First case of Anaplasma platys infection in a dog from Croatia
© Dyachenko et al; licensee BioMed Central Ltd. 2012
Received: 13 January 2012
Accepted: 9 March 2012
Published: 9 March 2012
Skip to main content
© Dyachenko et al; licensee BioMed Central Ltd. 2012
Received: 13 January 2012
Accepted: 9 March 2012
Published: 9 March 2012
It is known that Anaplasma (A.) platys, the causative agent of infectious canine cyclic thrombocytopenia, is endemic in countries of the Mediterranean basin. However, few reports are available from the Balkans. This case report describes a dog, which was imported from Croatia to Germany in May 2010. One month later the dog was presented to a local veterinarian in Germany due to intermittent/recurrent diarrhoea. Diagnostic tests were performed to identify infections caused by Anaplasma spp., Ehrlichia spp., Hepatozoon canis, Babesia spp., Leishmania spp., Borrelia burgdorferi and/or Dirofilaria immitis.
Haematological examination of a blood smear revealed basophilic inclusions in thrombocytes, which were confirmed as A. platys with a species-specific real-time PCR. Additionally, an infection with Babesia (B.) vogeli was also detected (PCR and serology). No specific antibodies against Anaplasma antigen were detectable. Although the dog showed no specific clinical signs, thrombocytopenia, anaemia and elevated C-reactive protein (CRP) were observed. Sequencing of a 1,348-bp partial ribosomal RNA gene revealed highest homology to A. platys sequences from Thailand, Japan and France.
A. platys was detected for first time in a dog imported from Croatia. As the dog was also co-infected by B. vogeli, unique serological and haematological findings were recorded. Thrombocytopenia, anaemia and elevated values of C-reactive protein were the laboratory test abnormalities observed in this case. A. platys infections should be considered in dogs coming from Croatia and adjacent regions.
Anaplasma platys (formerly Ehrlichia platys) was first identified and described in 1978 in Florida (USA) as a Rickettsia-like, platelet-specific organism in dogs with infectious canine cyclic thrombocytopenia (ICCT) . Based on morphology and serological cross-reactions with Ehrichia canis, the microorganism was first proposed as E. platys. Sequencing and phylogenic analysis of the 16S rRNA gene and GroESL operon showed that the pathogen was related to A. phagocytophilum and A. marginale, which led to reclassification and designation as A. platys[3, 4].
In dogs A. platys organisms infect peripheral blood platelets and form basophilic inclusions in the cells, so-called morulae, which contain one or more subunits [1, 5]. Both, the appearance of the pathogen in the platelets and the following thrombocytopenia are cyclic . The initial thrombocytopenias may develop primarily as a consequence of direct injury to platelets by replicating organisms. However, immune-mediated mechanisms of thrombocytopenia become more important in subsequent thrombocytopenic episodes . The fraction of infected platelets decreases dramatically in successive parasitaemias, but the associated thrombocytopenic episodes remain severe . In general, the infection is accompanied by unspecific and mild clinical manifestation including anorexia, depression, generalized lymph node enlargement, pale mucous membranes and elevated rectal temperatures [1, 7–9]. Nevertheless, a severe course of A. platys infection with ecchymotic haemorrhagia was reported to be caused by a Greek strain . The pathogen is assumed to be transmitted by Rhipicephalus sanguines, as in several studies A. platys-DNA was detected in this tick species and co-infections in dogs with E. canis and B. vogeli, two pathogens that share the same vector, reinforce this speculation [11–13]. The vector competence of R. sanguineus, however, could not be proven so far . Currently, A. platys has been described in both American continents (USA , Venezuela , Brazil ), Asia (China , Thailand , Taiwan , Japan ), Australia  and Africa . In Europe the occurrence of A. platys has been shown in Mediterranean countries: Italy , France , Spain , Portugal , Turkey  and Greece . Here the first case of a presumed autochthonous A. platys infection is described in a dog from Croatia.
A one-year-old male dog was imported from Croatia to Germany in May 2010 and, according to the owner declaration, has never been outside Croatia before. The dog was presented to a local veterinarian in Germany one month after the import due to intermittent/recurrent diarrhoea. Diagnostic tests for infections uncommon for the German area were requested (CBC with blood smear review, complete serum chemistry analysis as well as a "travel disease profile"). No abnormalities were found during the clinical assessment.
Patient's haematological and serum biochemical parameters at the time of clinical examination
white blood cells
red blood cells
mean corpuscular volume
mean corpuscular haemoglobin concentration
(0 - 0.6)
neutrophil granulocytes (band)
(0 - 0.3)
neutrophil granulocytes (segmented)
serum biochemical parameters
Based on abnormal clinicopathological findings in conjunction with the positive PCR results, specific therapy with doxycycline (10 mg/kg, orally, SID) was initiated for four weeks. After three weeks of Doxycycline therapy, a single injection of imidocarb dipropionate (6 mg/kg) was administered subcutaneously. In the following hours after the application the general health condition of the dog worsened, described as an anaphylactic reaction by the in-clinic veterinarian. Despite the immediate use of infusions and the administration of parenteral atropine, the dog died the next day.
Blood was collected on the day of the presentation in the clinic. CBC with blood smear review was performed on the sample paying particular attention to blood parasites and hemotropic bacteria, as was complete serum chemistry (IDEXX Vet Med Lab). CRP concentrations were measured by means of a validated CRP immunoturbidimetric assay.
Total DNA was extracted from whole blood by using QIAamp DNA Blood Mini kit (QIAGEN, Germany) according to the manufacturer's instructions. Real-time PCR at IDEXX Vet Med Lab was performed using the LightCycler 480 (Roche) with proprietary forward and reverse primers and hydrolysis probes. Target genes for pathogen detection using real-time PCR were as follows: A. platys (groEL), A. phagocytophilum (msp2), B. vogeli, B. canis, B. rossi, B. gibsoni (hsp 70), B. conradae (ITS2), H. canis (18S rRNA), Leishmania spp. (GP63), E. canis, E. chaffeensis, E. ewingii (dsb).
Serological examinations for antibody detection were performed using an A. phagocytophilum IFA (MegaScreen FLUOANAPLASMA ph., cut-off = 1:50, MegaCor, Hoerbranz, Austria), a microplate L. infantum ELISA (Leishmania-ELISA Dog, Afosa GmbH Dahlewitz, Germany; reference range: negative < 7, borderline 7-12 and positive > 12), an E. canis IFA (MegaScreen FLUOEHRLICHIA c., cut-off = 1:40, MegaCor, Hoerbranz, Austria), and a microplate B. canis ELISA (Babesia-ELISA Dog, Afosa GmbH Dahlewitz, Germany; reference range: negative < 14, borderline 14-19 and positive > 19). In regard to B. vogeli, positive reactions were documented earlier utilizing a B. canis IFA test . Furthermore, a rapid enzyme immunoassay assay test system (IDEXX SNAP® 4Dx®) was used as described elsewhere  following the manufacturer's directions. The IDEXX SNAP® 4Dx® detects antibodies against A. phagocytophilum (p44), Borrelia burgdorferi sensu lato (C6), E. canis (p30, p30-1) as well as antigen of Dirofilaria immitis.
A 1,400-bp fragment of the 16S ribosomal RNA gene was amplified from total blood DNA by using forward (AGAGTTTGATCCTGGCTCAG) and reverse (CGGCTACCTTGTTACGACTT) Anaplasmataceae-specific primers, which were modified according to a study published elsewhere . The reactions were prepared in total volumes of 50 μl with 15 pmol each primer, 0.2 mM each dNTP and 2 U Pwo DNA-polymerase in a 1× polymerase specific buffer complemented with MgSO4. After an initial denaturation step at 94°C the reactions cycled 30 times at 94°C for 30 sec, 58°C for 45 sec, 68°C for 90 sec and finally incubated at 68°C for 5 min in a Mastercycler pro (Eppendorf, Hamburg, Germany). The PCR products were purified with a peqGOLD Gel extraction Kit (PEQLAB, Erlangen, Germany) and submitted for sequencing to Eurofins MWG Operon (Ebersberg, Germany). The PCR product was sequenced three times in both directions. A 1,348-bp partial ribosomal RNA sequence was deposited to GenBank™ under accession number JQ396431.
With the exception of intermittent/recurrent diarrhoea the dog had no clinical signs, but anaemia and thrombocytopenia were evident in the haemogram (Table 1). Both are common, major abnormal clinicopathological findings observed in A. platys and B. vogeli- co-infected dogs  and are in concordance with the positive PCR results for A. platys and B. vogeli found in the dog. Infections with A. platys are difficult to detect, as in most cases the clinical manifestations are not clearly evident. B. vogeli infections usually cause subclinical to mild or moderate changes in adult dogs, however, concomitant conditions or additional pathogens present at the same time can exacerbate the course of infection .
Thrombocytopenia due to a monoinfection with A. platys has a cyclic character and is considered the result of the destruction of blood platelets by the proliferating pathogen during initial phase of infection, which probably triggers immunologic mechanisms in the subsequent course of the infection . In B. vogeli- infected dogs, regenerative haemolytic anaemia is a well-known feature, but not all naturally infected dogs become anaemic. Furthermore, B. vogeli infections do not show a homogenous clinicopathological pattern [29, 30]. Indeed, thrombocytopenia was reported to also be a haematological abnormality of B. vogeli infections, even though the decrease of platelets in blood is a consistent haematological abnormality in B. canis infections [30, 31]. The thrombocytopenia can be enhanced by mixed A. platys and B. vogeli infections leading to significantly lower numbers of platelets compared to single pathogen infection . Both, thrombocytopenia (2.5 fold below lower limit of normal range) and anaemia (low value of RBC count, haemoglobin and haematocrit), were found in the presented case (Table 1). Based on data generated with additional PCR and serology tests there was no evidence of co-infections with other pathogens such as A. phagocytophilum, H. canis, Ehrlichia spp. and Leishmania spp., which may occur also in Croatia and could take further influence on haematological parameters [32–34]. On the other hand, at the moment there are no available reports confirming the occurrence of E. canis in Croatia.
As representatives for acute phase proteins C-reactive protein (CRP, belonging to positive acute phase proteins) and albumin (negative acute phase protein) were evaluated, which in fact showed abnormal values. While CRP was elevated to up to 38.2 mg/L (reference range 0-9.7 mg/L), the serum albumin was decreased to 2.7 g/dL (reference range 3.2-4.7 g/dL) indicating systemic involvement likely due to a co-infection of A. platys and B. vogeli. Acute phase response as a part of the innate host defence system is linked to early innate responses for any pathological processes or diseases as reviewed elsewhere [35, 36]. In dogs, CRP is known to show the highest response among acute phase proteins and has been used as an early unspecific marker . An increase of CRP in dogs has been reported for many diseases caused for example by B. canis, E. canis, L. infantum and A. phagocytophilum. Thus far, no comprehensive investigations of acute phase response in tick-borne concurrent infections such as A. platys and B. vogeli exist.
The serological tests (IFA and SNAP 4Dx) based on A. phagocytophilum antigen failed to detect antibodies in the present case, but an ELISA based on B. canis antigen produced a positive result. However, the antibody level was low. It is known, that IFA tests with B. canis as a substrate cross-react with antibodies against B. vogeli and B. rossi, but homologous species antigens will cause stronger reactions . Furthermore, that exact time point of seroconversion is not known for a natural B. vogeli infection and follow-up serological examinations after initial PCR-positive results, as shown previously , was not possible in the present case. The attempt to detect antibodies against A. platys was based on cross-reaction with A. phagocytophilum antigen. It has previously been shown, that the serum samples from naturally infected A. platys dogs from USA and China react positively with A. phagocytophilum antigen on the SNAP 4Dx [41, 42]. Furthermore serum samples from A. platys infected animals in Portugal react well by means of IFA with A. phagocytophilum antigen . Therefore, strain differences do not appear be a reason for the negative serological tests in the present study. According to the literature, seroconversion as a consequence of an A. platys mono-infection occurs between 13 to 19 days post infection as shown with strains collected in US and Greece [2, 10]. As the dog was imported approx. 30 days before sampling to Germany and it is unlikely that the infection was acquired in Germany, an unusually long seronegative period can be inferred. It can be speculated that the tested blood samples were collected during the acute phase of infection probably before seroconversion occurred. This would be in concordance with negative IFA results as well as with high C-reactive protein and severe thrombocytopenia as described in A. phagocytophilum-infections previously . The other possible reason for the seronegativity could be the dual infection. In case of a simultaneous infection with A. platys and E. canis it was documented, that A. platys-specific antibodies were detectable for the first time on average 27 days (range 14-35 days) post infection . Consequently, the concurrent infections of A. platys and B. vogeli might have induced a delay in the humoral immunological response in this patient.
Intermittent diarrhoea was a reason for the presentation of the dog. The possible cause for the diarrhoea remains unknown, as no faecal sample was examined. But, it seems unlikely that intermittent diarrhoea alone could have led to hypoalbuminemia in the dog due to substantial gastrointestinal loss of fluids.
The dog died shortly after imidocarb dipropionate administration. However, the cause of death stays speculative, because no post-mortem examination was performed. An anaphylactoid reaction was observed in this case. Such reactions were described after imidocarb dipropionate administration in rare cases , while side effects of the drug (hypotension, hypersalivation, nasal discharge, lacrimation, diarrhoea, vomiting) can be reduced by atropine pretreatment. Furthermore, occasional renal tubular necrosis and hepatotoxicity after treatment have been described also [45, 46]. In this case the elevated urea nitrogen of 34.7 mg/dL (reference range 10-25 mg/dL) and creatinine within the reference range (0.9; reference range < 1.4 mg/dl) are indicative for azotaemia but not for a strong renal involvement, which could exacerbate the side effects of imidocarb dipropionate therapy. In a group of seven dogs with B. canis infection, which had been preselected due to renal involvement and treated with imidocarb dipropionate, four dogs died spontaneously and the kidneys of all animals of the group showed degenerative changes of mainly the proximal convoluted tubules as well as necrosis of the whole tubule in some cases . The authors reported that the histological alterations seen in the dogs were similar in dogs treated with imidocarb dipropionate and in the untreated animal. Hence, the pathological changes observed in the kidneys cannot be explained exclusively by the potential nephrotoxicity of imidocarb dipropionate . In the case presented here it remains open, whether the dog died because of the side effects or as a result of the combined impact of the observed B. vogeli and A. platys infections. Nevertheless, it is recommended to treat simultaneously even mildly azotaemic dogs with proper intravenous fluid therapy when imidocarb dipropionate is applied and lower the dose of the drug (e.g. 3 mg/kg) in patients suspected to have renal involvement, in order to decrease the risk of renal insufficiency .
Sequencing of the 16S rRNA gene produced a 1,348-bp sequence, which was identical to other A. platys 16S rRNA sequences generated in Thailand (EF139459), Japan (AY077619) and France (AF303467). The analysis of 16S rRNA sequences confirms the previous results, in which little genetic diversity was observed in 16S rRNA sequences of A. platys strains from different geographic locations [8, 22, 48].
The mixed A. platys and B. vogeli infection is well known from previous reports and represents a large fraction of co-infections among tick borne pathogens in dogs [20, 26, 49]. The examination of free-roaming dogs associated with remote Aboriginal communities in Australia showed 11% of dogs co-infected with both A. platys and B. vogeli. The occurrence of B. vogeli in Croatia and other southern European countries is well documented . The natural occurrence of A. platys in moderate climate zones and simultaneous coinfections with Babesia spp. allow the assumption that Rhipicephalus sp. e.g. Rhipicephalus sanguineus ticks may serve as the natural vector. On the other hand, this tick species has a wide genetic diversity  making it difficult to find the definitive vector for A. platys. The detection of A. platys in non-engorged questing adult Rhipicephalus turanicus tick in Israel  raises the question, whether other Rhipicephalus species actually serve as vectors for this agent.
To our knowledge this is the first report of A. platys infection in a dog imported from Croatia. Based on the owner's declaration, the dog has never travelled abroad before importation to Germany and the dog was presumably infected in Croatia. The infection in Germany is unlikely due to the following reasons: the vectors Rhipicephalus spp. ticks are commonly not present in Germany, even though ticks imported from abroad with travelled dogs can survive in homes with moderate temperatures . The dog owners in Germany, however, did not report any tick infestation and so far there are no reports of autochthonous occurrences of B. vogeli and A. platys from the region where the dog spent its final days.
A. platys infection was detected for the first time in a dog imported from Croatia. A co-infection with B. vogeli probably induced the untypical serological results. The major clinical manifestations were thrombocytopenia, anaemia and elevated values of C-reactive protein. A. platys infection should be considered in dogs living in or returning after travel from this area and showing abnormal clinicopathologic findings described in this report.
Infectious canine cyclic thrombocytopenia
Immunofluorescence antibody assay
White blood cell
Enzyme linked immunosorbent assay
Complete Blood Count
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.