Diagnostic accuracy of ultrasonography to detect hepatic and splenic lymphomatous infiltration in dogs and cats.

OBJECTIVES: To determine the diagnostic accuracy of ultrasonography in the detection of lymphomatous infiltration of the liver and spleen in a population of dogs and cats with lymphoma. To determine if specific ultrasonographic features of the liver and spleen in dogs are associated with lymphomatous infiltration or a specific immunophenotype of multi-centric lymphoma. MATERIALS AND METHODS: A blinded retrospective evaluation of ultrasonographic images of the liver and/or spleen in dogs and cats with cytologically or histologically confirmed lymphoma was performed by two board-certified veterinary radiologists. RESULTS: A total of 161 animals met the inclusion criteria, comprising 132 dogs and 29 cats. Ultrasonography had a sensitivity, specificity, accuracy, positive predictive value and negative predictive value of 16.7%, 91.0%, 55.9%, 62.5% and 55.0% for the detection of lymphomatous infiltration of the liver, and 73.1%, 93.9%, 82.6%, 93.4% and 74.7% for the spleen. In dogs, an ultrasonographically normal liver was associated with not having lymphomatous infiltration, leopard-spotted splenic parenchyma and splenomegaly were independently associated with lymphomatous infiltration and leopard-spotted splenic parenchyma was also associated with the B cell immunophenotype of multi-centric lymphoma. CLINICAL SIGNIFICANCE: Ultrasonography of the spleen and liver is specific but not sensitive in the detection of lymphomatous infiltration. A leopard-spotted splenic parenchyma in dogs is highly specific for lymphomatous infiltration and in this population predicted a specific immunophenotype of multi-centric lymphoma.

Clinical and diagnostic imaging findings in dogs with atrial appendage aneurysm: 7 cases (2014-2020).

INTRODUCTION/OBJECTIVE: Atrial appendage aneurysm in dogs is a rare condition and has not been well described. The aim of this study is to describe clinical and diagnostic imaging findings, especially computed tomography (CT), of atrial appendage aneurysms in dogs. ANIMALS: Seven client-owned dogs with a presumptive diagnosis of left or right atrial appendage aneurysm. MATERIALS AND METHODS: Retrospective study. Medical records were searched to identify dogs with a presumptive diagnosis of left (LAAA) or right atrial appendage aneurysm (RAAA). Signalment, history, examination findings, diagnostic test results, and imaging procedures were reviewed. Archived diagnostic images were retrieved and evaluated by two board-certified radiologists and a board-certified cardiologist. Data analysis was descriptive. RESULTS: Six cases were diagnosed with RAAA and one with LAAA with a median age of 8 years. Five affected dogs were small to medium-breed male dogs. All dogs underwent a thoracic CT examination for various reasons and all cases of RAAA were incidental findings. CT was useful to identify and assess the atrial appendage aneurysm, as well as neighboring structures, although possible pericardial defects could not be visualized. Five dogs had a concurrent echocardiographic examination, which successfully identified the LAAA and two RAAA. CONCLUSIONS: This case series described the clinical and CT findings in seven dogs with atrial appendage aneurysm, as well as echocardiographic findings in five of these cases. Right atrial appendage aneurysms appear to be mainly incidental findings. CT seems to be more sensitive than echocardiography in the detection of atrial appendage aneurysm.

Preliminary investigation of serum cardiac troponin I in dogs with acute ischaemic stroke.

OBJECTIVES: To describe the incidence of elevated serum cardiac troponin I in dogs with acute ischaemic strokes, to evaluate its prognostic value in these patients and characterise a possible relationship between cardiac troponin I elevation in dogs with ischaemic strokes and underlying cardiac dysfunction. MATERIALS AND METHODS: Prospective study of 18 dogs with acute ischaemic stroke diagnosed by MRI of the brain. Serum cardiac troponin I concentration, trans-thoracic echocardiography and six-lead electrocardiography were performed and findings were compared between dogs with good and poor outcome. RESULTS: Serum cardiac troponin I was increased in 17 dogs (median 0.95 ng/mL; range 0.146 to 153). Focal hyperechoic regions of myocardium were visible in two dogs using trans-thoracic echocardiography and presumed to represent acute infarcts. A significant association was found between cardiac troponin I and creatinine concentrations. No difference in cardiac troponin I concentrations was detected between dogs that experienced good and poor outcomes. Clinically important cardiac dysfunction was identified in two dogs. CLINICAL SIGNIFICANCE: Cardiac troponin I is commonly elevated in patients diagnosed with acute ischaemic stroke but, in this small study population, did not have prognostic value. Larger studies (recruiting a study population of 98 dogs for a power of 0.8 and a 0.05 alpha/critical value) would aid in further investigation of these preliminary results.

A randomised controlled comparison of aspiration and non-aspiration fine-needle techniques for obtaining ultrasound-guided cytological samples from canine livers.

Ultrasound-guided fine-needle sampling to obtain cytological samples is a well-established technique. However, the application of suction during sampling is controversial. Evidence from the human literature and one previous veterinary study suggest that non-aspiration may be superior for a number of organs. This prospective study compared the quality and diagnostic value of cytological samples from canine livers obtained by fine-needle aspiration (FNA) and non-aspiration (FN-NA) techniques. A total of 119 dogs that required ultrasound-guided FNA of the liver as part of their clinical investigation were recruited and randomly assigned to either FNA (n=54) or FN-NA (n=65) sampling groups. Specimens were reviewed by external cytopathologists masked to the technique used. Cytological reports were reviewed for their overall diagnostic value, cellularity, cell preservation and haemodilution. Overall, 88.2% (95% confidence intervals [CI], 82.4-94.0) of samples were diagnostic. There was a significant difference, as demonstrated by Chi-squared statistical analysis, in the prevalence of diagnostic samples between the FNA (81.5%; 95% CI, 71.1-91.8) and FN-NA groups (93.9%; 95% CI, 88.0-99.7; P=0.037). Non-diagnostic samples were significantly associated with lower cellularity, poorer cell preservation and more severe haemodilution (P<0.001 for each). However, there were no significant differences in the frequency of these specific variables between the FNA and FN-NA groups. In this study, fine-needle non-aspiration was superior to an aspiration technique for sampling the canine liver, as it resulted in higher rates of diagnostic cytology samples, with greater cellularity, less haemodilution and better cytological preservation.

Emergence of carriage of CTX-M-15 in faecal Escherichia coli in horses at an equine hospital in the UK; increasing prevalence over a decade (2008-2017).

BACKGROUND: This study investigated changes over time in the epidemiology of extended-spectrum ß-lactamase (ESBL) producing Escherichia coli within a single equine referral hospital in the UK. Faecal samples were collected from hospitalised horses in 2008 and 2017, processed using selective media and standard susceptibility laboratory methods. A novel real-time PCR with high resolution melt analysis was used to distinguish blaCTX-M-1 and blaCTX-M-15 within CTX-M-1 group. RESULTS: In 2008, 457 faecal samples from 103 horses were collected, with ESBL-producing E. coli identified in 131 samples (28.7, 95% CI 24.6-33.1). In 2017, 314 faecal samples were collected from 74 horses with ESBL-producing E. coli identified in 157 samples (50.0, 95% CI 44.5-55.5). There were 135 and 187 non-duplicate ESBL-producing isolates from 2008 and 2017, respectively. In 2008, 12.6% of isolates belonged to CTX-M-1 group, all carrying blaCTX-M-1, whilst in 2017, 94.1% of isolates were CTX-M-1 group positive and of these 39.2 and 60.8% of isolates carried blaCTX-M-1 and blaCTX-M-15, respectively. In addition, the prevalence of doxycycline, gentamicin and 3rd generation cephalosporin resistance increased significantly from 2008 to 2017 while a decreased prevalence of phenotypic resistance to potentiated sulphonamides was observed. CONCLUSIONS: The real-time PCR proved a reliable and high throughput method to distinguish between blaCTX-M-1 and blaCTX-M-15. Furthermore, its use in this study demonstrated the emergence of faecal carriage of CTX-M-15 in hospitalised horses, with an increase in prevalence of ESBL-producing E. coli as well as increased antimicrobial resistance to frequently used antimicrobials.

Sequential bacterial sampling of the midline incision in horses undergoing exploratory laparotomy.

BACKGROUND: There is limited information about bacterial isolates that are present on the equine midline incision during and following exploratory laparotomy. OBJECTIVES: To investigate the bacterial species cultured from the ventral midline pre-, intra- and post- laparotomy, whether particular bacterial isolates are associated with the development of surgical site infections (SSIs) and to report the antimicrobial resistance phenotypes of these isolates. STUDY DESIGN: Prospective cohort study. METHODS: The ventral midline of 31 horses undergoing exploratory laparotomy was sampled for bacterial culture at set time-points pre, intra and post-operatively. Inclusion criteria were that horses must have undergone exploratory laparotomy within 90 min of the initial colic examination upon hospital admission and must not have been placed in a stable prior to surgery. SSI was defined as any purulent or serous discharge from the laparotomy incision of >24 h duration. RESULTS: Seven horses (22.6%) developed a SSI. None of the variables tested were associated with the altered risk of SSI. The prevalence of a positive bacterial culture from the incision increased progressively over time and a variety of bacteria were isolated. A positive intra-operative culture was not a predictor of SSI; and when a SSI did occur, it was due to a different bacterial isolate. MRSA and ESBL-producers were identified in the post-operative period in one and four different horses respectively, but none of these developed a SSI. MAIN LIMITATIONS: Sampling was limited to hospitalisation and no culture results were available for horses developing SSI following hospital discharge. CONCLUSIONS: A variety of bacterial species may be isolated from equine laparotomy incisions peri-operatively without development of SSI. SSI does not appear to be solely related to bacterial contamination of the incision peri-operatively and other mechanisms such as bacteraemia merit further investigation.

CT findings in Pneumocystis carinii pneumonia in five dogs.

Pneumocystis carinii pneumonia is a rare disease in dogs. It is primarily reported in cavalier King Charles spaniels and miniature dachshunds with suspected underlying immunodeficiency. This case series reports the findings in five dogs (four cavalier King Charles spaniels and one Bedlington terrier) with confirmed P. carinii pneumonia. Thoracic (CT) revealed ground glass opacity of the pulmonary parenchyma with a diffuse or multifocal distribution. The severity of this pattern was variable. Less consistent imaging findings included parenchymal bands, bronchial dilation and signs consistent with pulmonary hypertension. Four dogs recovered well with treatment and there was resolution of CT abnormalities in all dogs with follow-up.

Canine total hip replacement using a cementless threaded cup and stem: a review of 55 cases.

OBJECTIVES: To determine the long-term results and complications associated with the Helica® cementless hip endoprosthesis system. MATERIALS AND METHODS: Retrospective study of 55 consecutive Helica total hip replacements performed between January 2010 and February 2015. RESULTS: A total of 55 total hip replacements were performed in 50 dogs; 45 dogs had unilateral hip replacements, and five had staged bilateral replacements. A total of 23 first-generation short femoral stems were implanted in 22 dogs, of which nine cases (39%) experienced complications and 19 (86%) recovered satisfactorily when revisions were included; 31 second-generation femoral stems were implanted in 28 dogs, of which 10 cases (32%) experienced complications, and 26 (93%) made satisfactory recoveries, including revisions. In 36 dogs for which data were available, postoperative Liverpool Osteoarthritis in Dogs scores were significantly reduced compared to preoperative scores. Complications occurred in a total of 19 cases (34·5%; 95% confidence interval: 22 to 47·1%); the most common was aseptic loosening of the femoral stem [11 cases (20%; 95% confidence interval: 9·4 to 30·6%)]. Following multivariable analysis, no risk factors were identified for overall complications, but there was a significant association of implant type (first-generation short stem) with loosening (odds ratio 4·9, 95% confidence interval: 1·1 to 22·1, P=0·034). CLINICAL SIGNIFICANCE: This study found the Helica hip endoprosthesis system to be effective in the management of hip dysplasia and osteoarthritis in dogs. Aseptic loosening of the femoral stem remains the most common complication but appears to have been significantly reduced with the introduction of the second-generation stem.

Evaluation of a multi-agent chemotherapy protocol combining lomustine, procarbazine and prednisolone (LPP) for the treatment of relapsed canine non-Hodgkin high-grade lymphomas.

The standard of care treatment for canine lymphoma is multi-agent chemotherapy containing prednisolone, cyclophosphamide, vincristine and an anthracycline such as doxorubicin (CHOP) or epirubicin (CEOP). Lomustine, vincristine, procarbazine, and prednisone (LOPP) has been evaluated as a rescue, with encouraging results; however, resistance to vincristine is likely in patients relapsing on CHOP/CEOP, and this agent may enhance LOPP toxicity without improving efficacy. The aim of this study was to evaluate responses to a modified-LOPP protocol that does not include vincristine (LPP) and is administered on a 21-day cycle. Medical records of dogs with high-grade multicentric lymphoma from 2012 to 2017 were reviewed. Dogs with relapsed lymphoma that received LPP as a rescue protocol were enrolled. Response, time from initiation to discontinuation (TTD) and toxicity of LPP were assessed. Forty-one dogs were included. Twenty-five dogs (61%) responded to LPP including 12 complete responses (CR) and 13 partial responses (PR). Responders had a significantly longer TTD (P < .001) compared to non-responders with 84 days for CR and 58 days for PR. Neutropenia was documented in 20 dogs (57%): 12 grade I to II, 8 grade III to IV. Thrombocytopenia was infrequent (20%): 5 grade I to II, 2 grade III to IV. Twelve dogs developed gastrointestinal toxicity (30%): 10 grade I to II and 2 grade III. Nineteen dogs had elevated ALT (59%): 9 grade I to II, 10 grade III to IV. Treatment was discontinued due to toxicity in 8 dogs (19%). The LPP protocol shows acceptable efficacy and toxicity-profile and minimizes in-hospital procedures.

Percutaneous ultrasound-guided cholecystocentesis: complications and association of ultrasonographic findings with bile culture results.

OBJECTIVES: To retrospectively evaluate cases presented for percutaneous ultrasound-guided cholecystocentesis for associated complications, identify risk factors associated with complications and to assess ultrasonographic findings and relate these to bacterial culture results. METHODS: Data on 300 patients presented for percutaneous ultrasound-guided cholecystocentesis were retrospectively collected and ultrasonographic images were assessed for defined structural changes. The incidence of major complications was determined. Multi-variable multi-level logistic regression was used to investigate the association of ultrasonographic findings with positive bile culture. RESULTS: Three hundred percutaneous ultrasound-guided cholecystocentesis procedures performed in 201 dogs and 51 cats were included; 35 patients had the procedure performed more than once. The overall incidence of major complications was 8 of 300 procedures (2·7%). Bile peritonitis occurred in 2 of 300 procedures (0·7%). An ultrasonographically abnormal gall bladder was found in 52% of cases and had a sensitivity, specificity and accuracy of 82, 55·7 and 61·5% respectively, to predict a positive bile culture. CLINICAL SIGNIFICANCE: Percutaneous ultrasound-guided cholecystocentesis is overall a safe technique when carried out in selected patients. Abnormal ultrasonographic findings are only a fair predictor of a positive bile culture.

Use of ECG-gated computed tomography, echocardiography and selective angiography in five dogs with pulmonic stenosis and one dog with pulmonic stenosis and aberrant coronary arteries.

Pulmonic stenosis (PS) is the most common congenital cardiac disease in dogs. Boxers and English bulldogs are among the most commonly affected breeds and also commonly associated with an aberrant coronary artery (CA). If an aberrant CA is suspected and balloon valvuloplasty indicated, an intra-operative angiography is recommended prior to the procedure. ECG-gated computed tomography (CT) can be used to screen for CA anomalies in a quick and minimally-invasive way (preventing side effects associated with selective catheter angiography) and allowing early planning of the procedure. The aim of this case series was to report CT findings associated with PS diagnosed by echocardiography. Our database was retrospectively searched for cases of dogs with PS diagnosed by echocardiography, where an ECG-gated CT was performed. A total of six cases were retrieved: all were diagnosed with severe PS. Four dogs had concurrent congenital defects: two dogs had a patent ductus arteriosus, one dog had a ventricular septal defect and an overriding aorta, one dog had an aberrant CA. Detailed CT findings of all cases were reported, including one case of a patent ductus arteriosus and an overriding aorta not identified by transthoracic echocardiography. In addition, an abnormal single left coronary ostium, with a pre-pulmonic right CA was described. In conclusion, despite echocardiography remaining the gold standard for diagnosis and assessment of PS, ECG-gated-CT angiography is a complementary diagnostic method that may provide additional relevant information, shorten surgery/anaesthesia time and reduce the amount of radiation to which the clinician is subjected.

Antimicrobial resistance in bacteria from horses: Epidemiology of antimicrobial resistance.

Antimicrobial resistance poses a significant threat to the continued successful use of antimicrobial agents for the treatment of bacterial infections. While the epidemiology of antimicrobial resistance in bacteria from man has been studied extensively, less work has been undertaken in companion animals, particularly horses. Methicillin-resistant Staphylococcus aureus has been identified as a cause of infections, with a low prevalence of nasal carriage by horses in the community but higher for hospitalised horses. Molecular characterisation has shown methicillin-resistant Staphylococcus aureus strains either to be predominantly of types associated with horses or of sequence type ST398. Antimicrobial-resistant Escherichia coli (including multidrug-resistant and extended spectrum ß-lactamase-producing isolates) have caused infections and been documented in faecal carriage by horses, with many significant resistance mechanisms identified. More sporadic reports and molecular characterisation exist for resistance in other bacteria such as enterococci, Salmonella, Acinetobacter and Pseudomonas species. Limited work has been undertaken evaluating risk factors and much of the epidemiology of antimicrobial resistance in bacteria from horses remains to be determined.

Comparison between echocardiographic and non-ECG-gated CT measurements in dogs.

The aim of this study was to compare echocardiographic measurements with non-ECG-gated contrast-enhanced cardiac CT measurements in dogs. Fifty-seven dogs were included in the study. The following echocardiographic parameters were measured: M-mode interventricular septum in diastole and systole, left ventricular internal diameter in diastole and systole, left ventricular free wall in diastole and systole, 2D left atrial maximal diameter, 2D left atrium to aorta ratio in diastole, 2D aortic annulus in systole and 2D pulmonary annulus in diastole and systole. CT measurements were obtained from multiplanar reconstruction images, replicating the imaging planes used for 2D measurements on echocardiography. It was not possible to discriminate between systole and diastole. The results showed moderate Lin's concordance correlation coefficients between the left ventricular internal diameter in systole (0.77), the aortic annuli (0.84) and the pulmonary annuli in diastole (0.78) and systole (0.80). Low coefficients were obtained between the other parameters. Bland-Altman plots for the parameters with highest concordance correlation coefficients were calculated. They suggested equivalence between the measurements of the aortic annuli. Equivalence was not seen between the remaining echocardiographic and CT measurements. Therefore, non-ECG-gated CT is not a reliable way of quantitatively assessing cardiac size.

Late presentation of canine nasal tumours in a UK referral hospital and treatment outcomes.

OBJECTIVES: To determine the computed tomographic stage of dogs with nasal tumours in a UK referral population, and whether stage, time to referral and treatment correlates with outcome. METHODS: Retrospective review of clinical records and computed tomography scans of dogs with nasal tumours. RESULTS: Dogs (n=78) presented to a referral practice in the UK with suspected nasal tumours are presented with more late stage tumours than dogs in the USA and Japan. Length of time from initial presentation to referral did not correlate with tumour stage at diagnosis. Median survival times for radiotherapy-treated dogs in this population are equivalent to those previously reported for late stage nasal tumours. CLINICAL SIGNIFICANCE: Dogs with nasal tumours are presented late in the course of disease in the North West of England. Dogs with clinical signs consistent with a nasal tumour should have timely imaging and biopsy, in order to make prompt treatment decisions. Although survival times are comparable with previous reports and radiotherapy is a valid treatment option for dogs with late stage disease, better outcomes are likely to be achievable with earlier treatment.

Cross-sectional study of antimicrobial-resistant bacteria in horses. Part 1: Prevalence of antimicrobial-resistant Escherichia coli and methicillin-resistant Staphylococcus aureus.

REASONS FOR PERFORMING STUDY: The increasing prevalence of antimicrobial-resistant bacteria such as methicillin-resistant Staphylococcus aureus (MRSA) and antimicrobial-resistant Escherichia coli represents a significant problem. However, the carriage of such bacteria by horses in the UK has not been well characterised. OBJECTIVES: To estimate the prevalence of nasal carriage of MRSA and faecal carriage of antimicrobial-resistant E. coli amongst horses in the general equine community of the mainland UK. METHODS: A cross-sectional study of horses recruited by 65 randomly selected equine veterinary practices was conducted, with nasal swabs and faecal samples collected. Faecal samples were cultured for antimicrobial-resistant E. coli. Nasal swabs were cultured for staphylococcal species; methicillin-resistant isolates identified as S. aureus were characterised by SCCmec and spa gene typing. Multilevel logistic regression models were used to calculate prevalence estimates with adjustment for clustering at practice and premises levels. Spatial variation in risk of antimicrobial resistance was also examined. RESULTS: In total, 650 faecal samples and 678 nasal swabs were collected from 692 horses located on 525 premises. The prevalence of faecal carriage of E. coli with resistance to any antimicrobial was 69.5% (95% CI 65.9-73.1%) and the prevalence of extended-spectrum ß-lactamase (ESBL)-producing E. coli was 6.3% (95% CI 4.1-9.6%). The prevalence of nasal carriage of MRSA was 0.6% (95% CI 0.2-1.5%). Spatial analysis indicated variation across the UK for risk of carriage of resistant and multidrug-resistant (resistant to more than 3 antimicrobial classes) E. coli. CONCLUSIONS AND POTENTIAL RELEVANCE: Carriage of MRSA by horses in the community appears rare, but the prevalence of antimicrobial-resistant E. coli (including ESBL-producing E. coli) is higher. A high prevalence of antimicrobial-resistant bacteria could have significant health implications for the horse population of the UK.

Cross-sectional study of antimicrobial-resistant bacteria in horses. Part 2: Risk factors for faecal carriage of antimicrobial-resistant Escherichia coli in horses.

REASONS FOR PERFORMING STUDY: The increasing prevalence of antimicrobial resistant bacteria such as antimicrobial-resistant and extended spectrum ß-lactamase (ESBL)-producing Escherichia coli represents a significant problem for human and veterinary medicine. Despite this, the risk factors for faecal carriage of such bacteria by horses in the UK, particularly those in the wider community, have not been well described. OBJECTIVES: To characterise the risk factors for faecal carriage of antimicrobial-resistant E. coli amongst horses in the mainland UK. METHODS: A cross-sectional study of horses recruited by 65 randomly selected equine veterinary practices was conducted, with a faecal sample collected and self-administered questionnaire completed by the horse owner. Faecal samples were cultured for antimicrobial-resistant E. coli, with isolates confirmed as E. coli having their antimicrobial resistance profile determined. Multilevel, multivariable logistic regression models were used to investigate risk factors for the carriage of antimicrobial-resistant E. coli in the sample population. RESULTS: Faecal samples and completed questionnaires were obtained for 627 horses located on 475 premises. Recent hospitalisation, contact with specific types of nonequid animals, the type of premises, the surrounding land use, the reason for veterinary treatment received in the last 6 months and antimicrobial treatment in the previous 10 days were identified as risk factors for many of the antimicrobial-resistance outcomes considered. Being stabled on the same yard as a recently hospitalised horse was identified as a risk factor for increased risk of carriage of ESBL-producing E. coli. CONCLUSIONS AND POTENTIAL RELEVANCE: Increasing antimicrobial resistance may have significant health implications for the horse population of Great Britain. This form of epidemiological investigation highlights potential risk factors that may be controlled to limit the extent of the problem.

Prevalence of antimicrobial-resistant Escherichia coli in dogs in a cross-sectional, community-based study.

The prevalence of carriage of antimicrobial-resistant (AMR) and extended-spectrum ß-lactamase (ESBL)-producing Escherichia coli was determined in 183 healthy dogs from a semi-rural community in Cheshire. Isolates were tested against a panel of antimicrobials and by PCR to detect resistance genes. In the suspected ESBL-producing isolates, the presence of bla(SHV), bla(TEM), bla(CTX-M) and bla(AmpC) genes was determined by PCR and sequencing. A total of 53 (29 per cent, 95 per cent confidence interval [CI] 22.4 to 35.5 per cent) dogs carried at least one AMR E coli isolate. Twenty-four per cent (95 per cent CI 17.9 to 30.2 per cent) of dogs carried isolates resistant to ampicillin, 19.7 per cent (95 per cent CI 13.9 to 25.4 per cent) to tetracycline and 16.9 per cent (95 per cent CI 11.5 to 22.4 per cent) to trimethoprim. A bla(TEM) gene was detected in 39 of 54 ampicillin-resistant isolates, a tet(B) gene in 12 of 45 tetracycline-resistant isolates, and a dfr gene in 22 of 33 trimethoprim-resistant isolates. Multidrug-resistant isolates were demonstrated in 15 per cent (28 of 183; 95 per cent CI 10.1 to 20.5 per cent) of dogs. Nine suspected ESBL-producing E coli were isolated, of which only one was confirmed by double disc diffusion testing. Two of these isolates carried the bla(TEM-1) gene and seven carried the bla(CMY-2) gene.