Veterinarni Medicina, 52, 2007 (5): 207–212

Case Report

Infection with Anaplasma phagocytophilum in a young

dog: a case report
O. Melter1,3, I. Stehlik2, H. Kinska1, I. Volfova1, V. Ticha1, D. Hulinska3
1
Labvet, Clinical Veterinary Laboratory, Prague, Czech Republic
2
Veterinary Clinic, Prague, Czech Republic
3
National Institute of Public Health, Prague, Czech Republic

ABSTRACT: An 11-months-old male Golden Retriever occasionally found to have Ixodes ricinus ticks attached to
the skin developed the acute onset of fever, lameness and inappetence followed by rapidly progressive depression,
ataxia and reluctance to move. Inclusions (morulae) were observed in granulocytes. The blood analysis revealed
severe thrombocytopoenia, lymphopoenia, eosinopoenia, elevation of alkaline phosphatase and hypercholestero-
laemia, mostly suggestive of an Anaplasma phagocytophilum infection. The amplification of a DNA sequence
specific for Anaplasma phagocytophilum and detection of specific antibodies supported the diagnosis. Borrelia
burgdorferi, another tick-borne pathogen, or specific antiborrelial IgG antibodies were not detected. The dog was
treated with oral doxycycline for 14 days: clinical symptoms resolved within six days.

Keywords: anaplasmosis; infection; dog

A. phagocytophilum is an obligate intracellular gene of the agent (Kirtz et al., 2005). Although two
pathogen known to cause granulocytic infections pet dogs with granulocytic morulae in their venous
in both animals and humans (Dumler et al., 2001; blood were reported in the Czech Republic (Huml
Cohn, 2003). The most common vector of A. phago- et al., 1996), the causative species was identified
cytophilum in Europe is the tick Ixodes ricinus. Pet only in game animals and humans by Hulinska et
animals are not a source of infection for humans but al. (2002, 2004).
could serve as a sentinel (Bjoersdorff, 2002). Clinical symptoms of the infection with A. phago-
Canine granulocytic anaplasmosis has been cytophilum in dogs are not specific but usually
documented sporadically worldwide, with preva- include fever, lethargy, thrombocytopoenia, de-
lence varying widely among countries (Engvall et pression and anorexia (Engvall et al., 1996). In
al.; 1996; Pusterla et al., 1998; Liddell et al., 2003; experimentally infected dogs only mild symptoms
Bexfield et al., 2005; Shaw et al., 2005; Liebisch were described (Cohn, 2003).
et al., 2006). The detection of morulae in granu- Dogs might be coinfected with other tick-borne
locytes does not identify the agent to the species species such as Ehrlichia spp. or Borrelia spp. and
level and further testing is needed (Cohn, 2003; the disease manifests itself primarily as acute or
Sirigireddy and Ganta, 2005). Lester et al. (2005), subacute arthritis in contrast to the infection in
Kirtz et al. (2000, 2005) and Pusterla et al. (1997) humans (Straubinger, personal communication).
reported 5–37% of infected granulocytes in dogs Antiborrelial antibodies can first be detected
with anaplasmosis. The culture detection of the by ELISA in the serum of dogs between 4 and
agent is still not a routine method yet and serologic 6 weeks after infection, with their titre peaking in
testing alone is not conclusive for diagnosis. The 90 to 120 days and remaining constant for at least
diagnosis of A. phagocytophilum infection is mostly 22 months after exposure (Straubinger, personal
achieved by an indirect immunofluorescence assay communication). Although PCR is sensitive and
and by amplified DNA specific to the 16S rRNA specific, negative results do not exclude the infec-

207
Case Report Veterinarni Medicina, 52, 2007 (5): 207–212

tion with B. burgdorferi (Cohn, 2003). Twenty-five (Day 0), the dog showed the severe acute onset of
percent of dogs healthy or suspected of borrelial pyrexia (39.6°C), lethargy, inappetence and ataxia.
or anaplasmal infection (n = 731) had significant It was not able to either stand or walk. The general
titres for both infections. The coinfection with clinical examination including the inspection and
A. phagocytophilum and B. burgdorferi is more palpation of leg bones and joints did not reveal any
likely to induce illness in the dogs as compared to additional physical abnormalities.
the infection with either organism alone (Beall et The baseline biochemistry screening (Spotchem,
al., 2006). SP-4430, Arkray) showed hypercholesterolaemia
Infections with other tick-borne agents – Babesia and elevation of alkaline phosphatase level (Table 1).
spp. and tick-borne encephalitis (TBE) virus – with The remaining parameters tested (urea, creatinine,
typical clinical symptoms are also reported spo- ALT, AST, GMT) were within the reference ranges.
radically in dogs in the Czech Republic (Klimes Intravenous fluid therapy with saline solution (60
et al., 2001). ml/kg/24 h) was immediately initiated. Intravenous
We present a case report of a young dog that fully metamizol (25 mg/kg, Vetalgin inj., Intervet) and
recovered after developing febrile illness, lameness, subcutaneous amoxicillin/clavulanate (12.5 mg/kg,
depression and ataxia as the acute onset of sus- Synulox, RTU, Pfizer) were given to control fever,
pected tick-borne disease. pain and suspected infection.
On Day 2, the fever (40.5°C) and depression in-
creased. Intravenous fluid therapy with saline solu-
Case history tion was given again (80 ml/kg/24 h). Haematological
(AL Cell Counter 871, AL Systeme) and biochemi-
An 11-months-old pet male Golden Retriever cal (Vitalab Selectra, Merck; EasyLyte Plus, Medica)
found to have several Ixodes ricinus adult female analyses revealed some values outside the reference
ticks attached to the skin between March and June ranges (Table 1). Severe thrombocytopoenia, lym-
developed acute signs of infection at the begin- phopoenia and elevated erythrocyte sedimentation
ning of June. On admission to the veterinary clinic rate (+83/h, adjusted for hematocrit +77/h, refer-

Table 1. Selected haematological and biochemical parameters of blood samples from the studied dog infected with
A. phagocytophilum

Parameter At baseline Day 2 Week 3 Month 5 Reference range**

Thrombocyte (count) ND 29* 287 299 200–500 × 109 /l
Leukocyte (count) ND 10.1 10.6 12.3 6–17 × 109 /l
Band neutrophil (count) ND 0.05*/505* 0 0 rel. 0–0.03/abs. 0–300 mm3
Segmented neutrophil (count) ND 0.87*/8 787 0.57*/6 042 0.62/7626 rel. 0.60–0.77/abs. 3 000–11 500 mm3
Lymphocyte (count) ND 0.03*/303* 0.30/3 180 0.34/4182 rel. 0.12–0.30/abs. 1 000–4 800 mm3
Monocyte (count) ND 0.05/505 0.06/636 0.02*/246 rel. 0.03–0.10/abs. 150–1 350 mm3
Eosinophil (count) ND 0* 0.07/742 0.02/246 rel. 0.02–0.10/abs. 100–1 250 mm3
Phosphorus ND 1.33 ND 1.62* up to 1.50 mmol/l
Calcium ND 2.82* ND 2.70 up to 2.77 mmol/l
Cholesterol 11.0* 10.51* ND 10.91* 2.74–7.65 mmol/l
Total protein ND 64.0 ND 68.6 58–76 g/l
Albumin ND 36.75 ND 39.09* 23–36 g/l
ALP 4.29 5.04* ND 2.30 up to 2.60 μkat/l
Creatinine kinase ND 4.25* ND 1.16 up to 2.50 μkat/l

ND = not determined; *the value outside the reference range, **based on Bush (1996)

208
Veterinarni Medicina, 52, 2007 (5): 207–212 Case Report

a b c

Figure 1 a–c. Blood smears from the acute onset of the illness. Anaplasma phagocytophilum morulae are evident
within the dog granulocytes. Giemsa-Romanowski staining (magnification 1 000×)

ence range up to 10/1/h) were observed. Eosinophils tively. Template DNA was isolated from EDTA-
could not be detected and the total number of leu- collected blood specimens of the infected animal
kocytes was within the reference range (Table 1). and five clinically healthy dogs as negative controls
The other haematological parameters, i.e. haemo- by a QiaAmpTissue kit (Qiagene). DNAs from the
globin, hematocrit and calculated characteristics, strain EHR 02 of A. phagocytophilum (Hulinska et
were within reference ranges (MCV 72.1 fl; MHC al., 2004) and strain M192 of B. garinii (Hulinska et
14.5 fmol and MCHC 32.7g/dl). Giemsa-stained al., 1999) were used as positive controls. The 293 bp
blood smears revealed the presence of isolated bac- amplicon specific to A. phagocytophilum was de-
teria or inclusions (morulae) (Figure 1) in 36% of tected only in the infected dog and the positive con-
three hundred neutrophils. trol (Figure 2) but it was not found in the healthy
Hypercholesterolaemia and elevation of creati- dogs. No amplicon specific to B. burgdorferi was
nine kinase and alkaline phosphatase were observed detected in the blood specimens from either the
(Table 1). Amylase activity was reduced (7.10 μkat/l, infected animal or the five clinically healthy dogs.
reference range 10–30 μkat/l). The other biochemi- The ELISA test using whole cell B. afzelii antigen
cal parameters tested (urea, creatinine, phospho- (Test-Line, Brno, Czech Republic) detected IgM
rus, calcium, sodium, potassium, chloride, glucose, to B. burgdorferi (index of positivity 1.40; positive
total protein, albumin, bilirubin, ALT, AST, GMT, result higher than 1.15) and IgG negativity in the
and lipase) were in reference ranges. PCR analy- blood of the infected dog. Since A. phagocytophilum
ses of A. phagocytophilum and B. burgdorferi sensu and borrelial infection was suspected, the antibiotic
lato were performed as described by Jackson et was converted to 10 mg/kg doxycycline (Ronaxan,
al. (2002) and Marconi and Garon (1992), respec- Merial) given orally twice daily for 14 days.

M 1 2 3 M

500 bp

400 bp

300 bp Figure 2. Agarose gel electrophoresis of 293 bp ampli-

cons obtained with primers EHR521 and EHR790 from
the infected dog’s blood (1) and DNA extracted from
200 bp A. phagocytophilum control strain EHR 02 (2). DNA
extracted from five healthy dogs not included in the
figure and PCR mixture (3) were used as negative con-
100 bp trols analyzed with PCR. Lanes M show size standards
(100 bp DNA ladder)

209
Case Report Veterinarni Medicina, 52, 2007 (5): 207–212

On the following day the temperature decreased palpation indicated that the nervous system was
to 39.0°C. Intravenous glucose in a 5% solution not involved. Even if no special neurological and
(20 ml/kg/24 h) and fluid therapy with vitamins musculoskeletal examinations were performed, the
(Duphalyte, Fort Dodge, 5 ml/kg/24 h) were ad- reluctance to move, typically reported in acutely
ministered. infected animals (Engvall et al., 1996), seemed to
On Day 5, the temperature further decreased to be due to deep depression known to result from
38.5°C and the dog showed only slight lethargy. The stress as a consequence of homeostasis imbalance
clinical examination on Day 6 indicated full clinical due to inflammation.
recovery, with the temperature falling to 38.2°C. Inclusions were observed in 36% of neutrophils:
The therapy with doxycycline was continued for this proportion was close to the highest reported
another 14 days. figure, i.e. 37% (Kirtz et al., 2000). The persistence
Three weeks after the first examination all hae- of the infection was screened microscopically and
matological parameters were within reference rang- by the amplification of specific A. phagocytophilum
es. The recovered dog was followed up at monthly DNA in two consecutive blood samples collected
intervals. three weeks and five months after the acute at-
Five months later, the laboratory analysis con- tack with negative results. However, Egenvall et
firmed haematological and biochemical parameters al. (2000) detected A. phagocytophilum infection
to be in reference ranges, with the only exception in three dogs for six months but caution would be
of mild hypercholesterolaemia (Table 1). Neither advised in comparing their results obtained in the
anaplasmal nor borrelial DNA was detected in the experimentally infected and immunosuppressed
convalescent venous blood five months after the dogs with ours found in the naturally infected ani-
acute onset of infection. ELISA of IgM and IgG mal.
specific to B. burgdorferi gave the same result as The amplification of the specific fragment of
the baseline tests, i.e. the dog was IgM positive A. phagocytophilum and not of that of Borrelia spp.
and IgG negative again. Specific IgG antibodies to confirmed the infection with A. phagocytophilum.
A. phagocytophilum were detected in the conva- The positivity to antiborrelial IgM antibodies indi-
lescent serum by the indirect immunofluorescence cated either an exposure to Borrelia spp. or cross-
assay (IFA) at a titre of 1:320 five months after the reactivity with anaplasmal or other gram-negative
entry examination. bacterial surface antigens. The aetiological role of
Borrelia burgdorferi in the case presented is dis-
putable as no DNA and specific IgG antibodies
DISCUSSION were detected on admission and in convalescent
specimens, while ELISA antibodies to whole cell
Canine granuloc ytic anaplasmosis usually or recombinant Borrelia burgdorferi antigens were
presents as an acute febrile systemic illness (Stiles, reported to remain constant for nearly two years
2000). Specific diagnostic tests include visualisa- (Straubinger, personal communication).
tion of specific morulae, anti-Anaplasma IgM and The elevation of hepatic transaminase activity
IgG antibody detection and PCR analysis, which is presented in the literature (Bjoersdorff, 2002)
is most reliable for early diagnosis (Engvall et al., rather than increased alkaline phosphatase (ALP),
1996; Bjoersdorff, 2002). creatinine kinase (CK) and cholesterolaemia as
The dog presented the most typical systemic described in our case history. The precise mecha-
clinical symptoms, i.e. fever, ataxia, depression, nisms for the elevations remain unknown. Elevated
lameness, and laboratory results, i.e. – thrombo- ALP and CK could probably be connected with
cytopoenia, eosinopoenia, lymphopoenia and oc- septicaemia because of the effects of large num-
currence of band neutrophils (Bjoersdorf, 2002), bers of multiplying bacteria and endotoxaemia.
suggestive of acute granulocytic anaplasmosis. Bacterial endotoxins released in the blood could
However, other reported symptoms such as lym- evoke liver and cardiac cell damage (Bush, 1996).
phadenopathy, splenomegaly and hepatomegaly The reasons for stable elevated cholesterol levels
(Neer, 1998) were not detectable by palpation. remain unclear. Increased lipolysis at the time of
The inspection and palpation of leg bones, mus- acute stress or starvation could be omitted because
cles and joints did not reveal any physical abnor- the level persisted five months after the successful
malities in the limbs. The reactivity to pain during therapy. An idiopathically increased serum choles-

210
Veterinarni Medicina, 52, 2007 (5): 207–212 Case Report

terol level in the dog could be one reason explaining REFERENCES

the phenomena (Bush, 1996).
The therapy with doxycycline proved as effec- Beall M., Chandrashekar R., Eberts M. (2006): Borrelia
tive as found by others (Stiles, 2000; Cohn, 2003). burgdorferi and Anaplasma phagocytophilum: poten-
The temperature started to decrease on day 2 and tial implications of co-infection on clinical presenta-
all clinical symptoms resolved on day 6 after the tion in the dog. Journal of Veterinary Internal Medicine,
initiation of the therapy. At 3-week and 5-month 20, 713–714.
follow-ups no clinical or laboratory changes except Bexfield N.H., Villiers E.J., Herrtage M.E. (2005): Immune-
for elevated cholesterol were observed. mediated haemolytic anaemia and thrombocytopenia
It is not likely that the dog was coinfected with associated with Anaplasma phagocytophilum in a dog.
another tick-borne agent. Babesia spp. was not Journal of Small Animal Practice, 46, 543–548.
seen microscopically in red blood cells and most Bjoersdorff A. (2002): Canine granulocytic ehrlichiosis
of the typical clinical symptoms of babesios (anae- due to Anaplasma phagocytophila. In: Beugnet F. (ed.):
mia, icterus, haemoglobinuria, dyspnoea and a Guide to Major Vector-borne Diseases of Pets. Merial,
low body condition index) did not appear in the France. 123–127.
dog (Taboada, 1998). The coinfection with TBE Bush B.M. (1996): Interpretation of Laboratory Results
virus was possible but unlikely as the primary TBE for Small Animal Clinicians. Blackwell Science, New
symptoms typical of TBE infection (shyness and York. 273–276.
unwillingness, periods of excitement and irregu- Cohn L.A. (2003): Ehrlichiosis and related infections.
lar movements and high fever in the course of the Veterinary Clinics: Small Animal Practice, 33, 863–
disease) (Svoboda et al., 2001) were not observed 884.
and the dog responded quickly (within 24 hours) Dumler J.S., Barbet A.F., Bekker C.P., Dasch G.A., Palmer
to the specific antibiotic therapy suggesting bacte- G.H., Ray S.C., Rikihisa Y., Rurangirwa F.R. (2001):
rial infection. Reorganization of genera in the families Rickettsiaceae
All findings summarized above, particularly the and Anaplasmataceae in the order Rickettsiales: uni-
presence of specific morulae and DNA in neu- fication of some species of Ehrlichia with Anaplasma,
trophils, negative laboratory or clinical results for Cowdria with Ehrlichia and Ehrlichia with Neorick-
other tick-borne agents, thrombocytopoenia and ettsia, descriptions of six new species combinations
the fact that a high titre of specific IgG antibodies and designation of Ehrlichia equi and ’HGE agent’ as
was detected in the convalescent serum suggest subjective synonyms of Ehrlichia phagocytophila.
that A. phagocytophilum was the single causative Journal of Systematic and Evolutionary Microbiology,
agent of the acute infection. Other authors also 51, 2145–2165.
reported the detection of specific IgG antibodies Egenvall A., Lilliehook I., Bjoersdorff A., Engvall E.O.,
by IFA from 2 weeks up to more than six months Karlstam E., Artursson K., Heldtander M., Gunnarsson
after infection in dogs suffering from anaplasmosis A. (2000): Detection of granulocytic Ehrlichia species
(Kirtz et al., 2000, 2005; Poitout et al., 2005). To DNA by PCR in persistently infected dogs. Veterinary
the author’s knowledge this is the first confirmed Record, 146, 186–190.
case of canine anaplasmosis caused by A. phago- Engvall E.O., Petterson B., Person M., Artusson K., Jo-
cytophilum to be reported in the Czech Republic, hansson K.E. (1996): A 16S rRNA-based PCR assay
where the infection was already suspected (Huml for detection and identification of granulocytic Ehr-
et al., 1996). lichia species in dogs, horses, and cattle. Journal of
Clinical Microbiology, 34, 2170–2174.
Hulinska D., Votypka J., Valesova M. (1999): Persistence
Acknowledgements of Borrelia garinii and Borrelia afzelii in patients with
Lyme arthritis. Zentralblatt fur Bakteriologie, 289,
The authors thank Kamil Sedlak, DVM, PhD, 301–318.
from the State Veterinary Institute in Prague for Hulinska D., Votypka J., Plch J., Vlcek E., Valesova M.,
measuring anti-borrelial antibodies and for the Bojar M., Hulinsky V., Smetana K. (2002): Molecular
confirmation of anti-anaplasmal antibodies. Thanks and microscopical evidence of Ehrlichia spp. and Bor-
are also due to Eva Kodytkova from the National relia burgdorferi sensu lato in patients, animals and
Institute of Public Health in Prague for reviewing ticks in the Czech Republic. New Microbiologica, 25,
the manuscript. 437–448.

211
Case Report Veterinarni Medicina, 52, 2007 (5): 207–212

Hulinska D., Langrova K., Pejcoch M., Pavlasek I. (2004): 2830–2834. Erratum in: Journal of Clinical Microbiol-
Detection of Anaplasma phagocytophila in animals ogy, 1993, 31, 1026.
by real-time polymerase chain reaction. Acta Patho- Neer T.M. (1998): Canine monocytic and granulocytic
logica, Microbiologica et Immunologica Scandinavica, ehrlichiosis. In: Greene C.E. (ed.): Infectious Diseases
112, 239–247. of the Dog and Cat. 2nd ed. W.B. Saunders Co., Phila-
Huml O., Cada F., Bohm, K., Velebny H. (1996): Granu- delphia. 139–147.
locytic ehrlichiosis in a dog (in Czech). Veterinarstvi, Poitout F.M., Shinozaki J.K., Stockwell P.J., Holland C.J.,
11, 464–466. Shukla S.K. (2005): Genetic variants of Anaplasma
Jackson C.A., Lovrich S.D., Agger W.A., Callister S.M. phagocytophilum infecting dogs in Western Washington
(2002): Reassessment of Midwestern lyme disease focus State. Journal of Clinical Microbiology, 43, 796–801.
for Borrelia burgdorferi and the human granulocytic Pusterla N., Huder J., Wolfensberger C., Litschi B., Parvis
ehrlichiosis agent. Journal of Clinical Microbiology, A., Lutz H.J. (1997): Granulocytic ehrlichiosis in two
40, 2070–2073. dogs in Switzerland. Journal of Clinical Microbiology,
Kirtz G., Leidinger E., Moser V. (2000): Canine granulo- 35, 2307–2309.
cytare Ehrlichiose (CGE) bei einem Hund in Osterreich. Pusterla N., Pusterla J.B., Deplazes P., Wolfensberger C.,
Wiener Tierazliche Monatsschrift, 87, 241–246. Miller W., Horauf A., Reusch C., Lutz H. (1998): Se-
Kirtz G., Meli M., Leidinger E., Ludwig P., Thum D., roprevalence of Ehrlichia canis and of granulocytic
Czettel B., Kolbl S, Lutz H. (2005): Anaplasma phago- ehrlichia infection in dogs in Switzerland. Journal of
cytophilum infection in a dog: identifying the causative Clinical Microbiology, 36, 3460–3462.
agent using PCR. Journal of Small Animal Practice, Shaw S.E., Binns S.H., Birtles R.J, Day M.J, Smithson R.,
46, 300–303. Kenny M.J. (2005): Molecular evidence of tick-trans-
Klimes J., Juricova Z., Literak I., Schanilec P., Trachta e mitted infections in dogs and cats in the United King-
Silva E. (2001): Prevalence of antibodies to tick-borne dom. Veterinary Record, 157, 645–648.
encephalitis and West Nile flaviviruses and the clinical Sirigireddy K.R., Ganta R.R. (2005): Multiplex detection
signs of tick-borne encephalitis in dogs in the Czech of Ehrlichia and Anaplasma species pathogens in pe-
Republic. Veterinary Record, 148, 17–20. ripheral blood by real-time reverse transcriptase-
Lester S.J., Breitschwerdt E.B., Collis C.D., Hegarty B.C. polymerase chain reaction. Journal of Molecular
(2005): Anaplasma phagocytophilum infection (gran- Diagnostics, 7, 308–316.
ulocytic anaplasmosis) in a dog from Vancouver Is- Stiles J. (2000): Canine rickettsial infections. Veterinary
land. Canadian Veterinary Journal, 46, 825–827. Clinics of North America: Small Animal Practice, 30,
Liddell A., Stockham S.L., Scott M.A., Sumer J.W., Paddock 1135–1149.
C.D., Gaudrealt-Keener M., Arens M.Q., Storch G.A. Svoboda M., Pospisil Z., Svobodova V., Klimes J., Knotek
(2003): Predominance of Ehrlichia ewingii in Missouri Z., Smola J., Toman M., Matouch O., Literak I., Cizek
dog. Journal of Clinical Microbiology, 41, 4617–4622. A., Treml F., Celer V., Zendulkova D., Lany P., Doubek
Liebisch G., Thiet W., Liebisch A. (2006): Die canine J., Schanilec P., Vernerova E., Halouzka R., Kohout P.,
monozytare (CME) und die canine granulozytäre Ehr- Srenk P., Barton J., Hera A. (2001): Infectious and
lichiose (CGE), zwei durch Zecken ubertragene Infek- parasitic diseases. In: Svoboda M., Senior D.F., Doubek
tionskrankheiten bei Hunden in Deutschland. Der J., Klimes J. (eds.): Disease of the Dog and Cat (in
Praktische Tierarzt, 87, 342–353. Czech). Noviko a.s., Brno. 1840–1841.
Marconi R.T., Garon C.F. (1992): Development of Taboada, J. (1998): Babesiosis. In: Greene C.E. (ed.): In-
polymerase chain reaction primer sets for diagnosis fectious Diseases of the Dog and Cat. W.B. Saunders
of Lyme disease and for species-specific identification Company, Philadelphia. 473–481.
of Lyme disease isolates by 16S rRNA signature nu-
cleotide analysis. Journal of Clinical Microbiology, 30, Received: 2006–04–27
Accepted after corrections: 2007–04–11

Corresponding Author:
MVDr. Oto Melter, PhD., National Institute of Public Health, Srobarova 48, 100 42 Prague, Czech Republic
Tel. +420 267 082 255, e-mail: [email protected]

212