Keywords hepatic angiomyolipoma - hepatocellular carcinoma - contrast-enhanced ultrasound - Sonazoid
Background
Hepatic angiomyolipoma (HAML) is an uncommon mesenchymal liver tumor consisting of
smooth muscle cells, adipose tissue, and thick-walled blood vessels. It lacks
distinct features on clinical examination or laboratory results. For instance,
hepatitis B virus (HBV) infection often linked with other liver tumors shows no
clear relationship with HAML and alpha-fetoprotein (AFP) levels usually remain low
(AFP<15ng/ml) [1 ]
[2 ]. However, due to its benign nature and rich
blood supply, HAML should be distinguished from hepatocellular carcinoma (HCC),
which is the most common primary malignant liver cancer.
Typically, HCC presents distinctive clinical and laboratory traits, including a
significant correlation with HBV and elevated AFP levels, facilitating
straightforward diagnosis. Nonetheless, HCC with negative hepatitis virus and low
AFP levels (non-viral AFP- HCC) can easily be mistaken for HAML. The management
strategies of these two types of tumors differ significantly. Conservative treatment
is generally recommended for HAML, while HCC, as an aggressive cancer, necessitates
more comprehensive treatments like surgery and transhepatic arterial chemotherapy
[3 ]
[4 ]. In recent years, we have observed a global increase in HCC cases in
non-hepatitis individuals [5 ]
[6 ]. A survey from Japan reported that the
proportion of HCC patients with non-viral etiologies has continued to increase from
10.0% in 1991 to 32.5% in 2015 [7 ]. Among
individuals without HCC risk factors, HCC appears as the most frequent primary
malignant liver tumor [8 ]. Notably, non-viral
HCC cases with a low level of AFP may be relatively rarer, but it can be quite
easily confused with benign hyper-vascularized HAML. Therefore, understanding their
imaging features and distinguishing them carefully is extremely significant.
Imaging plays a crucial role in the diagnostic process as a noninvasive approach.
Contrast-enhanced ultrasound (CEUS) has demonstrated remarkable accuracy in
detecting focal liver lesions with high sensitivity and specificity [9 ]. SonoVue (SV; sulfur hexafluoride; Bracco
SpA, Milan, Italy) and Sonazoid (SZ; perflubutane; GE Healthcare, Oslo, Norway) are
second-generation ultrasound contrast agents. SV is a pure blood agent; SZ exhibits
an extra post-vascular phase (10 minutes after injection and lasting at least 1
hour) with a high affinity for Kupffer cells in the liver [10 ]. Previously, several research studies
described conventional ultrasound (US) and SV-CEUS manifestations of HAML in detail
[11 ]
[12 ]
[13 ]. However, it appears that
fewer studies have explored SZ-CEUS features of HAML. Hence, it is essential to
investigate SZ-CEUS traits of HAML to improve accurate diagnosis.
This retrospective study aimed to compare CEUS features between HAML and challenging
cases of HCC, mainly those with no hepatitis infection but also with a low level of
AFP (non-viral AFP- HCC). We also compared the CEUS features of two different
contrast agents, Sonovue and Sonazoid.
Methods
Patients
This retrospective study was approved by the institutional review board of our
hospital and informed consent was waived.
The inclusion procedures for the HAML group in our study involved conducting an
extensive search of the pathological databases at our hospital using the keyword
“hepatic angiomyolipoma” between January 2012 and May 2023. The inclusion
criteria for the HAML group were: 1) pathologic diagnosis of HAML confirmed by
resection or biopsy; 2) CEUS performed before any treatment; 3) clinical and
imaging data were available and of good quality. After adhering to these
criteria, we identified and included 50 patients (17 men, 37 women, aged 41±11.3
years) with pathologically confirmed cases of HAML. Of these patients, 12 (24%)
underwent SZ-CEUS.
For the non-viral AFP- HCC group, we conducted a similarly comprehensive search
for patients that met our criteria. The search spanned January 2012 to May 2023
using the keyword “hepatocellular carcinoma”. Non-viral AFP- HCC participants
were required to meet the following criteria: 1) resection or biopsy confirmed
HCC pathology; 2) AFP levels were not higher than 15ng/ml; 3) hepatitis B
surface antigen was negative; 4) hepatitis C antibody was negative; 5) images
and clinical data were available and of good quality. Based on these criteria,
we identified and included 88 pathologically confirmed HCC patients (81 men, 7
women, aged 61.4±11.2 years). SZ-CEUS was performed on 19 of these 88 patients
(21.6%). The flowchart in [Fig.1 ] shows
how patients were selected.
Fig. 1 Flowchart of the selection of patients. The patients with
hepatic angiomyolipoma (HAML) and hepatocellular carcinoma with negative
hepatitis virus and low AFP levels (non-viral AFP- HCC) were included
according to these criteria.
Ultrasound and CEUS Examination Technique
SV-CEUS examinations were performed on an Acuson Sequoia 512 (Siemens
Healthineers) with a 4C1 convex array probe and Acuson New Sequoia (Siemens
Healthineers) with a 5C1 convex array probe, while SZ-CEUS examinations were
performed on the Acuson New Sequoia (Siemens Healthcare) with a 5C1 convex array
probe. Conventional US examinations were performed on all patients. The contrast
pulse imaging mode was used, and the mechanical indices for SV and SZ were set
to 0.19 and 0.30, respectively. SonoVue was administered as a 1.5–2.0 mL bolus
injection, followed by a saline flush with a volume of 5.0 mL. The target lesion
was scanned continuously for the first 1 minute, and then intermittently
observed until 5 min. The arterial, portal vein, and late phases were defined as
0–30s, 31–120s, and 121–250s post-injection, respectively. Sonazoid was used at
a dose of 0.6–0.8 mL (0.015mL/kg) and injected into the cubital vein by the same
method as SV. The arterial, portal vein, and late phases were recorded by the
same procedure as SV, and the post-vascular phase was recorded after 10
minutes.
Imaging analysis
Two radiologists with 10 and 4 years of liver CEUS imaging experience
independently assessed the conventional US and CEUS images in this study, with
any disagreements being resolved through discussion and consensus. The
histopathological results were kept blinded from the radiologists. The
conventional US characteristics that were evaluated included maximum diameter,
shape (rounded, oval, or irregular), boundary (well-defined or ill-defined), the
presence of a hypoechoic halo, internal echogenicity (categorized as having
hyper- and hypoechoic separation, strong hyperechogenicity with attenuation,
hyperechoic, isoechoic, or hypoechoic), and liver parenchyma (homogeneous or
heterogeneous). A heterogeneous internal echotexture, in addition to hyper- and
hypoechoic separation and strong hyperechogenicity with attenuation patterns,
shall be categorized as hyperechoic, isoechoic, or hypoechoic based on their
primary echo features.
The following CEUS scan characteristics were recorded: the type of arterial phase
hyperenhancement (APHE, categorized as homogeneous or inhomogeneous), the onset
of washout (no washout,<60s, 60–120s, 120s-5min, or > 5min), and the
washout pattern during the late phase (LP) and post-vascular phase. The washout
pattern can be categorized as entirely no washout, entirely mild washout,
entirely marked washout, or partially with washout and partially with no washout
(PWNW).
Statistical Analysis
All statistical analyses were performed using the SPSS software package (version
26.0, IBM, Armonk, NY, USA). The quantitative data was calculated as the mean±
standard deviation. The categorical variables were expressed as count and
proportion. Differences in age and diameter were compared using independent
sample t-test, and Chi-test or Fisher’s exact test was used for categorical
variables to compare the clinical, ultrasound, and CEUS characteristics. A
p -value of<0.05 was considered statistically significant.
Results
Clinical characteristics
[Table 1 ] displays the clinical
characteristics of patients with HAML and non-viral AFP- HCC. HAML was more
prevalent in females (33/50, 66.0%), while non-viral AFP- HCC was more common in
males (81/88, 92.9%) (p<0.001). Patients with HAML had a lower mean age than
those with HCC (41.1±11.3 vs. 61.4±11.2, p<0.001). The epithelioid subtype of
HAML accounted for 54% of cases (27 of 50).
Table 1 Clinical features.
HAML (n+=+50)
Non-viral AFP- HCC (n+=+88)
P
Age in years
41.1±11.3
61.4±11.2
<0.001*
Sex
<0.001*
Male
17 (34.0%)
81 (92.9%)
Female
33 (66.0%)
7 (8.6%)
Hepatitis B virus
12 (24.0%)
0
<0.001*
AFP
0.972
0~7ng/ml
41 (82.0%%)
72 (81.8%)
7–15ng/ml
9 (18.0%)
16 (18.2%)
Fatty liver disease
10 (20.4%)
22 (25.0%)
0.504
Epithelioid subtype+
27 (54.0%)
--
–
Sonazoid
12 (24%)
19 (21.6%)
0.744
HAML: hepatic angiomyolipoma; HCC: hepatocellular carcinoma *indicates
significance with p <0.05. + only for hepatic
angiomyolipoma.
Conventional US image findings
[Table 2 ] presents the conventional US
characteristics of HAML and non-viral AFP- HCC. The echogenicity was quite
different between the two tumors. HAMLs demonstrated hyper- and hypoechoic
separation (38.0% vs. 20.5%, p+=+0.025), strong hyperechogenicity with
attenuation (13.6% vs. 0%, p+=+0.001), or hyperechogenicity (34.0% vs. 13.6%,
p+=+0.005) more frequently ([Fig.2 ]).
88.9% (24 of 27) of the cases of the epithelioid subtype of HAML exhibited these
above three conventional US patterns, and only 3 cases presented as hypoechoic.
On the other hand, non-viral AFP- HCC was commonly isoechoic (12.5% vs. 0%,
p+=+0.007) or hypoechoic (53.4% vs. 14.0%, p<0.001). Additionally, a
hypoechoic halo was more frequently observed in non-viral AFP- HCC (6% vs.
22.7%, p+=+0.011). No significant differences were noted in diameter, shape,
boundary, and liver parenchyma.
Fig. 2 Three types of hyperechoic conventional US patterns of
hepatic angiomyolipoma frequently displayed. (a ) Strong
hyperechogenicity with attenuation. (b ) Hyper- and hypoechoic
separation, a clear distinction exists between hyperechogenicity and
hypoechogenicity. (c ) Hyperechoic, heterogeneous (or homogeneous)
hyperechogenicity except for the above two patterns. (b ) and
(c ) could also be observed in non-viral AFP- HCC.
Table 2 Comparison of conventional US features between
HAML and non-viral AFP- HCC.
HAML (n+=+50)
Non-viral AFP- HCC (n+=+88)
p
Diameter-mm
46.4±37.8
51.2±24.6
0.375
Shape
0.633
Rounded
7 (14.0%)
18 (20.5%)
Oval
39 (78.0%)
64 (72.7%)
Irregular
4 (8.0%)
6 (6.8%)
Boundary
0.336
Well-defined
39 (78.0%)
62 (70.5%)
Ill-defined
11 (22.0%)
26 (29.5%)
Hypoechoic halo
0.011*
Yes
3 (6.0%)
20 (22.7%)
No
47 (94.0%)
68 (77.3%)
Echogenicity
Hyper- and hypoechoic separation
19 (38.0%)
18 (20.5%)
0.025*
Strong hyperechogenicity with attenuation
7 (14.0%)
0 (0%)
0.001*
Hyperechoic
17 (34.0%)
12 (13.6%)
0.005*
Isoechoic
0 (0%)
11 (12.5%)
0.007*
Hypoechoic
7 (14.0%)
47 (53.4%)
<0.001*
Liver parenchyma
0.308
Homogeneous
49 (98.0%)
83 (94.3%)
Heterogeneous
1 (2.0%)
5 (5.7%)
HAML: hepatic angiomyolipoma; HCC: hepatocellular carcinoma. *indicates
significance with p <0.05.
CEUS findings using Sonovue and Sonazoid
[Table 3 ] shows the comparisons of CEUS
features between HAML and non-viral AFP- HCC. The CEUS characteristics on
SZ-CEUS were similar to those on SV-CEUS to a certain extent. Homogeneous APHE
was more frequently observed in HAML with both CEUS techniques, but it was
significant when using Sonovue (78.9% vs. 42.0%, p<0.001), not obvious when
using Sonazoid (91.7% vs. 73.7%, p+=+0.217). Using both contrast agents, HAML
hardly started to wash out before 120s, while some non-viral AFP- HCCs started
to wash out before 120s (7.9% vs. 33.2% on SV-CEUS, 0% vs. 42.1% on SZ-CEUS). In
LP, HAMLs more frequently displayed no washout (50% vs. 5.8% on SV-CEUS, 66.7%
vs. 15.8% on SZ-CEUS) and PWNW (7.9% vs. 0% on SV-CEUS, 16.7% vs. 0% on
SZ-CEUS), whereas non-viral AFP- HCCs more often exhibited entirely mild washout
(82.6% vs. 36.8% on SV-CEUS, 84.2% vs. 16.7% on SZ-CEUS).
Table 3 Comparison of CEUS features between HAML and
non-viral AFP- HCC using Sonovue and Sonazoid.
Sonovue
Sonazoid
HAML (n+=+38)
Non-viral AFP-HCC (n+=+69)
p
HAML (n+=+12)
Non-viral AFP- HCC (n+=+19)
p
Homogeneous APHEH
30 (78.9%)
29 (42.0%)
<0.001*
11 (91.7%)
14 (73.7%)
0.217
Washout onset
No washout
19 (50.0%)
4 (5.8%)
<0.001*
5 (41.7%)
0 (0%)
0.002*
<60s
0 (0%)
7 (11.5%)
0.049*
0 (0%)
1 (5.3%)
0.419
60–120s
3 (7.9%)
16 (21.7%)
0.048*
0 (0%)
7 (36.8%)
0.026*
<120s+
3 (7.9%)
23 (33.2%)
0.003*
0 (0%)
8 (42.1%)
0.012*
120s-5min
16 (42.1%)
42 (60.9%)
0.062
4 (33.3%)
8 (42.1%)
0.625
>5min
N/A
N/A
…
3 (25.0%)
3 (15.8%)
0.527
Late phase
Entirely no washout
19 (50.0%)
4 (5.8%)
<0.001*
8 (66.7%)
3 (15.8%)
0.007*
Entirely mild washout
14 (36.8%)
57 (82.6%)
<0.001*
2 (16.7%)
16 (84.2%)
<0.001*
Entirely marked washout
2 (5.3%)
8 (11.6%)
0.282
0 (0%)
0 (0%)
>0.999
PWNW
3 (7.9%)
0 (0%)
0.043*
2 (16.7%)
0 (0%)
0.142
Post-vascular phase
Entirely no washout
N/A
N/A
…
5 (41.7%)
0 (0%)
0.005*
Entirely mild washout
N/A
N/A
…
1 (8.3%)
11 (57.9%)
0.008*
Entirely marked washout
N/A
N/A
…
1 (8.3%)
7 (36.8%)
0.108
PWNW
N/A
N/A
…
5 (41.7%)
1 (5.3%)
0.022*
HAML: hepatic angiomyolipoma; HCC: hepatocellular carcinoma; APHE:
arterial phase hyperenhancement; PWNW: partial washout with partial no
washout. Hindicates homogeneous APHE versus inhomogeneous
APHE. + Washout onset before 120s including washout
onset before 60s and 60~120s.*indicates significance with
p <0.05.
In the distinct post-vascular phase of SZ-CEUS, HAMLs still more commonly showed
no washout (41.7% vs. 0%, p+=+0.005) and PWNW (41.7% vs. 5.3%, p+=+0.022).
Despite the fact that 36.8% of non-viral AFP- HCCs transformed to marked washout
in the post-vascular phase, non-viral AFP- HCCs still more frequently exhibited
mild washout (57.9% vs. 8.3%, p+=+0.008). All cases of non-viral AFP- HCC
exhibited washout, thereby facilitating differentiation from no-washout HAMLs.
This is a slightly superior to SV-CEUS, which only provides an LP where some
non-viral AFP- HCCs still exhibited no washout that consequently could not be
distinguishable from no-washout HAMLs.
It is worth noting that regardless of the contrast agent being used, PWNW was
exclusively found in the nodules exhibiting hyper- and hypoechoic separation
with a conventional US pattern ([Table.
4 ]), with partial washout in the hypoechoic part and partial no washout in
the hyperechoic part. However, not all nodules with hyper- and hypoechoic
separation showed PWNW. In the LP, PWNW was discernable solely in HAMLs, but the
frequency of PWNW in nodules with hyper- and hypoechoic separation was low, only
21.4% (3 of 14) on SV-CEUS and 40% (2 of 5) on SZ-CEUS. Interestingly, the HAMLs
with hyper- and hypoechoic separation in the post-vascular phase on SZ-CEUS all
demonstrated PWNW (5 of 5, 100%), more frequently compared to the LP (2 of 5,
40%), which can potentially aid in distinguishing nodules with hyper- and
hypoechoic separation as either HAML or non-viral AFP- HCC ([Table. 5 ]).
Table 4 The interaction of conventional US features and
CEUS washout pattern (Sonovue late phase vs. Sonazoid late phase)
between HAML and non-viral AFP- HCC.
No. of cases*
Hyper- and hypoechoic separation
Strong hyperechogenicity with attenuation
Hyperechoic
Isoechoic
Hypoechoic
Entirely no washout
4/0|2/1
3/0|3/0
10/2|3/1
0/0|0/0
2/2|0/1
Entirely mild washout
6/11|1/5
1/0|0/0
3/6|0/1
0/7|0/3
4/33|1/7
Entirely marked washout
1/1|0/0
0/0|0/0
1/2|0/0
0/1|0/0
1/4|0/0
PWNW
3/0|2/0
0/0|0/0
0/0|0/0
0/0|0/0
0/0|0/0
Total
14/12|5/6
4/0|3/0
14/10|3/2
0/8|0/3
7/39|1/8
PWNW: partial washout with partial no washout. *the number of cases is
listed as HAML on Sonovue CEUS/ HCC on Sonovue CEUS | HAML on Sonazoid
CEUS/ HCC on Sonazoid CEUS.
Table 5 The interaction of conventional US features and
Sonazoid CEUS washout pattern (late phase vs. post-vascular phase)
between HAML and non-viral AFP- HCC.
No. of cases*
Hyper- and hypoechoic separation
Strong hyperechogenicity with attenuation
Hyperechoic
Isoechoic
Hypoechoic
Entirely no washout
2/1|0/0
3/0|2/0
3/1|3/0
0/0|0/0
0/1|0/0
Entirely mild washout
1/5|0/3
0/0|1/0
0/1|0/1
0/3|0/2
1/7|0/5
Entirely marked washout
0/0|0/2
0/0|0/0
0/0|0/1
0/0|0/1
0/0|1/3
PWNW
2/0|5/1
0/0|0/0
0/0|0/0
0/0|0/0
0/0|0/0
Total
5/6|5/6
3/0|3/0
3/2|3/2
0/3|0/3
1/8|1/8
PWNW: partial washout with partial no washout. *the number of cases is
listed as HAML in late phase/ HCC in late phase | HAML in post-vascular
phase/ HCC in post-vascular phase.
Discussion
HAML, an unusual liver tumor, might not have any noticeable clinical or lab features.
For example, HBV doesn’t seem to have too much of a connection with HAML, and AFP
levels usually stay pretty normal. Generally speaking, this can be distinguished
from a typical HCC, which has a strong link with HBV and an elevated AFP. However,
there are also HCC cases that don’t involve HBV and have low AFP levels. It is a bit
tricky to tell the difference between these two since the clinical and laboratory
clues are lost. Previous studies have found that Sonovue CEUS was useful for
differentiating HAML from HCC [1 ]
[14 ]
[15 ],
but no studies have explored the diagnostic potential of Sonazoid CEUS in this
respect. So far, only two case reports have demonstrated the characteristics of HAML
on Sonazoid CEUS [16 ]
[17 ]. The purpose of this study was to compare
the features of HAML and non-viral AFP- HCC on Sonovue CEUS and Sonazoid CEUS, with
the goal of filling in existing gaps regarding the characteristics of HAML on
Sonazoid CEUS.
HAML presents diverse compositions and morphologies, resulting in varied
echogenicity. Although a US pattern showing strong hyperechogenicity with
attenuation seems to have an exclusive link to HAML [11 ]
[18 ], it may be challenging to
distinguish HAML from non-viral AFP- HCC utilizing the conventional US pattern
alone. The additional utilization of contrast agents becomes indispensable for
extracting enhancement information about nodules, thereby facilitating further
diagnostic assessment.
Both HAMLs and non-viral AFP- HCCs have abundant vascularity [19 ]
[20 ].
The pattern of APHE could not distinguish between HAML and non-viral AFP- HCC.
However, irrespective of contrast agent, HAMLs hardly started to wash out before
120s, and exhibited no washout more frequently, which could help to distinguish them
from non-viral AFP- HCCs, since most non-viral AFP- HCCs in our study still
maintained typical washout patterns as previously reported [21 ]. Notably, zero non-viral AFP- HCCs showed no
washout in the post-vascular phase, which can be attributed to the size of the
tumors. A portion of HCCs demonstrated no washout in the post-vascular phase, but it
was more frequently observed in nodules with a size of less than 30mm [22 ]. Generally speaking, as the size of an HCC
increases, there is a greater likelihood of washout during the post-vascular phase
[22 ]. Non-viral HCCs may be commonly
detected in more advanced stages, thus having a significantly larger tumor diameter
than viral HCCs, culminating in increased washout during the post-vascular phase,
aiding differentiation from cases of no-washout HAML.
Another interesting thing is that our study found PWNW was a relatively exclusive
feature of HAML, and more frequently occurred in the post-vascular phase on SZ-CEUS.
PWNW is a washout pattern solely observed in nodules with hyper- and hypoechoic
separation, with contrast agent partial washout in the hypoechoic part and partial
no washout in the hyperechoic part ([Fig. 3 ]
and [Fig. 4 ]). An earlier instance of this
phenomenon also surfaced in a case report [17 ],
yet it did not elicit substantial interest. This phenomenon may be attributed to the
following potential reasons: First, the washout pattern may be influenced by the
basic conventional US hyperechogenicity to an extent. Besides, the hyperechoic part
of HAMLs may be comprised of angiomatous tissue [11 ]
[18 ]
[23 ]
[24 ]
that could hinder contrast agent washout. Thirdly, the hyperechoic part of HAMLs may
contain macrophages with uptake of contrast agent in the post-vascular phase [16 ]
[25 ]
[26 ]. Conversely, the US pattern
with hyper- and hypoechoic separation in HCCs, the so-called “mosaic sign”,
representing the heterogeneity within HCCs, is probably due to the presence of
different grades of HCC differentiation [27 ] or
fatty metamorphosis [28 ], but this would not
change the washout pattern. In some cases of large HCCs, a necrotic part also
exhibited hyperechogenicity, which could be easily differentiated using CEUS ([Fig. 5 ]).
Fig. 3 Sonovue-CEUS images of hepatic angiomyolipoma. (a ) A
hypoechoic nodule demonstrated no washout in the late phase (LP). (b )
A hyper- and hypoechoic nodule displayed partial washout in the hypoechoic
part and partial no washout (PWNW) in the LP. The pathologies were both
hepatic angiomyolipoma.
Fig. 4 Sonazoid-CEUS images of hepatic angiomyolipoma with partial
washout in the hypoechoic part and partial no washout in the hyperechoic
part. A 35-year-old male presented a 32mm×29mm well-defined hyper- and
hypoechoic separation of the liver lesion (angle). The lesion showed
homogeneous arterial hyperenhancement, and began washout at 2 min 50 s. It
presented partial washout in the hypoechoic part and partial no washout
(PWNW) in the hyperechoic part at 20 min. The pathology was hepatic
angiomyolipoma.
Fig. 5 Sonazoid-CEUS images of hepatocellular carcinoma with hyper-
and hypoechoic manifestation. A 58-year-old male with normal level of AFP
and negative hepatitis B antigen, presented a 66mm×43mm well-defined hyper-
and hypoechoic separation of the liver lesion (angle). The lesion showed
heterogeneously arterial hyperenhancement with a non-enhanced necrotized
part in the hyperechoic part. The enhanced part stayed iso-enhanced before 5
min, and entirely markedly washed out at 23 min. The pathology was
hepatocellular carcinoma.
CT and MRI are valuable tools for distinguishing between benign and malignant liver
tumors. The specific features on CT for HAMLs include the presence of fat tissue
attenuation within tumors on unenhanced CT, demonstration of dysmorphic vessels,
heterogeneous arterial phase enhancement, and prolonged contrast enhancement [18 ]. The diagnostic signs of HAML on MRI exhibit
certain similarities, including: presence of fatty content in fat-saturated
T1-weighted sequences, detection of an internal vessel, hepatic vein drainage, and
remaining no washout [29 ]. The most important
sign on CT and MRI to differentiate HAML is the presence of fat content, but this
can be effective for lipomatous and mixed types of HAML, not for myomatous and
angiomatous types that lack lipid [18 ]
[30 ], resulting in varying accuracy when
diagnosing HAML. Nevertheless, these four types of HAML, irrespective of the
presence of fat, do exhibit hyperechogenicity on conventional US [18 ]. Similar results were found in our study.
Regardless of epithelioid subtype, a large percentage of HAMLs demonstrated
hyperechogenicity more frequently than non-viral AFP- HCCs, which may potentially
enhance sensitivity, compared to CT/MRI. Moreover, two head-to-head comparison
studies found that LP washout was present in more HAMLs on CT/MRI than on CEUS,
reinforcing the pivotal role of CEUS in accurate HAML diagnosis.
The study has some limitations that should be noted. First, the number of included
patients was small, due to the rarity of HAMLs and non-viral AFP- HCCs. However, it
was acceptable compared to prior studies. Second, we only compared non-viral AFP-
HCCs to HAMLs in this study, and other hypervascularized liver tumors were not
considered. Other types of hypervascularized liver tumors should be included in
future prospective studies.
Conclusion
Our study underscored the utility of Sonovue and Sonazoid CEUS for differentiating
HAMLs from non-viral AFP- HCCs, thereby serving to enrich our understanding of
distinctive features exhibited by HAMLs on Sonazoid CEUS.
Bibliographical Record
