J Vet Intern Med 2008;22:148–152
F e l i n e Vi s c e r a l H e m a n g i o s a r c o m a W.T.N. Culp, K.J. Drobatz, M.M. Glassman, J.L. Baez, and L.R. Aronson Background: Feline visceral hemangiosarcoma (HSA) is an uncommon tumor, and the clinical progression and outcome are rarely reported. Hypothesis: The prognosis of feline visceral HSA is poor because of severe clinical signs, anemia, and a high rate of metastasis. Animals: The medical records of 26 client-owned cats with visceral HSA were reviewed. Methods: Multi-institutional retrospective study. Results: The most common historical findings and clinical signs included lethargy, anorexia, respiratory difficulty, collapse, and vocalizing. Eighty-two percent of cats were anemic, and aspartate transaminase was increased in 53% of the study population. Metastatic lung disease was noted in 33% of affected cats. In 75% of the cats, abdominal ultrasonography identified a specific location of HSA. However, ultrasound identification of all multifocal lesions was successful only in 3/9 cats (33%). Tumor location was identified in the following organs: liver (35%), small intestine (31%), large intestine (31%), abdominal lymph node (31%), mesentery (27%), spleen (23%), lung (19%), omentum (12%), brain (8%), pancreas (8%), and diaphragm (8%). Multifocal HSA was noted in 77% of cats. Three cats received adjuvant chemotherapy (doxorubicin). Seventy-one percent of euthanized cats were euthanized within 1 day of diagnosis. The median survival time of the remaining cats (n 5 6) was 77 days (range, 23–296 days). Conclusion and Clinical Importance: Feline visceral HSA is most often multifocal at the time of diagnosis. The prognosis appears poor, and the number of cats receiving chemotherapy is low. Key words: Cat; Malignant; Neoplasia.
emangiosarcoma (HSA), a malignant neoplasm of endothelial cells, has been extensively described in dogs, but limited information is available for cats. In both species, visceral and nonvisceral forms of HSA have been described, with the visceral form more commonly being identified.1,2 Much of the literature on feline HSA consists of isolated case reports.3–10 Currently, information regarding clinical presentation, progression of disease, treatment options, and long-term outcome in cats diagnosed with visceral HSA is limited. In the largest retrospective study to date1 evaluating cats with HSA, 31 cases (both visceral and nonvisceral) were described. In that study, 17/31 cats were diagnosed with visceral HSA (15 abdominal and 2 thoracic). Of those cases, survival times were recorded in 8 abdominal cases and ranged from 0 to 35 weeks. Metastatic disease in the abdominal cavity was identified in cats with abdominal HSA, but no pulmonary metastases were noted. Metastasis was noted most commonly in abdominal lymph nodes (33%) and liver (27%).1 Survival times and prognosis based on specific locations of visceral HSA have not been extensively evaluated. The prognosis of cats with splenic HSA was
H
From the Department of Clinical Studies, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA (Culp, Drobatz, Baez, Aronson); and the Department of Veterinary Clinical Studies, College of Veterinary Medicine, Ohio State University, Columbus, OH (Glassman). Dr Baez is currently associated with the Center for Animal Referral and Emergency Services in Langhorne, PA. Dr Glassman is currently associated with Friendship Animal Hospital in Washington, DC. Presented at the American College of Veterinary Surgeons Meeting, Washington, DC, 2006. Corresponding author: William Culp, 3900 Spruce Street, Philadelphia, PA 19104; e-mail: wculp@vet.upenn.edu.
Submitted December 8, 2006; Revised March 4, 2007; Accepted July 24, 2007. Copyright r 2008 by the American College of Veterinary Internal Medicine 10.1111/j.1939-1676.2008.0022.x
similar to dogs based on 1 report.1 In another study,11 intestinal HSA was reported in 4 cats, of which 1 cat died during recovery from anesthesia, 1 was euthanized after recurrence of rectal prolapse, and 2 were euthanized at the time of tumor diagnosis. Of those cats, metastasis was noted in a mesenteric lymph node in 2 cats.11 Reports of chemotherapy for treatment of feline visceral HSA are rare. Some authors have advocated the use of doxorubicin because of its efficacy in dogs.12–13 The purpose of the present study was to evaluate the clinical presentation, progression, and outcome in a group of cats with visceral HSA. Chemotherapy protocols used are also described.
Materials and Methods Criteria for Selection of Cases A computer search of medical records was performed for cats coded with a diagnosis of HSA that were presented over a 16-year period (1990–2005) to the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania and over a 6-year period (1999– 2004) to the Ohio State University, College of Veterinary Medicine. Medical records of cats with HSA limited only to nonvisceral locations were excluded. In addition, medical records with incomplete data (specifically the location of HSA and treatments pursued) were excluded. The diagnosis of HSA was based on histopathology of surgically obtained tissues or post mortem samples. Histopathologic criteria for the diagnosis of HSA have been described previously.14 Data collected from the medical records included signalment, historical findings, physical examination findings, clinicopathologic data including CBC, serum biochemistry, FeLV/FIV status, and coagulation profile. Clinicopathologic data were obtained from 2 different clinics with different reference ranges, and comments about abnormalities of the data are specific to the individual laboratory. Additionally, results of diagnostic imaging, presurgical stabilization and therapy, the locations of the HSA, treatments administered, recurrence, survival time, and cause of death also were reviewed.
Feline Visceral Hemangiosarcoma
Statistical Analysis Continuous variables were assessed for normality by the Shapiro-Wilks test. Normally distributed variables were described with mean standard deviation, whereas median and range (minimum and maximum values) were used to describe nonnormally distributed variables. Discrete (categorical) variables were described with proportions and percentages and compared between groups by Fischer’s exact test. For all comparisons a P value of o.05 was considered significant. All statistical evaluations were performed by a commercially available statistical software program.a
Results Fifty-eight cats were coded with the diagnosis of HSA. Twenty records were excluded because they contained incomplete data, and 12 records were excluded because the cats had only nonvisceral HSA. The signalment of cats that were excluded was similar to that of cats that were included (data not shown). Twenty-six cats met inclusion criteria. The occurrence of feline visceral HSA in these hospital populations was 0.04%. There were 13 spayed females, 12 neutered males, and 1 intact male. The mean age at diagnosis was 10.2 years old ( 3.97 years). Breeds identified were domestic shorthair (21), domestic longhair (2), Maine Coon (2), and an unspecified breed (1). Historical findings included lethargy (11/26; 42%), anorexia (10/26; 38%), collapse (5/26; 19%), vocalizing (4/26; 15%), respiratory difficulty (4/ 26; 15%), vomiting (3/26; 12%), ataxia (1/26; 4%), and pain on abdominal palpation (1/26; 4%). Median duration of clinical signs was 12.4 days (range, 1–60 days). Abnormal physical examination findings (not all variables were recorded in all 26 cats) included tachypnea (17/20; 85%), dehydration (9/14; 64%), hypothermia (normal range: 100.5–102.51F) (9/20; 45%), depressed mentation (4/13; 31%), heart murmur (7/26; 27%), respiratory difficulty (5/26; 19%), tachycardia (5/26; 19%), bradycardia (3/26; 12%), decreased muscle mass (3/26; 12%), and hyperthermia (2/20; 10%). Only 1 cat (1/26; 3%) had a
149
palpable abdominal mass, and abdominal distension was not noted in any cat. Thoracic radiography was performed in 15/26 (58%) cases. Metastatic lung disease was noted in 5/15 cats (33%), sternal lymphadenopathy in 5/15 cats (33%), and pleural effusion in 2/15 cats (13%). All cats that did not have thoracic radiographs underwent necropsy, and no additional cats were noted to have pulmonary metastasis based on necropsy findings. Abdominal ultrasonography was performed in 12/26 (46%) cats. A specific abdominal tumor site was identified by ultrasonography in 9/12 cats (75%). Of those, ultrasonographic appearance of the mass was recorded in 6 cats and included heteroechoic echogenicity in 3 cases and hypoechoic echogenicity in 3 cases. In the cats with intraabdominal multifocal disease, ultrasonographic identification of all tumor locations was noted in 3/9 cats (33%). An abdominal effusion was also noted in 9 cats on ultrasonographic examination, and all were confirmed to have hemoperitoneum. Table 1 presents the clinicopathologic data. The most common findings were anemia (18/22; 82%), increased aspartate transaminase activity (AST) (8/15; 53%), increased BUN concentration (8/17; 47%), hyperglycemia (7/18; 39%), and increased serum creatinine concentration (5/17; 29%). Six of 7 (86%) cats with heart murmurs were anemic. Further characterization of the anemia was performed in 8 cats and included microcytic hypochromic anemia in 6 cats, macrocytic hypochromic anemia in 1 cat, and microcytic normochromic anemia in 1 cat. Of the cats that had coagulation profiles performed, 2/4 (50%) had prolongation of the prothrombin time and 1/4 (25%) had prolongation of the partial thromboplastin time. All cats tested for FeLV/FIV (11/26; 42%) were negative. In the subpopulation of cats that were euthanized on the 1st day of presentation (15/21; 71%), the mean PCV was 18% ( 5%). The mean PCV of those cats not euthanized on the 1st day of presentation was 31% ( 10%). The mean PCV of cats with hemoperitoneum
Table 1. Clinical laboratory findings. Parameter
n
Mean
Median
Range
Reference Range
% of Cats below Reference Range
% of Cats above Reference Range
PCV TP WBCa Platelets Glucosea BUN Creatinine Albumin ALT ASTa ALP GGT PT PTTa
22 16 17 6 18 17 17 16 16 15 8 6 4 4
21.60% 5.8 g/dL — 157,000/mL — 28 mg/dL 1.6 mg/dL 2.8 g/dL 57 U/L — 33 U/L 5.3 U/L 11.73 seconds —
19.50% 5.7 g/dL 13,800/mL 141,500/mL 157 mg/dL 32 mg/dL 1.7 mg/dL 2.6 g/dL 49.5 U/L 43 U/L 33 U/L 6 U/L 10.3 seconds 14 seconds
6–43% 4.2–10.9 g/dL 1,200–49,600/mL 48,000–330,000/mL 58–613 mg/dL 9–49 mg/dL 0.3–2.8 mg/dL 1.8–3.9 g/dL 16–131 U/L 15–952 U/L 14–65 U/L 0–8 U/L 7.7–100 seconds 7.7–100 seconds
25–40% 4.1–7.9 g/dL 4,004–18,700/mL 175,000–500,000 /mL 67–168 mg/dL 15–32 mg/dL 1–2 mg/dL 2.4–3.8 g/dL 33–152 U/L 1–37 U/L 22–87 U/L 5–19 U/L 6.8–10.2 seconds 10.7–16.4 seconds
82 0 12 33 6 6 18 25 13 0 25 17 0 0
5 19 24 0 39 44 29 6 0 53 0 0 50 25
a Parameters are not normally distributed. AST, aspartate transaminase.
150
Culp et al
(20 2%) was lower than cats without hemoperitoneum (25 4%) but this difference was not statistically significant (P 5 .1455). Similarly, 11/14 cats (79%) with hemoperitoneum were anemic at presentation, whereas 5/8 (63%) cats without hemoperitoneum were anemic, but this difference also was not statistically significant (P 5 .624). One of the cats with pleural effusion had thoracocentesis performed, which revealed a mixed hemorrhagic and chylous effusion. Abdominal effusion was noted in 14/26 (54%) cats at surgery or ultrasound examination and was identified as hemoperitoneum based on cytology or the hematocrit of the effusion in all cases. More cats (10/15; 67%) that were euthanized on the day of presentation had hemoperitoneum compared to cats that were not euthanized on the 1st day (3/7; 43%), but this difference was not statistically significant (P 5 .376). One cat diagnosed with hemoperitoneum was lost to follow-up and was not included in the statistical analysis. The use of antibiotics was recorded in 10 cats. Three of 10 cats (30%) received antibiotics: either doxycyclineb (10 mg/kg PO q24h) or ampicillinc (22 mg/kg IV q8h). Blood products were utilized in 11 cats. Six cats (23%) received fresh whole blood, 2 (8%) received packed red blood cells, 2 (8%) received oxyglobind (unknown dose), and 1 (4%) received fresh frozen plasma. Cats with hemoperitoneum more frequently (5/14; 36%) received blood products compared with cats without hemoperitoneum (2/11; 18%), but this difference was not statistically significant (P 5 .386). The locations of HSA were determined by surgery alone in 9 (35%) cats, by necropsy alone in 12 (46%) cats, and by both necropsy and surgery in 5 (19%) cats. Specific visceral sites of HSA included liver (9/26; 35%), small intestine (8/26; 31%), large intestine (8/26; 31%), lymph node (8/26, 7 mesenteric and 1 iliac; 31%), mesentery (7/26; 27%), spleen (6/26; 23%), lung (5/26; 19%), omentum (3/26; 12%), brain (2/26; 8%), pancreas (2/26; 8%), and diaphragm (2/26; 8%). Twenty of 26 (77%) cats had HSA in more than 1 location. Four cats had cutaneous HSA in addition to visceral HSA. The other locations in those 4 cats included liver and brain in case 1, lymph node and lung in case 2, brain and lung in case 3, and lung in case 4. Fourteen of 26 (54%) cats underwent exploratory laparotomy. The decision to perform exploratory laparotomy was based on the presence of a surgically resectable lesion identified on abdominal ultrasound examination or the presence of hemoperitoneum. Nine of the 14 (64%) cats that underwent surgery had hemoperitoneum. The surgical procedures performed included splenectomy (2), liver lobectomy (3), intestinal resection and anastomosis (3), and abdominal exploratory and biopsy (6). Multiple locations of HSA were identified intraoperatively in 7/20 (35%) cats that were eventually diagnosed with multifocal disease. The locations of HSA identified in the 14 cats that underwent surgery included liver (5/14; 36%), small intestine (3/14; 21%), large intestine (3/14; 21%), spleen (1/14; 7%), pancreas (1/14; 7%), diaphragm (1/14; 7%), and lymph node (1/14; 7%).
Six cats were euthanized intraoperatively. Of the 8 surviving cats, 2 were euthanized within the next day because of worsening of their condition, and 6 survived to be discharged. Of the cats not receiving surgery, 2 survived to be discharged. Chemotherapy (doxorubicin, 1 mg/kg) was started 7 days post operatively in 3 cats. The cats received 6 doses, 3 doses, and 1 dose, respectively. Two cats that underwent laparotomy had recurrence of HSA. Recurrence occurred in both cats in the same organ as the previously removed HSA: in the liver 3 months after surgery and in the duodenum 15 days after surgery. Recurrence was identified at necropsy. Twenty-one of 26 (81%) cats were euthanized, 1 cat died, and 4 cats were lost to follow-up. Cats lost to follow-up were censored from the survival data. Of those cats euthanized, 15/21 (71%) were euthanized within 1 day of diagnosis (day of presentation). The median survival time of the remaining euthanized cats (6 cats) was 77 days (range, 23–296 days). Two of the 3 cats receiving chemotherapy survived longer than 100 days.
Discussion The prevalence of HSA (both visceral and nonvisceral) has been reported to be 1.1 to 1.7% in studies evaluating large populations of cats.15–17 In this combination of hospital populations, the frequency of feline visceral HSA was 0.04%. Similar to previous studies, the cats in this study were older (mean age, 10.2 years) and were presented with relatively nonspecific clinical signs such as lethargy and anorexia.1,18–20 In the present study, cats also were brought to the veterinarian for signs such as collapse (19%), vocalizing (15%), and tachypnea (15%). Many of the common physical examination findings can be attributed to anemia (eg, tachypnea, hypothermia, heart murmur, respiratory difficulty, tachycardia, depressed mentation) or stress. Abnormalities in hydration status are likely associated with the lethargy, anorexia, or depressed mentation demonstrated by many of these cats. Dogs with hemoperitoneum tend to present with palpable abdominal masses and abdominal distension, but this was not the case in the cats described in this study. Eighteen of 22 (82%) cats were anemic on admission (Table 1). Anemia is a common finding in patients with aggressive tumors such as HSA, especially when hemoperitoneum is present. This finding is also consistent with the presence of a chronic disease process in these patients. Of the cats with a heart murmur, 6/7 (86%) were anemic. HSA was most commonly diagnosed in the liver (9/26; 35%) and intestines (small intestine, 8/26, 31%; large intestine, 8/26, 31%), which varies from other reports that cite the spleen as the most common visceral location in cats.1,16 Seventy-seven percent of the cats had HSA identified in more than 1 location, indicating a high metastatic potential. All of the cats with hepatic HSA had HSA identified in multiple locations, suggesting that hepatic HSA, in particular, is highly metastatic or that
Feline Visceral Hemangiosarcoma
the liver is a common site for metastasis in cats. Hepatic HSA was found in only 2 cats with increased AST. In our study population, 2 cats were found to have HSA in the brain and 2 were found to have pancreatic HSA. Pancreatic HSA in the cat is rare, with only 2 cases reported previously.16,21 Cerebral metastatic HSA has been documented previously in a cat and was believed to have a primary location in the subcutaneous tissue.8 Both of the cats with brain HSA in this study also had HSA identified in another location, including liver, muscle, and skin in 1 cat and lung and skin in the other cat. One of these cats had clinical signs (ie, ataxia) suggestive of brain disease. There were 2 cases of diaphragmatic HSA and 3 cases of omental HSA in our study population. To the authors’ knowledge, diaphragmatic and primary omental HSA in cats have not been reported previously. Both of the cats with diaphragmatic HSA had liver HSA, and 1 cat had pancreatic HSA as well. Both cats with diaphragmatic HSA also were diagnosed with hemoperitoneum, suggesting rupture of the primary site of HSA. The possibility of metastatic spread secondary to rupture of a primary abdominal HSA has been documented in the dog,22 and the presence of diaphragmatic HSA and hemoperitoneum is suggestive of a similar process in the cats reported here. One cat had HSA diagnosed only in the omentum and had hemoperitoneum secondary to bleeding from that site. Imaging studies were shown to be a beneficial component of the diagnostic evaluation of these cats. Thoracic radiographs correctly identified all 5 cats with pulmonary HSA. Of those, 3 cats had abdominal HSA (1, small intestine; 2, abdominal lymph nodes). This finding varies from a previous report1 that did not note pulmonary metastatic disease in any cats at the time of diagnosis of abdominal HSA. In addition, the use of abdominal ultrasonography should be considered before surgery in these cases because it will most likely assist in surgical planning. In 75% of the cats in which an abdominal ultrasound examination was utilized, a specific location of HSA was correctly identified. However, when employing abdominal ultrasonography, clinicians should recognize that multifocal locations in the abdomen may not be identified. In this study, two thirds of cats had at least 1 additional HSA location that was not identified by ultrasound. Surgical treatment was pursued in approximately half the cases. Nontraumatic hemoperitoneum in dogs is generally considered an indication for surgery and this rationale was most likely employed in the cats that were taken to surgery in our study. Of the cats taken to surgery, removal of macroscopic disease was attempted in 8 (ie, splenectomy, liver lobectomy, and intestinal resection and anastomosis). In 6 cats, the severity of the findings at surgery resulted in intraoperative euthanasia. Retrospectively, it is difficult to determine why the decision to euthanize was pursued early after the initial presentation. Pursuing treatment (both surgery and adjuvant chemotherapy) has prolonged survival times in dogs,12 and perhaps additional therapy in those cases
151
in which it is appropriate should be pursued more often in affected cats. In this study population, only 2 cats received chemotherapy. The efficacy of chemotherapy in cats with HSA is difficult to evaluate because there are very few reports on this subject in the veterinary literature. One study13 utilized chemotherapy (vincristine, doxorubicin) in 2 cats (both with distant metastasis but unidentified primary locations). The cat that received vincristine in that study survived for 77 days, and the cat that received doxorubicin died the day of treatment from respiratory arrest. Chemotherapy appears to increase median survival time in dogs with visceral HSA,23 and future studies on its use in cats with HSA are warranted. The limitations of this study include the small number of cases and its retrospective nature. In a small study population, survival data can be difficult to assess. In addition, 4 cats did not have necropsies performed, and the extent of the microscopic tumor location in those cats is not known. Visceral HSA appears to be highly metastatic in cats, and the prognosis is guarded. Future work should focus on adjuvant therapy (eg, chemotherapy, radiation therapy) after surgical excision or the use of chemotherapy as a sole treatment.
Footnotes a
STATA 8.0 for Windows, College Station, TX Sandoz Inc, Broomfield, CO c IVAX Pharmaceuticals Inc, Miami, FL d Biopure, Cambridge, MA b
References 1. Scavelli TD, Patnaik AK, Melhaff CJ, et al. Hemangiosarcoma in the cat: Retrospective evaluation of 31 surgical cases. J Am Vet Med Assoc 1985;187:817–819. 2. McAbee KP, Ludwig LL, Bergman PJ, et al. Feline cutaneous hemangiosarcoma: A retrospective study of 18 cases (1998–2003). J Am Anim Hosp Assoc 2005;41:110–116. 3. Esplin DG, Carr SH. Cutaneous hemangiosarcoma in a cat. Feline Pract 1986;16:38–40. 4. Swayne DE, Mahaffey EA, Haynes SG. Lymphangiosarcoma and hemangiosarcoma in a cat. J Comp Pathol 1989;100:91–96. 5. Mughannam A. Subcutaneous hemangiosarcoma in the cat. Calif Vet 1991;45:28–29. 6. Gilger BC, McLaughlin SA, Whitley RD, et al. Ortbital neoplasms in cats: 21 cases (1974–1990). J Am Vet Med Assoc 1992;201:1083–1086. 7. Lewis DC, Harari J. Peripheral arteriovenous fistula associated with a subcutaneous hemangiosarcoma/hemangioma in a cat. Feline Pract 1992;20:27–29. 8. Tudor K, Greenlee P. Cerebral metastatic hemangiosarcoma in the cat. Feline Pract 1994;22:20–21. 9. Merlo M, Bo S, Ratto A. Primary right atrium hemangiosarcoma in a cat. J Feline Med Surg 2002;4:61–64. 10. Multari D, Vascellari M, Mutinelli F. Hemangiosarcoma of the third eyelid in a cat. Vet Ophthalmol 2002;5:273–276.
152
Culp et al
11. Sharpe A, Cannon MJ, Lucke VM, et al. Intestinal hemangiosarcoma in the cat: Clinical and pathological features of four cases. J Small Anim Pract 2000;41:411–415. 12. Chun R. Feline and canine hemangiosarcoma. Compend Contin Educ Pract Vet 1999;21:622–629. 13. Kraje AC, Mears EA, Hahn KA, et al. Unusual metastatic behavior and clinicopathologic findings in eight cats with cutaneous or visceral hemangiosarcoma. J Am Vet Med Assoc 1999;214: 670–672. 14. Goldschmidt MH, Hendrick MJ. Tumors of the skin and soft tissues. In: Meuten DJ, ed. Tumors in Domestic Animals, 4th ed. Ames, IA: Iowa State Press; 2002:99–101. 15. Schmidt RE, Langham RF. A survey of feline neoplasms. J Am Vet Med Assoc 1967;151:1325–1327. 16. Engle GC, Brodey RS. A retrospective study of 395 feline neoplasms. J Am Anim Hosp Assoc 1969;5:21–31. 17. MacVean DW, Monlux AW, Anderson PS, et al. Frequency of canine and feline tumors in a defined population. Vet Pathol 1978;15:700–715.
18. Post G, Patnaik AK. Nonhematopoietic hepatic neoplasms in cats: 21 cases (1983–1988). J Am Vet Med Assoc 1992;201: 1080–1082. 19. Lawrence HJ, Erb HN, Harvey HJ. Nonlymphomatous hepatobiliary masses in cats: 41 cases (1972–1991). Vet Surg 1994;23:365–368. 20. Henry CJ, Turnquist SE, Smith A, et al. Primary renal tumors in cats: 19 cases (1992–1998). J Feline Med Surg 1999;1: 165–170. 21. Ottenjann M, Kohn B, Weingart C, et al. Rupturiertes ha¨mangiosarkom der milz als ursache eines ha¨moperiotneums bei 4 katzen. Kleinterpraxis 2003;48:345–352. 22. Zhao D, Yamaguchi R, Tateyama S, et al. Canine splenic hemangiosarcoma with abdominal dissemination. J Vet Med Sci 1994;56:753–755. 23. MacEwen EG. Hemangiosarcoma. In: Withrow SJ, MacEwen EG, eds. Small Animal Clinical Oncology, 3rd ed. Philadelphia, PA: WB Saunders Co; 2001:639–646.