Full text of DOI:10.1007/s10140-002-0198-7

Emergency Radiology (2002) 9: 79–81
DOI 10.1007/s10140-002-0198-7
ORIGINAL ARTICLE
Roslyn F. Schneider Æ Francis D. Ntimba
Aman Hourizadeh Æ Jonathan B. Schwartz
Corey D. Eber Æ Madhavi Patnana Æ Richard Goldfarb
Massive pulmonary embolism: a comparison of radiological
and clinical characteristics and outcomes
Published online: 17 April 2002
Ó ASER 2002
Abstract Study objectives: To describe the clinical fea- eventually discharged and 4 (7%) died. Two of the
tures of radiographically massive pulmonary embolism deaths were not attributable to MPE. Conclusions: Pa-
(MPE). Design: Retrospective analysis. Setting: A 1,368- tients with MPE usually present with dyspnea and
bed teaching hospital. Patients or participants: Patients hypoxemia, and most survive without thrombolytic
with pulmonary embolism between June 1997 and De- therapy.
cember 1999. Interventions: Radiographic reports of
Keywords Venous thromboembolism Æ Massive
pulmonary embolism Æ Right ventricular
dysfunction Æ Hypoxemia Æ Clinical presentation
patients with a radiographic diagnosis of pulmonary
embolism were reviewed to determine whether MPE
(>50% vascular occlusion) was present. For patients
with MPE, vital signs, respiratory and cardiac symp-
toms, medical history, arterial blood gases, electro-
cardiographic (ECG) and echocardiographic results,
treatment, and hospital mortality were recorded. Mea-
surements and results: Fifty-four patients with MPE
were identified. Patient age range was 28–91 years (mean
71 years). Symptoms were: dyspnea in 38 (70%), chest
pain in 21 (38%), syncope in 12 (22%), palpitations in 6
(11%), systolic blood pressure <90 mmHg in 12 (22%),
tachycardia (>120 beats/min) in 15 (28%) and tachyp-
Pulmonary embolism is still not well understood, and
while previous reports suggested that the mortality was
not changing, more recent reports suggest a decline in
mortality [1]. Most patients with pulmonary embolism
do not undergo pulmonary angiography, and the clinical
presentations and hospital outcomes of patients with
massive pulmonary embolism (MPE) demonstrated by
other modalities have not been systematically reported.
MPE is defined angiographically as anatomic
obstruction of 50% or more of the pulmonary artery
circulation [2]. The extent of disease on imaging studies
has been suggested by some to predict severity [2, 3, 4,
5]. A reduction of at least 50% of the cross-sectional
area of the pulmonary artery is said to cause significant
hemodynamic instability, altered mental status, syncope,
marked hypoxia and dyspnea, and chest pain. The
degree of obstruction was an inclusion criterion for
participation in multicenter trials of thrombolytic
therapy in pulmonary embolism [6, 7]. We reviewed the
clinical presentation and hospital outcome of all patients
with radiographically defined MPE at Beth Israel
Medical Center during a 30-month period.
nea(respiratory rate >30) in 15 (28%). Pa
O (arterial
2
partial pressure of oxygen) was less than 60 mmHg in 28
(71%) and the alveolar–arterial oxygen gradient was
always greater than 20. ECG had an S1Q3T3 pattern in
6 (12%). Echocardiography revealed right ventricular
dilatation in 12/31 (38%). Forty-nine patients received
anticoagulation treatment, 4 (7%) received thrombolytic
therapy with anticoagulation, 5 had inferior vena cava
filters (IVC) alone, 6 received IVC filters with antico-
agulation, and 2 received thrombolytic therapy, antico-
agulation, and IVC filters. Eighteen (33%) patients were
treated in the intensive care unit, 3 (5.5%) with
mechanical ventilation. Fifty (93%) patients were
R.F. Schneider (&) Æ F.D. Ntimba Æ A. Hourizadeh
J.B. Schwartz Æ C.D. Eber Æ M. Patnana Æ R. Goldfarb
Division of Pulmonary and Critical Care Medicine,
Beth Israel Medical Center,
1st Avenue at 16th Street,
New York, NY 10003, USA
E-mail: RSchneider@BethIsraelNY.org
Tel.: +1-212-4204016
Fax: +1-212-4204615
Materials and methods
All V/Q scans, spiral CT angiograms (SCTA), and pulmonary
angiograms performed between June 1997 and December 1999 for
the evaluation of suspected pulmonary embolism were reviewed.
The pulmonary angiograms and SCTA were graded for severity for
the purpose of this study by chest radiologists. The extent of vas-
cular occlusion was determined according to the Miller scoring

80
system for angiography, adapted to spiral CT [8]. One point was
assigned for each involved segmental artery, with a range of values
from 1 to 16. A massive pulmonary embolus was defined as a filling
defect affecting 50% or more of the pulmonary arterial distribu-
tion. Any score equal to or greater than 7 was considered to rep-
resent MPE with the potential to affect the left main pulmonary
artery, and thus the entire left lung. The V/Q scans, reviewed by a
nuclear medicine radiologist, were graded using the percentage of
perfusion defects involving both lungs, with MPE defined as
greater than 50% perfusion defect.
Table 1. Symptoms in patients with MPE
Symptom
No. of patients (%)
Dyspnea38 (70)
Chest pain
Syncope
21 (38)
12 (22)
6 (11)
Palpitations
The medical records of those determined to have MPE were
reviewed with regard to clinical presentation and vital signs. Other
data retrieved included arterial blood gas measurements, ECG,
echocardiography, treatment, and hospital outcomes.
Table 2. Physical findings and diagnostic test results in patients
with MPE
Total no.
with data
available
No. of
patients
(%)
Results
Systolic BP <90 mmHg
Heart rate >120
Respiratory rate >30
PaO <60 mmHg,
2
FIO = 0.21
2
Alveolar–arterial
gradient >20 FIO = 0.21
2
Sinus tachycardia
Right bundle branch block
S1Q3T3 pattern
Right ventricular dilation
54
54
54
39
12 (22)
15 (28)
15 (28)
28 (71)
We reviewed the reports of 1,768 V/Q scans, 466 SCTA,
and 36 pulmonary angiograms performed during the
study period. Seventy high-probability V/Q scans, 50
SCTA, and 10 pulmonary angiograms demonstrated
pulmonary emboli; 54 studies met our criteria for MPE,
made up of 24 V/Q scans, 29 SCTA, and one pulmonary
angiogram. Four patients (7%) died during their hos-
pitalization. In two cases death was not thought to be
attributable to the pulmonary embolism: these patients
had multiorgan system failure, were septicemic, had in-
ferior vena cava filters in place, and died more than 2
weeks following their pulmonary embolism. In the third
case the patient was an 81-year-old man after hip frac-
ture surgery with hypotension on presentation. He was
treated with heparin but suffered progressive hemody-
namic compromise and died the following day. The
fourth patient who died was 40 years old, had widely
metastatic breast cancer, and was normotensive and
tachycardic on presentation with an oxygen saturation
of 88%. She had an inferior vena cava filter placed
within the first 24 h of presentation and died on the
fourth subsequent day, presumably from pulmonary
embolism.
39
39 (100)
50
50
50
31
27 (54)
7 (14)
6 (12)
12 (38)
Table 3. Treatment regimens among patients with MPE
Treatment
No. of patients
Heparin alone
39
2
Tissue plasminogen activator
followed by heparin and inferior
venacavafilter
Tissue plasminogen activator
followed by heparin
2
6
Inferior vena cava filter and anticoagulation
Inferior venacavafilter
5
Hypotension and right ventricular dysfunction were the
variables found to correlate with higher mortality. The
combination of both was a stronger predictor of excess
mortality than either one alone. They concluded that
mortality was not related to the embolic load per se but
rather to whether or not the patient was in shock. De-
spite this, two multicenter thrombolytic trials required
Miller index rather than hemodynamic criteria for entry
to select the most severely affected patients [6, 7].
Bankier et al. demonstrated that the SCTA could be
used to score the radiographic severity using a modifi-
cation of the Miller score [8]. They too discovered a poor
correlation between the radiographic score and clinical
severity.
The ICOPER data [12] showed that patients with
right ventricular dysfunction have a higher (20.9% vs.
14.8%) 3-month all-cause mortality than those without
right ventricular dysfunction. Specifically, the patients
who were hypotensive on presentation had a 58.3%
mortality, compared with a 15.1% mortality in those
who were normotensive [12]. Most patients who die
The symptoms documented in all patients studied are
shown in Table 1. Physical findings and other objective
data are shown in Table 2 and treatment regimens in
Table 3.
Discussion
Massive pulmonary embolism accompanied by hemo-
dynamic compromise calls for prompt intervention that
usually includes thrombolytic therapy [9, 10]. The use of
the term ‘‘massive’’ to describe pulmonary embolism
implies to clinicians the need for more aggressive treat-
ment.
Alpert et al. in 1976 noted a higher hospital mortality
in patients with emboli causing more than 50% ob-
struction than in those with lesser degrees of obstruction
[11]. Further analyzed, their data revealed that those
who were not hypotensive had hospital mortality rates
similar to the entire group regardless of embolic load.

81
from pulmonary embolism do so within the first few
hours of symptom onset [11]. Those who survive to be
diagnosed and treated for at least 3 months with anti-
coagulation are unlikely to die from the pulmonary
embolism. For those treated, the reported mortality rate
attributable to pulmonary embolism is reported to be
below 0.5% [13] regardless of the size of the emboli or
the extent of hemodynamic compromise from the initial
embolic event.
Miller et al. studied hemodynamically stable patients
without prior cardiopulmonary disease and with acute
pulmonary embolism [14]. They showed that there was
no association between right ventricular dysfunction and
the extent of pulmonary vascular obstruction graded by
V/Q scan. Their study group had a mean degree of ob-
struction of 37%. In our study group greater than 50%
obstruction was found in 40% of all patients diagnosed
radiographically with pulmonary embolism. Our pa-
tients with nondiagnostic radiographic studies followed
by leg studies diagnostic for deep vein thrombosis and
probable pulmonary embolism were not included. This
may account for our rather high proportion of patients
with MPE.
Conclusion
Our data suggest that the size of the pulmonary emboli as
determined radiographically does not predict the clinical
course or risk of hospital morbidity and mortality. The
term ‘‘massive’’ pulmonary embolism should be reserved
for use in reference to those patients with significant
hemodynamic compromise. The decision to use thromb-
olytic therapy in practice and in scientific trials should be
based on clinical rather than radiographic criteria.
References
1. Lilienfeld DE (2000) Decreasing mortality from pulmonary
embolism in the United States, 1979–1996. Int J Epidemiol
29:465–469
2. Miller GAH, Sutton GC, Kerr IH, et al (1971) Comparison of
streptokinase and heparin in treatment of isolated acute
massive pulmonary embolism. Br Med J 2:681–684
3. Shih WJ, Pulmano C (1996) Massive pulmonary embolism
without symptoms demonstrated by radionuclide imaging with
thromboemboli in both main pulmonary arteries. Clin Nucl
Med 21:465–468
4. Veith FV, Hobson RW, Williams RA, et al (1994) Venous and
lymphatic disorders. Vascular surgery: principles and practice,
2nd edn. McGraw–Hill, New York, pp 860–864
5. Moser KM (1997) Pulmonary embolism – state of the art. Am
Rev Resp Dis 115:829–852
Only 31 of the 54 patients had echocardiography
performed. This may or may not be due to greater
severity of presentation as there are no accepted stan-
dard indications for echocardiography in this setting.
Physicians vary considerably in their use of echocardi-
ography in pulmonary embolism.
6. Sors H, Pacouret G, Azarian R, et al (1994) Hemodynamic
effects of bolus vs. 2-h infusion of alteplase in acute massive
pulmonary embolism. Chest 106:712–717
Hospital mortality may not be the best endpoint but
is nevertheless a useful one. Deaths from pulmonary
embolism usually occur early [15]. The rationale for
using thrombolytic therapy is to improve hemody-
namic function acutely. Death rates due to pulmonary
embolism during the 3 months following MPE would
be another possible endpoint. However, when patients
die after discharge from hospital, it can be quite dif-
ficult to know whether to attribute the death to the
pulmonary embolism or to comorbid conditions. Our
sample size was small, which limits the confidence in
the findings. The patients are, however, representative
of a 30-month experience in a large urban teaching
hospital.
In our series, only 12 of 54 patients (22%) with
radiographically defined MPE were hypotensive.
Among all patients with MPE, the mortality attributable
to pulmonary embolism was 3.7% (two patients). It is
likely that the morbidity and mortality and immediate
hemodynamic effect are dependent on the previous
cardiopulmonary status of the patient as well as the
embolic burden. The most common presenting symp-
toms were dyspnea and chest pain and most patients
were treated with anticoagulation alone.
7. Dalla-Volta S, Palla A, Santolicandro A, et al (1992) PAIMS 2:
Alteplase combined with heparin versus heparin in the treat-
ment of acute pulmonary embolism. Plasminogen activator
Italian multicenter study 2. J Am Coll Cardiol 20:520–526
8. Bankier AA, Karin J, Fleischmann D, et al (1997) Severity
assessment of acute pulmonary embolism with spiral CT:
evaluation of two modified angiographic scores and compari-
son with clinical data. J Thorac Imaging 12:150–158
9. Hyers TM, Agnelli G, Hull RD, et al (1998) Antithrombotic
therapy for venous thromboembolic disease. Chest 114:561S-
578S
10. Arcasoy SM, Kreit JW (1999) Thrombolytic therapy of
pulmonary embolism. Chest 115:1695–1707
11. Alpert JS, Smith R, Carlson J, et al (1976) Mortality in patients
treated for pulmonary embolism. JAMA 236:1477–1480
12. Goldhaber SZ, Visani L, De Rosa M (1999) Acute pulmonary
embolism: clinical outcomes in the International Cooperative
Pulmonary Embolism Registry (ICOPER). Lancet 353:1386–
1389
13. Douketis JD, Kearon C, Bates S, et al (1998) Risk of fatal
pulmonary embolism in patients with treated venous throm-
boembolism. JAMA 279:458–462
14. Miller RL, Das S, Anandarangam T, et al (1998) Association
between right ventricular function and perfusion abnormalities
in hemodynamically stable patients with acute pulmonary
embolism. Chest 113:665–670
15. The PIOPED investigators (1990) Value of the ventilation/per-
fusion scan in acute pulmonary embolism. JAMA 263:2753–
2759