Thrombocytopenia in Cats:

Transcription

1 Thrombocytopenia in Cats: A Retrospective Study of 41 Cases Holly L. Jordan, DVM, Carol B. Grindem, DVM, PhD, and Edward B. Breitschwerdt, DVM The prevalence of feline thrombocytopenia (<200,000 platelets/pl) at North Carolina State University, College of Veterinary Medicine Teaching Hospital, from January 1985 to March 1990, was 1.2% (41/3300). Cats were divided into six categories based on clinical diagnoses: 29% (I 2/41) had infectious disease, 20% (8/41) had neoplasia, 7% (3/41) had cardiac disease, 2% (1/41) had primary immune-mediated disease, 22% (9/41) had multiple diseases, and 20% (8/41) had disorders of unknown etiology. The mean platelet count for all thrombocytopenic cats was 52,00O/pL f 46,OOO/pL (I SD) with a range of ,000/pL. No significant differences were found between groups with respect to platelet count, packed cell volume, or white blood cell count, though anemia and leukopenia were common among the cats as a whole. Bleeding disorders (hemorrhage or thrombosis) were observed in 29% (12/41) of thrombocytopenic cats and were more likely to be associated with neoplasia, cardiac disease, and platelet counts less than or equal to 30,00O/pL. Disseminated intravascular coagulopathy was diagnosed in 12% (5/41) of the cats. Infections and/or neoplasia affecting the bone marrow were the most common diseases associated with thrombocytopenia. Feline leukemia virus and myeloproliferative neoplasia accounted for approximately 44% (18/41) of the specific diagnoses in thrombocytopenic cats. (Journal of Veterinary Internal Medicine 1993; 7: Copyright by the American College of Veterinary Internal Medicine.) ALTHOUGH thrombocytopenia is believed to be the most common platelet disorder recognized in feline medicine,' the actual prevalence of low platelet counts is unknown, and diseases associated with thrombocytopenia in cats have not been characterized. The literature on thrombocytopenia in cats consists of individual case report~~-~ and studies related to particular experimental condition^.^,' Less is known about the general causes of thrombocytopenia in feline populations, such as disseminated intravascular coagulopathy (DIC), or the problems associated with platelet deficiency, such as hemorrhage. To better understand the clinical relevance of thrombo- From the Departments of Microbiology, Pathology, and Parasitology (Jordan, Grindem), and Companion Animal and Special Species Medicine (Breitschwerdt), North Carolina State University, College of Veterinary Medicine, Raleigh, NC. Accepted February 26, Presented at the 25th Annual Meeting of the American Society for Veterinary Clinical Pathology. Phoenix. AZ, December I I Reprint requests: Carol B. Grindem, North Carolina State University. College of Veterinary Medicine, 4700 Hillsborough St, Raleigh, NC Copyright by the American College of Veterinary Internal Medicine /93/ I $3.00/0 cytopenia, a retrospective study was conducted specifically examining the prevalence of thrombocytopenia and associated diseases in a feline hospital population. Materials and Methods Feline hematology reports from the North Carolina State University, College of Veterinary Medicine (NCSU-CVM) from January 1985 to March 1990 were reviewed. A platelet count of less than 200,00O/pL (NCSU-CVM reference range = 300, ,000/pL) and a complete medical record were required for inclusion in this study. Medical records were reviewed to summarize the clinical diagnoses, signalment, and results of the complete blood cell count, coagulation profile, tests for feline leukemia virus (FeLV) antigen or antibody, and tests for feline immunodeficiency virus (FIV) antibody, and the clinical manifestations of bleeding and thrombosis. Based on clinical diagnoses, cats were divided into six disease categories for data analysis: infectious disease, neoplastic disease, cardiac disease, primary immune-mediated disease, multiple diseases, and unknown (Table 1). Platelet counts were performed on whole blood anticoagulated with tripotassium ethylenediaminetetraacetic 26 1

2 262 JORDAN, GRINDEM, AND BREITSCHWERDT Journal of Veterinary Internal Medicine TABLE I. Thrombocytopenic Cats by Disease Group Number of Purebreds Age in Years Disease Group Cats (96) (70) (Mean * SD) Males (7%) Specific Diagnosis Infectious 12 (29) 4/12 (33) 3.3 * 2.3 7/12 (58) FeLV (6), FIV (2), FIPV (2), FPLV (I), T. gvndii ( I ) Neoplastic 8 (20) 018 (0) 10.0 k 3.2 8/8 (100) Leukemia (2), HSA (2), Pulmonary ACA and Leukemia (I), LSA (I), Hepatic ACA (I), LSA and Leukemia ( I ) Cardiac 3 (7) 113 (33) 4.8 * (33) HCM (2). Eisenmenger s syndrome ( I) Immune-mediated I (2) 0/I (0) 0.7 1/1 (100) Primary immune-mediated thrombocytopenia Multiple 9 (22) 319 (33) 5.0 * 4.4 7/9 (78) All cats had two or more of the following: Leukemia (7), FeLV (5). LSA (2). H. fdis (2), AIHA ( I), HCM (I), Myasthenia gravis (I), Trauma ( I) Unknown 8 (20) 018 (0) 8.4 k 3.4 1/8 (13) None FPLV = feline panleukopenia virus. HSA = hemangiosarcoma. ACA = adenocarcinoma, LSA = lymphosarcoma, HCM = hypertrophic cardiomyopathy, AIHA = autoimmune hemolytic anemia. acid. Hemacytometer counts were made on a phase-contrast microscope using an ammonium oxylate dilution system.* When hemacytometer counts were unavailable, estimated platelet counts were made from blood smears: the number of platelets observed in 25 IOOOX fields was averaged, then multiplied by a factor of 20,000 to obtain an estimated platelet count/pl. One-way analysis of variance was used to test for differences between the diagnostic groups with respect to age, platelet count, white blood cell count (WBC), and packed cell volume (PCV). Chi-square analysis was used to test for differences in breed, sex, and occurrence of bleeding disorders (hemorrhage or thrombosis). A P value of.05 or less was considered significant. Results Forty-one of 3300 (1.2%) feline patients examined at NCSU-CVM between January 1985 and March 1990 had platelet counts less than 200,OOO/pL (Fig 1). The average platelet count for all 41 cats was 52,OOO/pL * 46,0OO/pL (1 SD) with a range of ,0OO/pL and a median of 40,OOO/pL. Most cats (88%) had less than 100,000 platelets/pl. Thrombocytopenic cats were similar to the total hospital cat population with respect to age, sex, and breed. Among the six diagnostic categories (Table I), animals in the infectious group were younger on average (mean = 3.3? 2.3 years), whereas cats with neoplasia were significantly older (mean = 10.0 k 3.2 years). Cats having neoplasia or multiple disorders tended to be male ( 100% and 78%, respectively). Cats with unknown diagnoses tended to be female (88%). There were no significant differences between disease categories with respect to breed, platelet count, WBC count, or PCV (Table 2), * Thrombo-Plus, Sarstedt, Germany, or Unopette 5855, Becton- Dickinson. Rutherford, NJ. though anemia and leukopenia were common among the cats as a whole. The largest group with thrombocytopenia included those cats with infectious diseases. Thrombocytopenia was associated with four viral infections (Table l), FeLV, FIV, feline infectious peritonitis virus (FIPV), and panleukopenia; and one protozoal agent, Toxoplasma gondii. Infectious agents in cats from the multiple diseases group included FeLV and Hemobartonellajelis. Thirtyseven of the 41 thrombocytopenic cats had been tested for FeLV antigen or antibody; 11 (30%) were infected. Testing for FIV antibody was not begun until 1988, so only 13 thrombocytopenic cats were tested for this viral infection; 2 ( 15%) had positive results. This is in contrast to a much lower prevalence of FeLV and FIV in the total hospital cat population during this time, of approximately 3.4% and 1%, respectively. Leukopenia was a common finding in cats with infections and was most likely related to the predominance of viral etiologies. Sixteen of the 41 thrombocytopenic cats had one or more neoplastic disorders (Table 1). Thirteen cats had round-cell neoplasia (leukemia, lymphosarcoma). Two cats had hemangiosarcoma, and 2 cats had adenocarcinomas (hepatic or pulmonary). Three cats in the neoplasia group and 2 cats with neoplasia in the multiple diseases group showed hemorrhagic problems, including bruising, epistaxis, hemorrhagic enteritis, hematemesis, and retinal hemorrhage (Table 2). Cats with neoplasia were frequently anemic; most had chronic nonregenerative anemia. No cats received chemotherapy or radiotherapy prior to blood sampling. Four of 4 1 thrombocytopenic cats had cardiac disease (Table 1). Two cats had hypertrophic cardiomyopathy; 1 had a congenital heart disorder, Eisenmenger s syndrome; and 1 cat in the multiple diseases group had hypertrophic cardiomyopathy and leukemia. All 3 cats in the cardiac diseases group had hemostatic disorders, including aortic or pulmonary thromboembolism, hemothorax, and DIC (Table 2).

3 VOl. 7. NO FELINE THROMBOCYTOPENIA 263 Number of Cats Platelets X looo/ul FIG. I. Frequency distribution of platelet counts in 41 thrombocytopenic (< 200,000 platelets/bl) feline patients examined at NCSU- CVM between January 1985 and March Primary immune-mediated thrombocytopenia was diagnosed in one cat (Table 1). This cat s platelet count was one of the lowest ( 1 OOO/pL) of all 4 1 cats. He had bleeding from multiple sites. Immunosuppressive doses of prednisone were curative. Two cats in the multiple disease group had a concurrent immune-mediated disease: autoimmune hemolyticanemia with FeLV and myasthenia gravis with lymphosarcoma. Of the nine cats in the multiple disease group, three had both FeLV and myeloproliferative disease (Table I). Hemic neoplasia and/or FeLV occurred with autoimmune hemolytic anemia, myasthenia gravis, cardiomyopathy, hemobartonellosis, or trauma (car accident). Eight cats presented with a variety of complaints, including anorexia, weight loss, dermatitis, neurologic dysfunction, urinary tract infection, hyphema, and idiopathic pleural effusion; however, no cause of thrombocytopenia was identified. Two of these cats presented in DIC of unknown origin. One cat developed jugular thrombosis. In all, 14 cats had one or more signs of abnormal hemostasis (Table 2). Hemorrhage was observed in nine cats (i.e., epistaxis, bruising, retinal hemorrhage, hyphema, hemothorax, hematemesis, hemorrhagic enteritis). Petechiation was observed in only two cats. Three cats were diagnosed with thrombotic problems (jugular, pulmonary, or aortic). DIC was diagnosed in five cats, based on clinical presentation and/or coagulation data. Though often noted in dogs with DIC, schistocytes were not observed in any ofthese cats. One cat with DIC had hemangiosarcoma, one had hepatic adenocarcinoma, two were not diagnosed, and one had Eisenmenger s syndrome. Discussion Thrombocytopenia was observed in 1.2% (41/3300) of cats in this study. This prevalence is at best an approximation because obtaining accurate platelet counts with feline blood is difficult. Automated counters frequently fail to accurately count cat platelets because of the large platelet volume, overlap in platelet and erythrocyte sizes, and the tendency for cat platelets to clump. 2 Manual counts using either a hemacytometer or a blood smear remain the most practical techniques for many veterinary laboratories. Infectious diseases, particularly viral diseases, were associated with feline thrombocytopenia more often (46% or I9/4 1 of thrombocytopenic cats at NCSU-CVM) than other types of disease in this study. This is in contrast to nonviral inflammatory etiologies that predominate in thrombocytopenic dog^'^,'^ and people. Feline leukemia virus infections were identified in 27% (11/41) of thrombocytopenic cats. This virus has been associated with both decreased platelet numbers and increased platelet v~lume.~ Most bone marrow cells, including megakaryocytes and platelets, can be infected.16 Platelets infected with FeLV have a reduced life span17 potentially TABLE 2. Hematologic Data of Thrombocvtouenic Cats PlateletsluL Cats With Mean f SD Anemic Leukopenic Hemostatic Disease Group (Range) Cats ((51) Cats (%) Disorders ((5)) Infectious 6 I.000 i 43,000 5/12 (42) 8/12 (67) 1/12 (8) ( I ,000) Neoplastic & 3 1,000 ( ) 618 (75) 218 (25) 318 (38) Cardiac f (9, ,000) 1/3 (33) 0/3 (0) 313 ( 100) Immune-mediated I (100) 011 (0) l/l (100) Multiple i 33,000 ( I00,000) 719 (78) 319 (33) 219 (22) Unknown 55,000 i ( I ) 318 (38) 218 (25) 418 (50) Anemia defined as PCV < 24% Leukopenia defined as WBC < 550O/pL. Hemostatic disorders defined as clinically evident hemorrhage or thrombosis or DIC.

4 264 JORDAN, GRINDEM, AND BREITSCHWERDT Journal of Veterinary Internal Medicine related to surface immune complex deposition via platelet Fc receptors.18 Infected megakaryocytes may produce larger platelets via alterations in the demarcation membrane ~ystem.~ The effect of FeLV infection on platelet function and hemostasis appears uncertain because bleeding is uncommon in FeLV-infected cats, despite thrombocytopenia. Two of I 1 FeLV-infected cats in this study developed hemorrhage (platelet counts were 84,00O/pL and 30,00O/pL), perhaps because of platelet dysfunction or other coagulopathy. Thrombocytopenia is a variable feature of infection with other feline viruses such as FIPV,6 FIV,I9 and panleukopenia.20 Like FeLV, FIPV is internalized by cat platelets in vitro.21 Widespread vasculitis and inflammation in cats experimentally infected with FIPV reduce platelet numbers and can culminate in DIC6 In functional studies, platelets from cats infected with FIPV show enhanced aggregation responses.22 Such platelet hyperaggregability may play a key role in promoting intravascular coagulation. Neoplasia induces thrombocytopenia by a variety of mechanisms, including myelophthisis, hemorrhage, microangiopathies, DIC, and immune-mediated mechanism~.*~,*~ Thirty-nine percent ( 16/4 1) of the thrombocytopenic cats had neoplasia; nearly a quarter had circulating neoplastic cells. Myelophthisis is a common cause of thrombocytopenia in neoplastic conditions, especially leukemia;24 four of seven bone marrow aspirates obtained from the cats with neoplasia contained abundant neoplastic cells with decreased numbers of normal blood cell precursors. Hemorrhage contributed to platelet utilization in many (31% or 5/16) of the cats with neoplasia. Vascular lesions and microangiopathies such as those observed with hemangiosarcoma may cause platelet consumption and destruction. Endothelial cell damage, platelet disruption, and tumor-induced procoagulant factors may combine to precipitate DIC and hemorrhage. Both cats with hemangiosarcoma had cutaneous hemorrhages, and one had recurrent episodes of acute DIC. Immune-mediated mechanisms are potentially important in this group; antimegakaryocyte antibodies have been shown in dogs with a variety of tumors.25 Thrombocytopenia has not been previously reported to be a feature of feline cardiomyopathy, although platelet functional abnormalities have been identified. Platelets from cardiomyopathic cats26 and taunne-deficient cats2 aggregate more readily in vitro than platelets from unaffected cats. Only one cardiomyopathic cat in this study had plasma taurine concentration measured, and it was within normal limits. Coagulation abnormalities, including DIC, occur in cardiomyopathic cats with aortic thromboembolism.28 Two of three cardiomyopathic cats in this study developed clinically evident thromboses. Neither cat was diagnosed with DIC, although the cat with Eisenmenger s syndrome developed DIC and hemothorax. Immune-mediated thrombocytopenia (IMT) was the least common cause of low platelet counts in this study. The inciting factor is often unknown, and it is a difficult diagnosis to verify because it requires antiplatelet or antimegakaryocyte antibody determination. Other reports of autoimmune3 or idiopathic thromb~cytopenia~ in cats are rare. Suspected IMT was observed in a cat with systemic lupus erythematosus2 and in cats receiving propylthi~uracil.~~ Thrombocytopenia occurs sporadically in myasthenic human patient^.^' The myasthenic cat in this study also had lymphosarcoma, making the cause of thrombocytopenia uncertain. Hemorrhage was observed more often in cats with platelet counts of 30,0OO/pL and lower. One of the most common sites for hemorrhage in thrombocytopenic cats is the gastrointestinal tract;31 however, no particular site predisposition was observed in this study. Low platelet counts alone do not appear to cause bleeding, because four of nine cats with counts less than 10,000 platelets/ pl did not bleed. One explanation is that sick cats tend to be inactive and are, therefore, less likely to cause vascular trauma, which could initiate hem~rrhage.~~ Other factors involved in susceptibility to bleeding include the rate of platelet loss, platelet function, and stability of endothelial membranes. Though 12% (5/ 12) of the cats had DIC, this number may have been higher if coagulation profiles had been obtained consistently (only nine cases had coagulation data). Disseminated intravascular coagulopathy was not necessarily a terminal event in the affected cats in this study; three of the five cats with DIC survived with treatment. In one study of thrombocytopenic dogs, the prevalence of DIC was approximately 36%. This species difference may in part be caused by the predominance of vascular diseases (i.e., dirofilariasis, ehrlichiosis) asso- ciated with canine thromb~cytopenia,~~ but comparison is hampered by the lack of coagulation data in cats. In our hospital, the prevalence of feline thrombocytopenia was low ( 1.2%). Infections and neoplasia affecting the bone marrow were the most common diagnoses in thrombocytopenic cats. Feline leukemia virus and/or myeloproliferative neoplasia specifically accounted for nearly 44% ( 18/4 1 ) of the diagnoses. Less common disorders included cardiac disease, immune-mediated disorders, and trauma. The cause of thrombocytopenia could not be determined in 20% (8/41) of the cats. Disseminated intravascular coagulation was observed in 12% (5/41) of the cats and was associated with neoplasia, cardiac disease, and unknown etiologies. Hemostatic disorders (hemorrhage or thrombosis) were present in 29% (12/41) ofthrombocytopenic cats, and when etiologies could be identified, were more often associated with neoplasia and cardiovascular disease. Hemorrhage in thrombocytopenic cats is difficult to predict from platelet counts, though cats with platelet counts less than or

5 VOl. 7. NO FELINE THROMBOCYTOPENIA 265 equal to 30,00O/pL were more likely to have hemostatic disorders. Platelet numbers and morphology should be included in the hematology profile for any sick cat. For thrombocytopenic cats, coagulation panels should be obtained to identify possible DIC. The most common potential etiologies of thrombocytopenia may be identified with FeLV tests and bone marrow aspirates. Careful examination for neoplasia and evaluation of cardiac function may also be warranted. Because thrombocytopenic cats with neoplasia, cardiac disease, or immune-mediated thrombocytopenia appear to be at risk for hemostatic disorders, these cats should be monitored closely. References I. Rebar AH. The feline hemogram. In: Feline Medicine 111, Proceedings ofthe 3rd Annual Kal Kan Seminar, Eastern States Veterinary Conference. Orlando, FL, 1987; Gabbert NH. Systemic lupus erythematosus in a cat with thrombocytopenia. Vet Med Small Anim Clin 1983: 1: Joshi BC, Raplee RG, Powell AL, et al. Autoimmune thrombocytopenia in a cat. J Am Anim Hosp Assoc 1979: 15: Harvey JW, Gaskin JM. Idiopathic thrombocytopenia and hemorrhage followed by thrombocytosis in a cat. Feline Pract 1980; 10: Cain GR, Cain JL. Turrel JM, et al. Immune-mediated hemolytic anemia and thrombocytopenia in a cat after bone marrow transplant. Vet Pathol 1988; 25: Weiss RC, Dodds WJ, Scott FW. Disseminated intravascular coagulation in experimentally induced feline infectious peritonitis. Am J Vet Res 1980; 41: Boyce JT, Kociba GJ, Jacobs RM, et al. Feline leukemia virus-induced thrombocytopenia and macrothrombocytosis in cats. Vet Pathol 1986; 23: Boon GD, Rebar AH. The clinical approach to disorders of hemostasis. In: Ettinger SJ, ed. Textbook of Veterinary Internal Medicine, 3rd ed. Philadelphia: WB Saunders. 1989: Snedecor GW. Cochran WG. One-way classifications: Analysis of variance. In: Statistical Methods. Ames, IA: Iowa State University Press, 1989: Snedecor GW, Cochran WG. The binomial distribution. In: Statistical Methods. Ames, IA: Iowa State University Press, 1989; I I. Weiser MG. Comparison of two automated multi-channel blood cell counting systems for analysis of blood of common domestic animals. Vet Clin Pathol 1983: 12: Schalm OW, Jain NC, Carroll EJ. Veterinary Hematology, 3rd ed. Philadelphia: Lea and Febiger. 1975; I I I. 13. Cockburn C, Troy GC. A retrospective study of sixty-two cases of thrombocytopenia in the dog. Southwest Vet 1986; 37: Grindem CB, Breitschwerdt EB, Corbett WT. et al. Epidemiologic survey of thrombocytopenic dogs: A report of 987 cases. Vet Clin Pathol 1991; 20: Teo CP, Kueh YK. Incidence of thrombocytopenia in an acute care hospital. Ann Acad Med, Singapore 1989; 18: Hoover EA, Olsen RG, Mathes LE, et al. Relationship between feline leukemia virus antigen expression and viral infectivity in blood, bone marrow, and saliva of cats. Cancer Res 1977; 37: Jacobs RM, Boyce JT. Kociba GJ. Flow cytometric and radioisotopic determinations of platelet survival time in normal cats and feline leukemia virus-infected cats. Cytometry 1986; 7: Day NK, O'Reilly-Felice C, Hardy WD, Jr., et al. Circulating immune complexes associated with naturally occurring lymphosarcoma in pet cats. J Immunol 1980; 126: Shelton GH, Linenberger ML, Grant CK, et al. Hematologic manifestations of feline immunodeficiency virus infection. Blood 1990; 76:l104-ll Greene CE. Scott FW. Feline panleukopenia. In: Greene CE. ed. Infectious Diseases of the Dog and Cat, 2nd ed. Philadelphia: WB Saunders, 1990; Boudreaux MK, Weiss RC, Toivio-Kinnucan M, et al. Potentiation of platelet responses in vim by feline infectious peritonitis virus. Vet Pathol 1990; 27: Boudreaux MK. Weiss RC. Toivio-Kinnucan M, et al. Enhanced platelet reactivity in cats experimentally infected with feline infectious peritonitis virus. Vet Pathol 1990; 27: Hefland SC. Platelets and neoplasia. Vet Clin North Am Small Anim Pract 1988; 18: Kruth SA, Cater RF. Laboratory abnormalities in patients with cancer. Vet Clin North Am Small Anim Pract 1990; 20: Hefland SC, Couto CG, Madewell BR. Immune-mediated thrombocytopenia associated with solid tumors in dogs. J Am Anim Hosp Assoc 1985; 2 1 : Helenski CA, Ross JN, Jr. Platelet aggregation in feline cardiomyopathy. J Vet Intern Med 1987; 1: Hayes KC, Pronszuk A, Addesa AE. et al. Taurine modulates platelet aggregation in cats and humans. Am J Clin Nutr 1989; 49:l2l Fox PR, Dodds WJ. Coagulopathies observcd with spontaneous aortic thromboembolism in cardiomyopathic cats. Proceedings Am Coll Vet Intern Med 1982; Peterson MC, Hurvitz AI. Leib MS, et al. Propylthiouracil-associated hemolytic anemia, thrombocytopenia and antinuclear antibodies in cats with hyperthyroidism. J Am Vet Med Assoc 1984: I84: Jansen PHP, Renier WO, de Vaan G, et al. Effect of thymectomy on myasthenia gravis and autoimmune thrombocytopenia purpura in a 13-year old girl. Eur J Pediatr 1987; 146: I. Couto CG. Oncology. In: Sherding RG, ed. The Cat: Diseases and Clinical Management. New York: Churchill Livingstone. 1989: Jain NC. Qualitative and quantitative disorders of platelets. In: Jain NC, ed. Veterinary Hematology, 4th ed. Philadelphia: Lea and Febiger, 1986;