Cardiac MRI Findings in a Dog with a Diffuse Pericardial Mesothelioma and Pericardial Effusion

Introduction

Pericardial effusion is a common acquired cardiac disease in dogs. The most common causes of pericardial effusion in dogs include hemangiosarcoma of the right atrium, idiopathic pericarditis, and chemodectoma of the heart base. Other less common reported causes include infectious pericarditis, mesothelioma, neuroendocrine tumors, or ectopic thyroid neoplasia.¹

A mesothelioma is a tumor of mesodermal origin that arises from the serosal lining of body cavities, most commonly pleural, peritoneal, or pericardial.² Mesotheliomas have always been considered rare causes of pericardial effusion, but a recent study suggested that mesotheliomas may be more common than originally thought.¹ In that study, 15 of 71 dogs with pericardial effusion of neoplastic origin were diagnosed with pericardial mesothelioma, and only 5 of those 15 cases had focal masses (4 of the heart base and 1 of the right atrium).

Identifying the etiology of pericardial effusion can be challenging. Thoracic radiographs and echocardiography are typically used to detect masses in the right atrium, pericardium, or heart base, but those masses can be small and difficult to detect.

Cardiac MRI (cMRI) is currently used in human medicine because it provides excellent images of the entire pericardium without the need for either IV ionic contrast administration or ionizing radiation.³ The multiplanar imaging achievable with MRI, the outstanding soft tissue contrast (far superior to either computed tomography or ultrasound), the ability to acquire images synchronized with the cardiac cycle (cardiac gating), and the ability to assess both hemodynamics and myocardial contractility are some of the advantages of cMRI over other imaging techniques.

The purpose of this report is to illustrate the potential of cMRI in the management of patients with pericardial effusion when echocardiographic findings are equivocal, and to illustrate cMRI findings in a case of diffuse pericardial mesothelioma.

Case Report

An 8 yr old Pit bull terrier presented to the emergency service of the Ryan Veterinary Hospital of the University of Pennsylvania with a 2-day history of lethargy and abdominal distension. On physical examination, the dog was lethargic, had muffled heart sounds, and presented with a pulsus paradoxus. His abdomen was markedly distended with a positive wave sign. An emergency ultrasound examination revealed a large amount a fluid in the abdomen and pericardium. A pericardiocentesis was performed, and 500 mL of hemorrhagic fluid were removed from the pericardial sac. In addition, 1,500 mL of clear fluid were removed from the abdomen.

The analysis of the pericardial fluid revealed 2.8 × 10⁹ nucleated cells/L, 4.11 × 10¹² red blood cells/L, and total protein of 31 g/L. Cytology revealed mesothelial hyperplasia and dysplasia, considered most likely reactive to the effusion. The analysis of the peritoneal fluid showed 0.2 × 10⁹ nucleated cells/L, 0.01 × 10¹² red blood cells/L, and total protein of 36 g/L, consistent with a modified transudate, likely secondary to cardiac tamponade. A complete blood cell count revealed a 4.46 × 10⁶ red blood cells/μL (reference range, 5.83–88.87 × 10⁶/μL), hemoglobin of 108 g/L (reference range, 133–205 g/L), and a hematocrit of 29.9% (reference range, 40.3–60.3%), consistent with mild nonregenerative anemia. The remainder of the blood work was within normal limits. The dog tested negative to Dirofilaria immitis, Ehrlichia canis, Borrelia burgdorferi, and Anaplasma phagocytophilum^a, and the serology titer of Rickettsia rickettsii was 1:40.

A full echocardiogram was performed the following day, which revealed a small amount of residual pericardial effusion, mild left atrial enlargement, and mild mitral and tricuspid valve insufficiency. A questionable small mass lesion was suspected in the area of the aorta at the heart base, but it was not possible to definitely confirm it.

The animal was stable and was sent home with a tentative diagnosis of idiopathic hemorrhagic pericarditis. He was scheduled for a follow-up examination 5 days later.

At the follow-up visit, the dog presented with mild abdominal distension, peripheral edema, and muffled heart sounds. On echocardiography, he had recurrent severe pericardial effusion. The quick recurrence of the pericardial effusion was suggestive of an underlying problem. A pericardiectomy was offered as an alternative treatment, but because a small periaortic mass lesion had been suspected on echocardiography, further diagnostics were recommended prior to surgery to rule out a mass that could be causing the pericardial effusion.

A cMRI was performed using a 1.5 Tesla MRI unit^b with the patient in dorsal recumbency under general anesthesia. The protocol used was as previously described for canine cMRI.⁴ A black blood series was obtained using an oblique fast spin-echo double inversion-recovery sequence (repetition time, 1,200 msec; echo time, 40.9 msec; echo train length, 16; blood suppressed [inversion time, automatic]). The triggering parameters were single cardiac cycle phase acquisition with a minimum trigger delay and an arrhythmia rejection window of 20 msec. Bright blood cine imaging was obtained using an oblique gradient-echo cine stack sequence (repetition time, 60 msec; echo time, 3.3 msec; flip angle, 30°). The acquisition was prospectively gated with 30 segments of the cardiac cycle reconstructed/location and two views of k-space encoded per segment per cardiac slice. Contrast-enhanced MRI was performed with the three-dimensional (3D) volume prescribed in the dorsal plane and positioned to cover the entire thorax using a multiphase MRI angiography protocol previously described.^4,5 Briefly, a 3D fast spoiled gradient-recalled echo sequence with elliptic centric view ordering of k-space was used with parallel acquisition. Four consecutive 3D volumes were acquired starting immediately after the injection of gadolinium^c. A precontrast mask was acquired with the same parameters just before manual injection of gadolinium (0.3 mmol/kg). The total acquisition time of the study was approximately 30 min.

On the black blood transverse series, along the left ventral pericardial space (midventricular level), there was a 63 mm long, crescent-shaped region that was slightly hyperintense to the myocardium and heterogeneous that was thought to represent either focally thickened pericardium or accumulation of fibrin and clots in the pericardial sac (Figure 1). Based on the postcontrast images, no enhancing material was seen within the pericardial sac at that location, which was more consistent with accumulation of blood clots and fibrin in the pericardial sac as opposed to either an active inflammatory or neoplastic infiltrative lesion. A small amount of hypointense pericardial fluid was present. The heart base, peri- and para-aortic regions, and right auricular appendage area were well seen on the various MRI series and no mass was identified (Figure 2). The postcontrast images revealed a diffuse thickening (up to 7 mm) of the pericardium with strong enhancement. The differential diagnoses for the diffuse thickening of the pericardium with strong enhancement included pericarditis and diffuse neoplastic infiltration such as mesothelioma (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

After the MRI, a pericardiectomy was performed after median sternotomy. The only finding during surgery was diffuse pericardial thickening. As had been shown by cMRI, no heart base mass was present. Samples of the thickened pericardium were submitted for culture and histopathology. The dog recovered well from the surgery, and the clinical signs resolved. He was sent home 2 days after surgery with oral tramadol (3.3 mg/kg q 8 hr).

Histopathologic examination of the pericardium revealed malignant neoplastic cells consistent with either carcinoma or mesothelioma. Immunohistochemistry was performed and showed that the neoplastic cells were positive for cytokeratin^d and vimentin^e, which was consistent with a diagnosis of malignant pericardial mesothelioma.

Two wk postsurgically, the patient was presented to the oncology service to discuss treatment options. A moderate amount of pleural effusion was detected on thoracic radiographs and ultrasound. A thoracocentesis was performed, and cytology confirmed the presence of neoplastic mesothelial cells in the pleural space. Based on that finding, intracavitary chemotherapy with carboplatin was selected as the best treatment option. Three courses of chemotherapy separated by 3-wk intervals were initially planned. The pleural fluid was drained by thoracocentesis, and an intracavitary injection of carboplatin^f (300 mg/m²) was administered. The carboplatin was diluted in 5% dextrose solution^g to 10 mg/mL and diluted again prior to injection in 0.22 mL/kg of the dextrose solution as described in a previous study.⁶ The same protocol was repeated for the next two treatments. Thoracic radiographs repeated before each treatment showed moderate recurrent pleural effusion, but the dog was reported to be doing well at home during the first 6 wk of chemotherapy.

One wk after the last treatment (9 wk after pericardiectomy), the animal was presented with lethargy and anorexia. One wk later, the owners decided to euthanize the dog due to progressive worsening of clinical status. Necropsy was not performed.

Discussion

cMRI is the reference imaging technique for assessment and characterization of either suspected cardiac or juxtacardiac masses in human medicine.⁷ The excellent multiplanar assessment of the anatomy, tissue composition, and functional impact makes cMRI the technique of choice for early detection of cardiac tumors and treatment planning.

In veterinary medicine, cMRI has been recently introduced in small animal medicine and has proven to be excellent for imaging normal anatomy and cardiac neoplasia and detect either congenital or degenerative abnormalities.^4,8,9 In a recent study, cMRI has been used in dogs trying to differentiate neoplastic from nonneoplastic pericardial effusions, but did not include cases of mesotheliomas.¹⁰ In that study, cMRI did not improve identification of cardiac tumors compared with transthoracic echocardiography; however, cMRI was useful for characterizing the extent, location, and potential tumor type. The reported lack of improvement in the identification of cardiac masses with cMRI could be due to the low number of cases included. That study included eight cases, and only six of those cases had a detectable cardiac mass. Four of the masses were identified with both transthoracic echocardiography and cMRI. In those four cases, cMRI was useful to both confirm and better characterize the lesions that were detected with transthoracic echocardiography. The other two masses were initially missed with both techniques. Once the authors had acquired more experience and had re-evaluated the cMRIs, the authors detected small cardiac masses that had been overlooked during the initial evaluation. That suggested that specific training could be necessary for a correct interpretation of cMRI. In addition, cMRI was useful in one of the cases to differentiate between a thrombus and neoplastic infiltration.

In this case report, cMRI was a useful diagnostic tool to rule out a small cardiac mass that was questionably present on echocardiography. In cases of hemorrhagic pericardial effusion, it is important to determine presurgically if a mass is present. Such a mass could continue to bleed into the pleural space after pericardiectomy. In addition, the presence of a neoplastic infiltration may change the prognosis of the patient and could alter the treatment options. To the authors’ knowledge, this is the first case report describing the MRI findings of a pericardial mesothelioma in a dog.

A pericardial mesothelioma is a rare cardiac neoplasia that has been sporadically reported in veterinary medicine.^1,2,11,12 The diagnosis of pericardial mesothelioma can be challenging because it can present either as a discrete mass or a more diffuse infiltration, as was seen in this case report. Cytology of the effusion is often not conclusive to differentiate between idiopathic pericarditis and mesothelioma. Mesothelial cell reactivity has been found in up to half of the cases of pericardial effusion independent of the cause. A definitive diagnosis can be difficult to achieve even with histopathologic evaluation, and special immunohistochemical stains are often necessary.¹ In a previous study, immunochemistry showed that 74% of neoplastic effusions had been considered nonneoplastic on cytology, and 13% of the nonneoplastic effusions had been cytologically misdiagnosed as being neoplastic.¹³

The differentiation between neoplastic and nonneoplastic pericardial effusion is important for prognosis and treatment. Two previous studies showed that the prognosis for neoplastic causes was poor, with mean survival times ranging from 26 days to 56 days, compared with survival times of 790–1,068 days for nonneoplastic causes.^14,15 However, according to a recent publication, cardiac mesotheliomas have a better prognosis than cardiac hemangiosarcomas and can have a similar prognosis compared with nonneoplastic causes of pericardial effusion.^1,14

This case report illustrates that cMRI can be useful in ruling out the presence of either a cardiac or pericardial mass lesion; however, further studies are needed to see if cMRI could have some potential in differentiating mesotheliomas from idiopathic pericarditis. In people, pericardial thickening with enhancement can be seen with mesotheliomas, pericarditis, and lymphoma.^7,16

Conclusion

cMRI in small animals can be used as an alternative method to rule out cardiac masses when echocardiography is inconclusive. A mesothelioma should be a differential diagnosis in dogs with diffuse thickening and strong enhancement of the pericardium.