii in questing nymphs of I. ricinus (data for years 2010–2017). The size and the levels of grey of the plots in the legend define the prevalence at the coordinates of the points

Fig. 4

The reported distribution and prevalence of B. valaisiana in questing nymphs of I. ricinus (data for years 2010–2017). The size and the levels of grey of the plots in the legend define the prevalence at the coordinates of the points

Fig. 5

a The reported distribution and prevalence of B. lusitaniae in questing nymphs of I. ricinus (data for years 2010–2017). The size and the levels of grey of the plots in the legend define the prevalence at the coordinates of the points. b The reported distribution and prevalence of B. burgdorferi s.s. in questing nymphs of I. ricinus (data for years 2010–2017)

Fig. 6

The reported distribution and prevalence of B. burgdorferi s.l. in questing nymphs of I. ricinus (data for years 2000–2017). The size and the levels of grey of the plots in the legend define the prevalence at the coordinates of the points

Fig. 7

The percentage of prevalence classes of B. burgdorferi s.l. in questing nymphs of I. ricinus associated with the climate zones in the target region (data for years 2000–2017). The X (categorical) axis shows the types of climate according to the definition of LANMPA2. The Y axis shows the percent of the prevalence categories as distributed across the climate categories

Fig. 8

The projection of the prevalence categories of B. afzelii (a), B. garinii (b), and B. valaisiana (c) modelled using support vector machine on the climate categories (LANMAP2 scheme) of the target territory

Fig. 9

The distribution of major classes of prevalence of B. afzelii in the environmental axes of a mean annual values of temperature (in Kelvin) versus the slope of temperature in spring (unitless), and b mean annual values of temperature versus the slope of NDVI in spring for the period 2002–2016. Points of the category “absent” were not plotted to improve readability

Fig. 10

The distribution of major classes of prevalence of B. garinii in the environmental axes of a mean annual values of temperature (in Kelvin) versus the slope of temperature in spring (unitless), and b mean annual values of temperature versus the slope of NDVI in spring for the period 2002–2016. Points of the category “absent” were not plotted to improve readability

Fig. 11

The distribution of major classes of prevalence of B. valaisiana in the environmental axes of a mean annual values of temperature (in Kelvin) versus the slope of temperature in spring (unitless), and b mean annual values of temperature versus the slope of NDVI in spring for the period 2002–2016. Points of the category “absent” were not plotted to improve readability