Evaluation of the presence of selected viral and bacterial nucleic acids in pericardial samples from dogs with or without idiopathic pericardial effusion

Many viruses have been identified in pericardial fluid and in tissue samples from humans with pericarditis by means of molecular diagnostics. In canine idiopathic pericardial effusion there is as yet no conclusive evidence to support the involvement of an infectious agent. This study was designed to investigate a possible relationship between idiopathic pericardial effusion in dogs and viruses most commonly encountered in humans affected with viral pericarditis. Coxsackievirus B3 RNA, influenza virus type A RNA, human adenovirus type 2 DNA, human cytomegalovirus DNA, and parvovirus B19 DNA were investigated using PCR on pericardial effusion samples and pericardial tissue specimens collected from 14 dogs with idiopathic pericardial effusion. PCR was also used to test for two bacteria, Borrelia burgdorferi and Chlamydia pneumoniae. The same microorganisms were also looked for in pericardial effusions or pericardial washes from 10 dogs with neoplastic pericardial effusion, and in samples collected from 10 dogs which died of a non-cardiac disease. One pericardial effusion sample from a dog with the idiopathic form of the disease tested positive for influenza virus type A and sequencing of the amplicon confirmed the PCR result. In another dog from the same group a cytomegalovirus was detected by PCR in the effusion, but sequencing showed this to be a false-positive result. The genomes of the microorganisms investigated were not detected in neoplastic effusions or pericardial washes. The results indicate that viral and bacterial DNA/RNA of relevance for human pericarditis is rare in pericardial samples from dogs with idiopathic pericardial effusion. The finding of influenza type A viral RNA in pericardial fluid from one dog with the idiopathic form of the disease warrants further investigation.     

Detection of Borrelia burgdorferi DNA in tissues from dogs with presumptive Lyme borreliosis

OBJECTIVE: To develop a quantitative PCR assay for detection of Borrelia burgdorferi DNA in formalin-fixed, paraffin-embedded tissues; compare results of this assay with results of immunohistochemical staining of tissues from seropositive dogs; and determine whether B burgdorferi DNA could be detected in renal tissues from dogs with presumptive Lyme nephritis. DESIGN: Cohort study. SAMPLE POPULATION: Archived tissue samples from 58 dogs. PROCEDURES: A quantitative PCR assay was performed on formalin-fixed, paraffin-embedded tissue sections from the dogs. Results were compared with results of immunohistochemical staining, B burgdorferi serostatus, clinical signs, and necropsy findings. RESULTS: 38 dogs were classified as having positive or equivocal results for Lyme borreliosis, and 20 were classified as having negative results on the basis of clinical signs, serologic findings, and pathologic abnormalities. Borrelia burgdorferi DNA was amplified from tissue samples from only 4 (7%) dogs, all of which had been classified as having positive or equivocal results for Lyme borreliosis and had signs of presumptive Lyme nephritis. Results of PCR assays of renal tissue were positive for only 1 dog, and there was no agreement between results of immunohistochemical staining (ie, detection of B burgdorferi antigen) and results of the PCR assay (ie, detection of B burgdorferi DNA) for renal tissues. CONCLUSIONS AND CLINICAL RELEVANCE: Results indicated that detection of B burgdorferi DNA in formalin-fixed, paraffin-embedded tissues is feasible, but that intact B burgdorferi DNA is rarely found in tissues from naturally infected dogs, even tissues from dogs with presumptive Lyme borreliosis. Further, findings support the contention that Lyme nephritis may be a sterile, immune complex disease.     

Failure to identify an association between serologic or molecular evidence of Bartonella infection and idiopathic rhinitis in dogs

OBJECTIVE: To determine whether infection with or exposure to Bartonella spp was associated with idiopathic rhinitis in dogs. DESIGN: Case-control study. ANIMALS: 44 dogs with idiopathic nasal discharge and 63 age- and weight-matched control dogs without nasal discharge and no clinical signs of bartonellosis. Procedures-Serum was tested for antibodies against Bartonella henselae and Bartonella vinsonii subsp berkhoffii with indirect fluorescent antibody assays. Blood was tested for Bartonella DNA with a PCR assay. RESULTS: Results of the antibody and PCR assays were negative for all 44 dogs with idiopathic nasal discharge. One control dog had antibodies against B henselae; a second control dog had positive PCR assay results. We did not detect a significant association between assay results and group designation. CONCLUSIONS AND CLINICAL RELEVANCE: The present study failed to confirm an association between idiopathic rhinitis and exposure to or infection with Bartonella spp in dogs. Findings do not rule out the possibility that Bartonella infection may cause nasal discharge in some dogs, but the failure to find any evidence of exposure to or infection with Bartonella spp in dogs with idiopathic nasal discharge suggested that Bartonella infection was not a common cause of the disease.     

Plasmapheresis in five dogs with systemic immune-mediated disease

Five dogs with signs referable to systemic immune-mediated disease, four with systemic lupus erythematosus, and one with probable lupus myopathy were treated with plasmapheresis in combination with low-dose immunosuppressive drug therapy. Previous treatment with conventional dosages of prednisone was not satisfactory and was associated with adverse side effects. Two dogs had short-term responses to combined therapy, and 3 dogs had sustained responses. Clinical remission was associated with normalization of serum complement levels and decreases in antinuclear antibody titers. Toxicosis potentially related to plasma component depletion was observed in 2 dogs. Acute clinical illness and disease states refractory to conventional immunosuppressive therapy should be considered indications for plasmapheresis.     

Perinuclear antineutrophil cytoplasmic autoantibodies in dogs infected with various vector-borne pathogens and in dogs with immune-mediated hemolytic anemia

Objective-To determine the prevalence of perinuclear antineutrophil cytoplasmic autoantibodies (pANCA) in dogs with confirmed or suspected immune-mediated hemolytic anemia (IMHA) or dogs infected with various vector-borne pathogens, including Rickettsia rickettsii, Bartonella henselae, Bartonella vinsonii subsp berkhoffii, Ehrlichia canis, Borrelia burgdorferi, and Leishmania infantum. Animals-55 dogs with confirmed or suspected IMHA, 140 dogs seroreactive for vector-borne pathogens, and 62 healthy dogs and dogs seronegative for vector-borne pathogens. Procedures-Samples were allocated to subgroups on the basis of the health status of the dogs and the degree of seroreactivity against various vector-borne pathogens. Serum samples were tested retrospectively via indirect immunofluorescence assay to determine pANCA status. Results-26 of 55 (47%) dogs with confirmed or suspected IMHA and 67 of 140 (48%) dogs seroreactive for vector-borne pathogens had positive results when tested for pANCA. Serum samples with the highest antibody concentrations against L infantum antigen had the highest proportion (28/43 [65%]) that were positive for pANCA. One of 20 (5%) dogs seronegative for tick-borne pathogens and 8 of 22 (36%) dogs seronegative for L infantum had positive results for pANCA. One of 20 (5%) healthy dogs had serum antibodies against pANCA. Conclusions and Clinical Relevance-pANCA were detected in a high percentage of dogs with IMHA and vector-borne infectious diseases. Therefore, pANCA may be a relatively nonspecific marker for dogs with inflammatory bowel disease, although they could represent a biomarker for immune-mediated diseases and infections.     

Papular dermatitis due to Leishmania spp. infection in dogs with parasite-specific cellular immune responses

Papular dermatitis due to Leishmania spp. infection was diagnosed in three boxers and two Rottweilers with Leishmania-specific cellular immunity. Diagnosis was based on histological and immunohistochemical examination of papules in four dogs and on cytological examination in one dog. Serum protein electrophoresis was within reference ranges and low antibody levels to Leishmania infantum were detected. Delayed-type hypersensitivity (DTH) reaction to leishmanin was evaluated before treatment in three dogs with positive results. After meglumine antimoniate therapy for 3 to 4 weeks and allopurinol treatment for 6 to 10 months, all dogs were clinically normal, had positive DTH reactions to leishmanin and reduced antibody titres. In conclusion, we suggest that this previously unreported cutaneous presentation of canine leishmaniosis appears to be associated with specific immunocompetence and, consequently, with a favourable prognosis.     

Evaluation of clinicopathologic features, response to treatment, and risk factors associated with idiopathic neutropenia in dogs: 11 cases (1990-2002)

OBJECTIVE: To evaluate the clinicopathologic features, response to treatment, and risk factors associated with idiopathic neutropenia in dogs. DESIGN: Retrospective case series. ANIMALS: 11 dogs. PROCEDURES: Medical records of dogs with idiopathic neutropenia were reviewed. Signalment, history, clinical signs, and response to treatment were recorded and compared with that in dogs with neutropenia attributable to known causes and to dogs without neutropenia (controls). RESULTS: Compared with dogs with neutropenia attributable to known causes, dogs with idiopathic neutropenia had lower neutrophil counts and were younger. When compared with control dogs, age < 4 years was identified as a risk factor for developing idiopathic neutropenia. In all dogs with idiopathic neutropenia, remission of neutropenia occurred within 18 days after administration of prednisone (2 to 4 mg/kg [0.9 to 1.8 mg/lb], PO, daily) and no serious complications or infections developed. CONCLUSIONS AND CLINICAL RELEVANCE: An immune-mediated pathogenesis should be considered for dogs with idiopathic neutropenia in which the cause is not known. Severe neutropenia and young age were significantly associated with idiopathic neutropenia in dogs. Prognosis appeared to be excellent with prednisone treatment.     

Suppurative, nonseptic polyarthropathy in dogs

The goals of this study were to determine the historical, physical examination, and clinicopathologic findings in dogs with suppurative, nonseptic polyarthropathy and to identify concurrent disorders associated with this syndrome. Medical records of 52 dogs with cytologic evidence of suppurative inflammation in two or more joints were examined retrospectively. Age of dogs was 4.8 years (median, range: 0.5-12 years). There was no clear breed or sex predilection, but most were large-breed dogs (body weight > or = 20 kg [44.4 lbs] in 40/52). Body temperature was 103.0 degrees F (39.4 degrees C) (median, range: 100.0-105.9 degrees F), with 29 of 52 dogs having a body temperature > or = 103 degrees F (39.4 degrees C). Lameness was identified in 42 of 52 dogs. Erosive changes were found in only 1 of 37 dogs that had radiography performed. A clear underlying disease process was not identified in 34 of 52 dogs. Seven dogs had evidence of infectious or inflammatory processes at extra-articular sites; 4 dogs were diagnosed with systemic lupus erythematosus (SLE); 2 dogs had gastrointestinal disease; 2 dogs had been vaccinated within 1 month before onset of polyarthritis; 1 dog had cancer; 1 dog had polyarthritis and meningitis; and 1 dog had erosive polyarthritis. Of the 44 dogs tested, 25 had antibodies to Borrelia burgdorferi, detected by an ELISA assay, which was significantly greater than the general hospital population (P = .007). Antibodies against Rickettsia rickettsiae and Ehrlichia canis were not definitively identified in the sera of any dog tested in this study (45 and 44 dogs, respectively). We conclude that an underlying disease process is not identified in most cases of suppurative polyarthropathy in dogs and that intestinal disease, neoplasia, and SLE are uncommon causes of polyarthritis. While seropositivity against the causative agent of Lyme disease was common and possibly a cause of polyarthritis in some dogs of our study, evidence of other vector-borne infection was not identified.     

Hemophagocytic syndrome in dogs: 24 cases (1996-2005)

OBJECTIVE: To determine the frequency, potential causes, and clinical and clinicopathologic features of hemophagocytic syndrome in dogs. DESIGN: Retrospective study. ANIMALS: 24 client-owned dogs. PROCEDURES: Records for dogs in which diagnostic bone marrow specimens (including an aspiration smear and core biopsy material) were obtained from 1996 to 2005 were reviewed. Inclusion criteria were presence of bicytopenia or pancytopenia in the blood and > 2% hemophagocytic macrophages in the bone marrow aspirate. RESULTS: Of 617 bone marrow specimens evaluated, evidence of hemophagocytic syndrome was detected in 24 (3.9%). The Tibetan Terrier breed was overrepresented among dogs with hemophagocytic syndrome. Clinical signs associated with hemophagocytic syndrome included fever, icterus, splenomegaly, hepatomegaly, and diarrhea. Hemophagocytic syndrome was associated with immune-mediated, infectious, and neoplastic-myelodysplastic conditions and also occurred as an idiopathic condition. Overall, dogs with infection-associated hemophagocytic syndrome had better 1-month survival rates than dogs with immune-associated and idiopathic hemophagocytic syndrome. CONCLUSIONS AND CLINICAL RELEVANCE: Results indicated that hemophagocytic syndrome may occur more frequently in dogs than has previously been suspected on the basis of the paucity of reported cases. Although most dogs had definable underlying disease conditions, idiopathic hemophagocytic syndrome was also identified. Hemophagocytic syndrome of any cause is potentially life-threatening; however, the prognosis should be adjusted on the basis of the associated disease process and potential for successful treatment.     

Intravenous Human Immunoglobulin for the Treatment of Immune-Mediated Hemolytic Anemia in 13 Dogs

Intravenous immunoglobulin (IVGG) was administered to 13 of 37 dogs with immune-mediated hemolytic anemia. All dogs received concurrent prednisone therapy, 14 dogs also received cyclophosphamide; and a single dog each received cyclosporine, azathioprines, and danazol. Dogs that responded to prednisone therapy without IVGG generally did so within 7 days (mean ± standard deviation = 5.6 ± 2.9 days). Intravenous immunoglobulin was administered after 10.4 ± 6.6 days of prednisone therapy as an intravenous infusion of 0.5 g/kg (range 0.25 to 0.73 g/kg). Eleven dogs received a single treatment, 2 dogs each received 2 treatments. No relevant adverse effects were noted. Eleven dogs had an increase in PCV of at least 4% 2.2 ±1.5 days after IVGG infusion. In 10 of these dogs, the PCV continued to increase until the time of hospital discharge. One responder died 1 hour after the increase in PCV, 1 dog was euthanized within 24 hours of IVGG administration, and 1 dog had no response over a period of 13 days. Results of this study suggest that IVGG therapy may be of value in dogs with immune-mediated hemolytic anemia that do not respond within 7 days of appropriate corticosteroid therapy.     

Relationship of toxoplasma infections to other diseases in dogs

Toxoplasma infection as detected by the immunofluorescent antibody (IFA) test was found in 13.2% of 448 dogs examined at the Purdue University Small Animal Clinic. Only titres of 1/64 and above were considered positive, but lesser titers were encountered more frequently. The highest titer observed was 1/1024 in 2 dogs one of which had idiopathic epilepsy and the other had dirofilariasis as well as hip dysplasia. The incidence of positive sera from male and female dogs were 11.7 and 14.5%, respectively. The highest frequency of infection occurred in young dogs up to 4 years of age and the lowest was in older animals. The presence of the high toxoplasmosis IFA titers in the sera of dogs seems to be associated with other disease conditions. Stress such as joint dislocations, fractures, helminth parasitism, and concomitant bacterial and viral infections, etc., was observed in 62.7% of the positive cases. Presence of tumors and renal disease conditions were associated with 28.8 and 5.2%, respectively, of the positive sera.     

Development of canine systemic lupus erythematosus model

The purpose of this study was to develop a model for canine systemic lupus erythematosus. Systemic lupus erythematosus (SLE) is a systemic autoimmune syndrome defined by clinical and serological features, including arthritis, glomerulonephritis, dermatitis and autoantibodies. SLE was induced in eight normal dogs by immunization with heparan sulphate, the major glycosaminoglycan of the glomerular basement membrane. All the heparan sulphate-immunized dogs showed mild-to-moderate levels of proteinuria and skin disease. Cutaneous signs associated with SLE including alopecia, erythema, crusting, scaling and seborrhoea were observed. Immunohistological examination of the skin lesions revealed deposition of immunoglobulin M and complement in the dermal-epidermal junction. Three of eight dogs showed lameness. The antinuclear antibody tests were positive with the antibody titres higher than 1:128. Therefore, this experimental SLE model could be useful for studying immune-mediated skin disease and autoimmunity.     

Clinical and serologic studies of canine borreliosis

During 1984 and 1985, blood samples were obtained from 271 dogs that were suspected of having borreliosis. The dogs lived in areas known to be infested with ticks and had been examined because of limb/joint disorders or for unknown illnesses marked by fever, anorexia, or fatigue. Lameness had been the most frequently reported clinical manifestation. Analyses of serum specimens, by an indirect fluorescent antibody (IFA) method or by an ELISA, detected antibodies to Borrelia burgdorferi, the etiologic agent of borreliosis in dogs and of Lyme disease in human beings. Antibody to B burgdorferi was detected in 76.3% of 114 specimens from dogs living in the lower Hudson Valley region of New York State (predominantly Westchester County), in 66.5% of 155 specimens from dogs from southern Connecticut, and in single specimens from dogs from Rhode Island and California. Geometric mean antibody titers peaked during the winter. Results of IFA tests and ELISA were in agreement, but the latter method yielded less variable results, had greater sensitivity, and was more easily standardized. Five dogs from New York State and Connecticut seropositive to B burgdorferi had developed kidney disorders during or after episodes of intermittent lameness. Application of murine monoclonal antibody in an IFA procedure verified the presence of B burgdorferi in renal cortical tissues from one dog.     

Human adenovirus antibody in sera of normal and tumour-bearing dogs.

Serum specimens from 42 normal dogs and 42 with untreated malignant tumours were assayed for the presence of antibodies to human adenovirus types 5, 21 and 31 and to infectious canine hepatitis (ICH) virus. Radioimmunoassays using human adenovirus antigens showed that 71 per cent (30/42) of all dogs with tumours, but only 19 per cent (8/42) of all normal dogs, were positive for human adenovirus antibody. Canine sera reactive with antigens of one human adenovirus type in radioimmunoassays were also reactive with antigens of the other two types. Dogs bearing malignant lymphoma or squamous cell carcinoma tumours had higher levels of antibody against adenovirus type 5 antigens. Human adenovirus type 5 was neutralised by sera from four tumour-bearing and two normal dogs, while sera from two normal and five tumour-bearing dogs were positive in immunodiffusion tests with human adenovirus antigens. Levels of ICH antibody in sera of normal adult dogs and adult dogs with tumours were not markedly different when measured by radioimmunoassays. Likewise, sera from these two groups of dogs had similar ranges of ICH neutralising antibody titres. In contrast, levels of ICH antibody detected by the serological assays in sera from non-pet, non-vaccinated pups were either markedly low or absent. Possible explanations for the observed increased levels of human adenovirus antibody in sera of tumour-bearing canine pets are discussed.     

Sterile nodular panniculitis and pansteatitis in three weimaraners

The clinical and pathological findings in three unrelated weimaraners with pyrexia and multiple subcutaneous nodules are reported. Abdominal pain was an additional feature in two of the dogs and clinical investigations revealed inflammation of subcutaneous, mesenteric and falciform fat. Histopathological findings were consistent with pansteatitis. In the third dog, lesions were apparently limited to the subcutis and, hence, a diagnosis of nodular panniculitis was made. Microbiological examination of tissues was negative in all dogs, and there was no evidence of pancreatic disease. This report thus describes a presumed sterile and idiopathic panniculitis/pansteatitis complex in weimaraner dogs. Although the aetiology is unknown, this may represent an immune-mediated disorder.     

Influence of drug treatment on survival of dogs with immune-mediated hemolytic anemia: 88 cases (1989-1999)

OBJECTIVE: To evaluate association of various treatments for immune-mediated hemolytic anemia with survival to discharge in dogs. DESIGN: Retrospective cross-sectional analysis. ANIMALS: 88 dogs with idiopathic immune-mediated hemolytic anemia. PROCEDURE: Medical records of dogs with immune-mediated hemolytic anemia treated between August 1989 and August 1999 were examined. Survival to discharge, PCV at referral, autoagglutination, and drug treatment and dosage were recorded. RESULTS: Treatments included administration of prednisone, dexamethasone, azathioprine, danazol, cyclosporine, cyclophosphamide, bovine hemoglobin solution, and human immunoglobulin. Overall mortality rate was 50.5%. Significant associations with death were not detected for use of azathioprine, cyclosporine, danazol, or human immunoglobulin. A significant difference in mortality rate was not detected between use of multiple immunosuppressive drug treatments and use of single immunosuppressive drugs. Use of cyclophosphamide and bovine hemoglobin solution were associated with significant increases in relative risk of death CONCLUSIONS AND CLINICAL RELEVANCE: Results suggest that use of cyclophosphamide and bovine hemoglobin solution in treatment of idiopathic immune-mediated hemolytic anemia may be associated with increased risk of death.     

Borreliosis in dogs from southern Connecticut

Blood samples were obtained from dogs in tick-infested regions of southern Connecticut to assess canine exposure to Borrelia burgdorferi, the etiologic agent of Lyme disease in human beings. An indirect fluorescent antibody test detected immunoglobulin (Ig)M antibodies at titers of 1:64 to 1:512 in 22 of 84 serum samples previously shown to be positive with a polyvalent rabbit anti-dog total Ig conjugate. Analyses of paired serum samples from 20 seropositive dogs revealed temporal differences in titers; changes occurred during brief (1 month) or extended (greater than 4 years) sampling periods. Clinical records for 52 seropositive dogs indicated a history of intermittent lameness in 19 of these. Limb/joint disorders typically developed in dogs without IgM antibodies, suggesting manifestation during later phases of illness. A microscopic-agglutination test was used to assess cross reactivity between B burgdorferi and 20 serovars of Leptospira interrogans and biflexa. Analyses of 63 dog serum specimens with antibodies to B burgdorferi and a series of reference rabbit sera revealed minor antigenic relatedness. There was geographic clustering of dogs with antibodies to B burgdorferi in areas of south-central and southeastern Connecticut, where human Lyme disease also occurs.     

Immunopathogenesis of canine aural haematoma

Data are presented from 15 dogs with aural haematoma. The series included six Labrador retrievers and four golden retrlevers and the mean age was 8.0 ± 3.02 years. Five dogs had evidence of pruritic skin disease and five further cases had other concurrent disease. Haematoiogy and serum biochemistry were normal in 12 and 13 of the 15 dogs, respectively. Ail dogs were Coombs' negative and serum antinuclear antibody had negative or low titres in ail the 11 cases tested. Histopathological examination of biopsies from the affected ears revealed variable degrees of erosion of auricular cartilage with fibrovascular granulation tissue filling the cartilage defects. There was minimal perichondrai inflammation. The biopsies were studied by immunohistochemlstry for deposition of lmmunoglobulln G (IgG), immunogiobulln M (IgM) and complement C3. In one dog there was basement membrane zone deposition of IgG and in another there was focal interepithellai deposition of both igG and IgM. The findings of this study do not support an autoimmune pathogenesis for canine aural haematoma, but suggest that an early immunological event may underlie the observed cartilage erosion.     

Azathioprine therapy for acquired myasthenia gravis in five dogs.

Five dogs with acquired myasthenia gravis (MG), verified via positive serum acetylcholine (ACh) receptor antibody concentrations, were treated with a drug protocol including azathioprine (AZA). Four of the five dogs were concurrently treated with pyridostigmine. Azathioprine was used as the sole immunosuppressive agent in four dogs. One dog was temporarily treated with a combination of an immunosuppressive dose of prednisone and AZA, then maintained on AZA as the sole immunosuppressive drug. Three patients experienced complete remission of clinical signs within three months of therapy. In the four dogs for which follow-up serum ACh receptor antibody concentrations were available, initial versus final concentrations decreased substantially (81%), coincident with clinical improvement. One dog died suddenly due to a suspected myasthenic crisis before attaining the target dose of AZA. Two of the four surviving dogs were euthanized approximately one and seven years after diagnosis. One of these two dogs was euthanized because of a rib osteosarcoma, and the other dog was euthanized because of paraparesis of undetermined cause. The remaining two dogs were alive and doing well at the time of final follow-up evaluation, approximately six months and one year after diagnosis. The use of AZA as a therapeutic agent for acquired canine MG has not been investigated. The cases presented in this report suggest a potentially important role for AZA in the treatment of acquired MG in dogs.     

Anaplasma phagocytophilum in dogs in Germany

A total number of 111 dogs were included in the present prospective study investigating the prevalence of Anaplasma phagocytophilum in dogs in Germany. Dogs were divided into two groups. Dogs of group 1 (n = 49) showed clinical and/or haematological signs seen in infections with A. phagocytophilum, whereas those of group 2 (n = 62) did not have any evidence of anaplasmosis. For each dog, an A. phagocytophilum 16S rRNA-nested polymerase chain reaction (PCR) of ethylenediaminetetraacetic acid (EDTA)-anticoagulated whole blood analysis, a microscopic evaluation of a buffy coat and a serum indirect fluorescent antibody test (IFAT) were performed. Forty-eight seroreactive dogs were identified altogether, which amounts to an overall point prevalence of 43.2%. There was no significant difference between the seroreactivity to A. phagocytophilum antigens among group 1 (44.9%) and 2 (41.9%) (P > 0.5). Seven dogs (6.3%) had positive PCR results. All of them were seroreactive. Six belonged to group 1. Morulae in neutrophilic granulocytes were found in two dogs of group 1 but in none of group 2. Both dogs were seroreactive. Very high antibody titres (> or =1:1024) were detected significantly more frequently in dogs with clinical signs attributable to infection with A. phagocytophilum (group 1) than in those without (group 2) (P < 0.001). There was no significant correlation of overall positives or antibody titres to age, breed, sex, or whether the dogs were family or working dogs. Dogs with high tick infestation were significantly more often seroreactive to A. phagocytophilum than those with no or low tick infestation (P = 0.007). In conclusion, there seems to be a high risk of infection with A. phagocytophilum in Germany. Results of this study suggest that severe illness solely caused by A. phagocytophilum may be possible although definitive evidence does not exist. Very high antibody titres (>1:1024) may be associated with clinical anaplasmosis.