Introduction

Cardiac tumors are uncommon in dogs; in a large retrospective study of 729,265 dogs, the incidence of cardiac tumors was reportedly 0.19%.¹ Cardiac hemangiosarcoma (HSA) is the most common neoplasia (69%) and causes pericardial effusion.^1,2 The incidence of HSA is 25–100-fold greater in dogs than in humans, and breed predilections also exist.³ For example, older German shepherd dogs and golden retrievers are predisposed to cardiac HSA.^4–7 Males also tend to be more predisposed, although sex predilection is controversial.^4,8 Additionally, HSA can potentially occur at any vascularized site in the body because it arises from the vascular endothelium or endothelial precursor cells; however, pluripotent bone marrow progenitor cells have been implicated as the origin for HSAs in the past 10 yr, and HSAs have been shown to occur most frequently in the spleen, right atrium/auricle, subcutaneous tissues, and liver.^3,9–13 Furthermore, HSA is aggressive, and malignant tumors commonly metastasize to the lungs at an early stage.^7,14 In dogs with cardiac and splenic HSAs, metastases are observed not only in the lungs but also in the liver, kidneys, great omentum, and serosa of the jejunum.⁸

Aggressive treatment including surgical tumor resection and subtotal pericardiectomy is considered palliative in dogs with cardiac HSA because most affected dogs die of metastases. Therefore, clinicians recommend only pericardiectomy for palliation of recurrent cardiac tamponade associated with right atrial bleeding from tumor rupture.¹⁵ However, a combination of tumor resection, pericardiectomy, and adjuvant chemotherapy significantly prolonged the survival time from a median of 42 days to one of 175 days in dogs with cardiac HSA.¹³ In addition, autogenous pericardial patch graft reconstruction was previously reported in three dogs with both right auricle and right atrial afflictions or with uncontrolled bleeding secondary to right auricular HSA resection.^16–18 Of these three dogs, one who had uncontrolled bleeding secondary to right auricular HSA resection underwent a broad autogenous pericardial patch graft reconstruction, which included the right atrium as well as the coronary groove and right ventricle.¹⁶ In that case, transient arrhythmia occurred after the surgery, although detailed information was not available. In addition, the effect on right ventricular function was not examined. Nevertheless, right myocardial ischemia may occur with this treatment because the broad suturing area includes the coronary groove and right ventricle.

Autogenous pericardium is an ideal material for heart reconstruction in dogs because it is an autologous tissue and is cost-effective.^19,20 However, autogenous pericardium has potential size and thickness limitations, and cartilaginous metaplasia and dystrophic calcification are likely to occur in the body.^21,22 In contrast, artificial patches are available in various sizes and thicknesses. In particular, expanded polytetrafluoroethylene (ePTFE) is most frequently used for heart reconstruction because it can be easily handled and has long-term durability.^23,24 Therefore, we used the ePTFE patch for right atrioventricular reconstruction instead of the autogenous pericardium in this study.

In this study, our purpose was to elucidate the effects of ePTFE patch graft reconstruction over a broad area, including the right atrium, coronary groove, and right ventricle, on arrhythmogenicity and right ventricular function.

Materials and Methods

Right Atrioventricular ePTFE Patch Graft

All four dogs underwent right atrioventricular ePTFE patch graft reconstruction (RAPP; Figure 1). Atropine sulphate hydrate^c was preoperatively administered (0.025 mg/kg subcutaneously [SQ]) prior to anesthesia induction with 6–8 mg/kg IV propofol^d. Anesthesia was maintained with inhalation of 1.5–2.0% isoflurane^e in 100% oxygen. The dogs were restrained in the supine position. Pre- and perioperative pain was controlled with 0.2 mg/kg of SQ meloxicam^f and 0.01 mg/kg of IV buprenorphine hydrochloride^g. Cefazolin sodium hydrate^h (25 mg/kg IV) was administered as a prophylactic antibiotic, which was repeated every 2 hr during surgery. Intraoperative fluid infusion was administered with lactated Ringer's solutionⁱ at 5 mL/kg/hr IV.

View Figure

• Download as Powerpoint
• Download Figure

The sternotomy incision method was used to access the right atrium and the right ventricle. After pericardial tenting, the right auricle was grasped using a Satinsky blood vessel clamp, and its tip was completely resected to generate a bleeding model. The bleeding site was reconstructed with an ePTFE patch^j. The ePTFE patch graft was visually trimmed according to the size of the right atrium and ventricle and continuously sutured with a 5-0 double-ended, nonabsorbable monofilament suture^k. The coronary groove site was shallowly sutured at a depth of 1–2 mm on the epicardial adipose tissue and reinforced using surgical fibrin sealant^l. After the subtotal pericardiectomy, bupivacaine hydrochloride^m was administered intramuscularly at the incision site to relieve local pain, and a chest tube was inserted into the pleural cavity until pleural effusion decreased. The sternum, thoracic musculature, subcutaneous tissue, and skin were closed in a routine manner. After the surgery, cefazolin sodium hydrate (25 mg/kg, q 8 hr) and enrofloxacinⁿ (5 mg/kg, q 24 hr) were administered parenterally for 3 days. Tramadol hydrochloride^o (4 mg/kg) was administered orally q 12 hr for 3 days to relieve postoperative pain. In the clinical case, the procedure was similar to the method mentioned above; the details are described in the “Results” section.

Results

None of the four healthy intact male dogs exhibited any abnormalities in the preoperative examination. Their mean age was 9.0 ± 1.8 yr, their mean body weight was 9.7 ± 1.3 kg, and they all underwent RAPP.

Using Holter monitoring, SVPCs, PVCs, PVC couplets, and short-run values were evaluated before surgery and at 1, 7, and 30 days after surgery (Table 1). Although the numbers of SVPCs, PVC couplets, and short-run events significantly differed among the evaluations according to Friedman’s test (P = .019, P = .005, and P = .013, respectively), there were no significant differences according to Scheffe's method. Although the number of PVC events tended to increase 1 day after RAPP, no significant difference was observed in the number of events at each evaluation. All arrhythmias described above tended to increase 1 day after RAPP; however, the number of events was decreased or diminished 7 and 30 days after RAPP. Short runs were observed in three of the four dogs, and their maximum heart rates were 112, 126, and 149 bpm. No dog had clinical findings associated with arrhythmia that required antiarrhythmic drugs.

TABLE 1 Comparison of Arrhythmias as Measured with Holter Electrocardiography in the Four Healthy Dogs Before and After Surgery

View Fullscreen

In the echocardiographic examinations, the average values of three consecutive beats were compared before surgery and at 1, 7, and 30 days after surgery (Table 2). Although S’ significantly differed among the evaluations according to Friedman’s test (P = .032), there were no significant differences according to Scheffe's method. None of the other values significantly changed after RAPP.

TABLE 2 Comparison of TAPSE and TDI Values Using Echocardiography in the Four Healthy Dogs Before and After RAPP

View Fullscreen

Based on the above results, we performed RAPP on the golden retriever (Figure 2). The 9 yr old male castrated dog weighed 33.45 kg. Radiographic/ultrasonographic examinations and computed tomography revealed no abnormality of other organs including the lungs and spleen. The anesthesia and RAPP procedures were as described above, although fentanyl^q (5 μg/kg IV, followed by 5 μg/kg/hr constant-rate infusion) and robenacoxib^r (2 mg/kg SQ) were used for intraoperative analgesia. Oral tramadol was used for postoperative analgesia. The tumor was not limited to the right auricle and had visibly spread to the left atrium. The tumor of the right auricle was grasped with a Satinsky blood vessel clamp and resected to the fullest extent possible; however, complete resection was determined to be difficult. Therefore, RAPP followed by pericardiectomy was performed to prevent bleeding from the future recurrent tumor after continuous suturing of the resection side of the right auricle with 5-0 monofilament sutures^k. Chest closure was performed, and a chest tube was placed in the pleural cavity until the effusion decreased. The pathological diagnosis was HSA, and tumor cells were not found in the atrial muscle of the resected side. Clinical signs of arrhythmia were not found; however, lidocaine^s (2 mg/kg by gradual IV infusion, followed by 20 μg/kg/min constant-rate infusion) was administered 2 days after surgery as a result of ventricular tachycardia, with the QRS complex of the PVC close to the preceding T wave (not R-on-T phenomenon). The ventricular tachycardia disappeared ∼24 hr after lidocaine administration for 8 hr followed by oral sotalol^t (1.1 mg/kg q 12 hr). However, sotalol was administered for 2 days only in order to maintain the antiarrhythmic effect. Thereafter, the antiarrhythmic drugs were discontinued, but no arrhythmia was observed. The dog was discharged 7 days after surgery and received adjuvant chemotherapy with doxorubicin at a referral animal hospital. The dog survived for 434 days without cardiac tamponade or right heart failure after RAPP and died from lung metastasis according to the telephonic follow-up. Echocardiographic data regarding right ventricular function were not available because the dog was a significant distance away from our university and could not be brought in for follow-up.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

There were several important findings in our study. First, transient arrhythmia, including SVPCs, PVCs, PVC couplets, and short runs that did not require antiarrhythmic drugs, tended to increase 1 day after RAPP and diminished or decreased 7 and 30 days after RAPP in the healthy dogs. However, transient antiarrhythmic treatment was required in our case of HSA as a result of more serious PVC than that observed in the healthy dogs. Second, the right ventricular function, examined using TAPSE and TDI, was not severely affected by RAPP, although S’ significantly differed among the evaluations. Finally, the ePTFE patch was shown to be a suitable alternative material to the pericardial autograft method and could be used for right atrioventricular patch graft reconstruction.

Dogs who are affected by cardiac HSA have a very poor prognosis, typically as a result of acute cardiac tamponade secondary to right atrial bleeding from tumor rupture or from metastases that primarily arise in the lungs, spleen, and liver.^{1,4,7–10,14} Palliative pericardiocentesis is performed to relieve acute cardiac tamponade because acute cardiac tamponade can cause collapse, syncope, or sudden death. However, pericardiocentesis is occasionally required several times per day or week. Therefore, cardiac HSA is a stressful disease not only for the affected dogs but also for the owners and veterinarians.

Although various treatment options can be attempted separately, outcomes with this approach are less than optimal. For example, pericardiectomy has been performed in dogs with recurrent acute cardiac tamponade secondary to cardiac HSA. In a previous study, two dogs who underwent pericardiectomy alone survived for 1 and 4 mo, respectively, after surgery.¹⁵ However, pericardiectomy alone is not a common procedure in dogs with cardiac HSA because it is associated with the possibility of lethal blood loss from the intrapericardial space accumulating in the thoracic cavity. In addition, 64 dogs who underwent chemotherapy alone with doxorubicin survived for 116 days after diagnosis.²⁸

Tumor resection is a therapeutic option for cases in which the tumor is localized to the right auricle and is performed with thoracotomy or a thoracoscope.^13,15,29 Fifteen dogs treated by a combination of transthoracic tumor resection and partial pericardiectomy survived for 46 days, and 8 dogs treated by a combination of thoracoscopic tumor resection and partial pericardiectomy survived for 90 days (2 of the 8 dogs received chemotherapy).^13,29 In addition, dogs treated with adjuvant chemotherapy combined with doxorubicin, cyclophosphamide, or vincristine following right auricular resection and partial pericardiectomy survived for 175 days after treatment.¹³ Therefore, a combination of tumor resection, partial pericardiectomy, and adjuvant chemotherapy is the best therapeutic treatment option for dogs with cardiac HSA to achieve longer survival times and avoid pericardiocentesis. However, tumor resection is not always feasible because cardiac HSA occurs not only at the right auricle but also at the right atrial body, occasionally neighboring the coronary groove.⁷ In addition, the closed site of the tumor is likely fragile because a visually normal right atrium was reported to rupture immediately after right auricle resection with vascular staples in a dog who required repeated pericardiocentesis.¹⁶ In that case, the dog survived for 260 days after the surgery, in which the right atrium, right auricle, and right ventricle—including the coronary groove—were broadly patched with a pericardial autograft, although an undetailed, transient arrhythmia was described.¹⁶ In addition, cardiac HSA occasionally vastly increased in size and could not be removed without causing a disruption in the right atrial body.⁷ Autogenous pericardial patch graft reconstruction was previously reported in dogs with uncontrolled right atrial bleeding or with large defects in the right atrial body.^16,17 Autogenous pericardial patch grafts are useful for various heart disease reconstructions in dogs, including pulmonic stenosis and tetralogy of Fallot.²⁰ However, an ePTFE patch was used as a right atrioventricular patch graft instead of a pericardial autograft in our study because the ePTFE patch is available in various sizes and thicknesses, and the bleeding from suture-induced pinholes likely arises because of the thinness of the pericardial autografts in healthy dogs. In our study, the ePTFE patch was reinforced with a surgical fibrin sealant because the epicardial adipose tissue was shallowly sutured. As a result, no dogs experienced lethal bleeding at the surgical site in our study.

In a dog who underwent autogenous pericardial right atrioventricular patch graft reconstruction that included the coronary groove, transient arrhythmia occurred as a result of uncontrolled bleeding from the right atrium following right auricular resection, but its course was not reported in detail.¹⁶ In our study, transient arrhythmias including SVPCs, PVCs, PVC couplets, and short runs occurred in the healthy dogs without clinical findings after surgery, but no treatments were required. Although all the arrhythmias tended to increase 1 day after the surgery, they decreased or diminished 7 and 30 days after the surgery. RAPP may induce self-limited arrhythmias; however, similar arrhythmias have been observed in clinically healthy beagles but were less severe than those in our dogs.³⁰ Even in the clinical case, only short-term use of antiarrhythmic drugs was required, although the arrhythmias were more serious than those in the healthy dogs.

In our study, TAPSE was used for the examination of right ventricular systolic function; it measures the apical displacement of the lateral portion of the tricuspid valve annulus during systole from an M-mode recording.³¹ In dogs with pulmonary hypertension, the TAPSE values significantly decreased and were associated with right ventricular dysfunction.²⁶ TDI is a quantitative assessment of segmental myocardial motion for right ventricular function, and TDI values reportedly decreased in dogs with right ventricular dysfunction associated with pulmonary hypertension.²⁷ In our study, E’, A’, and E’/A’ were used for the examination of right ventricular diastolic function, S’ was used for right ventricular systolic function, and global TDI derived from S’*E’/A’ was used for the entire right ventricular function. However, the TAPSE and TDI values, not involving S’, in which a significant difference was revealed with the Friedman’s test but not with the post hoc test, did not significantly change in the dogs who underwent RAPP, although ischemia-induced right ventricular deterioration was predicted. Therefore, RAPP did not appear to be the cause of lethal arrhythmia and right ventricular deterioration in the healthy dogs. Indeed, the clinical case underwent RAPP and experienced no clinical symptoms associated with right heart failure for 434 days.

Our study has some limitations. First, the sample sizes were very small because only four healthy dogs and one client-owned dog with cardiac HSA underwent RAPP. Second, pathological examination of the right ventricle and coronary vessels was not conducted in the four healthy dogs, although it was necessary to confirm the presence of ischemia in those lesions. Finally, right ventricular function was examined using only TAPSE and TDI; however, other echocardiographic evaluations such as the Tei-index should have been conducted.

Conclusion

An ePTFE patch can serve as a suitable alternative material for right atrioventricular patch graft reconstruction, but RAPP might lead to transient arrhythmias, including SVPCs, PVCs, PVC couplets, and short runs, which may or may not require short-term antiarrhythmic drug treatment. In addition, RAPP did not induce significant right ventricle functional deterioration. Therefore, RAPP may be a therapeutic option for uncontrolled bleeding from a right atrial rupture secondary to cardiac HSA or incomplete tumor resection. In our study, the detailed procedure for right atrioventricular patch graft reconstruction using the ePTFE patch along with its impact on right ventricular function and arrhythmogenicity were elucidated. However, it was necessary to study the effectiveness in clinical cases because only one clinical case was included in this study.