The Role of Prophylactic
Anticoagulation in the Surgical
here is an ever-increasing awareness of the importance of the
T thrombotic process on patient outcomes after a variety of inter-
ventions. Thrombosis is a complicated process that involves not
only the traditional clotting cascade but also the interaction of the
platelets, whose role in the progression of the clotting process is now
appreciated. We use the term prophylactic to include anticoagulant
therapy administered to prevent a thrombotic event before, during, or
after an intervention. Data will be offered to help the surgeon decide when
to use antithrombotic agents, when to discontinue them, and what the
relative advantages are of the drugs currently available. The shear number
of studies about the use of anticoagulants suggests that the evidence that
has been provided so far is inadequate to make evidence-based decisions
regarding their usage. Many of the studies are small and not powered to
identify small differences in effect. The technique of meta-analysis is
often used to circumvent this difﬁculty. Best evidence and recommenda-
tions from panels of experts are provided.
Evaluating the Evidence
Statistical Problems with Small Trials
There are more than 50 clinical trials that just compare low molecular
weight heparin (LMWH) with unfractionated heparin (UFH), and there
are several meta-analyses. Meta-analyses are associated with publication
bias and oversimpliﬁcation.1 In general, when meta-analyses are limited
to randomized, double-blinded clinical trials, the evidence is better than
that from observational studies but not equal to a randomized clinical trial
(RCT) with adequate power. Speciﬁc problems in the analysis of data
from the multitude of thromboprophylaxis trials are the variations in the
timing of drug administration and the variability in the duration of
Curr Probl Surg 2003;40:81-130.
0011-3840/2003/$30.00 0 40/2/022222
92 Curr Probl Surg, February 2003
TABLE 1. Evaluation of the evidence2
Grade Method Recommendation
1A Clear RCT with no limitations Strong-No reservation
1B Clear RCT with ﬂaws Strong-Most patients
1C Clear Overwhelming evidence from Strong-Most patients
1C Clear OS Intermediate-More data may
2A Unclear RCT with important limitations Intermediate-Best action may differ
2B Unclear RCT with inconsistent results Weak-Alternative approaches better
2C Unclear OS Very weak
RCT, Randomized clinical trial; OS, observational study.
Grading system for recommendations
A strong recommendation (grade 1) occurs when experts believe that
beneﬁts do or do not outweigh risks. If experts are less certain, they make
a weaker recommendation (grade 2). RCTs with consistent results provide
grade A recommendations; RCTs with inconsistent results or methods
ﬂaws provide grade B recommendations. Observational studies provide
grade C results. When evidence from observational studies is overwhelm-
ingly compelling, a grade of C is given.2 Table 1 summarizes the
grading methods used throughout this monograph.
The Hypercoagulable State
Thrombosis of arteries and veins accounts for more than 2 million
deaths each year in the United States, and 50% of these events can be
attributed to a congenital or acquired coagulation or platelet abnormali-
Clinical suspicion should be high for a hypercoagulable state when a
patient experiences recurrent thromboses, there is a family history of
thrombosis, a history of spontaneous abortions, there is thrombosis in an
unusual site, a young person has a thrombotic disorder, and a patient has
a history of multiple exposures to heparin (Table 2). Patients with
hypercoagulable states have a 6-fold increased risk of failure when
arterial reconstructions are performed.5 Long-term anticoagulation re-
duces the risk for thrombosis.
Activated Protein Resistance. Activated protein C (APC) resistance is
the most common risk factor for the development of venous thrombosis
and accounts for more than 50% of thromboses from inherited causes.6 In
Curr Probl Surg, February 2003 93
TABLE 2. Clinical concerns that should initiate an evaluation of hypercoagulability7
Thrombus in unusual site
Young patient with thrombosis
Family history of thrombotic disorder
Premature or unexplained failure of vascular intervention
Multiple exposures to heparin
Spontaneous mid-term abortions
the normal state, APC is a naturally occurring anticoagulant that degrades
factors Va and VIIIa. Patients with functional APC-R do not demonstrate
in vitro evidence of anticoagulation in the presence of exogenous APC,
even with adequate levels of normal protein C. Most, but not all, of these
patients have an inborn error of metabolism that results in an abnormal
factor V (Leiden). In the setting of the factor V (Leiden) mutation, factor
Va is resistant to degradation, and the altered factor V remains a
procoagulant. Heterozygous individuals have a 7-fold increased risk for
DVT. Homozygous individuals have an 80-fold increased risk and will
likely have at least 1 thrombotic episode during their lifetime.7 APC-R is
also a factor in arterial occlusive disease in the younger patient population
and is a common ﬁnding in patients with failed arterial reconstructions.8
The diagnostic approach to this condition should begin with the func-
tional assay for APC-R that is far less costly than the polymerase chain
reaction that detects the factor V (Leiden). The prothrombin gene variant
(20210A) is another risk factor for venous and arterial thrombosis. Its
mechanism is unclear. Patients with these conditions require long-term
Antiphospholipid Syndrome. Patients with the antiphospholipid syn-
drome experience the development of antibodies against a variety of
protein-phospholipid complexes that result in a hypercoagulable state.
This diagnosis should be suspected in a patient with recurrent deep
venous thrombosis (DVT), with thrombotic complications after an arterial
reconstruction, arterial occlusions in unusual sites, recurrent mid preg-
nancy abortions, and/or thrombocytopenia. The diagnosis is conﬁrmed by
assays for either the lupus anticoagulant or anticardiolipin antibodies.
Studies show that patients with antiphospholipid antibodies who undergo
arterial reconstruction have a higher incidence of disease progression that
those patients without the antibodies.10 The treatment includes warfarin
therapy for as long as antibodies persist. Pregnant women with this
syndrome should be treated with LMWH.
94 Curr Probl Surg, February 2003
Congenital Deﬁciency of Antithrombin III. Patients with congenital
deﬁciency of antithrombin III (AT III) are at increased risk for thrombo-
embolism. The prevalence of this genetic abnormality is 1:5000, and
thrombotic events usually occur in the second decade and are associated
with surgery, trauma, or pregnancy. The risk for DVT is 24% during the
period after operation.11 This is the only inherited disorder of coagulation
for which there is an Food and Drug Administration (FDA)–approved
replacement therapy. Commercially available antithrombin concentrates
have been available in the United States since 1990 (ATnativ; Kabi
Pharmacia, distributed by Baxter Healthcare Corp, Hyland Division,
Glendale, Calif). The goal of therapy is to increase the AT III levels to
more than 80% (protective levels) at all times during therapy. An initial
dose that is calculated to increase the AT III activity to 120% is
administered. The dose is calculated with the following formula: Units
required ([desired % AT III activity] [baseline % AT III activity])
weight (kilograms)/1.2%.12 If AT III activity cannot be measured,
the formula can be used with the assumption that AT III activity is
50%, with a second dose administered 12 hours later that is 50% of the
initial dose. This regimen, in combination with prophylactic antico-
agulation, should provide protection against thromboembolism. AT III
treatment can be discontinued once warfarin therapy is at an appro-
Protein C and S Deﬁciencies. The proteins C and S are vitamin-K
dependent and may be decreased secondarily in patients with chronic
hepatic and renal failure. Protein C is activated by thrombin and becomes
a major anticoagulant and ﬁbrinolytic agent. Protein S serves as a cofactor
to C and has no anticoagulant activity by itself. Deﬁciencies of protein C
usually cause venous thromboses at an early age. Patients with peripheral
arterial disease younger than 50 years of age have a 15% to 20%
incidence of protein C and S deﬁciencies.13 The treatment includes
anticoagulants and factor replacement with fresh frozen plasma during
times of risk. Life-long anticoagulation is indicated in patients with a
history of thrombosis.
Hyperhomocystinemia. Elevated plasma homocystine levels are asso-
ciated with increased risks of arterial and venous thrombosis. Mild
hyperhomocystinemia occurs in 7% of the population.14 The coagulant
effect of homocystine relates to the product of its oxidation. Once
oxidized, potent free radicals form that cause endothelial damage and
activation of platelets and leukocytes. The treatment is folate in doses
sufﬁcient to restore homocystine levels to normal.
Curr Probl Surg, February 2003 95
TABLE 3. Screening for hypercoagulability
A good history, including that of family
Tests during episode
Heparin-induced antiplatelet antibodies
Functional activated protein C resistance
Prothrombin gene variant
Tests after anticoagulation has been stopped (at least 3 days after heparin is stopped)
Proteins C and S
Screening for patients suspected of having a hypercoagulable state
(Table 3) should be individualized. The best test is the patient’s personal
and family history. The platelet count should be determined in every
patient who receives heparin therapy. Certain conditions can be evaluated
at the time of the thrombosis; other conditions cannot be evaluated at the
time because the clotting process and/or anticoagulants may consume
certain factors. Heparin-induced antiplatelet antibodies, the lupus antico-
agulant, anticardiolipin antibodies, homocystine levels, the Leiden muta-
tion, and the prothrombin variant can be assessed during the thrombotic
episode. The remaining tests (protein C, S, antithrombin III) are measured
when the patient is no longer receiving anticoagulation therapy with
warfarin and the heparin has been stopped for at least 3 days. This time
delay allows for replenishment of clotting factors. Table 4 summarizes the
diagnosis and treatment of the common hypercoagulable states.
Prophylaxis of Venous Thromboembolism
Acceptance of the Concept
The concept of prophylaxis has been accepted as valid by European
surgeons more readily than by North American surgeons. Bergqvist15
analyzed the results of 3 surveys that were conducted in Sweden and
found that the proportion of clinics that used some form of prophylaxis
increased from 81% in 1977 to 100% in 1988. A survey of 3500 Fellows
of the American College of Surgeons in 1993 that analyzed attitudes
towards the prevention of postoperative DVT found that 86% of North
American surgeons used some form of prophylaxis.16 Those surgeons did
96 Curr Probl Surg, February 2003
TABLE 4. The common syndromes of hypercoagulability
Syndrome Diagnosis Treatment
Antiphospholipid10 Lupus anticoagulant or Warfarin or LMWH as long as
anticardiolipin antibodies persist
Antithrombin III deﬁciency11 AT III grade 80% Antithrombin concentrate
(ATnativ, Kabi Pharmarcia) in
combination with heparin
Protein C and S Protein C and S levels Anticoagulation for life
deﬁciency13 combined with replacement of
factors during periods of risk
Activated protein-C Measure functional APC- Lifetime anticoagulation
resistnace (APC-R)6–8 R and then PCR for factor
V Leiden mutation
Hyperhomocystinemia14 Homocystine levels Folate
Prothrombin gene variant9 Gene 202210A Lifetime anticoagulation
Heparin-induced Thrombocytopenia, Remove all exposure to heparin,
thrombocytopenia76–83 antibodies to platelets aspirin; alternate
anticoagulant if thrombosis
diagnosed with Orgaran or
LMWH, Low molecular weight heparin; ATIII, antithrombin III; PCR, polymerase chain reaction.
not rationalize the decision on the basis of the risk of complications, vis
a vis the risk of DVT. Prophylaxis is used in a higher proportion of
patients in teaching hospitals. The increased use of prophylaxis is due to
an increased awareness of the problem, improved mechanical devices,
and improved drug therapies. There is an increased usage of mechanical
devices, particularly intermittent pneumatic compression, in the United
States as compared with Europe. North American general surgeons are
particularly reluctant to use heparin as a method of prophylaxis on a
routine basis, reserving it for very high-risk patients in combination with
mechanical methods. A signiﬁcant number of patients still receive aspirin
therapy for venous prophylaxis, despite the negative data and the lack of
endorsement from the National Institutes of Health.
Factors associated with higher risk for DVT after surgical procedures
are increased age, obesity, cancer, a history of DVT, recent surgical
procedures, and hypercoagulable states.17 The development of DVT in
pediatric patients is unusual, and prophylaxis is unnecessary unless 3 or
more of the mentioned risk factors are present.18 The second most
common cause of death in patients with cancer is thromboembolism.19
Curr Probl Surg, February 2003 97
The highest risk occurs in patients with carcinoma of the pancreas, female
reproductive system, breast, colon, lung, and prostate.20 The risk for
embolic complications after DVT appears to be greatest in the ﬁrst 5 days
Most authorities agree that spontaneous DVT begins within the valve
cusp.21 One theory is that venous dilatation is responsible for local
hypoxic injury at the site of the early thrombus formation.22 Further
development of this hypothesis has focused on the role of vasoactive
mediators that are produced at the site of tissue injury, enter the
circulation, and have a vasodilatory effect on distant veins, which results
in endothelial damage.
Intermittent Pneumatic Compression
A meta-analysis of all original articles from June 1966 to June 1996
evaluated the clinical effectiveness of intermittent pneumatic compres-
sion (IPC) devices in the prevention of DVT and pulmonary embolism
(PE) and compared these results to those with knee-high sleeves and
thigh-high sleeves.23 IPC devices reduced the relative risk of DVT by
62% when compared with placebo, by 47% when compared with
graduated compression stockings, and by 48% when compared with
low-dose heparin. IPC devices signiﬁcantly reduced the risk of DVT
in high-risk patients, such as patients who undergo orthopedic and
general surgical procedures. When compared with warfarin in ortho-
pedic patients, IPC was better at preventing calf vein thrombosis, and
warfarin was better at preventing proximal DVT. IPC devices were not
protective against PE. IPC appears to have a beneﬁcial effect beyond
the mechanical effect of compression on venous ﬂow. There are 2
beneﬁcial hematologic effects of IPC. First, IPC stimulates endoge-
nous ﬁbrinolytic activity. The precise mechanism of this effect is not
clear, although tissue plasminogen activator is increased with IPC.24
The second hematologic effect of IPC is the stimulation of tissue
factor pathway inhibitor.25
Prevention of venous thromboembolism requires that every hospital
have a written policy and formal strategy (Table 5).26 Aspirin should
not be used as a sole prevention strategy (evidence grade 1A). General
surgical patients who are at low risk for DVT are those who are
younger than 40 years of age and undergoing minor procedures. They
98 Curr Probl Surg, February 2003
TABLE 5. Recommendations for venous prophylaxis15,17,23,26
Surgical risk Recommendation
Low: 40 yrs, minor procedure Early ambulation (grade 1C)
Moderate: minor procedures with Elastic stockings or IPC or low-dose UFH-
thrombosis risk factors, age 40–60 5000 IU subcutaneously q12h (grade
yrs, major operation in patients 40 1A)
High: 60 yrs, 40 yrs with major LMWH* or IPC (grade 1A)
procedures and risk factors for
Very high: 60 yrs with associated risk LMWH* and IPC (grade 1A)
factors for thrombosis
Total joint replacement32 LMWH* or warfarin started before or
immediately after operation and
continued (grade 1A)
Very high with additional high risk for IPC (grade 1C)
See Table 6.
IPC, Intermittent pneumatic compression; UFH, unfractionated heparin; LMWH, low molecular
should have early ambulation as the sole recommended prophylaxis
(grade 1C). Minor surgical procedures in patients with additional risk
factors for thrombosis and every patient younger than 40 years of age
who undergoes major procedures and who have no additional risk
factors compose a moderate risk group. This group should have
low-dose UFH (5000 IU, subcutaneously, q12h), LMWH (Table 6),
elastic stockings (ES), or an IPC device (grade 1A). Higher risk
patients older than 60 years of age or patients older than 40 years of
age who undergo major procedures or with additional risk factors
should be treated with LMWH or intermittent compression (grade 1A).
Higher risk patients with an increased risk for bleeding should receive
mechanical prophylaxis with ES or IPC (grade 1C). Very high-risk
patients should receive heparin combined with ES or IPC (grade 1C).
Patients who undergo total hip replacement or total knee replacement
should have LMWH or adjusted-dose warfarin therapy started before
or immediately after the operation (grade 1A). These recommenda-
tions are suitable for all surgical specialties, with anticoagulants added
to mechanical devices as the risk of thrombosis increases. Duplex
scanning should be used liberally, particularly when anticoagulants are
not administered because of the risk of bleeding. All anticoagulants
should be used with caution in cases that involve either spinal or
epidural anesthesia (grade 1C ).
Curr Probl Surg, February 2003 99
TABLE 6. Low molecular weight heparin preparations
Drug Dosage for prophylaxis Dosage for treatment
Ardeparin (Normiﬂo) 50 anti-Xa IU/kg, NA
Dalteparin 2500-5000 IU, 200 IU/kg subcutaneously daily
(Fragmin) subcutaneously daily
Enoxaparin 30 mg, subcutaneously 1 mg/kg, subcutaneously q12h*
(Lovenox) twice daily, or 40 or 1.5 mg/kg daily (inpatients
mg, subcutaneously only)*
Tinzaparin NA 175 anti-factor Xa IU/kg daily*
NA, Not approved.
Anti-factor Xa activity should be measured to guide dosing for patients who weigh more than
Critically Ill Patients. Acutely ill and/or critically ill patients are at
particular risk for venous thromboembolic complications and can not
afford the additional stress of thrombosis and PE. Two reports have
focused on the problem of venous thromboembolism in this group of
patients. In the ﬁrst group, no patient received speciﬁc prophylaxis, and
all patients underwent venography to assess the incidence of DVT.27 In
the second group, patients were assigned randomly to receive either UFH
(5000 IU, subcutaneously q12h) or LMWH (enoxaparin 30 mg, subcu-
taneously q12h).28 The incidence of DVT was reduced by 30% in the
patients who received LMWH, as compared with the group who received
UFH. There were more major hemorrhages in the LMWH group. The
relative risk for hemorrhage with the use of LMWH has been studied in
trials that involved patients with serious medical illnesses. The FRAMI
(fragmin in acute myocardial infarction) study showed that a higher dose
grade was associated with increased efﬁcacy, but a higher risk of
bleeding.29 The TIMI II (thrombolysis in myocardial infarction) study
demonstrated more bleeding complications when the higher of 2 doses of
LMWH was used.30 Some caution must be exercised before these results
are applied to critically injured patients. The latter tend to be younger than
patients with serious medical illness. Patients with traumatic injuries
should have prophylaxis with LMWH as soon as it is considered safe;
when this is contraindicated, ES or IPC should be used (grade 1C).
Screening with duplex ultrasound scans is indicated if prophylaxis is
suboptimal. If anticoagulation is contraindicated and DVT is diagnosed,
100 Curr Probl Surg, February 2003
an inferior vena cava (IVC) ﬁlter should be inserted (grade 1C ).31 IVC
ﬁlters should not be used as primary prophylaxis (grade 1C).
Colorectal Surgery. Colorectal surgery is associated with a higher risk
of thromboembolism than general surgical procedures. A review of the
literature by Wile-Jorgensen and colleagues32 identiﬁed 3 studies that
focused exclusively on colorectal procedures. These authors then suc-
cessfully contacted 16 other authors who included patients with colorectal
problems in their series. The ensuing analysis demonstrated that any type
of heparin is superior to no treatment or placebo in the prevention of
DVT, that LMWH and UFH were equally effective, and that the addition
of mechanical compression to any form of heparin is superior to heparin
Patients with Cancer. The hypercoagulable tendencies of patients with
cancer and the increased risk for thromboembolism must be balanced
with increased risk of bleeding complications in patients undergoing
anticoagulation therapy.33 The risk of bleeding in patients with cancer
increases as the INR reaches supratherapeutic levels, and some investi-
gators suggest IVC ﬁlter placement as the primary means of prophylax-
is.34 Bergqvist and colleagues35 examined the optimal duration of therapy
for patients with cancer who undergo planned curative resections. Patients
(n 332) were assigned randomly to receive enoxaparin (40 mg,
subcutaneously daily) for 6 to 10 days, followed by either placebo or
enoxaparin for another 21 days. Bilateral venograms were performed
between days 25 and 31 in every patient. The primary efﬁcacy end point
was DVT between days 25 and 31, and the primary safety end point was
bleeding during the 3-week period after randomization. The authors found
that the rates of DVT in the placebo group were signiﬁcantly higher
(13.8% vs 5.5%; P .01), with no differences noted in bleeding or other
complications, and concluded that treatment should continue for 4 weeks
after planned curative operations for abdominal or pelvic cancer. These
data are comparable to those data that document the efﬁcacy of prolonged
treatment with LMWH after orthopedic procedures36 and noncardiac
thoracic surgery37 and do not support the placement of IVC ﬁlters as the
primary means of prophylaxis. A meta-analysis of studies that compared
anticoagulation with IVC ﬁlters in patients with malignant disease
concluded that both treatments were equally effective but that ﬁlter
placement was less expensive and provided a greater gain in quality-
adjusted life expectancy, when anticoagulation monitoring was consid-
ered.38 Nonetheless, the preponderance of opinion favors a selective
approach to IVC ﬁlter placement, with anticoagulation used as primary
Curr Probl Surg, February 2003 101
Peripheral Vascular Patients
Infrainguinal Bypass. Leg swelling is a common occurrence after
infrainguinal bypass, but the incidence of DVT is not well documented.
Certainly the immobility, limb edema, and associated hypercoagulable
conditions would seem to put this patient population at risk. Passman and
colleagues39 prospectively evaluated 74 patients who had undergone
infrainguinal bypass grafting with bilateral venous duplex scanning
before the operation and then 1 week and 6 weeks after the operation.
These investigators found that the risk of DVT was low (2.8%) and that
routine prophylaxis was not indicated.
Amputations. The association between DVT and patients who undergo
amputation has been investigated, with conﬂicting results. One of the
problems with most of the earlier studies is the method of diagnosis of the
DVT. Now that the accuracy of duplex scanning for DVT has been
validated, an accurate assessment of the relationship can be determined.
Yeager and colleagues40 used duplex scanning of both lower extremities
to evaluate prospectively 31 patients with above-knee amputations and 41
patients with below-knee amputations. The overall incidence of DVT was
11%. These investigators concluded that patients with a history of DVT
were at higher risk, that the DVT often occurred in the contralateral leg,
and that DVT may be present before the amputation. Their recommen-
dations include routine duplex scanning in any patient before amputation,
but DVT prophylaxis with anticoagulants is not routinely pursued.
Duration of Prophylaxis
Hull and colleagues41 evaluated extended out-of-hospital LMWH pro-
phylaxis after elective hip arthroplasty. The authors reviewed data from
206 relevant studies and concluded that the evidence supported that the
extended use of LMWH as both effective and safe. The optimal duration
of anticoagulant therapy was not deﬁned in these studies, and extended
therapy was associated with a risk of major bleeding that approached 3%
per year.42 Kearon and colleagues43 examined the optimal duration of
anticoagulant therapy after a ﬁrst episode of DVT in a multicenter
randomized trial. These authors found that there is a high rate of
recurrence in those patients with idiopathic DVT when anticoagulation is
discontinued after 3 months of therapy. Patients who receive placebo had
a 27.4% incidence of recurrence as compared with an incidence of 1.3%
in patients who received warfarin therapy. The only patients at a higher
risk for a recurrence were those patients with the lupus anticoagulant.
Even the presence of factor V (Leiden) did not predict recurrence.
102 Curr Probl Surg, February 2003
Low Molecular Weight Heparin
Mechanism of Action
LMWHs are at least as effective and safe as UFH and more convenient
to use. They can be used without laboratory monitoring, and there is a
more predictable relationship between dose and response. The anticoag-
ulant effects of the 2 types of heparin are different. The anticoagulant
effect of UFH is largely due to the inhibition of thrombin by the formation
of a 3-way complex of thrombin, antithrombin, and heparin. LMWH
(fractions and fragments of UFH) has less antithrombin activity, and the
anticoagulant effect is due to anti-factor Xa inhibition. The increased
bioavailability and plasma half-life enable once-per-day dosing as com-
pared with multiple injections for UFH. Dosing is based on body weight,
and monitoring is not necessary, except in patients with renal failure
because the excretion of drug is through the kidneys. Several experimen-
tal and clinical studies suggest that an improved cost/beneﬁt ratio can be
expected with LMWH compared with UFH.
Comparisons to Unfractionated Heparin
A recent meta-analysis was performed to (1) verify the clinical effect of
LMWH in the prevention of venous thromboembolism compared with
UFH, placebo, or no treatment; (2) to examine the effect of LMWH in a
subset of patients with cancer, and (3) to compare results with low and
high prophylactic dosages.44
Prevention. Data from 59 RCTs and 48,624 patients qualiﬁed for use in
this analysis. Primary end point veriﬁcation with objective tests (such as
duplex ultrasonography, ﬁbrinogen uptake, impedance plethysmography,
or venography) were required. Patients who underwent orthopedic,
noncancer thoracic, or noncancer laparoscopic procedures were excluded
from the analysis. Secondary end points included bleeding and thrombo-
The incidences of venous thromboses, PE, and death for the no
treatment and placebo groups were 14.5%, 0.5%, and 0.9%, respectively.
These data put the patient who underwent general surgical procedures at
comparable risk for thromboembolic complications as patients with acute
myocardial infarction45 but at less risk than those patients who underwent
orthopedic procedures.46 The incidence of DVT was reduced by 72% in
the LMWH group (P .001). The incidence of PE was reduced by 71%
in the LMWH group (P .018), and the overall mortality rate was
reduced by 46% (P .09). Hemorrhage was more frequent in the
anticoagulated group (103% for major hemorrhage, 106% for total
Curr Probl Surg, February 2003 103
hemorrhage, 88% for wound hematoma, and 53% for transfusion require-
ments; P .001 for all criteria). Wound hematomas occur more
frequently than major hemorrhage (10.3% vs 2.8%). When LMWH was
compared with UFH, LMWH was statistically superior for the prevention
of DVT, PE, all thromboembolism, major hemorrhage, total hemorrhage,
and wound hematoma. UFH was superior in reduced mortality rates and
transfusion requirements. When only the well-identiﬁed double-blind
series were analyzed the trends in favor of LMWH disappeared.
Efﬁcacy in Patients with Cancer. When the subset of patients with
cancer was analyzed, the efﬁcacy and safety of LMWH relative to UFH
were similar to those in patients with nonmalignant disease. Low-dose
LMWH (3400 anti-Xa units or less) did not signiﬁcantly reduce the
incidence of venous thromboembolism compared with UFH but did
reduce the incidence of hemorrhagic risk (P .005). High-dose LMWH
(greater than 3400 anti-Xa units) was more effective than UFH, particu-
larly in regard to the prevention of PE, but this beneﬁt was counterbal-
anced by a signiﬁcant increase in the risk of major hemorrhage. These
data are similar to those reported by Kakkar47 in general surgical patients.
Best evidence therefore suggests a close similarity between LMWH and
UFH with regard to both thromboembolic and hemorrhagic events, but
the ease of use of LMWH makes it preferable in contemporary practice.
Dosing Information. Low-dose LMWH provides comparable efﬁcacy
and increased safety over UFH; high-dose LMWH does not improve
efﬁcacy, and safety is strongly decreased.48 Dosing information for
commercially available LMWHs is listed in Table 6.
The Use of Oral Anticoagulants49
The Initiation of Therapy
Warfarin therapy should begin with a daily maintenance dose of 5 mg.
Lower initial doses should be considered in elderly patients, patients with
hepatic dysfunction or malnutrition, and patients at high risk for bleeding.
Warfarin should not be administered to pregnant patients during the ﬁrst
and third trimester, to patients with alcoholism or drug abuse, and patients
with unsupervised dementia. A loading dose is not necessary and, in
certain situations, can be dangerous. Warfarin can be started simulta-
neously with heparin, unless an underlying hypercoagulable state is
suspected. In those situations, heparin anticoagulation should precede the
administration of warfarin. All dose adjustments should be based on the
international normalized ratio (INR) level. The appropriate INR targets
for a variety of conditions are listed in Table 7.
104 Curr Probl Surg, February 2003
TABLE 7. Recommended international normalized ratio (INR) levels116
DVT prophylaxis 2.5 (2.0-3.0) 4-6 Weeks after operation
DVT, episode 1 2.5 (2.0-3.0) 3-6 Months
PE, treatment 2.5 (2.0-3.0) 3-6 Months
Arterial embolus, AF 2.5 (2.0-3.0) Life
Arterial embolus, MI 3.0 (2.5-3.5) Variable
Bioprosthetic heart valves 2.5 (2.0-3.0) 3 Months after operation
Mechanical heart valves 3.0 (2.5-3.5) Life
Prosthetic bypass grafts89 3.0-4.5 Life of bypass graft
DVT, Deep venous thrombosis; PE, pulmonary embolism; AF, atrial ﬁbrillation; MI, myocardial
Indications and International Normalized Ratio
Warfarin in adjusted doses to maintain the INR between 2.0 and 3.0 is
indicated for the following conditions: prophylaxis of DVT of high-risk
surgery, treatment of DVT, treatment of PE, prevention of arterial
embolism after acute myocardial infarction, in patients with atrial
ﬁbrillation and tissue heart valves, and prevention of the propagation of
existing thrombus. The INR should be maintained in the range of 2.5 to
3.5 for patients with mechanical heart valves and for patients with
antiphospholipid syndrome. When warfarin is started with a dosage of 5
mg, the INR should reach the desired level in 4 to 5 days. If the INR on
day 2 of therapy exceeds 1.5, the dose of warfarin should be reduced. If
the INR is less than 1.5 on day 3, a higher maintenance dose will be
necessary. An INR between 1.5 and 2.0 suggests that the maintenance
dose will be 5 mg; an INR of more than 2.0 suggests that a lower dose will
The Treatment of Patients with Prolonged International
Normalized Ratio Levels
The treatment of patients with excessive prolongation of the INR
depends on the level, whether the patient is bleeding, and whether the
patient is to undergo an invasive procedure. If the INR is less than 5.0 and
there is no bleeding or anticipated intervention, the warfarin should be
withheld until the INR is therapeutic. Patients with an INR of more than
5.0 but less than 9.0 who require rapid reversal before an intervention
should receive vitamin K1 (2-4 mg orally) and another 1 to 2 mg of
vitamin K1in 24 hours if the INR remains high. Warfarin can be resumed
Curr Probl Surg, February 2003 105
when the INR is at target levels. Patients with an INR of more than 9.0
and no signiﬁcant bleeding should receive vitamin K1 (3-5 mg orally)
with additional vitamin K1 after 24 hours, as necessary. For an INR of
more than 20 that is associated with serious bleeding, patients should
receive vitamin K1 10 mg by slow intravenous infusion that is supple-
mented with fresh frozen plasma (15 mL/kg) or prothrombin complex,
depending on the urgency.
The risk of bleeding complications in patients who are treated with
long-term anticoagulants has been evaluated in numerous RCTs. A
meta-analysis of the trials that evaluated anticoagulation for atrial
ﬁbrillation, recent myocardial infarction, and prosthetic heart valves by
Gullov and colleagues50 determined that the incidence of fatal and major
hemorrhage was reduced signiﬁcantly by better control because of the
adoption of the INR. Regardless of the duration of therapy, patients who
receive warfarin anticoagulation therapy must be monitored closely.
Patients who are enrolled in the Veterans Affairs Stroke Prevention Trial
in nonrheumatic atrial ﬁbrillation were monitored weekly for 12 weeks
and then monthly. Considerable dose adjusting was required, with only
28% of the patients in the desired range at 1 week and to 65% in the
desired range at week 12. Dosing adjustments were required in 40% of the
patients in the monthly surveillance period.51
The most serious nonhemorrhagic complication of warfarin therapy is
skin necrosis that occurs in patients with protein C and S deﬁciency. In
this condition, a hypercoagulable state occurs after 2 to 3 days of warfarin
therapy. This complication can be avoided if all patients are treated with
heparin for the ﬁrst 4 days of warfarin therapy.
Efﬁcacy in Patients with Coronary Artery Disease
The value of aspirin in the secondary prevention of thrombotic
complications from coronary artery disease (CAD) has been established
ﬁrmly.52 There is emerging evidence that the simultaneous alteration in
platelet activity and ﬁbrin formation may be more effective than either
strategy alone.53 The ﬁrst Northwick Park Heart Study found an associ-
ation between the level of factor VII activity and CAD. Those patients in
whom CAD developed had elevated levels of factor VII (117%) as
compared with patients who were at low risk whose factor VII levels were
below normal (70%).54 This reduction in factor VII corresponds to raising
the INR to 1.5. This is less than the level necessary for efﬁcacy in venous
thromboprophylaxis and likely to cause less major bleeding. A trial that
106 Curr Probl Surg, February 2003
evaluated low intensity oral anticoagulation with warfarin and low-dose
aspirin in the primary prevention of CAD was conducted in the United
Kingdom.55 A total of 5499 men who were at high risk for CAD were
assigned randomly into 4 groups: active warfarin and active aspirin
(group WA), active warfarin and placebo aspirin (group W), placebo
warfarin and active aspirin (group A), and placebo warfarin and placebo
aspirin (group P). Patients in the active aspirin groups received 75 mg
daily in a controlled release formulation. The warfarin dosage was
adjusted to maintain the INR at 1.5. The WA group experienced
signiﬁcantly fewer ischemic cardiac events that the P group (8.7 vs 13.3
per 1000 person years; P .006). Comparisons among the W and A
groups with the P group showed beneﬁts (20% reduction in CAD events),
but these differences were not statistically signiﬁcant. These data have
implications in high-risk patients who undergo operation, providing that
the beneﬁts of cardiac protection exceed the risks of major hemorrhage.
Estrogen Therapy and Pregnancy56
Estrogen and Thrombosis
Many women receive chronic estrogen therapy either as a contraceptive
or later in life as replacement therapy, although the latter practice is likely
to decrease with the new information from the Women’s Health Initiative
Trial.57 This study, which evaluated the use of estrogen and progestin in
healthy postmenopausal women, found that the overall health risks
exceeded the beneﬁts of replacement therapy over an average period of
5.2 years. This trial will result undoubtedly in fewer women receiving
hormone therapy and less risk for thromboembolic problems that are
associated with surgical interventions. There is ample evidence that
indicates an increased risk for venous thromboembolism in women who
are receiving estrogen therapy, which ranges from 2 times to 11 times
normal.58 The addition of an inherited or acquired coagulation disorder
raises the risk of DVT 50-fold in a patient receiving estrogen therapy.59
The incidence of DVT is 5 times higher in pregnant patients than in
nonpregnant patients. Factors that may make pregnant women more
susceptible to the development of a DVT are varicose veins, iliac vein
compression, and stasis. In addition, levels of factors I, VII, VIII, IX, X,
XI, XII, and platelets are increased; antithrombin III and protein S are
decreased. These changes revert to baseline within 2 months of deliv-
ery.60 The Leiden mutation in the factor V gene predisposes patients to
thrombotic events and is relatively common in the obstetric population.
Curr Probl Surg, February 2003 107
The prothrombin gene variant (20210A), protein C and S deﬁciencies,
and anticardiolipin antibodies are also relatively common.61
Venous Prophylaxis during Pregnancy
Heparin does not cross the placental barrier and has become the gold
standard anticoagulant during pregnancy. Patients who receive long-term
warfarin anticoagulation therapy should be counseled about the risks of
pregnancy. If pregnancy is elected, patients can be switched to adjusted-
dose UFH or LMWH. Pregnant patients with mechanical heart valves
present an added challenge because they require therapeutic anticoagu-
lation that can be achieved in several ways throughout the pregnancy.
Adjusted-dose UFH can be used throughout the pregnancy. Alternatively,
heparin can be used until week 13; warfarin can be used until the middle
of the third trimester, and then heparin can be restarted until delivery.
Pregnant patients who are at moderate risk are those patients with a
previous DVT that was associated with a transient risk factor and no
current risk factors (such as morbid obesity or strict bed rest). Duplex
surveillance and postpartum anticoagulants are indicated in this group
(grade 1C). Patients with a history of an unexplained DVT but who are
not receiving long-term anticoagulation therapy should have duplex
surveillance, low-dose UFH (5000 IU, subcutaneously q12h), or prophy-
lactic LMWH (Table 6) in addition to postpartum anticoagulants (grade
1C ). Patients who are at very high risk for recurrent thromboembolism
should be treated with intravenous UFH to maintain the activated partial
thromboplastin time (APTT) within the therapeutic range (grade 1C). If
LMWH is chosen in place of UFH, the dosing should be adjusted
according to body weight. Heparin therapy must be discontinued 4 to 6
hours before delivery. Women with more than 3 miscarriages and women
with a history of preeclampsia or abruption and antiphospholipid antibody
should receive prophylactic LMWH (grade 1A). Anticoagulation should
be continued for 6 weeks or 3 to 6 months if a diagnosis of venous
thromboembolism is veriﬁed. Warfarin can be used in the postpartum
period, initially overlapped with LMWH or UFH until the INR is more
The role of antiplatelet agents has been investigated in arterial throm-
bosis and embolism, occlusive and hemorrhagic strokes, angina and
myocardial infarctions, venous thromboprophylaxis and PE, and resteno-
108 Curr Probl Surg, February 2003
ses and occlusions after arterial reconstructions. It is now evident that the
platelet plays a role of varying signiﬁcance in all of these areas. Platelets
are derived from the bone marrow megakaryocytes and have a life span
of 10 days. The intact endothelium is nonreactive to the platelet, but in the
setting of an interruption of a blood vessel or certain immune complexes,
the platelet becomes activated. When this occurs, the platelet changes
shape from a disc to a spiny sphere, and the surface membrane develops
the capacity to catalyze interactions between activated coagulation
proteins. The platelet thus interacts with the coagulation cascade in the
early stages of thrombus formation to form thrombin that generates the
ﬁbrin that is necessary to stabilize the hemostatic plug.
Mechanisms of Blocking Platelet Function
There are 5 avenues for blocking platelet function and aggregation: (1)
blocking cyclooxygenase and thromboxane A2 production, (2) blocking
thromboxane synthase, (3) blocking the thromboxane A2-receptor, (4)
inhibiting exposure of Gp IIb/IIIa receptor sites, or (5) directly blocking
these receptors. The ﬁnal common pathway by which all platelet agonists
ultimately effect platelet aggregation is the transformation of the Gp
IIb/IIIa receptor site on the platelet membrane.
Mechanism of Action. Aspirin, the most commonly used anti-platelet
agent, irreversibly inhibits the formation of thromboxane A2 and rein-
forces the effects of other platelet antagonists.62 Aspirin doses between 75
and 162.5 mg are effective for all indicated uses (Table 8). Regular
aspirin is absorbed in the stomach with peak plasma levels, which occur
in 30 minutes. Enteric coated aspirin is absorbed more slowly with
evidence of antiplatelet activity in 3 to 4 hours after administration. The
antiplatelet effect lasts for the life of the platelet.
Indications for Use. Aspirin as a primary preventive measure for CAD
is not recommended for patients younger than 50 years of age with no
other risk factors (grade 2B). Aspirin is recommended for patients older
than 50 years of age with a major risk factor for CAD such as cigarette
smoking, hypertension, diabetes mellitus, elevated serum cholesterol, and
family history (Grade men 2A, grade women 2C). Aspirin is also
recommended for patients with noncardiac transient ischemic attack
(TIA) but in this situation is more effective when combined with
dipyridamole (grade 1A). Aspirin should be administered to patients
before and after carotid endarterectomy (CE; grade 1A) and femoral
Curr Probl Surg, February 2003 109
TABLE 8. Recommendations for the use of aspirin52,62– 65,117
Indication Strength of recommendation
Primary prevention of CAD in patients 50 yrs of Not indicated (grade 2B)
age without risk factors
Primary prevention of CAD with risk factors55 Indicated in men (grade 2A)
Indicated in women (grade 2C)
TIA, noncardiac Indicated, but more effective with
dipyridamole (grade 1A)
Before and after operation for patients who Indicated (grade 1A)
undergo carotid endarterectomy70,71
Patients with angina pectoris and/or myocardial Indicated (grade 1A)
Patients with atrial ﬁbrillation Not indicated (grade 1A)
Patients with prosthetic arterial bypass grafts87–89 Indicated - started pre-op (grade
1A); combine with warfarin for
high-risk grafts (grade 1A)
Patients with saphenous vein grafts85 Indicated (grade 1C )
Venous thromboprophylaxis46 Not indicated (grade 1A)
CAD, Coronary artery disease; TIA, transient ischemic attack.
popliteal bypass grafts (grade 1C for saphenous vein, grade 1A for
prosthetic). Any patient with the clinical impression of an acute myocar-
dial infarction or angina pectoris should receive aspirin, which should be
continued indeﬁnitely (grade 1A). Aspirin should not be administered
concurrently with warfarin, except for patients with very high risk for
thromboembolism or failure of either drug alone (grade 2C). Patients with
atrial ﬁbrillation are better treated with warfarin, and aspirin does not add
therapeutic beneﬁt. The use of aspirin with warfarin is indicated in certain
situations such as arterial embolization from mechanical heart valves and
atrial ﬁbrillation despite adequate anticoagulation. When administered
together, the risk of bleeding is increased signiﬁcantly.63 The bleeding
risk can be lowered by the use of low doses of both aspirin and warfarin
(INR, 1.5-2.0), but this may not be effective in these difﬁcult to treat
patients. Although aspirin reduces the incidence of vascular occlusions by
25%, most arterial thrombotic events still occur in patients who take
Side Effects. The most common side effects of aspirin are gastrointes-
tinal and range from mild erosions to frank ulceration of the gastric
mucosa. The most alarming side effect, however, is the small, but real,
incidence of intracranial hemorrhage in healthy men who receive aspirin
for prophylaxis against heart attack and stroke.65 When aspirin is used as
secondary prophylaxis (ie, in higher-risk patients), the risk of intracranial
bleeding is superceded because of the relatively greater beneﬁt of aspirin
in this setting.
110 Curr Probl Surg, February 2003
Adenosine Diphosphate Receptor Antagonists
A group of antiplatelet agents that act as speciﬁc adenosine diphosphate
receptor antagonists has been approved. The 2 drugs that are available
currently are ticlopidine hydrochloride (Ticlid) and clopidogrel bisulfate
(Plavix). Ticlid is indicated for patients who have had a stroke or TIA and
are unable to take aspirin. The effect of Ticlid reaches 50% of maximum
after 4 days of treatment. Problems with neutropenia (1.6%) and agran-
ulocytosis (0.8%) require regular monitoring and have limited its usage.
Plavix is indicated in patients at high risk for myocardial infarction and/or
stroke and has fewer serious side effects. Its effect is rapid, achieving
platelet inhibition within 2 hours of dosing. The most common side effect
is rash (0.3%). This agent can also be used when aspirin cannot be used
and in cases of aspirin failure. The clopidogrel versus aspirin in patients
at risk of ischemic events (CAPRIE) trial that involved 19,000 patients
showed that clopidogrel was superior to aspirin in reducing arterial
thrombotic events and was safer with fewer hemorrhages.66
A study that examined whether the combination of aspirin and clopi-
dogrel increased the risk of bleeding was performed in response to the
increasing use of combination therapy in contemporary practice.67 The
hematologic end point in this trial was the measurement of bleeding
times, which were elevated after combination therapy. This does not
correlate necessarily with surgical bleeding but raises some concerns,
because combination therapy will be used increasingly as the results of
the clopidogrel in unstable angina to prevent recurrent events trial
(CURE) study are made available. This trial showed a signiﬁcant beneﬁt
for patients with acute coronary syndromes who were receiving combined
Gp IIb/IIIa Inhibitors
Antiplatelet agents directed against the ﬁnal common pathway, the
ﬁbrinogen receptor on the platelet membrane, are extremely potent and
are used with a greater risk of bleeding complications. Abciximab
(ReoPro) is a monoclonal antibody directed against the Gp IIb/IIIa
receptor sites. This and several other agents in various stages of
development have the net effect of reducing the amount of platelet
membrane available to form thrombus. Abciximab blocks the platelet
aggregation seen in patients with heparin-induced thrombocytopenia
(HIT) but is not a sufﬁcient anticoagulant as a sole agent in this condition.
ReoPro can be used with heparin and aspirin to decrease cardiac ischemic
Curr Probl Surg, February 2003 111
TABLE 9. Antiplatelet agents (non-aspirin)
Drug Indication for use Action/timing
Abciximab (ReoPro)74,75 Used with heparin and Group IIb/IIIa inhibitor
aspirin to reduce Onset of action within
complications after minutes, half-life 24–48
PTCA and unstable hours
Clopidogrel (Plavix)66–68 Reduce atherosclerotic ADP receptor antagonist
events (MI, stroke, Onset of action within 24
vascular death) in high- hours - duration 5 days after
risk patients stopping
Dipyridamole (Persantine)63 Additive effect with aspirin Inhibits phosphodiesterase
in patients with TIA
Eptiﬁbatide (Integrelin) Acute coronary syndrome Group IIb/IIIa antagonist
with heparin and aspirin Has a reversible effect when
PTCA drug is discontinued and
action is very brief
Ticlopidine (Ticlid)94 CVA prophylaxis in ADP receptor antagonist
patients who cannot Onset of action within 4 days
take aspirin Duration 2 weeks after
PTCA, Percutaneous transluminal coronary angioplasty; MI myocardial infarction; ADP,
adenosine diphosphate; TIA, transient ischemic attack; CVA, cerebrovascular accident.
complications before and after percutaneous coronary interventions.
Table 9 lists the indications for use of the nonaspirin antiplatelet agents.
CE is the most frequently performed vascular procedure in the United
States. In the last decade, there has been a large increase in case volume
after conﬁrmation of the efﬁcacy of CE by the North American Symp-
tomatic Carotid Endarterectomy Trial and ACAS trials. Despite the
efﬁcacy of CE over medical treatment for extracranial cerebral vascular
disease, the risk of an ipsilateral perioperative stroke after CE remains at
2% to 7%.69,70 There are 2 main types of operation-related stroke. The
intraoperative stroke, apparent on recovery from anesthesia, is attribut-
able directly to intraoperative ischemia or embolization. The postopera-
tive stroke that occurs sometime after an uneventful recovery from
operation is due to vessel occlusion or embolization from the thrombo-
genic endarterectomized surface. Evidence suggests that platelets adhere
to the exposed surface collagen within minutes of restoring ﬂow.71
Microparticulate embolization can be detected by transcranial Doppler
ultrasound with good correlation between the detection of 25 or more
emboli in the middle cerebral artery distribution during any 10-minute
period after operation and an increased occurrence of TIA and/or stroke.72
Infusion of dextran 40 has been shown to decrease the rate of emboliza-
112 Curr Probl Surg, February 2003
tion of microparticulate debris,73 but no data are available on the role of
Gp IIb/IIIa antagonists in this setting.
Recent success in decreasing thrombotic events in both percutaneous
coronary interventions and acute myocardial infarction with platelet Gp
IIb/IIIa antagonists has stimulated interest in furthering the applications
of these drugs. A recent RCT that used abciximab in acute ischemic
stroke documented safety when it was administered up to 24 hours after
stroke onset, with a trend toward improvement in functional outcomes.74
Abciximab has also been used in conjunction with heparin and aspirin in
carotid artery percutaneous angioplasty and stenting and is thought
potentially to reduce recurrent stenoses.75 Unfortunately, the use of
platelet Gp IIb/IIIa inhibitors with open surgical procedures has not been
studied because of concerns of bleeding.
Occasionally patients who require anticoagulation during a surgical
procedure will have a history of HIT. This immune disorder occurs in 2%
to 4% of patients who receive heparin therapy, and these patients have a
2.6-fold increase in the incidence of perioperative thromboses.76 Patients
with heparin-associated antiplatelet antibodies that are exposed to heparin
may have platelet aggregation and thrombosis. When the diagnosis is
suspected (falling platelet count, resistance to heparin anticoagulation,
unexplained thrombosis), all sources of heparin must be discontinued; the
plasma should be tested for antiplatelet antibodies, and platelet function
should be blocked with aspirin.
When rapid anticoagulation is required, there are 3 classes of drugs that
can be used: (1) LMWH, (2) heparinoids, or (3) thrombin inhibitors.
Although there are reports of the successful use of LMWH, there is a
signiﬁcant risk of cross-reactivity, and the use of these drugs should be
avoided in this situation, unless studies reveal no cross-reactivity with the
patient’s antibodies. The likelihood of cross-reactivity approaches 33%
against enoxaparin (Lovenox) and dalteparin (Fragmin). The heparinoid,
danaproid (Orgaran), is approved for patients with HIT who require
anticoagulation therapy during a surgical procedure. Orgaran is puriﬁed
from porcine intestinal mucosa and has been used as the sole anticoag-
ulant for hemodialysis, cardiopulmonary bypass, and DVT.77–79 Cross-
reactivity is approximately 10%, and patients with HIT who receive
Curr Probl Surg, February 2003 113
Orgaran therapy should undergo platelet reactivity testing.80 Some
caution must be used when Orgaran is used as an anticoagulant during a
surgical procedure because the incidence of bleeding after cardiopulmo-
nary bypass has been reported to be 23%.81 The high incidence of
bleeding is due to the long anticoagulant effect (plasma anti-factor Xa
activity is 25 hours) and the absence of a reversing agent. Protamine
sulfate reverses the antithrombin effect, but not the anti-factor Xa activity.
The intraoperative treatment is further complicated because the activated
clotting time and the activated partial thromboplastin time have no value
in the determination of the anticoagulant effect. The only reliable
intraoperative monitoring assay for Orgaran effect is the anti-factor Xa
assay that is measured every 20 minutes.82 Intraoperative dosing should
be titrated, and only the amount required should be administered. An
anti-factor Xa grade of more than 1.0 U/mL is necessary to prevent clots
in the ﬁlters of the cardiopulmonary bypass machine, but levels of more
than 1.50 U/mL have been associated with excessive postoperative
Lepirudin (Reﬂudan) a recombinant hirudin, has also been used in
patients with HIT. A prospective multicenter trial that involved 82
patients with HIT was conducted to evaluate Reﬂudan.83 There were 51
patients with thrombosis, 5 patients with thrombosis who received
thrombolysis therapy, 18 patients without thrombosis, and 8 patients with
thrombosis during cardiopulmonary bypass. Outcomes were compared
with a historic group of 120 patients. The incidences of death, amputation,
and new thromboembolic events were reduced signiﬁcantly in the patients
who were treated with Reﬂudan; bleeding complications were similar in
both groups. The drug has a half-life of 1.3 hours in patients with normal
renal function and is associated with a rapid recovery of platelet counts
and prolongation of the activated partial thromboplastin time. These are
major advantages when compared with Orgaran. Patients with HIT who
require prolonged anticoagulation therapy should be converted to warfa-
rin therapy as soon as possible.
Antithrombotic Therapy in Peripheral Arterial
Systemic heparin followed by oral anticoagulation is indicated in
patients with atrial ﬁbrillation or post–myocardial infarction and
arterial embolization (grade 1C). The purpose of anticoagulation in
114 Curr Probl Surg, February 2003
TABLE 10. Antithrombotic treatment in peripheral arterial disease 85–90
Arterial embolism Antiplatelet - No
Warfarin - Yes
Aortofemoral bypass Antiplatelet - No
Warfarin - No
Carotid endarterectomy Antiplatelet - Yes
Warfarin - No
Prosthetic femoropopliteal bypass Antiplatelet - Yes
Warfarin - No
Saphenous vein femoropopliteal bypass Antiplatelet - Yes
Warfarin - No
Infrageniculate bypass Antiplatelet - Yes
Warfarin - Yes, with INR of
INR, Internaitonal normalized ratio.
this setting is both to prevent further embolic episodes and to
minimize proximal and distal propagation of thrombus (Table 10).84
Data do not support the use of prophylactic anticoagulation for vascular
reconstructions of high-ﬂow, low-resistance arteries, such as aortofemoral
bypass grafts (grade 1C). Antiplatelet therapy has been shown to reduce the
risk of arterial graft failure in the femoropopliteal segment (grade 1A).85 A
meta-analysis of 5 RCTs that compared antiplatelet therapy with placebo
shows that 12 patients need treatment for 1 to 3 years to prevent 1 graft
occlusion. Antiplatelet therapy is more effective in the prevention of
occlusion of prosthetic grafts compared with vein grafts. Many investigators
believe that oral anticoagulation improves the patency rates after infraingui-
nal reconstruction, but the evidence that supports this belief is weak. The only
RCT that compared oral anticoagulation with placebo showed that 6 patients
required treatment to prevent 1 occlusion and 7 patients required treatment to
prevent 1 amputation.86 The effect of a combination of antiplatelet and oral
anticoagulants on graft occlusion and other vascular complications was
reviewed in a meta-analysis by Tangelder and colleagues.87 Outcomes
measures were occlusion of bypass grafts, the occurrence of stroke and/or
myocardial infarction, and death. The results conﬁrmed the efﬁcacy of
antiplatelet and anticoagulant therapy in prolonging graft patency without
any secondary beneﬁt in other variables.
Veteran Affairs Cooperative Study no. 362 examined the beneﬁts of
long-term warfarin therapy in patients with lower extremity arterial bypass
grafts.88 The results show that the addition of warfarin does not add
Curr Probl Surg, February 2003 115
protection beyond that provided by aspirin, the exception being
prosthetic above-knee prosthetic bypasses. Conﬂicting evidence was
presented by the investigators in the Dutch Oral Anticoagulant or
Aspirin study who found that, when the INR is targeted at 3.0 to 4.5,
warfarin provides added beneﬁt to aspirin.89 Both trials determined,
however, that aspirin treatment provided the major beneﬁt in terms of
graft patency. The explanation for the different effects of warfarin in
these 2 prospective trials probably had something to do with the INR
targets. In the Veterans Affairs study, the INR range that was selected
was 1.4 to 2.8, considerably lower than the bypass oral anticoagulants
or aspirin trial. The Veterans Affairs study also excluded high-risk
patients for thrombosis who were included in the bypass oral antico-
agulants or aspirin study. In another evaluation of warfarin therapy,
Sarac and colleagues90 found that the combination of drugs improved
patency rates in high-risk venous bypass grafts from 56% in patients
who received aspirin therapy alone, as compared with 77% in patients
on the combination over a 3-year period of follow-up. Most of these
studies point out that the long-term administration of warfarin in-
creases the risk of lethal intracranial hemorrhage.
Jackson and colleagues91 examined the effects of warfarin anticoagulation
on the ischemic sequalae of graft occlusion. These investigators found that
the ischemic consequences of polytetraﬂuoroethylene (PTFE) bypass failure
were more severe than after vein graft failure. Warfarin lessened the ischemic
impact of PTFE grafts without an effect in patients with failed vein grafts but
did not lower the incidence of severe ischemia in failed PTFE to the level
seen with failed vein grafts.
McMillen and colleagues92 evaluated the use of LMWH in the
perioperative period after 69 infrageniculate bypass grafts with PTFE.
LMWH was compared with UFH. This was a nonrandomized observa-
tional study that used UFH between April 1993 and October 1994 and
LMWH between October 1994 and January 1996. Coagulation studies
and the mean number of postoperative days were signiﬁcantly lower in
the LMWH group, but there were no signiﬁcant differences in graft
thrombosis or mortality rates.
Antithrombotic Treatment after Catheter-based
The Role of Warfarin
The rapid advances in balloon and stent technology for the treatment of
arterial and venous lesions have been abetted by the use of a variety of
116 Curr Probl Surg, February 2003
anticoagulants. Warfarin was the ﬁrst agent used after a canine experi-
ment by Palmaz and colleagues93 that showed decreased platelet deposi-
tion on intravascular stents in the group treated with a combination of
heparin, aspirin, dipyridamole, and dextran compared with the groups
treated with heparin alone or heparin with aspirin. Somehow, the authors
concluded that warfarin should be added to patients as a “safety net.” As
a result, the early stent trials and the FDA incorporated warfarin into their
guidelines. In 1991 the focus changed away from warfarin in an attempt
to reduce bleeding complications, shorten hospital stays, and improve
outcomes towards aspirin and ticlopidine.94 Data were collected largely in
registries; a large meta-analysis of 2295 patients revealed a statistically
signiﬁcant and clinically meaningful improvement in results after coro-
nary stenting in patients who received ticlopidine therapy. Because the
comparisons were uncontrolled, the transition from warfarin to ticlopidine
was slow in the United States, as reﬂected in a consensus document from
the American College of Cardiology in 1996 that still recommended
warfarin anticoagulation for high-risk stent deployments.95
The Role of Anti-Platelet Agents
The recommendation for warfarin, however, passed because data from
several randomized, placebo-controlled trials unequivocally showed the
superiority of ticlopidine over warfarin, as measured by the mortality rate
at 6 weeks, myocardial infarction, and the need for repeat revasculariza-
tion.96 At this time, clopidogrel has become the agent of choice in
combination with aspirin.97 The ﬁnal chapter has not been written,
however, because new evidence that was presented in the enoxaparin and
ticlopidine after elective stenting trial suggests that LMWH (enoxaparin)
may be the optimal anticoagulant after coronary artery stenting.98
The best available evidence suggests that the following protocol is
appropriate for patients who undergo percutaneous arterial interventions.
All patients should be pretreated with 80 to 325 mg of aspirin (grade 1A).
Plavix (300 mg loading dose followed by 75 mg/day) should be
administered to those patients who cannot tolerate aspirin (grade 1A).
Plavix can be added to aspirin for those patients who undergo stent
placement, but the supporting data are not as strong (grade 2A). Gp
IIb/IIIa inhibitors should be administered to reduce the incidence of
ischemic complications after percutaneous transluminal coronary angio-
plasty (grade 1A). There are no data to support the use of the Gp IIb/IIIa
inhibitors during noncardiac interventions. One of the concerns about the
Curr Probl Surg, February 2003 117
TABLE 11. Indications for perioperative anticoagulation49
Arterial embolism atrial ﬁbrillation 1C
Mitral valve prolapse with documented embolization 1A
Mobile aortic atheroma 4 mm by TEE 2C
Unexplained pulmonary embolism in patient with 1C
patent foramen ovale
Bileaﬂet mechanical aortic valve, tilting disk valve 2A
in mitral position, caged ball or disk valve53
Recent DVT44 1A
Hypercoagulable state58 1A
TEE, Transesophageal echocardiogram; DVT, deep venous thrombosis.
aggressive combination of Gp IIb/IIIa inhibitors and the obligatory
periprocedural use of heparin is the incidence of access site complica-
tions, which can be reduced by the reduction of the heparin dose to 50 to
70 IU/kg when Gp IIb/IIIa agents are used and the removal of the access
sheath when the activated clotting time falls below 150 seconds (grade
Operation in the Patient Who Undergoes
Which Patients Require Perioperative Anticoagulation
Surgeons are often presented with a patient who is receiving anticoag-
ulation therapy who requires an intervention, and the surgeon must decide
whether the anticoagulant can be discontinued safely for the immediate
perioperative period. Indications and guidelines for perioperative antico-
agulation are listed in Tables 11 and 12. If there is a low risk of
thromboembolism, as in patients who are undergoing anticoagulation
therapy for chronic atrial ﬁbrillation or for a history of DVT, the
anticoagulation may be discontinued 3 to 4 days before the procedure,
which allows the INR to return to near-normal values. Warfarin therapy
can then be resumed in these patients when oral intake is started. In
patients with atrial ﬁbrillation, the risk of postoperative thromboembolism
is increased from 4.5% to 17.6% when 2 or more of the following factors
are present: a TIA or stroke or other thromboembolic events within the
past several months, hypertension, or recent congestive heart failure.99
These patients should receive short-term low-dose heparin (5000 IU
UFH, subcutaneously) during the procedure and shortly thereafter, and
118 Curr Probl Surg, February 2003
TABLE 12. Operating on a patient who has received anticoagulation therapy
Low risk (eg, history of DVT, Discontinue anticoagulants 4–5 days before operation
chronic atrial ﬁbrillation) and let INR return to normal; resume warfarin after
Moderate risk (eg, atrial ﬁbrillationLow-dose UFH (5000 IU subcutaneously q12h) during
with recent history of embolic and shortly thereafter the procedure until warfarin is
event, congestive heart failure) resumed; LMWH is also a good choice here
High risk (eg, mechanical heart Full anticoagulation with UFH
Endovascular intervention in patient Reduce heparin dose to 50–70 mg/kg and remove
receiving Gp IIb/IIIa inhibitor or access sheath when ACT 150 seconds
Open procedure in patient receiving Be prepared to use Tisseel for topical hemostasis
antiplatelet agents, particularly
DVT, Deep venous thrombosis; INR, international normalized ratio; UFH, unfractionated
heparin; LMWH, low molecular weight heparin; ACT, activated clotting time.
warfarin therapy can be resumed after the operation. LMWH is a good
choice here, except for patients with mechanical heart valves who require
full anticoagulation with UFH. Both warfarin and heparin pose an
increased risk for bleeding complications.100 When patients who take
warfarin therapy need dental work and there is a risk of oral hemorrhage,
the patient can be given a mouthwash with epsilon aminocaproic acid, and
the anticoagulant therapy can be continued (grade 2B). There are several
studies that demonstrate the safety of operating in patients who receive
Is LMWH a Better Intraoperative Anticoagulant?
Although most surgeons continue to use UFH in the operating room, it
is well recognized that there is considerable variability between individ-
uals to ﬁxed dosages. After an intravenous dose, the disappearance curve
of UFH follows nonlinear kinetics and has an initial disappearance phase
followed by a convex curve.103 There are data to suggest that LMWH has
signiﬁcant advantages in terms of predictable levels of anticoagulant
activity. The pharmacokinetics of intravenously administered UFH and
LMWH during aortofemoral bypass grafting were analyzed and com-
pared.104 The anticoagulant effect, as measured by anti-factor Xa activity,
was more sustained in the LMWH group, and the interindividual variation
of anti-factor Xa activity in the LMWH group was smaller. When using
LMWH, the surgeon must recognize that the anticoagulant effect does not
correlate with the APTT but rather with the anti-factor Xa activity and
Curr Probl Surg, February 2003 119
that the latter effect is not reversed with protamine. The ability to monitor
anti-factor Xa activity is essential if one wishes to use LMWH in the
operative setting because concentrations that exceed 0.4 anti-factor Xa
produce a critically enhanced risk of hemorrhage.105
The anticoagulant effect of heparin is reversed by protamine, but many
studies have concluded that the coagulation status does not need to return
to baseline normal to insure hemostasis.106 In a study that evaluated the
need for protamine after vascular reconstructions, Dorman and col-
leagues107 found that routine use was not necessary and that the incidence
of wound hematomas and reoperations was not increased in the group of
patients who did not receive protamine therapy. Protamine itself can
result in a number of adverse effects, particularly in patients with diabetes
mellitus who receive protamine-insulin injections.
Heparin versus Warfarin: Is One Safer?
An experiment to evaluate the bleeding tendencies of various strategies
of periprocedural anticoagulation was performed by Caliendo and col-
leagues.108 White rabbits were divided into 3 groups: rabbits in group I
received warfarin to raise the INR to 2.5 to 3.5, rabbits in group II
received heparin to raise the partial thromboplastin time to 1.5 to 2.0
times control, and rabbits in group III received no anticoagulation. Group
I animals then had warfarin discontinued until the INR was 2.0 to 2.5, and
group II animals had heparin discontinued 6 hours before the proposed
surgical procedure and the prothrombin time returned to baseline. All
animals underwent laparotomy and a controlled 1-cm full-thickness
laceration of the right lobe of the liver that was then repaired. The extent
of bleeding and the time to hemostasis was measured in each animal.
Group I animals were restarted on warfarin on the evening of the
procedure. Group II animals had a continuous intravenous heparin
infusion that was started 6 hours after operation. All animals were killed
24 hours after the operation. The heparin-treated animals had signiﬁcantly
more blood loss than the other 2 groups. There were no differences
between the control and warfarin-treated animals.
The results of this animal study prompted these investigators to evaluate
the same regimens in 40 patients who were undergoing anticoagulation
therapy with mechanical heart valves, atrial ﬁbrillation, an active DVT, a
hypercoagulable state, or a previous peripheral bypass. Operations were
performed in 29 patients (60%) who received warfarin therapy. The other
patients were either converted from warfarin to heparin therapy or
underwent anticoagulation primarily with heparin. There were thrombotic
complications that were related to the cessation of warfarin therapy in 5
120 Curr Probl Surg, February 2003
of 15 patients. Four of the 5 thromboses occurred in patients who were not
converted to heparin therapy because of a perceived low risk of
thromboembolic events. Four of these 5 patients required operation for
the complication. There were 12 patients with bleeding complications, 5
of whom required reoperation. Bleeding was more common in the
patients who received heparin therapy (8/20, 40%) as compared with
patients who received warfarin therapy (4/30, 13%). Bleeding in the
warfarin group correlated with the INR. If the INR was less than 3, 2 of
24 patients had bleeding complications, whereas an INR of more than 3
was associated with bleeding in 2 of 4 patients. Bleeding complications
were also correlated with the age of the patient. No patient younger than
62 years of age had a bleeding problem, whereas 12 of the 34 patients
who were older than 62 years of age had bleeding problems, regardless of
the type of anticoagulation therapy. These data suggest that it may not be
necessary to stop warfarin in the perioperative period for a variety of
operations, provided that the INR is less than 3, whereas periprocedural
heparin is associated with bleeding complications. Caution must be used,
however, in older patients because age appears to be an independent
variable that is associated with bleeding complications.
Topical hemostatic agents are useful particularly in patients who receive
anticoagulation therapy and who undergo operative procedures. Oxidized
cellulose, puriﬁed gelatin, and bovine collagen serve as templates on
which thrombus forms and require the formation of ﬁbrin from circulating
blood. More recently ﬁbrin sealants that are prepared from thrombin and
ﬁbrinogen have been approved for clinical use and have been useful in
vascular procedures109 and splenic trauma.110 Tisseel VH (Baxter Health-
care Corporation, Deerﬁeld, IL) is a 2-component ﬁbrin sealant that
contains thrombin, calcium chloride, and aprotinin (ﬁbrinolysis inhibitor
solution). The reconstituted components are mixed and applied topically.
The thrombin transforms the ﬁbrinogen in the solution into ﬁbrin, and the
aprotinin prevents early degradation of the ﬁbrin. When Tisseel was
compared with historic controls in patients who undergo laparotomy for
blunt or penetrating injuries to the spleen and/or liver, patients who were
treated with Tisseel had signiﬁcantly fewer splenectomies and hepatec-
tomies (1/45 vs 33/56; P .001).111
Agents that are focused on the inhibition of thrombin (such as hirudin and
argatroban) have been approved for use in patients with HIT. Several
Curr Probl Surg, February 2003 121
other agents are currently in trials that target other factors in the clotting
cascade. These include agents that target clotting enzymes that are higher
in the pathways (such as factor Xa, factor IXa, and the factor VIIa/tissue
Direct Thrombin Inhibitors
Fondaparinux is a new class of antithrombotic agent that binds
selectively to antithrombin, which causes it to inhibit factor X. This agent
was found to be more effective than enoxaparin in a prospective
randomized trial that was conducted in 64 centers after elective knee
surgery. A total of 1049 patients were assigned randomly to receive either
enoxaparin (Lovenox) 30 mg twice daily or 2.5 mg of fondaparinox
(Arixtra). Treatment was continued for 5 to 9 days, and primary efﬁcacy
(DVT) was assessed between days 5 and 11. The study demonstrated that
fondaparinox was more effective than enoxaparin in the prevention of
DVT after major elective knee surgery, although major bleeding compli-
cations were more frequent.113
Hirudin is a polypeptide found in the saliva of leeches that inhibits
thrombin and prevents the conversion of ﬁbrinogen to ﬁbrin, activation of
factors V, VIII, and XIII, and thrombin-induced platelet activation. Direct
thrombin inhibitors are active against both circulating and thrombus-
bound thrombi, unlike heparin that is active only against the former.
Several recombinant analogs are available. In a multicenter trial that
compared the efﬁcacy of desirudin (Revasc, CGP 39393, 15 mg,
subcutaneously twice daily) with UFH (5000 IU, subcutaneously three
times daily) in patients who underwent total hip arthroplasty, the former
was found to be superior in the prevention of proximal DVT by 79%
without any increase in bleeding complications.114 Anticoagulant therapy
was continued for 8 to 11 days. During a 6-week period of follow-up,
there were documented pulmonary emboli in 4 patients, each of whom
received heparin therapy. The anticoagulant effects of argatroban have
also been compared with UFH. Both produced dose-related anticoagulant
effects (increased activated clotting time, APTT). Argatroban had a 4-fold
shorter dissipation (half-life, 18 to 41 minutes vs 23 to 134 minutes for
Ancrod is a thrombin-like enzyme that is found in the venom of the
Malayan pit viper. The anticoagulant effect of ancrod is the result of a
122 Curr Probl Surg, February 2003
decrease in the ﬁbrinogen levels, the production of ﬁbrin monomers, and
the subsequent stimulation of tissue plasminogen activator. The level of
ﬁbrinogen must be closely monitored, however, lest bleeding complica-
Oral heparins, for many reasons, are the ideal long-term anticoagulant.
Several delivery systems have been evaluated and are in various stages of
the FDA approval process. Baughman and colleagues116 report the results
of a RCT in humans using sodium N-[8(-2-hydroxybenzoyl)amino]
caprylate and LMWH. Measurements of the anticoagulant effect included
APTT, tissue factor pathway inhibitor, and factors I, Ia, and Xa 1 hour
after dosing in normal volunteers. Heparin that was administered orally
with sodium N-[8(-2-hydroxybenzoyl)amino] caprylate produced signif-
icant elevations of all the clotting parameters and was well tolerated.
1. Boissel J-P, Haugh MC. The iceberg phenomenon and publication bias: The
Editors’ fault? Clinical Trials and Meta-analysis 1993;28:309-15 [editorial].
2. Guyatt G, Schunemann HJ, Cook DJ, Jaeschke R, Pauker SG, Bucher H. Grades of
recommendation for antithrombotic agents: Sixth ACCP Consensus Conference on
Antithrombotic Therapy. Chest 2001;119(suppl):1-3.
3. Bick RL, Kaplan H. Syndromes of thrombosis and hypercoagulability: congenital
and acquired causes of thrombosis. Med Clin North Am 1998;82:409-58.
4. Stassen JM, Nystrom A. A historical review of hemostasis, thrombosis, and
antithrombotic therapy. Ann Plast Surg 1997;39:317-29.
5. Levy PJ, Gonzalez FM, Rush DS, Haynes JL. Hypercoagulable states as an
evolving risk for spontaneous venous and arterial thrombosis. J Am Coll Surg
6. Grifﬁn JH, Evatt B, Wideman C, Fernandez JA. Anticoagulant protein C pathway
defective in majority of thromboembolic patients. Blood 1993;82:1989-93.
7. Silver D, Vouyouka A. The caput medusae of hypercoagulability. J Vasc Surg
8. Ouriel K, Green RM, DeWeese JA, Cimino C. Activated protein C resistance:
prevalence and implications in peripheral vascular disease. J Vasc Surg 1996;23:
9. Ferraresi P, Marchetti G, Legnani C, Cavallari E, Castoldi E, Mascoli F. The
heterozygous 20210 G/A prothrombin genotype is associated with early venous
thrombosis in inherited thrombophilias and is not increased in frequency in arterial
disease. Arterioscler Thromb Vasc Biol 1997;17:2418-22.
10. Lam EY, Taylor LM, Landry GL, Porter JM, Moneta GL. Relationship of
antiphospholipid antibodies and progression of lower extremity arterial occlusive
disease after lower extremity bypass operations. J Vasc Surg 2001;33:976-82.
11. Tengborn L, Bergvist D. Surgery in patients with congenital antithrombin III
deﬁciency. Acta Chir Scand 1988;154:179-83.
Curr Probl Surg, February 2003 123
12. Jackson MR, Olsen SB, Gomez ER, Alving BM. Use of antithrombin III
concentrates to correct antithrombin III deﬁciency during vascular surgery. J Vasc
13. Eldrup-Jorgensen J, Flanigan DP, Brace L, Sawchuck AP, Mulder SG, Anderson
CP. Hypercoagulable states and lower limb ischemia in young adults. J Vasc Surg
14. Kang SS, Wong PW, Malinow MR. Hyperhomocystinemia as a risk factor for
occlusive vascular disease. Annu Rev Nutr 1992;12:279-98.
15. Bergqvist D. Prophylaxis against postoperative venous thromboembolism: a survey
of surveys. Thromb Haemorrh 1990;2:69-73.
16. Caprini JA, Arcelus JI, Hoffman K, et al. Prevention of venous thromboembolism
in North America: results of a survey among general surgeons. J Vasc Surg
17. Haas S. European consensus statement on the prevention of venous
thromboembolism: European Consensus Conference. Windsor UK, November
1991. Blood Coagul Fibrinolysis 1993;4(suppl):S5-10.
18. Rohrer MJ, Cutler BS, MacDougall E, Herrmann JB, Anderson FA, Wheeler HB.
A prospective study of the incidence of deep venous thrombosis in hospitalized
children. J Vasc Surg 1996;24:46-50.
19. Agnelli G. Venous thromboembolism and cancer: a two-way clinical association.
Thromb Haemostas 1997;78:117-20.
20. Aderka D, Brown A, Zelikovski A, Pinkhas J. Idiopathic deep vein thrombosis in
an apparently healthy patient as a premonitory sign of occult cancer. Cancer
21. Sevitt S, Gallagher N. Venous thrombosis and pulmonary embolism: a clinicopath-
ological study in injured and burned patients. Br J Surg 1961;48:479-89.
22. Stone EA, Stewart GJ. Architecture and structure of canine veins with special
reference to conﬂuences. Anat Rec 1988;222:154-63.
23. Vanek VW. Meta-analysis of effectiveness of intermittent pneumatic compression
devices with a comparison of thigh-high to knee-high sleeves. Am Surg 1998;64:
24. Comerota AJ, Chouhan V, Harada RN, Sun L, Hosking J, Veermansunemi R. The
ﬁbrinolytic effects of intermittent pneumatic compression. Ann Surg 1997;226:306-
25. Chouhan VD, Comerota AJ, Sun L, Harada R, Gaughan JP, Rao AK. Inhibition of
tissue factor pathway during intermittent pneumatic compression: a possible
mechanism for antithrombotic effect. Arterioscler Thromb Vasc Biol 1999;19:
26. Geerts WH, Heit JA, Clagett GP, Pineo F, Colwell CW, Anderson FA. Prevention
of venous thromboembolism: Sixth ACCP Consensus Conference on Antithrom-
botic Therapy. Chest 2001;119(suppl):12-20.
27. Geerts WH, Code KI, Jay RM, Chen E, Szalai JP. A prospective study of venous
thromboembolism after major trauma. N Engl J Med 1994;331:1601-6.
28. Greets WH, Jay RM, Code KI, Chen E, Szalai EA, Hamilton PA. A comparison of
low-dose heparin with low molecular weight heparin as prophylaxis against venous
thromboembolism after major trauma. N Engl J Med 1996;335:701-7.
29. Kontny F, Dale J, Abildgaard U, Pedersen TR. Randomized trial of low molecular
weight heparin (Dalteparin) in prevention of left ventricular thrombus formation
124 Curr Probl Surg, February 2003
and arterial embolism after acute myocardial infarction: The Fragmin in Acute
Myocardial Infarction (FRAMI) Study. J Am Coll Cardiol 1997;30:962-7.
30. The TIMI 11A Investigators. Dose-ranging trial of enoxaparin for unstable angina:
results of TIMI 11A: the Thrombolysis in Myocardial Infarction (TIMI) 11A Trial
Investigators. J Am Coll Cardiol 1997;29:1474-82.
31. Proctor MC. Indications for ﬁlter placement. Semin Vasc Surg 2000;13:194-8.
32. Wille-Jorgensen, P.; Rasmussen, MS.; Anderson, BR.; Borly, L. Heparins and
mechanical methods for thromboprophylaxis in colorectal surgery. Cochrane
Database Syst Rev 2001;3:CD001217.
33. Krauth D, Holden A, Knapic N, Liepman M, Ansell J. Safety and efﬁcacy of
long-term oral anticoagulation in cancer patients. Cancer 1987;59:983-5.
34. Calligaro K, Bergen WS, Haut MJ, Savarese RP, DeLaurentis DA. Thromboem-
bolic complications in patients with advanced cancer: anticoagulation versus
Greenﬁeld ﬁlter placement. Ann Vasc Surg 1991;5:186-9.
35. Bergqvist D, Agnelli G, Cohen AT, et al. for the ENOXSCAN II Investigators.
N Engl J Med 2002;346:975-80.
36. Dahl OE, Andreassen G, Aspelin T. Prolonged thromboprophylaxis following hip
replacement surgery- results of a double-blind, prospective, randomized, placebo-
controlled study with dalteparin (Fragmin). Thromb Haemost 1997;77:26-31.
37. Lausen I, Jensen R, Jorgensen LN. Incidence and prevention of deep venous
thrombosis occurring late after general surgery: randomized controlled study of
prolonged thromboprophylaxis. Eur J Surg 1998;164:657-63.
38. Serasin FP, Eckman MH. Management and prevention of thromboembolic events
in patients with cancer-related hypercoagulable states: a risky business. J Gen
Intern Med 1993;8:476-86.
39. Passman MA, Farber MA, Marston WA, et al. Prospective screening for postop-
erative deep venous thrombosis in patients undergoing infrainguinal revasculariza-
tion. J Vasc Surg 2000;32:669-75.
40. Yeager RA, Moneta GL, Edwards JM, Taylor LM, McConnell DB, Porter JM.
Deep vein thrombosis associated with lower extremity amputation. J Vasc Surg
41. Hull RD, Pineo GF, Stein PD, et al. Extended out-of-hospital low-molecular-weight
heparin prophylaxis against deep venous thrombosis in patients after elective hip
arthroplasty: a systematic review. Ann Intern Med 2001;135:858-69.
42. Schulman S, Granqvist S, Holmstrom M. The duration of oral anticoagulant therapy
after a second episode of venous thromboembolism. N Engl J Med 1997;336:393-8.
43. Kearon C, Gent M, Hirsh J, et al. Comparison of three months of anticoagulation
with extended anticoagulation for a ﬁrst episode of idiopathic venous thromboem-
bolism. N Engl J Med 1999;340:901-7.
44. Mismetti P, Laporte S, Darmon Y, Buchmuller A, Decousus H. Meta-analysis of
low molecular weight heparin in the prevention of venous thromboembolism in
general surgery. Br J Surg 2001;88:913-30.
45. Collins R, MacMahon S, Flather M, Baigent C, Remvig L, Mortenson S. Clinical
effects of anticoagulant therapy in suspected myocardial infarction: systemic
overview of randomized trials. Br Med J 1996;313:652-9.
46. Bergquist D, Bonnar J, Clarke-Pearson D, Collins R, Comerota A, Eckloff B.
Prevention of venous thromboembolism: international consensus statement. Int
Curr Probl Surg, February 2003 125
47. Kakkar VV. Efﬁcacy and safety of clivarin and other LMWHs in general surgery:
a meta-analysis. Blood Coagul Fibrinolysis 1993;4(suppl I):S23-7.
48. Kcoh A, Bouges S, Ziegler S, Dinkel H, Daures JP, Victor N. Low molecular
weight heparin and unfractionated heparin in thrombosis prophylaxis after major
surgical intervention: update of previous meta-analyses. Br J Surg 1997;84:750-9.
49. Hirsch J, Dalen JE, Anderson DR, et al. Oral anticoagulants: mechanism of action,
clinical effectiveness, and optimal therapeutic range: Sixth ACCP Consensus
Conference on Antithrombotic Therapy. Chest 2001;119(suppl):37-42.
50. Gullov AL, Koefoed BG, Petersen P. Bleeding complications to long-term oral
anticoagulant therapy. J Thromb Thrombolysis 1994;1:17-25.
51. Ezekowitz MD, Radford JK, Rickles FR, Redmond N. Initiating and maintaining
patients on warfarin anticoagulation: the importance of monitoring. Cardiovasc
Pharmacol Ther 1999;4:3-8.
52. Antiplatelet Trialists’ Collaboration. Collaborative overview of randomized trials
of antiplatelet therapy-1: prevention of death, myocardial infarction, and stroke by
prolonged antiplatelet therapy in various categories of patients. Br Med J 1994;
53. Turpie AGG, Gent M, Laupacis A. A comparison of aspirin with placebo in patients
treated with warfarin after heart valve replacement. N Engl J Med 1993;329:524-9.
54. Meade TW, Mellows S, Brozovic M. Haemostatic function and ischaemic heart
disease: principal results of the Northwick Park Heart Study. Lancet 1986;2:533-7.
55. The Medical Research Council’s General Practice Research Framework. Throm-
bosis prevention trial: randomized trial of low-intensity oral anticoagulation with
warfarin and low-dose aspirin in the primary prevention of ischaemic heart disease
in men at increased risk. Lancet 1998;351:233-41.
56. Ginsberg JS, Greer IA, Hirsch J. Pregnant patients: Sixth ACCP Consensus
Conference on Antithrombotic Therapy. Chest 2001;119(suppl):26-9.
57. Writing Group for the Women’s Health Initiative Investigators. Risks and beneﬁts
of estrogen plus progestin in healthy postmenopausal women: principal results from
the women’s health initiative randomized controlled trial. JAMA 2002;288:321-33.
58. Schafer AI. The hypercoagulable states. Ann Intern Med 1985;102:814-28.
59. Bennet L, Odeberg H. Resistance to activated protein C, highly prevalent amongst
users of oral contraceptives with venous thromboembolism. J Intern Med 1998;
60. Sipes SL, Weiner CP. Venous thromboembolic disease in pregnancy. Semin
61. Dizon-Townson DS, Nelson LM, Jang H, Varner MW, Ward K. The incidence of
the factor V Leiden mutation in an obstetric population and its relationship to deep
vein thrombosis. Am J Obstet Gynecol 1997;176:883-6.
62. Patrono C. Drug therapy: aspirin as an antiplatelet agent. N Engl J Med
63. Cheeseboro JH, Fuster V, Elveback LR. Trial of combined warfarin plus dipyrid-
amole or aspirin therapy in prosthetic heart valve replacement: danger of aspirin
compared with dipyridamole. Am J Cardiol 1983;51:1537-45.
64. Antiplatelet Trial Collaboration. Collaborative overview of randomized trials of
antiplatelet therapy: I. Prevention of death, myocardial infarction, and stroke by
prolonged antiplatelet therapy in various categories of patients. Br J Med 1994;
126 Curr Probl Surg, February 2003
65. Steering Committee of the Physicians’ Health Study Research Group. Final Report
on the aspirin component of the ongoing Physicians’ Health Study. N Engl J Med
66. CAPRIE Steering Committee. A randomized, blinded trial of clopidogrel versus
aspirin in patients at risk of ischemic events (CAPRIE). Lancet 1996;348:1329-39.
67. Payne DA, Hayes PD, Jones CI, Belham P, Naylor AR, Goodhall AH. Combined
therapy with clopidogrel and aspirin signiﬁcantly increases the bleeding time
through a synergistic antiplatelet action. J Vasc Surg 2002;35:1204-9.
68. CURE trial investigators. Effects of clopidogrel in addition to aspirin in patients
with acute coronary syndromes without ST elevation. N Engl J Med 2001;345:
69. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Bene-
ﬁcial effect of carotid endarterectomy in symptomatic patients with high-grade
carotid stenosis. N Engl J Med 1991;325:445-53.
70. European Carotid Surgery Trialists’ Collaborative Group. MRC European Carotid
Surgery Trial: interim results for symptomatic patients with severe (70 –99%) or
with mild (0 –29%) carotid stenosis. Lancet 1991;337:1235-43.
71. Stratton JR, Zierler RE, Kazmers A. Platelet deposition at carotid endarterectomy
sites in humans. Stroke 1987;18:722-7.
72. Gaunt ME, Martin PJ, Smith JL, et al. Clinical relevance of intraoperative
embolization detected by transcranial Doppler ultrasonography during carotid
endarterectomy: a prospective study of 100 patients. Br J Surg 1994;81:1435-9.
73. Lennard N, Smith J, Dumville J, et al. Prevention of postoperative thrombotic
stroke after carotid endarterectomy: the role of transcranial Doppler ultrasound. J
Vasc Surg 1997;26:579-84.
74. The Abciximab in Ischemic Stroke Investigators. Abciximab in acute ischemic
stroke: a randomized, double-blind, placebo- controlled, dose-escalation study.
75. Schneiderman J, Morag B, Gerniak A, et al. Abciximab in carotid stenting for
postsurgical carotid restenosis: intermediate results. J Endovasc Ther 2000;7:263-
76. Calaitges JG, Liem TK, Spadone D, Nichols WK, Silver D. The role of
heparin-associated antiplatelet antibodies in the outcome of arterial reconstruction.
J Vasc Surg 1999;29:779-86.
77. Ortel TL, Gockerman JP, Califf RM, McCann RL, O’Connor CM, Metzler DM.
Parenteral anticoagulation with the heparinoid Lomoparen (Prg 10172) for extra-
corporeal procedures in patients with heparin induced thrombocytopenia and
thrombosis. Thromb Haemost 1992;67:292-6.
78. DeValk HW, Banga JD, Wester JW, Brouwer CB, van Hessen MW, Meuwissen
OJ. Comparing subcutaneous danaproid with intravenous unfractionated heparin
for the treatment of venous thromboembolism: a randomized controlled trial. Ann
Intern Med 1995;123:1-9.
79. Henny CP, Ten Cate H, Ten Cate TW, Surachno S, Van Bronswijk H, Wilmink JM.
Use of a new heparinoid as anticoagulant during acute haemodialysis of patients
with bleeding complications. Lancet 1983;1:890-3.
80. Tardy/Poncet B, Mahul P, Beraud AM, Favre JP, Tardy B, Guyotat D. Failure of
Organon therapy in a patient with a previous heparin-induced thrombocytopenia
syndrome. Br J Haemotol 1995;90:969-70.
Curr Probl Surg, February 2003 127
81. Magnani HN, Beijering RJ, Ten Cate JW, Chong BH. Organon anticoagulation for
cardiopulmonary bypass in patients with heparin induced thrombocytopenia. In:
Pifarre R, editor. New anticoagulants for the cardiovascular patient. Philadelphia:
Hanley & Belfus; 1997. p. 487-500.
82. Gitlin SD, Deeb M, Yann C, Schmater AH. Intraoperative monitoring of danaproid
sodium anticoagulation during cardiovascular operations. J Vasc Surg 1998;27:
83. Greinacher A, Volpel H, Janssens U, et al. Recombinant hirudin (lepirudin)
provides safe and effective anticoagulation in patients with heparin-induced
thrombocytopenia: a prospective study. Circulation 1999;99:73-80.
84. Jackson MR, Clagett GP. Peripheral arterial occlusive disease: Sixth ACCP
Consensus Conference on Antithrombotic Therapy. Chest 2001;119(suppl):20-1.
85. Antiplatelet Trialists’ Collaboration. Collaborative overview of randomized trials
of antiplatelet therapy: II. Maintenance of vascular graft or arterial patency by
antiplatelet therapy. Br Med J 1994;308:159-68.
86. Kretschmer G, Herbst F, Prager M, Sautner T, Wenzl E, Berlakovich GA. A decade
of oral anticoagulant treatment to maintain autologous vein grafts for femoropop-
liteal atherosclerosis. Arch Surg 1992;127:1112-5.
87. Tangelder MJD, Lawson JA, Algra A, Eikelboom BC. Systematic review of
randomized control trials of aspirin and oral anticoagulants in the prevention of
graft occlusion and ischemic events after infrainguinal bypass surgery. J Vasc Surg
88. and members of the Department of Veteran Affairs Cooperative Study #362,
Johnson WC, Williford WO. Beneﬁts, morbidity, and mortality associated with
long-term administration of oral anticoagulant therapy to patients with peripheral
arterial bypass procedures: a prospective randomized study. J Vasc Surg 2002;35:
89. Dutch Bypass Oral Anticoagulants or Aspirin (BOA) Study Group. Efﬁcacy of oral
anticoagulants compared with aspirin after infrainguinal bypass surgery: a random-
ized trial. Lancet 2000;355:346-51.
90. Sarac TM, Huber TS, Back MR, Ozaki CK. Warfarin improves the outcome of
infrainguinal vein bypasses at high risk for failure. J Vasc Surg 1998;28:446-57.
91. Jackson MR, Johnson WC, Williford WO, Valentine RJ, Clagett CP. The effect of
anticoagulation therapy and graft selection on the ischemic consequences of
femoropopliteal bypass graft occlusion: results from a multicenter trial. J Vasc Surg
92. McMillan WD, McCarthy WJ, Lin SJ, Pearce WH, Yao JS. Perioperative low
molecular weight heparin for infrageniculate bypass. J Vasc Surg 1997;25:796-801.
93. Palmaz JC, Garcia OJ, Kapp DT. Balloon-expandable intraarterial stents: effect of
anticoagulation on thrombus formation. Circulation 1987;76:IV-45A.
94. DeCaterina R, Sicari R, Bernini W. Beneﬁt/risk proﬁle of combined antiplatelet
therapy with ticlopidine and aspirin. Thromb Haemost 1991;65:504-10.
95. Pepine CJ, Holmes DR. ACC expert consensus document: coronary artery stents.
J Am Coll Cardiol 1996;28:782-94.
96. Tcheng JE, Kong DF. Vale, warfarin: a stentorian farewell. Am Heart J 1999;138:
97. Caman CP. Effectiveness of clopidogrel versus aspirin in preventing acute myocar-
128 Curr Probl Surg, February 2003
dial infection in patients with symptomatic athorothrombosis. (CAPRIE trial) Am J
98. Zidar JP. Low-molecular weight heparins in coronary stenting (the ENTICES
Trial): enoxaparin and ticlopidine after elective stenting. Am J Cardiol 1998;82:
99. Lowson SM, Hanson EW. Anticoagulation and elective surgery. N Engl J Med
1997;337:939 [letter to the editor].
100. Tinker JH, Tarhan S. Discontinuing anticoagulant therapy in surgical patients with
cardiac valve prostheses. JAMA 1978;238:729-38.
101. Francis CW, Marder VJ, McCollister E, Yaukoolbodi S. Two step warfarin therapy:
prevention of postoperative venous thrombosis without excessive bleeding. JAMA
102. Dietrich W, Dilthey G, Spannagl M, Richter JA. Warfarin pretreatment does not
lead to increased bleeding tendency during cardiac surgery. J Cardiothorac Vasc
103. DeSwart CAM, Nijmeijer B, Roelofs JMM, Sixma JJ. Kinetics of intravenously
administered heparin in normal humans. Blood 1982;60:1251-8.
104. Kroneman H, Eikelboom BC, Knot EAR, et al. Pharmacokinetics of low-molecular
weight heparin and unfractionated heparin during elective aortobifemoral grafting.
J Vasc Surg 1991;14:208-14.
105. Hirsch J, Ofosu FA, Levine M. The development of low molecular heparins for
clinical use. In: Verstraete M, Vermylen J, Lijnen HR, Arnout J, editors.
Thrombosis and haemostasis. Belgium, Leuven: Leuven University Press; 1987. p.
106. Casteneda AR. Must heparin be neutralized following open-heart operations?
J Thorac Cardiovasc Surg 1966;52:716-9.
107. Dorman BH, Elliott BM, Spinale FG, et al. Protamine use during peripheral
vascular surgery: a prospective randomized clinical trial. J Vasc Surg 1995;22:248-
108. Caliendo FJ, Halpern VJ, Marini CP, et al. Warfarin anticoagulation in the
perioperative period. Is it safe? Ann Vasc Surg 1999;13:11-6.
109. Matthew TL, Spotnitz WD, Kron I, Daniel TM, Tribble CG, Nolan SP. Four years’
experience with ﬁbrin sealant in thoracic and cardiovascular surgery. Ann Thorac
110. Ochsner MG, Maniscalco T, Champion HR. Fibrin glue as a hemostatic agent in
hepatic and splenic trauma. J Trauma 1990;97:194-220.
111. Shoemaker W, Kram HB, Fleming AW, editors. Use of ﬁbrin sealant to decrease
mortality and morbidity from traumatic injuries to the liver and spleen: internal
study no. 021. Glendale (CA): Baxter Healthcare; 1991.
112. Weitz J, Hirsch J. New anticoagulant drugs: Sixth ACCP Consensus Conference on
Antithrombotic Therapy. Chest 2001;119(suppl):42-3.
113. Lormeau JC, Heralt JP. The effect of the synthetic pentasaccharide SR 90197/
ORG31540 on thrombin generation ex vivo is uniquely due to ATIII-mediated
neutralization of factor Xa. Thromb Haemost 1995;74:1474-7.
114. Eriksson BI, Ekman S, Lindbratt S, et al. Prevention of thromboembolism with use
of desirudin (Revasc) with that of unfractionated heparin in patients having a total
hip replacement. J Bone Joint Surg 1997;79:326-33.
115. Swan SK, Peter JV, Lambrecht LJ, Hursting MJ. Comparison of anticoagulant
Curr Probl Surg, February 2003 129
effects and safety of argatroban and heparin in healthy subjects. Pharmacotherapy
116. Baughman RA, Kapoor SC, Agarwal RK, Kisicki J, Catella-Lawson F, Fitzgerald
GA. Oral delivery of anticoagulant doses of heparin: a randomized, double-blind,
controlled studying humans. Circulation 1998;98:1610-5.
117. Ansell, J.E.; Becker, R.C. Guidelines for the initiation, monitoring and clinical use
of antithrombotic therapy in venous and arterial disease. Anticoagulation Forum
118. Patrono C, Coller BS, Dalen JE, et al. Platelet-active drugs: the relationships among
dose, effectiveness, and side effects: Sixth ACCP Consensus Conference on
Antithrombotic Therapy. Chest 2001;119(Suppl):44.
130 Curr Probl Surg, February 2003