DOI:10.2214/AJR.04.1218
AJR 2006; 186:763-773
© American Roentgen Ray Society
Contrast-Enhanced Versus Conventional and Color Doppler Sonography for the Detection of Thrombosis of the Portal and Hepatic Venous Systems
Sandro Rossi1,
Laura Rosa1,
Valentina Ravetta1,
Alessandro Cascina1,
Pietro Quaretti2,
Andrea Azzaretti2,
Paola Scagnelli2,
Carmine Tinelli3,
Paolo Dionigi4 and
Fabrizio Calliada2
1 Department of Internal Medicine VI, IRCCS Policlinico "S. Matteo,"
viale Golgi 19, Pavia 27100, Italy.
2 Department of Radiology, IRCCS Policlinico "S. Matteo," Pavia
27100, Italy.
3 Department of Biometrics and Clinical Epidemiology, IRCCS Policlinico
"S. Matteo," Pavia 27100, Italy.
4 Department of Surgery, IRCCS Policlinico "S. Matteo," Pavia 27100,
Italy.
Received August 2, 2004;
accepted after revision February 3, 2005.
Address correspondence to S. Rossi
(s.rossi{at}smatteo.pv.it).
Abstract
OBJECTIVE. We conducted a prospective study to compare sonography,
color Doppler sonography, and contrast-enhanced sonography for the detection
and characterization of portal and hepatic vein thrombosis complicating
hepatic malignancies.
SUBJECTS AND METHODS. Three hundred sixteen patients with
biopsy-proved hepatic tumors were studied at baseline and 3 months later with
sonography, color Doppler sonography, and contrast-enhanced sonography.
Thrombosis was defined as the presence of intraluminal echogenic material at
sonography, absence of intraluminal color signals at color Doppler sonography,
and presence of nonenhancing intraluminal area at contrast-enhanced
sonography. Thrombi were considered malignant if they displayed continuity
with tumor tissue at sonography, intrathrombus color signals at color Doppler
sonography, and enhancing signals at contrast-enhanced sonography, both having
arterial waveforms at Doppler spectral examination. Definitive diagnoses were
obtained by sonographically guided biopsy except for thrombi displaying at
conventional sonography unequivocal continuity with tumor tissue.
RESULTS. Thrombosis was detected in 79 (25.0%) of 316 patients at
baseline and in 83 (26.3%) of 316 patients after 3 months. Eighty-one (97.6%)
of the 83 thrombi were malignant. Definitive diagnosis was performed by
imaging in 60 (72.3%) of the 83 cases and by biopsy in 23 cases (27.7%). For
thrombus detection, contrast-enhanced sonography displayed significantly
higher sensitivity than color Doppler sonography (p = 0.004) and
borderline superiority over sonography (p = 0.058). For thrombus
characterization, contrast-enhanced sonography was significantly more
sensitive than color Doppler sonography (p < 0.0005) and
conventional sonography (p = 0.02).
CONCLUSION. Contrast-enhanced sonography is superior to sonography
and color Doppler sonography for the detection and characterization of portal
and hepatic vein thrombosis complicating hepatic malignancies.
Keywords: contrast medium • liver cancer • microbubble contrast agent • sonography • thrombosis
Introduction
Hypercoagulability states, myeloproliferative disorders, inflammatory and
neoplastic diseases, portal hypertension, and percutaneous or endoscopic
injection therapies can all be complicated by thrombosis of the portal or
hepatic veins
[1–5].
The detection and etiologic characterization of these thrombi are essential
for treatment planning, particularly in patients with hepatic tumors because
malignant thrombosis is a negative factor in terms of prognosis
[6–8].
Sonography is generally the first imaging technique used in patients with
liver disease, and its ability to detect portal vein thrombosis is reportedly
comparable to that of other imaging techniques
[9–12].
Thrombi appear on sonography as solid intraluminal material that may have a
hypo-, iso-, or hyperechoic pattern
[13]. Findings of this type
are nonspecific, however, and benign and malignant thrombi cannot be
distinguished unless there is evidence of direct intraluminal tumor extension
[9,
13–15].
On color Doppler sonography the presence of intrathrombus signals with
arterial waveforms on Doppler spectral examination is considered a highly
specific sign of thrombus malignancy, although its sensitivity is only
moderate [5,
14]. However, thrombi located
in the left hepatic lobe or deep inside the right lobe cannot be adequately
explored with color Doppler sonography; in cases of this type, sonographically
guided biopsy of the thrombus is the only way to obtain definitive proof of
malignancy [16].
Preliminary clinical studies indicate that contrast-enhanced sonography can
be highly useful for the characterization of focal hepatic lesions based on
their vascularization
[17–20].
The improved visualization of vascular structures offered by contrast-enhanced
sonography and its ability to reveal tiny vessels in the tissue are
characteristics that could be particularly helpful in evaluating portal and
hepatic vein patency and the vascularization of any thrombi that might be
detected. The prospective study described here was a preliminary attempt to
evaluate the usefulness of contrast-enhanced sonography for the detection and
characterization of portal or hepatic vein thrombi in patients with liver
tumors and to compare its performance with those of conventional B-mode
sonography and color Doppler sonography with Doppler spectral examination.
Subjects and Methods
Subjects
The study protocol, which was fully concordant with the ethical principles
of the Declaration of Helsinki, was approved by the institutional ethics
committee. Each participant provided written informed consent for all study
procedures.
Between January 2002 and June 2003, sonography, color Doppler sonography,
and contrast-enhanced sonography studies of the hepatic and portal veins were
performed on 324 consecutive patients with malignant hepatic tumors. All
diagnoses of malignancy had been confirmed by sonographically guided biopsy
using a 21-gauge cutting needle (Biomol, Hospital Service), and all patients
underwent abdominal helical CT for tumor staging. The latter was done using a
third-generation single-detector CT scanner (Somatom Plus, Siemens Medical
Solutions) with conventional technique
[21].
Eight patients whose livers could not be adequately visualized because of
anatomic peculiarities (n = 2) or substantial fatty degeneration
(n = 6) were excluded from the study. The study population was
composed of the remaining 316 patients. They included 199 men and 117 women
who ranged in age from 26 to 88 years (mean ± SD, 67 ± 10
years). Of these patients, 220 had hepatocellular carcinoma (HCC), 14 had
cholangiocarcinoma, and 82 had hepatic metastases. The latter were from
colorectal cancer (n = 52), breast cancer (n = 10), stomach
cancer (n = 6), carcinoid tumors (n = 3), or other
epithelial tumors (n = 11). Multiple tumor nodules were detected in
169 (53.5%) of the 316 patients, and the maximum nodule diameter was greater
than 3.0 cm in 162 patients (51.3%). At recruitment and during their
participation in the study, none of the patients was taking any drugs that
affected coagulation.
Sonography, color Doppler sonography, and contrast-enhanced sonography were
also performed on 10 patients with benign inflammatory bowel disease
(n = 7) or Budd-Chiari syndrome (n = 3). All 10 were known
to have long-standing benign portal thrombosis that had remained unchanged for
at least 1 year. These patients included five men and five women who ranged in
age from 41 to 82 years (mean ± SD, 57 ± 14 years). The purpose
of these studies was to establish with greater certainty the contrast-enhanced
sonography patterns associated with benign thrombi. They were not included in
analyses of the sensitivity and specificity of the three sonography
techniques.
Sonography Techniques and Diagnostic Criteria
All sonography examinations were performed on a Prosound SSD 5500 ePHD
(extended Pure Harmonic Detection) scanner (Aloka) and multi-frequency convex
array transducers (3.0–6.0 MHz) with suitable technology for microbubble
detection—for example, mechanical index (MI) settings of 0.04 or lower,
second harmonic filtering, and phase detection. Each examination was digitally
recorded (Premium Digital Videocassettes and DVCAM model DSR-20 DMP recorder,
both from Sony).
The examination began with a preliminary gray-scale sonography examination
of the upper abdomen. Sonography scans were obtained in the sagittal,
transverse, oblique, and intercostal planes using conventional technique and
second harmonic filtering. The intrahepatic branches of the portal vein,
splenic and mesenteric veins, hepatic veins, and inferior vena cava were then
examined with color Doppler sonography. The color Doppler sonography images,
in which red and blue indicated flow toward and away from the transducer,
respectively, were displayed on-screen with simultaneous B-mode gray-scale or
Doppler spectral examination images. A number of flow settings were used
depending on the underlying flow velocity, and color gain was adjusted during
each examination to select the highest value allowing artifact-free images. If
a thrombus was detected, the Doppler-encoded area was restricted to maximize
color sensitivity and frame rate, and the thrombus was carefully examined for
internal color signals. Any signal detected was subjected to Doppler spectral
examination using a sample volume of 1.5–3.0 mm without angle
correction.
Each patient then underwent contrast-enhanced sonography using an aqueous
suspension of stabilized sulfur fluoride microbubbles (SonoVue, Bracco) as a
contrast agent, in accordance with the manufacturer's instructions. The
product consists of 25 mg of lyophilized powder, which is reconstituted in 5.0
mL of 0.9% sodium chloride solution to produce a solution containing sulfur
hexafluoride microbubbles at a concentration of 8 µL/mL. Shortly after
preparation, this solution was administered as a 2.4-mL bolus via a 19-gauge
IV cannula in an antecubital vein. The injection was immediately followed by a
bolus of 5.0 mL of 0.9% sodium chloride solution. A chronometer displayed on
the screen was used to determine the temporal characteristics of flow
enhancement. Contrast-enhanced sonography scans were obtained in a harmonic
mode with an MI of 0.04 or less. If a thrombus was detected
(Fig. 1), up to two additional
boluses of SonoVue were injected, and the thrombus was examined more closely
for pulsating, enhancing signals within its boundaries
(Fig. 2A). When present, these
signals were subjected to Doppler spectral examination, as described
(Fig. 2B).
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —Contrast-enhanced sonography appearance of portal vein thrombi in
76-year-old man. Intercostal scan with longitudinal view of portal vein during
portal phase reveals enhanced lumen containing two nonenhanced areas that are
nonocclusive thrombi.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —Doppler spectral examination in 52-year-old man of enhancing
intrathrombotic signals detected during contrast-enhanced sonography. Oblique
subcostal scan with longitudinal view of common trunk of portal vein during
portal phase of contrast-enhanced sonography. Within enhanced lumen,
nonenhancing area (thrombus) measuring about 4.0 mm in diameter can be seen
adhering to wall of vein. At center of thrombus, pulsating punctate enhancing
signal reflects intrathrombotic vascularization.
|
|
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —Doppler spectral examination in 52-year-old man of enhancing
intrathrombotic signals detected during contrast-enhanced sonography. Doppler
spectral examination of pulsating punctate enhancing signal within thrombus
reveals arterial waveform.
|
|
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —Doppler spectral examination in 52-year-old man of enhancing
intrathrombotic signals detected during contrast-enhanced sonography. Helical
CT scans. In arterial phase (C), hyperdense area that represents
hepatocellular carcinoma (HCC) nodule is seen in segment IV of liver. On same
scan, in portal (D) and late (E) phases, branch of portal vein
adjacent to HCC nodule appears to be fully patent, with no sign of small
neoplastic thrombus seen on contrast-enhanced sonography.
|
|
View larger version (157K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —Doppler spectral examination in 52-year-old man of enhancing
intrathrombotic signals detected during contrast-enhanced sonography. Helical
CT scans. In arterial phase (C), hyperdense area that represents
hepatocellular carcinoma (HCC) nodule is seen in segment IV of liver. On same
scan, in portal (D) and late (E) phases, branch of portal vein
adjacent to HCC nodule appears to be fully patent, with no sign of small
neoplastic thrombus seen on contrast-enhanced sonography.
|
|
View larger version (155K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2E —Doppler spectral examination in 52-year-old man of enhancing
intrathrombotic signals detected during contrast-enhanced sonography. Helical
CT scans. In arterial phase (C), hyperdense area that represents
hepatocellular carcinoma (HCC) nodule is seen in segment IV of liver. On same
scan, in portal (D) and late (E) phases, branch of portal vein
adjacent to HCC nodule appears to be fully patent, with no sign of small
neoplastic thrombus seen on contrast-enhanced sonography.
|
|
At the end of the baseline examination, digitally recorded images obtained
with each of the three sonography techniques were subjected to independent
frame-by-frame review by a team of two reviewers who have been trained in and
are experienced with Doppler techniques. For each method, a diagnosis of
thrombosis or nonthrombosis was recorded; in the former case, the thrombus was
also diagnosed as benign or malignant. All diagnoses represent consensus
decisions reached by the two reviewers and were based on the criteria shown in
Table 1.
View this table:
[in this window]
[in a new window]
|
TABLE 1: Diagnostic Criteria Used for Detection and Characterization of Portal
and Hepatic Vein Thrombosis by Different Sonography Techniques
|
|
In cases with sonographic diagnoses of thrombosis at baseline, we also
analyzed the helical CT scans obtained for tumor-staging purposes. This
analysis was limited to those cases in which the baseline sonography studies
and the helical CT examination had been performed within 5 days of one
another. The scans that satisfied these criteria were reviewed by two
experienced radiologists. Using the criteria recommended by Tublin et al.
[22], the reviewers classified
each CT scan as positive or negative for thrombosis
(Fig. 3). The sole purpose of
this analysis was to obtain a preliminary estimate of the relative sensitivity
of CT in detecting portal or hepatic vein thrombosis. The CT diagnoses had no
impact on the diagnoses of thrombosis made by sonography, color Doppler
sonography, or contrast-enhanced sonography or on the definitive
diagnoses.
Three months after the baseline examinations, sonography, color Doppler
sonography, and contrast-enhanced sonography were repeated on all patients to
verify the initial diagnosis. The second examination was performed and
evaluated as described.
Definitive Diagnoses of Thrombosis and Thrombus Malignancy
At the end of the study, patients were definitively classified as negative
for thrombosis if none of the three sonography techniques had revealed any
evidence of thrombosis at the baseline or 3-month study. In the remaining
cases, one or more of the baseline or 3-month examinations was positive for
thrombi. In these cases, the definitive diagnoses of thrombosis and thrombus
malignancy were based on sonographically guided aspiration biopsy performed
using a 21-gauge Chiba needle (Ecojekt, Hospital Service)
[16]
(Fig. 4). The exception to this
rule was for thrombi that showed unequivocal evidence of continuity with the
tumor tissue (Fig. 5). In these
cases the nature of the thrombus was assumed to be the same as that of the
biopsied tumor. This feature is widely accepted in CT and MRI studies as
sufficient proof of portal thrombus malignancy
[14,
15].
View larger version (104K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5 —Intercostal sonography scan reveals direct extension of
hepatocellular carcinoma tissue into main portal vein in 71-year-old man. This
finding represents unequivocal evidence of continuity between thrombus and
tumor tissue and is a reliable indicator of thrombus malignancy.
|
|
Statistical Calculations
The sonography, color Doppler sonography, and contrast-enhanced sonography
diagnoses regarding thrombus detection and characterization made using the
criteria in Table 1 were
checked against the definitive diagnoses performed as described, and the
sensitivity [true-positives / (true-positives + false-negatives)] and
specificity [true-negatives / (false-positives + true-negatives)] of each
technique was computed. In both cases, exact 95% confidence intervals (CIs)
were calculated using software (version 7.0, Stata Statistical Software 2002,
Stata Corp.). Chi-square and Fisher's exact tests were used to evaluate
differences in the sensitivity and specificity of the different techniques, as
appropriate [23]. A p
value of less than 0.05 was considered indicative of statistical significance.
All tests were two-sided.
Results
Thrombus Detection
Portal or hepatic vein thrombosis was definitively excluded in 233 (73.7%)
of the 316 patients, and definitive diagnoses of thrombosis were made in the
remaining 83 patients (26.3%): 76 (34.5%) of 220 with HCC; three (21.4%) of 14
with cholangiocarcinoma; and four (4.9%) of 82 with metastases. Thrombosis was
significantly more common in patients with tumor nodules larger than 3.0 than
in those with smaller nodules (72/162 [44.4%] vs 11/154 [7.1%], respectively;
p < 0.00001). The portal vein or its collaterals were involved in
78 (94.0%) of 83 patients with thrombosis, including 73 (96.0%) of the 76 with
HCC; two (66.7%) of the three with cholangiocarcinoma; and three (75.0%) of
the four with metastases. Only five (6.0%) of the 83 thrombi involved the
hepatic veins: three (3.9%) of 76 found in patients with HCC, one (33.3%) of
three patients with cholangiocarcinoma, and one (25.0%) of four in patients
with metastases. Fifty-five (66.3%) of the 83 thrombi were occlusive and 60
(72.3%) were continuous with the tumor tissue.
Seventy-nine (95.2%) of the 83 thrombi were detected during the baseline
examination, using the diagnostic criteria listed in
Table 1: 72 (86.7%) of the 83
thrombi were detected on sonography; 67 (80.7%) of the 83, on color Doppler
sonography; and 79 (95.2%) of the 83, on contrast-enhanced sonography. The
remaining four thrombi (4.8%) were first detected on one or more of the
sonography studies performed during the 3-month examination. On the hypothesis
that these thrombi might have been present at the baseline examination in a
smaller, less detectable form, we classified these four cases as
false-negative detections for all three sonography techniques. There were no
false-positive detections.
As shown in Figure 6,
contrast-enhanced sonography was more sensitive in thrombus detection than
either color Doppler sonography (p = 0.004) or sonography (p
= 0.058), although the latter displayed only borderline significance. With
respect to color Doppler sonography, contrast-enhanced sonography provided
significantly higher sensitivity for the detection of thrombi with and without
continuity with the tumor tissue (p < 0.05) and occluding and
nonoccluding thrombi (p < 0.05). All three sonography techniques
were characterized by 100% specificity.
Forty-eight of the 79 patients whose thrombi were detected at the baseline
sonography examination had undergone helical CT within 5 days of sonography
studies. Review of these 48 CT scans revealed evidence of thrombosis in only
20 (41.7%). The thrombi visualized on the CT scans included two (11.1%) of 18
that had appeared to be nonocclusive on sonographic studies; 18 (60%) of 30
that were fully occlusive; 13 (35.1%) of 37 that showed clear continuity with
the tumor tissue; and seven (63.6%) of 11 that were noncontinuous with the
tumor tissue.
Thrombus Characterization
Eighty-one (97.6%) of the 83 thrombi were definitively classified as
malignant. In 60 (74.1%) of 81 cases, this diagnosis was based on clear
sonography evidence of thrombus continuity with the tumor tissue. In these 60
thrombi, color Doppler sonography revealed intrathrombus vascular signals in
52 cases (86.7%) but Doppler spectral examination visualization of arterial
waveforms was possible in only 47 (78.3%) of 60 cases. In contrast,
contrast-enhanced sonography detected pulsating enhanced signals whose
arterial waveforms were confirmed by Doppler spectral examination in 60 of 60
cases. In the remaining 21 (25.9%) of 81 cases, the diagnosis of thrombus
malignancy was based on sonographically guided biopsy. All 21 of these thrombi
were visualized on contrast-enhanced sonography and 14 (66.7%) of 21 on color
Doppler sonography, but only four (19.0%) of 21 met the criteria for
malignancy on both sonography studies. In the other 17 (81.0%) of 21 cases,
color Doppler sonography failed to reveal any color signals within any of the
10 thrombi it detected; in contrast, contrast-enhanced sonography detected
enhancing pulsating signal within 11 (64.7%) of 17 thrombi and no enhancing
pulsating signals in the remaining six (35.3%).
View larger version (35K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8 —Schematic representations of thrombus patterns observed on
contrast-enhanced longitudinal sonography scan (A) of venous vessel and
orthogonal sonography scan (B) of venous vessel. = Pattern 2 can evolve into
pattern 1, 3, or 4 during portal and late phases. White lines into thrombus
represent visible arterial neovascularization. NO = nonocclusive thrombi, O =
occlusive thrombi.
|
|
The other two (2.4%) of 83 thrombi were ultimately classified as benign. In
both cases, the definitive diagnosis was based on cytologic examination of
sonographically guided biopsies because neither of these thrombi had presented
continuity with the tumor tissue on gray-scale sonography. In one case, the
color Doppler sonography and contrast-enhanced sonography findings at baseline
and 3 months were also negative for thrombus malignancy. In contrast, the
second benign thrombus met both the color Doppler sonography and
contrast-enhanced sonography criteria for malignancy
(Table 1) on the baseline
examination, in clear contradiction with the biopsy, which showed no cytologic
evidence of tumor cells. Three months later, a repeat contrast-enhanced
sonography examination revealed that the vascular signals, which had
previously appeared to be intrathrombotic, were actually related to a small
branch of the left hepatic artery crossing the portal vein at the site of the
thrombus. This case was thus classified as a false-positive characterization
for color Doppler sonography and contrast-enhanced sonography.
The relative sensitivities of the three sonography techniques in thrombus
characterization are shown in Figure
7. Contrast-enhanced sonography was significantly superior to
color Doppler sonography for the characterization of malignant thrombi
(p < 0.0005). Compared with sonography, it displayed higher
sensitivity for identifying malignancy in all thrombi (p = 0.02) and
in those without continuity with tumor tissue (p < 0.001). Because
only two thrombi proved to be benign, the specificities of the three
sonography techniques could not be determined in this study.
Thrombus Patterns on Contrast-Enhanced Sonography
Figure 8 shows schematic
representations of the different thrombus patterns detected on
contrast-enhanced sonography both in patients with liver tumors and in those
with long-standing benign portal thrombosis. Clinical examples of each pattern
are provided in Figures 9A,
9B,
10A,
10B,
11A,
11B,
11C,
12A,
12B,
13A, and
13B. Pattern 1 is typical of
thrombi without internal vascularization—that is, benign thrombosis
(Figs. 9A,
9B,
10A, and
10B). Thrombus enhancement is
absent in all three phases. Pattern 2 (blooming) consists of diffuse thrombus
enhancement that is visible only during the early arterial phase. It reflects
diffuse vascularization of a thrombus similar to that of the tumor tissue from
which it originates (Figs.
11A,
11B, and
11C). Patterns 3 (linear or
punctate) and 4 (multilinear or multipunctate) can be observed during either
the arterial or portal and late phases of contrast-enhanced sonography and are
indicative of thrombus vascularization (Figs.
12A,
12B,
13A, and
13B). The
linear-versus-punctate appearance depends on the orientation of the
intrathrombotic vessel or vessels with respect to the scan angle. Patterns may
be combined: For example, in the arterial phase, certain areas of a thrombus
may display diffuse enhancement (pattern 2), whereas other areas present
linear or punctate enhancement (patterns 3 or 4). The latter enhancement
generally persists during the portal and late phases after the blooming has
disappeared. In all patterns, during the portal and late phases, luminal
enhancement appears around the thrombus if the latter is nonocclusive
(Fig. 1); if the thrombus is
occlusive, only the proximal lumen will be enhanced (Fig. 13C).
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A —Nonenhancing contrast-enhanced sonography pattern in benign
nonocclusive thrombus in 48-year-old woman. Oblique sonography scan with
longitudinal view of stent inserted between portal vein and inferior vena cava
for treatment of Budd-Chiari syndrome.
|
|
View larger version (90K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B —Nonenhancing contrast-enhanced sonography pattern in benign
nonocclusive thrombus in 48-year-old woman. Oblique contrast-enhanced
sonography scan with longitudinal view of stent. During portal phase of
contrast-enhanced sonography, uniformly nonenhancing area about 5.0 mm in
diameter is visible within enhanced lumen of stent.
|
|
View larger version (138K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10A —Nonenhancing contrast-enhanced sonography pattern in benign
occlusive thrombus in 63-year-old man. Simultaneous intercostal
(left) and color Doppler (right) sonography scans with
longitudinal view of portal vein. On sonography, lumen of portal vein is
completely filled with hypoechoic material representing occlusive thrombus.
Color Doppler sonography revealed no color signals within lumen of vein or
within thrombus.
|
|
View larger version (96K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10B —Nonenhancing contrast-enhanced sonography pattern in benign
occlusive thrombus in 63-year-old man. Intercostal contrast-enhanced
sonography scan with longitudinal view of portal vein. During portal phase,
uniformly nonenhancing area perfectly reproducing shape of vessel is visible
within portal vein.
|
|
View larger version (140K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11A —Blooming contrast-enhanced sonography pattern evolving into
multipunctate pattern in intensely vascularized malignant thrombus in
75-year-old man. Simultaneous oblique (left) and color Doppler
(right) sonography scans with longitudinal view of left branch of
portal vein. Sonography reveals echogenic area (thrombus) within vessel lumen.
Color signals are visible within thrombus.
|
|
View larger version (71K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11B —Blooming contrast-enhanced sonography pattern evolving into
multipunctate pattern in intensely vascularized malignant thrombus in
75-year-old man. Oblique contrast-enhanced sonography scan with longitudinal
view of portal vein during arterial phase. Within poorly enhanced lumen of
left branch of portal vein is diffusely enhanced area (arrows)
representing thrombus with rich internal neovascularity (blooming
pattern).
|
|
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11C —Blooming contrast-enhanced sonography pattern evolving into
multipunctate pattern in intensely vascularized malignant thrombus in
75-year-old man. During portal phase of contrast-enhanced sonography scan
shown in B, lumen of portal vein is enhanced and thrombus appears as
predominantly nonenhancing area (arrows) containing multiple punctate
enhancing signals.
|
|
View larger version (163K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12A —Pulsating linear or punctate contrast-enhanced sonography pattern in
occlusive malignant thrombus in 63-year-old woman. Oblique color Doppler
sonography scan with longitudinal view of left branch of portal vein. Vessel
lumen is completely filled with hypoechoic material. No color signals were
detected within lumen of portal vein or within thrombus.
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12B —Pulsating linear or punctate contrast-enhanced sonography pattern in
occlusive malignant thrombus in 63-year-old woman. Oblique contrast-enhanced
sonography scan with longitudinal view of left branch of portal vein during
arterial phase. Thrombus appears as predominantly nonenhancing area that
reproduces shape of vessel lumen; within its borders is pulsating linear
signal, which is indicative of arterial neovascularization.
|
|
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13A —Pulsating multilinear or multipunctate contrast-enhanced sonography
pattern in occlusive malignant thrombus in 68-year-old man. Simultaneous
oblique subcostal (left) and color Doppler (right)
sonography scans show longitudinal views of right branch of portal vein.
Sonography scan reveals isoechoic area within vessel lumen (arrow)
representing thrombus; on color Doppler sonography, some color signals are
visible within boundaries of thrombus.
|
|
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13B —Pulsating multilinear or multipunctate contrast-enhanced sonography
pattern in occlusive malignant thrombus in 68-year-old man. Oblique subcostal
contrast-enhanced sonography scan with longitudinal view of right branch
portal vein during initial portal phase. Thrombus (arrow) appears as
predominantly nonenhancing area reproducing shape of vessel lumen. Multiple
pulsating punctate and linear signals indicative of arterial
neovascularization are visible within its borders. Only proximal lumen of
portal vein appears enhanced.
|
|
All 10 patients with long-standing benign thrombi presented pattern 1 in
both the arterial and portal phases of the contrast-enhanced sonography
examination. However, this pattern was also detected in six cases of malignant
thrombosis (false-negative contrast-enhanced sonography characterizations).
The other 75 thrombi definitively classified as malignant presented pattern 2,
3, or 4. Their relative frequencies in the arterial and portal phases are
reported in Table 2.
View this table:
[in this window]
[in a new window]
|
TABLE 2: Patterns Displayed on Contrast-Enhanced Sonography by the 81 Malignant
Portal or Hepatic Vein Thrombi Detected in Patients with Hepatic
Tumors
|
|
Contrast-Enhanced Sonography Characteristics and Complications
The mean duration of the contrast-enhanced sonography examination,
including Doppler spectral examination when possible, was 15 min (range,
11–21 min), and each patient received from one to three boluses of
SonoVue (mean, 2.5 boluses). The mean estimated cost of each contrast-enhanced
sonography study was 
140, including the costs of contrast medium and
digital cassettes, use of the sonography room, and medical and nursing staff
and the amortization of sonography and recording equipment.
Portal blood flow enhancement was evident 22 ± 4 sec (mean ±
SD) after injection of the contrast agent and remained at useful levels for a
mean of 240 ± 30 sec; enhanced flow in the hepatic veins appeared after
22 ± 5 sec and remained useful for a mean of 180 ± 25 sec. When
present, enhanced pulsating intrathrombus signals appeared 14 ± 5 sec
after SonoVue injection and persisted for 175 ± 30 sec.
No major complications were observed. One patient experienced mild
transient hyperemia at the site of IV cannula (antecubital vein) that was
probably due to accidental extravasation of contrast medium.
Discussion
Our findings in the present study confirm that portal vein thrombosis is a
frequent complication of primary liver tumors, especially those with diameters
larger than 3.0 cm [24]. In
patients with HCC, these thrombi are almost always malignant and, in most
cases, direct invasion of the veins by neoplastic tissue can be
sonographically shown.
In this preliminary clinical experience, contrast-enhanced sonography
appears to be a reliable technique for evaluating the patency of the veins of
the portal and hepatic systems. It was significantly more sensitive than color
Doppler sonography for both the detection and characterization of thrombi, and
it was also more sensitive than conventional sonography for thrombus
characterization. In overall thrombus detection, the superiority of
contrast-enhanced sonography over sonography displayed only borderline
significance. Nonetheless, it allowed detection of thrombi that were missed on
conventional sonography (n = 7) and on color Doppler sonography
(n = 12). In fact, contrast-enhanced sonography was significantly
more sensitive for the detection of nonocclusive thrombi, which represented
fewer than 30% of those found in our study population. In a larger patient
series, it is thus likely that the superiority of contrast-enhanced sonography
over sonography would have been more significant. Furthermore, the detection
sensitivity of contrast-enhanced sonography reported here may well be
underestimated. For the four thrombi detected for the first time during the
second examination, the baseline diagnoses were classified as false-negatives,
although it is possible that some of these delayed diagnoses were actually new
thrombi that had not been present 3 months earlier. When these four cases were
reclassified as true-negatives for all three techniques, contrast-enhanced
sonography still proved to be superior to both color Doppler sonography and
conventional sonography.
The high sensitivity of contrast-enhanced sonography in thrombus detection
has already been noted with the use of first-generation microbubble contrast
agents [15,
25]. Second-generation agents,
such as SonoVue, produce an even better optical gradient between vessel
lumens, which are strongly enhanced, and thrombi, which appear as nonenhancing
areas. Furthermore, because the signals are generated by the oscillation,
rather than the rupture, of the bubbles induced by low-MI insonation,
real-time imaging can be achieved. The useful enhancement lasts for 3–4
min, during which an experienced sonographer can easily examine all portal and
hepatic veins. No other imaging technique currently offers an observation
period of this length. These factors markedly improve diagnostic yield,
allowing detection of thrombi with diameters of approximately 3.0 mm and of
hypoechoic thrombi, which are difficult to visualize with conventional
sonography and are not always detectable even with Doppler techniques. In
addition, the absence of artifacts caused by heart-related movement greatly
facilitates exploration of vessels in the left hepatic lobe and segment VIII
of the liver, which cannot be adequately studied with color Doppler sonography
[26].
As for thrombus characterization, the main advantage of contrast-enhanced
sonography over conventional sonography techniques is its ability to detect
even tiny vessels within tissue. Arterial neovascularization within a
neoplastic thrombus appears as enhancing signals, which are easily
distinguishable from those of venous flow by their pulsation. The high
visibility of these signals also facilitates Doppler spectral examination
confirmation of their arterial nature. In fact, Doppler spectral examination
during color Doppler sonography was sometimes impossible because of the
location of the thrombus (e.g., left hepatic lobe or deep in the right hepatic
lobe); in other cases, it was difficult because of a lack of patient
cooperation. Doppler spectral examination during contrast-enhanced sonography
was successful in a significantly higher percentage of cases. The only
false-positive contrast-enhanced sonography characterization was caused by the
presence of vascular signals from a small branch of the left hepatic artery
overlaying the thrombosed segment of the portal vein, and the same error was
also made on color Doppler sonography. Errors of this type have also been
reported by other groups
[5].
Helical CT is frequently proposed for noninvasive assessment of the portal
venous system [22], although
its sensitivity in thrombus characterization is reportedly low and its
comparative efficacy has never been evaluated, to our knowledge, in
large-scale prospective studies
[27]. Our experience in this
study suggests that helical CT may be less sensitive than contrast-enhanced
sonography or even sonography for the detection of venous thrombosis in the
liver. However, it is important to recall that, although state-of-the-art
sonography equipment was used in this study, the CT studies were performed
using a third-generation single-slice helical CT scanner, and in any case the
two methods were compared in only a limited number of cases.
One of the main limitations of our study is the absence of pathologic
confirmation for some diagnoses. Sonographically guided biopsy is undoubtedly
the best method for characterizing malignant thrombosis
[16]. It was deferred in those
60 cases in which gray-scale sonography revealed unequivocal evidence of
continuity between thrombus and tumor tissue on both the baseline and 3-month
studies. CT or MRI visualization of thrombus–tumor continuity is widely
accepted as a reliable indicator of thrombus malignancy
[14,
15], and we think that there
is no reason to assume that the significance of this finding is different when
it is clearly shown by sonography. Furthermore, in most cases,
thrombus–tumor continuity was accompanied by clear color Doppler
sonography–Doppler spectral examination evidence of intrathrombus
vascularization, which is associated with a specificity of close to 100%
[5,
14]. For these reasons, we
think that all of the definitive diagnoses reported in this study can be
considered fully reliable.
Our results were obtained with state-of-the-art equipment, which may not be
available in many centers. This factor might explain the excellent performance
of gray-scale sonography and second harmonic filtering, which, in our study,
was surprisingly almost as informative as color Doppler sonography. Moreover,
the success of the contrast-enhanced sonography examinations, like other
sonographic studies, depends on the skill, experience, and motivation of the
sonographers and on the characteristics of the patient. Used correctly,
however, the technique we evaluated seems to offer several potential
advantages. It not only improves detection of small thrombi, but can also be
used to reliably identify malignant thrombi even when there is no continuity
with tumor tissue. Intrathrombus neovascularity can certainly be shown with
conventional color Doppler sonography and Doppler spectral examination, and if
this is the case, there is no indication for contrast-enhanced sonography.
However, this process is considerably easier in the presence of contrast
enhancement. Furthermore, the absence of artifacts related to heart movements
facilitates exploration of thrombi located in left hepatic lobe, which are
notoriously difficult to examine with color Doppler sonography. With respect
to Doppler spectral examination performed during color Doppler sonography,
Doppler spectral examination performed during contrast-enhanced sonography can
thus eliminate the need for biopsy in a considerably higher percentage of
cases.
Preliminary studies have already highlighted the potential value of
contrast-enhanced sonography in diagnosing primary hepatic tumors and in
detecting liver metastases
[18–20].
Although thrombosis is less common with the secondary hepatic lesions,
contrast-enhanced sonography exploration of the portal and hepatic veins could
also be used in cases of this type to provide more complete staging. The
higher cost of contrast-enhanced sonography should therefore be weighed
against potential benefits in terms of the elimination of additional
diagnostic procedures, particularly invasive ones such as liver biopsy, and
also the avoidance of inappropriate therapies.
In conclusion, our preliminary findings are promising and suggest an
important future role for contrast-enhanced sonography in staging hepatic
tumors. However, further prospective blinded studies are necessary to confirm
its apparent superiority to conventional sonography studies and to compare it
with helical CT and MRI, which also appear to be promising tools
[28] for the assessment of the
veins of the portal and hepatic systems.
References
- McNamara C, Juneja S, Wolf M, Grigg A. Portal or hepatic vein
thrombosis as the first presentation of a myeloproliferative disorder in
patients with normal peripheral blood counts. Clin Lab
Haem 2002; 24:239
-242[CrossRef][Medline]
- Jansen HLA, Garcia-Pagan JC, Elias E, Mentha G, Hadengue A, Valla
DC. Budd-Chiari syndrome: a review by an expert panel. J
Hepatol 2003; 38:364
-371[CrossRef][Medline]
- [No authors listed]. Primary liver cancer in Japan:
clinicopathologic features and results of surgical treatment. Liver Cancer
Study Group of Japan. Ann Surg 1990;211
: 277-287[Medline]
- Atri M, de Stempel J, Bret PM, Illescas FF. Incidence of portal
vein thrombosis complicating liver metastasis as detected by duplex
ultrasound. J Ultrasound Med 1990;9
: 285-289[Abstract]
- Lencioni R, Caramella D, Sanguinetti F, Battolla L, Falaschi F,
Bartolozzi C. Portal vein thrombosis after percutaneous ethanol injection for
hepatocellular carcinoma: value of color Doppler sonography in distinguishing
chemical and tumor thrombi. AJR 1995;164
: 1125-1130[Abstract/Free Full Text]
- Tessler FN, Gehring BJ, Gomes AS, et al. Diagnosis of portal vein
thrombosis: value of color Doppler imaging. AJR1991; 157:293
-296[Abstract/Free Full Text]
- Janssen HLA, Wijnhoud A, Haagsma EB, et al. Extrahepatic portal
vein thrombosis: aetiology and determinants of survival.
Gut 2001; 49:720
-724[Abstract/Free Full Text]
- Nonami T, Yokoyama I, Iwatsuki S, Starzl TE. The incidence of
portal vein thrombosis at liver transplantation.
Hepatology 1992;16
: 1195-1198[CrossRef][Medline]
- Tanaka K, Numata K, Okazaki H, Nakamura S, Inoue S, Takamura Y.
Diagnosis of portal vein thrombosis in patients with hepatocellular carcinoma:
efficacy of color Doppler sonography compared with angiography.
AJR 1993; 160:1279
-1283[Abstract/Free Full Text]
- Finn JP, Kane RA, Edelman RR, et al. Imaging of the portal venous
system in patients with cirrhosis: MR angiography versus duplex Doppler
sonography. AJR 1993;161
: 989-994[Abstract/Free Full Text]
- Naik KS, Ward J, Irving HC, Robinson PJ. Comparison of dynamic
contrast-enhanced MRI and Doppler ultrasound in the pre-operative assessment
of the portal venous system. Br J Radiol1997; 70:43
-49[Abstract]
- Bach AM, Hann LE, Brown KT, et al. Portal vein evaluation with US:
comparison to angiography combined with CT arterial portography.
Radiology 1996;202
: 149-154
- Van Gansbeke D, Avni EF, Delcour C, Engelholm L, Struyven J.
Sonographic features of portal vein thrombosis. AJR1985; 144:749
-752[Abstract/Free Full Text]
- Dodd GD 3rd, Memel DS, Baron RL, Eichner L, Santiguida LA. Portal
vein thrombosis in patients with cirrhosis: does sonographic detection of
intrathrombus flow allow differentiation of benign and malignant thrombus?
AJR 1995; 165:573
-577[Abstract/Free Full Text]
- Ricci P, Cantisani V, Biancari F, et al. Contrast-enhanced color
Doppler US in malignant portal vein thrombosis. Acta
Radiol 2000; 41:470
-473[CrossRef][Medline]
- Vilana R, Bru C, Bruix J, Castells A, Sole M, Rodes J. Fine-needle
aspiration biopsy of portal vein thrombosis: value in detecting malignant
thrombosis. AJR 1993;160
: 1285-1287[Abstract/Free Full Text]
- Correas JM, Bridal L, Lesavre A, Méjean A, Claudon M,
Hélénon O. Ultrasound contrast agents: properties, principles of
action, tolerance, and artifacts. Eur Radiol2001; 11:1316
-1328[CrossRef][Medline]
- Leen E, Angerson WJ, Yarmenitis S, et al. Multi-centre clinical
study evaluating the efficacy of SonoVue (BR1), a new ultrasound contrast
agent in Doppler investigation of focal hepatic lesions. Eur J
Radiol 2002; 41:200
-206[CrossRef][Medline]
- Hohmann J, Albrecht T, Hoffmann CW, Wolf KJ. Ultrasonographic
detection of focal liver lesions: increased sensitivity and specificity with
microbubble contrast agent. Eur J Radiol2003; 46:147
-159[CrossRef][Medline]
- Quaia E, Calliada F, Bertolotto M, et al. Characterization of focal
liver lesions with contrast-specific US modes and a sulfur
hexafluoride–filled microbubble contrast agent: diagnostic performance
and confidence. Radiology 2004;232
: 420-430[Abstract/Free Full Text]
- Rossi S, Garbagnati F, Lencioni R, et al. Percutaneous
radio-frequency thermal ablation of nonresectable hepatocellular carcinoma
after occlusion of tumor blood supply. Radiology2000; 217:119
-126[Abstract/Free Full Text]
- Tublin ME, Dodd GD 3rd, Baron RL. Benign and malignant portal vein
thrombosis: differentiation by CT characteristics. AJR1997; 168:719
-723[Abstract/Free Full Text]
- Campbell MJ, Machin D. Medical statistics: a commonsense
approach, 3rd ed. Chichester, England: Wiley, 1999:6
: 85-89, Appendix
I:155-158
- Calvet X, Bruix J, Bru C, et al. Natural history of hepatocellular
carcinoma in Spain: five years' experience in 249 cases. J
Hepatol 1990; 10:311
-317[CrossRef][Medline]
- Marshall MM, Beese RC, Muiesan P, Sarma DI, O'Grady J, Sidhu PS.
Assessment of portal venous system patency in the liver transplant candidate:
a prospective study comparing ultrasound, microbubble-enhanced colour Doppler
ultrasound, with arteriography and surgery. Clin
Radiol 2002; 57:377
-383[CrossRef][Medline]
- Choi BI, Kim TK, Han JK, Chung JW, Park JH, Han MC. Power versus
conventional color Doppler sonography: comparison in depiction of vasculature
in liver tumors. Radiology 1996;200
: 55-58[Abstract/Free Full Text]
- Taylor CR. Computed tomography in the evaluation of the portal
venous system. J Clin Gastroenterol 1992;14
: 167-172[Medline]
- Kreft B, Strunk H, Flacke S, et al. Detection of thrombosis in the
portal venous system: comparison of contrast-enhanced MR angiography with
intraarterial digital subtraction angiography.
Radiology 2000;216
: 86-92[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
A. H. M. Caiado, R. Blasbalg, A. S. Z. Marcelino, M. da Cunha Pinho, M. C. Chammas, C. da Costa Leite, G. G. Cerri, A. C. de Oliveira, T. Bacchella, and M. C. C. Machado
Complications of Liver Transplantation: Multimodality Imaging Approach
RadioGraphics,
September 1, 2007;
27(5):
1401 - 1417.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
