Observations on dogs found naturally infected with Hepatozoon canis in Italy

Transcription

1 CAS CLINIQUE Observations on dogs found naturally infected with Hepatozoon canis in Italy A. GAVAZZA, M. BIZZETI and R. PAPINI Dipartimento di Clinica Veterinaria, Viale delle Piagge 2, Pisa, Italy Corresponding author : Dr Roberto Papini, Dipartimento di Clinica Veterinaria, Facoltà di Medicina Veterinaria, Viale delle Piagge, Pisa, Italia Phone +39/ ; Fax +39/050/542892; rpapini@vet.unipi.it SUMMARY Clinical, haematological and laboratory findings recorded in 16 dogs found naturally infected with H. canis in Italy were reported. Clinical signs included depression, anorexia/dysorexia and weight loss (100 % each), slight and serious lymphadenomegaly (37.5 % and 25 %, respectively), fever (25 %), pale mucous membranes and pain on palpation of muscles (6.25 % each). Parasitaemia values ranged from < 1 % to 70 %. Haematological abnormalities were eosinophilia (87.5 %), leukocytosis (43.75 %), neutrophilia (31.25 %), decreased values of HCT (31.25 %), HGB and MCV (25 % each), monocytosis (25 %), lymphocytosis (18.75 %), thrombocytopenia and lymphopenia (6.25 % each), as well as normochromic anaemia, which was mild and microcytic (25 %) or serious, non-regenerative and normocytic (6.25 %). Serum biochemical analyses demonstrated increased CPK (25 %) and total protein (12.5 %), or low A/G ratio (37.5 %) and hypoalbuminaemia (12.5 %). Serum protein electrophoresis revealed β 2 - (100 %) and β 1 -hyperglobulinaemia (68.75 %), α 1 (68.75 %) and α 2 -hypoglobulinaemia (25 %), and γ-hyperglobulinaemia (37.5 %). Concurrent pathologies were E. canis or L. infantum (12.5 % each). KEY-WORDS : Hepatozoon canis - clinical signs - haematological findings - biochemical findings - Italy - epidemiology. RÉSUMÉ Observations réalisées chez des chiens retrouvés naturellement infectés par Hepatozoon canis en Italie. Par A. GAVAZZA, M. BIZZETI et R. PAPINI. Les données cliniques, hématologiques et biochimiques collectées chez des chiens naturellement infectés par H. canis en Italie sont rapportées. Les symptômes cliniques étaient : abattement, anorexie/dysorexie, amaigrissement (100 %), adénomégalie modérée ou sévère (37.5 % et 25 %, respectivement), fièvre (25 %), pâleur des muqueuses et douleur à la palpation des membres (6.25 %). Le niveau de parasitémie variait de < 1 % jusqu à 70 %. Les résultats hématologiques ont montré une éosinophilie (87.5 %), une leucocytose (43.75 %), une neutrophilie (31.25 %), une diminution des valeurs de l hématocrite (HCT) (31.25 %), de l hémoglobinémie (HGB) et du Volume Corpusculaire Moyen (MCV) (25 %), une monocytose (25 %), une lymphocytose (18.75 %), une thrombocytopénie et lymphopénie (6.25 %), ainsi qu une anémie normochrome et microcytaire modérée (25 %) ou arégénérative et normocytaire sérieuse (6.25 %). Les examens biochimiques ont révelé l augmentation de la créatine phospho-kinase (CPK) (25 %) et des protéines totales (12.5 %), une baisse du rapport albumine/globulines (37.5 %) et une hypoalbuminémie (12.5 %). L électrophorèse des protéines sériques a montré une augmentation des fractions β 2 (100 %), β 1 (68.75 %) et γ (37.5 %), et une diminution des fractions α 1 (68.75 %) et α 2 (25 %). Des infections simultanées par E. canis ou L. infantum (12.5 % chacun) ont été mises en évidence. MOTS-CLÉS : Hepatozoon canis - symptômes cliniques - données hématologiques, données biochimiques - Italie - épidemiologie. Introduction Canine hepatozoonosis is a tick-borne infection and was originally thought to be caused by a single Hepatozoon species but at least two are now recognized, belonging to the family Hepatozoïdae (phylum Apicomplexa) : Hepatozoon americanum and Hepatozoon canis. On the basis of phylogenetic analysis, geographic distribution, vectors and pathogenesis, these two etiological agents have been shown to be different [2]. The more pathogenic H. americanum is the cause of canine hepatozoonosis throughout the southern gulf states of the USA and is transmitted by Amblyomma maculatum ticks. This pathogen may cause a serious clinical syndrome in dogs characterised by fever, lethargy, weight loss, stiffness, signs of pain, paralysis, and ocular discharge [32]. Diagnosis of American hepatozoonosis is made by identifying organisms in muscle biopsies [27]. This clinical picture is in marked contrast to that caused by a morphologically identical parasite, H. canis, that is the causative agent of canine hepatozoonosis recognised in most other parts of the world where

2 566 GAVAZZA (A.) AND COLLABORATORS its vector is enzootic, and has been reported in Mediterranean countries such as Italy, France, Greece, Spain, Portugal and Israel, as well as in South American, Asian and African countries [1, 4, 6, 8, 9, 11-13, 17, 18, 20, 22, 24, 25, 28-31]. Moreover, H. canis has been recorded in Germany [14, 16] and Northwest Europe [19] as an imported infection, and its distribution may eventually extend into other regions because of the widespread diffusion of the invertebrate hosts. The main vector is the brown dog tick, Rhipicephalus sanguineus, although the infection may be also transmitted by other tick species [26]. Pathogenesis of H. canis is thought to be weak, because subclinical infections are common, usually causing a mild disease that affects the spleen, lymph nodes, and bone marrow, resulting in anaemia and lethargy. The disease is easily diagnosed by observation of the blood smear, and gametocytes in circulating white blood cells are often an incidental finding in dogs without clinical signs [1, 3, 11]. Protozoan parasites of the genus Hepatozoon have a coccidial-type life cycle with schizogony and gametogony occurring in vertebrate hosts, while sexual stages and sporulation occur in invertebrate hosts. Briefly, infections in dogs are due to ingestion of ticks containing sporulated oocysts. After ingestion, sporozoites are released in the dog s intestinal tract, penetrate the gut wall and are carried by blood or lymph stream to the liver, spleen, lymph nodes, bone marrow or muscle, where schizogony occurs. Some merozoites enter neutrophils or monocytes and develop into gametocytes. When a tick ingests a blood meal from an infected dog, the gametocytes are released within the gut of the tick. Subsequently, oocysts containing sporozoites are formed in the tick body cavity, and the cycle is continued [9]. H. canis can be transmitted transplacentally too [24]. There are also emerging diseases caused by parasitic agents that have been introduced from another geographic region locality, transcending national boundaries in some cases, to give rise to completely new diseases for that region. In addition, the geographical range of some arthropod vectors is expanding due to a variety of climatic and ecological factors. This facilitates the transmission of various clinically important vector-borne pathogens that were previously absent from specific region. Therefore, the ranges of vectorborne infections such as hepatozoonosis and others are extending and enhanced by animals with subclinical infections returning home from endemic areas. This is of particular concern to clinicians as the diseases caused by these organisms will be unfamiliar to them and may go undiagnosed, misdiagnosed or inappropriately treated, due to either the absence of clinical suspicion or because of their subclinical nature [21]. In order to provide further insight on canine hepatozoonosis, the purpose of the present study is to describe clinical, haematological and laboratory findings in a group of dogs found naturally infected with H. canis in Italy. Materials and methods Our study included 16 naturally occurring cases of canine hepatozoonosis referred to our teaching veterinary clinic (Dipartimento di Clinica Veterinaria, University of Pisa, Italy) on June A detailed description, clinical history and physical examination were recorded for each dog as shown in Table I. All of them were sexually intact and were pure breed hunting dogs : 8 Griffons Vendéen (GV), 5 Roughhaired Italian Hound (RIH), and 3 Shorthaired Italian Hound (SIH). There were 13 males and 3 females, Legend : SIH = Shorthaired Italian Hound ; GV = Griffon Vendéen; RIH = Roughhaired Italian Hound ; M = male ; F = female ; yr = year ; mons = months. TABLE I. Distribution of breed, sex, age, clinical signs recorded at history and physical examination, parasitaemia levels and associated pathologies in 16 dogs with Hepatozoon canis infections.

3 OBSERVATIONS ON DOGS FOUND NATURALLY INFECTED WITH HEPATOZOON CANIS IN ITALY 567 aged between 9 months and 7 years. All of the animals were of known provenance. They were allowed outdoors, living in a kennel located in a rural area of the Tuscany region (central Italy) near hills, and were part of a cohort of 39 dogs used for wild-boar hunting. They were referred to the hospital because of depression, anorexia/dysorexia and weight loss (Table I). The duration of the clinical signs at the time of presentation ranged from one to two weeks. Dogs had a history of recurrent tick infestation and recent treatments against ectoparasites. Vaccinations were regularly administered, first at approximately 2 months of age, against canine distemper, canine viral hepatitis, canine parvovirus and leptospirosis. Dogs were last vaccinated approximately 1 year before. Blood samples were collected into K 3 EDTA-coated blood tubes by cephalic venipuncture from all the examined dogs. Diagnosis of the H. canis infections was made by identifying gametocytes of the parasite within leukocytes in stained buffy coat smears, after centrifugation of whole blood samples at 2,700 rpm for 10 min. Buffy coat and whole blood samples smears were made, air dried, stained with May- Grunwald-Giemsa, and examined by light microscopy at 20x and 100x. Microscopic examinations of buffy coat smears revealed large intracytoplasmatic inclusion bodies inside circulating neutrophils and monocytes. Inclusion bodies appeared as ellipsoid, elongated and pale-staining, measuring about 6-11 microns in length by 3-5 microns in width. These bodies had a variably shaped, central or eccentric nucleus stained dark blue or violet. On the basis of their morphological characteristics, inclusion bodies were identified as H. canis gametocytes. Parasitaemia levels were directly calculated in whole blood smears by counting number of parasitized neutrophils and monocytes out of 200 total leukocytes. Haematological examinations were performed by a standard electronic impedance cell counter (Genius, SEAC, Calenzano, FI, Italy) and included counts of white blood cells (WBC), red blood cells (RBC), haemoglobin (HGB), haematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular haemoglobin concentration (MCHC), platelets (PLT), neutrophils (NS), eosinophils (EOS), lymphocytes (LYMP) and monocytes (MON). When HCT value was < 30 %, whole blood smears were stained with new methylene blue to evaluate regenerative or non-regenerative anaemia. The following serum biochemical assays were determined for each dog by spectrophotometric method (Slim, SEAC, Calenzano, FI, Italy) : glutamic oxoalacetic transaminase (GOT), glutamic pyruvate transaminase (GPT), alkaline phosphatase (ALP), creatine phosphokinase (CPK), total bilirubin (TB), blood urea nitrogen (BUN), cholesterol, conjugated bilirubin (CB), total plasma proteins (TP), and albumin/globulin ratio (A/G). Also, serum protein electrophoresis was performed on cellulose acetate agar (Microtech 648R, Interlab, Roma, Italy) using Ponceau Red dye. Other routine laboratory assays included : urine analysis (Microaution MA4260, Menarini diagnostic, Firenze, Italy), coprological examinations by flotation technique, serological screenings for Leishmania infantum, Ehrlichia canis, Rickettsia rickettsii and Borrelia burgdorferi by immunofluorescent antibody test, direct detection of circulating microfilariae on blood smear and diagnosis of occult dirofilariasis by a commercial kit (Canine Heartworm Antigen, Petchek, Idexx, Maine, USA). Results The clinical signs recorded in the 16 dogs examined are shown in details in Table I. All dogs revealed poor haircoat conditions. Serious lymphadenomegaly was observed in 6 (37.5 %) of the subjects and slight lymphadenomegaly in 4 (25 %). Fever was recorded in 4 (25 %) cases. Only in 1 (6.25 %) case, severe lymphadenomegaly and fever were associated with pale mucous membranes and pain on palpation of muscles of the neck, trunk and legs. Parasitaemia values ranged from < 1 % to 70 % as shown in Table I. Results of haematological examinations are presented in Table II. Leukocytosis was noted in 7 (43.75 %) cases, ranging from mild to severe. Different cell counts revealed eosinophilia in 14 (87.5 %) of the cases, and this was the most common haematological abnormality observed. Neutrophilia was present in 5 (31.25 %) animals, monocytosis in 4 (25 %) and lymphocytosis in 3 (18.75 %), respectively. Also, haemograms showed decreased values of HCT (5/16, %), HGB (4/16, 25 %) and MCV (4/16, 25 %). One (6.25 %) dog had serious normocytic, normochromic, non-regenerative anaemia, and 4 (25 %) showed mild microcytic, normochromic anaemia. Thrombocytopenia or lymphopenia were observed in 1 (6.25 %) case. Results of serum biochemical analyses demonstrated alterations in many parameters, as reported in details in Table III. Six (37.5 %) dogs showed low A/G ratio and 4 (25 %) high CPK values. High concentration of TP and hypoalbuminaemia were recorded in 2 (12.5 %) of the animals, respectively. Electrophoresis revealed β 1 and β 2 -hyperglobulinaemia in 11 (68.75 %) and 16 (100 %) of the dogs, respectively, as well as α 1 and α 2 -hypoglobulinaemia in 11 (68.75 %) and 4 (25 %) of them. Also, γ-hyperglobulinaemia was observed in 6 (37.5 %) of the animals. In 4 (25 %) cases, H. canis infection was associated with other concurrent pathologies. Two (12.5 %) subjects proved to be positive for E. canis or L. infantum antibodies at 1 : 64 and 1 : 40 dilution titres, respectively. Urine analysis were within the reference ranges. Coprological results and the other routine laboratory assays were negative. Discussion Habitat, environmental conditions and the presence of ticks are essential factors in the development of canine hepatozoonosis. All 16 dogs examined were living in the same kennel, which would suggest a common source of infection. No tendency was evident in the breed distribution of dogs. However, the presence of infections in hunting dogs may be attributed to their outdoor lifestyle, which probably enhances their opportunities for contact with ticks. In the present study,

4 568 GAVAZZA (A.) AND COLLABORATORS Legend : WBC = white blood cells ; RBC = red blood cells ; HGB = haemoglobin ; HCT = haematocrit ; MCV = mean corpuscular volume ; MCHC = mean corpuscular haemoglobin concentration ; PLT = platelets ; NS = neutrophils ; EOS = eosinophils ; LYMP = lymphocytes ; MON = monocytes ; Ref. Val. = reference values ; and = above and below reference values, respectively. TABLE II. Haematological parameters in dogs found naturally infected with Hepatozoon canis. Legend : GOT = glutamic oxoalacetis transaminase ; GPT = glutamic pyruvate transaminase ; ALP = alkaline phosphatase ; CPK = creatine phosphokinase ; TB = total bilirubin ; CB = conjugated bilirubin ; BUN = blood urea nitrogen ; Chol =cholesterol ; TP = total plasma proteins ;Alb = albumin ; Glob = globulin ; A/B = albumin/globulin ratio ; Ref. Val. = reference values ; and = above and below reference values, respectively. TABLE III. Serum biochemistry parameters in dogs naturally infected by Hepatozoon canis.

5 OBSERVATIONS ON DOGS FOUND NATURALLY INFECTED WITH HEPATOZOON CANIS IN ITALY 569 male dogs appeared to be more susceptible than female dogs (81.25 % vs %). This is not in agreement with a previous report on 15 cases of American hepatozoonosis involving eight female dogs and seven male dogs [7]. However, the prevalence of males in our study is probably due solely to the preponderance of males in the group of dogs examined. Gametocytes of H. canis were observed in dogs of all ages, since the infection was found in animals from 9 months to 7 years of age. However, they were most prevalent in young dogs approximately 1 year old. This finding is in agreement with those of other authors. EZEOKOLI et al. [13] found that the infection was distributed amongst all ages, although it was most prevalent in young subjects. MUNDIM et al. [23] reported that the prevalence was higher in dogs less than 1 year of age. BANETH and WEIGLER [3] observed that the infection was mostly prevalent in animals under 6 months of age and from 5 to 10 years old. O DWYER et al. [26] indicated that animals of all ages are equally infected. Since hepatozoonosis is a chronic disease and can remain in remission for varying periods of time with intermittent parasitaemia, it is possible that higher frequencies observed in young dogs are due to an acute phase of the disease, when more gametocytes can be detected in the peripheral blood stream. However, it is assumed that dogs of all ages can be infected. The detection of H. canis infections during the month of June is not surprising. Since ticks are more active in the summer and hepatozoonosis is recognized as a primarily tickborne disease in dogs, it can be easily expected to find parasitaemia cases during the warmer months of the year. The parasitaemia levels ranged from < 1 % to 70 % of neutrophils and monocytes. Levels of H. canis parasitaemia previously detected varied from as low as 0.5 % [17] to as high as 60 % of neutrophils [4]. ELIAS and HOMANS [12] reported a variation from 6 to 43 %, whilst BANETH et al. [6] found 1-10 % of the circulating neutrophils to be infected. O DWYER et al. [26] showed that the number of infected neutrophils/monocytes did not exceed 3 %. In American hepatozoonosis, very low parasitaemias have been reported, affecting no more than 0.1 % of leukocytes [32]. In the present investigation, the dog with the highest parasitemia showed the greatest disease severity with fever, lymphadenomegaly, pale mucous membranes, muscle pain, leukocytosis and non-regenerative anaemia. Probably, the high percentage of parasitized neutrophils was a manifestation of the widespread dispersion of H. canis schizonts in tissues of the dog. BANETH and WEIGLER [3] suggested a relationship between severity of clinical signs and parasitaemia levels. Also, it has been suggested that immune mechanisms may play an important role in the control of H. canis parasitism : immunosuppression induced by administration of corticosteroids has been reported to induce rapid detection of parasitaemia and rapid appearance of gamonts in treated dogs [5]. However, pathogenicity and relationship between parasitaemia and disease severity need to be further evaluated. Concurrent infections with H. canis and E. canis or L. infantum have been reported [8, 10] and may be explained by several mechanisms. R. sanguineus is the common vector of H. canis and E. canis, and may allow for simultaneous exposure to both the organisms. H. canis may increase the susceptibility to or promote the clinical manifestation of erlichiosis. Infection of a single monocyte with one gametocyte of H. canis and a morula of E. canis has also been described [8]. Immune suppression caused by severe clinical diseases may predispose to H. canis infection or allow expression of a subclinical infection. Therefore, hepatozoonosis can be considered as an opportunistic infection indicative of an immunodeficiency status caused by leishmaniasis. However, it is impossible to exclude that H. canis can predispose dogs to infection with L. infantum. Almost all the animals showed a weak form of the disease and the common clinical signs in these cases were depression, anorexia/dysorexia, weight loss and lymphadenomegaly. A more serious form was observed in 4 dogs and was characterized by the association of these clinical signs with fever, and with pale mucous membranes and pain on palpation of muscles of the neck, trunk and legs in one of them. Clinical signs appear to be due to the release of merozoites from schizonts in tissues. Canine hepatozoonosis due to H. canis is generally not associated with clinical signs. Some investigators considered H. canis to be non-pathogenic and, when signs of disease were evident in infected dogs, they were attributed to other causes such as dirofilariasis, distemper, generalized demodicosis, or leishmaniasis. For example, in 8 naturally occurring cases in Brazil, all the dogs had concurrent diseases which could be implicated in the severity of the pathology observed [17]. In our study, 4 out of 16 dogs with confirmed hepatozoonosis had other concurrent infections (2 had L. infantum and 2 had E. canis infections). Some of the clinical signs and laboratory abnormalities may be attributed to coexistent diseases, and thus the contribution of H. canis to the clinical disease in these 4 dogs cannot be determined. However, in 12 dogs clinical signs were associated with primary H. canis infection. None of the signs were attributable to coexisting disease states. Also, the remarkable similarity of clinical signs and laboratory findings led us to conclude that the changes encountered were directly related to H. canis infection. Therefore, there is no doubt that H. canis can be a primary pathogen in dogs, causing depression, anorexia/dysorexia and weight loss, as well as fever and lymphadenomegaly, or, in the most serious cases, pale mucous membranes and muscular pain. The pathology of the dogs in the present report was clinically similar in many respects to worldwide reports of canine hepatozoonosis. ELIAS and HOMANS [12] observed fever, anaemia, anorexia and weakness of the hind legs in six H. canis infected dogs from different parts of southern Israel. In one report from Senegal, clinical signs included weight loss, dermatitis, pruritus and mild lymphadenomegaly [31]. In a study from Nigeria, fever, anorexia, loss of weight, depression, limping, abnormal gait, oculonasal discharge and conjunctivitis were recorded [20]. Clinical cases with lethargy, weight loss, obvious signs of pain, or periosteal newborn proliferation, similarly as for American type of hepatozoonosis, were also reported in Japan [25]. In different surveys from Italy, most of the dogs showed various clinical signs including weight loss, anorexia/dysorexia, hyperthermia alternate to hypothermia, icterus, lymphadenomegaly, abdominal pain on palpation, sple-

6 570 GAVAZZA (A.) AND COLLABORATORS nomegaly, and diffuse muscular tenderness [1, 30]. Abortion was also recorded [28]. Otherwise, H. canis was occasionally found in asymptomatic dogs [1, 11]. Unfortunately, clinical signs of canine hepatozoonosis are rather unspecific and can be associated with other clinical diseases which frequently occur in dogs. The exact factors governing the release of merozoites from schizonts and appearance of clinical signs are unknown at this time. For this reason H. canis infection may be easily underestimated during clinical investigation and thus additional diagnostic investigations should be performed to rule out the presence of H. canis. In most of the cases, eosinophil counts were greatly increased. It is possible that blood eosinophils responded to tick infestations and developed a response of great magnitude following repeated and recent tick feeding. High prevalence of eosinophilia was not described in previous studies on canine hepatozoonosis. The reason for this difference still remains not completely clear. Instead, anaemia, leukocytosis, neutrophilia, monocytosis, lymphopenia, thrombocytopenia, and decreased values of HGB and HCT have been previously reported [3, 17]. However, in the present study, incidence and severity of these haematological findings were lower than those reported by other authors [3, 17]. Also, in contrast to other reports [12], neutropenia was not found in any of the dogs. Reports of serum chemistry results in canine hepatozoonosis are fragmentary. Our results showed that serum biochemical values had a similar trend in all the cases and included alterations in many parameters. Serum protein electrophoresis showed increase in the β- and decrease in the α-globulin range. Beta 2 -hyperglobulinaemia was a constant laboratory finding and was almost always associated with β 1 -hyperglobulinaemia and α 1 -hypoglobulinaemia. Alpha 2 -hypoglobulinaemia and γ-hyperglobulinaemia were less frequently detected. It is likely that the involvement of these proteins may represent a response to the persistent inflammatory changes associated with tissue stages of the parasite. Also, the increase of β- and in some cases of γ- fractions may be due to polyclonal lymphatic activation. Hyperglobulinaemia with polyclonal gammopathy has been previously reported in two highly parasitemic dogs from Israel and in one case of American hepatozoonosis, and can probably be considered as a symptom of a chronic state of the disease [4, 7]. Other biochemical abnormalities in serum included an increase in the values for CPK and TP, as well as a decrease in the values for A/G ratio and, more rarely, in albumin concentration. Low A/G ratio could be related to the β- and/or γ-hyperglobulinaemia, and, to a lesser extent, to concurrent hypoalbuminaemia. Elevation of CPK was probably due to muscular damage which can be associated with hepatozoonosis, as shown by other authors [4, 15]. With the increase in CPK values occurring in four of the cases, one might expect evident and diffuse pain on palpation of muscles. However, this clinical sign was demonstrated in only one of the cases in which elevated CPK was detected. This discrepancy between CPK levels and muscular tenderness has been previously recorded. In a retrospective study, BARTON et al. [7] observed that CPK resulted to be elevated in only 1 of 15 dogs found naturally suffering from American hepatozoonosis. For contrast, pain on palpation of muscles, limb and spine was present in 14 of them. Elevation of TP was probably related to the increase of β- and/or γ-globulin fractions. High TP have been observed in different cases of canine hepatozoonosis [4, 7]. Hypoalbuminaemia was probably due to anorexia and chronic inflammatory disease. Low concentrations of albumin have been previously reported in canine hepatozoonosis [4, 15]. The typical decrease in albumin is attributed to decreased protein intake, prolonged anorexia, chronic inflammation, or renal loss and proteinuria caused by glomerulonephritis. Mild to severe proteinuria (indicating possible glomerulonephritis) was found in 8 of 11 dogs in a study from Greece [22] ; moreover, renal involvement has been recorded in American hepatozoonosis [7] and in reports from Israel and Spain [4, 18]. In the present report, however, none of the dogs examined had demonstrable proteinuria, and kidney involvement was not recorded, but we cannot exclude that dogs could develop renal involvement later in the course of the infection. Consistent increases in ALP activity and low BUN values have been observed in various cases of canine hepatozoonosis [5, 7, 17]. However, in contrast to other reports, these findings were not detected in the present study. Canine illnesses caused by tick-borne pathogens may be considered among the most significant groups of emerging diseases worldwide. Haemoparasites deserve particular attention since they are an important cause of disease in dogs as many have long incubation periods or may exist in a subclinical state. Hepatozoonosis is of particular concern among these infections. In fact, hepatozoonosis is a tick-borne disease of increasing importance in dogs in regions that have previously been considered free from this infection [14, 16, 19]. Like many emerging diseases in dogs caused by protozoa, hepatozoonosis has only recently attracted more intensive research interest. Clinical signs and laboratory findings in canine hepatozoonosis are unspecific and similar to those observed in other diseases commonly found in dogs. The present study suggests that hepatozoonosis must be added to the differential diagnosis and adequate haematological investigations must be performed by practising veterinarians, mostly when the presence of ticks or a history of tick infection is recorded. Because of the widespread distribution of R. sanguineus and the easy movement of dogs from one country to another, there is a great possibility for the disease to spread into new geographic areas. For example, H. canis has been recorded in Germany as an imported infection [14, 16] and it is also regularly introduced into Northwestern Europe by dogs bitten by infected R. sanguineus and Dermacentor spp. during visits to endemic regions [19].

7 OBSERVATIONS ON DOGS FOUND NATURALLY INFECTED WITH HEPATOZOON CANIS IN ITALY 571 Conclusion Infections with H. americanum have been well described and are characterised by fever, lethargy, weight loss, stiffness, signs of pain, paralysis, ocular discharge [27, 32]. Reports of infection with H. canis have described non-specific clinical signs such as depression, anorexia/dysorexia and weight loss. Attempts to investigate the pathogenicity and clinical manifestations of H. canis have been complicated by finding infections in apparently healthy dogs [1, 3, 11] and by the common occurrence of mixed infections with H. canis and other pathogens [8, 10]. However, it has been shown that mild cases of infection can develop into severe infections. We believe that our results could be useful both for monitoring the clinical development and for improving the epidemiological approach to the disease. It is advisable that all dogs with H. canis parasitaemia are treated and that treatments are followed by repeated blood counts and thorough examination of blood smear, until no gametocytes are found in 2 or 3 consecutive samples taken 2 weeks apart. A combination of tetracycline hydrochloride (22 mg/kg, three times daily) and imidocarb dipropionate (6 mg/kg, by subcutaneous injection, repeated after 14 days) can successfully clear gamectocytes from the blood and resolve clinical signs in dogs with hepatozoonosis [3, 12]. References 1. ARRU E., PAU S. and LEONI A. : Segnalazione di Hepatozoon canis in Sardegna. Riv. Parassitol., 1982, 43, BANETH G., BARTA J.R., SHKAP V., MARTIN D.S., MACIN- TIRE D.K. and VINCENT-JOHNSON N. : Genetic and antigenic evidence supports the separation of Hepatozoon canis and Hepatozoon americanum at the species level. J. Clin. Microbiol., 2000, 38, BANETH G. and WEIGLER B. : Retrospective case-control study of hepatozoonosis in dogs in Israel. J. Vet. Int. Med., 1997, 11, BANETH G., HARMELIN A. and PRESENTEZ B.Z. : Hepatozoon canis in two dogs. J. Am. Vet. Med. Assoc., 1995, 206, BANETH G., SAMISH M., ALEKSEEV E., AROCH I. and SHKAP V. : Transmission of Hepatozoon canis to dogs by naturally-fed or percutaneously-injected Rhipicephalus sanguineus ticks. J. Parasitol., 2001, 87, BANETH G., SHKAP V., PRESENTEY B.Z. and PIPANO E. : Hepatozoon canis : the prevalence of antibodies and gametocytes in dogs in Israel. Vet. Res. Commun., 1996, 20, BARTON C.L., RUSSO E.A., CRAIG T.M. and GREEN R.V. : Canine hepatozoonosis : A retrospective study of 15 naturally occurring cases. J. Am. Vet. Med. Assoc., 1985, 21, BEAUFILS J.P. and LEGROUX J.P. : Présence simultanée d Ehrlichia sp. et d Hepatozoon canis dans des granulocytes de chien : A propos de 2 cas. Prat. Méd. Chir. Anim. Comp., 1992, 27, BEAUFILS J.P. and MARTIN-GRAVEL J. : L hépatozoonose canine. Première partie : étude bibliographique. Prat. Méd. Chir. An. Comp., 1988, 23, CIARAMELLA P., OLIVA G., DE LUNA R., GRADONI L., AMBROSIO R., CORTESE L., SCALONE A. and PERSECHINO A. : A retrospective clinical study of canine leishmaniasis in 150 dogs naturally infected by Leishmania infantum. Vet. Rec., 1997, 141, DOMINA F. and CATARSINI O. : Hepatozoon canis : rilievo occasionale di 9 casi. Atti Soc. It. Sc. Vet., 1982, 36, ELIAS E. and HOMANS P.A. : Hepatozoon canis infection in dogs : clinical and haematological findings ; treatment. J. Small Anim. Pract., 1988, 29, EZEOKOLI C.D., OGUNKOYA A.B., ABDULLAHI R., TEKDEK L.B., SANNUSI A. and ILEMOBADE A.A. : Clinical and epidemiological studies on canine hepatozoonosis in Zaria, Nigeria. J. Small Anim. Pract., 1983, 24, FISCHER S., HARTMANN K. and GOTHE R. : Hepatozoon canis : eine importierte parasitare Infektion bei Hunden. Tier. Prax., 1994, 22, GAUNT P.S., GAUNT S.D. and CRAIG T.M. : Extreme neutrophilic leukocytosis in a dog with hepatozoonosis. J. Am. Vet. Med. Assoc., 1983, 182, GLASER B. and GOTHE R. : Importierte arthropodenübertragene Parasiten und parasitische Arthropoden beim hund. Erregerspektrum und epidemiologische Analyse der 1995/96 diagnostizierte Fälle. Tier. Prax., 1998, 26, GONDIM L.F.P., KONYAGAWA A., ALENCAR N.X., BIONDO A.W., TAKAHIRA R.F. and FRANCO S.R.V. : Canine hepatozoonosis in Brazil : description of eight naturally occurring cases. Vet. Parasitol., 1998, 74, HERVÁS J.L., CARRASCO J.C., GOMEZ-VILLAMANDOS J.C., MENDEZ A. and SIERRA M.A. : Acute fatal hepatozoonosis in a puppy : histopathological and ultrastructural study. Vet. Rec., 1995, 137, HOLLAND M. : Emerging diseases in Northern Europe. J. Small Anim. Pract., 2001, 42, IBRAHIM N.D.G., RAHAMATHULLA P.M. and NJOKU C.O. : Neutrophil myeloperoxidase deficiency associated with canine hepatozoonosis. Int. J. Parasitol., 1989, 19, IRWIN P.J. : Companion animal parasitology : a clinical perspective. Int. J. Parasitol., 2002, 32, KONTOS V.J. and KOUTINAS A.F. : [Canine hepatozoonosis : a review of 11 naturally occurring cases]. Bull. Hell. Vet. Med. Assoc., 1990, 41, MUNDIM A.V., MUNDIM M.J.S., JENSEN N.M.P. and ARAÜJO S.F. : Hepatozoon canis : estudo retrospectivo de 22 casos de in fecçao natural em cães de Uberlândia. MG Rev. Cent. Cienc. Biomed. Univ. Fed. Uberlândia, 1994, 10, MURATA T., INOUE M., TATEYAMA S., TAURA Y. and NAKAMA S. : Vertical transmission of Hepatozoon canis in dogs. J. Vet. Med. Sc., 1993, 55, MURATA T., SHIRAMIZU K., HARA Y., INOUE M., SHIMODA K. and NAKAMA S. : First case of Hepatozoon canis infection of a dog in Japan. J. Vet. Med. Sci., 1991, 53, O DWYER L.H., MASSARD C.L. and PEREIRA DE SOUZA J.C. : Hepatozoon canis infection associated with dog ticks of rural areas of Rio de Janeiro State, Brazil. Vet. Parasitol., 2001, 94, PANCIERA R.J., EWING S.A., MATHEW J.S., LEHENBAUER T.W., CUMMINGS C.A. and WOODS J.P. : Canine hepatozoonosis : comparison of lesions and parasites in skeletal muscle of dogs experimentally and naturally infected with Hepatozoon americanum. Vet. Parasitol., 1999, 82, PAU S. and ZEDDA M.T. : Osservazioni sulla presenza di Hepatozoon canis a livello placentare. Atti Soc. It. Sc. Vet., 1983, 37, REBELO M.E. : Hepatozoon canis (James, 1905) em canideos domesticos na zona do Alto-Douro, concelho de Alijo. Rep. Trab. LNIV, 1989, 21, ROSATI S. : Segnalazione di Hepatozoon canis nella provincia di Imperia. Ann. Fac. Med. Vet. Torino, 1986, 31, VERCRUYSSE J. and PARENT R. : Note sur deux cas d hépatozoonose à Dakar. Rev. Med. Vét., 1982, 133, VINCENT-JOHNSON N., MACINTIRE D.K. and BANETH G. : Canine hepatozoonosis : pathophysiology, diagnosis, and treatment. Comp. Cont. Educ. Pract. Vet., 1997, 19,