Case 1. A forty-six-year-old woman who weighed 55 kg had
osteoarthritis of the hip secondary to developmental dysplasia. She had no
history of bleeding predisposition or other medical problems, and she had
never received any form of heparin. An uncemented total hip replacement was
performed through a posterior approach. Postoperatively, the patient received
a transfusion of two units of packed red-blood cells. An antithrombotic
regimen of 40 mg of enoxaparin once daily was started six hours after surgery
and was to be continued for four weeks. There were no bleeding complications
in the wound or any other site during hospitalization. The patient was
discharged on the sixth postoperative day but returned to the hospital on the
ninth day complaining of headache and motor disturbances in the left arm.
Neurological examination revealed no pathological signs or reflexes, except
for reduced strength in the left arm. The blood pressure was 180/90 mm Hg, and
the platelet count was 126,000/mm3; the platelet count had been
250,000 preoperatively and 180,000 postoperatively (normal values at our
hospital are 150,000 to 300,000/mm3). Computed tomography of the
brain performed on that day showed no notable findings. She was admitted to
the neurology department of our hospital. On the following day (postoperative
day 10), the platelet count fell to 35,000 and paresis developed in the left
arm. The low-molecular-weight heparin was discontinued, but tests for
antibodies to heparin were not performed. A repeat computed tomography scan
was performed on the eleventh postoperative day and revealed bilateral
parasagittal hemorrhages (Fig.
1). She was admitted to the intensive care unit because of a rapid
deterioration in her neurological status and died later the same day. An
autopsy was performed and confirmed brain hemorrhage as the cause of
death.
Case 2. A forty-seven-year-old woman who weighed 65 kg had
osteoarthritis of the left hip secondary to developmental dysplasia, for which
an osteotomy of the hip had been previously performed. She had no history of
coagulopathy or other medical problems and had no known allergies; previous
exposure to heparin was not reported. An uncemented total hip arthroplasty was
performed through a lateral approach, and she received two units of packed
red-blood cells postoperatively. A regimen of 40 mg of enoxaparin once daily
was started twenty-four hours after surgery and was to be continued for four
weeks. The patient was discharged on the ninth day with no bleeding
complications, but she returned to the hospital on the thirteenth
postoperative day with a headache and sensory disorders of the left arm. The
platelet count was 70,000. The low-molecular-weight heparin was discontinued,
and the presence of antibodies to heparin was investigated. Magnetic resonance
imaging showed a small focus of high intensity in the brain
(Fig. 2). This was considered
to be a thrombotic focus, and treatment with unfractionated heparin was
started for thrombolysis. On the following days, the platelet count and
hemoglobin decreased gradually (reaching a lowest point of 40,000 for the
platelet count) and thrombolysis treatment was discontinued. The presence of
antiheparin antibodies was confirmed two weeks after readmission. Paresis of
the arms and legs developed progressively
(Fig. 3), as a result of large
cerebral hemorrhages. Decompression of the cerebral hematomas was performed
twice. After forty-five days in the intensive care unit, the patient was
discharged but was quadriparetic. She was readmitted one year later for a
resection arthroplasty of the hip because of recurrent dislocations,
discomfort, and consequent difficulties in nursing care. There had been no
improvement in her neurological status at that time.
Bleeding in association with the administration of a low-molecular-weight
heparin in total joint replacement has been reported to occur in various
anatomical sites, including the operative site and the epidural, intrahepatic,
and retroperitoneal sites as well as in the gastrointestinal
tract6,7,10-12.
Low-molecular-weight heparin has also been implicated in abdominal wall
hematomas in general
surgery8 and in
intracranial hemorrhage in surgery for brain
tumors13. This
complication of the use of a low-molecular-weight heparin, however, has not
been previously reported in orthopaedic patients, to our knowledge.
Low-molecular-weight heparin is associated with two forms of
thrombocytopenia: type-I and type-II heparin-induced (or heparin-associated)
thrombocytopenia. Type-I heparin-induced thrombocytopenia is associated with
mild thrombocytopenia (100,000 to 130,000 platelets per microliter). It
typically occurs one to four days after the initiation of heparin therapy, and
screening tests for heparin-induced thrombocytopenia antibody activity are
negative9,14.
We are aware of no data suggesting that patients with type-I heparin-induced
thrombocytopenia have an increased risk of
thrombosis15 and,
in fact, platelet counts may increase during subsequent heparin
therapy9. Type-I
heparin-induced thrombocytopenia is attributed to a direct, reversible,
proaggregatory effect of
platelets16.
Patients do not require specific
treatment17.
Type-II heparin-induced thrombocytopenia appears typically five or more
days after the start of heparin
therapy18,19.
It may, however, develop rapidly (mean, 10.5 hours) in patients who have
received heparin within the previous 100
days20, as previous
exposure to heparin is a risk factor for type-II heparin-induced
thrombocytopenia18,20.
Conversely, a delayed-onset heparin-induced thrombocytopenia and thrombosis,
developing after heparin discontinuation, has also been
described19. The
syndrome has an onset that is independent of heparin type, dosage, or route of
administration21,
and its diagnosis requires both thrombocytopenia and the presence of
antibodies to
heparin14. A
modified definition of type-II heparin-induced thrombocytopenia, in reference
particularly to postoperative orthopaedic patients, requires a decrease of 50%
in the platelet count from the postoperative
peak22. After
discontinuation of heparin, the platelet count increases to normal levels
usually within five to seven
days17. A prolonged
recovery of the platelet count to normal levels should thus prompt
investigation of other causes of
thrombocytopenia14.
Clinical scoring systems have been proposed to evaluate the probability of
the syndrome. They are based on the severity of the thrombocytopenia, recovery
following heparin withdrawal, the onset of thrombotic complications, and the
exclusion of other causes of
thrombocytopenia23,24.
It is estimated that type-II heparin-induced thrombocytopenia develops in 1%
to 5% of patients who receive
heparin25, with the
occurrence varying depending on the clinical status of the patient (with the
greatest risk, in descending order, after surgery, medical treatment, and
pregnancy), the type of heparin preparation (bovine unfractionated heparin is
associated with a greater risk than porcine unfractionated heparin, which
carries a greater risk than low-molecular-weight heparin), the route of
administration (intravenous administration is associated with a greater risk
than subcutaneous injection), and the definition of thrombocytopenia
used17,26.
However, the proportion of patients who have antibodies to heparin but do not
have thrombocytopenia develop is larger.
Arterial and venous thromboses are the major clinical complications, as
they cause ischemia to the limbs, with the potential for limb loss, and to the
vital organs, with the potential for organ failure or death. They occur in up
to 30% of patients with type-II heparin-induced
thrombocytopenia15,
which is then commonly referred to as the "white clot syndrome."
Common locations for arterial thromboses are the lower limb, the brain
(thrombotic stroke), and the heart (myocardial infarction), while the most
common complications for the venous system are deep venous thrombosis and
pulmonary embolism. Thromboses may be noted in unusual locations (the
mesenteric or renal artery, vascular graft occlusion of an artery, adrenal
hemorrhagic infarction, or cerebral vein thrombosis), and disseminated
intravascular coagulation can
occur14,27.
Other complications, such as skin lesions at injection sites, may also
occur27. Rates of
thrombotic events of 5% to 10% in the first one to two days have been
observed, suggesting that many of these patients had subclinical thromboses at
the time of
diagnosis28. The
rate of mortality has been reported to be 15% (eight) in a series of
fifty-three patients with laboratory-confirmed type-II heparin-induced
thrombocytopenia21
(or 19% if only the patients with heparin-induced thrombocytopenia complicated
with thrombosis were counted), two deaths in a series of twelve patients with
delayed-onset heparin-induced thrombocytopenia and thrombosis, and almost 30%
(twenty-five) in an earlier series of eighty-five patients with
heparin-induced thrombocytopenia complicated with
thrombosis17.
The mechanism for thrombocytopenia in type-II heparin-induced
thrombocytopenia is antibody-induced platelet activation, leading to platelet
aggregation and a decrease in the platelet count. The principal antigen is a
complex of heparin and platelet factor 4, and the antibodies can be detected
by serological
assays29,30.
They form more frequently with the use of unfractionated heparin than with
low-molecular-weight heparin. In a randomized study of 209 hospitalized
patients receiving treatment with heparin, the prevalence of antibodies to
heparin was 17% in those treated with unfractionated heparin and 8% in those
treated with low-molecular-weight
heparin31. In
another randomized, double-blind study of 665 patients who received heparin as
prophylaxis after hip surgery, antibodies to heparin developed in 7.8% of the
patients treated with unfractionated heparin compared with 2.2% of the
patients treated with low-molecular-weight
heparin9.
In the first patient (Case 1) in this report, thrombocytopenia developed on
the ninth postoperative day and led to brain hemorrhage. The presence of
antibodies to heparin was not investigated, as heparin-induced
thrombocytopenia was not included in our differential diagnosis; however, the
clinical course of the syndrome was highly suggestive of type-II
heparin-induced thrombocytopenia as there were no other obvious reasons for
thrombocytopenia9,22,26.
In the second patient (Case 2), type-II heparin-induced thrombocytopenia was
established as the cause of the initial thrombotic incident. The
administration of unfractionated heparin led to severe brain hemorrhages as
there is cross-reactivity between the two
heparinoids5,9.
The current recommendations of the American College of Chest Physicians for
prevention therapy with low-molecular-weight heparin in patients undergoing
total hip replacement are to initiate therapy twelve hours before surgery or
twelve to twenty-four hours after surgery at the usual high-risk dose, or four
to six hours after surgery at half the usual high-risk dose and then increase
to the usual high-risk dose the following
day3. The
recommendations for the prevention of type-II heparin-induced thrombocytopenia
in postoperative patients receiving a prophylactic dose of
low-molecular-weight heparin suggest monitoring of the platelet count every
two or three days from day 4 to day 14 (or until heparin is stopped, whichever
occurs first), when
practical26.
Screening for antibodies is not recommended. For patients with strongly
suspected or confirmed heparin-induced thrombocytopenia, whether complicated
by thrombosis or not, the immediate cessation of the heparinoid is mandated
and the introduction of an alternative anticoagulation therapy, such as a
direct thrombin inhibitor or danaparoid, is
recommended26.
Routine ultrasonography of the lower-limb veins is also recommended to
identify deep-vein thrombosis in these patients, even if there is no clinical
evidence to suggest thrombosis.
We believe that type-II heparin-induced thrombocytopenia is a syndrome that
occurs in orthopaedic patients more frequently than is currently believed. We
think that it remains underdiagnosed, reflecting a lack of awareness about the
condition among orthopaedic surgeons. Its diagnosis requires clinical
alertness, laboratory confirmation, and prompt treatment in order to prevent
disastrous complications such as those seen in the two patients in the present
report. ?