This is the first study documenting seroprevalence of T. gondii, N. caninum, L. infantum and Anaplasma spp., and the molecular detection and prevalence of B. henselae and three haemotropic mycoplasmas in cats in Tirana, Albania.
Toxoplasma gondii is one of the most prevalent cosmopolitan parasites, which can infect a wide spectrum of warm-blooded vertebrates. Based on its zoonotic nature, toxoplasmosis is one major public health issue worldwide and thus monitored closely in human medicine, but it is also considered as an important cause of reproductive disease in small ruminants. Felids, including domestic cats, are the only definitive hosts of this parasite. As cats are shedding Toxoplasma oocysts only for a short period, sero-surveys are suitable measures to study the epidemiology of the pathogen, and they can be used as an indicator of the environmental contamination [17–19]. Toxoplasmosis is widespread in south-east Europe . The average prevalence of IgG antibodies was reported to be approximately 50% in pregnant woman in Albania  and among Albanians who migrated to Italy  and this percentage of exposure represents the upper limit of the range of currently reported data from the Balkans [23–25]. The overall seroprevalence of 62.3% in the cats from Tirana is among the highest infection percentage observed in Europe where T. gondii antibodies were detected in up to 70% of the cats . Recently seroprevalences of 47% to 81% were reported in cats in Romania [27, 28] or of almost 48% in Hungary . Consistent with several other studies, the T. gondii seroprevalence in the cats from Tirana was positively associated with the age of the cats [28, 29]. The association between seropositivity and gender, however, is unclear: in most surveys, including this study, the overall prevalence of infection in male and female cats did not differ significantly [27, 28]; but there are some studies that found T. gondii exposure significantly more often among female cats [29, 30]. The high prevalence of T. gondii antibodies in the cats from Tirana, Albania, is certainly related to their origin from suburban habitats with constant access to the outdoors, which has been identified as a risk factor of the infection . The prevalence of antibodies to T. gondii in pregnant women in Albania was similar to the proportion observed in other parts of Europe. Beside consumption of undercooked meat, ‘direct soil contact’ has been thought to be associated with the seropositivity . Therefore, given the high percentage of exposure to T. gondii in cats from Tirana, the seroprevalence of T. gondii in cats should be monitored further as it gives an indication of the contamination of the environment with oocysts and thus the potential implication for public health. Furthermore, the public should be informed about the risks of undercooked meat consumption.
Neospora caninum is primarily associated with dogs and cattle, and neosporosis continues to be an important cause of abortion in cattle. Felines belong to the wide spectrum of domestic and wild animals that are exposed to N. caninum, but neither Neospora have been isolated from them nor were clinical signs of disease reported . For Albania, there is no information on the prevalence of N. caninum in cattle, but a recently conducted study estimated a seroprevalence of 18.3% in dogs (Hamel et al., unpublished). Sero-surveys in cats indicated 0 to almost 25% prevalence in domestic cats [29, 32–36]. The seroprevalence observed in the Tirana cats was thus in the range of results of other surveys, and consistent with a study from Italy .
Seropositivity was not associated to the age or sex of the cats. However, other studies found a positive correlation of the anti-N. caninum antibody prevalence and the cats’ age [33, 35, 36]. As indicated through the significant association of positive samples for N. caninum and T. gondii antibodies in this study (including the considerable number of cats with borderline anti-N. caninum titres) and similar observations in other studies [34, 35], serological cross-reactions between these two closely related apicomplexan protozoa cannot be ruled-out because N. caninum and T. gondii share common antigens .
None of the blood smears were positive for morulae, and no A. phagocytophilum- DNA was detected in any blood sample, but exposure to Anaplasma infection has been demonstrated through low titres in the serum of three cats (2.1%) using the commercial A. phagocytophilum- IFAT. Similar results were obtained previously when cat sera from Spain were tested [38, 39]. In contrast, 40% of 30 dogs from Tirana tested positive with an A. phagocytophilum-IFAT , and in a recently completed study with 602 dogs under veterinary care from Albania, a seropositivity of 24% was established (Hamel et al., unpublished). The main vector of A. phagocytophilum in Europe, the tick Ixodes ricinus, is abundant in Albania and has been recorded on dogs in the city and district of Tirana in the past . However, recently conducted surveys revealed R. sanguineus as the predominating species of ticks parasitizing dogs from Tirana while I. ricinus was recovered from single animals in low numbers only . On cats from Tirana, however, no I. ricinus at all, but few R. sanguineus ticks were found previously  as well as in the present study. The occurrence of two anaplasmas, A. phagocytophilum and A. platys, has been confirmed in blood samples of dogs from Albania with the latter species dominating (Hamel et al., unpublished). A. platys is thought to be transmitted by R. sanguineus and feline A. platys infections have been reported previously . Cross-reactivity of serum samples between A. phagocytophilum and A. platys antigens have been described in dogs [43–45]. Thus, seropositivity detected with the commercial A. phagocytophilum-IFAT in the cats in this study may be attributed to infection with either of the two or both anaplasmas.
Albania belongs to the countries where zoonotic visceral leishmaniosis is endemic, and the disease is of great relevance in humans [46–48]. Competent vectors of L. infantum, the causative agent of human and canine leishmaniosis in Albania , are abundant in the country , and their occurrence was also documented for the district and city of Tirana . Sero-surveys conducted in Albania indicated average anti-Leishmania antibody prevalences of 4–5.1% in dogs of different categories and geographic origin  (Hamel et al., unpublished). In contrast to the situation in dogs, feline leishmaniosis occurs sporadically and usually only in regions where canine leishmaniosis is endemic. The recently proven infectiousness of cats to sand-flies re-emphasized the question of the role of felines in the epidemiology of leishmaniosis which is still not clear [52, 53]. With only 1 out of 146 feline sera from Tirana testing as Leishmania-positive, the percentage of exposure of cats to the parasite in the present study was at the lower end of the wide range of findings reported from other countries in the south of mainland Europe [54–58] and is much lower than the 5.1% seroprevalence recorded in dogs from Tirana (Hamel et al., unpublished). For the Balkans, information on feline leishmaniosis is available apparently through two recently conducted sero-surveys in Greece only. For both studies, serum samples were collected from stray cats in the Thessaloniki area, but the seroprevalences determined differed substantially with 21.6%  or 3.9% , respectively.
Canine D. immitis is endemic in Albania [3–5, 7], and several of its incriminated vectors are common representatives of the culicid fauna in the country . However, microfilariae were not detected in any blood smears or by using the Knott’s test neither was heartworm antigen detected in any serum sample of the cats from Tirana. Although canine heartworm infection was reported with increasing frequency from most of the countries of the Balkans in the recent past [7, 62], the authors are not aware of any survey on feline dirofilariosis in the region. Similar to leishmaniosis, feline dirofilariosis is much less frequently diagnosed compared to canine dirofilariosis in a given region and, based on data collected in endemic areas, the prevalence of heartworm infection in cats can be expected to be 5% to 10% of that seen in dogs .
As the main reservoir for B. henselae, the cat is a host which is often associated with Bartonella infections and with their transmission to man, while fleas have been proven to serve as vectors for transmission of the agents among cats [64, 65]. Human disease associated to infection with Bartonella species as well as seropositivity to Bartonella spp. has been described from some countries in the Balkans [66, 67] but yet not from Albania. The prevalence of Bartonella infection in cats in the Balkan region is largely unknown. To the knowledge of the authors, there is only a small-scale survey from Serbia, which found that 57% of 40 cats had anti-B. henselae antibodies . Bartonella henselae and B. clarridgeiae were demonstrated in C. felis fleas collected from seven cats from Tirana  including five cats studied in this survey. The single flea collected from the four months old, female kitten whose blood tested positive for B. henselae in the present study, however, did not harbour Bartonella- DNA . In contrast, the positive fleas reported by Silaghi et al. (2012) were collected from cats negative for Bartonella spp. . Thus, further studies are needed to understand the involvement of cats in the epidemiology of Bartonella infections in the Balkans.
Four haemotropic mycoplasmas have been recognized in cats: M. haemofelis, Candidatus M. haemominutum, Candidatus M. turicenis and Candidatus M. haematoparvum-like. These mycoplasmas differ in their pathogenicity, with M. haemofelis being the most virulent one that may cause severe haemolytic anaemia. The others may cause changes in blood parameters, but rarely cause diseases without concurrent infections; for the latter one no data is available. Subclinical infections with cats acting as carriers with subclinical infection are common. Feline haemotropic mycoplasma infections are believed to have a worldwide distribution with variable prevalence in the different populations of cats [69, 70]. For Albania, this is the first study to document the occurrence and prevalence of three haemotropic mycoplasmas in cats, and it is the only other study on feline haemotropic mycoplasma infection from the Balkans beside one report from Greece . The overall proportion of haemotropic mycoplasma infection of almost 31% as well as the prevalence of the individual mycoplasmas, including the percentage of co-infections, were within the range of results obtained in other studies in Europe [71–75]. The highest overall prevalence of haemotropic mycoplasmas in Europe (43.4%) was found in a recently reported study from Portugal, where all four feline haemotropic mycoplasmas were identified, with all four mycoplasmas co-infecting one cat . In line with some other studies, male sex [69, 71, 72] and older age [16, 71, 77, 78] were found significantly associated with the detection of haemotropic mycoplasma DNA in the cats from Tirana. Contradictory to Gentilini et al. (2009), who suggested an association of summer season and higher prevalence of feline haemotropic mycoplasma infection because of a higher ectoparasite load which facilitates vectorial transmission of infections, own data and results of studies from Germany and Portugal [72, 76, 79] do not indicate a positive correlation of the presence of blood-feeding arthropods and feline haemoplasmosis. Furthermore, analysis for mycoplasma DNA of all fleas collected from the cats in this study revealed only DNA of Candidatus M. haemominutum in two single C. felis fleas from two cats (unpublished data). One of those cats was a kitten that tested positive for B. henselae but was negative for haemotropic mycoplasma infection and had one flea only; the other cat was infected with Candidatus M. haemominutum and had three fleas. Both, the lack of correlation of the presence of blood-feeding arthropods and detection of DNA of haemotropic mycoplasmas and the marked discrepancy between the detection percentages of haemotropic mycoplasmas in the cats and in their fleas support hypotheses of the existence of routes of transmission of these bacterial organisms other than fleas acting as vectors. Aggressive interaction, such as cat bites, has been suggested as the possible route of transmission and this is supported by the higher prevalence in older and male cats seen in the present study .