Review Article

Pulmonary Embolism in
Orthopaedic Patients: Diagnosis
and Management
Abstract
Paul Tornetta, MD
Yelena Bogdan, MD

Orthopaedic patients are at particularly high risk for pulmonary

embolism. There has been a trend recently toward overdiagnosis
of pulmonary embolism; thus, evaluation of the nature of a clinically
relevant pulmonary embolism is needed, as is assessment of the
timing, risks, and outcomes of therapeutic anticoagulation in
surgical patients. Recent literature shows the incidence of
pulmonary embolism to be increasing without a corresponding
increase in mortality, suggesting that not all emboli may be
clinically relevant and that increasingly sensitive tests may be
picking up small emboli. The size and location of a clot or clots
may matter when deciding on management. A risk-benefit
evaluation can assist in deciding treatment.

From Boston University Medical

Center, Boston, MA.
Dr. Tornetta or an immediate family
member serves as a paid consultant
to or is an employee of, and has
received royalties from, Smith &
Nephew, and serves as a board
member, owner, officer, or
committee member of the American
Orthopaedic Association and the
Orthopaedic Trauma Association.
Neither Dr. Bogdan nor any
immediate family member has
received anything of value from or
has stock or stock options held in a
commercial company or institution
related directly or indirectly to the
subject of this article.
J Am Acad Orthop Surg 2012;20:
586-595
http://dx.doi.org/10.5435/
JAAOS-20-09-586
Copyright 2012 by the American
Academy of Orthopaedic Surgeons.

586

rthopaedic patients are at high

risk for both pulmonary embolism (PE) and deep vein thrombosis
(DVT). These disorders have been
well studied in the arthroplasty population; an American Academy of
Orthopaedic Surgeons guideline is
available for public access.1 However, this guideline does not apply
to the nonarthroplasty orthopaedic
population and addresses prophylaxis rather than management. Here,
we discuss the diagnosis and management of PE in general orthopaedics, particularly trauma.
Symptomatic PE occurs in 2% to
10% of pelvic trauma patients, and
fatal PE, in 0.5% to 2%.2 Deep vein
thrombi appear to occur with greater
incidence in proximal rather than
distal lower extremity fractures, although the clinical significance of
this finding is unclear.3 The most
common signs of PE are tachycardia,
low oxygen saturation, and shortness of breath; however, the clinical
presentation of PE is notoriously un-

reliable, and many embolisms are silent. In a review of 695 patients, Kim
et al4 found a 27.8% rate of positive
CT results for PE in postoperative
orthopaedic patients. In that study, a
prior history of thromboembolic disease was the only significant predictor of a positive scan; a high body
mass index was a marginal predictor.
Thus, a combination of patient characteristics and symptoms may predict PE.
The magnitude of PE is a spectrum
ranging from central large clots to
tiny subsegmental clots. The clinical
relevance of this spectrum of condition is also quite varied. It is well
known that untreated emboli can
lead to death; however, many have
no clinical sequelae. How, then,
should management such as aggressive anticoagulation, which has its
own risks and costs, be chosen? As
methods of embolus detection become more sensitive for smaller occlusions, the clinical significance of
these small emboli is called into

Journal of the American Academy of Orthopaedic Surgeons

Paul Tornetta, MD, and Yelena Bogdan, MD

question. Is the management we are

instituting rational, or are we overtreating some percentage of patients?

Diagnostic Methods and

Time Trends
PE is diagnosed in several ways.
Clinically, many patients do not
present with the classic symptoms of
pleuritic chest pain and shortness of
breath. Often, unexplained tachycardia in a postoperative patient, combined with low oxygen saturation, is
all that is needed to prompt a search
for PE. D-dimer, a blood test that
measures fibrin degradation, may be
elevated in patients with PE, but it is
also elevated in postoperative patients generally and is thus unreliable
as a standalone diagnostic test. The
Wells score has undergone numerous
iterations and remains one of the
most popular clinical prediction
models for PE diagnosis.5 The Wells
score assigns points based on patient
characteristics and, when combined
with the D-dimer level, can be a very
useful predictive tool (Tables 1 and
2). However, this score has been developed primarily from a nonsurgical
patient population and thus may not
apply to orthopaedic injury patients.
With regard to imaging studies for
the diagnosis of PE, pulmonary angiography is the benchmark, but it is
expensive and invasive. The two ma-

jor and most common imaging diagnostic modalities are CT pulmonary

angiogram (CTPA) and ventilation
perfusion (VQ) scanning. The advantages of CTPA are direct visualization of the clot, high sensitivity, identification of alternative pathology,
and rapid testing time. The advantages of VQ scanning are low radiation exposure and low cost. A randomized trial comparing the two
methods in 1,417 patients showed
that CTPA had a higher positive predictive value than did VQ scanning
but that the two methods were similar in ruling out PEs.6 The negative
predictive value of CTPA appears to
be adequate; a meta-analysis of 23
studies showed that the 3-month rate
of recurrent, symptomatic PE in
4,657 patients with high clinical suspicion for PE and a negative CTPA
was 1.4%.7 One study in patients
with moderate to high probability
for PE and/or elevated D-dimer levels
gave the negative predictive value of
CTPA as 99.5%.8 CTPA is now the
most common diagnostic method for
PE, but it is used as part of an overall approach that includes clinical
prediction, ultrasonography, and
other modalities.
General time trends from the past
decade show that the incidence of PE
is increasing. In a sample of more
than 1 million patients from the Nationwide Inpatient Sample database,

the number of PE diagnoses increased from 126,546 cases in 1998

to 229,637 in 2005, but fatal PE
rates dropped from 12.3% to 8.2%.9
The decreases in mortality rate may
suggest that management of PE beTable 1
Wells Score for Predicting
Pulmonary Embolisma
No. of
Points
Assigned

Variable
Clinical signs and symptoms of DVT (ie, minimum leg swelling, pain
on palpation of deep
veins)
An alternative diagnosis
is less likely than PE
Heart rate >100 BPM
Immobilization or surgery
in previous 4 weeks
Previous DVT/PE
Hemoptysis
Malignancy (on treatment, treated in last 6
months, or palliative)

3
1.5
1.5
1.5
1
1

DVT = deep vein thrombosis,

PE = pulmonary embolism
a
The number of points is added. The
higher the score, the greater the likelihood
of clinical probability. (See Table 2 for percentages.) >6 = high, 2 to 6 = moderate,
<2 = low probability
Adapted with permission from Wells PS,
Anderson DR, Rodger M, et al: Derivation
of a simple clinical model to categorize
patients probability of pulmonary embolism: Increasing the models utility with the
SimpliRED D-dimer. J Thromb Haemost
2000;83(3):416-420.

Table 2
Pulmonary Embolism Rates in a Validation Group of Patients Using the Wells Score Model and D-dimer
Pulmonary Embolism Rate (%)

Wells Score
<2
26
>6

Normal D-dimer
(95% CI)

Elevated D-dimer
(95% CI)

Overall
(95% CI)

2.7 (0.39)
2.9 (0.410)
20 (5.171.6)

0 (013.2)
37.3 (2550.9)
60 (32.383.7)

2 (0.27.1)
18.8 (12.426.6)
50 (27.272.8)

CI = confidence interval
Adapted with permission from Wells PS, Anderson DR, Rodger M, et al: Derivation of a simple clinical model to categorize patients probability
of pulmonary embolism: Increasing the models utility with the SimpliRED D-dimer. J Thromb Haemost 2000;83(3):416-420.

September 2012, Vol 20, No 9

587

Pulmonary Embolism in Orthopaedic Patients: Diagnosis and Management

Figure 1

Clinically Relevant
Pulmonary Embolism

Line graph illustrating deaths from pulmonary embolism (PE), 1992 to 2006.
PEDeaths = number of deaths resulting from PE, TotPE = total number of
cases of PE. (Reproduced with permission from Burge AJ, Freeman KD,
Klapper PJ, Haramati LB: Increased diagnosis of pulmonary embolism
without a corresponding decline in mortality during the CT era. Clin Radiol
2008;63[4]:381-386.)

came more effective over time; however, better treatment modalities cannot also explain the increased PE
incidence. The other explanation is
that our PE tests are overly sensitive.
This would result in increased diagnosis of small emboli, which may
have fewer clinical sequelae. In other
words, the diagnosis exists but may
not be clinically significant. This
trend has been termed overdiagnosis
by Wiener et al.10
Several studies have focused on
overdiagnosis of PE. An analysis of
27 million patients over a 10-year
period (1994 to 2004) showed a
doubling of PE diagnoses without a
change in mortality rate,11 which
suggested a strong association of the
increased use of CT to explain these
changes (Figure 1). This trend leads
to the issue of appropriate selection
of patients for these imaging studies,
which are not without adverse effects
(eg, radiation exposure equivalent of

588

100 to 400 chest radiographs). Wiener et al10 looked at trends before

and after the advent of CTPA and reported an 81% increased incidence
of PE after CTPA with little change
in mortality, as well as a 71% increase in complications from anticoagulation. This suggests that overdiagnosis occurs and is prevalent; in
addition, the lack of change in mortality implies that previously missed
small emboli picked up by CTPA
may not be clinically important. Furthermore, overdiagnosis is potentially harmful to patients when they
are anticoagulated for clinically insignificant emboli. In one study, a
comparison of clinical indicators of
PE compared with results of CTPA
showed that 25% of positive CTPA
findings were not associated with
high clinical probability of PE;
awareness of this fact could assist in
patients not receiving unnecessary
anticoagulation.12

Symptomatic PE can present in different ways. One study describes PE

according to three syndromes that
increase in severity, from pulmonary
infarction syndrome (least severe) to
isolated dyspnea to circulatory collapse (most severe). Patients with less
severe syndromes are more likely to
have a normal electrocardiogram
and PaO2 >80 mm Hg but are less
likely to have tachypnea, dyspnea, or
a high-probability VQ scan.13 This
suggests that PE can be stratified in
terms of clinical severity. It follows
that some PEs are completely asymptomatic and found incidentally. The
prevalence of incidental PE is approximately 2.6% according to one
meta-analysis of patients undergoing
CT for reasons other than PE diagnosis (eg, evaluation of metastatic
disease).14 Incidental PE also is more
common in hospitalized patients and
those with cancer. Incidental PE is
more likely to occur in lobar and
segmental arteries. The clinical history of asymptomatic PE has been
evaluated by several studies; however, the studies had small patient
sample sizes, limited clinical history
of the patient populations, or limited
follow-up time. It is clear that more
evidence is needed to assess the potential clinical relevance of asymptomatic PE.
Several studies have investigated
the size and location of the embolism
(central versus segmental or subsegmental) (Figure 2) in relation to clinical significance. Although some authors have found no correlation
between signs and symptoms and
clot size or location,15 others agree
that clot characteristics make a difference. In a 5-year study of postoperative cancer patients, Auer et al16
found a 7.8% average annual increased incidence in segmental and

Journal of the American Academy of Orthopaedic Surgeons

Paul Tornetta, MD, and Yelena Bogdan, MD

Figure 2

Axial CT pulmonary angiograms of the thorax demonstrating types of pulmonary embolism (PE). A, Right main central
PE (arrow). B, Segmental PE in right lower lobe (arrow). C, Subsegmental PE (arrow). D, Very small dot PE (arrow).
Note the progressive decrease in size from main to subsegmental and dot PEs, which may not be clinically relevant.
(Images courtesy of Akira Murakami, MD, Boston University Medical Center Radiology Department, Boston, MA.)

subsegmental PE, no change in overall incidence of central PE, and no

change in overall incidence of fatal
PE. Central PE was more severe than
peripheral and was associated with
hypoxia, tachycardia, a higher Wells
score, a greater number of symptoms, higher 30-day mortality
(33.3%, central PE, versus 5%, peSeptember 2012, Vol 20, No 9

ripheral PE), and a higher rate of admission to the intensive care unit. Le
Gal et al17 also reported that subsegmental PE presented with less dyspnea and less chance of being characterized as having a high clinical
probability of PE.
Small emboli are being detected
with increased frequency, but this

does not appear to influence PE

outcome. Carrier et al18 found an
increased detection rate of subsegmental PE in multidetector
CTPA compared with single-detector
CTPA. However, the 3-month risk of
venous thromboembolism (VTE) in
untreated patients with suspected PE
and negative CTPA was similar be-

589

Pulmonary Embolism in Orthopaedic Patients: Diagnosis and Management

Table 3
Characteristics of Patients With Pulmonary Embolus of Different Sizesa
Factor

Subsegmental (n = 22)

Segmental (n = 67)

Central (n = 245)

14 (63.6%)b
4 (18.2%)
14 (63.6%)
9 (40.9%)
15 (68.2%)
None
None

36 (53.7%)
16 (23.9%)
28 (41.8%)
41 (61.2%)
54 (80.6%)
None
3 (4.5%)

45 (18.4%)
117 (47.8%)
44 (18.0%)
177 (72.2%)
229 (93.5%)
6 (2.4%)
7 (2.9%)

Incidentally found
Dyspnea
Surgery or trauma
Coexisting DVT
Received anticoagulation
PE recurrence
PE-related death

DVT = deep vein thrombosis, PE = pulmonary embolism

a
Note the higher incidence of recurrence and death in larger clot sizes, despite receiving anticoagulation at a higher rate than that of the
smaller clots. (Where applicable, all P values are <0.05.)
b
Parentheses denote percentages of total patients in group.
Adapted with permission from Cha SI, Shin KM, Lee JW, et al: Clinical characteristics of patients with peripheral pulmonary embolism. Respiration 2010;80(6):500-508.

tween single and multidetector

CTPA. Thus, the authors suggested
that subsegmental emboli, even untreated, may not be clinically significant. In addition, subsegmental emboli tend to be less associated with
DVT,17,19,20 which suggests a difference from the more dangerous embolic events. Subsegmental emboli
also have a rounded dot-like appearance, suggesting that they are part of
a normal physiologic process by
which lung fibrinolytic activity dissolves emboli to protect the greater
systemic circulation.19 Further, untreated subsegmental emboli have
good outcomes with respect to recurrent VTE and cardiopulmonary complications. According to the Prospective Investigation of Pulmonary
Embolism Diagnosis, subsegmental
emboli are more likely to be classified as low probability on a VQ
scan.21 Perrier et al22 showed that a
low-probability VQ scan combined
with low clinical probability results
in only a 1% to 2% risk of subsequent thromboembolic events (DVT
or PE) for up to 6 months; in addition, not instituting anticoagulation
appears to be safe for a lowprobability scan. In a study of 334
patients (median follow-up, 8
months),
subsegmental
emboli

590

showed no hemodynamic instability

on echocardiogram, and there was
no PE recurrence or PE-related death
despite 30% of patients not receiving
anticoagulation therapy.20 Patients
with segmental emboli also had no
PE recurrence, although 80% were
anticoagulated; however, they did
have a 4.5% rate of PE-related
deaths. Thus, subsegmental emboli
seem to have a better prognosis than
do the larger segmental emboli (Table 3).
Currently, the American College of
Chest Physicians guidelines do not
mention clot size as a variable. Physicians often do not take size or location of the clot into account before
instituting anticoagulant therapy,
and radiologists frequently do not
discuss their findings with treating
physicians. As one author notes, It
is not always clear whether small PE,
in the absence of demonstrable DVT,
justify the expense, mortality, and serious morbidity associated with anticoagulation.23 He suggests that
groups of people exist in which risks
of anticoagulation for PE may outweigh the benefits and that there are,
as well, certain scenarios in which
small emboli should be treated (Table 4). These studies suggest that
such patient stratification may be

necessary, that not all emboli are

equal, and that management strategies should be altered, particularly in
surgical patients.

Anticoagulation for
Pulmonary Embolism
It is clear that anticoagulation is of
paramount importance in treating
clinically relevant PE to prevent
death. In a landmark study, Barritt
and Jordan24 conducted a randomized controlled trial of patients with
PE. The group that received unfractionated heparin with vitamin K antagonist (n = 16) experienced no PE
recurrences or fatalities, whereas the
control group (n = 19) had 10 recurrences and 5 fatalities. Since then,
multiple studies have shown benefits
for anticoagulation and, based on
their results, societies have released
guidelines for anticoagulation therapy for PE.25,26 The American College of Chest Physicians provides the
following grade 1 (strong) recommendations: a confirmed PE is
treated with low-molecular-weight
heparin (LMWH), monitored intravenous or subcutaneous unfractionated heparin, weight-based subcutaneous unfractionated heparin, or

Journal of the American Academy of Orthopaedic Surgeons

Paul Tornetta, MD, and Yelena Bogdan, MD

Table 4
Suggestions for Anticoagulation Therapy for Small Pulmonary Emboli23
Cardiopulmonary Reserve

Coexisting DVT

Other Factor(s)

Anticoagulate

No
No
No
Yes

Symptomatic
Asymptomatic/Incidental finding
Anticoagulation contraindicated

Recurrent PE

No
No
No
Yes
Yes
Yes

Adequate
Adequate
Adequate
Adequate
Adequate
Inadequate

DVT = deep vein thrombosis, PE = pulmonary embolism

Table 5
American College of Chest Physicians Recommendations for Treatment of Pulmonary Embolism25
Type of
Pulmonary
Embolism
Acute

Management

Goals of Management/
Rationale

LMWH, UFH, or fondaparinux 5 d

LMWH: prevent formation of new

thrombi27
UFH better for increased bleeding
risk because rapidly reversed28
Initiation of VKA on day 1, discontinu- Prevent extension of thrombus and
ation of heparin with INR 2 24 h
disease recurrence28
Depends on clotting factors II
and X depletion27
With transient risk factors VKA 3 mo

Unprovoked/No risk fac- VKA 3 mo

tors
Evaluation for lifetime VKA
Incidence of recurrence higher at 2 yr
than provoked PE29
First unprovoked PE, no Lifetime VKA, target INR = 2.5

bleeding risk, patient

preference
PE in patients with canLMWH 3 mo
LMWH more effective in preventing
cer
recurrence than VKA30
LMWH or VKA as long as cancer re- LMWH more effective in preventing
mains active
recurrence than VKA30

Grade of
Recommendationa
1C

1A

1A
1A
1C
1A

1A
1C

INR = international normalized ratio, LMWH = low-molecular-weight heparin, PE = pulmonary embolism, UFH = unfractionated heparin,
VKA = vitamin K antagonist
a
A = high-quality evidence, C = low-quality evidence

subcutaneous fondaparinux for 5

days, with concurrent initiation of
vitamin K antagonists (eg, warfarin)
until the international normalized ratio is 2. For PE in a patient with
transient risk factors, 3 months of
treatment is recommended; for PE in
a patient without risk factors, 3
months plus evaluation for lifetime
anticoagulation is recommended,
September 2012, Vol 20, No 9

with a target international normalized ratio of 2.525 (Table 5). The British Thoracic Society has similar
guidelines, adding that thrombolytic
therapy should be used only in massive PE and that oral anticoagulation
should not be started unless PE is
confirmed with imaging. These
guidelines emphasize the risk of anticoagulation in certain patient sub-

groups, such as pregnant patients

and those with cancer, but postoperative patients are not mentioned.26
At least one response to the current
anticoagulation guidelines questions
whether they are appropriate for orthopaedic patients.31 Each of these
medications has drawbacks. Warfarin requires close monitoring and is
not predictable in its pharmacokinet-

591

Pulmonary Embolism in Orthopaedic Patients: Diagnosis and Management

Table 6
Summary of Outcomes of Anticoagulation Therapy for Pulmonary Embolism
Study
Nijkeuter et al35
Douketis et al36
Palla et al37

Patients
673 (PE)
2,052 (310 PE,
292 DVT + PE)
497 (PE)

Treatment Duration
(mo)

Follow-up (mo)

Recurrence Rate

3
6

3
54

2%
0.20.5/100 person-yearsa

12

12

9.6%

DVT = deep vein thrombosis, PE = pulmonary embolism

a
Risk of fatal pulmonary embolism

ics. LMWH and fondaparinux require nonoral administration. The

newly introduced direct thrombin inhibitor ximelagatran is an oral agent
and requires no monitoring; however, it causes elevation of transaminase levels and an increased rate of
coronary events of unclear significance.32
The timing of therapeutic anticoagulation for PE in postsurgical patients has not been thoroughly researched. In a study of spine surgery
patients with PE, three patients were
treated with inferior vena cava (IVC)
filters because the PE presented
within the first week of surgery, and
three were treated with anticoagulation because the PE presented 1 week
after surgery. Neither group of patients had severe complications from
the treatment.33 The authors recommended a waiting period of at least 8
postoperative days before instituting
a full dose of anticoagulation. Another study of neurosurgical patients
cited a high mortality rate (15%) associated with anticoagulation therapy; the authors could not comment
on safe timing.34
Several studies focus on the outcomes of anticoagulation therapy for
PE (Table 6). In a study of 673 patients with PE (of whom only 10
were surgical) who were treated with
anticoagulants for 3 months with
complete follow-up, Nijkeuter et al35
found a 3% recurrence rate of
thromboembolic events (20 patients)

592

and a 2% recurrence rate of PE (14

patients), of which 79% were fatal,
mostly in the first week. Immobilization for >3 days posed significant
risk for both recurrent VTE and fatal
recurrent PE (odds ratio, 2.79). The
overall 3-month mortality risk factors in the entire PE cohort were age,
immobilization, cancer, and inpatient
status. Douketis et al36 researched a
longer treatment period; in 2,052 patients (310 with PE, 292 with both
DVT and PE) who were anticoagulated for 6 months, with an average
54-month follow-up, the risk of fatal
PE after stopping therapy was 0.2 to
0.5 events per 100 person-years (case
fatality rate, 4% to 9%). For an even
longer treatment period, Palla et al37
looked at 497 patients with PE (33%
surgical) anticoagulated for 1 year.
Forty-eight patients (9.6%) had recurrent PE, which was fatal in 36 of
48 cases. Thirty-nine of 48 recurrences (81.2%) occurred within 10
days of diagnosis, and 2 patients had
recurrent nonfatal PE between 6 and
12 months. Finally, Stein et al27 reviewed several studies of timing of
heparin versus vitamin K antagonist
in DVT patients. Most VTE (DVT or
PE) recurrences took place after 5
days, even in those who were not
treated with heparin; a therapeutic
level of heparin in 24 hours resulted
in fewer recurrent events. These
studies suggest that the greatest risk
of PE arises within the first 2 weeks
and that these early recurrences have

a high fatality rate. It follows that

anticoagulation therapy is most important in the first days of diagnosis,
precisely when it is most dangerous
to a surgical patient. However, these
studies did not delineate the nature
of the embolus, such as size and location; this must be taken into account, given the earlier stated research on the effect of embolus size.
It is also important to remember that
the studies mentioned above focus
mostly on medical rather than surgical or orthopaedic patients, and so
their results may not be applicable.
In particular, the risks of anticoagulation on postoperative surgical site
bleeding are not addressed.
Other studies focus on the mortality of untreated PE, although there
are, to our knowledge, no randomized controlled trials of treated versus untreated PE. One study reviewed data from the Prospective
Investigation of Pulmonary Embolism Diagnosis of 20 patients who
had PE but had not received anticoagulation therapy. These patients
were more likely to have segmental
perfusion defects than were the
treated patients. One patient died,
and one had recurrent PE. The authors concluded that mild PE has
a low mortality risk.38 Nielsen et al39
presented a trial of 87 patients with
DVT, 43 of whom had a silent,
asymptomatic PE on lung scan. All
87 patients were randomized to receive or not receive anticoagulation.

Journal of the American Academy of Orthopaedic Surgeons

Paul Tornetta, MD, and Yelena Bogdan, MD

Anticoagulation therapy did not influence PE resolution on lung scan at

3-month follow-up.
The type of anticoagulation may
have an effect on outcomes, as well.
In a randomized trial, Hull et al40
looked at patients with both DVT
and nonmassive PE who received either LMWH (97 patients) or unfractionated heparin (103 patients).
Rates of recurrent VTE events were
zero versus 7%, respectively, suggesting that LMWH is at least as good a
therapy, if not better, than standard
intravenous heparin.
Therapeutic anticoagulation involves several risks: bleeding, thrombocytopenia, and osteoporosis, as
well as skin necrosis and, in the case
of warfarin, teratogenicity. Of these,
bleeding is most important and relevant for orthopaedic patients. Bleeding can lead to pain, prolonged rehabilitation, compartment syndrome,
return to the operating room, anemia
and transfusion, and wound infection. Bleeding is a strong predictor of
mortality in hospitalized patients.41
Even prophylactic anticoagulation
carries a bleeding risk at the surgical
site: a 6.38 relative risk for LMWH
compared with aspirin and a 4.88
relative risk for warfarin compared
with aspirin in one study.42
Much of the data on bleeding complications come from total joint literature. The rates quoted for bleeding
range from minimal to as high as
50%.43-46 None of these studies is
randomized; however, they provide
useful data on risk. In a review of
112 arthroplasty patients who were
treated for DVT or PE with intravenous heparin, Patterson et al43 found
an overall bleeding rate of 30%,
with decreasing risk over time (50%
if anticoagulated within 5 days of
surgery, 40% if within 7 days, and
15% if >1 week). In 41 patients
(35%), heparin was discontinued because of complications; 38 of those
patients received warfarin instead
September 2012, Vol 20, No 9

and did well. Twenty-three patients

treated for mild PE with warfarin
alone had no bleeding or PE-related
complications. The authors reached
several important conclusions: that a
bolus heparin dose may be too aggressive; that a PE should be confirmed before instituting treatment;
and, most importantly, that anticoagulation in the first postoperative
week carries a very high risk and
should be avoided. These findings
are echoed in other studies, which
show an increased bleeding rate in
the first month of anticoagulation.44
Supratherapeutic levels of international normalized ratio or partial
thromboplastin time are also associated with bleeding complications, as
well as higher transfusion requirements and longer hospitalizations.44,45
Anticoagulation therapy for even
small pulmonary emboli carries
risks. In one study of 43 falsenegative CT scans that eventually
were read as positive for PE, 21 patients did not receive therapeutic anticoagulation. Patients who received
no anticoagulation had a significantly lower rate of hemorrhage, renal failure, and early death than did
those who received anticoagulation
therapy.47 Another study of 71 patients who received anticoagulation
for subsegmental PE and 22 who did
not showed 8 instances of hemorrhage, including 5 major events, all
in the anticoagulated patients. Interestingly, no patient in either group
died of PE, and there was one PE recurrence in the anticoagulated
group.48 These findings emphasize
the potential harm of anticoagulant
therapy and warn against a one size
fits all approach to PE treatment.
IVC filters are an alternative for
patients with PE who have a high
risk of bleeding (ie, recent surgery)
or who cannot tolerate anticoagulants. They are also indicated for patients who experience a recurrent PE

despite therapeutic anticoagulation.

When placed, these filters block the
passage of emboli from the lower extremities to the pulmonary circulation. They can be permanent or retrievable (for up to 1 year); removal
carries a risk of iatrogenic IVC injury, which increases with the time
since filter placement. Anticoagulation therapy is begun, either concurrently with the filter or just before
removal, when the patients bleeding
risk is acceptable. Additionally, in
multiple-trauma patients, particularly those with long bone injury,
spinal cord injury, or pelvic fractures, IVC filters are often placed in
an attempt to prevent massive PE.
One review of 9,348 orthopaedic patients found a 1% total rate of filter
placement (90 patients), with 61%
of 90 filters placed prophylactically.49
The ratio of prophylactic-based to
treatment-based filters was 3.25 in
fractures and 2.1 in joint arthroplasties. In that study, 10% of the retrievable filters were not able to be
removed, and a further 11% had
complications during removal. Despite these risks, filters are commonly used in orthopaedics and provide another weapon in the arsenal
of PE prevention and treatment.

Summary
Orthopaedic patients present a dilemma for treating clinicians: they
are at high risk for both PE and
bleeding events. The orthopaedic
community has extensively debated
anticoagulation therapy for prophylaxis of PE, and the American Academy of Orthopaedic Surgeons has
presented clinical practice guidelines
to help guide care. However, data
suggest that we may be overdiagnosing PE, that not all pulmonary emboli are the same, that small emboli
are of questionable clinical relevance, and that the risks of anticoag-

593

Pulmonary Embolism in Orthopaedic Patients: Diagnosis and Management

ulation therapy are not minor.

We believe that the risks and benefits of mangement of the diagnosis of
PE should be reviewed with the patient and that a spectrum of treatment should be based on the size and
clinical presentation of the PE, as
should the magnitude and timing of
any orthopaedic surgical procedures.
At this juncture, we are unable to
make recommendations for treatment of small emboli. We feel that
the subject merits additional clinical
study and call for development of
guidelines for anticoagulation in orthopaedic patients with PE, specifically including the size and location
of the embolus and the potential risk
of bleeding.

postoperative orthopedic patient using

spiral CT scans. HSS J 2010;6(1):95-98.
5.

6.

References printed in bold type indicate those published within the past
5 years.
1.

American Academy of Orthopaedic

Surgeons: Preventing Venous
Thromboembolic Disease in Patients
Undergoing Elective Hip and Knee
Arthroplasty: Evidence-Based Guideline
and Evidence Report. Available at:
http://www.aaos.org/Research/guidelines/
VTE/VTE_full_guideline.pdf. Accessed
July 26, 2012.

16.

Auer RC, Schulman AR, Tuorto S, et al:

Use of helical CT is associated with an
increased incidence of postoperative
pulmonary emboli in cancer patients
with no change in the number of fatal
pulmonary emboli. J Am Coll Surg 2009;
208(5):871-880.

17.

Le Gal G, Righini M, Parent F, van

Strijen M, Couturaud F: Diagnosis and
management of subsegmental pulmonary
embolism. J Thromb Haemost 2006;
4(4):724-731.

18.

Carrier M, Righini M, Wells PS, et al:

Subsegmental pulmonary embolism
diagnosed by computed tomography:
Incidence and clinical implications. A
systematic review and meta-analysis of
the management outcome studies.
J Thromb Haemost 2010;8(8):17161722.

19.

Suh JM, Cronan JJ, Healey TT: Dots are

not clots: The over-diagnosis and overtreatment of PE. Emerg Radiol 2010;
17(5):347-352.

20.

Park B, Messina L, Dargon P, Huang W,

Ciocca R, Anderson FA: Recent trends in
clinical outcomes and resource
utilization for pulmonary embolism in
the United States: Findings from the
nationwide inpatient sample. Chest
2009;136(4):983-990.

Cha SI, Shin KM, Lee JW, et al: Clinical

characteristics of patients with peripheral
pulmonary embolism. Respiration 2010;
80(6):500-508.

21.

Wiener RS, Schwartz LM, Woloshin S:

Time trends in pulmonary embolism in
the United States: Evidence of
overdiagnosis. Arch Intern Med 2011;
171(9):831-837.

The PIOPED Investigators: Value of the

ventilation/perfusion scan in acute
pulmonary embolism: Results of the
prospective investigation of pulmonary
embolism diagnosis (PIOPED).
JAMA 1990;263(20):2753-2759.

22.

Perrier A, Miron MJ, Desmarais S, et al:

Using clinical evaluation and lung scan
to rule out suspected pulmonary
embolism: Is it a valid option in patients
with normal results of lower-limb venous
compression ultrasonography? Arch
Intern Med 2000;160(4):512-516.

23.

Goodman LR: Editorial: Small

pulmonary emboli: What do we know?
Radiology 2005;234(3):654-658.

24.

Barritt DW, Jordan SC: Anticoagulant

drugs in the treatment of pulmonary
embolism: A controlled trial. Lancet
1960;1(7138):1309-1312.

25.

Kearon C, Kahn SR, Agnelli G, et al:

Antithrombotic therapy for venous
thromboembolic disease: American
College of Chest Physicians EvidenceBased Clinical Practice Guidelines (8th
Edition). Chest 2008;133(6 suppl):454S545S.

26.

British Thoracic Society Standards of

Care Committee Pulmonary Embolism
Guideline Development Group: British
Thoracic Society guidelines for the
management of suspected acute
pulmonary embolism. Thorax 2003;

Anderson DR, Kahn SR, Rodger MA,

et al: Computed tomographic pulmonary
angiography vs ventilation-perfusion
lung scanning in patients with suspected
pulmonary embolism: A randomized
controlled trial. JAMA 2007;298(23):
2743-2753.

7.

Moores LK, Jackson WL Jr, Shorr AF,

Jackson JL: Meta-analysis: Outcomes in
patients with suspected pulmonary
embolism managed with computed
tomographic pulmonary angiography.
Ann Intern Med 2004;141(11):866-874.

8.

Subramaniam RM, Blair D, Gilbert K,

Coltman G, Sleigh J, Karalus N:
Withholding anticoagulation after a
negative computed tomography
pulmonary angiogram as a stand-alone
imaging investigation: A prospective
management study. Intern Med J 2007;
37(9):624-630.

References
Evidence-based Medicine: Levels of
evidence are described in the table of
contents. In this article, references 6,
21, 24, 36, 37, and 39 are level I
studies. References 3, 7, 12, 14, 1720, 22, 27, 32, 33, 35, 44, 45, and
47-49 are level II studies. References
4, 15, 16, and 34 are level III studies.
Reference 13 is a level IV study.

Wells PS, Anderson DR, Rodger M, et al:

Derivation of a simple clinical model to
categorize patients probability of
pulmonary embolism: Increasing the
models utility with the SimpliRED
D-dimer. Thromb Haemost 2000;83(3):
416-420.

9.

10.

11.

Burge AJ, Freeman KD, Klapper PJ,

Haramati LB: Increased diagnosis of
pulmonary embolism without a
corresponding decline in mortality
during the CT era. Clin Radiol 2008;
63(4):381-386.

12.

Ranji SR, Shojania KG, Trowbridge RL,

Auerbach AD: Impact of reliance on CT
pulmonary angiography on diagnosis of
pulmonary embolism: A Bayesian
analysis. J Hosp Med 2006;1(2):81-87.

2.

Montgomery KD, Geerts WH, Potter

HG, Helfet DL: Thromboembolic
complications in patients with pelvic
trauma. Clin Orthop Relat Res 1996;
329:68-87.

13.

Stein PD, Henry JW: Clinical

characteristics of patients with acute
pulmonary embolism stratified according
to their presenting syndromes. Chest
1997;112(4):974-979.

3.

Abelseth G, Buckley RE, Pineo GE, Hull

R, Rose MS: Incidence of deep-vein
thrombosis in patients with fractures of
the lower extremity distal to the hip.
J Orthop Trauma 1996;10(4):230-235.

14.

Dentali F, Ageno W, Becattini C, et al:

Prevalence and clinical history of
incidental, asymptomatic pulmonary
embolism: A meta-analysis. Thromb Res
2010;125(6):518-522.

Kim HJ, Walcott-Sapp S, Leggett K, et al:

Detection of pulmonary embolism in the

15.

4.

594

embolus after total joint arthroplasty:

Do size and location of embolus matter?
Am J Orthop (Belle Mead NJ) 2010;
39(4):185-189.

Pulido L, Grossman S, Smith EB, et al:

Clinical presentation of pulmonary

Journal of the American Academy of Orthopaedic Surgeons

Paul Tornetta, MD, and Yelena Bogdan, MD

58(6):470-483.
27.

28.

29.

30.

31.

32.

33.

34.

Stein PD, Hull RD, Matta F, Yaekoub

AY: Anticoagulant therapy for acute
venous thromboembolism: What we
think we know and what the data show
for the timing of recurrent events. Clin
Appl Thromb Hemost 2009;15(6):609612.
van Es J, Douma RA, Gerdes VE,
Kamphuisen PW, Bller HR: Acute
pulmonary embolism: Part 2. Treatment.
Nat Rev Cardiol 2010;7(11):613-622.
Baglin T, Luddington R, Brown K,
Baglin C: Incidence of recurrent venous
thromboembolism in relation to clinical
and thrombophilic risk factors:
Prospective cohort study. Lancet 2003;
362(9383):523-526.
Lee AY, Levine MN, Baker RI, et al:
Low-molecular-weight heparin versus a
coumarin for the prevention of recurrent
venous thromboembolism in patients
with cancer. N Engl J Med 2003;349(2):
146-153.
Callaghan JJ, Dorr LD, Engh GA, et al:
Prophylaxis for thromboembolic disease:
Recommendations from the American
College of Chest Physicians. Are they
appropriate for orthopaedic surgery?
J Arthroplasty 2005;20(3):273-274.
Huisman MV, Bounameaux H: Treating
patients with venous thromboembolism:
Initial strategies and long-term secondary
prevention. Semin Vasc Med 2005;5(3):
276-284.
Schizas C, Neumayer F, Kosmopoulos V:
Incidence and management of
pulmonary embolism following spinal
surgery occurring while under chemical
thromboprophylaxis. Eur Spine J 2008;
17(7):970-974.
Swann KW, Black PM, Baker MF:
Management of symptomatic deep

September 2012, Vol 20, No 9

venous thrombosis and pulmonary

embolism on a neurosurgical service.
J Neurosurg 1986;64(4):563-567.
35.

Nijkeuter M, Shne M, Tick LW, et al:

The natural course of hemodynamically
stable pulmonary embolism: Clinical
outcome and risk factors in a large
prospective cohort study. Chest 2007;
131(2):517-523.

36.

Douketis JD, Gu CS, Schulman S,

Ghirarduzzi A, Pengo V, Prandoni P:
The risk for fatal pulmonary embolism
after discontinuing anticoagulant therapy
for venous thromboembolism. Ann
Intern Med 2007;147(11):766-774.

37.

Palla A, Ribas C, Rossi G, Pepe P,

Marconi L, Prandoni P: The clinical
course of pulmonary embolism patients
anticoagulated for 1 year: Results of a
prospective, observational, cohort study.
J Thromb Haemost 2010;8(1):68-74.

38.

Stein PD, Henry JW, Relyea B: Untreated

patients with pulmonary embolism:
Outcome, clinical, and laboratory
assessment. Chest 1995;107(4):931-935.

39.

Nielsen HK, Husted SE, Krusell LR,

Fasting H, Charles P, Hansen HH: Silent
pulmonary embolism in patients with
deep venous thrombosis: Incidence and
fate in a randomized, controlled trial of
anticoagulation versus no anticoagulation. J Intern Med 1994;235(5):457-461.

40.

Hull RD, Raskob GE, Brant RF, et al:

Low-molecular-weight heparin vs
heparin in the treatment of patients with
pulmonary embolism. Arch Intern Med
2000;160(2):229-236.

41.

Eikelboom JW, Quinlan DJ, ODonnell

M: Major bleeding, mortality, and
efficacy of fondaparinux in venous
thromboembolism prevention trials.
Circulation 2009;120(20):2006-2011.

42.

Brown GA: Venous thromboembolism

prophylaxis after major orthopaedic

surgery: A pooled analysis of
randomized controlled trials.
J Arthroplasty 2009;24(6 suppl):77-83.
43.

Patterson BM, Marchand R, Ranawat C:

Complications of heparin therapy after
total joint arthroplasty. J Bone Joint Surg
Am 1989;71(8):1130-1134.

44.

Zidane M, Schram MT, Planken EW,

et al: Frequency of major hemorrhage in
patients treated with unfractionated
intravenous heparin for deep venous
thrombosis or pulmonary embolism: A
study in routine clinical practice. Arch
Intern Med 2000;160(15):2369-2373.

45.

Neviaser AS, Chang C, Lyman S, Della

Valle AG, Haas SB: High incidence of
complications from enoxaparin
treatment after arthroplasty. Clin
Orthop Relat Res 2010;468(1):115-119.

46.

Spencer FA, Emery C, Joffe SW, et al:

Incidence rates, clinical profile, and
outcomes of patients with venous
thromboembolism: The Worcester VTE
study. J Thromb Thrombolysis 2009;
28(4):401-409.

47.

Engelke C, Rummeny EJ, Marten K:

Pulmonary embolism at multi-detector
row CT of chest: One-year survival of
treated and untreated patients.
Radiology 2006;239(2):563-575.

48.

Donato AA, Khoche S, Santora J,

Wagner B: Clinical outcomes in patients
with isolated subsegmental pulmonary
emboli diagnosed by multidetector CT
pulmonary angiography. Thromb Res
2010;126(4):e266-e270.

49.

Bass AR, Mattern CJ, Voos JE, Peterson

MG, Trost DW: Inferior vena cava filter
placement in orthopedic surgery. Am J
Orthop (Belle Mead NJ) 2010;39(9):435439.

595