Overall, 175/624 cats (28%) were tested positive for at least one vector-borne pathogen. Within the group of cats with a history of any time spent abroad, this prevalence was 30% (110/363 cats). Previous studies in dogs living in Germany have found prevalences of 35% (imported dogs ), 13% (travelling dogs ), and 44% (any history of time spent abroad ). Any comparison of the prevalence of infection with vector-borne pathogens in dogs and cats is of limited value, for several reasons which include the following: different prevalence rates of some pathogens in dogs and in cats in endemic countries, variation in study design, difference in immune responses to infection in dogs and cats, different host preferences of specific pathogens, and inborn resistance mechanisms for some pathogens . Moreover, cats exhibit a more thorough cleaning behaviour than dogs, which may cause them to remove a potential vector and therefore inhibit any possible disease transmission . Cats far more rarely accompany their owners on travels abroad, and consequently there was a higher ratio of imported cats (98%, 356/363 cats) compared to cats which had been travelled with their owners (2%, 6/363 cats). Since all 6 cats had outdoor access, their risk of coming into contact with a relevant vector is similar to that of travel companion dogs. However, due to the small number of cats which had travelled, any attempts at interpretation of this data is not feasible. Beside that, the prevalence of vector-borne infections in imported cats and dogs are approximately the same.
The prevalence of vector-borne infections varies not only among countries but also within the countries themselves, as it is determined largely by geographical and climatic conditions, as well as the presence of suitable vectors and reservoirs for the pathogen [17, 18]. Not only the import of cats from abroad but also international travel and conveyance of goods is increasing in frequency. Coupled with the change in climate in many parts of Europe, this could contribute to an increased spread of pathogens and their potential vectors into previously non-endemic areas such as Germany, where they may spread further and form reservoirs for infection. Under suitable conditions, pathogens transmitted via imported vectors may cause infection in competent hosts endemic to Germany, of which cats are one example. Moreover, endemic but potentially competent vectors may be infected with these previously non-endemic pathogens during a blood meal on infected cats, and proceed to contribute to the spread of these pathogens [2, 19, 20]. One example are isolated cases of autochthonous infections with D. repens [21–23] and L. infantum  in dogs in Germany, which has not been described in cats.
Direct detection methods demonstrate the presence of deoxyribonucleic acid or the antigen of a pathogen. PCR assays are used primarily in acute or peracute infections prior to seroconversion, or in the case of kittens due to the presence of maternal antibodies . Despite the high sensitivity of the PCRs used in this study, false negatives are not uncommon in cats due to their propensity for having comparatively low pathogen concentrations in blood. This is suspected to be the case in Rickettsia spp., A. phagocytophilum or Ehrlichia spp. infections in cats [25, 26].
Indirect detection methods demonstrate the presence of antibodies after contact with the pathogen. This does not correlate with the presence of disease, as seroconversion may not occur until two to three weeks after exposure, and antibodies may be detectable for up to several years after disease resolution, both depending on the pathogen. Generally, it is possible to distinguish more recent infections from those which date further back by means of simultaneous Immunoglobulin M levels, or serum pairs taken at intervals of 2 to 4 weeks. However, the former is unusual in any routine diagnostics for the pathogens discussed, while the latter is often not feasible in practice. The indirect IFAT utilised in this study detected Immunoglobulin G antibodies for all pathogens. Additionally, due to the subjective microscopic assessment of samples, there is a possibility of human error negatively influencing the sensitivity in cases with low antibody titres. Further limitations may be due to cross reactivity with other pathogens, false negative results in very young animals or those which are immunosuppressed, as well as in those cases in which testing was done too early in the natural history of the disease and therefore prior to any seroconversion .
This study included detection assays for Leishmania spp., Hepatozoon spp., Ehrlichia spp., Rickettsia spp., and Dirofilaria spp. This selection was due to the framework of the corresponding testing panel offered by LABOKLIN, which facilitated the uniform testing of a population of cats by means of a set testing panel for a defined spectrum of pathogens. Due to the relatively late seroconversion of Leishmania spp. and the long prepatency of Dirofilaria spp., the prevalence of both pathogens may be higher than reported (Leishmania spp.: 4% (IFAT); Dirofilaria spp.: 0.2% (PCR)). In the following, every pathogen considered in this study will be discussed individually.
Cats in the Mediterranean countries are infected by the same Leishmania spp. as dogs in these regions, primarily L. infantum. There is much variation in the reported prevalence of Leishmania spp. in cats tested by indirect assays not only among different European countries but also across different regions within one country, ranging from 0.1–60% [1, 15, 28–52]. To the knowledge of the authors, there are no data on cats in Germany at this point in time. Utilising IFAT, this study found antibodies to Leishmania spp. in 22/624 cats (4%), and in 13 of the 363 cats (4%) with a history of any time spent abroad. Contrary to dogs or humans, in which horizontal or vertical transmission is possible, cats seem to be infected solely by vector transmission . The prevalence of Leishmania spp. is lower in cats than in dogs, and cats are less likely to develop clinical signs if they are infected [1, 54]. Dogs are currently the only known primary reservoir of infection . While cats are presumed to also be reservoir, this has not yet been proven . There is little evidence on the susceptibility or resistance of cats to natural infection. Cats exhibit a more efficient T-helper cell-1 immune response than dogs, which may explain the lower prevalence of the pathogen in cats . However, the pathogenesis of feline leishmaniasis remains unclear, as well as the role of cats in the life cycle of the pathogen. It has been shown that sandflies may become infected with L. infantum during a blood meal on an infected cat, and consequently cats may be instrumental in the spread of the pathogen in areas with a high prevalence . It is therefore possible that the 22 cats in this study which were tested positive for antibodies might transmit L. infantum further within Germany, provided they are still infected with the pathogen. Suitable competent vectors like P. perniciosus have been described in the South of Germany , as has P. mascitti, a potentially competent vector [59, 60].
Depending on the specific test utilised, it is usually recommended to use a titre cut-off of 1:80 when performing Leishmania spp. IFAT in cats . In reference to this and according to manufacturer guidelines, this study used a cut-off of 1:64. Cross reactivity between different Leishmania spp. are probable in the 22 cats which tested positive in this study. Twelve of the 22 cats which tested positive (55%) were imported into Germany from Mediterranean countries and Southeast Europe, where L. infantum is endemic. One out of the 22 cats (5%) was imported from Brazil, where cats may be infected with not only L. infantum but also L. amazonensis or L. braziliensis [62–65]. In the remaining 9/22 cats (41%), it was not possible to obtain a travel or import history.
Infections with H. felis, H. canis, and H. silvestris have been described in cats. The prevalence of Hepatozoon spp. detected by PCR in Europe is between 0% and 38%, and all three Hepatozoon spp. which may infect cats in Europe have been previously described [1, 6, 28, 66–71]. To the knowledge of the authors, the prevalence of Hepatozoon spp. infections in cats in Germany is unknown. In this study, the pathogen was detected by PCR in 53/618 cats (9%). In 7 of these 53 cats (13%), which had been imported from Spain (n = 5), Greece, and Malta (n = 1, respectively), it was possible to detect H. felis via species differentiation. This result is in accordance with previous studies which have determined H. felis to be the primary infecting pathogen in cats [66–71]. In 39/53 cats which were tested positive for Hepatozoon spp. (74%), there was a history of travel/import consistent with an infection in an endemic area abroad. There is no evidence of autochthonous infections in cats within Germany, and therefore it is most likely, that the remaining 14/53 cats (26%) were also infected in an endemic region abroad. The only feline case report of an autochthonous infection with H. felis in Central Europe so far was from Austria .
There is little knowledge about the pathogenesis, replication cycle, host spectrum, and modes of transmission of Hepatozoon spp. in cats. In addition to vector transmission, there are reports of transplacental transmission in cats in the case of H. canis and H. felis [5, 72]. Therefore, any female cat which was tested positive in this study and had not been spayed (n = 7) might transmit the pathogen in Germany to their kittens, regardless of any contact with a vector. In the Hepatozoon spp. infected cats of this study, we detected coinfections with Leishmania spp. (n = 4), Rickettsia spp. (n = 3), and Ehrlichia spp. (n = 2).
E. canis or E. canis-like pathogens can infect cats [73, 74]. The prevalence of Ehrlichia spp. in the Mediterranean as tested by indirect detection methods (IFAT) in cats ranged between 1% and 18% [31, 32, 34, 36, 39, 75–79]. There does not seem to be any data on the prevalence of antibody testing in cats in Germany to this date. A study in 479 cats in South Germany did not demonstrate any Ehrlichia spp. DNA . Rhipicephalus sanguineus, which is a potential vector for E. canis, is only found in Germany for short durations in specific temperatures, or as populations in constantly heated buildings . Therefore, autochthonous natural infections in cats in Germany are unlikely.
Cross reactivity in indirect detection methods may occur with E. chaffensis (found in cats in the United States and Brazil) and E. ewingii (found in cats in the United States), as well as with A. phagocytophilum and A. platys at lower titres. Cross-reactivity due to contact with A. phagocytophilum in Germany cannot be excluded, especially in the group of 44 cats with a titre of 1:40.
Cats may be instrumental in the transmission cycle of some rickettsia of the spotted fever group (SFG), especially of R. conorii and R. felis [81, 82]. Dogs are a known reservoir for R. conorii and have been demonstrated to exhibit a clinical infection [83, 84]. This pathogen also has zoonotic potential. In cats, antibody titres to R. conorii have been shown after infections with Rhipicephalus sanguineus [75, 82, 85]. Seroprevalence has been examined in cats in Italy, Spain, and Portugal (IFAT/ELISA: 0-48.7%) [15, 31, 32, 34, 75, 85, 86]. Rickettsia felis is an established cause of the emerging flea-borne spotted fever, of which there have been several cases described in humans worldwide [87, 88]. Cats will have antibodies for Rickettsia spp. after infection (either natural or within the framework of an experiment) with fleas of the species Ctenocephalides felis . The pathogen has also been detected by PCR in previously non-infected fleas after a blood meal on infected cats . Consequently, Ctenocephalides felis is a competent vector and therefore autochthonous infections within Germany are possible.
This study utilised IFAT to detect antibodies, which is regarded as the gold standard for serological confirmation of pathogen contact in dogs and cats. There are however cross reactions between any of the more than 20 species in the spotted fever group . We detected antibodies to Rickettsia spp. in 52/467 cats (11%). In those 29 cats which were seropositive and had reportedly been imported from abroad, there is a possibility of infection with rickettsia in either Germany or their home country. In the four cats which had never left Germany, an infection with R. felis is most likely. Species differentiation by PCR was not performed.
Furthermore, the clinical importance of Rickettsia spp. infections in cats is still unknown. For example, one study evaluated clinically ill cats for evidence of rickettsial infections, but no association between positive antibody titres and fever could be shown and no febrile cat had a positive PCR result for R. felis or R. rickettsii .
Infection with Dirofilaria spp. occurs primarily in dogs but has also been described in cats . The prevalence of Dirofilaria spp. in cats varies between 0% and 33% across Europe [10, 31, 52, 91–99]. There does not seem to be any data on prevalence in Germany, specifically. A first case report in Europe describes a cat in Poland which was infected with D. repens and Wolbachia spp. . Only one of the 618 cats tested for microfilaria by means of PCR (0.2%) was tested positive, and species differentiation was not performed. A travel history was not available for this cat. It seems most likely that it was infected abroad in an endemic country, especially considering the existing coinfection with L. infantum. Cats are in general more resistant to Dirofilaria spp. infections compared to dogs . Additionally, some mosquito species which could function as vectors seem to prefer dogs to cats for their blood meals , which may explain the lower prevalence in cats. However due to some diagnostic peculiarities, the true prevalence in cats may be higher than that found in this study. A large fraction of the not yet mature pathogens is destroyed shortly after reaching the pulmonary arteries in cats, and consequently the duration of life of these pathogens is far shorter in cats (2–4 years) than it is in dogs (5–7 years) . Cats are rarely infected with more than five roundworms, which may be overlooked even in a post-mortem examination . Additionally, female roundworms are seen more in cats. Therefore microfilaraemia is rare in cats, as no male worms are available . Antigen testing is prone to false negative results due to the low level of pathogens in cats, and therefore direct detection methods should only be utilised coupled with at least one specific antibody test as well as imaging modalities [9, 104]. Another possibility to increase the sensitivity is the heat pre-treatment of feline serum and/or plasma samples before analysis , which was not carried out through.
In total, coinfections were detected in 22 out of 624 cats (4%) in this study. As it is known in dogs, coinfections may complicate the diagnoses and treatment in infected animals and may worsen the prognosis . Coinfections with multiple vector-borne pathogens occur in cats as well as in dogs and humans, but the clinical consequences are still unknown and have to be evaluated in further studies, especially in cats .
Leishmania and Ehrlichia spp. infections, which were present in 12 positive tested cats each, may cause an immunosuppression, possibly making infected animals more susceptible for infections with other pathogens . In 5 Ehrlichia spp. positive tested cats with low titres of 1:40, possible cross-reactions with A. phagocytophilum in Germany have to be taken in consideration.
Limitations of this study
Limitations of this study are mainly its retrospective design (e.g. no consistent histories) and the limit of pathogens included. Moreover, certain vector-borne infections as e.g. Cytauxzoon spp. could not be included. Furthermore, species differentiation for specific pathogens included in the study was not performed, except in the case of seven cats which were tested positive for H. felis. There was also no information on the presence or absence of ectoparasite prophylaxis in the cats, which may impact the prevalence of certain vector-borne pathogens. In the cats which had been travelled with their owners it was not possible to reliably document the duration of time or the time of the year spent in endemic countries. As many of the relevant vectors show pronounced seasonality, the time of year may significantly influence both incidence and prevalence of the pathogens they may transmit. The histories taken from the veterinarians only included the countries of stays abroad.