Hemostatic Abnormalities in Dogs With

Hemangiosarcoma
A. S. Hammer, DVM, C. G. Couto, DVM, C. Swardson, DVM,
and D. Getzy, DVM

The hemostasis profiles of 24 dogs with histologically confirmed hemangiosarcoma were prospectively
evaluated. Microangiopathic hemolysis was defined as the presence of schistocytes; disseminated intra-
vascular coagulation was defined as I) thrombocytopenia, 2) fibrin(ogen) degradation products > 10
pg/mL, 3 ) prolongation of one or more coagulation times (activated partial thromboplastin time or
one-stage prothrombin time) by greater than 25% of the control, 4) fragmented red blood cells (>l+
based on a semiquantitative grading scale), and 5) fibrinogen I 80 mg/dL. Three of the five criteria
listed above had to be met for disseminated intravascular coagulation to be diagnosed. Fifty percent of
the dogs were considered to have disseminated intravascular coagulation at presentation. Thrombocy-
topenia was present in 75% of the dogs and was the most common abnormality. The mean platelet count
was 137,80O/pL. Twenty-five percent of the dogs died as a result of the hemostatic abnormalities. Only
12%of the dogs had microangiopathic hemolysis without other evidence of disseminated intravascular
coagulation. Hemostatic abnormalities are present in many dogs with hemangiosarcoma a t the initial
clinical presentation and represent an important clinical finding. (Journal of Veterinary Internal Medi-
cine 1991; 511-14)

HEM, NGIOSARCOMA in the dog has been asso-
ciated with hemostatic abnormalities such as dissemi-
nated intravascular coagulation (DIC) and microangio-
'-'
pathic hemolysis (MAH). To our knowledge, there
are only nine published cases of DIC in dogs with he-
mangiosar~oma.'-~ Microangiopathic hemolysis has
been associated with both spontaneous and experimen-
tally induced hemangiosarcoma in this species."* The
red blood cell fragmentation is believed to result from
distortion and disruption of erythrocytes as they pass
through fibrin deposits within vessels and abnormal
tumor vasculature.',2,8Another report of ten dogs with
hemangiosarcoma revealed that 90% were thrombocy-
topenic, and 80% had fragmented red blood c e k 6 Our
clinical impression has been that dogs with hemangio-
sarcoma have a high prevalence of hemostatic abnor-
malities. We prospectively evaluated the hemostasis
profiles of 24 dogs with histopathologically confirmed
hemangiosarcorna to determine the prevalence of these
abnormalities. We concluded that hemostatic abnor-
malities are present in a large number of dogs with he-
From the Department of Veterinary Clinical Sciences (Hammer,
Couto) and the Department of Veterinary Pathobiology (Swardson,
Getzy), the Ohio State University, College of Veterinary Medicine,
Columbus, Ohio.
Reprint requests: Alan Hammer, 1935 Coffey Road, Columbus, OH
43210.
mangiosarcoma and are of clinical importance due to
the high mortality associated with DIC.7
Methods
Thirty-five dogs examined at the Veterinary Teaching
Hospital-Ohio State University (VTH-OSU) between
January 1983 and March 1989 were diagnosed by histo-
pathology as having hemangiosarcoma. The histopatho-
logic sections were retrospectively reviewed and 24 dogs
were prospectively evaluated for hemostatic abnormali-
ties at presentation. The coagulation tests included one-
stage prothrombin time (OSPT) using rabbit brain
thromboplastin* and activated partial thromboplastin
time (APTT) using rabbit brain cephalint; the end-
points for these tests were determined by fibrometer.
The OSPT and APTT, respectively, were reported as a
percent prolongation above the concurrently run con-
trol. Controls consist of pooled citrated plasma from a
donor colony frozen and stored at -70°C. The reference
ranges for the OSPT and APTT are 10 to 12 seconds and
18 to 24 seconds, respectively. Also determined were
platelet count, fibrinogen concentration (heat precipita-
* Baxter Healthcare Corporation,
Dade Division, Miami, FL.
t Actin Activated Cephaloplastin Reagent, American Dade,
Aguada, PR.

11
 Journal of Veterinary
12 HAMMER ET AL. Internal Medicine

TABLE1. Individual Data for Dogs With Hemangiosarcoma and Hemostasis Profiles

OSPT APTT Fragmented

(% prolongation (% prolongation Fibrinogen Red Blood
Tumor Site Platelets/pL above control) above control) FDP (mg/dL) Cells Outcome Diagnosis

1 right atrium 32,000 56 52 40 m / m L I00 0 euthanized DIC

2 subcutaneous, kidney, liver 28,000 32 21 10-40 &g/mL 300 I+ alive DIC
3 liver 3 1,000 38 54 10-40 pg/mL 200 2+ euthanized DIC
4 spleen 148,000 0 8 negative 200 NE alive thrombocytopenia
5 subcutaneous 349,000 -9 I negative 400 0 alive normal
6 liver 45,000 6 I38 negative I00 0 died in DIC thrombocytopenia
7 omentum 344.000 21 -17 negative 110 I+ a1i ve MAH
8 subcutaneous 92,000 -5 4 negative 200 0 euthanized thrombocytopenia
9 spleen 75,000 25 76 40 pglmL 600 2+ euthanized DIC
I0 spleen 28,000 21 242 10-40 pg/mL 200 I+ died in DIC DIC
II right atrium 16,000 190 172 40 pg/mL 25 I+ euthanized DIC
12 spleen 3 1,000 101 I35 negative 155 2+ died* DIC
13 spleen 64,000 17 14 >40 pg/mL 94 NE alive thrombocytopenia
14 spleen 370,000 13 6 negative NE 2+ euthanized MAH
15 spleen 430,000 -13 55 negative NE 0 euthanized prolonged APTT
16 liver, right atrium 45,500 20 40 >40 pg/mL I10 0 died in DIC DIC
17 spleen 140,000 -5 41 negative I20 2+ euthanized DIC
I8 liver, spleen 43,000 10 I06 10-40 pg/mL I00 NE died in DIC DIC
19 right atrium, spleen 133,000 -3 12 negative 300 I+ alive MAH
20 subcutaneous I2 1,000 6 8 negative I00 0 alive thrombocytopenia
21 spleen 96,000 15 23 negative 200 0 alive thrombocytopenia
22 subcutaneous. lungs 76,000 27 89 40 pg/mL 200 0 died in DIC DIC
23 subcutaneous 332,000 4 3 negative 200 0 alive normal
24 right atrium, lung 238,000 31 58 10-40 &mL 300 I+ died in DIC DIC

NE: not evaluated.

* Died due to causes unrelated to hemostatic abnormalities

tion method'), and fibrin(ogen) split products (FDP) did not fulfill the above criteria were described. The
concentration (latex agglutination method$). The refer- patients' immediate outcome (i.e., within 4 days of the
ence range for platelet counts is 150,000to 400,OOO/pL hemostasis screen) was determined and recorded as
and for fibrinogen is 80 to 160 mg/dL; normal fi- alive, death related to hemostatic abnormalities, death
brin(ogen) degradation product concentration is less unrelated to hemostatic abnormalities, or euthanized.
than 10 pg/mL. Fibrin(ogen) split products were re-
ported as < I 0 pg/mL, 10 to 40 pg/mL, and >40 pg/mL.
Results
Microscopic examination of blood smears was per-
formed to identify red blood cell fragmentation using
The data for individual dogs are listed in Table 1. There
the semiquantitative system of Weiss et a1.l' This system
were 12 dogs (50%)with laboratory evidence of DIC and
classifies schistocytosis based upon the average number
three dogs ( 12%) with microangiopathic hemolysis
of schistocytes per IOOOX microscopic field. One to two
alone. Thrombocytopenia was the most common ab-
schistocytes per field are graded 1+; three to eight schis-
normality and was present in 18 dogs (75%). The mean
tocytes per field are graded 2+; nine to 20 are graded as
platelet count for the 24 dogs was 137,8OO/pL. Six dogs
being 3+; and greater than 20 schistocytes are 4+. The
(25%) had isolated thrombocytopenia without evidence
average of five fields (lOOOX) was reported.
of DIC or MAH. The distribution of the platelet counts
The dogs were classified as having DIC if at least three
is depicted in Figure 1. Two dogs (8%) were considered
of the following five criteria were present: thrombocyto-
to have normal hemostasis profiles. In one dog, the only
penia (platelets < 15O,OOO/pL), FDPs > 10 pg/mL, pro-
abnormality was a prolonged APTT. Fragmented red
longation of the APTT and/or OSPT by greater than
blood cells ( 2 1 +) were present in 1 1 dogs (46%)
25% of the control, fragmented red blood cells 2 I + ,
and represented the only abnormality in two of the
and fibrinogen concentration I80 mg/dL. The diag-
dogs (8%).
+
nosis of MAH was made if 2 1 schistocytes were pres-
Six dogs (25%) died as a result of DIC and eight dogs
ent. Disseminated intravascular coagulation and MAH
(33%) were euthanized shortly after diagnosis and could
could both coexist in the same patient. If no abnormali-
not be evaluated for DIC as a cause of death. Nine dogs
ties were found, the dog's hemostatic status was classi-
(37%) survived the immediate four-day, postoperative
fied as normal. Any other hemostatic abnormalities that
period following the initial hemostasis profile and one
dog died due to causes unrelated to hemostatic abnor-
$ Thrombo-Wellcotest, Wellcome Diagnostics, Dartford, England. malities.
Vol. 5 . No. 1 , 1991 HEMOSTATIC ABNORMALITIES
30
FIG. 1. Distribution of 20
platelet counts in 24 dogs
with hemangiosarcoma.
10
0
<25 26-50
Discussion
The most clinically relevant finding of this study was
that 50% of the dogs with hemangiosarcoma evaluated
for hemostatic abnormalities by routine laboratory tests
were diagnosed as having DIC based upon laboratory
evidence alone. This confirms our clinical impression
that dogs with hemangiosarcoma have a high prevalence
of hemostatic abnormalities despite the paucity of re-
ported cases in the veterinary literature. A survey of
coagulation abnormalities in 100 dogs with neoplastic
disease revealed that 8% had increased FDPs and 36%
'
were thrombocytopenic.' There were only four dogs
with hemangiosarcoma included in that survey. Al-
though dogs with DIC were not specifically counted, the
investigators implied that only 8% to 16% of the dogs
studied had DIC. Based on the hemostasis results of the
24 dogs reported herein, dogs with hemangiosarcoma
appear to have a greater risk for DIC than dogs with
other types of neoplasia.
Disseminated intravascular coagulation is important
to recognize due to the associated high mortality rate.'
Twenty-five percent of the dogs reported in this study
died as a consequence of DIC (i.e., bleeding or thrombo-
sis with attendant organ failure). This figure is probably
falsely decreased by the fact that 33% of dogs were eu-
thanized shortly after diagnosis. Other studies have
found a mortality rate exceeding 80% in dogs with DIC.7
Thrombocytopenia was the most common abnormal-
ity, being present in 75% of the patients. The percentage
of dogs with thrombocytopenia reported here is similar
to that reported by Ng and Mills in ten dogs with he-
mangiosarcoma (90% of the dogs were thrombocyto-
13
51-75 76-100 101-125 126-150 >150
Platelet Count (xl OOO/pL)
penic).6 Despite the mean platelet count of 137,8OO/~L,
the degree of thrombocytopenia was variable as depicted
in Figure 1. Greater than one third of the dogs had
platelet counts of less than 50,00O/pL. In addition to the
thrombocytopenia, it is conceivable that at least some of
the dogs probably had some degree of platelet dysfunc-
tion induced by the circulating FDPs."
Approximately two thirds (24/35) of the dogs with
hemangiosarcoma diagnosed during the study period
had hemostasis profiles performed. Some bias based
upon the history and physical examination was un-
doubtedly introduced as to which dogs were evaluated.
However, the prevalence of DIC in the total population
of dogs with hemangiosarcoma remains high (12 of 35
or 34%).
Our data on the prevalence of MAH are similar to
those reported in experimentally induced hemangiosar-
coma. In our study, almost 50% of the dogs had MAH
either directly or indirectly (i.e., DIC) related to heman-
giosarcoma. In the dogs with experimentally induced
hemangiosarcoma, 45% had anemia that was attributed
primarily to MAH.' Five dogs in that study were evalu-
ated for hemostatic abnormalities and only one had evi-
dence of DIC.' Fifty-eight percent of the dogs with ex-
perimentally induced hemangiosarcoma had frag-
mented red blood cells in circulation.' This is similar to
the 46% of dogs in this study with fragmented red blood
cells. Other investigators found schistocytosis in 23% of
dogs with splenic neoplasia. l 3
In summary, we conclude that DIC is a likely compli-
cation of hemangiosarcoma in dogs. Due to the high
morbidity and mortality in dogs with DIC,7 hemostasis
profiles should be performed to evaluate dogs suspected

14 HAMMER ET AL
of having hemangiosarcoma. Measures to control or
prevent DIC (e.g., heparin, blood component therapy)
in dogs with hemangiosarcoma may increase survival
times by decreasing early mortality.
References
1. Rebar AH, Hahn FF, Halliwell WH, et al. Microangiopathic he-
molytic anemia associated with radiation-induced hemangio-
sarcomas. Vet Pathol 1980; 17:443-454.
2. Madewell BR, Feldman BF. Characterization of anemias asso-
ciated with neoplasia in small animals. J Am Vet Med Assoc
1980; 176:419-425.
3. Legendre AM, Krehbiel JD. Disseminated intravascular coagula-
tion in a dog with hemothorax and hemangiosarcoma. J Am
Vet Med Assoc 1977; 171:1070-1071.
4. Zenoble RD, Gabbert NH. A possible case of disseminated intra-
vascular coagulation and splenic hemangiosarcoma. Canine
Practice 1977; 4:52-55.
5. Crow SE, Bell TG, Wortman JA. Hematuria associated with renal
hemangiosarcoma in a dog. J Am Vet Med Assoc 1980;
1761531-533.
Book Review
Veterinary Pediatrics
Johnny D.Hoskins
As Dr. Hoskins stated in the preface for Veterinary Pedi-
atrics, review papers and research articles have been
published on the diagnostic, medical, and surgical dis-
orders of the young dog and cat. Additionally, puppies
and kittens have been studied as models for under-
standing the physiology, pathology, and therapy utilized
in human neonates. Unfortunately, access to this infor-
mation has been limited to prac’ticing veterinarians
since the data have been scattered in various research
journals and textbooks. Pertinent information has been
accumulated in Veterinary Pediatrics to provide an ex-
cellent guide to the health needs of dogs and cats from
birth to 6 months of age.
Topics in veterinary Pediatrics include physical ex-
amination and radiography, behavioral development
and disorders, drug and blood component therapy, the
cardiovascular system, the respiratory system, the ner-
vous system, the digestive system, the liver and pan-
creas, endocrine and metabolic systems, the urinary sys-
tem, the hematopoietic system, the immune system, the
musculoskeletal system, the skin, the lymphoid system,
the eye, the ear, nutrition and nutritional disorders, toxi-
cology, environmental injuries, and preventive health
programs. The chapter on the liver and pancreas by
Journal of Veterinary
Internal Medicine
6. Ng CY, Mills JN. Clinical and haematological features of hae-
mangiosarcoma in dogs. Aust Vet J 1985: 62:l-4.
7. Drazner FH. Clinical implications of disseminated intravascular
coagulation. Compend Contin Educ Pract Vet 1982; 4:974-
981.
8. Bull BS, Kuhn IN. The production ofschistocytes by fibrin strands
(a scanning electron microscope study). Blood 1970; 35: 104-
1 1 1.
9. Jain NC. Schalm’s Veterinary Hematology. 41h ed. Philadelphia:
Lea and Febiger, 1988: 58-59.
10. Weiss DJ. Uniform evaluation and semiquantitative reporting of
hematologic data in veterinary laboratories. Vet Clin Pathol
1988; I3:27-3 1.
1 1. Madewell BR, Feldman BF, O’Neill S. Coagulation abnormalities
in dogs with neoplastic disease. Thrombosis and Haemostasis
1980; 44135-38.
12. Bick RL. Diseminated intravascular coagulation and related syn-
dromes: A clinical review. Semin Thromb Hemost 1988;
14:299-338.
13. Johnson KA, Powers BE, Withrow SJ, et al. Splenomegaly in dogs:
Predictors of neoplasia and survival after splenectomy. Journal
of Veterinary Internal Medicine 1989; 3: 160- 166.
Sharon A. Center, William E. Hornbuckle, and Johnny
D. Hoskins provides new information, i.e., normal
values for routine biochemical indicators of hepatobili-
ary disorders in puppies and kittens during the first 4
weeks of life. The chapter on drug and blood component
therapy by Craig E. Greene, Johnny Hoskins, and Jen-
nifer M. Authement provides much needed information
on therapeutic considerations for young animals.
Informative discussions on the cause and therapy of
neonatal hypoglycemia are given. Readers should be
aware, however, that the formula for making 20% dex-
trose solution (p. 266) is wrong. The solution made by
adding 30 ml of 50% dextrose to 500 ml of 5% dextrose
is 7.5% dextrose, not 20% dextrose. I would not recom-
mend that 20% dextrose be given intravenously to pup-
pies; administering 7.5% dextrose is safer (see “Compli-
cations of Drug Therapy”, p. 34). The formula given for
making 10%dextrose is also wrong.
Noticeably missing from this textbook is a chapter on
the reproductive system. Topics such as cryptorchidism,
effects of prepubertal gonadectomy, and puppy vaginitis
would be useful information to the practicing veterinar-
ian. Nevertheless, Veterinary Pediatrics is an excellent
book. It is hoped that a second edition will follow.
Patricia N. Olson, DVM, PhD
Fort Collins, Colorado