DOI:10.2214/AJR.07.3516
AJR 2008; 191:885-889
© American Roentgen Ray Society
Venous Thromboembolism and Occult Malignancy: Simultaneous Detection During Pulmonary CT Angiography with CT Venography
Guillaume Bierry1,
Nathalie Holl1,
Frauke Kellner2,
Sophie Riehm1,
Marie-Noelle Roedlich1,
Michel Greget1 and
Francis Veillon1
1 Department of Radiology 1, University Hospital of Strasbourg, 10 Ave.
Molière, 67098 Strasbourg Cedex, France.
2 Department of Radiology, Inselspital Bern, Bern, Switzerland.
Received December 8, 2007;
accepted after revision March 17, 2008.
Address correspondence to G. Bierry
(guillaume.bierry{at}chru-strasbourg.fr).
Abstract
OBJECTIVE. We explored the potential for patients with proven venous
thromboembolism or pulmonary embolism (PE) to have occult malignancies
detected during the same CT examination. To verify this, we compared the
presence of occult malignancies identified on pulmonary artery CT angiography
(CTA) and CT venography (CTV) when venous thromboembolism (VTE) was
present.
SUBJECTS AND METHODS. Pulmonary artery CTA combined with CTV was
performed on a 16-MDCT scanner on 186 adult patients suspected of having
pulmonary embolism without any known malignancies. CTV was performed from the
diaphragm to the knee 180 seconds after CTA. Two radiologists evaluated the
presence of VTE, that is PE or deep venous thrombosis (DVT), and tumor lesions
on both examinations in consensus. The malignant nature of the possibly
identified tumors was confirmed by pathologic examination.
RESULTS. VTE was found in 49 patients (26%). Malignant tumors were
detected in 24 patients (13%). Eleven patients with malignant tumors had VTE
(46% of patients with malignant tumors; 22% with VTE and 6% of all patients).
There was correlation with presence of malignancies between both and DVT and
DVT associated with PE but not between presence of malignancies and PE only.
Patients with DVT and those with DVT associated with PE had a risk ratio of
3.2 and 3.3, respectively, for having a malignant tumor discovered
simultaneously.
CONCLUSION. A high number of malignant tumors can be incidentally
discovered on pulmonary artery CTA, even more so with additional CTV.
Radiologists should scrutinize scans to pick up unknown malignancies,
especially in patients with identified VTE.
Keywords: CT venography • occult malignancy • pulmonary CT angiography • venous thromboembolism
Introduction
CT angiography (CTA) of the pulmonary artery is well accepted as the
technique of choice for the diagnosis of pulmonary embolism (PE). A recent
protocol is to combine pulmonary artery CTA and CT venography (CTV), which is
scanned between the level of the abdomen to the lower limbs, during the venous
phase, to detect simultaneous deep venous thrombosis (DVT)
[1]. CTV is as accurate as
sonography of the lower limb for the diagnosis of DVT. It even increases
venous thromboembolism (VTE) detection when associated with pulmonary artery
CTA [2].
A near full-body scan of the patient, performed in one session, gives
emergency and chest radiologists a huge volume of images to analyze.
Incidental tumor detection has been reported in other CT procedures such as CT
coronarography, CT colonography, or CT angiography of the great vessels
[3]. VTE is known to occur more
often in patients with malignancies. Conversely, we suggest that a malignant
disease is also likely to be discovered during pulmonary artery CTA and CTV in
patients with simultaneously identified VTE.
This study was performed prospectively, over a 12-month period, with
examination of inpatients who underwent combined pulmonary artery CTA and CTV
for detection of suspected VTE. We determined the occurrence of simultaneously
discovered mal ignancies and VTE.
Subjects and Methods
Patient Selection
We performed a prospective study between January 2006 and January 2007; all
inpatients suspected of having pulmonary embolism (PE) or DVT referred to our
unit for pulmonary artery CTA with CTV were included. Indications for CT
examinations were considered appropriate in patients with high clinical
probability of VTE: presence of tachycardia > 100 beats per minute (bpm),
low grade fever, hemoptysis, chest pain, unilateral edema and pain on deep
venous palpation, unilateral lower-limb pain, hypoxemia, abnormal ECG (right
heart failure sign), abnormal blood gas test (shunt effect), and postsurgical
period. CT was also considered in patients with low to moderate clinical
probabilities with elevated D-dimer serum levels
[4].
Patients with a history of malignancy or of VTE were excluded. Patients
below 40 years were excluded because venous sonography is preferred over CTV
in young patients due to the lesser radiation exposure. A total of 186
patients were finally included in our study.
CT Acquisition
The thorax was scanned from apex to base on a 16-MDCT scanner (Sensation
16, Siemens Medical Solutions) with a 0.75-mm-collimation, helical
acquisition, 1-mm reconstruction, 0.7-mm reconstruction increment, 0.5-second
rotation time, 100–140 mAs, and 120 kVp. Acquisition start was automated
with a Care Bolus system (Siemens Medical Solutions) after injection of 140 mL
of iodinated contrast medium in an antebrachial vein with a flow rate of 3
mL/s. The abdomen and lower limbs were scanned from the diaphragm down to the
subpopliteal recess with a 1.5-mm collimation, 5-mm reconstruction, 5-mm
reconstruction increment, 100 mAs, 120 kVp, 180 seconds after the start of
contrast medium injection.
View larger version (60K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —Simultaneously discovered pancreatic carcinoma, deep venous
thrombosis, and pulmonary embolism in 45-year-old woman presenting with sudden
shortness of breath and elevated D-dimers. Pulmonary artery CT angiography
image shows left pulmonary artery thrombosis (arrow).
|
|
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —Simultaneously discovered pancreatic carcinoma, deep venous
thrombosis, and pulmonary embolism in 45-year-old woman presenting with sudden
shortness of breath and elevated D-dimers. On CT venography (CTV) image,
tissue lesion of pancreas head (arrow) proven to be pancreatic
carcinoma with mild biliary and pancreatic tract dilatation was
identified.
|
|
View larger version (144K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —Simultaneously discovered pancreatic carcinoma, deep venous
thrombosis, and pulmonary embolism in 45-year-old woman presenting with sudden
shortness of breath and elevated D-dimers. Partial thrombosis of right iliac
vein (arrow) was simultaneously detected during CTV.
|
|
View larger version (78K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —Lung carcinoma with deep venous thrombosis in 66-year-old man
who presented with atypical chest pain. Pulmonary artery CT angiography image
shows tissue lesion of right lower lobe (arrow) that was confirmed to
be non-small-cell lung carcinoma at transbronchial biopsy.
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —Lung carcinoma with deep venous thrombosis in 66-year-old man
who presented with atypical chest pain. CT venography image shows thrombosis
of superficial and deep right femoral veins appearing as intraluminal contrast
media void with surrounding contrast rim (arrow).
|
|
CT Interpretation
CT examinations were prospectively analyzed by two radiologists working in
consensus. The workstation used was Leonardo (Siemens Medical Solutions).
Analysis was performed using standard mediastinal windows, lung windows, and
bone wind ows. One of the two radiologists was a radiology resident with more
than 3 years of experience, the other a staff radiologist with 20 years of
experience.
In each patient, the presence of PE, DVT, and tumoral lesion was recorded.
Thrombi were identified as low attenuating, partial, or complete intraluminal
filling defects surrounded by a high-attenuation ring of enhanced blood.
Thrombus length and location were not considered for analysis. Any tumorlike
lesion identified by pulmonary artery CTA or CTV that presented radiologic
findings suggestive of malignancy was pointed out in the radiologic
report.
Statistical Analysis
Analysis was performed with SPSS software, version 13.0. Data were compared
by using the chi-square test of independence
[5,
6]. A p value <
0.05 was considered statistically significant.
Results
A total of 186 patients were included in the study (84 men and 102 women;
mean age [±SD], 62 ± 17 years). VTE was present in 49 patients
(26.3%): 25 had both PE and DVT (13.4%), 11 (5.9%) patients had PE, and 13
(7%) had DVT (Figs. 1A,
1B,
1C,
2A,
2B,
3A,
3B,
3C). The mean age of patients
with PE or DVT was 63.6 ± 17.8 years. Occult malignancy was detected in
24 patients (10 men and 14 women, 12.9% of all patients) (Tables
1 and
2).
View larger version (64K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —Hepatic metastases from occult breast carcinoma with
concomitant deep venous thrombosis and pulmonary embolism in 76-year-old woman
who presented with right upper quadrant pain with elevated D-dimers. Pulmonary
artery CT angiography image shows bilateral pulmonary artery embolism
(arrows).
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —Hepatic metastases from occult breast carcinoma with
concomitant deep venous thrombosis and pulmonary embolism in 76-year-old woman
who presented with right upper quadrant pain with elevated D-dimers. CT
venography (CTV) image shows left popliteal venous thrombosis
(arrow).
|
|
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —Hepatic metastases from occult breast carcinoma with
concomitant deep venous thrombosis and pulmonary embolism in 76-year-old woman
who presented with right upper quadrant pain with elevated D-dimers. CTV also
detected multiple hypovascular hepatic nodules with metastases
(arrows).
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2: Size and TNM Staging of Malignant Tumors Incidentally Discovered on
Pulmonary Artery CT Angiography and CT Venography
|
|
Eleven patients with occult malignancies presented with PE or DVT (6% of
all patients, 45.8% of patients with malignancies) (Figs.
1A,
1B,
1C,
2A,
2B,
3A,
3B,
3C)
(Table 1). Of these patients,
four were men and seven were women, with a mean age of 69.7 ± 10.7
years. Eight patients had PE and DVT, two had DVT without PE, and one had PE
without DVT (Table 1).
Malignant tumors were incidentally detected in an additional 13 patients
without PE or DVT (7% of patients).
Statistical analysis showed a significant correlation between discovered
malignant tumors and presence of DVT (p = 0.01). Patients with DVT
had a risk ratio of 3.2 to have a malignant tumor simultaneously discovered.
No statistically significant cor relation was present between presence of
malignant tumors and PE only (p = 0.06). A significant correlation,
however, was found between the presence of a malignant tumor and PE when DVT
was present (p = 0.02). The risk ratio was 3.3.
Discussion
A malignant disease is widely accepted to be a major risk factor for VTE
[7]. The pathogenic mechanisms
of thrombosis involve a complex interaction between tumor cells and various
hemostatic components, such as the coagulation and fibrinolytic systems, the
vascular endothelium, leukocytes, and platelets
[8]. Patients with malignancies
have a fourfold risk of developing VTE and a sixfold risk when receiving chemo
therapy. Depending on the type of neoplasm and chemotherapy, VTE is present in
0.2% to 43% of patients. Incidentally discovered PE occurs in 0.6% to 5% of
patients with malignancies [9,
10].
VTE is a known indicator of occult malignancies, and about 10% of patients
with idiopathic VTE actually have an under lying identifiable malignancy. In
this respect, VTE can precede the diagnosis of malignancy by months
[11]. The literature reports
an odds ratio of 3.2–4.8 for having a newly diagnosed malignancy in
patients with VTE compared with patients without
[12,
13]. Routine cancer screening
has been proposed in patients with idiopathic VTE
[8,
11,
14,
15].
In our series, occult cancer was simultaneously discovered on CT
examinations in 11 patients with DVT or PE, corresponding to 45.8% (11/24) of
patients. Patients presenting with both cancer and VTE represented
approximately 6% of all patients, a result that is in accordance with the
literature [16]. In addition,
the main cancer types reported to be associated with VTE are cancer of the
pancreas, kidney, ovary, and stomach, all cancer types that were detected in
our series [16]. Combined
pulmonary artery CTA and CTV are therefore simultaneous cancer screening
procedures that are close to being a VTE diagnostic technique.
We found a correlation between the presence of malignancies and presence of
DVT and between the presence of malignancies and DVT with PE. Patients with
DVT or with DVT and PE had risks 3.2 and 3.3 times higher of having an occult
malignancy than patients without VTE. Rather surprisingly, no correlation was
present between presence of malignancy and pulmonary embolism alone. This may
have several different reasons.
Some PEs, especially subsegmental ones, may not have been detected in
patients with malignancies. Because the statistical test is based on
comparison between patients with PE and patients without, the misdiagnosis of
one or two patients with PE and simultaneous malignancies would have modified
the statistical results significantly. Because of the relatively small study
population, missing one or two PEs in this study would explain why the link
between PE and malignancy could not be found. Also DVT and PE are considered a
continuum in the thrombo embolic pathology, and PE is potentially present in
all patients with DVT. A study based on a larger population would have
probably showed a statistical link between PE only and malignancies. In
addition, our study only analyzed the presence of malignancy in patients with
clinical symptoms of thromboembolism, although it has been reported that
silent PEs are more frequent than clinically apparent ones in patients with
DVT, and a wider screening of patients with malignancies may have increased
the frequency of detected thromboembolism
[17]. Because malign ancies
are likely to be diagnosed in hospitalized patients with thromboembolism with
combined pulmonary artery CTA and CTV, these CT procedures should be
considered a simultaneous cancer screening.
We only included hospitalized patients, who have an increased risk for
developing VTE. Patients from the emergency department were not included. The
frequency of occult malignancies detected in our series was particularly high
compared with what has been reported in the literature. In our study, thoracic
malignancies were present in 5.9% (11 of 186) of patients, whereas the
incidence has been reported to be less than 1% in CT cardiac imaging or
pulmonary artery CTA [11,
18,
19]. CT colonography has been
reported to show that less than 2% of patients have malignant tumoral lesions
(colonic lesions excluded) whereas we discovered abdominal malignancies in
6.5% (12 of 186) of patients
[20].
Several factors can explain the high frequency of malignancies observed in
our study. One is that we included only hospitalized patients, whereas other
reports of inciden tally detected malignancies during CT exami na tions
considered outpatients in lung screen ing, CT colonography, and coronary CT
angio graphy or patients from emergency departments
[19]. Inpatients when admitted
because of nonmalignant diseases present with several pathologies and risk
factors that are also risk factors for the development of malignant tumors
(i.e., smoking, toxic exposure, poor nutrition).
Another factor is the age of our patients (mean age, 62 years). Patients
younger than 40 years old were excluded from the study because the relative
benefit from CTV and radiation exposure was considered too poor to be applied
to a population of reproductive age. The inclusion criterion of patients older
than 40 years represents a bias of this study and very likely increased the
probability of detecting malignancies.
A third factor is the protocol used for the non invasive CTV. The use of
contrast medium, which is not necessary in CT colonography and lung cancer
screening, may have improved the detection of malignancy. Detection of
malignant tumors is reported to be higher with contrast-enhanced CT
examinations in comparison with unenhanced examinations
[21]. Patients were explored
from the diaphragm to the level of the knees. Any abdominal or pelvic tumors
could potentially be detected. The use of continuous thin-section slices (5
mm) is now rec om mended for accurate thrombus detection
[2,
22]. This allowed an
exploration with the same accuracy obtained in routine abdominal CT, with even
small tumors possibly identified. The literature reports that intraabdominal
veins (inferior vena cava, renal veins) account for less than 5% of the
thrombosed veins detected during CTV. For this reason, scanning downward
starting below the level of the iliac crest has been suggested
[22,
23].
Among the 17 primary solid tumors, two cases of lung cancer presented with
metastases or nodal invasion that precluded surgical management
(Table 2). In one patient with
hepatic metastases from colon carcinoma, lesions were not included in the
field of the thoracic CT examination and were only observed on the CTV. Only
the rectal carcinoma was positive on physical examination, and one renal tumor
had microscopic hematuria. In both cases, such observations have been made
after the radiologic diagnosis. A total of eight abdom inal malignancies
(ovary, pancreas, ampullary adenocarcinoma, hepatic meta stasis, colon,
stomach, and kidney) would have only been detected by CTV. Identification of
those abdominal tumors undoubtedly increased the radiation dose delivered to
the patient. But, once a VTE is diagnosed, several additional investigations
are routinely performed to detect associated malignancies
[14]. Most notably, abdominal
CT is routinely performed. Thus the benefit of dose reduction scanning below
the iliac crest is only temporary.
The study had some limitations. First, CT examinations were performed on a
16-MDCT unit. The use of a more modern 64-MDCT unit might have detected more
thromboembolisms. Second, patients with recurrent thromboembolisms were not
included in our study. In such patients, an even higher risk of occult
malignancies is expected, and a similar study based on this population may be
the subject of another investigation.
In conclusion, when performing pulmonary artery CTA and CTV we are aware of
the necessity to analyze all information produced by the imaging examination.
An attentive analysis of the entire thoracic and abdominal structures on all
pulmonary artery CTA and CTV examinations should be routine. An even more
careful evaluation should be made in hospitalized patients in whom thrombo
embolic disease is discover ed. In such patients, pulmonary artery CTA and CTV
should also be considered a cancer screening procedure with an increased
likelihood of finding an occult malignancy. When the CTV exam ination begins
at the level of the diaphragm instead of below the level of the iliac crest,
it not only permits the detection of venous thrombosis but also serves as a
simultaneous screening for underlying malignant disease. An attentive analysis
of thoracic and ab dominal struc tures can in some cases obviate further
investigation.
References
- Cham MD, Yankelevitz DF, Shaham D, et al. Deep venous thrombosis:
detection by using indirect CT venography. The Pulmonary
Angiography–Indirect CT Venography Cooperative Group.
Radiology 2000;216
: 744–751[Abstract/Free Full Text]
- Ghaye B, Nchimi A, Noukoua CT, Dondelinger RF. Does multi-detector
row CT pulmonary angiography reduce the incremental value of indirect CT
venography compared with single-detector row CT pulmonary angiography?
Radiology 2006;240
: 256–262[Abstract/Free Full Text]
- Budoff MJ, Fischer H, Gopal A. Incidental findings with cardiac CT
evaluation: should we read beyond the heart? Catheter Cardiovasc
Interv 2006; 68:965
–973[CrossRef][Medline]
- Le Gal G, Righini M, Roy PM, et al. Prediction of pulmonary
embolism in the emergency department: the revised Geneva score. Ann
Intern Med 2006; 144:165
–171[Abstract/Free Full Text]
- Sistrom CL, Garvan CW. Proportions, odds, and risk.
Radiology 2004;230
: 12–19[Abstract/Free Full Text]
- Applegate KE, Tello R, Ying J. Hypothesis testing III: counts and
medians. Radiology 2003;228
: 603–608[Abstract/Free Full Text]
- Heit JA, Silverstein MD, Mohr DN, Petterson TM, O'Fallon WM, Melton
LJ 3rd. Risk factors for deep vein thrombosis and pulmonary embolism: a
population-based case-control study. Arch Intern Med2000; 160:809
–815[Abstract/Free Full Text]
- Prandoni P, Falanga A, Piccioli A. Cancer and venous
thromboembolism. Lancet Oncol 2005;6
: 401–410[CrossRef][Medline]
- Gosselin MV, Rubin GD, Leung AN, Huang J, Rizk NW. Unsuspected
pulmonary embolism: prospective detection on routine helical CT scans.
Radiology 1998;208
: 209–215[Abstract/Free Full Text]
- Sebastian AJ, Paddon AJ. Clinically unsuspected pulmonary embolism:
an important secondary finding in oncology CT. Clin
Radiol 2006; 61:81
–85[CrossRef][Medline]
- Lee AY. Thrombosis and cancer: the role of screening for occult
cancer and recognizing the underlying biological mechanisms.
Hematology Am Soc Hematol Educ Program2006
: 438–443
- Prins MH, Hettiarachchi RJ, Lensing AW, Hirsh J. Newly diagnosed
malignancy in patients with venous thromboembolism: search or wait and see?
Thromb Haemost 1997;78
: 121–125[Medline]
- Monreal M, Fernandez-Llamazares J, Perandreu J, Urrutia A,
Sahuquillo JC, Contel E. Occult cancer in patients with venous
thromboembolism: which patients, which cancers. Thromb
Haemost 1997; 78:1316
–1318[Medline]
- Piccioli A, Prandoni P, Ewenstein BM, Goldhaber SZ. Cancer and
venous thromboembolism. Am Heart J 1996;132
: 850–855[CrossRef][Medline]
- Piccioli A, Lensing AW, Prins MH, et al. Extensive screening for
occult malignant disease in idiopathic venous thromboembolism: a prospective
randomized clinical trial. J Thromb Haemost2004; 2:884
–889[CrossRef][Medline]
- Khorana AA, Francis CW, Culakova E, Kuderer NM, Lyman GH.
Frequency, risk factors, and trends for venous thromboembolism among
hospitalized cancer patients. Cancer2007; 110:2339
–2346[CrossRef][Medline]
- Cronin CG, Lohan DG, Keane M, Roche C, Murphy JM. Prevalence and
significance of asymptomatic venous thromboembolic disease found on oncologic
staging CT. AJR 2007;189
: 162–170[Abstract/Free Full Text]
- Hunold P, Schmermund A, Seibel RM, Gronemeyer DH, Erbel R.
Prevalence and clinical significance of accidental findings in electron-beam
tomographic scans for coronary artery calcification. Eur Heart
J 2001; 22:1748
–1758[Abstract/Free Full Text]
- Kino A, Boiselle PM, Raptopoulos V, Hatabu H. Lung cancer detected
in patients presenting to the emergency department studies for suspected
pulmonary embolism on computed tomography pulmonary angiography.
Eur J Radiol 2006;58
: 119–123[CrossRef][Medline]
- Yee J, Kumar NN, Godara S, et al. Extracolonic abnormalities
discovered incidentally at CT colonography in a male population.
Radiology 2005;236
: 519–526[Abstract/Free Full Text]
- Iezzi R, Cotroneo AR, Filippone A, Di Fabio F, Merlino B, Bonomo L.
Extravascular incidental findings at multislice CT angiography of the
abdominal aorta and lower extremity arteries: a retrospective review study.
Abdom Imaging 2007;32
: 489–494[CrossRef][Medline]
- Cham MD, Yankelevitz DF, Henschke CI. Thromboembolic disease
detection at indirect CT venography versus CT pulmonary angiography.
Radiology 2005;234
: 591–594[Abstract/Free Full Text]
- Nchimi A, Ghaye B, Noukoua CT, Dondelinger RF. Incidence and
distribution of lower extremity deep venous thrombosis at indirect computed
tomography venography in patients suspected of pulmonary embolism.
Thromb Haemost 2007;97
: 566–572[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
