In this study I. ricinus was the predominant tick species infesting companion animals, followed by the hedgehog tick, I. hexagonus. This is in analogy with other studies in North-Western Europe. Similar results were reported from The Netherlands , Germany  and the UK [17, 18]. In the UK and Ireland, the fox tick I. canisuga was also frequently recovered from dogs [17, 18]. Ogden et al.  found I. ricinus on a significantly higher proportion of dogs than cats, while I. hexagonus was more frequently found on cats. They suggested that differences in behaviour between dogs and cats could affect their likelihood of encountering both Ixodes species. This hypothesis was not supported by our data, with similar proportions of dogs and cats carrying I. ricinus and I. hexagonus. Equal proportions of both Ixodes species were also recovered from dogs and cats in Germany  and The Netherlands . The recovered Ixodes ticks were mainly adults, as was also observed in other studies in dogs and cats [16, 18, 19]. Although it cannot be excluded that some nymphs and larvae were overlooked in the clinical inspections, adult Ixodes ticks are known to attach preferably to large or medium-sized mammals, including dogs and cats .
Generally, nymphs and adults of I. ricinus show a marked seasonal variability in their questing activity, with a first peak in late spring and early summer and a second peak in autumn [21, 22]. This seasonal pattern was also observed in the number of ticks recovered from dogs by  and . In this study, we observed an obvious spring peak, but only low numbers of I. ricinus were submitted in autumn. Although participation rates of the veterinarians may have introduced a bias, tick questing activity may be variable from year to year and a single spring or summer peak has previously been observed for questing I. ricinus nymphs  and for adult I. ricinus on dogs . Most I. hexagonus were also recovered in the spring in our study, but a second, smaller peak occurred in October. Only limited published information is available on the seasonal abundance of I. hexagonus ticks, but seasonal fluctuations in numbers of I. hexagonus collected from dogs and hedgehogs are generally weaker compared to I. ricinus[16, 24].
The majority of infected ticks were found in Northern Belgium (Flanders). The north-eastern part of Belgium (the Campine) is known to be heavily infested with Ixodes and has a relatively high incidence of Lyme borreliosis . Although the south-eastern part of Belgium has a lot of forest and would also be expected to have a lot of good Ixodes habitats, the number of ticks positive for tick-borne diseases from that area was lower in our study. The reason for this remains unclear.
There was no significant difference between I. ricinus and I. hexagonus in the proportion of ticks that contained DNA from B. burgdorferi s.l., A. phagocytophilum or R. helvetica. Although both I. ricinus and I. hexagonus can be vectors of Borrelia spp. and A. phagocytophilum[26, 27], I. ricinus is considered to be the principal vector of these pathogens [27, 28]. As a nest dwelling species, I. hexagonus will have little direct contact with humans. Nevertheless, I. hexagonus could be a vector for the transmission of Borrelia spp. and A. phagocytophilum to hedgehogs, and then via co-infections with I. ricinus indirectly to humans [24, 27].
The percentage of ticks positive for Borrelia spp. was within the range of infection rates (2.3-22%) in ticks collected from dogs in other European countries [3, 16, 29–31]. Similar infection rates were also reported in questing I. ricinus in Belgium (12-23%) [32, 33]. Within the B. burgdorferi sensu lato group, B. afzelii and B. garinii are the most common species in The Netherlands, Belgium and northern France, while B. burgdorferi s.s. is less common in this region . B. valaisiana has also been repeatedly found in The Netherlands (e.g. 3, 29). Together with other recent studies in Belgium and Luxembourg [33, 35] our results show that besides B. valaisiana, B. spielmanii and B. lusitaniae are also present in Ixodes in the Benelux region. B. garinii, B. afzelii and B. burgdorferi s.s. are well known to be pathogenic for humans, but the pathogenic significance of the other species is still unclear [36, 37].
The majority of dogs and cats that are exposed to Borrelia infections remain clinically normal , which was also the case in this study. Most cases of canine Lyme borreliosis are associated with B. burgdorferi s.s. Whilst B. burgdorferi s.s. is the only Borrelia species in the USA, prevalences of B. burgdorferi s.s. in Europe are much lower , as was also the case in this study. This may explain why lyme borreliosis is frequently diagnosed in dogs in endemic regions in the USA, but less frequently in European dogs.
Although Borrelia can be transmitted from dogs to ticks, dogs are not considered as important reservoir hosts  and studies examining seropositivity in dogs, their owners and other local residents found no correlation between dog ownership and infection risk [38–40]. Nevertheless, dogs and cats can be used as sentinels for lyme borreliosis. Serological studies in the USA showed that exposure of dogs to B. burgdorferi mimics the geographical distribution of reports of Lyme borreliosis in humans  and the use of dog sera to detect and quantify the risk of Lyme borreliosis for humans in a certain region is considered to be more sensitive than the use of incidence reports of human clinical cases . Seroprevalence has been found greater in dogs than in humans due to their greater habitat exposure, lack of protective clothing and inability to check themselves from ticks . The use of dog sera also has the advantage over human serology that the seroprevalence among dogs is more likely to reflect the actual environmental risk of Lyme borreliosis, because of the short half-life of canine antibodies against B. burgdorferi. However, serological studies are often limited by small sample sizes, and false positive results are possible. Detection of Borrelia DNA in ticks collected from dogs and cats can be a valuable alternative .
The A. phagocytophilum infection rate in this study was much higher than previously reported point prevalences in Ixodes ticks from dogs in The Netherlands and Poland (1.6-2.9%) [3, 30]. Seroprevalences in dogs in Europe are also very variable, with prevalences from < 5% to > 50% . The percentage of seropositive dogs depends on the dog population sampled (e.g. healthy dogs vs. dogs with signs of tick-borne diseases) and geographical variation in exposure to ticks and reservoir hosts. The high infection rate of A. phagocytophilum in ticks collected from dogs in this study is in contrast with the low incidence of human granulocytic anaplasmosis (HGA) in Belgium (< 100/year) . However, there is a discrepancy between the official (low) incidence rates  and the high number of HGA cases that are detected in specific surveys [42, 43]. In a recent 10-year serological survey in patients with symptoms of tick-borne infections, 31% of the samples were positive, and 111 cases of HGA were confirmed . These data suggest that Belgium is a hot spot for HGA and that many cases of HGA probably remain undiagnosed .
Most dogs infected with A. phagocytophilum probably remain healthy . The most common clinical signs in dogs that develop illness are lethargy, fever and lameness. In the present study, no association was found between clinical signs and the presence of A. phagocytophilum in ticks collected from these animals. Although identical 16S rRNA gene sequences have been found in canine and human isolates of A. phagocytophilum in Europe [44, 45], dogs are not thought to be important reservoirs for A. phagocytophilum, since bacteremia is of short duration in this species . In Slovenia, no difference in seroprevalence was observed between people with or without exposure to dogs .
Ixodes spp. can transmit several Rickettsia species belonging to the ‘spotted fever’ group, such as Rickettsia helvetica and R. monacensis. We found R. helvetica DNA in 14.1% of Ixodes ticks collected from dogs and cats in Belgium. In the Netherlands, 24.7% I. ricinus ticks and 0.8% I. hexagonus ticks collected from dogs, cats and a hedgehog were infected with R. helvetica. In Switzerland, 40.9% and 17.6% of Ixodes ticks collected from cats and dogs, respectively, tested positive for R. helvetica with a gltA-specific TaqMan PCR system . R. helvetica is a suspected pathogen in humans. Symptoms that have been associated with R. helvetica infections include fever, headache, arthralgia, myalgia and perimyocarditis . The high prevalence of R. helvetica in Ixodes spp. and the high abundance of these tick species suggest that the likelihood of transmission of R. helvetica to humans should be high . However, despite a high infection rate (19%) of R. helvetica in ticks collected from humans in the Netherlands, no association between symptoms and R. helvetica was found .
The clinical importance of R. helvetica in domestic animals is as yet uncertain and it is also unknown whether dogs and cats can serve as a reservoir after infection. The fact that the estimated prevalence of R. helvetica in ticks collected from dogs, cats and roe deer was higher than in ticks collected from the vegetation [48, 50] may indicate that large animals act as a reservoir for R. helvetica.
Next to I. ricinus and I. hexagonus, small numbers of R. sanguineus and D. reticulatus were collected from dogs. All submitted R. sanguineus ticks were considered to be imported, since they were all collected from dogs with a history of travelling abroad, mostly to Southern Europe. One R. sanguineus tick contained DNA from R. massiliae. R. massiliae is suspected to be the main cause of Mediterranean spotted fever in Spain .
Although most submitted Dermacentor ticks were also from dogs with a travel history, D. reticulatus ticks were repeatedly sampled from one particular dog that had never been outside Belgium. Flagging confirmed the presence of questing Dermacentor ticks in the area where the dog was walked daily . This was the first finding of an indigenous population of D. reticulatus in Belgium. Further investigations have revealed the presence of at least 4 other foci of D. reticulatus in Belgium (51, M. Madder, unpublished results).