Although hepatozoonosis of domestic cats was initially reported in 1908, the same year when the type species Hepatozoon muris was described from a laboratory rat and its life cycle described , it has almost been overlooked since and little has been published on its pathogenesis. This study aimed to integrate new histopathologic, hematologic, clinical, epidemiological and genetic findings on feline hepatozoonosis and promote the understanding of this infection. The results clearly indicate that feline infection is primarily caused by a morphologically and genetically distinct species, H. felis, which has predilection to infecting muscular tissues, and is highly prevalent in the cat population studied. The lack of previous comprehensively integrated data merits the redescription of this parasite elucidating its parasitological characteristics.
Surveys of Hepatozoon sp. infection in domestic cats describe variable rates of infection in different areas. A study of myocardium specimens from 100 cats brought to necropsy in Israel found that 36% of cats harbored cardiac Hepatozoon-like meronts . Interestingly, this was the same rate of infection found in Israel almost 40 years later in the current study, indicating that feline hepatozoonosis is not a new emerging infection in this country. Studies using PCR detection from Spain have shown diverse prevalence rates with 0.6% in one study , 16% in a cat colony from Barcelona , and 4% in cats from the Barcelona area . A comparative study carried out in several districts of Bangkok, Thailand, where both canine and feline hepatozoonosis are prevalent, has reported a high infection rate of 32% in 300 cats by PCR. A positive association was found between the rates of infected dogs and cats in the same districts and 18S rDNA sequences from cats and dogs were closest to H. canis. These findings encouraged the authors to hypothesize that H. canis was the cause of both canine and feline infections. However, the genetic analyses in this study were probably made before the sequences designated as H. felis were deposited in GenBank . Therefore, the identity of some feline sequences may have been misinterpreted. A study from Brazil evaluated 200 blood samples from Sao Luis in Brazil and found only one cat infected with a Hepatozoon sp. which clustered with H. felis on a phylogenetic analysis .
The lack of association between age and infection found in the study supports the possibility of transplacental transmission and kittens being born already infected. It may also suggest intensive exposure at a young age. The highly significant association found between infection and access to outdoors suggests the possibility of transmission by arthropod vectors, such as fleas, ticks or mites, which are common ectoparasites of cats globally, or by carnivorism as described for a number of Hepatozoon spp. [2–5]. So far, no arthropod vector has been described for H. felis. Other Hepatozoon spp. have been demonstrated to be transmitted by fleas, ticks, mites, lice, mosquitoes and sandflies  and it is therefore expected that H. felis would also be transmitted by an hematophagous arthropod. The existence of more than one route of transmission for H. felis is also optional as other Hepatozoon spp. have been shown to be transmitted both by arthropod vectors and by additional routes, for instance H. canis is transmitted by the tick Rhipicephalus sanguineus as well as transplacentally [3, 29], and H. americanum is transmitted by the tick Amblyomma maculatum and by carnivorism [5, 30].
The level of parasitaemia is usually low in feline hepatozoonosis with less than 1% of the neutrophils and monocytes containing gamonts . A survey from Thailand found that parasitemia was detected by light microscopy of blood smears in only 0.7% of 300 cats, while 32% were positive by sensitive PCR . Furthermore, none of the blood smears from 100 Israeli cats of which 36% were positive for cardiac Hepatozoon sp. meronts, had evident gamonts . This study also found a higher infection rate in the myocardium of apparently healthy cats compared to sick cats . Feline hepatozoonosis seems to be mostly sub-clinical as a high proportion of the population is infected with no apparent overt clinical manifestations. The seven cats described in a case series of feline hepatozoonosis suffered from various other infections including FIV, FeLV and hemotropic mycoplasmosis . Although no significant association was found in the present study between FIV infection and Hepatozoon infection, this could be due to the small number of samples included, as proportionally, more FIV+ cats were infected with Hepatozoon than FIV negative cats. FIV or FeLV infections have also been described in conjunction with feline hepatozoonosis in other studies [11, 14]. It is therefore probable that Hepatozoon infection may escape control by the immune-system in immune-suppressed cats, allowing the intensification of parasitemia and increasing the likelihood of detection by blood smear microscopy [10, 11]. The reason for not testing the cats in the study for FeLV infection is the relatively low prevalence of this infection in local Israeli cats .
Feline hepatozoonosis is associated mostly with infection of muscle tissues. Hepatozoon sp. meronts have been reported in the muscles of domestic cats with hepatozoonosis [9, 11], and elevated activities of the muscle enzyme creatine kinase were found in the majority of cats with hepatozoonosis in a retrospective study of this infection . The genetic and morphologic findings of this study clearly showed that it is H. felis which infects myocardial and skeletal cat muscles, and not another Hepatozoon sp. However, other tissues such as the lungs were also infected with meronts, and PCR detected the presence of the parasite’s DNA in hemolymphoid organs such as the spleen, bone marrow and lymph nodes, and in the liver, pancreas, and kidney, as well as in fetal lungs and amniotic fluid. No substantial inflammatory response surrounding meronts was seen in the muscles examined in this study, as well in the other reports [9, 11]. This is in agreement with the generally sub-clinical nature of this infection. Infection of muscle tissues by Hepatozoon sp. has also been reported in wildlife felids and carnivores where H. felis or closely related species are responsible for myositis and myocarditis [32–35]. Hepatozoon americanum infection of dogs and wildlife in the USA also has a predilection to muscle tissue, like H. felis, differing substantially from H. canis in dogs, which is found mostly in the hemolymphoid tissues and does not directly infect muscle. However, H. americanum induces the formation of muscle cysts which are much larger than H. felis meronts, are composed of concentric layers of muco-polysaccharide material surrounding a core zoite in a formation described as “onion skin” cysts, and elicit severe and painful pyogranulomatous myositis following merogony [24, 36, 37].
The phylogenetic placement of 18S rRNA Hepatozoon sequences amplified from domestic cats in this study revealed that both H. felis and H. canis infect cats in Israel, although H. felis is by far more common. Israeli H. felis sequences are indistinguishable from those reported from Spain and Brazil, and closely related to those reported in wildlife felids and carnivores from India, Korea, Japan, Tanzania, Brazil and Argentina [15, 16, 33–35, 38–40]. It is obvious that H. felis clusters away from H. canis and other Hepatozoon species, and is responsible for infection of domestic cats and likely also of other carnivores.
It is most likely that H. felis is the predominant species of Hepatozoon that infects domestic cats and wild felids globally. Its wide geographic distribution could be due to transmission by some ubiquitous vector such as a common flea, mite or tick species, or to highly successful alternative routes of transmission such as transplacental transmission or carnivorism of a yet unknown wildlife intermediate host. Hepatozoon canis has been shown to spread rapidly in a young dog shelter population, and it is possible that H. felis may also spread rapidly under similar conditions .