Concurrent Splenic and Right Atrial Mass at Presentation in Dogs with HSA: A Retrospective Study

Introduction

Canine hemangiosarcoma (HSA) is a sarcoma that arises from either the vascular endothelium or endothelial precursor cells.¹ This creates a situation where tumor cells readily access the systemic circulation, accounting for the high metastatic rate of this tumor type.^1,2 Common sites of metastasis include the lungs and liver via hematogenous metastasis, and the omentum via direct implantation due to rupture of a splenic or hepatic HSA.³ The right atrium is reported to be a site of concurrent disease for splenic HSA.⁴

A study by Waters et al. (1988) evaluated the metastatic pattern for dogs with splenic HSA. This was a retrospective study that evaluated the necropsy findings of 25 dogs that presented for clinical signs associated with rupture of a splenic HSA.⁴ In that study, four of the dogs were treated surgically and necropsied when they succumbed to their disease. The other 21 dogs were euthanized or died at the time of presentation. In that study, 6/25 dogs had gross evidence of HSA of the right atrium.⁴ Of those dogs, 5/6 had metastasis to other sites concurrently, including the lungs (5/6), liver (3/6), kidney (2/6), omentum (2/6), mesentery, and jejunal serosa. That study also reported intraperitoneal or extraperitoneal metastasis in 22/25 dogs, with the liver, omentum, and mesentery being the most common metastatic sites overall.⁴ Waters et al. (1988) concluded that 25% of dogs that present with splenic HSA have concurrent disease of the right atrium.⁴ This figure has been subsequently reported in multiple papers and textbooks, and is commonly cited to rationalize the need for preoperative cardiac ultrasound for patients with either a presumptive or suspect splenic HSA.^3,5–7 This has resulted in the recommended staging of suspect splenic HSA cases with three-view thoracic radiography and abdominal and cardiac ultrasound.

Three-view thoracic radiography and abdominal ultrasound are generally readily available prior to exploratory laparotomy and splenectomy in these patients. A cardiac ultrasound can be more challenging to coordinate at most centers because they rely on the availability of a cardiologist or a veterinarian with advanced training in cardiac ultrasound. Because most cardiac ultrasounds are typically scheduled procedures and splenectomy for hemoabdomen is often an emergency or urgent procedure, performing a cardiac ultrasound is not always possible preoperatively.

The rate of concurrent splenic and cardiac HSA at the time of staging may not be consistent with Waters’ necropsy study. It has also been suggested by other authors that this rate may be lower in clinical cases.³ The purpose of this study was to retrospectively evaluate the presence of concurrent splenic and right atrial HSA in canine patients at the time of presentation. Specifically, this study aimed to establish whether a cardiac ultrasound was an essential part of preoperative staging for a suspect splenic HSA and if an abdominal ultrasound was an essential part of staging for a suspect cardiac HSA.

Materials and Methods

The medical records for dogs presenting to the Veterinary Teaching Hospital at the Ontario Veterinary College from Jan 1, 1998 to Jan 1, 2009 were reviewed. There were two groups of patients that were evaluated in this study. Group 1 included dogs that presented with a splenic mass, either with or without hemoabdomen, that were fully staged for HSA. Inclusion criteria for dogs in group 1 were: canine patients with a presenting complaint that was referable to splenic HSA; a histologic diagnosis of HSA; and full staging for gross metastasis with thoracic radiographs, abdominal ultrasound, and a cardiac ultrasound. Group 2 included dogs that presented with a right auricular or right atrial HSA that were evaluated for metastatic disease. Inclusion criteria for dogs in group 2 were dogs that presented with clinical signs associated with pericardial effusion and cardiac tamponade due to right atrial HSA, a histologic diagnosis of right atrial HSA, and subsequent evaluation for metastatic disease either antemortem or postmortem.

For dogs in group 1, recorded data included: age; breed; sex; presenting complaint; the presence or absence of hemoabdomen, a cardiac mass, and abdominal metastasis; the site of abdominal metastasis (if present); the site of a cardiac mass (if present); and the presence or absence of pericardial effusion, radiographic signs of pericardial effusion, and clinical signs of pericardial effusion. The presence of abdominal metastasis was determined by abdominal ultrasound by either a radiologist or surgical exploration. A histologic diagnosis of the primary mass and abdominal metastasis (if present) was achieved either by samples submitted postoperatively or by postmortem in cases that were euthanized. A board-certified pathologist evaluated all samples. The presence or absence of a right atrial mass and pericardial effusion was determined by cardiac ultrasound that was performed by a board-certified cardiologist. Radiographic signs of pericardial effusion included an enlarged cardiac silhouette and a globe-shaped heart. Clinical signs of a pericardial effusion included a combination of signs at presentation including: irregular, weak, or thready pulses; muffled heart sounds; jugular venous distension; and electrocardiogram results consistent with pericardial effusion (such as attenuated R wave amplitude and electrical alternans).

For dogs in group 2, recorded data included: age; breed; sex; presenting complaint; the presence or absence of a splenic mass; the presence or absence and location of another metastatic site; and treatment administered. A histologic diagnosis was achieved by either surgical removal of the cardiac mass or by postmortem examination by a board-certified pathologist.

Results

Discussion

In this study, 8.7% (2/23) of the dogs that presented with signs referable to splenic HSA had a concurrent right atrial mass that was found on cardiac ultrasound. Neither of these two dogs had clinical signs of pericardial effusion associated with the cardiac mass. Both dogs were golden retrievers. Golden retrievers were found to be 10.56 times more likely to have a concurrent cardiac mass when presenting for signs associated with a splenic HSA. It is difficult to interpret these findings because of the small number of dogs with concurrent splenic HSA and a cardiac mass in this study. Further, there were no dogs with a splenic HSA and concurrent cardiac HSA that were not golden retrievers. With the small percentage of dogs with concurrent disease, it is hard to make a definitive conclusion regarding breed predisposition. A larger case series would be necessary to determine if golden retrievers are at a higher risk for concurrent cardiac and splenic HSA. Although statistically significant, these results should be interpreted with caution to some degree with respect to breed prevalence. These findings also suggest that when dogs present with clinical signs due to splenic HSA, the rate of a concurrent cardiac mass is lower than previously reported.

It should be noted that the presence of a cardiac mass was considered equivocal in one of the dogs at the time of diagnosis, and the mass was not found on a subsequent ultrasound 4 wk postoperatively. In this same patient, a right atrial mass was found 5 mo postoperatively and was considered unequivocal at that time. This finding highlights the difficulty in definitively determining the presence or absence of a right atrial mass on cardiac ultrasound. In a study of the outcome of surgical treatment of dogs with a right atrial HSA, Weisse et al. (2005) reported that of the 23 dogs that had a preoperative cardiac ultrasound, only 11/23 (48%) of the dogs had a definitive mass diagnosed with cardiac ultrasound.⁸

The findings in this study contrast those reported by Waters et al. (1988), which evaluated the metastatic pattern in dogs that presented with clinical signs associated with a splenic HSA. In that study, 25% (6/25) of the dogs had a concurrent right atrial mass and 4/6 of these dogs were German shepherd dogs.⁴ The increased prevalence noted in the study by Waters et al. (1988) compared with the current study could be due to several reasons. First, both studies have relatively small sample sizes, and it is possible that more accuracy of the prevalence of a concurrent splenic and cardiac mass would be achieved with a larger sample size in both studies. Second, the two groups of dogs are similar, but the inclusion criteria for the studies were different. For example, both studies required that dogs presented with signs referable to a splenic HSA and a histologic diagnosis of HSA; however, Waters et al. (1988) included dogs that were examined with a full necropsy either at the time of diagnosis or the time of death.⁴ In the current study, dogs were included if they had full staging with thoracic radiographs, abdominal and cardiac ultrasound, and a histologic diagnosis of splenic HSA. It is possible that because preoperative staging was done sequentially (compared with a necropsy in which the evaluation for metastasis is done simultaneously), there may have been cases in this study that were not staged with a cardiac ultrasound because of evidence of gross metastasis on thoracic radiographs and abdominal ultrasound. These dogs may have also had a cardiac mass that was undetected and/or not included in the current study, which would have been found on post mortem examination. From a practical standpoint, the inclusion criteria in this study may be a more accurate reflection of the population of dogs that veterinarians see clinically because it is not judicious to continue with cardiac staging in a patient with evidence of gross metastasis on thoracic radiographs or abdominal ultrasound. Finally, this study was attempting to establish the utility of cardiac ultrasound to screen for metastasis prior to surgery, not to determine the absolute presence or absence of cardiac HSA on postmortem examination. The reason for the low percentage of cases with a positive cardiac ultrasound in this study, compared with the percentage of dogs with a cardiac HSA on postmortem in the study by Waters et al. (1988), may be because cardiac ultrasound is an insensitive test compared with postmortem and/or because the absolute number of dogs with a cardiac mass is lower than what was presented in the study by Waters et al. (1988).

Both Waters et al. (1988) and the current study's authors made efforts to incorporate dogs into their respective groups based on clinical signs referable to a splenic mass (Waters et al. [1988] and group 1 in the current study) or a right atrial mass (group 2 in the current study). However, there was definitely a significant amount of overlap in the presenting complaints and clinical signs of dogs that present with a splenic HSA and a right atrial HSA. Presenting complaints such as lethargy, weakness, and anorexia and clinical signs such as collapse, tachycardia, thready pulses, pale mucous membranes, and tachypnea could be attributable to a mass in either location. In Waters et al.’s study (1988), two of the six cases that had a concurrent splenic and cardiac HSA had no evidence of hemoabdomen, and their clinical signs were listed as abdominal mass (n=2), weight loss, and chronic diarrhea (n=1). It is possible that these two dogs actually had clinical signs associated with the cardiac mass, and the splenic mass was the metastatic site. In the current study, a much higher percentage of dogs that presented for a cardiac mass had a concurrent splenic mass than vice versa.

Another difference between the findings in this study compared with Waters et al. (1988) was the breed prevalence. This study found that golden retrievers were significantly more likely to have a concurrent splenic and cardiac mass. Waters et al. (1988) reported that 4/6 dogs with a concurrent splenic and cardiac mass were either German shepherd dogs or German shepherd mixed-breed dogs.⁴ The difference was not statistically significant compared with breeds that did not have a concurrent cardiac and splenic mass.⁴ Variations may reflect differences in the time frame, geography, and breed prevalence of the two studies. Both German shepherd dogs and golden retrievers have been shown to be predisposed to HSA in other studies.^2,8–10

Concurrent intra-abdominal metastasis was found in 7/23 dogs that presented with a splenic mass. This figure may actually underestimate the degree of intra-abdominal metastasis in dogs that present with a splenic HSA. This is because one of the inclusion criteria for dogs included in group 1 was evaluation with a cardiac ultrasound. It is possible that if a dog had gross evidence of intra-abdominal metastasis on abdominal ultrasound, a cardiac ultrasound and treatment may not have been pursued based on these findings, and this case would have been ineligible for this study. This finding does emphasize the fact that widespread metastasis is common with splenic HSA and that, even with staging with an abdominal ultrasound, metastatic lesions may be found on exploratory laparotomy.

Group 2 in this study included dogs that presented with a right atrial HSA and clinical signs associated with pericardial effusion. The metastatic rate in this group of dogs was high. This group of dogs was a different population of dogs compared with group 1 because most of the evaluations were done by postmortem because of the high proportion of dogs that were euthanized or died at presentation. Full antemortem staging was not required for inclusion in group 2. In this population of dogs, the rate of concurrent cardiac and splenic lesions was much higher, with 29% (9/31) of the dogs having lesions at both sites. Of these nine cases, none of the dogs had a hemoabdomen at the time of presentation, and none of the clinical signs could be directly attributable to the splenic mass. The reason for the higher rate of concurrent splenic and right atrial disease in this group of dogs is likely that these dogs represent both cases of right atrial HSA that have metastasized to the spleen via a hematogenous route and cases of synchronous disease.

Potential causes of an HSA of both the spleen and the right atrium concurrently include: metastasis of a primary right atrial HSA to the spleen; metastasis of a splenic HSA to the right atrium; and synchronous tumor development at both sites. If metastatic, given the hematogenous route, it is more plausible that a mass of the right atrium would metastasize to a parenchymatous organ, rather than the other way around. Two of the mechanisms of the arrest of circulating tumor cells into the target tissues support this theory. One mechanism is that tumor cells are arrested at distant vascular beds due to size-dependent trapping of tumor cells. The other mechanism is ligand-dependent adhesion of tumor cells to the endothelial basement membrane, which is exposed between the endothelial cells in small vessels.¹¹ The pressure and velocity of blood flow through the right atrium and auricle and the barrier provided by the endothelial lining of the heart create a barrier to metastasis by the hematogenous route.

In contrast, dogs with a primary splenic HSA are unlikely to metastasize to the right atrium. Because of the small number of cases that were evaluated by postmortem in group 1, it is also possible that small lesions that may not have been noted on radiographs or ultrasound would have been found on postmortem examination. In this study, a much smaller proportion of dogs were examined by postmortem in group 1 compared with group 2.

In this study, five dogs with a right atrial HSA underwent treatment. Two of the dogs were treated with intermittent pericardiocentesis and died within 3–4 mo of presentation. Three of the dogs were treated with surgical resection of the right auricular mass and chemotherapy. These dogs died 3 mo (n=1) and 4 mo (n=2) postoperatively. It is interesting to note that the survival times of these dogs were similar despite the fact that two received palliative treatment and three were treated aggressively. Weisse et al. (2005) recently reported mean and median survival times for dogs undergoing right atrial HSA resection and chemotherapy were 164 days and 175 days, respectively. Without chemotherapy, the dogs in that study had mean and median survival time of 46 days and 42 days, respectively.⁸

In the group 2 dogs in the current study, there was widespread metastasis to other sites. Forty-two percent of the dogs had evidence of metastasis to a site other than the spleen. These sites included the lungs, liver, mesentery, and omentum. Age was found to be a significant risk factor for the development of metastasis to nonsplenic sites, with the risk of nonsplenic metastasis decreasing by 0.457 times with every 1 yr increase in age. This finding is difficult to explain. It is possible that with age, dogs are less likely to develop synchronous disease of the spleen and right atrium, but are more likely to develop right atrial HSA alone. HSA of the right atrium has a high risk of widespread metastasis and may not necessarily affect the spleen. Overall, this finding, although statistically significant, does not have a lot of clinical value.

Although dogs included in group 1 and group 2 were different populations with different inclusion criteria, if all of the dogs in this study were assessed for the presence of a concurrent splenic and cardiac mass, the overall rate would be 11/54 (20.4%). The majority of these cases were right atrial HSA that had presumably metastasized to the spleen and other sites.

Conclusion

Dogs with a splenic HSA that are staged by cardiac ultrasound have a low incidence of a concurrent cardiac mass. Full staging of right atrial HSA with thoracic radiographs and abdominal ultrasound is recommended due to the propensity for widespread metastasis.