Content uploaded by Bastiaan G. Meerburg
Author content
All content in this area was uploaded by Bastiaan G. Meerburg
Content may be subject to copyright.
Critical Reviews in Microbiology, 2009; 35(3): 221–270
REVIEW ARTICLE
Rodent-borne diseases and their risks for public health
Bastiaan G Meerburg1, Grant R Singleton2, and Aize Kijlstra3,4
1Wageningen University & Research Centre, Plant Research International, Wageningen, e Netherlands, 2International
Rice Research Institute, Metro Manila, Philippines, 3Wageningen University & Research Centre, Animal Sciences
Group, Lelystad, e Netherlands, and 4University of Maastricht, Faculty of Medicine, Department of Ophthalmology,
Maastricht, e Netherlands
Address for Correspondence:Bastiaan G Meerburg, Wageningen University & Research Centre, Plant Research International, P.O. Box 616, 6700 AP
Wageningen, e Netherlands. E-mail: bastiaan.meerburg@wur.nl
(Received 28 November 2008; revised 19 April 2009; accepted 22 April 2009)
Introduction
During the last decades we have seen a rise in human
diseases that are associated with small-mammal res-
ervoirs. In 1998, Mills and Childs (1998) outlined steps
towards understanding the link between vertebrate host
ecology and human disease. A decade has passed since
their study, and therefore it is now time to look back and
see what are the current scientic issues in rodent-borne
diseases and their eects on public health.
e intention of this vast review is to contribute to
a better understanding of rodents, rodent reservoirs,
their contribution to disease and a quantication of
the risks per disease. is information is important for
clinicians, researchers and professionals in the eld. In
Table 1, a summary is provided of the contribution of
rodents to transmission of dierent pathogens. In the
text, more information can be found about each of these
pathogens.
Rodentia is the most abundant and diversied order
of living mammals, representing about 43% of the total
number of mammalian species (Huchon et al. 2002;
Wilson and Reeder 1993). Its species are distributed on
every continent except Antarctica and include many of
the most abundant and taxonomically diverse mammals.
In many places rodents live in close contact with human
populations, their farm animals or pets. In other places,
peri-urban rodents provide a nexus between wildlife
communities and humans, exposing humans to some
zoonoses circulating in these natural ecosystems.
Taking food from our table
Rodents are important competitors globally with humans
for food, particularly through the pre-harvest damage
they cause to cereals (Stenseth et al. 2003). For example,
across Asia, pre-harvest losses of rice range from 5% in
Malaysia to 17% in Indonesia. To put this into perspec-
tive, a loss of 6% in Asia amounts to enough rice to feed
220 million people, roughly the population of Indonesia,
for 1 year. Rat damage is often patchy and family rice
plots small, so it is not uncommon for farmers or villag-
ers to lose half of their entire rice crop to rats (Singleton
2003). On a global scale, it was recently estimated that
almost 280 million undernourished could additionally
benet if more attention were paid to reducing pre- and
post-harvest losses by rodents (Meerburg et al. 2009).
In Southeast Asia, rats are the number one pre-harvest
pest in Indonesia and are in the top three pests in
ISSN 1040-841X print/ISSN 1549-7828 online © 2009 Informa UK Ltd
DOI: 10.1080/10408410902989837
Abstract
Rodents are the most abundant and diversified order of living mammals in the world. Already since the
Middle Ages we know that they can contribute to human disease, as black rats were associated with dis-
tribution of plague. However, also in modern times rodents form a threat for public health. In this review
article a large number of pathogens that are directly or indirectly transmitted by rodents are described.
Moreover, a simplified rodent disease model is discussed.
Keywords: Rodents; rats; mice; disease; health; pathogens
http://www.informapharmascience.com/mcb
222 B G Meerburg et al.
Vietnam. In the uplands of Laos, Myanmar, Vietnam,
and parts of India, rat populations occasionally erupt
and cause massive problems (Singleton 2003). In 2007,
Mizoram, in northeastern India, and Myanmar experi-
enced such an outbreak. In Mizoram, a previous mas-
sive plague in the 1950s led to famine conditions and
triggered a change of government.
In many areas, farmers actually abstain from planting
a second or third rice crop because of the expectation of
severe rodent damage. is “forgone” loss in productiv-
ity is rarely taken into account.
Rodents are also seen as pests because of their gnaw-
ing habit, which can cause economic losses, spoilage
of food and lead to structural damages (Brown et al.
2008). However, rodent presence can also have serious
implications for public and veterinary health. Rodents
are hazardous, as they can amplify pathogens from the
environment and form reservoirs of (zoonotic) disease
(Gratz 1994; Webster and Macdonald 1995).
Rodents as transmitters of pathogens to humans
Rodent-borne diseases can be spread via two dierent
pathways (Figure 1).
e rst pathway is a direct route. Rodents can spread
pathogens to humans, e.g., by biting them or because
humans consume food products or water that is con-
taminated with rodent feces. Moreover, humans can
come in contact with surface water that is contaminated
with rodent urine (e.g., leptospirosis) or we breathe in
germs that are present in rodent excrements (e.g., hanta-
viruses). Also, rodents are sometimes mentioned in rela-
tion to horizontal transmission of pathogens that cause
animal diseases, thus causing huge economic damages
and image losses for animal husbandry. Beside highly
contagious viral pathogens such as classical swine fever,
and foot and mouth disease also bacterial infections
(e.g., Mycobacterium avium) need to be addressed in
this respect.
Rodent-borne pathogens can also be spread indi-
rectly to humans. en, rodents can serve as amplifying
hosts of the pathogens and can bring them into direct
contact with humans by mean of ectoparasitic arthro-
pod vectors (ticks, mites, eas). Rodents that are acci-
dentally or on purpose ingested by livestock can transfer
pathogens which can result in human morbidity if these
food products are not-thoroughly cooked. Moreover,
rodents can help to maintain pathogen transmission
cycles in a number of dierent environments, varying
from densely populated urban areas to rural areas and
in the wilderness.
e number of dierent pathogens to whose life cycle
rodents contribute in one way or another, is impressive
(Singla et al. 2008). e goal of this article is to give a
broad overview of the various diseases that humans may
acquire from infection with these pathogens, although
the authors are aware that the list is non-exhaustive.
Table 1 contains an overview of the dierent diseases
in humans, their impact on the human populations, the
severity of the disease for human health (measured in
its mortality) or the economy (measured in morbidity
and production losses). Clearly, rodents can account for
huge health and economic losses.
While the numbers of people and animals that are
aected are variable in both time and place, the conse-
quences at an individual level can be large. For example,
patients hospitalized in the United Kingdom after a rat
bite had to stay on average 11.2 days (Hospital Episode
Statistics, Department of Health, England, 2002–03).
Moreover, global climate change and changing human
settlement patterns (especially in developing countries)
could lead to increased problems with rodent-borne
pathogens as the distribution of rodent species, arthro-
pods and thus also pathogens linked to these species
could be inuenced (Githeko et al. 2000).
Below we discuss the dierent diseases in humans
(and sometimes in livestock) that may be the result of
pathogen transmission by rodents. An overview of all
the diseases that are discussed in this review is provided
in Table 1.
Viruses
Hantavirus Pulmonary Syndrome
Hantavirus Pulmonary Syndrome was recognized in
1993 in the South Western parts of the United States
as an acute disease caused by several strains of the
related strains of viruses in the genus of Hantavirus,
Bunyaviridae family (Nichol et al. 1993). Most remark-
ably, the sudden appearance of this rodent-borne virus
in the arid US Southwest was accompanied by anoma-
lous weather patterns (Epstein 1995).
Rodent
Human
Rodent
Human
Arthropod Livestock
Food
products
Figure 1. Two dierent pathogen transmission pathways: on the left
the direct route, on the right the indirect route. e pathogen is the
arrow in the owchart.
Rodent-borne diseases and their risks for public health 223
Table 1. Overview of dierent pathogens that may be transmitted by rodents and their consequences.
Disease Agent Carrier/Reservoir Population at-risk Chance
Severity
Human Health Economy
Hantavirus Pulmonary
Syndrome
Virus, Bunyaviridae Carrier 2 1 3 1
Hemorrhagic Fever with
renal syndrome (+ other
hemorrhagic fevers)
Virus, Bunyaviridae Carrier 2 2 2 2
Nephropathia
epidemica
Virus, Bunyaviridae Carrier 1 1 1 1
Crimean-Congo hemor-
rhagic fever
Virus, Bunyaviridae Reservoir 1 1 3 1
Borna disease Virus, Bornaviridae Reservoir 1 1 1 2
Omsk hemorrhagic
fever
Virus, Flaviviridae Reservoir 1 1 1 1
Kyasanur Forest Disease Virus, Flaviviridae Reservoir 1 1 1 1
Apoi Virus Disease Virus, Flaviviridae Unknown Unknown Unknown Unknown Unknown
Tick-borne encephalitis Virus, Flaviviridae Reservoir 2 1 3 1
Powassan encephalitis Virus, Flaviviridae Reservoir 1 1 1 1
Lymphocytic
Choriomeningitis virus
(LCMV)
Virus, Arenaviridae Reservoir 1 1 1 1
Lassa Fever Virus, Arenaviridae Carrier 2 2 3 2
South American
arenaviruses (Junin,
Mapucho etc.)
Virus, Arenaviridae Carrier 2 2 3 1
North American
arenaviruses
Virus, Arenaviridae Carrier 1 1 Unknown Unknown
Colorado Tick Fever Virus, Reoviridae Reservoir 1 1 1 1
Venezuelan quine
encephalitis
Virus, Togaviridae Reservoir 2 2 2 2
Western equine
encephalitis
Virus, Togaviridae Reservoir 1 1 1 1
Hepatitis E Virus, Caliciviridae Reservoir 1 1 1 1
Cowpox Virus, Poxviridae Reservoir/
carrier
1 1 1 1
Contagious viral animal
diseases (Classical
Swine Fever, Foot and
Mouth Disease)
Virus, Picornaviridae
(FMD); Flaviviridae
(CSF)
Reservoir? 0 1 0 3
Leptospirosis (Weils’
disease)
Bacteria,
Spirochaetes
Carrier 2 2 3 2
Lyme disease Bacteria,
Spirochaetes
Reservoir 3 2 1 2
Tick-borne relapsing
fever
Bacteria,
Spirochaetes
Reservoir 2 1 1 1
Scrub typhus Bacteria,
Alphaproteobacteria
Reservoir 2 1 3 1
Murine typhus Bacteria,
Alphaproteobacteria
Reservoir 3 1 1 1
Sylvatic epidemic
typhus
Bacteria,
Alphaproteobacteria
Reservoir 1 1 1 1
Queensland tick typhus
or spotted fever
Bacteria,
Alphaproteobacteria
Reservoir 1 1 1 1
Rocky Mountain spot-
ted fever
Bacteria,
Alphaproteobacteria
Reservoir 1 1 3 1
Rickettsialpox Bacteria,
Alphaproteobacteria
Reservoir 2 1 0 1
Bartonella Illnesses Bacteria,
Alphaproteobacteria
Reservoir 2 2 1 1
Table 1. Continued on next page.
224 B G Meerburg et al.
Table 1. Continued.
Disease Agent Carrier/Reservoir Population at-risk Chance
Severity
Human Health Economy
Human granulocytic
anaplasmosis
Bacteria,
Alphaproteobacteria
Reservoir 2 1 1 1
Q-fever Bacteria,
Gammaproteobacteria
Reservoir 3 2 3 2
Salmonellosis Bacteria,
Gammaproteobacteria
Carrier 3 1 1 3
Tularemia Bacteria,
Gammaproteobacteria
Carrier 2 1 3 1
E. coli 0157/VTEC Bacteria,
Gammaproteobacteria
Carrier 2 1 3 2
Plague (Yersina pestis) Bacteria,
Gammaproteobacteria
Reservoir 2 2 2 2
Campylobacteriosis Bacteria,
Epsilonproteobacteria
Carrier 3 1 1 3
Rat-bite fever and
Haverhill fever
Bacteria, Fusobacteria Reservoir 2 1 3 1
Listeriosis Bacteria, Bacilli Carrier 3 1 3 2
Toxoplasmosis Parasite, Sporozoea Reservoir 3 2 2 3
Babesiosis Parasite, Sporozoea Reservoir 3 2 1 1
Cryptosporidiosis Parasite, Sporozoea Reservoir 3 2 1 3
Chagas disease Parasite,
Zoomastigophorea
Reservoir 3 1 3 2
Leishmaniasis Parasite,
Zoomastigophorea
Reservoir 3 2 3 2
Giardiasis Parasite,
Zoomastigophorea
Reservoir 3 2 1 2
Taeniasis Parasite, Cestoda Reservoir 1 1 1 1
Rodentolepiasis Parasite, Cestoda Reservoir 1 1 1 1
Echinococcosis Parasite, Cestoda Reservoir 2 1 3 1
Schistosomiasis Parasite, Trematoda Reservoir 3 2 1 3
Human fasciolosis Parasite, Trematoda Reservoir 3 1 1 3
Brachylaimiasis Parasite, Trematoda Reservoir 1 1 2 1
Alariasis Parasite, Trematoda Reservoir 1 1 0 1
Echinostomiasis Parasite, Trematoda Reservoir 1 1 0 1
Trichinosis Parasite, Nematoda Reservoir 3 2 1 2
Capillariasis Parasite, Nematoda Carrier 3 1 1 1
Angiostrongylosis Parasite, Nematoda Reservoir 2 1 3 1
Toxascariasis Parasite, Nematoda Carrier 1 2 0 2
Baylisascariasis Parasite, Nematoda Carrier 1 2 1 2
Aelurostrongylosis Parasite, Nematoda Reservoir 0 0 0 0
Amoebic dysentery Parasite, Lobosea Reservoir 3 1 3 1
Neosporosis Parasite, Conoidasida Reservoir 0 1 0 2
Reservoir: rodents harbor disease-causing organisms and thus serve as potential sources of disease outbreaks, but always via a vector (tick,
sand-y etc.)
Carrier: rodent that shows no or limited symptoms of a disease but harbors the disease-causing agent and is capable of passing it directly onto
humans
Population at-risk: focal = 1, regional = 2, more than 2 continents = 3
Chance: chance of contracting the disease (all pathways, not only via rodents): small chance = 1, moderate chance = 2, high chance = 3
Human health: Mortality without treatment <5%=1, 5 to 10% = 2, >10% = 3. No mortality = 0.
Economy: losses in terms of morbidity combined with other losses (e.g. in animal productivity): small losses = 1,
moderate losses = 2, huge losses = 3.
e deer mouse (Peromyscus maniculatus) is the pri-
mary reservoir of the Sin Nombre Virus (SNV) in Canada,
the United States of America and Mexico (Abbott et al.
1999; Boone et al. 1998; Childs et al. 1994; Mills et al.
1998; Otteson et al. 1996; Song et al. 1996). In a recent
study from the United States, a large number (1700)
of individual deer mice were captured and tested for
SNV, revealing an average SNV antibody prevalence of
Rodent-borne diseases and their risks for public health 225
approximately 11% (Lonner et al. 2008). Also other spe-
cies of rodents, such as the California mouse (Peromyscus
californicus), cactus mice (Peromyscus eremicus), har-
vest mice (Reithrodontomys megalotis) and California
voles (Microtus californicus), were found positive when
checked for SNV-antibodies (Bennett et al. 1999),
although cross-reactions with other viruses may have
occurred as these can be encountered in this area
(Rowe et al. 1995). e cotton rat (Sigmodon hispidus),
rice rat (Oryzomys palustris), the white-footed mouse
(Peromyscus leucopus), and the Cloudland deer mouse
(Peromyscus maniculatus nubiterrae) can be competent
hosts for hantaviruses that can cause HPS (e.g., Black
Creek Canal virus (BCCV), Bayou virus (BAYV), New
York virus (NYV), Monongahela virus (MONV) (Peters
and Khan 2002; Rhodes III et al. 2000; Schmaljohn and
Hjelle 1997; Song et al. 1996) in North America. In South
America, HPS can be caused by the long-tailed pygmy
rice rat (Oligoryzomys longicaudatus) which can carry
the Andes virus (ANDV) in Chile, Uruguay and Argentina
(Padula et al. 2004; Torres-Perez et al. 2004), and the
vesper mouse (Calomys laucha) that is also capable to
carry an SNV-like virus (Laguna Negra Virus, LNV) in
Brazil, Paraguay (Chu et al. 2003; Yahnke et al. 2001),
and some parts of Bolivia (Schmaljohn and Hjelle 1997;
Williams et al. 1997; Yahnke et al. 2001). In Bolivia and
Argentina, the presence of LNV is also reported in the
small-eared pigmy rice rat (Oligoryzomys microtis) and
the large vesper mouse (Calomys callosus) (Carroll et al.
2005; Levis et al. 2004). From Honduras it was reported
that the Coues’ rice rat (Oryzomys couesi) (Milazzo et al.
2006) can carry a Bayou-like virus, for which the name
Catacamas virus (CATV) was proposed. In Mexico, Playa
de Oro virus (OROV) was detected in both Coues’ rice rat
(Oryzomys couesi) and the Jaliscan cotton rat (Sigmodon
mascotensis) (Chu et al.). Choclo virus (CHOV) was
reported in fulvous pygmy rice rats (Oligoryzomys fulves-
cens) in Panama, while Lechiguanuas virus (LECHV) was
reported in the same rodent species in Central Argentina
(Maes et al. 2004). Moreover, Rio Mamoré Virus (RMV)
was reported in small-eared rice rats (Oligoryzomys
microtis) in Bolivia and Peru, Bermejo Virus (BRMV)
in Chacoan pygmy rice rats (Oligoryzomys chacoensis),
Oran virus (ORNV) in long-tailed rice-rats (Oligoryzomys
longicaudatus) in North Western Argentina and Maciel
virus (MACV) in dark eld mice (Necromys benefactus)
in Central Argentina (Maes et al. 2004). An overview of
all these dierent locations is provided in Table 2.
Hantavirus pulmonary syndrome (HPS) outbreaks in
humans are reported in Paraguay (Williams et al. 1997),
Argentina (Padula et al. 1998), Chile (Torres-Perez et al.
2004), and Panama (Bayard et al. 2004). Humans can
acquire HPS through inhalation of aerosolized virus par-
ticles, rodent bites, or direct contact with rodent drop-
pings or urine. Beside the general public, several groups
are at higher risks of contracting the disease: mammalo-
gists, public health workers, rodent trappers, farmers,
and military personnel (Childs et al. 1995; Jonsson et al.
2008; Zeits et al. 1997). HPS is characterized by bilateral
interstitial pulmonary inltrates, respiratory compro-
mise usually requiring the administration of supple-
mental oxygen and clinical symptoms resembling those
of acute respiratory distress syndrome (ARDS). HPS can
be divided into two phases: a prodromal phase, which
usually lasts 3–5 days, and a cardiopulmonary stage
marked by diuse pulmonary edema and hypotension
within 2–5 days after the onset of pulmonary symptoms.
e rapid progression of interstitial pulmonary edema
to alveolar edema, with severe bilateral involvement
and the accumulation of pleural eusion, accounts for
the 30–40% mortality associated with HPS (Peters et al.
1999). Due to the high morbidity and mortality (fatal-
ity rate of 30–40%, CDC, Atlanta, http://www.cdc.gov)
and the fact that except from supportive care, no treat-
ment exists for hantavirus infection, hantavirus pul-
monary syndrome is one of the most serious diseases
that rodents can cause in humans, and its impact may
even increase in the future as human-to-human trans-
mission of ANDV was reported from Argentina, caus-
ing high mortality (Padula et al. 1998). As mentioned
before, some authors (Epstein 1995) have claimed that
climatic factors have played an important role in the rst
emergence of the disease in the USA. According to these
authors, the likely scenario in the southwestern United
States is that a period of 6 years of drought caused pine
nuts and grasshoppers to ourish, thereby nourishing
deer mice. en, driven from underground burrows by
oodings caused by heavy rains, an increased popula-
tion of natural hosts enhanced the chance for the virus
to thrive and be passed on (Epstein 1995; Kolivras and
Comrie 2004).
A variety of vaccines has been developed using both
killed virus and recombinant DNA technology (Custer
et al. 2003; Hooper et al. 1999; Hooper et al. 2001; Maes
et al. 2004). Naked DNA vaccines, usually based on
plasmid DNA, have been demonstrated to be promising
vaccine approaches for various viral infections (Maes
et al. 2004), such as ANDV, HTNV, PUUV, SEOV, and
SNV. Currently, there are no therapeutics against HPS,
although several approaches are being explored in order
to change this (Jonsson et al. 2008; Maes et al. 2004).
Hemorrhagic fever with renal syndrome
e genus Hantavirus, Bunyaviridae family, also con-
tains viruses that can cause hemorrhagic fever with renal
syndrome or HFRS (Schmaljohn and Hjelle 1997). is
denotes a group of similar illnesses throughout Eurasia
and adjacent territories (Scandinavia, China, Russia,
Korea, Balkans, Western Europe) (Vapalahti et al. 2003).
226 B G Meerburg et al.
HFRS includes diseases that are alternatively known
as Korean hemorrhagic fever, epidemic hemorrhagic
fever and nephropathia epidemica (Plyusnin et al. 1999;
Schmaljohn and Hjelle 1997). e striped eld mouse
(Apodemus agrarius) distributes the Hantaan virus
(HTNV) in China, Russia, and Korea (Lee et al. 1981; Lee
et al. 1981). In Central and Eastern Europe (Vapalahti
et al. 2003), the yellow-necked eld mouse (Apodemus
avicollis) and the striped eld mouse (Apodemus agrar-
ius) distribute two closely related viruses: the Dobrava-
Belgrade Virus (DOBV, also described as DOBV-Af or
Belgrade virus) and Saaremaa Virus (SAAV, previously
described as DOBV-aa). DOBV (carried by A. avicol-
lis) has been associated with severe HFRS, especially in
the Balkans (Avsic-Zupanc et al. 2000; Brus et al. 2002;
Maes et al. 2004; Sibold et al. 1999; Vapalahti et al. 2003).
e Norway rat (Rattus norvegicus) (Heyman et al.
2004; Reynes et al. 2003) and the black rat (Rattus rat-
tus) (Reynes et al. 2003; Wang et al. 2000) are mentioned
as worldwide spreaders of the Seoul virus (SEOV). In
Korea, the Soochong (SOO) or Amur Virus (Baek et al.
2006) was found in the Korean eld mouse (Apodemus
peninsulae) and was recently also encountered in the
North-Eastern part of China (Jiang et al. 2007).
In Europe, the common vole (Microtus arvalis) car-
ries the Tula virus (Heyman et al. 2002; Plyusnin et al.
1994; Vapalahti et al. 2003). In the Balkans, the pine
vole (Pitymys subterraneus) is also linked with Tula
virus (Song et al. 2002), although there it is considered
to be a spill-over reservoir (which does not contribute
to the spreading of the virus). In Germany, a 43-year-
old man became ill with fever, renal syndrome, and
pneumonia (Klempa et al. 2003). Typing revealed the
presence of neutralizing antibodies against TULV, while
TULV genetic material was detected in common voles
that were trapped in the neighbourhood of the patients’
home. is was the rst case of hemorrhagic fever with
renal syndrome and pulmonary involvement associated
with TULV infection (Klempa et al. 2003). Moreover,
there has been some evidence of human infection with
TULV after a rodent bite (Schultze et al. 2002). However,
evidence that TULV can cause symptoms in humans
remains far from proven.
HFRS is characterized by systemic involvement of
the capillaries and small vessels, which causes capillary
leakage and hemorrhagic manifestations. Renal involve-
ment leading to acute renal dysfunction as a result of
interstitial hemorrhage and interstitial inltrates is
also common. After the prodromal period, the clinical
course of patients with severe disease can be divided
into ve phases: febrile, hypotensive, oliguric, diuretic,
and convalescent (Peters et al. 1999). Approximately
60.000– 150.000 cases of HFRS involve hospitalization,
with the majority (90%) in China, Russia, and Korea.
e fatality rates range from 5–10% if HFRS is caused by
Hantaan virus (Schmaljohn and Hjelle 1997). e mech-
anism of transmission to man indicates a principal role
for respiratory infection from aerosols of infectious virus
from rodent urine, feces, and saliva. Interhuman, sec-
ondary spread of infection does not occur (Tsai 1987).
Climate inuences the emergence and re-emergence
of infectious diseases, which is also happening for HFRS.
In a particular study in China the impact of climatic,
reservoir and occupational variables on the transmis-
sion of HFRS in low-lying parts of the country were
assessed using empirical data over the period 1980–1996.
Table 2. Overview of genus Hantavirus, Bunyaviridae family in New World rodents that are linked to human disease.
Virus Rodent Region
Sin Nombre Virus (SNV) Deer mouse (Peromyscus maniculatus)
Cactus mouse (P. californicus)
Harvest mouse (Rheitrodontomys megalotis)
North America
Black Creek Canal Virus (BCCV)
Bayou Virus (BAYV)
New York Virus (NYV)
Monongahela Virus (MONV)
Cotton rat (Sigmidon hispidus)
Rice rat (Oryzomys palustris)
White-footed mouse (P. leucopus)
Cloudland deer mouse (P. maniculatus)
North America
Andes Virus (ANDV) Long-tailed pigmy rice rat (Oligoryzomys longicaudatus) Chile, Uruguay, Argentina
Laguna Negra Virus (LNV) Vesper mouse (Calomys laucha)
Small-eared pigmy rice rat (O. microtis)
Large vesper mouse (Calomys callosus)
Brazil, Paraguay, Bolivia
Catacamas Virus (CATV) Coues’ rice rat (O. couesi) Honduras
Playa de Oro Virus (OROV) Coues’ rice rat (O. couesi)
Jaliscan cotton rat (S. mascotensis)
Mexico
Choclo Virus (CHOV) Fulvous pygmy rice rat (O. fulvescens) Panama
Lechiguanuas Virus (LECHV) Fulvous pygmy rice rat (O. fulvescens) Central America
Rio Mamoré Virus (RMV) Small-eared pigmy rice rat (O. microtis) Bolivia, Peru
Bermejo Virus (BRMV) Chacoan pygmy rice rat (O. chacoensis) Argentina
Oran Virus (ORNV) Long-tailed pigmy rice rat (O.longicaudatus) Argentina
Maciel Virus (MACV) Dark eld mouse (Necromys benefactus) Argentina
Rodent-borne diseases and their risks for public health 227
e seasonal amount of precipitation, the density of
mice and the level of crop production in autumn could
be used as predictors of HFRS transmission in the low-
lying area of HFRS foci (Bi et al. 2002).
Ribavirin was tested for ecacy in HFRS patients in
China and shown to have a statistically signicant bene-
cial eect if initiated early in the disease course (Huggins
et al. 1991; Jonsson et al. 2008). Other compounds, such
as 1--d-ribofuranosyl-3-ethynyl-[1,2,4]triazole (ETAR)
seem also promising (Chung et al. 2008).
In North-Western Europe where HFRS is absent,
there are hantavirus types that cause a mild variant of
this disease. Nephropathia epidemica is a virus-infection
caused by the Puumala virus (PUUV). e Puumala
virus is spread by the bank vole (Myodes glareolus,
earlier Clethrionomys) in Europe (Sibold et al. 1999),
Scandinavia (Olsson et al. 2005), and Russia.(Bernshtein
et al. 1999; Lundkvist et al. 1997), while the grey-sided
vole (Myodes rufocanus) was associated with Puumala
virus in Japan (Kariwa et al. 1995). For Puumala, there
are also signs that the virus can survive for prolonged
times outside the host, thus causing indirect transmis-
sion via the environment (Kallio et al. 2006). About
80% of the infected individuals are asymptomatic or
develop only mild symptoms, and the disease does not
spread from human to human. Mortality rate is below
1%. e incubation period of Nephropathia epidemica is
between two and six weeks. It has a sudden onset with
fever, headache, back pain and gastrointestinal symp-
toms, but sometimes worse symptoms such as internal
hemorrhage can occur and it can even lead to death.
e bank vole is the reservoir for the virus, which is con-
tracted from aerosolized droppings. Recently, human
cases of Nephropathia epidemica due to Puumala virus
infection in Europe have increased (Tersago et al. 2008).
Following the hypothesis that high reservoir host abun-
dance induces higher transmission rates to humans,
explanations for this altered epidemiology must be
sought in factors that cause bank vole (Myodes glare-
olus) abundance peaks (Tersago et al. 2008), such as the
predator-prey cycles in Northern-Europe (Hanski and
Henttonen 1996; Hanski et al. 2001) and mast years with
heavy seed crops of oak and beech in Central-Europe
(Jensen 1982; Jensen 1985).
Other hantaviruses not yet linked with human disease
Rodents also carry a number of other hantaviruses,
although these are not (yet) linked to disease in humans.
For example, in Africa, the African wood mouse
(Hylomyscus simus) carries the Sangassou virus (Klempa
et al. 2006), the bandicoot rat (Bandicota indica) carries
the ailand virus (Hugot et al. 2006; Pattamadilok et al.
2006), the meadow vole (Microtus pennsylvanicus) car-
ries Prospect Hill virus, the reed vole (Microtus fortis)
the Khabarovsk (KBRV) virus in Russia (Horling et al.
1996) and the western harvest mouse (Reithrodontomys
megalotis) El Moro Canyon (ELMC) virus in the USA and
Mexico (Calisher et al. 2005).
e impact of these hantaviruses on human health
is yet unknown. Moreover, there will also be other
hantaviruses in circulation that are not yet discovered
and hantaviruses are constantly coevoluting with their
hosts. Also, new hosts are discovered that are also
capable to carry hantaviruses: such as happened with
insectivores (Okumura et al. 2007; Song et al. 2007). As a
consequence, more research is needed on the presence
and distribution of hantaviruses in the world and their
potential impact on human health.
Crimean-Congo hemorrhagic fever
Crimean-Congo hemorrhagic fever (CCHF) is a tick-
borne hemorrhagic fever with documented person-to-
person transmission and a case-fatality rate of between
3-30% (Ergonul 2006). Ticks of the genus Hyalomma
are responsible for the transmission of the Crimean-
Congo hemorrhagic fever virus, genus Nairovirus, family
Bunyaviridae. is widespread virus has been found
in Africa, Asia, the Middle East, and Eastern Europe.
Outbreaks have been reported in many countries,
including the former Soviet Union, China, Pakistan, Iraq,
Iran, South Africa, Mauritania, Uganda, Burkina Faso,
the Democratic Republic of Congo, Kosovo, Albania,
Bulgaria, and Turkey (Ergonul 2006). According to this
author, over 3400 humans were aected during these
outbreaks since the 1940s (Ergonul 2006), mainly agri-
cultural workers. e length of the incubation period
for the illness depends on the acquisition pathway. It
can vary between 1–9 days (usually between 1–3 days)
in case of infection via a tick bite to 5–13 days (but usu-
ally between 5–6 days) following contact with infected
tissues or blood.
According to the WHO (WHO, Geneva, http://www.
who.int), the onset of symptoms is sudden, with fever,
myalgia (aching muscles), dizziness, neck pain and
stiness, backache, headache, sore eyes, and photo-
phobia. ere may be nausea, vomiting, and sore throat
early on, which may be accompanied by diarrhoea and
generalised abdominal pain (Ergonul 2006).
en, the patient can experience sharp mood swings,
and may become confused and aggressive. After two to
four days, the agitation may be replaced by sleepiness,
depression and lassitude, and the abdominal pain may
localize to the right upper quadrant, with detectable
hepatomegaly (liver enlargement). Other clinical signs
which emerge include tachycardia, lymphadenopathy,
and a petechial rash, both on internal mucosal sur-
faces, such as in the mouth and throat, and on the skin.
e petechiae may give way to ecchymoses and other
228 B G Meerburg et al.
hemorrhagic phenomena such as melena, hematuria,
epistaxis, and bleeding from the gums (Swanepoel et al.
1989). ere is usually evidence of hepatitis. e severely
ill may develop hepatorenal and pulmonary failure after
the fth day of illness (Swanepoel et al. 1989).
Various rodents have been found to carry antibod-
ies to the virus (Nalca and Whitehouse 2007; Shepherd
et al. 1987), and can be a host for the immature stages
of the tick vectors (Ergonul 2006). us, rodents have an
important role in the tick-cycle and the conservation of
the pathogen. ere is no evidence of direct transmis-
sion of the pathogen between rodents and humans.
According the fact sheet of Crimean-Congo hem-
orrhagic fever on the WHO website (www.who.int),
general supportive therapy is the mainstay of patient
management. Intensive monitoring to guide volume
and blood component replacement is required. e
antiviral drug ribavirin has been used in treatment
of established CCHF infection with apparent benet
(Fisher-Hoch et al. 1995). Both oral and intravenous
formulations seem to be eective (Fisher-Hoch et al.
1995; Saijo et al. 2004).
An inactivated mouse brain-derived vaccine against
CCHF has been developed and used on a small scale in
Eastern Europe (Papa et al. 2002), but there is no safe
and eective vaccine widely available for human use.
Control with acaricides (chemicals that kill ticks) is the
only realistic preventive option.
Borna disease
Borna disease virus (BDV) is an enveloped virus with
a negative-stranded non-segmented RNA genome,
which has been classied as the prototype virus of a
newly established family, Bornaviridae, within the order
Mononegavirales ( Staeheli et al. 2000). BDV infections can
result in neurological disease that mainly aects horses
and sheep in certain areas of Germany, Switzerland,
Austria, and the Principality of Liechtenstein (Staeheli
et al. 2000). ere are also reports from Finland that the
Borna virus is prevalent in wild mammals (Kinnunen
et al. 2007). e detection of BDV-specic antibodies in
psychiatric patients (Bode et al. 1995; de la Torre et al.
1996; Rott et al. 1985) suggests that Borna disease (BD)
may be a zoonosis.
Rodents are mentioned as a possible source of virus
transmission to farm animals (Durrwald et al. 2006;
Sauder and Staeheli 2003). Experimental infection of
rodents has resulted in persistence of BDV and is associ-
ated with the presence of viral gene products in rodent
saliva, urine, and feces (Sauder et al. 1996). us, animal
feed contaminated with urine of persistently infected
rats or other rodents represent a source of infectious
BDV (Sauder and Staeheli 2003; Staeheli et al. 2000).
e role of rodents in transmission of BDV has also been
mentioned in relation to feline infections in Sweden
(Berg et al. 1998). As the pathways of human infection
remain unclear, more research is needed on this topic.
Omsk hemorrhagic fever
Omsk hemorrhagic fever is a disease that is endemic
in Western-Siberia (Gajdusek 1956; Holbrook et al.
2005). It is caused by Omsk hemorrhagic fever virus
(OHFV), a member of the genus Flavivirus of the fam-
ily Flaviviridae (Li et al. 2004). e virus is transmitted
through direct contact of humans with infected animals
and bites from infected ticks (Dermacentor sp p. ( Li et al.
2004)). In humans, OHFV is the only known tick-borne
avivirus that always causes hemorrhagic disease in
the absence of encephalitis (Lin et al. 2003), although
cases are known where related tick-borne aviviruses
such as Kyasanur Forest Disease and Alkhurma viruses
also occur without encephalitis. Clinical signs of OHF
are fever, headache, myalgia, dehydration, and hem-
orrhage (Pavri 1989). OHF has a case fatality rate of
0.5–3% (Li et al. 2004). ere are no reported cases of
person-to-person transmission or nosocomial spread
of OHFV.
OHFV is maintained in nature through circulation
among ticks and rodents including water voles (Arvicola
terrestris) and muskrats (Ondata zibethica) (Kharitonova
and Leonov 1985; Li et al. 2004). Infection of humans is
frequently associated with occupation, as muskrat trap-
pers are common victims of OHF.
Kyasanur Forest Disease
Kyasanur Forest Disease (KFD) is caused by a OHFV
related tick-borne Flavivirus of the family Flaviviridae
found in forest rodents in south-western India (Solomon
and Mallewa 2001), the Kyasanur Forest Disease Virus
(KFDV). Rats (Rattus rattus) are known hosts for a species
of tick (Haemaphysalis spinigera) that is the main vector
of the pathogen (Saxena 1997), although other tick spe-
cies are capable of transmitting the disease (Hoogstraal
1966). In humans, KFD can cause hemorrhagic fever
(Hoogstraal 1966). Since its discovery in the 1950s,
KFDV has caused epidemic outbreaks of hemorrhagic
fever annually, aecting 100–500 people with a case
fatality rate of 2–10% (Gould and Solomon). However,
the endemic area of KFD is rather limited. Recently, a
variant of KFDV, characterized serologically and geneti-
cally as Alkhurma hemorrhagic fever virus (AHFV), has
been identied in Saudi Arabia (Gould and Solomon;
Priyabrata 2006) and very recently in China (Wang et al.
2009). Tick control could be a preventive measure for
reducing the transmission of KFD to humans (Ghosh
et al. 2007). Moreover, a vaccine candidate against KFD
has been developed (Dandawate et al. 1994).
Rodent-borne diseases and their risks for public health 229
Apoi Virus Disease
A relative of Kyasanur Forest Disease Virus, Apoi Virus
(APOIV), was isolated from spleens of healthy Apodemus
mice in Japan (Karabatsos 1995). Infection causes
encephalitis in newborn mice. A case of laboratory
infection resulted in CNS signs, fever, headache, myal-
gia and arthralgia with residual leg paralysis (Karabatsos
1995). No vector could be identied. Exact risk for public
health remains unknown.
Tick-borne encephalitis
Tick-borne encephalitis virus (TBEV) is a zoonotic arbo-
virus infection of the tick-borne avivirus group, family
Flaviviridae, genus Flavivirus. e TBEV species consists
of three sub-types, namely Far Eastern, Siberian (previ-
ously west-Siberian) and European (previously Central
European Encephalitis, CEE virus) (Gritsun et al. 2003).
e European lineage is transmitted by the tick Ixodes
ricinus and both the Far Eastern and Siberian subtypes
by the tick Ixodes persulcatus (Han et al. 2005).
TBEV is endemic in Russia and Eastern and Central
Europe (Dumpis et al. 1999), China (Lu et al. 2008) and
in Western Europe and Scandinavia, resulting in 11,000
annual cases of tick-borne encephalitis (TBE) in Russia
and 3,000 cases in the rest of Europe (Gritsun et al. 2003).
Several rodents such as the bank vole (Myodes glare-
olus), eld mouse (Apodemus agrarius Pallas), and red
voles (Myodes rutilus Schreber) (Bakhvalova et al. 2006;
Weidmann et al. 2006) are known hosts of ticks that can
cause TBE. Rodents play an essential role in the trans-
mission of the tick-borne encephalitis virus between
ticks (Randolph et al. 1999). ese natural hosts have
neutralizing antibodies to TBEV and no detectable
viremia. However, they can still support virus trans-
mission between infected and uninfected ticks feeding
closely together on the same animal (Labuda et al. 1997).
TBEV is transmitted to humans usually by the bite of a
tick (either Ixodes persulcatus or Ixodes ricinus); occa-
sionally, cases occur following consumption of infected
unpasteurized milk (Dumpis et al. 1999).
Transmission is seasonal and occurs in spring, sum-
mer or autumn, particularly in rural areas favored by
the tick vector: e.g., in forest foci with enhanced moist
vegetation where mammals might provide a blood meal
for them. TBEV can cause acute central nervous system
disease, which may result in death or in 30% of the cases
in long-term neuropsychiatric sequelae. TBEV can pro-
duce a variety of clinical symptoms: febrile, meningeal,
meningoencephalitic, poliomyelitic, polyradiculone-
uritic, and chronic (Gritsun et al. 2003). Usually the
febrile form is the rst infection phase, after which the
second phase with neurological manifestations may or
may not occur. Incubation time is between 7 and 14 days
(Gritsun et al. 2003). Mortality rates vary from 0.5-2% for
the European TBEV-genotype, to mortalities between
20-35% or even higher for Far Eastern TBEV-genotypes
(Anonymous 2006; Haglund and Gunther 2003; Lu et al.
2008). Eective vaccines against TBE are reported in
Europe (Gritsun et al. 2003), but are also available in
China and Russia. Some have mentioned that the risk
for travelers of acquiring TBEV has increasing with
the recent rise in tourism to areas of endemicity dur-
ing spring and summer (Dumpis et al. 1999), although
in some areas the peak of the amount of cases is in the
autumn.
Powassan encephalitis
Powassan virus (POW) is a North American tick-borne
avivirus (Hoogstraal 1966; McLean et al. 1970), related
to the TBE virus in Eurasia (Gholam et al. 1999). It was
rst isolated from a patient with encephalitis in 1958.
From 1958–1998, 27 human POW encephalitis cases
were reported from Canada and the northeastern
United States (McLean et al. 1970; Ralph 1999). From
September 1999–July 2001, four Maine and Vermont res-
idents with encephalitis were found to be infected. e
incubation period is on average 7–14 days. It is dicult
to distinguish the symptoms of Powassan encephalitis
from those caused by the herpes simplex virus (Ralph
1999). Moreover, it is likely that not every infection leads
to disease: serologic surveillance studies in Canadian
communities showed positive tests in up to 3% of the
population, suggesting that infection without encepha-
litis can occur in humans (McLean et al. 1962).
e vector of POW is the woodchuck tick (Ixodes
cookei), which is hosted by a variety of rodents (mar-
mots, sciurid rodents, groundhogs) (Hardy et al. 1974;
Hoogstraal 1966), but also weasels, skunks, raccoons,
coyotes, and foxes (Johnson 1987; Main et al. 1979;
Ralph 1999). Moreover, the scope of transmission of
the virus may be broadened by domestic cats and dogs,
which can act as harbingers of infected ticks and thereby
expose humans (Ralph 1999).
Transmission of the pathogen to humans can take
place by bites from an infected tick or mite (Ralph 1999)
or by consumption of food products from infected
animals (e.g., raw milk) (Woodall and Roz 1977). As
rodents play a role in the life cycle of the tick, Powassan
encephalitis was incorporated in this review, although it
is dicult to quantify the exact contribution of rodents
to human morbidity and mortality.
Lymphocytic choriomeningitis
Lymphocytic choriomeningitis virus (LCMV) is a
rodent-borne virus belonging to the family Arenaviridae,
genus Arenavirus. LCMV can cause a variety of human
230 B G Meerburg et al.
diseases which are known as lymphocytic choriomen-
ingitis (LCM), ranging in severity from u-like illness
to meningitis and encephalitis (Lourdes Lledó 2003).
Moreover, intrauterine infections also occur and can
result in chorioretinitis, hydrocephalus, microcephaly
or macrocephaly, mental retardation, and fetal death
(Barton and Hyndman 2000; Jahrling and Peters 1992;
Mets and Chhabra 2008; Rawlinson et al. 2008). Since
1955, 54 cases of congenital LCMV have been reported,
with 34 of the cases diagnosed since 1993 (Jamieson
et al. 2006).
Infections have been reported in the Americas,
Europe, Australia, and Japan. Seroprevalence studies
in humans have shown that the prevalence of LCMV in
humans lies between 2–5%, which suggest that many
cases are clinically inapparent. e disease is contracted
by humans through breathing air that is contaminated
with rodent excrements, especially from the domestic
house mouse Mus domesticus (Childs et al. 1991; Childs
et al. 1992). A case-report from France in which LCMV
was detected in 14 out of 20 mice trapped at a patient’s
home reports that patient and mouse LCMVs were iden-
tical (Emonet et al. 2007). However, this LCMV strain
was highly divergent from previously characterized
LCMV (Emonet et al. 2007). Also, laboratory animals
(Dykewicz et al. 1992) and pet rodents (Amman et al.
2007) can transfer the disease to humans.
Infected wild rodents can be encountered in several
areas of the world, including Argentina (Laura Riera
2005), the USA (Childs et al. 1992), and Spain (Lourdes
Lledó 2003). Currently, there is no specic treatment
available against LCM, which is especially problematic
in pregnant women. As ribivarin works against LCMV in
cell culture, this is recommended under specic circum-
stances, such as presentations resembling hemorrhagic
fever or in immunosuppressed cancer patients (Peters
1994). e great majority of LCMV-infected patients sur-
vive without residua and the mortality rate is about 1%
(Jahrling and Peters 1992). Interestingly, there have been
recent reports of human to human transmission through
organ transplants that have led to severe complications
for recipients of organs carrying LCMV (Anonymous
2008; Fischer et al. 2006; Palacios et al. 2008).
Lassa fever
Lassa fever is an acute viral illness that occurs endemically
in West Africa and a signicant cause of mortality and
morbidity. e disease is caused by a single-stranded
RNA virus (Lassa Virus, LASV) belonging to the virus
family Arenaviridae. e disease was isolated in the
multimammate rat (Mastomys natalensis) (Fichet-Calvet
et al. 2008; Green et al. 1978; Johnson et al. 1981; Meulen
et al. 1996; Monath et al. 1974), a rodent with a perido-
mestic dispersal pattern. Humans acquire the infection
by breathing air that is contaminated with rodent excre-
ments, by direct contact with rodent droppings or urine,
by consumption of food that is contaminated by rodents,
by bite wounds, or by close contact with other humans
that have contracted the disease.
e number of LASV infections per year in West Africa
is crudely estimated at 100,000 to 300,000, with approxi-
mately 5,000 deaths (Khan et al. 2008). In some areas of
Sierra Leone and Liberia, it is known that 10–16% of peo-
ple admitted to hospitals have Lassa fever, which indi-
cates the serious impact of the disease on the population
of this region (Birmingham and Kenyon 2001). Fatality
rate lies between 15–20%. Lassa fever occurs in all age
groups and in both men and women, but death rates
are highest for pregnant women in the third trimester of
their pregnancy and unborn children in uteri of infected
mothers. Lassa fever begins after an incubation period
of 7–18 days, with fever, weakness, malaise, and severe
headache and throat ache (Fisher-Hoch and McCormick
2001). Up to a third of the hospitalized patients progress
to a prostrating illness 6–8 days after onset of fever, with
persistent vomiting and diarrhea (Fisher-Hoch and
McCormick 2001). Bleeding is only seen in 15–20% of
the patients, primarily aecting the mucosal surfaces
(Fisher-Hoch and McCormick 2001).
Persons at greatest risk of contracting the virus are
those living in rural areas where Mastomys are usu-
ally encountered, especially in areas of poor sanitation
or crowded living conditions. Estimates of antibody
prevalence range from 4–6% in Guinea to 15–20% in
Nigeria, though in some villages in Sierra Leone as
many as 60% of the population have evidence of past
infection (Fisher-Hoch and McCormick 2001). Health
care workers are at risk if proper barrier nursing and
infection control practices are not maintained (Fisher-
Hoch et al. 1985).Ribivarin has been successfully used
in Lassa Fever patients in combination with supportive
care, but also new anti-arenavirus drugs are still being
developed as fatalities still occur even while ribi-
varin is used (Khan et al. 2008). A number of groups
are currently developing a vaccine for Lassa fever
(Fisher-Hoch et al. 2000; Fisher-Hoch and McCormick
2001; Fisher-Hoch and McCormick 2004; Geisbert
et al. 2005).
Other
Beside Lassa virus (LASV) and lymphocytic choriomen-
ingitis virus (LCMV) the lymphocytic choriomeningitis-
Lassa (Old World) complex also includes the Ippy virus
(IPPV), the Mobala virus (MOBV), and the Mopeia
virus (MOPV); viruses that have not yet been associated
with human morbidity. IPPV was found in Arvicanthus
spp. (e.g., Nile grass rats) in Central Africa, MOBV was
reported from soft-furred rats (Praomys spp.) in the
Rodent-borne diseases and their risks for public health 231
Central African Republic, while MOPV is present in
multimammate mice, M. natalensis in South Africa.
Recently, also a novel ‘‘killer’’ arenavirus was found in
South Africa, which was directly associated with rodent
presence (Zeller et al. 2008).
South American hemorrhagic fevers
Of all 15 New World (North and South American)
arenaviruses only ve South American arenaviruses
can cause hemorrhagic fevers. ese are Bolivian
Hemorrhagic Fever (caused by Machupo virus, MACV),
Argentinean Hemorrhagic Fever (caused by Junin
virus, JUNV), Venezuelan Hemorrhagic Fever (caused
by Guanarito virus, GTOV), Tacaribe fever (caused by
Tacaribe virus (Salazar-Bravo et al. 2002), TACV), and
Sabia Hemorrhagic Fever (Sabia virus, SABV) in Brazil
(Gonzalez et al. 1996). Transmission of four of these
South American hemorrhagic fever viruses (MACV,
JUNV, GTOV, SABV) are associated with a primary rodent
reservoir, and their transmission to humans is believed
to involve mechanisms similar to human infection with
hantaviruses in North-America (Doyle et al. 1998).
Comparisons of arenavirus phylogeny with rodent host
phylogeny and taxonomic relationships provide several
examples in which virus–host cospeciation is potentially
occurring (Bowen et al. 1997): this means that host and
pathogen have adapted to each other. Both arena- and
hantaviruses have coevolved with their rodent hosts.
In Argentina, JUNV was encountered for the rst time
in 1958 in the small vesper mouse (Calomys laucha) and
the corn mouse (Calomys musculinus) (Doyle et al. 1998;
Mills et al. 1991; Mills et al. 1992). In another study, it
was found that seropositive C. musculinus were pre-
dominantly males in the oldest age and heaviest body
mass classes, and that these animals were twice as likely
to have body scars as seronegative males. is suggests
that most infections were acquired through horizontal
transmission among dominant males through scar-
ring and biting (Mills et al. 1994). Other rodent species
that can be aected are the grass eld mouse (Akodon
azarae) and the dark eld mouse (Bolomys obscurus). As
all of these rodents mainly live in grasslands, cultivated
elds and hedgerows, the highest chance for contact of
humans with these animals is during eld work. Human
activity (e.g., changes in land use) can have eects on the
number of cases. e expension of maize agriculture has
favored the growth of corn mouse populations and there
has been a correlated increase in human infections with
JUNV (De Villafañe et al. 1977).
In Bolivia, in 1963 the rst evidence for existence of a
potential non-human reservoir of MACV infection was
found (Johnson et al. 1965). Virus strains were recovered
from tissue samples of the wild vesper mouse Calomys
callosus captured in the area of San Joaquín, Bolivia
(Johnson et al. 1965). is rodent has a peridomestic life-
style, with its main habitat in grasslands. Human contact
with this animal takes place primarily in houses.
Venezuelan hemorrhagic fever (VHF) was initially rec-
ognized as a distinct clinical entity in 1989 and the etio-
logic agent, GTOV, was isolated and identied as a novel
arenavirus in 1991 (Weaver et al. 2000). Epidemiologic
eld studies in the VHF-endemic region of western
Venezuela indicated that two grassland rodent species,
the cane mouse (Zygodontomys brevicauda) and the
cotton rat (Sigmodon alstoni), are natural hosts of the
virus (Fulhorst et al. 1999; Tesh et al. 1993). e main
habitat of this rodent is in brushes and grasslands, and
it is thought that most people acquire infection when
working there (De Manzione et al. 1997). For SABV
that was rst isolated in 1990, no wild rodent reservoir
is known yet, although the disease was associated with
several human cases (including a laboratory worker in
Connecticut) (Doyle et al. 1998). For TACV, the wild res-
ervoir consists of bats of the genus Artibeus.
After an incubation period of 7–14 days, all South
American Arenavirus Hemorrhagic Fevers (AHF) begin
insidiously with progressive development of fever, chills,
malaise, anorexia, myalgia and sore throat. As the dis-
ease progresses, patients develop weakness, arthralgia,
back pain, nausea, vomiting, epigastric pain, dizzi-
ness, conjunctivitis, ushed face, and bleeding gums.
By the 6th or 7th day, the patients are usually acutely ill
with dehydration, disorientation and frequently hem-
orrhagic and/or neurological manifestations (Tesh
2002). If death occurs, it usually results from massive
hemorrhage and/or shock. If the patient survives this
period, improvement begins about the 10th or 12th day
and a slow convalescence ensues e mortality rates in
patients with untreated AHF range from about 10 to 33%
(Tesh 2002). Mortality rates of AHF decrease with early
hospitalization and intensive supportive care, including
immune human plasma or the antiviral drug ribavirin
(Tesh 2002). Moreover, an eective vaccine against AHF
is available (Enria and Maiztegui 1994; Enria and Barrera
Oro 2002).
Beside viruses that can cause hemorrhagic fevers,
rodents are also linked to a number of arenaviruses
that have not (yet) been associated with human dis-
ease. Examples are the Pirital virus (PIRV), which was
recently isolated from the cane mouse (Zygodontomys
brevicauda) and the cotton rat (Sigmodon alstoni) in
Venezuela (Fulhorst et al. 1999; Weaver et al. 2000),
the Amapari (AMAV) virus that was encountered in
the rice rat (Oryzomys goeldii) and the bristly mouse
(Neacomys guianae) in the tropical forests of Brazil,
the Oliveros virus (OLIV) of Northern Argentina that
was found in the dark bolo mouse (Bolomys obscurus),
the Parana virus (PARV) in the Paraguayan rice rat
(Oryzomys buccinatus), the Latino virus (LATV) in the
232 B G Meerburg et al.
large vesper mouse (Callomys callosus) in Bolivia, the
Flexal virus (FLEV) in Brazil (Oryzomys spp.), Pichindé
(PICV) virus in the Tomes’s rice rat (Oryzomys albigu-
laris) in Colombia (Peters 1998) and Allpahuayo virus
(ALLV) in arboreal rice rats (Oecomys bicolor, Oecomys
paricola) in the Peruvian Amazon (Moncayo et al.
2001). It is unknown whether all of these arenaviruses
have the potential to cause human disease under natu-
ral conditions.
North American arenaviral diseases
Until the mid 1990s the sole arenavirus that was recog-
nized in North America was the Tamiami virus (TAMV),
which was observed in the cotton rat Sigmodon hispidus
in Southern Florida (Calisher et al. 1970). From 1996
onwards the Whitewater Arroyo virus (WWAV) was
encountered in the white-throated woodrat (Neotoma
albigula) in New Mexico (Calisher et al. 2001; Fulhorst
et al. 1996). Later, antibodies to the virus were also
found in southern plains woodrats (Neotoma micro-
pus) in Texas (Fulhorst et al. 2002) and in the Riparian
wood rat (Neotoma fuscipes), the desert wood rat
(Neotoma lepida), the brush mouse (Peromyscus boylii),
the California mouse (Peromyscus californicus), the
cactus mouse (Peromyscus eremicus), the deer mouse
(Peromyscus maniculatus), and the Western harvest
mouse (Reithrodontomys megalotis) in Southern
California (Bennett et al. 2000).
While TAMV does not seem to have implications
for human health, the WWAV was linked to the death
of a 14-year-old girl in Northern California and two
other fatal cases in the same region (Anonymous 2000;
Enserink 2000). ese patients presented with febrile
illness and respiratory distress, with two developing
hemorrhagic symptoms and liver failure (Reignier et al.
2008). It was thought that human infection might occur
when humans inhale aerosolized rat urine (Anonymous
2000; Enserink 2000). However, only one victim reported
a possible contact with rodent droppings before becom-
ing ill, and no subsequent reports have appeared con-
rming the association of WWAV with these or any other
human infections (Reignier et al. 2008).
In 2002, a new arenavirus in Northern America was
discovered: the Bear Canyon virus or BCNV (Fulhorst
et al. 2002). is virus was rst found in California in the
California mouse (Peromyscus californicus) (Fulhorst
et al. 2002), but the principal host is the large-eared
woodrat (Neotoma macrotis). is thus represents a
successful host-jumping event of the Bear Canyon Virus
from the large-eared woodrat to the California mouse
(Cajimat et al. 2007). More recently, in Texas, a close
relative of Whitewater Arroyo virus was discovered in
southern plains woodrats (Neotoma micropus): the
Catarina virus or CTNV (Cajimat et al. 2007).
In Arizona, an arenavirus was isolated from a cap-
tured wild Mexican woodrat (Neotoma mexicana) that is
considered a strain of a novel species, the Skinner Tank
Virus (SKTV) (Cajimat et al. 2008). Very recently, 2 novel
virus species were encountered from white-throated
woodrats (Neotoma albigula) captured in Arizona, Big
Brushy Tank virus (BBTV) and Tonto Creek virus (TTCV)
(Milazzo et al. 2008).
In general, the implications of these viruses for public
health and their association with neotomine or sigmo-
dontine rodents in North America are subject of ongoing
research (Cajimat et al. 2007).
Colorado Tick Fever
Colorado Tick Fever Virus (CTFV) is a double-stranded
RNA-virus (arbovirus) representing the genus Coltivirus
of the family Reoviridae (Attoui et al. 2005; Leiby and
Gill 2004). Field studies in small mammals in the Rocky
Mountains, USA established that Eutamias minimus
and Spermophilus lateralis were the most important
hosts for CTFV and were the source of virus for imma-
ture stages of the tick vector, Dermacentor andersoni
(Bowen et al. 1981; Carey et al. 1980). Also porcupines
(also a member of the Rodentia) are an important host
(McLean et al. 1993). Beside the main tick vector, also
other tick species (D. occidentalis, D. albopictus, D.
arumapertus, Haemaphysalis leporispalustris, Otobius
lagophilus, Ixodes sculptus, and I. spinipalpis) can be
infected with the virus (Attoui et al. 2005). Humans can
acquire infection by bites from a tick through infected
saliva. Seasonal occurrence of infections is correlated
with the presence and host-seeking activities of the tick
vector. ere is evidence of an endemic area where CTFV
circulates between ticks and mammalian hosts, namely
in the Western United States and in Western Canada
(British Columbia). In the past, unconrmed reports of
CTFV presence in natural cycles came from other parts
of the USA (Long Island) and eastern Canada (Florio
et al. 1950; Newhouse et al. 1964).
In Colorado, at the end of the 1970s about 180 cases
of human infection with CTFV were reported each year
(op cit. Carey et al. 1980). However, actual incidence is
unknown owing to insucient physician knowledge of
the disease, diculties in establishing the diagnosis,
and deciencies in, or for some diseases absence of,
reporting systems (Walker 1998).
Although subclinical infections might occur (Carey
et al. 1980; Emmons 1988), most infections result in
mild to moderately severe symptoms rst appearing
several days or 2 weeks post-infection. Onset is typically
sudden, with chilly sensations, high fever, severe head-
ache, retrobulbar pain, photophobia, lethargy, myalgia,
and arthralgia (Emmons 1988). e spleen and liver
are sometimes palpable. ere may also be anorexia,
Rodent-borne diseases and their risks for public health 233
nausea, vomiting, abdominal pains, neurologic or
encephalitic signs (disorientation, hallucinations, and
sti neck), and a variety of other rare or unusual compli-
cations (Emmons 1988). Mother-to-child transmission
was reported in pregnant women (Attoui et al. 2005). In
children, severe encephalitis and hemorrhages do occur
and viremia can be persistent up to 120 days because
of virus survival within maturing erythrocytes (Hughes
et al. 1974). Pericarditis and myocarditis have also been
reported (Emmons 1988). Fatalities seldom occur, espe-
cially since PCR techniques were developed that allow
early diagnosis (Klasco 2002). Under specic clinical
conditions, provision of ribavirin to patients can be con-
sidered (Klasco 2002). Symptomatic treatment includes
acetaminophen for relief of fever and pain (Attoui et al.
2005).
Patients infected with CTFV show long-lasting immu-
nity. An experimental vaccine was developed in the
1960s and produced long-lasting immunity, but produc-
tion was stopped in the 1970s (Attoui et al. 2005).
Venezuelan equine encephalitis (VEE)
Venezuelan equine encephalitis virus (VEEV) is an
emerging mosquito-borne pathogen of equids and
humans that occurs in the USA (mainly in Texas and
Florida), Central America (Mexico, Panama, Guatemala)
and parts of South America (Peru, Colombia,Venezuela)
(Aguilar et al. 2004). e virus is a member of the
Togaviridae group, genus Alphavirus (Aguilar et al.
2004; Lukaszewski and Brooks 2000), and a number of
dierent variants have been identied (Rico-Hesse et al.
1995). Since its discovery in the 1920s VEEV has caused
periodic epidemics among human beings and equines
in Latin America from the 1920s to the early 1970s
(Weaver et al. 1996). In 1973 there was a large outbreak
of VEE in Venezuela (Weaver et al. 1996). Consequences
of these outbreaks for both human and domestic ani-
mal populations can be high: an outbreak in Colombia
in 1995 caused an estimated 75,000 human cases, 3000
with neurologic complications and 300 fatal (Rivas et al.
1997). Of the state’s estimated 50,000 equines, 8% may
have died (Rivas et al. 1997).
Overall, human death rates have generally been
estimated at approximately 0.5% during these epidem-
ics, with most of the neurologic disease and fatal cases
reported in children (Aguilar et al. 2004). e worldwide
number of human cases of VEE is estimated to exceed
100,000 (Ferro et al. 2003).
Rodents are the reservoir host for VEEV. In the USA
(Chamberlain et al. 1964)and Panama (Grayson and
Galindo 1968), wild cotton rats (Sigmidon hispidus) are
a host species for the mosquito (Culex species) that can
also transfer the virus to humans. In Venezuela, anti-
bodies were present in the Guaira spiny rat (Proechimys
guairae), mouse oppossums (Marmosa spp), and com-
mon oppossums (Didelphis marsupialis) (Salas et al.
2001). In Peru, the recently identied VEEV subtype
IIID strain was isolated from spiny rats (Proechimys
spp.), Culex (Melanoconion) spp., mosquitoes and from
a patient with fever, chills, and malaise (Aguilar et al.
2004). Besides contributing as reservoir hosts to the
transmission cycle, the further role of rodents seems
limited.
Currently, trials with a vaccine candidate to prevent
infection with VEEV in horses look promising (Fine et al.
2007). Moreover, a series of nonclinical studies showed
a vaccine candidate to be eective in protecting rodent
and nonhuman primates against virulent challenge
with several subtypes of VEEV (Fine et al. 2007). Others
also report positive results in the battle against VEE
(Lukaszewski and Brooks 2000; Riemenschneider et al.
2003). ese studies could lead to the development of an
eective human vaccine against the disease.
Western equine encephalitis
e Western equine encephalitis virus (WEEV) is an
Alphavirus in the family of Togaviridae. It can cause the
relatively rare viral disease Western equine encephali-
tis (WEE). Transmission patterns are similar to that of
VEEV: the virus is transmitted by mosquitoes of the gen-
era Culex and Culiseta. Especially gray and California
ground squirrels were mentioned as host species for
these mosquitoes (Hardy et al. 1974). More recently,
antibodies were found in the cotton mouse (Peromyscus
gossypinus), and the cotton rat (Sigmidon hispidus) in
Florida (Day et al. 1996).
WEEV can cause sporadic and epidemic equine and
human CNS infections (Castorena et al. 2008; Earnest
et al. 1971). ere have been about 700 conrmed human
cases in the United States since 1964, but the disease is
also present in countries in South America.
Overall mortality is approx. 4% and is associated
mostly with infection in the elderly. ere is no com-
mercial vaccine against WEEV and there are no licensed
therapeutic drugs in the United States for this infection
(Wu et al. 2007). Consequently, scientists are working
hard to develop a vaccine (Barabé et al. 2007; Wu et al.
2007).
Hepatitis E
Hepatitis E virus (HEV) is a virus in the genus Hepevirus
of the family Hepeviridae (Ahn et al. 2005; Denise Goens
and Perdue 2004). Recent identication of HEV antibod-
ies in pigs (Kase et al. 2008; Kulkarni and Arankalle 2008;
Li et al. 2008; Zhang et al. 2008), horses, ducks (Zhang
et al. 2008), sheep, chickens, and cattle suggested that
animal reservoirs existed for this virus. In a study that
234 B G Meerburg et al.
tried to reveal these reservoirs (Favorov et al. 2000) it was
found that there is a widespread HEV or HEV-like infec-
tion in rodents, especially for those that live in urban
habitat. e highest prevalence of antibody (59.7%) was
found in the genus Rattus (Favorov et al. 2000); the high
prevalence was conrmed in another study (Kabrane-
Lazizi et al. 1999). Also, there are indications that rodents
form a reservoir for HEV in other parts of the world, such
as Nepal (He et al. 2002), where Hodgon’s rat (R. rattus
brunneusculus) and the lesser bandicoot rat (Bandicota
bengalensis) displayed presence of antibodies.
Humans can acquire infection through direct contact
with infected animals or through consumption of meat
of animals that is not thoroughly cooked (Li et al. 2005).
For example, pigs are frequently mentioned as natural
hosts of HEV and are a likely source for human infec-
tion (Clayson et al. 1995; Li et al. 2005; Van der Poel et al.
2001). Transfer of the disease from rodents to these food
animals (e.g., pigs) is a possibility, but how large risks are
remains unknown. In humans, there are reports of verti-
cal transmission (Khuroo et al. 1995; Tsega et al. 1992),
but person-to-person transmission of HEV appears to be
uncommon (Myint et al. 1985). Transfusion-transmitted
infections of HEV due to infected blood do occur
(Matsubayashi et al. 2008). Typical clinical signs and
symptoms in patients with symptomatic HEV infection
are similar to those of other types of viral hepatitis and
include malaise, fever, anorexia, abdominal pain, nau-
sea/vomiting, and hepatomegaly (Khuroo 1980; Myint
et al. 1985). Hepatitis E is fatal in about 2% of all cases,
whilst high case-fatality rates (15–25%) occur among
pregnant women (Tsega et al. 1992).
Cowpox
Cowpox virus (CPXV) is a member of the genus
Orthopoxvirus in the family Poxviridae, which is present
in Eurasia. Being closely related to smallpox viruses and
vaccinia virus (VACV), these viruses all induce cross-
protection against each other. Vaccinia virus is one of the
consequences of the discovery of vaccination by Edward
Jenner, who used CXPV to induce human immunity
against smallpox. In his time, poxvirus occurred natu-
rally in livestock (Moss and Flexner 1987). His vaccine
was passaged initially in humans and subsequently in
cattle and sheep. In some places it was even mixed with
smallpox virus, thus probably leading to the emergence
of the ‘new’ vaccinia virus (Moss and Flexner 1987). In
some regions of the world such as Brazil, this vaccinia
virus has escaped to nature (Da Fonseca et al. 2002), thus
circulating in rodents (Fonseca et al. 1998) and causing
natural outbreaks.
Natural infection and disease with CXPV occurs
primarily in domestic cats (Tryland et al. 1998) (rarely
in man and cattle), and wild rodents are generally
accepted as reservoir hosts (Boulanger et al. 1996).
Antibodies have been detected in wild ground squir-
rels (Citellus fulvus) and gerbils (Rhombomys opimus,
Meriones libicus, and Meriones meridianus) in Georgia
(Tsanava et al. 1989) and Turkmenistan (Marennikova
et al. 1977), in root voles (M. oeconomus) in northern
Russia (Lvov et al. 1988) and from various rodents in
Norway (Tryland et al. 1998). In the United Kingdom,
antibodies have been found in house mice, but the
highest seroprevalence was encountered in bank voles
(M. glareolus), wood mice (A. sylvaticus), and eld
voles (M. agrestis) (Bennett et al. 1997; Chantrey et al.
1999; Crouch et al. 1995). In a study in bank voles and
wood mice it was found that seroprevalences can vary
considerably over time at the same site and among the
same species (Hazel et al. 2000). Moreover, although
the rodents do not show obvious signs of disease, dur-
ing another study it was demonstrated that cowpox
virus can reduce the fecundity of infected bank voles
and wood mice by increasing the time to rst litter by
20–30 days (Feore et al. 1997). Interestingly, it was found
that bank voles that had high probabilities of infection
survived better than uninfected individuals, some-
thing which was not found in wood mice. is suggests
that each species has its own role in the transmission
dynamics of cowpox virus (Telfer et al. 2002).
Moreover, in this mixed populations of hosts it was
propagated that the best description for the transmis-
sion dynamics is frequency dependent, which means
that each host makes a xed number of contacts with
other hosts, independent of the population size (Begon
et al. 1999). On the contrary, previously it was thought
that the best descriptive model was density dependent:
susceptible hosts were assumed to contact other hosts
throughout the whole of their population at random.
e number of these contacts then rose in proportion to
the size of the population.
Also, hosts can inuence each other: in a study on 14
islands in Northern England (Begon et al. 2003) it was
found that in the case of cowpox dynamics, wood mouse
density thresholds were inuenced at least as much by
the bank vole thresholds as they were by the dynamics
within the wood mouse populations themselves.
Rodents can also transfer the disease directly to
humans, as was reported some years ago when a Norway
rat (R. norvegicus) transmitted the disease to a woman
(Wolfs et al. 2002). Patients show painful, hemorrhagic
pustules or black eschars, usually on the hand or face,
accompanied by edema, erythema, lymphadenopathy,
and systemic involvement (Baxby et al. 1994). Fatalities
mainly occur in the immunosuppressed (Baxby et al.
1994).
Via induced cross-immunity the WHO elemina-
tion campaign for smallpox with worldwide vaccina-
tion also had positive results for the suppression of
Rodent-borne diseases and their risks for public health 235
cowpox (Pelkonen et al. 2003; Vorou et al. 2008). Since
the elimination of smallpox, vaccination practices were
abandoned, resulting in decreased immunity of younger
unvaccinated age groups against other orthopoxviruses
such as monkeypox and cowpox (Pelkonen et al. 2003;
Vorou et al. 2008). As eective treatments do not exist,
some argue that cowpox can therefore be considered as
an emerging zoonotic health threat (Vorou et al. 2008).
Contagious viral animal diseases
Although not important from the perspective of human
health, the consequences of an outbreak of conta-
gious viral animal diseases (classical swine fever, foot
and mouth disease, avian inuenza) for the economy
(particularly the livestock sector) are signicant. e
2001 outbreak of foot and mouth disease in the United
Kingdom is estimated to have caused economic losses
in the order of 3.1 billion pounds (ompson et al. 2002).
e eect on the image of the animal husbandry sector
are more dicult to measure.
During recent classical swine fever virus (CSFV)
outbreaks, rodents were mentioned in direct connec-
tion with the large numbers of secondary infections
that were observed in the vicinity of primary infected
herds (Elbers et al. 1999; Hughes and Gustafson 1960;
Mintiens et al. 2003; Terpstra 1988; Westergaard 1996).
is was due to the fact that in many of these outbreaks
none of the ‘traditional’ transmission routes for CSFV,
for example, direct animal contact, swill feeding, or
transport contact, were responsible for the virus spread
(Terpstra 1988).
Although often linked, there are only a few experi-
ments that describe the role of rodents in transmission of
contagious animal diseases (DeWulf et al. 2001; Hughes
and Gustafson 1960; Terpstra 1987). An experiment by
Terpstra (Terpstra 1987) revealed that rats which were
fed in close contact with CSFV-infected pigs were not
able to transmit the infection to susceptible animals.
Another study provides evidence that rats are unlikely
to represent signicant biological reservoirs of CSFV
(DeWulf et al. 2001). e same authors suggest that the
likelihood of mechanical spread is dicult to assess and
claim that the mechanical spread of CSFV by pets and
rodents remains a possibility. is opinion is shared by
other authors (Elbers et al. 1999), who explain that it is
conceivable under eld conditions that rodents with a
sub-optimal health will have less-ecient grooming of
their fur, which may therefore increase the possibility to
mechanically transmitted CSFV. On the other hand, in a
epidemiological study in which the 1997-1998 outbreak
of CSFV in the Netherlands was studied no associations
between the presence of rats or mice around the premises
and increased risk of infection with CSFV was found
(Elbers et al. 2001). is may be because farmers had to
respond in a questionnaire whether there were rodents
present on their farms. From earlier studies (Meerburg
et al. 2004), we know that this formulation generally
leads to an underestimation of rodent presence.
Rodents may be involved to some degree in the
epidemiology of foot-and-mouth disease because the
brown rat Rattus norvegicus is susceptible to foot-and-
mouth disease and can excrete the virus over long
periods. erefore rodents may play a signicant part
in the dissemination of the disease (Capel-Edwards
1970).
Moreover, rodents are also mentioned in association
with a number of other animal diseases such as por-
cine parvovirus (Joo et al. 1976; Joo et al. 1976) which
can lead to reproductive failure in pigs and Aujeszky’s
disease virus (Maes et al. 1979), a disease that is also
known as pseudorabies or ‘mad itch.’ Rodents also are
associated with the horizontal transmission of clini-
cal encephalomyocarditis fever virus (ECMV) between
farms (Knowles et al. 1998; Maurice et al. 2007; Spyrou
et al. 2004), thus leading to economic losses.
Bacteria
Leptospirosis
Rodents are carriers of spirochetes of the genus
Leptospira throughout the world (Boqvist et al. 2002;
Bunnell et al. 2000; iermann 1977; Webster et al. 1995;
Wisseman et al. 1955) and are important reservoirs
of infection for man and domestic animals. Several
Leptospira strains are directly linked to rodents, such
as L. arborea, L. copenhagi, L. icterohaemorrhagiae,
L. bim, and L. ballum (Bharti et al. 2003; Collares-
Pereira et al. 1997; Collares-Pereira et al. 2000). e
complex taxonomy of dierent Leptospira species is
explained in an excellent review by Barthi and col-
legues (Bharti et al. 2003).
Humans acquire infection through consumption
of food or water that is contaminated by rodents or by
contact through skin or mucous membranes with soil or
water that is contaminated by rodent urine. Handling of
dead infected rodents may also form a source of infec-
tion. It can lead to aseptic meningitis or Weil’s disease,
which is characterized by lymphadenopathy, jaundice,
renal failure, and hemorrhages (Arean 1962; Bharti et al.
2003; Plank and Dean 2000; Steele 1958). e mortal-
ity rate associated with severe leptospirosis may be as
high as 15% (Ko et al. 1999). Although there are some
positive reports on the use of antileptospiral vaccines in
humans, long-term ecacy studies of vaccines have not
been published (Bharti et al. 2003) and many questions
remain on virulence factors and pathogenesis (Koizumi
and Watanabe 2005).
236 B G Meerburg et al.
e number of human cases worldwide is not well-
documented and suers from consequent under-
reporting in many areas of the world (WHO, Geneva,
http://www.who.int). It probably ranges from 0.1 to 1
per 100 000 per year in temperate climates to 10 or more
per 100 000 per year in the humid tropics. During out-
breaks (often associated with disasters) and in high-risk
groups, 100 or more per 100 000 may be infected (WHO,
Geneva, http://www.who.int).
Leptospirosis has a major impact on rural communi-
ties in developing countries in Asia. An epizootic of lept-
ospirosis in humans occurred in NE ailand from 1995
to 2003. In 1996, 398 cases were reported in 4 provinces,
with a peak in 2000 of 14,285 cases and 362 deaths, and
cases reported across 16 provinces in 2001. e number
of human cases remained high until 2003, with 171
deaths in 2001 and 95 in 2002. Most cases (ranging from
72% to 94% of those reported in a year) occurred among
rice farmers (Phulsuksombati et al. 2001; Tangkanakul
et al. 2005), resulting in a severe impact on rural as well
as urban communities in the region (Tangkanakul et al.
2001). Information on leptospirosis in other regions in
Asia is extremely limited. e symptoms are u-like and
are often mistakenly diagnosed and neglected in the
rural areas until serious clinical damage has occurred.
In the Philippines, >1,000 people are hospitalized
annually with leptospirosis. Fatality is high rate rang-
ing from 11% to 20%. is appears to be in part because
people go to hospital only if severe symptoms develop
because they cannot aord to pay for long stays in a
hospital. erefore, it is the urban and rural poor who
are at greatest risk because of higher rates of exposure to
infectious bacteria and little available income for early
medical intervention.
Lyme disease
Rodents play an important role in the spreading of the
Lyme disease spirochetes Borrelia burgdorferi (Shih and
Chao 1998), B. garinii, and B. afzelli (Parola and Raoult
2001). e enzootic rodent-tick cycle maintains Borreliae
specically (Nakao et al. 1994), although the relative
potential as reservoir diers between rodent species
(Brown and Lane 1996; Burkot et al. 1999; Humair et al.
1999; Mather et al. 1989; Sinski et al. 2006). During a study
in Switzerland, Borrelia infection was more prevalent in
Myodes than in Apodemus (Humair et al. 1999). Also in
an earlier study it was demonstrated that M. glareolus
plays a dierent role as reservoir host species compared
with two Apodemus species (Kurtenbach et al. 1995).
ese authors also found that prevalence of Borrelia
burgdorferi in host-seeking ticks (Ixodes ricinus L.) was
1% for larvae, 5% for nymphs, and 10–20% for adults.
In the USA (California), it was shown that dusky-footed
woodrats and California kangaroo rats showed infection
prevalences of 85.7% and 78.6%, respectively. In contrast,
only 22.2% of brush mice (Peromyscus boylii) and 7.1%
of pinyon mice (P. truei) were infected (Brown and Lane
1996). During another study in the USA (New York), the
determinants of Lyme-disease risk (density and Borrelia
burgdorferi-infection prevalence of nymphal Ixodes
scapularis ticks) were assessed (Ostfeld et al. 2006).
It was shown that the strongest predictors of a current
year’s risk were the prior year’s abundance of mice and
chipmunks and abundance of acorns 2 years previously
(Ostfeld et al. 2006).
From a study in Germany we know that immunity
to B. burgdorferi in natural reservoir hosts is an impor-
tant regulatory factor in the horizontal transmission of
B. burgdorferi in nature (Kurtenbach et al. 1994). Voles
carry a larger number of ticks than mice, while on the
other hand mice are more frequently infected with the
pathogen. us, voles are high responders to the vector
(higher infectivity), while mice are high responders to
the microparasite (higher % infected) (Kurtenbach et al.
1994).
is demonstrates that mice and voles play dierent
quantitative roles in the ecology of Lyme borreliosis in
Europe. To make things even more complicated, the
presence of predatory vertebrates may indirectly protect
human health by reducing the population size of rodent
reservoirs (Ostfeld and Holt 2004). On the other hand,
some researchers describe zooprophylaxis (Matuschka
et al. 1991). ese researchers think that because rodents
are present, the rate at which ticks bite other hosts
(including humans) may be reduced, and thus the likeli-
hood of infection is also reduced (Matuschka et al. 1991;
Schmidt and Ostfeld 2001). Some say that there is proof
that increased biodiversity (i.e., an increasing number of
potential tick host species) may lead to a dilution eect
(Dobson et al. 2006; Schmidt and Ostfeld 2001).
In the USA, Ixodes dammini and pacicus ticks are
the most important vectors of the pathogen, while in
Europe Ixodes ricinus is the responsible vector. In Asia,
particularly China, B. burgdorferi was encountered in
the pygmy wood mouse (Apodemus uralensis) and the
long-tailed dwarf hamster (Cricetulus longicaudatus)
and in I. persulcatus ticks (Takada et al. 2001).
Humans are infected by the bite of a tick that acquired
the infection from an enzootic reservoir. Infection can
lead to Lyme’s disease, which is characterized by cutane-
ous manifestations (for example, Erythema chronicum
migrans, a rash that is an early sign of infection which
helps early diagnosis). If the disease remain unnoticed,
it can aect the nervous system, heart, eye, and joints
in variable combinations. Fatalities due to Lyme’s dis-
ease seldom occur. However, there are dierences in the
clinical presentations of Lyme borreliosis between con-
tinents, e.g., between the USA and Europe. Patients in
Europe may develop certain skin manifestations, such
Rodent-borne diseases and their risks for public health 237
as borrelial lymphocytoma and acrodermatitis chronica
atrophicans) that are either rare of non-existent in the
USA (Strle et al. 1999). is might be caused by strain
variation: in United States patients, members of the
genomic group Borrelia burgdorferi sensu stricto were
identied, whereas European isolates have included two
additional genospecies, B. garinii and B. afzelii (Picken
et al. 1998; Strle et al. 1999).
Although a vaccine based on the recombinant outer-
surface protein A (OspA) was available (Wormser et al.
2000), this was withdrawn from the market due to its
side eects.
In recent decades the tick that is responsible for
spreading of the pathogen in Europe has spread into
higher latitudes and has become more abundant, pos-
sibly associated with climate change (Lindgren et al.
2000; Randolph 2005). According to the WHO (WHO,
Geneva, http://www.who.int), this will also contribute
to extended and more intense transmission seasons for
all tick-borne diseases, such as Lyme’s disease and tick-
borne encephalitis (TBE).
Tick-borne relapsing fever
Tick-borne relapsing fever (TBRF), an underrecognized
and underreported disease, is caused by numerous
species of the spirochete Borrelia and occurs through-
out the world in many discrete enzootic foci where the
spirochetes are maintained primarily in rodents and
soft-body ticks of the genus Ornithodoros (Schwan and
Hinnebusch 1998), with the exception of Australia, New
Zealand, and Oceania (Johnson 1977).
In North America, several relapsing fever spiro-
chetes, Borrelia hermsii, Borrelia turicatae, and possibly,
Borrelia parker (Fritz et al. 2004), alternate infections
between rodents and tick vectors (Ornithodoros spp.)
(Boyer et al. 1977; Dworkin et al. 2002; Dworkin et al.
2002). Many residents and visitors are exposed in the
endemic regions of the western United States to the vec-
tors of TBRF. ere are approximately 25 cases of TBRF in
the United States each year (CDC, Atlanta, http://www.
cdc.gov). e pathogen can be found primarily in chip-
munks (Tamias spp.) and pine squirrels (Tamiascuirus
spp.) above elevations of 1,000 m (Fritz et al. 2004).
In Israel, cases of TBRF are thought to be the result of
the pathogen B. persica that is transmitted by the tick
Ornithodoros tholozani (Sidi et al. 2005). is pathogen
is also present in Syria, Egypt, Iran, and Central Asia
(Rebaudet and Parola 2006). Recurrent fever due to B.
hispanica is reported sporadically in Spain, Portugal,
Cyprus, Greece and North Africa. A new species patho-
genic to humans was isolated and characterized in Spain
in 1996 in Ornithodoros erraticus (Anda et al. 1996).
West African tick-borne relapsing fever (TBRF) is
caused by the microorganism, Borrelia crocidurae. If
untreated, mortality rates are up to 5% (Southern Jr. and
Sanford 1969). A survey in Senegal (Godeluck et al. 1994)
demonstrated that the African grassrat (Arvicanthis
niloticus) and Mastomys huberti (Hubert’s Mastomys)
function as hosts for the pathogen. Besides Senegal, this
pathogen also occurs in Morocco, Libya, Egypt, Iran,
and Turkey (Rebaudet and Parola 2006).
Borrelia caucasica, present in Iraq and the Caucasus,
is transmitted by Ornithodoros asperus, another
tick of rodents. Borrelia latyschevii is transmitted by
Ornithodoros tartakovskyi in Central Asia, Russia, and
Iran (Estrada-Peña and Jongejan 1999).
Humans acquire infection when infected ticks that
live in the burrows or nests of rodents within a human
dwelling search for an alternative host to feed upon.
Such infection can result in recrudescent illness with the
characteristic clinical syndrome of periodic fevers. e
rst attack is generally the most severe in terms of height
and duration of fever, while succeeding attacks are usu-
ally milder. Severity is highest in (pregnant) women and
children (Barbour 1999; Jongen et al. 1997). Incubation
time after the tick bite is between 4 and 14 days (Johnson
and Golightly 2000).
Scrub typhus
Scrub typhus (also named tsutsugamushi disease) can
be the result of infection of humans with the bacterium
Orientia tsutsugamushi. is pathogen is transmit-
ted by bites from infected trombiculid mites (‘chig-
gers’) that are hosted by various rodent species, such
as the Norway rat (R. norvegicus), the lesser bandicoot
(Bandicoota bengalensis), the bandicoot rat (Bandicoota
indica), house mouse (Mus musculus), the striped-eld
mouse (Apodemus agrarius), and the Japanese grass
vole Microtus montebelli (Milne-Edwards) (Kumar et al.
2004; Liu et al. 2003; Takahashi et al. 2004). Cases of
scrub typhus have been reported from Southern and
Eastern-Asia (Japan, Indonesia, Korea, ailand, Asiatic
Russia, the Indian subcontinent, Pakistan, China), and
Australia (Graves et al. 2006; Khuntirat et al. 2003; Kim
et al. 2007; Liu et al. 2003; Sin et al. 2000; Takahashi et al.
2004; Traub et al. 1954; Yahnke et al. 2001). In Australia,
hosts of chiggers include introduced rodent species,
native rodent species and small marsupials, particularly
the long-nosed bandicoot, Perameles nasuta, and scrub
typhus has been reported down the east coast of Australia
(Spratt 2005). Similar to murine typhus, people acquire
the infection through the bite of an infected mite. e
estimated incidence of the disease in endemic areas is
one million cases each year (op cit. Fournier et al. 2008).
Incubation time varies between 6 and 12 days (Varghese
et al. 2006). Symptoms include fever, headache, muscle
pain, cough, and gastrointestinal symptoms. More viru-
lent strains can cause hemorrhages and intravascular
238 B G Meerburg et al.
coagulation. If untreated, mortality rates can mount up
to 40% (Varghese et al. 2006), but with use of antibiotics
it is below 5% (Dupon et al. 1992). ough infection can
be easily treated with doxycycline/azithromycin, clini-
cal diagnosis is often dicult because the characteristic
eschar at the site of the bite is present only in the minor-
ity of the patients.
In northern ailand in 2000, more than 3,900 cases of
scrub typhus were reported, with most cases being male
rice farmers (in northeast ailand in 2000, the morbid-
ity rate was 8.7 per 100,000 people) (W. Tangkanakul,
personal communication). A 10-year study in northern
ailand identied 9 murid rodents as carriers of scrub
typhus, with the main carriers being R. rattus (23%, 419
of 1,855), R. argentiventer (2%, 5 of 23), B. berdmorei
(22%, 2 of 9), R. losea (13%, 82 of 638), and B. indica (9%,
52 of 564) (Coleman et al. 2003).
In Northeast ailand in 1997, nine of 22 human cases
of leptospirosis were diagnosed as positive to scrub
typhus. is is the rst report of a high incidence of co-
infection with leptospirosis and typhus. Twenty of the 22
people screened were rice farmers, a group known to be
at high risk in ailand for both diseases (Watt et al. 2003).
e authors highlighted the importance of this nding
with the reported death of a patient who was diagnosed
with leptospirosis and was treated with the appropriate
antibiotic. However, the patient was not diagnosed as a
possible typhus victim (combined infection) and scrub
typhus was not sensitive to the antibiotic administered
(intravenous penicillin). In Asia, doctors who diagnose
leptospirosis in agricultural workers are recommended
to also consider co-infections with typhus in patients
that do not respond to treatment for leptospirosis (Watt
et al. 2003).
Murine typhus
Rodents, more specically rats, are associated with the
worldwide distribution of the bacterium Rickettsia typhi
(Chaniotis et al. 1994; Graves et al. 2006; Gray et al. 2007;
Kim et al. 2007; Letaief 2006; Reeves et al. 2006), which
can cause murine typhus. e vector is the rat ea,
Xenopsylla cheopis. e triad of rickettsia-ea-rat seems
to be a true commensalism, because the rickettsiae
harm neither rat nor ea (Azad 1990). Humans acquire
murine typhus from an infected ea: most eas defecate
while biting and their feces can contain the bacteria that
cause the disease. Although uncommon, it is also pos-
sible for humans to contract murine typhus by inhaling
contaminated dried ea feces. e incubation period
for the disease lies on average between 6 to 14 days.
Symptoms include headache, fever, nausea, and body
aches. Five or six days after the initial symptoms, 54% of
the patients get a rash that starts on the trunk of their
bodies and then spreads to arms and legs (Elston 2005).
e full triad of fever, headache, and rash occurs in only
12.5% of the patients (Elston 2005). Respiratory and
gastrointestinal symptoms are frequent and may result
in confusion with a viral illness. A worse prognosis is
noted in those with renal dysfunction, leukocytosis, and
hypoalbuminemia (Elston 2005). Adverse outcomes are
also associated with advancing age, and therapy with
sulfa antibiotics (Dumler et al. 1991). Mortality rate for
murine typhus is between 1–4%. No eective vaccine is
available (Baxter 1996).
Murine typhus is likely to persist in endemic areas
where rat populations remain high (Azad 1990), espe-
cially in coastal areas in the vicinity of ports. is was also
demonstrated by a survey in Africa, which demonstrated
that prevalence of antibodies against R. typhi in humans
was higher in coastal areas (Tissot-Dupont et al. 1995).
In some countries (Pakistan, ailand, India, Myanmar,
and Southern USA) the infection has a wider distribu-
tion than only the coastal areas (Azad 1990).
Because the disease is frequently underreported, it is
dicult to provide reliable incidence rates. In the United
States about 42,000 cases were reported between 1931
and 1946, but incidence rapidly declined due to rat con-
trol programs. Since 1961 the number of reported cases
per year has remained well under 100 (Azad 1990; Traub
et al. 1978).
Recently, the pathogen which causes murine typhus
has been shown to have an alternative peridomestic
animal cycle that does not involve rats and X. cheopis.
is new cycle was reported in the USA involving cats,
dogs and opossums, and the cat ea Ctenocephalides
felis (Azad et al. 1997).
Sylvatic epidemic typhus
Sylvatic epidemic typhus, caused by Rickettsia prow-
azekii, is transmitted by ectoparasites (lice and eas)
of rodents to humans. In large parts of the world (South
and Central America, Africa, Asia, Mexico), R. prowazekii
are also transmitted between humans (Yu and Walker
2006). is is done by the body louse Pediculus humanus
humanus, that lives in clothes and multiplies when such
conditions as cold weather, lack of hygiene, or war are
present (Raoult and Roux 1999). Its prevalence can be
considered as an indicator for the socioeconomic level
of the society (Raoult and Roux 1999).
In the wild, the pathogen is maintained in a south-
ern ying squirrel (Glaucomys volans) sylvatic cycle in
the eastern United States (Foley et al. 2007; Reynolds
et al. 2003). Only a few cases have been reported since
1985 (Reynolds et al. 2003). In one third of these cases,
contact with ying squirrels or with ying squirrel nests
occurred before disease onset (Reynolds et al. 2003). In
general, all louse-borne diseases can be transmitted by
contact of broken skin or wounds with infected louse,
Rodent-borne diseases and their risks for public health 239
eas, mites or their droppings, or by inhalation of their
aerosolized feces. However, the transmission mechanism
of R. prowazekii from ying squirrels to humans is less
well understood. Although various transmission modes
are mentioned, none has been empirically conrmed.
However, one species of eas (Orchopeas howardii) that
is hosted by ying squirrels is known to opportunisti-
cally bite humans, thereby playing an important role in
pathogen transmission (Reynolds et al. 2003).
Queensland tick typhus
Queensland tick typhus or spotted fever has been
reported along the eastern coast of Australia. It is caused
by Rickettsia australis and was rst described in soldiers
posted to northern Queensland during World War II.
e vectors have been described as Ixodes holocyclus,
Ixodes tasmani and, probably, Ixodes cornuatus. e
reservoir hosts are rodents and marsupial bandicoots
(Spratt 2005).
Rocky Mountain spotted fever
Rocky Mountain spotted fever (RMSF) is the most severe
and most frequently reported rickettsial illness in the
United States. Spotted fever also occurs in Mexico and
in Central and South America (Fuentes 1986; Hoogstraal
1967), but is sometimes known under dierent names
there (e.g., Brazilian Spotted Fever). e causative agent
is Rickettsia rickettsii, a bacterium of the Rickettsiaceae
family, that spreads to humans by ixodid ticks. Rodents
can serve as an reservoir of the infectious agent (Atwood
et al. 1965), e.g., white-footed mice (Peromyscus leu-
copus) (Magnarelli et al. 1981), the hispid cotton rat
(Sigmodon hispidus), the cotton mouse (Peromyscus
gossipinus, Le Conte), the golden mouse (Ochrotomys
nutalli, Harlan), the woodland vole (M. pinetorum), the
prairie vole (Microtus ochrogaster, Wagner), and the
meadow vole (M. pennsylvaticus, Ord) are mentioned as
possible hosts (Kollars Jr. 1996).
e American dog tick (Dermacentor variabilis) and
Rocky Mountain wood tick (Dermacentor andersoni) are
the primary vectors of the pathogen in North-America,
while the brown dog tick (Rhipicephalus sanguineus) can
also be a vector (Demma et al. 2005; Hoogstraal 1967). In
Central and South America, the tick Amblyomma cajen-
nense can be considered a main vector (Figueiredo et al.
1999).
Initial signs and symptoms of spotted fever include
sudden onset of fever, headache, and muscle pain, fol-
lowed by development of rash. e disease is sometimes
dicult to diagnose in the early stages, and without
prompt and appropriate treatment with doxycycline
fatalities do occur. e incubation period is between 3
and 12 days (Walker 1995).
Rickettsialpox
A disease that is associated with urban environments
is rickettsialpox (Comer et al. 2001; Comer et al. 2001).
is disease is caused by the worldwide distributed
(Comer et al. 2001) pathogen Rickettsia akari, which is
transmitted to humans by mites of rodents. e patho-
gen is maintained by vertical transmission in the house
mice mite (Liponyssoides sanguineus) and by horizontal
transmission between the mite and its main host: the
house mouse (M. musculus domesticus) (Radulovic et al.
1996). e possibility exists, however, that other species
of mites are also involved in the cycle (Bennett et al.
2007). Moreover, the pathogen was also isolated from
commensal rats in Ukraine, from black rats (R. rattus),
dusky-footed woodrats (Neotoma fuscipes) and deer
mice (P. maniculatus) in the USA (Bennett et al. 2007)
and from Korean reed voles (Microtus fortis) in Korea
(Jackson et al. 1957). is suggests that R. akari can adapt
to other rodent hosts. Human infections were reported
in the USA (Bennett et al. 2007; Kass et al. 1994; Paddock
et al. 2003; Paddock et al. 2006), in South Africa (Gear
1954), Turkey (Ozturk et al. 2003), in Croatia (Radulovic
et al. 1996), in Bosnia and Herzegovina (Terzin et al.
1956), and in Ukraine (Eremeeva et al. 1995). Infection
of humans occurs only if mice or other preferred hosts
are not available (Heymann 1996).
e clinical portrait of rickettsialpox in humans con-
sists of a cutaneous lesion at the site of inoculation by
the mite: rst a papule appears and later this evolves
into an eschar (Yu and Walker 2006). After about 1 week,
patients develop fever, chills, malaise, and headache,
followed shortly by a secondary papulovesicular cuta-
neous eruption (Heymann 1996). Although the symp-
toms look severe, patients will usually recover within 1
to 2 weeks. No fatalities have been reported (Heymann
1996). However, rickettsialpox is infrequently reported
and underdiagnosed at present.
Bartonella illnesses
Of the 19 described Bartonella species that infect a wide
variety of domestic and wild animals such as cats, dogs,
mice, rats, squirrels, deer, and moose, 7 have been associ-
ated with human disease (Alsmark et al. 2004). Fleas and
lice are ecient vectors of rodent bartonellae (Boulouis
et al. 2005; Bown et al. 2004) and are involved in the
transmission to humans. For this reason, disease often
occurs in people with a low-living standard such as the
homeless or intravenous drug users (Comer et al. 2001;
McGill et al. 2003; McGill et al. 2003; Smith et al. 2002).
Rodents can form a reservoir for several Bartonella
species such as B. elizabethae, B. washoensis, B. gra-
hamii, B. taylorii, and B. vinsonii. Evidence of host
specicity suggests the possibility of a long-term
co-evolution or co-speciation of Bartonella with their
240 B G Meerburg et al.
rodent hosts (Kosoy et al. 2000). Prevalences in rodent
populations can be quite high (11.1–62.2%) with vari-
ance between dierent rodent species as was demon-
strated in studies from the United Kingdom, the south-
eastern USA and Poland (Birtles et al. 1994; Easterbrook
et al. 2007; Kosoy et al. 1997; Welc-Faleciak et al. 2008).
In a review from some years ago (Boulouis et al. 2005) it
is described that Norway rats (R. norvegicus) constitute
the main reservoir of B. elizabethae, while Bartonella
grahamii has been mainly isolated from bank voles
(M. glareolus) in the UK and Poland and yellow-necked
mice in Sweden. However, this Bartonella species was
also isolated in rats and a domestic mouse in the USA.
According to these authors, California ground squir-
rels (Spermophilus beecheyi) are the main reservoir
of B. washoensis (Boulouis et al. 2005). Isolation of B.
henselae (the causative agent of cat’s scratch disease in
humans) was recently reported in three long-tailed eld
mice (Apodemus sylvaticus) in Denmark (Engbaek and
Lawson 2004). In Spain, two dierent Bartonella geno-
types were detected in M. spretus, and one genotype that
was corresponding with B. tribocorum in the Norway
rat (R. norvegicus) (Márquez et al. 2008). However, it
remains questionable to what extent rodents contribute
to Bartonella transmission to humans (Anderson and
Neuman 1997). e probable mode of transmission is
via feces that is deposed by rodent ectoparasites on the
broken human skin.
Some years ago, B. grahamii was isolated for the
rst time from the eye of a patient with neuroretinitis
(Kerkho et al. 1999; Meerburg and Kijlstra 2007) in
Europe and possibly linked to rodent presence, although
no clear evidence could be provided. But if transmission
occurs, consequences may be serious, as in humans
Bartonella spp. can lead to endocarditis, cardiac dis-
ease and febrile illness respectively (Breitschwerdt and
Kordick 2000; Iralu et al. 2006; Welc-Faleciak et al. 2008).
In the immunodepressed (e.g., AIDS-patients), B. quin-
tana has been related with chronic lymphadenopathy
and numerous atypical manifestations (English 2006).
Human granulocytic anaplasmosis
Humans can get infected with Anaplasma phagocy-
tophilum, the causative agent of human granulocytic
anaplasmosis (HGA), previously known as human gran-
ulocytic ehrlichiosis, when they infringe on tick-small
mammal habitats (Dumler and Bakken 1998). HGA was
rst recognized in 1994 (Bakken et al. 1994). HGA is
characterized by headache, high fever, chills, and myal-
gias. In the early stage, the disease can be readily treated
with antibiotics. However, if untreated, severe illness can
occur, including secondary infections due to renal dys-
functions, respiratory failures, and occasionally this may
even result in death (Fritz and Glaser 1998). Infected ticks
(Ixodes spp.) are hosted by rodents (primarily wood rats
and white-footed mice, but also other rodents such as
squirrels are capable of carrying these ticks (Levin et al.
2002). If these infected ticks then bite a human, they fre-
quently cause HGA-fever in several parts of the world,
such as the USA (Upper Midwest, New England, parts of
the mid-Atlantic states, northern California), many parts
of Europe and sometimes even Asia.
HGA is clinically variable. Recent sero-epidemiological
surveys suggested that many infections go unrecognized,
since in endemic areas as much as 15–36% of the popu-
lation has been infected (Bakken et al. 1998; Mattson
et al. 2002). Most patients have a moderately severe
febrile illness with headache, myalgia, and malaise. e
estimated fatality rate is 0.7% as determined from sur-
veillance data, and elderly patients are more prone to
severe infections and death (Demma et al. 2005). From
the period 1986–2001 in total 1540 cases of HGA were
reported in the USA. e average reported annual inci-
dences for HGA in the USA during 2001–2002 was 1.4
cases per million people. However, because of the non-
specic nature of clinical signs, human granulocytic
anaplasmosis is most likely underrecognized and under
reported (Demma et al. 2005).
Q-fever
Q-fever is caused by the intracellular gram-negative
bacterium Coxiella burnetii. Morphologically, the genus
Coxiella is similar to the genus Rickettsia, but with a vari-
ety of genetic and physiological dierences.
Coxiella bacteria are present throughout the world,
except for Antarctica and probably New Zealand
(Fournier et al. 1998). Cattle, sheep and goats are the most
common animal reservoir of this zoonotic disease.
Rodents also are suspected as a reservoir, but their
role in transmission to humans might be limited com-
pared to other pathways. Nevertheless, rodents can
be infected. In the United Kingdom, wild Norway rats
(R. norvegicus) displayed seroprevalences between
7-53% (Webster et al. 1995), while in the USA low
seroprevalences were detected in muskrats (Ondatra
zebethica), rats (Rattus spp.), Beechey ground squir-
rels, (Otospermophilus beecheyi), wood rats (Neotoma
fuscipes), and deer mice (Peromyscus spp) (Riemann
et al. 1979). In Japan, another rodent species, Myocastor
coypus, has shown moderate (13%) seroprevalence for
C. burnetii, (Ejercito et al. 1993).
People acquire infection through inhalation of con-
taminated dust, contact with contaminated animal
products (milk, meat, wool) and particularly birthing
products (amniotic uid etc.). Moreover, ticks can trans-
fer the pathogenic agent to other animals, but contrary
to other tick-borne pathogens, not to humans (op cit.
Fournier et al. 1998).
Rodent-borne diseases and their risks for public health 241
If untreated, the disease is usually deadly, but with
appropriate treatment the death rate is around 10%.
During its course, the disease can progress to an atypi-
cal pneumonia, which can result in a life threatening
acute respiratory distress syndrome (ARDS), which is
frequently found in North America. Symptoms usually
occur during the rst 4–5 days post infection. However, at
other locations in the world (e.g. Europe) Q-fever causes
(granulomatous) hepatitis which becomes symptomatic
with malaise, fever, liver enlargement (hepatomegaly),
pain in the right upper quadrant of the abdomen and
jaundice (icterus). If a person is chronically infected,
the disease can lead to endocarditis. Q-fever also causes
abortions and reproductive problems in livestock, and
thus economic losses for farmers. In livestock, a natural
cycle of transmission without involvement of ticks has
been postulated (Krauss 1989).
Salmonellosis and campylobacteriosis
Salmonella and Campylobacter are generally regarded as
the most important food-borne pathogens in the world.
Reduction or elimination of these pathogens in the rst
part of the food chain (on the farm) is important to prevent
disease among consumers of animal products. Previous
research has proven that wild rodents and house mice
are able to amplify these pathogens in the environment
and are probably capable of transmitting them to food
animals (Davies and Wray 1995; Evans and Sayers 2000;
Garber et al. 2003; Henzler and Opitz 1992; Meerburg et al.
2006; Meerburg 2007). If products of these animals (e.g.,
their meat) are then improperly cooked, they could lead
to human infection. It is known also that rodents can be
long-term sources of infection: e.g., a study demonstrated
rodents are still capable to transfer Salmonella enteriditis
to chicks after two and ve months post infection (Davies
and Wray 1996). Unfortunately, it is not yet known to what
extent human cases of salmonellosis or campylobacterio-
sis can be related to rodents, but other sources are pre-
sumably more important. e consequences of infection
with these pathogens can be severe: diarrhea, headache,
vomiting, and sometimes even death.
Tularemia
Tularemia is caused by the intracellular gram-negative
bacterium Francisella tularensis, a member of the
Francisellaceae family. Humans can acquire infection
through direct contact with infected animal carcasses,
consumption of food or water that is contaminated by
rodents, after a bite of an infected mammal (Friedl et al.
2005), tick, deery, or another insect or by breathing aer-
osols containing the bacteria (Christova and Gladnishka
2005; Hörnfeldt 1978; Pape et al. 2005; Petersen and
Schriefer 2005; Wobeser et al. 2007). Tularemia mainly
occurs in the northern hemisphere and most frequently
in Scandinavia, Central Europe, Northern America,
Japan, and Russia, although F. tularensis subsp. novicida
has been reported in Australia (Hollis et al. 1989).
Until now, four subspecies of the pathogen have been
discovered, which each exhibit distinct biochemical
and viral proles (Farlow et al. 2005). Human disease is
primarily associated with two of these subspecies: the
highly virulent F. tularensis subsp. tularensis (type A),
which can only be encountered in North America and
the moderately virulent F. tularensis subsp. holarctica
(type B), which is endemic throughout the Northern
Hemisphere (Farlow et al. 2005). While type A is reported
to have a terrestrial cycle with the main reservoirs being
cottontail rabbits (Sylvilagus spp.) and ticks (Mörner
1992). Recently, molecular subtyping has further divided
type A into 2 subpopulations, A1 and A2 (Kugeler et al.
2009). Type B is reported to have a mainly water-borne
cycle with aquatic rodents as reservoirs, e.g., muskrats
(Ondatra zibethicus) and beaver (Castor canadensis) in
North America, and ground voles (Arvicola terrestris)
in the former Soviet Union (Mörner 1992). In Europe,
tularaemia is most frequently seen in hares (Lepus spp.)
(Mörner 1992).
Humans contract tularaemia mostly through mos-
quito bites or by handling infected animals (Ikaheimo
et al. 2000). e disease occurs in several forms in
humans, depending to a large extent on the bacterial
entry route into the body. Most common is ulceroglan-
dular tularemia, which usually occurs after a bite from
an arthropod vector which has previously fed on an
infected animal (Ellis et al. 2002). Sometimes cases of
ulceroglandular tularemia occur in hunters and trappers
as a consequence of the handling of infected meat, with
infection via cuts or abrasions. Although the ulceroglan-
dular form of tularemia even without treatment is rarely
fatal (mortality rate less than 3%), patients may take a
signicant time (2–3 months) to heal.
Mortality ranges depend strongly on the subspe-
cies of the infection (Kugeler et al. 2009). Very recently,
pulsed-eld gel electrophoresis typing identied 4 dis-
tinct type A genotypes, A1a, A1b, A2a, and A2b, as well
as type B (Kugeler et al. 2009). Human infections due to
A1b resulted in signicantly higher mortality (24%) than
those caused by A1a (4%), A2 (0%), and type B (7%).
e worldwide incidence of this disease is unknown,
(WHO, Geneva, http://www.who.int), but several
thousands of cases each year have been estimated.
Nevertheless, some data on the incidence of disease are
available. In Japan 1355 cases were reported between 1924
and 1987 (Ohara et al. 1991), while in Sweden, the annual
number of reported case from 1973 to 1985 ranged from
less than 5 cases to over 500 (Mörner 1992). In 2003, 823
human cases of tularemia were reported in Finland, 698
in Sweden, and 22 in Norway (Bystrom et al. 2005).
242 B G Meerburg et al.
In Turkey, 205 cases were reported over the period
1988 to 1998 (Helvaci et al. 2000), while 126 cases of
disease were reported in Slovakia during the period
1985 to 1994 (Gurycová 1997). In Bulgaria, a tularemia
outbreak aected 285 people from 1997 to 2005
(Kantardjiev et al. 2006). In the USA, 316 isolates from
human cases were collected in the period 1964-2004
(Staples et al. 2006).
E. coli 0157/VTEC
e Shiga toxin-producing Escherichia coli (STEC/
VTEC) of the O157 serotype can cause a serious human
food-borne disease, which can lead to hemorrhagic or
watery diarrhea. Particularly in children, this can be
accompanied by the life-threatening hemolytic uremic
syndrome. e mortality rate lies between 3–17% and up
to 30% during outbreaks (Todd and Dundas 2001).
Most human cases are caused by the consump-
tion of raw cow milk or undercooked meat, while also
food products or drinks that are contaminated with
cow manure can be contaminated by the pathogen.
Sporadic cases have been shown to originate from
direct contact with cattle or the contaminated envi-
ronment (Lahti et al. 2002; Moller Nielsen et al. 2005).
In an on-farm trial, the persistence of E. coli 0157/H7
in cattle and the farm environment was investigated
on eight Ontario dairy farms positive for E. coli O157
during the previous year (Rahm et al. 1997). VTEC was
found in composite samples from calf feeders, calf barn
surfaces, cow feeders, ies, cow barn surfaces, and
individual milk lters (Rahm et al. 1997). ese authors
did not encounter the pathogen in other environmen-
tal samples, including rodent feces. Nevertheless, it
is thought that rodents can form a pathogen reser-
voir (Cizek et al. 2000; Van Donkersgoed et al. 2001).
In an on-farm study from the Czech Republic, fresh
droppings of wood mice (A. sylvaticus), house mice
(Mus musculus), and Norway rats (R. norvegicus) were
examined for VTEC. In 40% of the rats, E. coli was
demonstrated (Cizek et al. 1999). In the USA however,
samples from 300 rodents (species not specied) from
12 cattle farms were tested and no E. coli was found
among them, while herd prevalences varied from 1.1
to 6.1% (Hancock et al. 1998).
In Denmark, wild animals living close to cattle and
pig farms (four each) were examined for VTEC (Nielsen
et al. 2004). Among the 260 samples from wild animals
(including birds, rodents and insects) the prevalence
was generally low. However, rodents carried VTEC and
VTEC isolates from a Norway rat (R. norvegicus) were
identical to cattle isolates(Nielsen et al. 2004). is sug-
gest that rodents can carry VTEC, but it remains unclear
which role they play in the transmission cycle of VTEC.
Plague
e most famous disease associated with rodent presence
is probably the infection of rodent eas with bubonic
plague (caused by Yersinia pestis, a member of the family
Enterobacteriaceae), resulting in many millions of casu-
alties during its rst (6th and 7th century AD), second
(14th to 17th century AD) and third (late 19th and early 20th
century AD) pandemics (Perry and Fetherston 1997).
Natural transmission of plague to humans remains a
possibility in many regions of the world, where foci exist
in sylvatic rodent populations. Even today, there are an
estimated 1000–3000 cases of the bubonic plague each
year worldwide, mainly in Africa, the Americas, and Asia
(Keeling and Gilligan 2000). Between 1967 and 1993 the
fatality rate was around 10% (Perry and Fetherston 1997).
A recent review highlights that although in a historical
sense the number of plague cases is relatively low, the
disease is still widely distributed globally, has an innate
ability to spread rapidly and clinical symptoms can unfold
quickly (Stenseth et al. 2008). Moreover, the eects of the
previous pandemics are strongly etched in our memo-
ries and therefore the disease still generates a high fear
factor. In 1994, a localized outbreak of plague in Surat,
India, led to 50 deaths (Ganapati 1995) and national and
international fear resulting in a collapse of tourism and
trade in at a national level at an estimated cost of US$600
million (Fritz et al. 1996). erefore the authors of this
review conclude that “it would be a mistake to overlook
its threat to humanity” (Stenseth et al. 2008).
e oriental rat ea (Xenopsylla cheopis) is the clas-
sic vector for plague. However, also other eas are
sometimes mentioned in connection with plague
(Adjemian et al. 2007). Most rodents show moderate
resistance to infection (as a result of previous exposure
or possibly an inherent characteristic of the species or
subspecies), relatively mild signs, and low mortality rate
(Perry and Fetherston 1997). Some species of Microtus
and Peromyscus have been suggested as maintenance
hosts in western North America, while some types of
mice (in Africa and Russia), gerbils (in Russia, India,
Iran, South Africa, Syria, and Turkey), and voles (in
Russia and Mongolia) are relatively resistant to plague
and are suspected enzootic hosts (Perry and Fetherston
1997). One type of highly resistant rat (Dipodomys spp.)
seroconverts, with few animals becoming ill and rarely
dying, despite evidence of bacterial presence (Perry and
Fetherston 1997). Field data from Kazakhstan showed
that the incidence of clinical cases correlated with uc-
tuations in the abundance of its main reservoir host, the
great gerbil (Rhombomys opimus) (Davis et al. 2004).
Clinical symptoms of bubonic plague are the fol-
lowing (Homan 1980): an abrupt onset, where a chill
is followed by a temperature rise, often to 40ºC, and
accompanied by headache, backache, restlessness, and
Rodent-borne diseases and their risks for public health 243
rapid pulse. is is followed by extreme postration and
central nervous systern manifestations such as anxiety,
delirium, and coma or convulsions. By this time, 75–90%
of patients have an enlarged, tender, nonuctuant, hard
lymph node, the bubo (Homan 1980). Patients should
be isolated and treated with gentamicin or doxycycline
(Mwengee et al. 2006), as streptomycin, tetracycline,
and chloramphenicol have mostly become outdated or
unavailable.
Rat-bite fever and Haverhill fever
Both rat-bite fever and Haverhill fever are caused by
Streptobacillus moniliformis. In Asia, also Spirillum
minus (or: sodoku) can cause rat-bite fever. Haverhill
fever was originally recognized in 1926 and is an infec-
tion transmitted to humans through the consumption
of contaminated water, milk, or food that has previously
been in contact with rats. is disease is characterized
by a high incidence of pharyngitis and pronounced
vomiting. More common is rat-bite fever, which is char-
acterized by abrupt high fevers, headaches, migratory
ancylosis, vomiting and skin rash (2–4 days post infec-
tion) (Elliott 2007). Despite its worldwide distribution,
rat-bite fever is rarely diagnosed by physicians. Estimates
of the mortality rate of untreated cases approach 10–13%
(Elliott 2007; Graves and Janda 2001) and fatality is often
associated with cases in infants and patients with endo-
carditis. In previous studies (Graves and Janda 2001) it
was demonstrated that not only rat bites, but also owners
of pet rats are vulnerable of contracting the disease. e
pathogen can be transferred via scratches or kissing of
the animal. Most reports of S. moniliformis originate from
the United States, although reports have also come from
Brazil, Canada, Mexico, Paraguay, the United Kingdom
and France, but sporadic reports from Norway, Finland,
Germany, Spain, Italy, Greece, Poland, Denmark, and
e Netherlands also exist. Australia has also demon-
strated some cases where it has been reported in eld
populations of house mice (Mus muculus domesticus)
(Taylor et al. 1994). Only few cases have been reported
in Africa (Elliott 2007).
Listeriosis
Listeria are gram-positive bacteria and are named after
their discoverer, the English surgeon Joseph Lister. e
most renown species is L. monocytogenes, the causa-
tive agent of listeriosis. Indeed, this pathogen can fre-
quently be encountered among rodents. A recent study
from eastern Russia (Zaytseva et al. 2007) revealed
that 1.1% of rodents screened carried this pathogen.
Unfortunately, the rodent species were not specied in
this study. In an earlier (military) study from the USA
(Olsufev and Emelyanova 1968), it was found that the
pathogen aected wild voles (Microtus arvalis Pall.) and
water rats (Arvicola terrestris L.). In Japan, wild black
rats (Rattus rattus) were collected in the city centre of
Tokyo and it was demonstrated that 6.5% were infected
with L. monocytogenes (Iida et al. 1998).
Food contaminated with L. monocytogenes is a sig-
nicant cause of human illness and death worldwide
(Mead et al. 1999). e pathogen can cause granulo-
matous septicemia in newborns and fatal meningitis in
adults (Farber and Peterkin 1991), but is opportunistic
and frequently encountered in the elderly, pregnant
mothers, and AIDS patients. e case-fatality rate in
recent outbreaks and sporadic cases is around 20%-30%
(WHO, Geneva, http://www.who.int). Antibiotics eec-
tive against Listeria species include vancomycin, ampi-
cillin, linezolid, azithromycin, and ciprooxacin.
Parasites
Toxoplasmosis
Toxoplasma gondii is a protozoan parasite with a com-
plex life cycle. Cats play a key-role as denitive hosts:
they acquire infection from their prey species, such as
rodents (intermediate hosts). Infected cats shed oocysts
with their feces. Livestock animals that take up oocysts
from their environment can form tissue-cysts in their
meat and organs (Van Knapen 1989). If their meat is
consumed without proper cooking, the parasite may
still be viable and is transferred to humans (Kijlstra and
Jongert 2008). is process is called acute infection and
most cases via this route are asymptomatic (although
the immunosuppressed form an exception). Congenital
infection is another route of transmission of the disease
and may have serious and sometimes even fatal conse-
quences. is occurs when a woman is infected for the
rst time during pregnancy and the parasite crosses the
placenta and invades the tissues of the fetus (Gilbert
2000). e severity of infection depends on the time of
infection during the pregnancy: severity is the greatest
during the early stage of pregnancy and infection may
then result in hydrocephalus, mental retardation or
even spontaneous abortion. Infection at a later stage
during gestation may lead to milder symptoms such as,
e.g., ocular toxoplasmosis. e risk of transmission from
mother to fetus is highest during the third trimester of
the gestation, when contact of maternal and fetal circu-
lation is more likely to occur (Rothova and Kijlstra 1989).
Of these infected children, only a part will perceive clini-
cal symptoms at birth. At a later age however, the disease
may manifest itself and result in eye lesions.
Rodents have been shown to carry Toxoplasma gondii
(Dubey et al. 1995; Hejlicek et al. 1997; Kijlstra et al. 2008;
Smith and Frenkel 1995; Tizard et al. 1978), and they are
244 B G Meerburg et al.
therefore associated with transmission of this parasite
to pigs (Kijlstra et al. 2004; Meerburg 2006; Meerburg
and Kijlstra 2006; Weigel et al. 1995). It remains unclear
whether the role of rodents is to provide a constant res-
ervoir of infection for cats or whether they directly trans-
mit the disease to pigs, if the latter consume dead rodent
carcasses or even live rodents (Kijlstra et al. 2008).
Babesiosis
Babesiosis is an uncommon parasitic disease caused by
piroplasms, protozoan parasites of the Babesia genus
(Homer et al. 2000). About 70 dierent types of Babesia
protozoa exist. e pathogen is usually transmitted by
ticks, sometimes in conjunction with Lyme disease as is
the case with Babesia microti that uses the same tick vec-
tor, Ixodes scapularis (Benson et al. 2004). Moreover, in
babesia-endemic areas, the organism can also be trans-
mitted by blood transfusion (Leiby 2006; Wei et al. 2001).
Depending on the Babesia strain, infection may vary
from aymptomatic (as is often the case with B. microti in
the USA and Japan), or cause a mild non-specic illness
(B. divergens in Europe). Sometimes, infection results in
severe disease, especially in young (neonatals), the elder
and immunosuppressed people, and fatalities do occur
(Hatcher et al. 2001). In regions where malaria is present,
symptoms are often confused with that of Plasmodium
infection. Due to its clinical appearance, the disease is
often underdiagnosed and underreported.
Rodents can be reservoirs of the infection, especially
for infection with B. microti. is species is commonly
distributed throughout North America (Goethert et al.
2003), Eurasia (Zamoto et al. 2004), Europe (Casati et al.
2006; Duh et al. 2003), and Japan (Tsuji et al. 2001), and
can cause human disease. After six human cases of clini-
cal babesiosis were diagnosed, examination of rodents
on Nantucket Island, USA disclosed infections with
Babesia microti in eld mice (Microtus pensylvanicus)
and in white-footed mice (Peromyscus leucopus) (Healy
et al. 1976). After a human case of babesiosis in Taiwan,
three species of rats (black rats, R. rattus, Norway rats, R.
norvegicus, and spiny rats, R. coxinga) from areas around
the patient’s neighborhood were trapped and examined
for the prevalence of B. microti infection by routine tech-
niques of examining blood smears and inoculating blood
into hamsters. e infection rate was about 67% and 83%
as determined by examining blood smears and inocu-
lating hamsters, respectively (Shih et al. 1997). In Japan
(Shiota et al. 1984), parasite presence was demonstrated
in Japanese eld rodents, (A. speciosus) and (Apodemus
argenteus). In a study from Poland (Karbowiak 2004),
prevalence of infection for B. microti in Microtus arvalis
varied from 9 to 33%, in M. agrestis it was almost 50%
and in M. oeconomus between 7 and 50%. Myodes spp.
voles and Apodemus spp. mice played an inferior role as
zoonotic reservoir for Babesia microti. In Slovakia (Duh
et al. 2003), prevalence among yellow-necked mice (A.
avicollis) was 12% and prevalence among bank voles
(M. glareolus) 16%. Recently, in ailand a new type of
rodent babesia (Dantrakool et al. 2004) that seems to
be phylogenetically closest to the canine B. canis, was
found in bandicoot rats (Bandicota indica). Whether
this species can also infect humans, as is the case with B.
microti is not known yet.
Cryptosporidiosis
Cryptosporidiosis is a diarrheal disease that is caused
by microscopic parasites of the genus Cryptosporidium.
Cryptosporidium thrives in the intestinal tracts of
infected humans or animals. As a consequence, the
parasite is found in soil, food, water, or surfaces that
have been contaminated with infected human or animal
feces. Major outbreaks are often associated with drink-
ing of infected water.
Symptoms of cryptosporidiosis generally begin 2–10
days (average 7 days) after becoming infected with the
parasite. In persons with a healthy immune system,
symptoms (which could include watery diarrhea, stom-
ach cramps, dehydration, nausea, vomiting, fever, weight
loss) usually last about 1–2 weeks (WHO, Geneva, http://
www.who.int). However, in the immunosuppressed the
disease can lead to more severe problems, which is a
major treatment problem in AIDS patients.
Moreover, cryptosporidiosis has emerged as a major
cause of neonatal mortality in livestock, principally
in lambs and calves, thus causing economic losses for
farmers.
Rodents seem an important source of cryptosporidial
oocysts, especially in sylvatic and bushy areas. ere,
they impose the highest risk for infection in humans.
A survey in Spain showed that Cryptosporidium is
present in various rodents, such as the wood mouse
(A. sylvaticus), the yellow-necked mouse (A. avicol-
lis), the Algerian mouse (Mus spretus), the black rat (R.
rattus), and the bank vole (Myodes glareolus) (Torres
et al. 2000). Studies in the USA (Klesius et al. 1986) and
the United Kingdom (Chalmers et al. 1997) reported
that the house mouse (M. musculus) can be infected by
Cryptosporidium and that prevalences vary between
22–30%. In the UK, prevalence in wood mice (Apodemus
sylvaticus) was 22% (Chalmers et al. 1997). On Skomer
Island, a part of the UK, prevalence in Skomer bank voles
(Myodes glareolus skomerensis) for Cryptosporidium was
51% (Bull et al. 1998). In Finland, the prevalence of crypt-
osporidia was determined in high density populations
of Microtus agrestis and Myodes glareolus (Laakkonen
et al. 1994). Prevalence in eld voles (Microtus agres-
tis) and bank voles (Myodes glareolus) was 2%, while
in root voles (Microtus oeconomus) no infection was
Rodent-borne diseases and their risks for public health 245
found (Laakkonen et al. 1994). On farms in the United
Kingdom (Webster and Macdonald 1995) 63% of the
Norway rats (R. norvegicus) was infected, while in Japan
infection rates in the same species varied between
2-21% (Iseki 1986; Miyaji et al. 1989; Yamura et al. 1990).
In black rats in Japan, rates varied between 0–49% (Iseki
1986; Miyaji et al. 1989; Yamura et al. 1990). In Australia,
the ‘cattle’ genotype of Cryptosporidium was isolated in
5 mice trapped on wheat-sheep farms. ese and earlier
studies indicate that sheep and cattle may transmit the
‘cattle’ genotype to mice. Subsequently mice may trans-
mit Cryptosporidium to other domestic animals and to
humans (Morgan et al. 1999).
Rodents can share their habitat with farm animals,
or travel through grazing land used by them, enabling
ample opportunity for transmission of Cryptosporidia
from them to the farm animals. Moreover, rodents are
found commonly in urban areas, thus providing a link
between rural and urban disease foci. ey could con-
tribute to many of the sporadic human cases of crypt-
osporidiosis in towns and cities, by leaving their many
small droppings wherever they forage, thus contaminat-
ing human and animal feed stores and accommodation
(Sturdee et al. 1999).
Chagas’ disease
Chagas’ disease is a tropical parasitic disease which
occurs in the Americas, particularly in large parts of
South America. It is caused by a agellate Trypanosoma
cruzi that is transmitted to humans and other mam-
mals mostly by blood-sucking assassin triatomines of
the subfamily Triatominae (Family Reduviidae). ese
hemipteran insects deposits parasite-laden feces on the
skin, while feeding.
Chagas’ disease currently aects 16–18 million peo-
ple, with some 100 million (25% of population in Latin
America) at risk of acquiring the disease (WHO, Geneva,
http://www.who.int) killing around 50,000 people
annually. Secondary transmission routes include blood
transfusion, congenital infection, tissue transplants, and
food-borne transmission (WHO, Geneva, http://www.
who.int).
e initial phase is acute and can be characterized by
patent parasitemia during 40–60 days. Because clinical
symptoms are generally mild (except for severe neu-
rological complications in children) and atypical, the
infection with T. cruzi is often not recognized. After the
acute phase, patients enter the indeterminate form of
the chronic phase that may last for several years or per-
sist indenitely (Buscaglia and Di Noia 2003). During
this long interval, infected persons themselves also form
a parasite reservoir. Up to 20 years after infection, 35%
of the patients will develop pathological signs such as
cardiomyopathy, peripheral nervous system damage or
dysfunction of the digestive tract often leading to mega-
esophagus and/or megacolon (Buscaglia and Di Noia
2003). During this period, most fatalities occur.
It is thought that rodents can act as a disease reser-
voir, although the exact domestic/peridomestic and
sylvatic transmission cycles of Trypanosoma cruzi
remain unclear. However, the parasite was found in
many rodent species such as: black rats (R. rattus),
brown rats (R. norvegicus), house mice (M. musculus),
wood rats (Neotoma micropus canescens), the hispid
pocket mice (Perognathus hispidus), Mexican spiny
pocket mice (Liomys irroratus), Northern grasshopper
mice (Onychomys leucogaster), white troated wood rats
(Neotoma albigula albigula), rusty antelope squirrels
(Citellus leucurus cinnamoneus), pygmy mice (Baiomys
musculus), Jaliscan cotton rats (Sigmodon mascotensis)
and fulves, harvest mice (Reithrodontomys fulvescens),
and the punaré (Trichomys apereoides) (Burkholder
et al. 1980; Cortez et al. 2006; Herrera and Urdaneta-
Morales 1997; Herrera et al. 2005; Mota et al. 2007; Wood
1949; Wood and Wood 1961). Some triatomine species
are associated with wild nesting vertebrates (Ramsey
et al. 2000; Usinger et al. 1966).
Leishmaniasis
Leishmaniasis is caused by the protozoa Leishmania.
e leishmaniases can be divided into 3 major clinical
syndromes: cutaneous (CL), mucocutaneous (MCL),and
visceral leishmaniasis (VL), or kala-azar. e pathogen
is transmitted by the bite of an infected female sandy
(Phlebotomus sp. in the Old World and Lutzomyia sp.
in the New World), and humans are mostly incidental
hosts. e result of infection in humans can vary from
a chronic skin ulcer, to erosive mucosal disease with
progressive destruction of the nasopharynx and severe
facial disgurement in case of CL, to a life-threatening
systemic infection with hepato-splenomegly in case of
VL (Hepburn 2000).
VL is caused by Leishmania donovani in India
and Eastern Africa and by L. infantum chagasi in the
Mediterranean basin, western Africa and Latin America
(WHO, Geneva, http://www.who.int). In contrast to CL
and MCL, in which the protozoan parasite is histologically
localized, visceral disease is caused by parasite growth
within reticuloendothelial cells throughout the body.
Leishmaniasis is endemic from northern Argentina
to southern Texas (not in Uruguay, Chile, or Canada), in
southern Europe, Asia (not southeast Asia), the Middle
East, and Africa (particularly east and north Africa,
with sporadic cases elsewhere), but not in Australia or
Oceania (Herwaldt 1999). In total, about 350 million
people are at-risk (Herwaldt 1999).ere are about
500,000 cases of visceral leishmaniasis each year; over
90% of worldwide cases are in Bangladesh, northeastern
246 B G Meerburg et al.
India (particularly Bihar State), Nepal, and Sudan
(Old World), and in northeastern Brazil (New World)
(Herwaldt 1999). If clinically evident but untreated,
visceral leishmaniasis causes life-threatening systemic
infection (Herwaldt 1999).
ere are 1–1.5 million new cases of cutaneous lei-
shamaniasis each year (Desjeux 2004) of which more
than 90% occur in Afghanistan, Algeria, Iran, Iraq, Saudi
Arabia and Syria, in the ‘Old World,’ and Brazil and Peru
in the ‘New World’ (Hepburn 2000). In the Old World,
cutaneous leishmaniasis is caused by Leishmania
tropica in urban areas and Leishmania major in dry
desert areas. In the New World, Leishmania leishma-
nia (e.g., Leishmania amazonensis, Leishmania mexi-
cana, Leishmania chagasi) and Leishmania viannia
(e.g., Leishmania braziliensis, Leishmania guyanensis,
Leishmania panamensis) can cause the disease.
e pathogen is most prevalent in rural areas or for-
ests, with a moist climate in which the sand ies thrive.
Rodents can form a reservoir for the protozoa. In Mexico,
the protozoa was found in the Yucatan deer mouse
(Peromyscus yucatanicus) and the black-eared rice
rat (Oryzomys melanotis) (Chable-Santos et al. 1995).
Leishmania panamensis was also reported in black rats
(R. rattus) in Brazil (Vasconcelos et al. 1994). Moreover,
spiny rats (Proechimys spp.) appear to be reservoir
hosts of L. amazonensis in Brazil and French Guyana
(Ashford 1996). Also, the paca (Cuniculus paca) is said
to be the reservoir host of a little known pathogen spe-
cies L. lainsoni (Ashford 1996). In Belize, the climbing rat
(Ototylomys phyllotis) is the principal host of Leishmania
mexicana (Disney 1968), with 53–56% of the captured
rodents infected. In the southern USA, the woodrat
(Neotoma micropus) forms a reservoir for L. mexicana
in Texas (Kerr et al. 1995). In Jordan, 23% of the fat sand
rats (Psammomys obesus) were carrying the pathogen
L. major (Saliba et al. 1994). In Israel, the fat sand rat
(P. obesus), Sundevall’s gerbil (Meriones crassus) and
probably the short-tailed bandicoot rat (Nesokia indica)
function as reservoirs for L. major (Schlein et al. 1984).
e same is happening in the Nile grass rat (Arvicanthis
niloticus luctuosus) in Sudan (Hoogstraal and Heyneman
1969) and in Libyan jirds (Meriones libycus) and great
gerbils (Rhombomys opimus) in Iran (Yaghoobi-Ershadi
et al. 1996) In Egypt, the great Egyptian gerbil (Gerbillus
pyramidum) are carriers of L. major (Fryau et al. 1993).
In the old world, black rats (R. rattus) are mentioned as
incidental carriers of L. donovani (Ashford 1996). In con-
clusion, it can be said that rodents play an important role
both as reservoirs and incidental hosts for this protozoa.
Giardiasis
Giardiasis is an infection of the intestines that is caused
by Giardia lamblia, a agellated protozoan parasite
(Adam 1991) that is present throughout the world. One
of the important symptoms of the disease is diarrhea,
both in man and in animals. Other symptoms include
loss of appetite, lethargy, fever, stomach problems
(cramps), projectile vomiting (rare), bloating, and atu-
lence. Symptoms typically begin 1–2 weeks after infec-
tion and may wane and reappear cyclically (Adam 1991).
Symptoms are caused by the thick coating of Giardia
organisms coating the inside of the small intestine and
blocking nutrient absorption. Most cases are asymp-
tomatic; only about a third of infected people exhibit
symptoms. Untreated, symptoms may last for six weeks
or longer.
Humans can get infected through consumption of
contaminated food products or infected water, also dur-
ing swimming. Giardiasis usually occurs sporadically,
although outbreaks do occur.
Rodents may form a pathogen reservoir and could
cause contamination of the water with G. lamblia cysts
(Appelbee et al. 2005; Bednarska et al. 2007). Beavers
and muskrats have been reported as carriers of this
parasite (Frost et al. 1980). Voles also seem to be an
important reservoir (Wallis et al. 1984): red-back voles
(C. gapperi Vigors), meadow voles (M. pennsilvanicus),
water voles (M. richardsoni), and the long-tailed voles
(M. longicaudus) have been identied as carriers with
high parasite prevalence (Pacha et al. 1987; Wallis et al.
1984). Less important, though still carriers, are deer mice
(P. maniculatus) (Wallis et al. 1984) and yellow-necked
mice (A. avicollis) (Karanis et al. 1996).
Taeniasis
Taeniidae are highly characteristic tapeworms in the
subfamilies Taeniinae and Echinococcinae. Adults are
present in the intestine of carnivorous and omnivorous
mammals, including humans, throughout the world. Life
cycles for Taenia are indirect and consistently involve
two mammalian hosts, a carnivorous or omnivorous
denitive host (e.g. canids, felids, viverrids, mustelids,
hyaenids, and humans) and a herbivorous intermediate
host (principally artiodactyls, rodents, and lagomorphs)
(Hoberg 2002).
e mature tapeworms are 40– 100 cm long and
inhibit the small intestines of the denitive host (Krauss
et al. 2003). is infection is called taeniasis. Humans
are the denitive hosts for Taenia solium (the pork tape-
worm) and T. saginata (the beef tapeworm). Humans
are also the denitive hosts for T. asiatica, a newly rec-
ognized tapeworm found in Asia (Anonymous 2005).
Animals are the denitive hosts for T. crassiceps, T. ovis,
T. taeniaeformis, T. hydatigena, T. multiceps, T. serialis,
and T. brauni (Anonymous 2005). Taenia larvae are
found in the muscles, central nervous system (CNS),
and other tissues of the intermediate hosts. Larvae are
Rodent-borne diseases and their risks for public health 247
more likely to cause disease than the adult tapeworms.
ere are two forms of larval infection, cysticercosis and
coenurosis (Anonymous 2005).
Infection with the larvae of Taenia solium, T. saginata,
T. crassiceps, T. ovis, T. taeniaeformis, or T. hydatigena
is called cysticercosis. e larvae of these organisms
are called cysticerci (Anonymous 2005). Humans are
intermediate hosts for T. solium, T. crassiceps, T. ovis,
T. taeniaeformis, and T. hydatigena. T. solium is often
found in humans; the other four species are very rare
(Anonymous 2005). T. solium is the only Taenia species
for which humans are both the denitive and an inter-
mediate host. Animals can be intermediate hosts for
these ve species as well as for T. saginata and T. asiatica
(Anonymous 2005).
Infection with the larval forms of T. multiceps, T. seri-
alis and T. brauni is called coenurosis. Human coenurosis
is much less common than cysticercosis. Approximately
100 or more cases have been reported worldwide
(Anonymous 2005), mainly in Africa (Hoberg 2002) and
South America. Only a few cases have been documented
in the United States and Europe (Anonymous 2005).
Humans can be intermediate hosts for T. multiceps,
T. serialis, and T. brauni. Animals can also be intermedi-
ate hosts for these three species (Anonymous 2005).
In humans, taeniasis is caused by eating inadequately
cooked pork (T. solium and T. asiatica) or beef (T. sagi-
nata) (Anonymous 2005). In animals, taeniasis is caused
by T. crassiceps, T. ovis, T. taeniaeformis, T. hydatigena,
T. multiceps, T. serialis, and T. brauni and is acquired by
eating tissues from a variety of intermediate hosts includ-
ing ruminants, rabbits and rodents (Anonymous 2005).
Eggs are shed in proglottids with the feces of denitive
hosts and are ingested by the intermediate hosts, allow-
ing the oncospheres to hatch in the intestine (Krauss
et al. 2003). Humans usually ingest tapeworm eggs on
fruits and vegetables or acquire them directly from the
soil. ey can also acquire infection by consumption of
contaminated water (Anonymous 2005).
Worldwide, taeniasis and cysticercosis are common
parasitic infections: 2-3 million people are thought to
be infected with adult T. solium, 45 million with adult
T. saginata, and 50 million with T. solium cysticerci. An
estimated 50,000 people die annually from the CNS or
cardiac complications (Anonymous 2005).
e level of involvement of rodents in human infec-
tion depends on the Taenia species that is contracted.
Rodents (and also lagomorphs) are intermediate hosts
of metacestodes of T. multiceps and T. serialis and con-
tribute to the l