Вконтакте Одноклассники Фейсбук Гугл+ Английский Испанский Итальянский Русский Украинский


Pericardial effusion associated with systemic inflammatory disease in seven dogs (January 2006 - January 2012)

Pericardial effusion associated with systemic inflammatory disease in seven dogs (January 2006 - January 2012)

Author information

Covey H.L., Connolly D.J. Pericardial effusion associated with systemic inflammatory disease in seven dogs (January 2006 - January 2012).PRESS J Vet Cardiol. 2018 Apr;20(2):123-128.


Pericardial effusion (PE) is reported in dogs as a consequence of neoplasia, primary cardiac disease or as an idiopathic condition . We describe seven dogs with systemic inflammatory disease, PE without tamponade and increased cardiac troponin I concentrations. Echocardiographic findings and adjunctive testing did not identify other known causes of PE. Resolution of the PE was documented in five of seven dogs in which follow-up echocardiography was performed, often after anti-inflammatory therapy. Resolution of PE was associated with normalisation of cardiac troponin I levels. Clinical signs had not recurred in six dogs with follow-up for more than 12 months and up to 7 years. These findings suggest an association between systemic inflammation and PE in dogs.

KEYWORDS: Cardiac troponin I; Echocardiography; Pericarditis; Tamponade


Echocardiographic measurements were made in M-mode for left ventricular internal dimension at end-diastole (LVIDd), left ventricular internal dimension at end-systole (LVIDs) and fractional shortening (FS) [1] and in two-dimensional mode for the ratio of the left atrial dimension to the aortic annulus dimension [2]. Echocardiographic measurements were consistent with systolic dysfunction (FS 8.8%; RI 23-47%, normalised LVIDd 2.01; RI 1.27-1.85, normalised LVIDs 1.72; RI 0.71-1.26). Pericardial effusion and mild mitral insufficiency were noted. Diagnostic pericardiocentesis revealed a neutrophilic exudate with no bacterial growth.

Further investigations by computed tomography revealed pleural fluid and an area of hyperintensity in the cervical musculature. Pleocytosis (protein content 0.53 g/L; RI 0-0.3 g/L) was identified on cerebrospinal fluid analysis. Cervical muscle biopsy revealed panniculitis and myositis with no bacterial growth. A diagnosis of steroid-responsive meningitis-arteritis (SRMA) was reached.

Treatment with prednisolone (3.22 mg/kg PO q 24 h) was instituted on day 2 and repeat echocardiography after four days revealed resolution of the PE, with subjectively similar systolic dysfunc¬tion. Prednisolone therapy was tapered by 25% every 4 weeks, for a complete course duration of 8 months, and telephone follow-up revealed her to be alive and asymptomatic 3 years after presentation.

Case 2

An 8-year-old male neutered whippet presented with an 11-day history of pyrexia. At presentation, his temperature was 39.9 °C, HR 120 min1, RR 24 min-1, and a grade I/VI systolic murmur was auscultated. Physical evidence of cardiac tamponade was not evident. Hypoalbuminaemia (22 g/L) and neutrophilia (23.52 x 109/L) were present, and canine-specific pancreatic lipase immunoreactivity was increased (772 mg/L; RI < 200 mg/L). Cardiac troponin I was not measured. Echocardiography revealed decreased FS (12.7%) and increased normalised LVIDd (2.1; RI 1.27-1.85) and normalised LVIDs (1.71; RI 0.71-1.26).

Pericardial effusion was present with mild diastolic collapse of the right atrial free wall that did not affect cardiac filling or chamber dimensions. Diagnostic thoracocentesis revealed a neutrophilic exudate (total nucleated cell count 45 x 109/L and protein 33.5 g/L) with negative bacterial culture. Diagnostic ultrasound revealed abdominal lymph node enlargement and mild ascites but was otherwise unremarkable. Computed tomography confirmed PE, pleural effusion, and small volume ascites. Urinalysis was unremarkable. He was hospitalised with a diagnosis of systemic inflammatory response syndrome (SIRS) of unknown aetiology and treated with intravenous fluid therapy. Repeat echocardiography two days later revealed the resolution of the PE however fractional shortening remained reduced at 13%. Clinical signs, including abdominal effusion, resolved, and he was discharged with no medication. Follow-up confirmed continued survival 4 years after presentation.

Case 3

A 2-year-old female neutered mixed breed was presented for investigation of a 1 week history of pyrexia. At presentation, physical examination revealed evidence of neck pain, with a temperature of 39.1 °C, HR 120 min-1, and RR 30 min-1. Blood testing revealed hypoalbuminaemia (25.2 g/L) and neutrophilia (14.4 x 109/L). Cardiac troponin I was increased (31.2 mg/L). Echocardiography confirmed small volume PE without tamponade, and no structural cardiac abnormalities. Fractional shortening was decreased (16%), but normalised LVIDd (1.8) and normalised LVIDs (1.42) were within reference intervals. Abdominal ultrasonography revealed minimal peritoneal effusion but was otherwise unremarkable. Cerebrospinal fluid analysis revealed mononuclear pleocytosis. Serology for Toxoplasma sp., Neospora sp., Bartonella sp., and Ehrlichia sp., were negative. Canine-specific pancreatic lipase immunoreactivity and urinalysis were within reference intervals. Bacterial blood cultures were negative. A diagnosis of SRMA was reached.

Treatment with clindamycin (11.7 mg/kg PO q 12 h for 3 weeks) and prednisolone (2.6 mg/kg PO q 24 h) was initiated. Re-examination after 19 days revealed resolution of clinical signs and decreased cTnI (0.5 mg/L). Prednisolone was tapered by 25% every 4 weeks for 6 months. Echocardiography was not repeated but follow-up revealed survival, without clinical signs, 3 years and 5 months after presentation.

Case 4

An 8-year-old male neutered lurcher was presented with a 6-week history of pyrexia. Exami¬nation revealed neck pain, rectal temperature 39.8 °C, HR 120 min1, RR 30 min1, a healing abscess on the dorsum, and grade II-III/VI apical bilateral systolic heart murmur. Mild hypoalbuminaemia (27.3 g/L) and neutrophilia (15.76 x 109/L) were present. Cardiac troponin I was increased at 12.6 mg/L. Initial echocardiog¬raphy showed no PE, mild mitral regurgitation, and poor left ventricular systolic function (FS 12.6%, normalised LVIDd 2.48, normalised LVIDs 2.02).

Cerebrospinal fluid analysis showed pleocytosis with no microorganisms, consistent with a diagnosis of SRMA. Blood cultures and infectious dis¬ease screening (PCR testing for canine distemper, Neospora sp. and Toxoplasma sp., serology for Borrelia sp. and Ehrlichia sp.) were negative. Repeat echocardiography 2 days after the presentation revealed PE, and an electrocardiogram showed supraventricular tachycardia (230 min-1). Therapy was initiated with amoxicillin-clavulanate (20 mg/kg PO q 12 h) and diltiazem (1.8 mg/kg PO q 8 h). Prednisolone (2 mg/kg PO q 24 h) was added on day 5. Sinus rhythm was restored with occasional ventricular premature complexes. On day 7, diagnostic pericardiocentesis was carried out in response to an increase in PE. Fluid analysis revealed protein-rich modified transudate (protein 38.2 g/L, total nucleated cell count not recorded). A 14-day course of enrofloxacin (5 mg/ kg PO q 24 h) was added. Echocardiography on day 14 revealed resolution of the PE but continued poor systolic function of the left ventricle (FS 7.7%, normalised LVIDd 2.22, normalised LVIDs 1.92). An electrocardiogram showed increased frequency of ventricular premature complexes. Sotalol (1.5 mg/kg PO q 12 h) was added to the treatment regime. Cardiac troponin I had decreased to 0.51 mg/L.

Echocardiography after 24 days of first pre¬sentation revealed continued resolution of PE and improved FS (21.1%). Antibiotic therapy was dis¬continued after 5 weeks. Sotalol and diltiazem were continued for recurrent supraventricular tachycardia. Azathioprine (50 mg/m2 PO every second day) was added and prednisolone therapy was slowly tapered. Both therapies were eventually discontinued. Telephone follow-up revealed that the dog was euthanised 11 months after initial presentation due to urinary incontinence.

Case 5

A 3-year and 7-month-old female neutered cocker spaniel presented with a 2-month history of lethargy, stiff gait, and intermittent pyrexia. Examination revealed neck pain, temperature of 40.1 °C, HR 108 min-1, RR 28 min-1. Neutrophilia (22.3 x 109/L) was present. Cardiac troponin I was increased at 70.1 mg/L. Echocardiography revealed PE without tamponade and reduced left ventricular systolic function (FS 18.1%) with hyperechoic ventricular myocardium. Radiographs showed small volume pleural effusion. Cerebrospinal fluid analysis showed inflammatory cells with no microorganisms, consistent with a diagnosis of SRMA. Bacterial blood cultures were negative. Therapy was initiated with prednisolone (4 mg/kg PO q 24 h) and azathioprine (50 mg/m2 PO every second day) from day 2.

Echocardiography after 6 days of documented resolution of PE identified a small pedunculated mural mass (7 x 7 mm) in the right atrium suspected to be a thrombus. Repeat cTnI was decreased (2.3 mg/L). Aspirin (0.5 mg/kg/day PO q 24 h) was added. All medications were eventually tapered and discontinued over a 6-month period. Telephone follow-up to the referring veterinarian confirmed continued survival 7 years and 10 months after initial presentation.

Case 6

A 2-year 9-month-old female neutered cocker spaniel was presented with a 3-week history of lethargy, inappetence, and neck pain. Examination revealed neck and joint pain with pyrexia (40.5 °C), HR 100 min-1, RR 16 min-1, and neu¬trophilia (25.5 x 109/L). Cardiac troponin I was increased (6.8 mg/L). Echocardiography revealed small volume PE, but normal left ventricular chamber diameters (normalised LVIDd 1.57, normalised LVIDs 1.09) and FS (26.7%). Other cav- itatory effusions were not found. Cerebrospinal fluid analysis revealed moderate pleocytosis. Synovial fluid was considered unremarkable and a diagnosis of SRMA was reached.

Prednisolone therapy was initiated as a tapering course over 4 months starting at 2.1 mg/kg PO q 24 h. Repeat echocardiography performed 6 days later confirmed the resolution of PE. Attempts at follow-up communication were unsuccessful, and so survival could not be ascertained.

Case 7

A 1-year and 4-month-old male neutered Nova Scotia duck-tolling retriever presented with a 1- week history of pyrexia, lethargy, and cough. Examination revealed pyrexia (40.3 °C), HR 86 min1, RR 26 min1, and a harsh cough. Diag¬nostic testing revealed neutrophilia (26.3 x 109/L), mild hypoalbuminaemia (26.9 g/L), and increased cTnI (30.2 mg/L). Echocardiography showed small volume PE and marginal pleural effusion without evidence of left ventricular systolic dysfunction. Computed tomography revealed focal opacity of a lobar bronchus and a small volume peritoneal effusion. A purulent exudate was found in the affected bronchus on bronchoscopy. Cytology and/ or culture of joint aspirates, cerebrospinal fluid, bronchoalveolar lavage and blood were not indicative of active infection.

Clinical signs completely resolved with intravenous fluid therapy alone. The dog was discharged without medications, for monitoring at home. Repeated cTnI, at the referring veterinarian 19 days later was within the reference interval (0.2 mg/L). Follow-up confirmed continued survival with no clinical signs 1 year and 10 months later.


Pericardial effusion often presents with signs of cardiac tamponade, such as collapse due to acute reduction in cardiac output , or progressive exercise intolerance and respiratory difficulty [3]. Imaging of the thorax and heart is central in obtaining a provisional diagnosis of the underlying cause [4]. Pericardial effusion in dogs has been reported most often associated with cardiac neoplasia, or as idiopathic effusions in large breed dogs, and less often associated with congestive heart failure or bacterial, parasitic or fungal pericarditis [5—9].

This report documents PE without tamponade in dogs with evidence of systemic inflammatory dis¬eases. Similar observations have been reported in humans with several different manifestations of systemic inflammatory disease [10,11]. To our knowledge, this observation has not been pre¬viously reported in dogs. Diagnosis of systemic inflammatory response syndrome is based on the presence of two or more of the following criteria; increased HR, increased RR, increased or decreased body temperature and white blood cell count outside the reference range [12]. Adjusted criteria for dogs have varied, but the purposes of this report were defined as body temperature >39.7 °C or <37.8 °C and white blood cell count >12,000 WBCs/mL or <4000 WBCs/mL (with or without band neutrophils), RR > 40 min~1 and HR > 160 min1 [13]. Six of the seven dogs in this report fulfilled two or more of these criteria for SIRS. Five of those witha SIRS diagnosis, and the remaining case not consistent with SIRS, had an inflammatory disease focus identified during workup.

Known underlying causes of PE in dogs were not identified in the dogs in this report. Pericardial effusion in these was not associated with cardiac tamponade as is often the case with known causes of PE . Finally, the PE resolved without therapeutic pericardiocentesis, typically in response to antiinflammatory therapies, did not recur on longterm follow-up. Therefore, it seems reasonable to conclude that PE in these dogs was the result of systemic inflammation.

Increased cTnI concentration is indicative of damage to cardiac myocytes [14]. In the dogs in this report, secondary non-septic myocarditis through the action of circulating inflammatory mediators was the most likely cause of increased cTnI, though primary myocarditis cannot be ruled out. Increased serum cTnI has been documented in neoplastic (especially cardiac haemangiosarcoma) PE, but not idiopathic PE [15]. The majority of dogs showed evidence of impaired left ventricular systolic function that persisted even after resolution of the effusion, suggesting compromise of myocardial function by systemic inflammatory disease [16]. In one dog where longer-term echocardio- graphic follow-up was available, an improvement in systolic function was demonstrated over the 64- day period. Other dogs in this report may not have had sufficient follow-up to demonstrate this. The improvement in serum troponin levels and long¬term absence of cardiac-related clinical signs without inotropic support are supportive of a transient cardiac injury, but persistent structural myocardial disease was not ruled out in all of these dogs.

The formation of fluid in the pericardial sac, in the absence of abnormal production by infectious or neoplastic disease, is by passive filtration from plasma [17]. Disease states affecting vascular fluid balance; in terms of hydrostatic pressure, endo¬thelial membrane permeability or colloid osmotic pressure, would therefore be anticipated to affect and possibly to increase its formation. Although hypoalbuminemia was a common finding in these dogs, the changes were mild, attributable to acute phase response or third-space fluid shift, and unlikely to have contributed to fluid extravasation by alterations in oncotic pressure. Fluid exudation from affected vessels can be associated with inflammation [18] mediated by inflammatory cytokines resulting in altered vascular function and permeability [19].

The presence of multicavitary effusions in five of the seven dogs is consistent with vasculitis, associated with systemic inflammation, as a cause of the PE. Hyperechoic myo¬cardial changes on echocardiography in one dog could be attributed to myocardial inflammatory change, though other etiologies cannot be entirely ruled out. The pathogenesis of SRMA (the most common diagnosis in these dogs) has been linked to pro-inflammatory cytokine production from T- helper 2 cells [20]. Fibrinoid arteritis is the characteristic histological lesion of SRMA [21], with lesions commonly identified in the meninges, but interestingly also in coronary arteries [22]. Transepicardial fluid flow is one of the principal determinants of fluid balance between the pericardial space and the myocardium. Inflammatory medi¬ated oedema of the myocardium, as seen in some inflammatory diseases, would be expected to increase transudation of fluid and formation of PE [23]. Idiopathic PE could be considered as a possible cause in these dogs, however given their long survival, spontaneous long-term resolution of PE without therapeutic pericardiocentesis would be atypical of idiopathic PE. Furthermore, cTnI was not elevated in previously reported cases of idio¬pathic PE [15] and affected dogs do not generally have evidence of systemic inflammation.

Histological analysis of the pericardium was not performed in dogs in this report, making it difficult to exclude neoplasia, particularly mesothelioma, as a cause of PE. The reported median survival for dogs with pericardial mesothelioma is 13.6 months [24] and so the prolonged survival of most dogs in this case series, with complete resolution of their signs, would be atypical. Infectious disease testing included blood culture (five dogs) and serology or PCR testing (including Bartonella sp. [one dog], Neospora sp. and Toxoplasma sp. [one dog] and arthropod borne disease [three dogs]). All results were negative but not all dogs underwent infectious disease testing, and the difficulty in diagnosis of some pathogens, particularly Bartonella, which has been reported to cause myocarditis and pericarditis with effusion [25,26], make this difficult to entirely exclude. However, five dogs were given immunosuppressive treatment and would have been expected to worsen if infectious disease were present.

In conclusion, this report describes a series of dogs presenting with evidence of systemic inflammation, increased cTnI, left ventricular systolic dysfunction and non-tamponade PE. Pericardial effusion was typically associated with multicavitary effusion, was self-limiting and associated with long-term resolution in response to antiinflammatory therapies. These findings suggest an association between systemic inflammation and PE in dogs. The long-term importance of left ventricular systolic dysfunction in dogs with inflammation-associated PE remains unknown.

Conflicts of Interest Statement

The authors do not have any conflicts of interest to disclose.


Acknowledgements to Pierre Menaut for con¬tribution of cases to the series.



Полезно знать