 |
Epidemiological consequences of tick ecology |
 |
||||||||||||||||
|
Sarah Randolph, Department of Zoology, University of Oxford, UK A central puzzle in predicting the risk of European tick-borne zoonoses
is why, when tick-borne encephalitis virus (TBEv) and Lyme borreliosis
spirochaetes Borrelia burgdorferi s.l. (LB) share the same tick vector and
have rodent hosts in common, do these pathogens show very different
epidemiological patterns? The tick Ixodes ricinus occurs throughout
Europe, from Ireland to the Urals, from southern Sweden to Greece and
northern Spain. TBEv, however, occurs only in discrete foci within the
tick's distribution, whereas LB occurs extensively throughout Europe more
or less wherever ticks occur. Furthermore, the infection prevalence in
unfed I. ricinus and the basic reproduction number (R0) are both an order
of magnitude lower for TBEv than LB, suggesting that the conditions for
TBEv maintenance are much more narrowly defined than for LB. Tick-borne encephalitis risk maps Recent progress in explaining and predicting the distribution pattern of TBE foci has come from integrating two complementary approaches, biological process-based analysis and statistical pattern-matching analysis. The cellular basis of TBEv enzootic cycles is the transmission of non-systemic infections between co-feeding infected and uninfected ticks (Labuda et al. 1993). To permit this, larval and nymphal ticks must show synchronous seasonal feeding periods, which is indeed typical within TBE foci, but not outside foci (Randolph et al. 2000). At the same time, these two tick stages show coincident aggregated distributions amongst their rodent hosts so that each infected nymph can pass the virus to many larvae. Together, this transmission route and these tick-feeding patterns approximately treble the potential for TBEv survival (Randolph et al. 1999). The ideal way to generate risk maps, therefore, is to use a tick population model to predict geographically variable tick seasonal dynamics depending on local climatic conditions. Until we have such a model, we must turn from the biological to a statistical analysis, seeking correlations between the spatial patterns of TBEv distribution and spatial patterns of environmental conditions. By incorporating a digitized map of TBE foci (Immuno 1997) into a geographical information system with information on seasonal environmental conditions derived from NOAA's AVHRR satellites, we can predict the probability of TBE-presence in each pixel of the satellite images, and so derive a pan-European predictive risk map (Randolph 2000). The map captures the gross patterns of TBE distribution in central Europe and the Baltic region, the considerable heterogeneity within central Europe, and correctly predicts some of the new or reactivated foci recorded over the past few years, e.g. on Bornholm Island and in Sweden (http://www.tbe-info.com/reports/sweden_map.html). Satellite imagery not only allows accurate predictions of observed
patterns, it also helps to inform us of the biology underlying those
patterns. Of the satellite signals, the most significant predictors are
the Normalized Difference Vegetation Index (NDVI), which indirectly
reflects moisture conditions on the ground and therefore habitat
suitability for I. ricinus (Estrada-Peña 1999), and land surface
temperature (LST), which determines the tick's seasonal dynamics and
therefore the existence of the co-feeding transmission route for TBEv. TBE
foci are characterized by a significantly more rapid temperature decline
in the autumn (Randolph et al. 2000), although the precise biological link
between varying LST seasonal profiles and tick seasonal patterns is not
yet defined. Fragile TBE cycles at risk from climate change We can progress from this analysis of the spatial pattern of TBE to predictions about temporal dynamics, in particular the possible impact of climate change. Again using a pattern-matching exercise, but this time using climate itself rather than satellite imagery as the predictor variables, we can identify the multi-variate climatic predictors of present- areas of disease risk. The same variables are then applied to climate scenarios forecast for the future (based on the Hadley Centre models) to predict future TBEv distributions. TBE is predicted to be driven into higher latitude and higher altitude regions as summers are forecast to get hotter and drier (Randolph & Rogers 2000). By the 2020s, the southern and southwestern edges of the present range of TBEv may be cleared, and in northeast Europe the range may contract in the Baltic States, but move north and west in Sweden. Areas around the major lakes in southern Sweden, where foci have been identified recently, appear to become suitable for TBEv. This trend continues through the 2050s and 2080s, until TBEv is confined to parts of Scandinavia, with new foci in southern Finland. These rather extreme predictions are consistent with our new quantitative understanding of the natural ecology of TBEv cycles. They are inherently fragile and may be disrupted by changes in the limiting abiotic factors, notably the seasonal temperature profile and moisture conditions. Many of the recent changes in annual TBE incidence in Europe are
consistent with these predictions, although this does not mean that any
climate change to date is the sole cause of change. There has been least
change, or even a decrease, in countries at the southern edge of the
current distribution, Croatia, Slovenia and Hungary. At the other
latitudinal extreme, the northward extension of I. ricinus and increase in
number of TBE cases in Stockholm county since 1984 has been related to
milder winters and extended spring and autumn seasons, permitting
prolonged season of tick activity and hence pathogen transmission
(Lindgren 1998; Lindgren et al. 2000). Likewise, Finland shows a gradual
increase in TBE incidence since 1984, but in Estonia, Latvia, Lithuania
and Poland the increases have been far more dramatic, but not until 1990s.
In the Czech Republic, however, there was a more marked increase in 1953
than in 1993. These sudden but asynchronous increases suggest that, rather
than being explicable by a single regional climatic factor, site-specific
non-biological causes may be more important. Lyme disease - tick-host relationships In contrast to TBE, natural cycles of LB are robust and maintained
under a wide variety of conditions, but there is still marked spatial
variation in risk across Europe. The variable infection prevalence in
questing nymphal ticks, from 0 to c.25%, depends on biotic factors
operating against a generally permissive abiotic background. Specifically,
a genetically diverse array of Borrelia interacts with an even more
diverse array of mammalian and avian host species. Many of the major host
species for I. ricinus contribute to transmission of spirochaetes, but
they do so in different, complementary ways because they feed different
fractions of the tick population and they are differentially competent to
transmit the different genospecies of B. burgdorferi s.l. to ticks (Hu et
al. 1997; Humair & Gern 1998; Humair et al. 1995, 1998; Kurtenbach et
al. 1998a). The known reservoir status of the host species, revealed by
xenodiagnosis, is mirrored by the species-specific lethality of the host's
serum, mediated via the alternative complement system (Kurtenbach et al.
1998b). For example, rodent serum kills B. garinii and B. valaisiana, and
rodents only transmit B. afzelii and B. burgdorferi s.s. In contrast, bird
serum kills B. afzelii and B. burgdorferi s.s., and birds only transmit B.
garinii and B. valaisiana. These interactions have a marked impact on the
infection pattern in tick populations and therefore on the risk of
infection to humans. Parallels between TBE and Lyme borreliosis? Despite this interpretation that the risk of TBE is limited principally
by abiotic factors and that of LB by biotic factors, there are
nevertheless important parallels between them, and with other tick-borne
zoonoses in Europe and USA. All have shown marked increases over recent
decades. Is this due merely to raised awareness of ticks as vectors and
more intense surveillance, or have biological factors caused an increase
in real incidence? Only at the extreme northern limits of I. ricinus
distribution is there good evidence that climatic factors have played some
part. In many parts of the northern hemisphere, however, contact between
ticks and humans has increased. This is due largely to an increase in the
distribution and density of ticks, caused largely by human impact on the
habitat and wildlife hosts of ticks. For example, deer populations have
increased markedly. Changing agricultural and sociological factors also
take more people into tick-infested forests. In the Czech Republic, where
a long history of systematic registration of TBE should reduce the
surveillance bias to a minimum, the annual numbers of cases of TBE and LB
since 1986 are very closely correlated (Randolph 2001). The data suggest
that LB was under-reported only in the first year of records, and
thereafter the common principal risk factor, tick-human contact,
determined annual variations in both infections. Acknowledgements It is a pleasure to acknowledge the contributions to this paper made by
past and present members of the Oxford Tick Research Group. Rob Green and
David Rogers made significant contributions to the risk mapping of TBE. In
addition, Milan Labuda and Lise Gern have been a source of inspiration. References
CRTBI - Extended Abstracts Bunikis J et al.: Structure and Function of the Surface Proteins of Borrelia Spirochetes Dambrauskienè V et al.: Epidemiology of Tick-Born Encephalitis and it's Clinical Manifestations in Panevèzys City County Egenvall A et al.: A serosurvey of granulocytic Ehrlicha spp. and Borrelia burgdorferi sensu lato in 2018 Swedish horses Gray J S et al.: The biology of Ixodes ticks, with special reference to Ixodes ricinus Guillaume B et al.: Human Granulocytic Ehrlichia Infection in Belgium Haglund M et al.: Tick-borne encephalitis (TBE) - an overview Larsen K et al.: Tick species and arthropod-transmitted infections from Danish cats and dogs Lundkvist Å et al.: Characterization of Tick-borne enchephalitis virus from Latvia – evidence for co-circulation of three distinct subtypes Malmgren L et al.: A field trial of the effectiveness of 65% permetrin spot-on and 9.7% fipronil spot-on against ticks (Ixodes ricinus) on dogs Massung R F et al.: Genetic Variants of Ehrlichia phagocytophila in the United States Nilsson I et al.: Serological evidence of Lyme arthritis in Egypt Nyman D et al.: Ticks have preferences in choosing human hosts Ornstein K et al.: Quantification of spirochete burden in Borrelia burgdorferi infected ticks fed on OspA immunized mice by 16S rRNA RT real-time PCR Randolph S et al.: Epidemiological consequences of tick ecology Skarpaas T et al.: Tick-borne encephalitis in Norway Soutschek E et al.: A defined mixture of recombinant antigens from several Borrelia genospecies improves serodiagnosis of Lyme disease von Stedingk L V et al.: Recent research on human babesiosis – the Scandinavian perspective Stuen S et al.: Granulocytic Ehrlichia infection in domestic and wild ruminants in Norway |
|||||||||||||||||
 |
|
last modified
Friday, June 13, 2003
|
 | ||||||||||||||||
