Anaplasma phagocytophilum Typical and Atypical Manifestations of Infection in Dogs

Typical and Atypical Manifestations of Anaplasma
phagocytophilum Infection in Dogs
Matthew D. Eberts, DVM, Pedro Paulo Vissotto de Paiva Diniz, PhD, DVM*, Melissa J. Beall, DVM,
Brett A. Stillman, PhD, Ramaswamy Chandrashekar, PhD, Edward B. Breitschwerdt, DVM
Eighteen clinically ill dogs, naturally infected with Anaplasma phagocytophilum, were examined at a veterinary practice in
Baxter, Minnesota. A clinical examination, complete blood cell count, enzyme- linked immunosorbent assay (ELISA) for A
phagocytophilum, Borrelia burgdorferi, and Ehrlichia canis antibodies and Dirofilaria immitis antigen, and a polymerase chain
reaction test for A phagocytophilum DNA were obtained for all dogs. Physical examination findings included fever, arthropathy, lymphadenopathy, epistaxis, acute gastritis, cervical hyperpathia, and central nervous system dysfunction. Complete
blood cell count abnormalities included thrombocytopenia, morulae in neutrophils, anemia, leukopenia, eosinopenia, lymphopenia, and monocytosis. Seroreactivity to A phagocytophilum was found in 61%, B burgdorferi antibodies in 17%, and
D immitis antigen in 5% of the dogs. Fever, arthropathy, neurologic dysfunction, and epistaxis are clinical syndromes that can
be associated with A phagocytophilum infection. Treatment with doxycycline resulted in rapid resolution of clinical signs in all
dogs. (J Am Anim Hosp Assoc 2011; 47:e86–e94. DOI 10.5326/JAAHA-MS-5578)
dogs in 1996.5 Granulocytic anaplasmosis can cause hematologic
Anaplasma phagocytophilum is a gram-negative intracellular bac-
abnormalities, most typically thrombocytopenia, leukopenia, and
terium transmitted by Ixodes species ticks and is the causative
anemia.5–8 The mechanism of thrombocytopenia is not fully un-
agent of granulocytic anaplasmosis.1 In the United States, Ixodes
derstood. but based on mouse experimental models and the
scapularis is the vector for A phagocytophilum in the Northeast and
presence of antiplatelet antibodies in naturally infected dogs, both
Midwest, whereas Ixodes pacificus vectors the organism on the
consumptive- and immune-mediated mechanisms appear to be
west coast.1 Surveys identified a wide distribution in seropreva-
involved.6,9 Granulocytic intracytoplasmic morulae, which are
lence among dogs in the United States, and high rates of exposure
clusters of reproducing bacteria located in phagosomes, can be
corresponded to the distribution of I scapularis and I pacificus.2 In
visualized during blood smear evaluation.10 Serum biochemical
recent years, there were dynamic changes in the range of these tick
abnormalities can include increased serum alkaline phosphatase
vectors and their associated diseases. Climate change and changes
activity, hypoalbuminemia, elevated serum alanine transferase,
in wildlife distribution enabled expansion of I scapularis in North
and elevated total bilirubin.5,6,8
America, and this trend is predicted to continue.
Both enzyme-linked immunosorbent assay (ELISA) and in-
A phagocytophilum has been documented to cause disease in
direct fluorescent antibody tests are commercially available, and
a wide variety of species, including dogs, cats, horses, ruminants,
documentation of seroconversion can be used to confirm the
and humans. Canine granulocytic anaplasmosis was first de-
diagnosis of granulocytic anaplasmosis. Experimental evidence
scribed as an acute febrile disease in Minnesota and Wisconsin
demonstrated that dogs might seroconvert within 10 days of
From the Lakeland Veterinary Hospital, Baxter, MN (M.E.); Intracellular
Pathogens Research Laboratory College of Veterinary Medicine,
North Carolina State University, Raleigh, NC (P.D., E.B.); and IDEXX
Laboratories, Inc, Westbrook, ME (M.B., B.S., R.C.).
ELISA enzyme-linked immunosorbent assay; IFA immunofluorescent assay;
NCSU North Carolina State University; PCR polymerase chain reaction
*P. Diniz’s present affiliation is Small Animal Internal Medicine, Western University of Health Sciences College of Veterinary Medicine, Pomona, CA.
Correspondence: [email protected] (M.E.)
Nov/Dec 2011
ª 2011 by American Animal Hospital Association
Manifestations of Anaplasma phagocytophilum Infection
exposure.11 Clinical illness usually occurs within 1–2 wk after tick
NCSU). Cases were selected from 23 clinically ill dogs PCR tested
transmission of A phagocytophilum; however, based on the high
at NCSU and IDEXX Laboratories. Inclusion criteria for the cases
seroprevalence in historically healthy dogs in endemic regions,
selected for this study included molecular confirmation of A
most exposed dogs will not develop overt clinical disease. Due to
phagocytophilum infection using PCR in conjunction with the
the acute nature of this infection, approximately 40% of clinically
availability of a medical history, physical examination findings,
ill dogs will not produce a detectable level of A phagocytophilum
a complete blood count, an in-clinic serologic assaya, and access to
antibodies at the time of presentation when using current sero-
the clinical response to doxycycline, as follow-up data. Eighteen
logic tests.5–8 In contrast, A phagocytophilum polymerase chain
dogs fulfilled these criteria (three dogs were excluded due to lack
reaction (PCR) assays can be used to detect organism-specific
of complete blood count at initial presentation, and two dogs were
DNA sequences in the blood during the early stages of ill-
excluded for lack of follow-up information). Anaplasmosis was
With proper primer selection and validation, PCR testing
diagnosed in all 18 cases between April and November 2005.
provides a sensitive and specific modality for the diagnosis of
acute anaplasmosis in animals and human patients, and a positive
Complete Blood Counts
PCR result is considered indicative of infection.10 However, it is
Complete blood counts were performed at the time the dog pre-
important to note that A phagocytophilum PCR can be positive in
sented for diagnostic evaluation using an in-clinic laser flow
clinically normal dogs and periodically negative in actively
cytometry automated unitb. Certified veterinary technicians vi-
infected dogs, possibly due to variations in numbers of circulating
sually examined a blood smear to evaluate cell morphology,
organisms during infection.8,11,12
to perform differential cell counts, and to determine whether
As a vector-borne infectious pathogen, A phagocytophilum
morulae were present.
can induce a variety of disease manifestations in dogs, which
presents a challenge to practitioners attempting to accurately di-
agnose and determine effective treatment of their patients. Canine
All dogs were tested at the time of presentation with a qualitative
anaplasmosis is expanding geographically across North America.
ELISA assay that detects Ehrlichia canis and Borrelia burgdorferi
With increasing number of dogs exposed to this pathogen, natural
antibodies, and Dirofilaria immitis antigenc. Retrospectively, all
infection studies need to be disseminated to provide the veteri-
dogs were also tested with a qualitative ELISA assay detecting
nary community with a more complete definition of the spectrum
E canis, B burgdorferi, and A phagocytophilum antibodies and
of disease. It is important that practitioners be aware of potential
D immitis antigena. The latter testing platform detects antibodies
clinical signs and the tests available to diagnose anaplasmosis. The
directed against a synthetic A phagocytophilum peptide from the
purpose of this study is to describe clinical, hematologic, and
major surface protein (p44/MSP2) and reportedly has 99.1%
serologic findings and treatment outcomes for 18 dogs naturally
sensitivity and 100% specificity compared with an immunofluo-
infected with A phagocytophilum in an I scapularis–endemic
rescent assay (IFA) test.13 The B burgdorferi C6 peptide used in
this assay does not cross-react with vaccinal antibodies and is
considered to indicate active as opposed to previous infection.14
Materials and Methods
Molecular Analysis
Cases were retrospectively selected from a previous study involving
PCR testing was performed either by personnel in the Vector Borne
731 dogs, 81% of which were purebred (Eberts, unpublished data).
Disease Diagnostic Laboratory at NCSU or at IDEXX Laboratories
These dogs were examined between July 2004 and January 2006 at
in Westbrook, Maine. Eighteen ethylenediaminetetraacetic acid–
a companion animal hospital located in Baxter, Minnesota, an area
anticoagulated blood samples collected between April 2005 and
endemic for I scapularis.12 Based on resources available at the time
November of 2005 were processed for genomic DNA and analyzed
of the study, retrospective PCR testing was performed at North
by PCR to detect A phagocytophilum infection. NCSU tested
Carolina State University (NSCU) on 273 dogs, of which 51 had
a total of 14 samples, and IDEXX tested a total of 8 samples.
clinical signs consistent with anaplasmosis and/or borreliosis.12 Of
Four of the 18 samples were tested by both laboratories. DNA
these 51 dogs, 19 tested PCR positive for A phagocytophilum
was extracted from 200 mL of canine ethylenediaminetetra-
DNA.12 Further retrospective PCR testing was performed at
acetic acid whole-blood samples that were stored frozen at
IDEXX Laboratories on eight additional clinical cases (four al-
2708C with a commercially available kitd. The final eluted
ready PCR tested at NCSU and four not previously tested at
volume was 200 mL per sample. The DNA concentration was
quantified by spectrophotometry, and the absence of PCR
amplified PCR products were then evaluated on a 2% agarose gel
inhibitors was demonstrated by the amplification of a fragment
stained with ethidium bromide.
of the glyceraldehyde-3-phosphate dehydrogenase gene, as
previously described.15
Canine DNA from a healthy dog was used as a PCR negative
control. Plasmid clones of partial DNA sequences identical to the
following accession numbers were also used as positive controls:
North Carolina State University Protocol
A platys 16S rRNA (AY821826), A platys GroEL (AF478129),
PCR was performed using 16S rRNA oligonucleotide primers
A phagocytophilum 16S rRNA (AF507941), A phagocytophilum AnkA
that were designed to amplify all Anaplasma and Ehrlichia spe-
(CP000235), and A phagocytophilum msp2 (AY151054). The DNA
cies. Subsequently, the Anaplasma platys GroEL gene and the
concentration of the purified plasmid was determined by re-
A phagocytophilum AnkA gene were targeted as a secondary con-
striction enzyme digestion and agarose gel electrophoresis (2%).
firmation of the initial PCR result and to determine the infecting
Plasmid clones were diluted 10-fold in purified dog DNA, ranging
Anaplasma species (A phagocytophilum vs A platys). The oligonu-
from 1 million to 1 plasmid/mL. The limit of detection observed
cleotide primers were described elsewhere. The 16S rRNA gene
in PCR amplifications was 16S rRNA ¼ 10 copies per reaction,
amplifications were performed in a 25-mL final volume reaction
AnkA gene ¼ 25 copies per reaction, and GroEL gene ¼ 5 copies
containing 13buffer with 2 mM magnesium chloride, 0.625 U of
per reaction. The limit of detection was reached in 100% of tests.
Takara Ex Taq DNA polymerase, 12.5 pmol of each primer,
To prevent PCR amplicon contamination, sample extraction, re-
200 mM (each) of deoxyadenosine triphosphate, deoxythymidine
action setup, PCR amplification, and amplicon detection were
5’-triphosphate, deoxycytidine 5’-triphosphate, and deoxyguanosine
performed in separated areas.
triphosphate, and 1–5 mL of DNA template, according to the DNA
concentration determined for each sample (50–200 ng/reaction).
The GroEL and AnkA gene amplifications were performed as
previously described, with 7.5 pmol of each primer added in
Fifteen of the 18 dogs (83%) were purebred, and the median age of
each reaction. Conventional PCR was performed under the
dogs was 6 yr (range, 3 mo to 14 yr). There were three intact
following conditions: a single hot-start cycle at 958C for 2 min
females, seven spayed females, three intact males, and five castrated
followed by 55 cycles of denaturation at 948C for 15 sec,
males. The most common presenting complaints were lethargy (13
annealing temperature 628C for 15 sec, and extension at 728C for
dogs) and lameness (10 dogs), with circling, vomiting, and epi-
15 sec. Amplification was completed by an additional cycle at
staxis reported in 1 dog each. Median duration of clinical signs
728C for 1 min, and products were analyzed by 2% agarose gel
before presentation was 3 days (range, 1–14 days) (Table 1).
electrophoresis containing 0.2 mg of ethidium bromide/mL
under ultraviolet light.
Physical Examination
Sixteen of 18 dogs (89%) were febrile (rectal temperature .
IDEXX Protocol
39.28C), and the median temperature of febrile dogs was 40.28C
A conventional PCR for A phagocytophilum was performed
(range, 39.4–40.88C). Ten of 18 dogs (55%) had arthropathy,
according to a published protocol and used the following msp2
which was defined clinically as pain and swelling localized to
primer pair: msp2–3f (59-CCAGCGTTTAGCAAGATAAGAG)
joints on physical examination. Nine dogs (50%) had poly-
and msp2–3r (59-GCCCAGTAACATCATAAGC).16 Amplification
arthropathy, and one dog (5%) had monoarthropathy involving
was performed in a 25 mL volume reaction, containing a 13
the right radiocarpal joint. Lymphadenopathy, left cerebral dys-
buffer mix of 1.5 mM magnesium chloride, 200 mM each of
function, cervical hyperpathia, epistaxis, and acute gastritis were
deoxyadenosine triphosphate, deoxythymidine 5’-triphosphate,
found in one dog each (Table 1).
deoxycytidine 5’-triphosphate, and deoxyguanosine triphosphate,
2.5 U of Taq DNA polymerase, 2 mL DNA template, and 0.5 mM
Complete Blood Count
of each the forward and reverse msp2 primer. The conventional
Seventeen of 18 dogs (94%) were thrombocytopenic (normal range
testing was performed in an Applied Biosystems Thermocyclere
defined as 175,000/mL to 500,000/mL), with platelet counts ranging
with the following program: a single 4 min cycle at 948C followed
from 40,000/mL to 120,000/mL (mean 70,941/mL). Morulae were
by 40 cycles of denaturing at 948C for 30 sec, annealing at 558C
identified in neutrophils during blood smear evaluation in 17 of
for 30 sec, and extension at 728C for 1 min. Once the 40 cycles
18 dogs (94%). Twelve dogs (67%) had a nonregenerative normo-
were complete, the samples were held for 5 min at 728C. The
cytic normochromic anemia (normal range defined as hematocrit
| 47:6 Nov/Dec 2011
Manifestations of Anaplasma phagocytophilum Infection
Physical Examination Results
Dog No.
date (2005)
Duration of
complaint (days)
Cocker spaniel
5 mo
Springer spaniel
4 yr
Female spayed
Basset hound
13 yr
Male neutered
Lameness, lethargy
Labrador retriever
2 yr
Female spayed
Lameness, lethargy
Mixed breed
2 yr
Female spayed
Lameness, lethargy
Border collie
1 yr
Mixed breed
10 yr
14 yr
Female spayed
Male neutered
Lameness, lethargy
Cerebral dysfunction
Shih tzu
6 yr
Female spayed
Golden retriever
5 yr
Female spayed
Lameness, lethargy
Cocker spaniel
8 yr
Male neutered
Cervical hyperpathia
Mixed breed
6 mo
Lameness, lethargy
Shih tzu
6 yr
Male neutered
Labrador retriever
2 yr
Male neutered
Lameness, lethargy
Border collie
3 mo
Polyarthropathy, acute
German shepherd
9 yr
Female spayed
Golden retriever
12 yr
Lameness, lethargy
Polyarthropathy, mild
German shepherd
12 yr
Polyarthropathy, epistaxis
37–55%), with hematocrits ranging from 25.7% to 36.9% (mean
negative to A phagocytophilum (but morulae were observed in
32.5%). Ten dogs (55%) were leukopenic (normal range defined
neutrophils), and then tested negative to all analytes in June 2007
as 5,500–16,900/mL), with total white blood cell counts ranging from
(22 mo later) and June 2008 (34 mo later). Dog 6 initially tested
3,670/mL to 5,420/mL (mean 4,432/mL). Eight dogs (50%) were
positive for A phagocytophilum, and then tested positive for
eosinopenic (normal eosinophil count defined as 100–1,490/mL),
A phagocytophilum in May 2008 (36 mo later). Dog 10 initially
with eosinophil counts ranging from 20/mL to 80/mL (mean 55/mL),
tested positive to A phagocytophilum but tested negative to all
of which one dog had concurrent heartworm disease. Seven dogs
analytes in June 2007 (25 mo later) and tested negative again in
(39%) were lymphopenic (normal lymphocyte count defined as
June 2008 (37 mo later) (Table 2).
700–4,900/mL), with lymphocyte counts ranging from 200/mL to
680/mL (mean 512/mL), and one dog that presented for lameness
Molecular Analysis
and lethargy had a monocytosis (normal monocyte count de-
As part of the retrospective evaluation, blood samples from all
fined as 100–1,400/mL) of 1,430/mL. This dog was concurrently
dogs obtained at the time of presentation tested positive for
infected with D immitis.
A phagocytophilum DNA, targeting either the AnkA gene (n¼14)
or the msp-2 gene (n¼8). Molecular evidence of infection cor-
related with the presence of morulae in all but one case where
Of the 18 dogs in this study, A phagocytophilum antibodies
morulae were not detected. Of the samples tested by both labora-
were found in 11 (61%) dogs at the time of initial presentation;
tories (n¼4), complete agreement was obtained. No molecular evi-
B burgdorferi antibodies were found in 3 (17%); D. immitis antigens
dence of A platys infection was detected in the 14 samples tested by
were found in 1 (5%); and no dog was E canis seroreactive (Table 2).
Four dogs, three of which were A phagocytophilum seroreactive by
SNAP 4Dx at the time of initial presentation, had follow-up
testing using the in-clinic ELISA testa in subsequent years. Dog
Treatment with doxycycline at a dose ranging from 6 to 8 mg/kg PO q
1 initially tested positive for A phagocytophilum but tested negative
12–24 hr for 14–30 days resulted in a positive therapeutic response
for all analytes in August 2007 (28 mo later). Dog 2 initially tested
in all dogs. Three dogs were treated for 14 days, 1 dog for 21 days,
Enzyme-Linked Immunosorbent Assay (ELISA) Results
28 mo later tested negative to all analytes
22 and 34 mo later tested negative to all analytes
Dog no.
11 (61%)
3 (17%)
0 (0%)
1 (5%)
No. positive (%)
Long-term follow up
36 mo later tested positive to Anaplasma
25 and 37 mo later tested negative to all analytes
and 14 dogs for 28–30 days. Treatment duration did not appear to
A phagocytophilum and B burgdorferi.12 The timing of the
influence outcome, as disease manifestations resolved within 24 hr
B burgdorferi infection could not be determined in these dogs,
in three dogs, including the dog with central nervous system
but it might represent a prior transmission event, because up to
dysfunction, 48 hr in seven dogs, 72 hr in four dogs, and 4–6 days
6 wk is generally required for the dog to develop a detectable
in the remaining four dogs (Table 3).
antibody response after tick attachment.21
B burgdorferi alters the host immune response, which can lead
For the dogs described in this study, fever accompanied by ar-
to increased severity of Lyme arthritis.22,23 In the original serosurvey,
thropathy was the most common clinical presentation associated
coinfected dogs were more likely to have lameness, joint pain,
with A phagocytophilum infection. On physical examination, 10 of
and joint effusion than dogs with single infections.12 In this
the 18 dogs had joint pain and effusion. Lyme disease, which is
study, 3 of the 18 dogs were coinfected with B burgdorferi (dogs
caused by B burgdorferi and transmitted by the same Ixodes ticks,
3, 7, and 14) (Table 2). All three presented with arthropathy
also often manifests as a febrile arthropathy.14 One challenge
(Table 1), and the severity of clinical signs did not appear to be
when trying to define the clinical disease that is associated with
worse than the other dogs with arthropathy. The coinfected dogs
canine anaplasmosis is concurrent or sequential transmission of
were thrombocytopenic; however, the degree of thrombocyto-
In mouse studies, coinfection with A phagocytophilum and
A phagocytophilum and B burgdorferi. Although it is clear that
penia was not more severe than dogs without borreliosis. When
dogs in nature can experience frequent and repeated tick
looking at response to doxycycline, 4 of the 18 dogs in this study
infestations, only a small percentage of infected ticks contain
took .3 days for clinical signs to completely resolve (Table 3). It
these two pathogens, yet individual ticks may be infected with
is interesting to note that, of those four dogs, three were coin-
one or both organisms. For example, depending on location,
fected with B burgdorferi. The limited number of coinfected dogs
studies from the Midwest identify B burgdorferi in 16.5–57% and
makes it difficult to determine whether this has true clinical
A phagocytophilum in 3.8–14% of collected I scapularis.17–20 In the
study from which these dogs were selected, disease manifestations
Although neurologic signs were reported for other rickettsial
were more often observed in dogs that were coinfected with
infections in the dog, such as Rocky Mountain Spotted Fever and
| 47:6 Nov/Dec 2011
Manifestations of Anaplasma phagocytophilum Infection
A retrospective Swedish study did not show an association be-
tween A phagocytophilum antibodies and the incidence of neu-
Treatment and Outcomes
Dog No.
rologic disease; however, because numerous tick exposed dogs in
endemic regions become infected without developing clinically
6 mg/kg q 12 hr for 21 days
7 mg/kg q 12 hr for 28 days
Resolution in 2 days
Resolution in 1 day
apparent illness, seroprevalence studies might have less utility
7 mg/kg q 12 hr for 28 days
Resolution in 4 days
7 mg/kg q 12 hr for 14 days
Resolution in 2 days
5 mg/kg q 12 hr for 14 days
Resolution in 2 days
6 mg/kg q 12 hr for 14 days
Resolution in 3 days
cervical hyperpathia was the only abnormality identified during
7 mg/kg q 12 hr for 28 days
Resolution in 5 days
physical and neurologic examinations. The anatomic source of the
7 mg/kg q 12 hr for 28 days
Resolution in 1 day,
with no further
neurologic dysfunction
neck pain was unclear, but meningitis was considered a diagnostic
than DNA-based testing modalities for establishing disease
Dog 11 presented with a history of acute-onset lameness;
possibility. As morulae were observed on the blood smear, cere-
8 mg/kg q 24 hr for 28 days
Resolution in 2 days
8 mg/kg q 24 hr for 28 days
brospinal fluid was not sampled for analysis; thus, pain could have
Resolution in 6 days
8 mg/kg q 12 hr for 28 days
originated from the cervical musculature, cervical joints, or the
Resolution in 3 days
7 mg/kg q 24 hr for 30 days
8 mg/kg q 24 hr for 28 days
Resolution in 2 days
Resolution in 3 days
8 mg/kg q 12 hr for 28 days
Resolution in 5 days
currence of disease reported, indicated that anaplasmosis could
6 mg/kg q 12 hr for 28 days
Resolution in 1 day
cause acute cervical pain in dogs.
7 mg/kg q 12 hr for 28 days
Resolution in 2 days
Bleeding problems were not reported with granulocytic an-
6 mg/kg q 12 hr for 28 days
Resolution in 2 days
8 mg/kg q 12 hr for 28 days
aplasmosis but were described for other rickettsial infections in
Resolution in 3 days
intervertebral discs. The rapid resolution of pain and lameness
after initiation of doxycycline, with no further problems or re-
dogs, such as monocytic ehrlichiosis (E canis), cyclic thrombocytopenia (A platys), and Rocky Mountain Spotted Fever (Rickettsia
ehrlichiosis, relatively few reports described central nervous system
rickettsii).1,26,27 In addition to rickettsial organisms, other infec-
disease resulting from A phagocytophilum infection. Dog 8
tions were associated with canine epistaxis, including aspergil-
presented with acute onset altered mentation and circling to the
losis, bartonellosis, and leshmaniaisis.28–30 Dog 18 presented
left. Cranial nerve examination, peripheral reflexes, postural
with epistaxis, accompanied by fever, joint pain, and thrombo-
reactions, and proprioception were normal. Neurologic exami-
cytopenia. The platelet count was 101,000/mL, which would
nation was consistent with left cerebral dysfunction. It is in-
suggest that factors other than thrombocytopenia were associated
teresting to note that this dog was geriatric (14 yr old) and did
with spontaneous bleeding in this dog. Although the mecha-
not present with fever or evidence of joint pain or effusion.
nism of epistaxis was not determined, rapid and sustained res-
Although fatal illness due to anaplasmosis has not been reported
olution of bleeding occurred after initiation of doxycycline
in dogs, elderly humans are more likely to develop severe and
therapy. Further diagnostic testing was not performed. In the
potentially life-threatening illnesses. The complete blood count
dog, it appears that polymicrobial infections might be an impor-
findings were consistent with anaplasmosis, and morulae were
tant factor to consider diagnostically when examining infectious
easily identified on the blood smear. Despite the advanced age
causes of epistaxis.24,28 Because of the atypical presentation in
of this dog and the neurologic presentation, rapid resolution of
this dog, it was an intriguing possibility that multiple vector
clinical signs occurred within 24 hr after the initiation of
borne infections might have been involved, but in the absence
doxycycline therapy. Further diagnostics, such as brain imaging
of further diagnostic testing, this conclusion remained purely
and cerebrospinal fluid collection, were not performed; thus,
other neurologic disorders could not be definitively ruled out.
The medical history describing a rapid onset illness, in
However, the dog had no history of neurologic disease, and no
conjunction with the temporal relationship of the clinical pre-
further neurologic problems were reported during the remaining
sentation with I. scapularis activity in central Minnesota, sup-
months of his life. Dog 8 was euthanatized 4 mo after presentation
ported an acute disease process rather than acute decompensation
for quality-of-life issues associated with progressive osteoarthritis,
after chronic A phagocytophilum infection. Dogs most often pre-
a pre-existing condition. Although neurologic dysfunction is
sented in the spring and fall with clinical disease at a time that
a seemingly uncommon manifestation of anaplasmosis, rapid
coincided with the highest numbers of I scapularis in the envi-
diagnosis and effective treatment can elicit a favorable outcome.
ronment.31 Of the 273 dogs tested by PCR in the original
serosurvey, no dog tested A phagocytophilum positive during the
cow gave birth to an infected calf, which had clinical signs con-
months of December, January, February, and March.12
sistent with anaplasmosis 13 days after birth.41 The 14 wk time
All of the dogs in this study responded clinically to
doxycycline therapy, and no dog developed subsequent illness
frame between illness of dog 6 and dog 15 made tick transmission
more likely than perinatal transmission.
consistent with granulocytic anaplasmosis. To date, there have
The complete blood count is a useful diagnostic test for
been no clinical reports that document chronic infection with
granulocytic anaplasmosis. Thrombocytopenia is a common he-
A phagocytophilum; however, there was experimental evidence that
matologic abnormality and evaluation of stained blood smears
supported this possibility. Two experimental studies demonstrated
facilitates visualization of granulocyte morulae in acutely infected
presence of A phagocytophilum DNA months after initial infection
dogs. For veterinary clinics that obtain in-house automated
and even after doxycycline therapy.32,33 Importantly, notable
complete blood counts, having a trained technician visually
clinical signs other than thrombocytopenia did not occur after
evaluate blood smears is critically important for the diagnosis of
the acute phase of infection, and viable organisms could not be
anaplasmosis, as well as other tick-borne infections, including
transmitted by blood transfusion from a PCR positive dog to
babesiosis, cytauxzoonosis, ehrlichiosis, and hepatzoonosis.42
a naïve dog. In addition to chronic infection, the possibility of
Morulae provide rapid diagnostic information but are not present
reinfection exists for dogs in highly endemic areas where dogs
in every infection. Another problem with morulae is that based on
may be re-exposed to anaplasmosis over months and years. There
microscopy alone A phagocytophilum morulae are indistinguish-
is evidence to support A phagocytophilum strain variation in
able from E ewingii morulae.10 Dependent on the skill level and
various regions of the United States and internationally.
experience of the technician, other granulocyte changes or arti-
are no data to define whether a dog can be reinfected with ho-
facts (e.g., other cytoplasmic structures or stain precipitate) have
mologous or heterologous A phagocytophilum strains. As in pre-
the potential to be misdiagnosed as A phagocytophilum inclusions.
vious studies, doxycycline was an effective treatment of canine
In cases where no morulae are seen, PCR testing can facilitate
anaplasmosis.5–8 The ideal duration of therapy has not been
an accurate diagnosis, which offers the advantage of differentiating
established, although a 28 day course of doxycycline at 10 mg/kg q
between other infectious and noninfectious diseases that have
24 hr PO has been recommended.36 In human medicine, current
similar hematologic and clinical presentations. In this study, a PCR
recommendations call for a 10 day treatment of doxycycline for
positive test was part of the inclusion criteria to insure that the dogs
granulocytic anaplasmosis.37 In this case series, 3 dogs were
had active infection at the time of presentation. This introduced
treated for 14 days, 1 dog for 21 days, 13 dogs for 28 days, and 1
a bias, because in the original study from which the cases were
dog for 30 days. All dogs responded to treatment in a similar
selected, not all suspected cases of anaplasmosis were PCR positive.
manner. In experimentally infected dogs, A phagocytophilum DNA
For example, there were 16 morulae positive dogs tested by PCR
persisted after doxycycline treatment of 14 and 28 days.33,38 Be-
and 14 were positive.12 The two PCR-negative, morulae-positive
cause dogs were not followed with sequential PCR testing, chronic
dogs might have represented false-negative PCR results or the
or reinfection in asymptomatic recovered dogs in this study could
morulae might have been misdiagnosed. Another potential pitfall
not be determined.
is that clinically normal dogs can test A phagocytophilum PCR
Three puppies, dog 1 (5 mo of age), dog 12 (6 mo of age), and
positive.8,12 In the original serosurvey, seven asymptomatic
dog 15 (3 mo of age) were described in this case series. Dogs 1, 12,
dogs were PCR positive, four of which were also seropositive.12
and 15 were treated with doxycycline for 21, 30, and 28 days,
These dogs likely represented asymptomatic acute infections;
respectively, with no evidence of tooth discoloration. It was in-
however, in the four seropositive PCR positive dogs, asymptomatic
teresting to note that dog 15, an offspring of dog 6, was born 2 wk
chronic infection was a possibility. These data suggested that an
before the onset of illness in the dam. None of the other puppies in
ill PCR positive dog might have clinical disease unrelated to
that litter had problems during nursing, weaning, or when placed
A phagocytophilum infection. Failure to achieve a prompt treatment
in new homes. Dog 6 remained in the same household as the dam
response to doxycycline would tend to rule out canine anaplas-
and developed clinical disease 14 wk after the dam’s illness.
mosis or could suggest coinfection with organisms that were not
Perinatal transmission of A phagocytophilum was reported in
doxycycline responsive.
a human infant who developed symptoms 9 days after birth. A
Seven dogs in this study had negative A phagocytophilum
recent study involving a bitch naturally infected just before
serologic test results at the time of diagnosis, which was not
whelping failed to demonstrate perinatal transmission to the
surprising because acute infection with vector-borne organisms,
puppies.40 In a bovine experimental infection study, an infected
including Anaplasma, Ehrlichia, and Rickettsia spp., can cause
| 47:6 Nov/Dec 2011
Manifestations of Anaplasma phagocytophilum Infection
clinical signs before the dog has a measureable antibody response.
Therefore, treatment decisions should not rely solely on antibody
testing, whether the results are negative or positive. The seronegative dog might have been recently infected, whereas the seropositive dog might have been exposed months earlier and
already eliminated the organism through innate immunity.
Serology results must be considered in the context of physical
examination, hematologic abnormalities, and potentially PCR
results. Long-term serological follow up using the in-clinic serologic screening test was available for four dogs, of which three
tested negative at subsequent testing intervals up to 3 yr. One dog
had a repeatable A phagocytophilum antibody response, which
might have been due to persistent antibody production or due to
re-exposure to A phagocytophilum infected ticks and an amanestic response after repeated transmission events. With the
limited number of dogs and testing time points in this study, it
was not possible to draw conclusions as to how long after infection an individual dog would retain an A phagocytophilum
antibody titer using the ELISA-based screening test. After natural infection in a group of 14 Swedish dogs, IFA titers decreased
below the diagnostic cutoff in some dogs by 4–12 mo after infection; however, variations in immune response in conjunction
with the possibility of reinfection made predictions for clinical
practice problematic.7
As diagnostic capabilities available to practicing veterinarians
continue to increase, the spectrum of clinical disease associated
with canine granulocytic anaplasmosis will be further refined.
Although an acute febrile illness accompanied by lameness or joint
effusion is seemingly the most typical presentation of canine
anaplasmosis, clinicians should be aware of other infrequent or
atypical presentations. Neurologic disease, lymphadenopathy,
vomiting, and epistaxis were the predominant abnormalities in
some dogs in this study. Prompt treatment was associated with
a rapid clinical response and the long-term prognosis appeared to
be excellent. In cases where anaplasmosis is suspected but the dog
does not respond to doxycycline, other disease processes should be
The costs of molecular analysis and retrospective serologic testing
were paid for by IDEXX Laboratories.
SNAP 4Dx Test; IDEXX Laboratories, Inc, Westbrook, ME
LaserCyte Hematology Analyzer; IDEXX Laboratories, Inc,
Westbrook, ME
SNAP 3Dx Test; IDEXX Laboratories, Inc, Westbrook, ME
DNeasy Blood & Tissue Kit; Qiagen, Inc, Valencia, CA
Applied Biosystems Thermocycler; Life Technologies Corp.,
Carlsbad, CA
1. McQuiston JH, McCall CL, Nicholson WL. Ehrlichiosis and related
infections. J Am Vet Med Assoc 2003;223(12):1750–6.
2. Bowman D, Little SE, Lorentzen L, et al. Prevalence and geographic
distribution of Dirofilaria immitis, Borrelia burgdorferi, Ehrlichia
canis, and Anaplasma phagocytophilum in dogs in the United
States: results of a national clinic-based serologic survey. Vet Parasitol 2009;160(1-2):138–48.
3. Madhav NK, Brownstein JS, Tsao JI, et al. A dispersal model for
the range expansion of blacklegged tick (Acari: Ixodidae). J Med
Entomol 2004;41(5):842–52.
4. Brownstein JS, Holford TR, Fish D. A climate-based model predicts
the spatial distribution of the Lyme disease vector Ixodes scapularis
in the United States. Environ Health Perspect 2003;111(9):1152–7.
5. Greig B, Asanovich KM, Armstrong PJ, et al. Geographic, clinical,
serologic, and molecular evidence of granulocytic ehrlichiosis,
a likely zoonotic disease, in Minnesota and Wisconsin dogs. J Clin
Microbiol 1996;34(1):44–8.
6. Kohn B, Galke D, Beelitz P, et al. Clinical features of canine granulocytic anaplasmosis in 18 naturally infected dogs. J Vet Intern
Med 2008;22(6):1289–95.
7. Egenvall AE, Hedhammar AA, Bjöersdorff AI. Clinical features and
serology of 14 dogs affected by granulocytic ehrlichiosis in Sweden.
Vet Rec 1997;140(9):222–6.
8. Granick JL, Armstrong PJ, Bender JB. Anaplasma phagocytophilum
infection in dogs: 34 cases (2000-2007). J Am Vet Med Assoc 2009;
9. Borjesson DL, Simon SI, Tablin F, et al. Thrombocytopenia in
a mouse model of human granulocytic ehrlichiosis. J Infect Dis 2001;
10. Greig B, Armstrong PJ. Canine granulocytotropic anaplasmosis
(A. phagocytophilum infection). In: Greene CE, ed. Infectious
diseases of the dog and cat. 3rd ed. St. Louis: Saunders; 2006:
11. Egenvall A, Bjöersdorff A, Lilliehöök I, et al. Early manifestations of
granulocytic ehrlichiosis in dogs inoculated experimentally with
a Swedish Ehrlichia species isolate. Vet Rec 1998;143(15):412–7.
12. Beall MJ, Chandrashekar R, Eberts MD, et al. Serological and molecular prevalence of Borrelia burgdorferi, Anaplasma phagocytophilum,
and Ehrlichia species in dogs from Minnesota. Vector Borne
Zoonotic Dis 2008;8(4):455–64.
13. Chandrashekar R, Mainville C, Daniluk D, et al. Performance of an
in-clinic test SNAP 4Dx, for the detection of antibodies to canine
granulocytic infection, Anaplasma phagocytophilum (abstract). J Vet
Intern Med 2007;21:626.
14. Greene C, Straubinger R. Borreliosis. In: Greene CE, ed. Infectious
diseases of the dog and cat. 3rd ed. St. Louis: Saunders; 2006:417–35.
15. Birkenheuer AJ, Levy MG, Breitschwerdt EB. Development and
evaluation of a seminested PCR for detection and differentiation of
Babesia gibsoni (Asian genotype) and B. canis DNA in canine blood
samples. J Clin Microbiol 2003;41(9):4172–7.
16. Massung RF, Slater K, Owens JH, et al. Nested PCR assay for
detection of granulocytic ehrlichiae. J Clin Microbiol 1998;36(4):
17. Layfield D, Guilfoile P. The prevalence of Borrelia burgdorfieri
(Spirochaetales: spirochaetaceae) and the agent of human granulocytic ehrlichiosis (Rickettsiaceae: Ehrlichieae) in Ixodes scapularis
(Acari:Ixodidae) collected during 1998 and 1999 from Minnesota.
J Med Entomol 2002;39(1):218–20.
18. Piesman J, Clark KL, Dolan MC, et al. Geographic survey of vector
ticks (Ixodes scapularis and Ixodes pacificus) for infection with the
Lyme disease spirochete, Borrelia burgdorferi. J Vector Ecol 1999;24
19. Michalski M, Rosenfield C, Erickson M, et al. Anaplasma
phagocytophilum in central and western Wisconsin: a molecular
survey. Parasitol Res 2006;99(6):694–9.
20. Steiner FE, Pinger RR, Vann CN, et al. Infection and co-infection
rates of Anaplasma phagocytophilum variants, Babesia spp., Borrelia
burgdorferi, and the rickettsial endosymbiont in Ixodes scapularis
(Acari: Ixodidae) from sites in Indiana, Maine, Pennsylvania, and
Wisconsin. J Med Entomol 2008;45(2):289–97.
21. Liang FT, Jacobson RH, Straubinger RK, et al. Characterization of
a Borrelia burgdorferi VlsE invariable region useful in canine Lyme
disease serodiagnosis by enzyme-linked immunosorbent assay.
J Clin Microbiol 2000;38(11):4160–6.
22. Thomas V, Anguita J, Barthold SW, et al. Coinfection with Borrelia
burgdorferi and the agent of human granulocytic ehrlichiosis alters
murine immune responses, pathogen burden, and severity of Lyme
arthritis. Infect Immun 2001;69(5):3359–71.
23. Zeidner NS, Dolan MC, Massung R, et al. Coinfection with Borrelia
burgdorferi and the agent of human granulocytic ehrlichiosis suppresses IL-2 and IFN g production and promotes an IL-4 response
in C3H/HeJ mice. Parasite Immunol 2000;22(11):581–8.
24. Bakken JS, Dumler JS. Clinical diagnosis and treatment of human
granulocytotropic anaplasmosis. Ann N Y Acad Sci 2006;1078:
25. Jäderlund KH, Egenvall A, Bergström K, et al. Seroprevalence of
Borrelia burgdorferi sensu lato and Anaplasma phagocytophilum in
dogs with neurological signs. Vet Rec 2007;160(24):825–31.
26. Wilson J. Ehrlichia platys in a Michigan dog. J Am Anim Hosp Assoc
27. Greene C, Breitschwerdt E. Rocky mountain spotted fever, murine
typhuslike disease, rickettsialpox, typhus, and Q fever. In: Greene
CE, ed. Infectious diseases of the dog and cat. 3rd ed. St. Louis:
Saunders; 2006:232–45.
28. Breitschwerdt EB, Hegarty BC, Maggi R, et al. Bartonella species as
a potential cause of epistaxis in dogs. J Clin Microbiol 2005;43(5):
29. Jüttner C, Rodríguez Sánchez M, Rollán Landeras E, et al. Evaluation of the potential causes of epistaxis in dogs with natural visceral
leishmaniasis. Vet Rec 2001;149(6):176–9.
| 47:6 Nov/Dec 2011
30. Mylonakis ME, Saridomichelakis MN, Lazaridis V, et al. A retrospective study of 61 cases of spontaneous canine epistaxis (1998 to
2001). J Small Anim Pract 2008;49(4):191–6.
31. Daniels TJ, Falco RC, Curran KL, et al. Timing of Ixodes scapularis
(Acari: Ixodidae) oviposition and larval activity in southern New
York. J Med Entomol 1996;33(1):140–7.
32. Egenvall A, Lilliehöök I, Bjöersdorff A, et al. Detection of granulocytic Ehrlichia species DNA by PCR in persistently infected dogs.
Vet Rec 2000;146(7):186–90.
33. Alleman A, Chandrashekar R, Beall M, et al. Experimental inoculation of dogs with a human isolate (Ny18) of Anaplasma
phagocytophilum and demonstration of persistent infection following doxycycline therapy (abstract). J Vet Intern Med 2006;
34. Barbet AF, Lundgren AM, Alleman AR, et al. Structure of the
expression site reveals global diversity in MSP2 (P44) variants
in Anaplasma phagocytophilum. Infect Immun 2006;74(11):
35. Foley J, Nieto NC, Madigan J, et al. Possible differential host tropism
in Anaplasma phagocytophilum strains in the Western United States.
Ann N Y Acad Sci 2008;1149:94–7.
36. Neer TM, Breitschwerdt EB, Greene RT, et al; American College
of Veterinary Internal Medicine. Consensus statement on ehrlichial
disease of small animals from the infectious disease study group
of the ACVIM. J Vet Intern Med 2002;16(3):309–15.
37. Wormser GP, Dattwyler RJ, Shapiro ED, et al. The clinical assessment, treatment, and prevention of lyme disease, human granulocytic anaplasmosis, and babesiosis: clinical practice guidelines by the
Infectious Diseases Society of America. Clin Infect Dis 2006;43(9):
38. Alleman A, Wamsley H, Abbott J, et al. Experimental Anaplasma
phagocytophilum infection of dogs by intravenous inoculation
of human and canine isolates and treatment with doxycycline
(abstract). Vet Pathol 2007;44:19.
39. Dhand A, Nadelman RB, Aguero-Rosenfeld M, et al. Human
granulocytic anaplasmosis during pregnancy: case series and literature review. Clin Infect Dis 2007;45(5):589–93.
40. Plier ML, Breitschwerdt EB, Hegarty BC, et al. Lack of evidence
for perinatal transmission of canine granulocytic anaplasmosis
from a bitch to her offspring. J Am Anim Hosp Assoc 2009;45(5):
41. Pusterla N, Braun U, Wolfensberger C, et al. Intrauterine infection
with Ehrlichia phagocytophila in a cow. Vet Rec 1997;141(4):
42. Walker D. Peripheral blood smears. In: Cowell R, Tyler R, Meinkoth J,
et al, eds. Diagnostic cytology and hematology of the dog and cat. 3rd ed.
St. Louis: Mosby; 2007:254–83.