598 ■■■■■■■■■■■■■■■■■■■■■■■■■
■ Original Article ■■■■■■■■■■■■■■■■■■■■■■■■■
Value of Laparoscopic Microwave Coagulation Therapy
for Hepatocellular Carcinoma in Relation
to Tumor Size and Location
T. Abe, H. Shinzawa, H. Wakabayashi, M. Aoki, K. Sugahara, A. Iwaba, H. Haga, S. Miyano, Y. Terui,
H. Mitsuhashi, H. Watanabe, T. Matsuo, K. Saito, T. Saito, H. Togashi, T. Takahashi
Second Dept. of Internal Medicine, Yamagata University School of Medicine, Yamagata, Japan
Background and Study Aims: The indications for laparo-
scopic microwave coagulation therapy (LMCT) for he-
patocellular carcinoma (HCC) have not yet been ade-
quately evaluated. This study investigated the value of
LMCT in the treatment of HCC.
Patients and Methods: Forty-three patients with liver
cirrhosis (including five patients in Child Pugh grade
C), with 56 HCC lesions, were enrolled in the study.
When dynamic computed tomography (CT) showed a
loss in HCC enhancement characteristics and a low
concentration area after LMCT, a lesion was judged to
have undergone complete necrosis.
Results: The rate of complete necrosis for lesions meas-
uring 40 mm or less was significantly higher (P < 0.01)
than for those measuring 41 mm or more. The rate of
complete necrosis for lesions located on the liver sur-
Introduction
Hepatectomy, percutaneous ethanol injection (PEI), and
transcatheter arterial embolization (TAE) are the preferred
major therapeutic modalities for hepatocellular carcinoma
(HCC). Hepatectomy is likely to result in complete recov-
ery, although it is not indicated for patients with low func-
tional hepatic capacity, as the surgical mortality is consid-
ered to be very high [1]. PEI is indicated for patients with
small HCCs measuring 30 mm or less in diameter. The
procedure has an adequate necrotizing effect on the tumor,
but PEI does not produce a satisfactory effect on fibrous
HCCs, as the ethanol is not distributed uniformly [2 – 6].
TAE is used when HCCs are large or multifocal, but a sin-
gle course of TAE is not sufficient to produce a satisfac-
tory necrotizing effect on the tumor. In addition, TAE is
Endoscopy 2000; 32 (8): 598 – 603
© Georg Thieme Verlag Stuttgart New York
• adequately evaluated. We have treated HCCs with LMCT,
ISSN 0013-726X
■
face, excluding those located close to the gallbladder or
in contact with the diaphragm, was also significantly
higher (P < 0.01) than for those situated deep within
Downloaded by: University of Pittsburgh. Copyrighted material.
the liver. The outcome for lesions of 40 mm or less was
favorable. Intra-abdominal hemorrhage occurred in
two patients, pneumothorax in three, and hepatic in-
farction in one, all associated with LMCT. However,
these patients did not suffer any sequelae of clinical sig-
nificance.
Conclusions: This study suggests that there is a strong
indication for LMCT for HCCs measuring 40 mm or
less in diameter and those located on the liver surface
even if they are as large as 50 mm, but not for those lo-
cated close to the gallbladder or in contact with the dia-
phragm. LMCT appears to be applicable in patients
with impaired liver function.
less effective in highly differentiated HCCs, since tumor
tissue cannot be reached with angiography [7].
Microwave coagulation therapy was recently introduced as
a method of treating HCCs [8 – 16]. The technique has ad-
vantages over PEI and TAE, as tumors with fibrous cap-
sules or septa and tumor vessels can be satisfactorily treat-
ed using thermal coagulation around an electrode [14]. Mi-
crowave coagulation therapy with laparotomy has been
used to treat HCC [8]. Saitsu et al. [9] reported the use of
laparoscopic microwave coagulation therapy (LMCT), and
Seki et al. [14] attempted percutaneous microwave coagu-
lation therapy (PMCT) under ultrasonographic guidance.
In the treatment of HCC, LMCT is less invasive than mi-
crowave coagulation using laparotomy [8, 9]; and in com-
parison with PMCT, it allows safer management of patients
who have larger tumors [13]. There have been several re-
ports of the clinical application of LMCT [9 – 13]; how-
ever, the indications for the procedure have not yet been
using laparoscopic ultrasonography (LUS). In this study,
Laparoscopic Microwave Coagulation Therapy for HCC
the value of LMCT in the treatment of HCC was evaluated,
focusing on its necrotizing effects on tumors and the inci-
dence of local recurrence.
Patients and Methods
Patients
Forty-three patients (27 males, 16 females; mean age
68.9 ± 6.7, range 55 – 81) who were diagnosed as having
HCC on histological or angiographic examinations were
enrolled in the study. They underwent LMCT at the Second
Department of Internal Medicine at Yamagata University
between March 1995 and September 1998. All of the pa-
tients had cirrhosis. The advantages and disadvantages of
the available therapeutic modalities for HCC, such as hepa-
tectomy, TAE, PEI, PMCT, and radiofrequency (RF) abla-
tion, in addition to LMCT, were explained to the patients.
We then recommended LMCT. Informed consent in writing
was obtained after all of the LMCT procedures had been
thoroughly explained.
Sixteen patients were graded A, 22 were graded B, and five
were graded C in the Child–Pugh system. One (2.3 %) was
positive for hepatitis B surface antigen (HBsAg), 39
(90.7 %) were positive for anti-hepatitis C virus (HCV), and
three (7 %) were negative for HBsAg and anti-HCV. Thirty-
three patients had a single lesion, while 10 had multiple le-
sions (eight with two lesions, one with three lesions, and the
other with four lesions). A total of 56 lesions measuring 7 –
70 mm in diameter (29.6 ± 14.3 mm, mean ± SD) were ex-
amined. Twenty-two lesions located on the liver surface were
observed under laparoscopy. Thirty-four lesions located
deep inside the liver could not be visualized under laparosco-
py, but were detected by LUS. Using Couinaud’s classifica-
tion [17], the lesions were located in segment 2 (n = 4), seg-
ment 3 (n = 7), segment 4 (n = 8), segment 5 (n = 14), seg-
ment 6 (n = 7), segment 7 (n = 1), and segment 8 (n = 15).
Methods
The course of LMCT was visualized using a digital 3
charge-coupled device (CCD) camera (OTV-SX, Olympus
Optical Co. Ltd,, Tokyo, Japan). A model SSA-270A ultra-
sound unit (Toshiba Inc., Tokyo, Japan) and a model PEF-
704LA ultrasound probe with a laparoscope (linear model,
7 MHz, Toshiba Inc., Tokyo, Japan) were used for LUS. A
model OT-110M microwave generator and model TMD-
16CB-10/250 percutaneous electrode for deep coagulation
(Heiwa Electronics, Inc., Osaka, Japan) were also used.
However, the electrode was modified to make it more easi-
ly recognizable on LUS – the electrode surface 10 mm
from the tip was serrated. This modified electrode was
adopted after pneumothorax had been observed in three pa-
tients.
With the patient under a combination of continuous epidur-
al anesthesia and local or general anesthesia, compressed
air was introduced into the abdomen through a pneumo-
Endoscopy 2000; 32 599
peritoneum needle. The first trocar was inserted through
the left side of the navel, and the second trocar for LUS
was guided into the abdomen through the right side of the
navel under laparoscopic observation. After the location of
the lesion had been confirmed with LUS, a 14-gauge nee-
dle was inserted and a microwave electrode was then guid-
ed along the sheath of the needle. Depending on the tumor
diameter, microwave irradiation was applied at least once
and up to a maximum of 40 times (16.9 ± 1.9 times) to ob-
tain an adequately coagulated area in a single course of
treatment.
Microwave irradiation (at an output of 60 W) was contin-
ued for 60 s at a time. After each 60-s coagulation proce-
dure, the electrode was moved. A piece of gauze soaked in
cold physiological saline was placed on the site at which
the electrode had been inserted, to cool it and prevent skin
burning due to the microwave irradiation.
When the HCC was located adjacent to the gallbladder, the
Downloaded by: University of Pittsburgh. Copyrighted material.
electrode was inserted 10 mm away from and parallel to
the wall of the gallbladder, to avoid burning the cystic
wall. In cases in which the liver surface was unusually un-
even, a physiological saline solution that had been heated
to approximately body temperature was injected into the
abdominal cavity. The liver was soaked in the solution so
that the entire contour of the HCC located deep inside the
liver and the insertion line of the electrode would be recog-
nizable on LUS.
Following the procedures described by Murakami et al.
[15], the effectiveness of LMCT was evaluated by dynamic
computed tomography (dynamic CT) seven days after
LMCT. When the dynamic CT showed a loss of the en-
hancement characteristic of HCC and the low-density area
that appears after LMCT, the lesion was judged to have un-
dergone complete necrosis. To detect local recurrences, dy-
namic CT examinations were carried out on principle every
four months after the LMCT procedure for one year. After
the first year, dynamic CT and abdominal ultrasound ex-
aminations were performed alternately every three months.
Serum alpha-fetoprotein (AFP) and protein induced by vi-
tamin K absence-II (PIVKA-II) were measured every
month. When the abdominal ultrasonography findings or
AFP or PIVKA-II values suggested recurrent tumor, dy-
namic CT was performed. The lesion was judged to have
undergone local recurrence when enhancement on the dy-
namic CT was detected in the LMCT-irradiated area or an
adjacent one. The therapeutic effects of LMCT were eval-
uated by tumor diameter, type of tumor (superficial or
deep-seated), and by the segment in which the lesion was
mainly situated.
Changes in aspartate aminotransferase (AST) and total
bilirubin levels were measured to determine the influence
of LMCT on hepatic function.
Fisher’s exact test was used to assess the significance of
differences in the proportions for the categorical data. The
600 Endoscopy 2000; 32 Abe T et al

log-rank test was used to examine the significance of dif-
ferences in the data calculated by the Kaplan–Meier meth-
od. For parameters, the significance level was set at 5 % on
two sides.
Results
Complete necrosis was achieved in 45 of the 56 lesions
(80.4 %) in a single course of treatment. The percentages
of complete necrosis relative to the diameter of each lesion
were: 100 % for 20 mm or less (11 of 11 lesions); 95 % for
21 – 30 mm (19 of 20 lesions); 83.3 % for 31 – 40 mm (10
of 12 lesions); 37.5 % for 41 – 50 mm (three of eight le-
sions); and 40 % for those 51 mm or larger (two of five).
Complete necrosis was achieved in 40 of 43 lesions
(93 %) measuring 40 mm or less and in five of 13 lesions
(38.5 %) measuring larger than 40 mm. The difference in
the rates of complete necrosis between the groups with a
tumor size of 40 mm or less and larger than 40 mm was
statistically significant (P < 0.01). In three lesions of
diameter). Local recurrences were observed in three of the
45 lesions (6.7 %) assessed as having complete necrosis,
with the recurrences taking place more than one year after
the LMCT procedure (13, 15, and 20 months later, respec-
tively). However, no metachronous lesions were observed
during the follow-up period. One patient died of liver fail-
ure, without recurrent HCC (Table 2).
The cumulative rate of lesions with no local recurrence
was 100 % after one year, and 82.8 % after both two years
and three years (Figure 1). All of the lesions measuring
20 mm or less in diameter, 81.8 % of those measuring
21 – 40 mm, and 50 % of those 41 mm or more remained
free of local recurrence for two years (Figure 2). Although
there were no significant differences between the lesions of
various diameters, the incidence of local recurrence within
two years was lower for tumors 40 mm or less in diameter
in comparison with those larger than 40 mm.
With regard to the procedure-related complications, intra-

Downloaded by: University of Pittsburgh. Copyrighted material.

40 mm or less in which complete necrosis was not abdominal hemorrhage occurred in two patients, pneumo-
achieved, one was located in segment 5 close to the gall- thorax in three, and hepatic infarction in one. Although
bladder (the minimum distance between the wall of the the hemorrhage involved arterial bleeding from the abdom-
gallbladder and the margin of the HCC was 1 mm) and inal wall or the liver surface, intra-abdominal hemorrhage
two were in segment 8 in contact with the diaphragm.
Table 1 The rate of complete necrosis of hepatocellular cancer
There were 22 lesions that were located on the liver surface after laparoscopic microwave coagulation therapy in relation to the
(observed under laparoscopy). Among these lesions, one tumor location (liver segment)
was 45 mm, one was 48 mm, and one was 50 mm in maxi-
mum diameter. In a single course of treatment, complete Liver segment Rate of complete necrosis
necrosis was achieved in 20 of these 22 lesions (90.9 %), % n/Σ*
while complete necrosis was obtained in 25 of 34 lesions
S2 75.0 3/4
(73.5 %) that were located deep in the liver. In two of the
S3 100.0 7/7
22 lesions that were located on the liver surface, one meas-
S4 100.0 8/8
ured 38 mm in diameter and was located in segment 5
close to the gallbladder (the minimum distance between S5 78.6 11/14
the cystic wall and the margin of the HCC was 1 mm). S6 85.7 6/7
The other was 35 mm in diameter and was located in seg- S7 100.0 1/1
ment 8 in contact with the diaphragm. Excluding these two S8 60.0 9/15
lesions, the difference in the rates of complete necrosis be- * Number of tumors with complete necrosis/number of tumors treated.
tween the groups (those in which the tumor was located on
the liver surface and those with deep tumors in the liver)
Table 2 The length of the folllow-up period in 45 hepatocellular
was statistically significant (P < 0.01). cancer lesions with complete necrosis after laparoscopic microwave
coagulation treatment (LMCT) in relation to tumor size
In a single course of treatment, complete necrosis was
achieved for all lesions in segments 3, 4, and 7, while the Tumor size
process was incomplete in one lesion in segment 2, another ≤ 20 mm 21 – 40 mm ≥ 41 mm
(n = 11) (n = 29) (n = 5)
in segment 6, three in segment 5, and six in segment 8.
Complete necrosis in segment 8 was calculated to be 60 % Mean follow-up 23.2 ± 12.8 15.1 ± 10.5 13.6 ± 8.6
(nine of 15 lesions) (Table 1). Of the six lesions located in (months) (range 1 – 38) (range 1 – 42) (range 1 – 28)
segment 8 that showed incomplete necrosis, two were in Min. Max
contact with the diaphragm and four were larger than < 12 2 lesions 10 lesions 1 lesion
40 mm in maximum diameter. ≥ 12 < 24 4 lesions 13 lesions* 3 lesions**
≥ 24 < 36 2 lesions 3 lesions 1 lesion
The average follow-up period for the 45 lesions with com-
≥ 36 3 lesions 3 lesions –
plete necrosis was 16.9 ± 11.5 months (mean ± SD). Thirty-
two lesions were followed up for longer than 12 months * Local recurrence was observed in two lesions (one 13 months after
(nine lesions of 20 mm or less in diameter, 19 lesions of LMCT and the other 20 months after LMCT), and one patient died
of liver failure without a recurrence of hepatocellular cancer.
21 – 40 mm in diameter, four lesions of 41 mm or more in ** Local recurrence was observed in one lesion 15 months after LMCT.
Laparoscopic Microwave Coagulation Therapy for HCC
%
100
50
0
0 1 2 3 4
Years
Figure 1 The cumulative rate of lesions without local recurrence
was 100 % after one year and 82.8 % after two years, remaining
at 82.8 % after three years. Local recurrences were observed in
three lesions more than one year (13, 15, and 20 months) after la-
paroscopic microwave coagulation therapy. (Kaplan–Meier curve
for the incidence of local recurrence in lesions assessed as having
undergone complete necrosis)
% under laparoscopy (superficial HCCs) was also high, even
100
50
≤ 20 mm (n = 11)
21 – 40 mm (n = 29)
≥ 41 mm (n = 5)
0 trode used in PMCT. Arterial hemorrhage occurred in two
0 1 2 3 4 cases, one from the liver at the point from which the elec-
Years
Figure 2 The cumulative rate of lesions without local recurrence
in relation to the size of the tumors. The two-year remission rate
after treatment was 100 % in lesions with a diameter of 20 mm
or less, 81.8 % in those measuring 21 – 40 mm, and 50 % in those
with diameters of 41 mm or more. (Kaplan–Meier curve for the in-
cidence of local recurrence in lesions assessed as having undergone
complete necrosis)
in these two patients was immediately controlled by micro-
wave coagulation therapy. The three patients with pneumo-
thorax were managed conservatively. The hepatic infarction
may have been caused by an artery and portal vein at seg-
ment 2 that were coagulated with LMCT; but the patient
did not have any clinically significant sequelae. The artery
and portal vein in question measured 3 mm and 5 mm in
diameter on an angiogram, respectively.
In the patients who underwent LMCT, aspartate amino-
transferase levels were elevated by 200 – 300 IU/l on the
day after LMCT; but the level returned almost to the base-
Endoscopy 2000; 32 601
line within seven days. The total bilirubin level increased
slightly the day after LMCT, but returned to the baseline
within seven days.
All of the patients had pain at the electrode insertion site
after LMCT, but this was tolerable for most and resolved
within seven days in all cases. Some of the patients ran a
fever, which subsided within three days after LMCT. All of
them were able to start with oral feeding on the day after
LMCT and were also able to walk the day after treatment.
The mean postoperative hospital stay was 14.3 days (range
8 – 21 days) after LMCT.
Discussion
Seki et al. [14] carried out PMCT under local anesthesia in
18 patients with HCCs of 9 – 20 mm. They reported that
PMCT was effective for small HCCs, but that a single
treatment was insufficient for lesions larger than 20 mm.
In the present study, it was shown that a high rate of com-
Downloaded by: University of Pittsburgh. Copyrighted material.
plete necrosis can be achieved in lesions of 40 mm or less.
LMCT can therefore be used to treat larger HCCs in which
PMCT is not practical. The rate of complete necrosis in le-
sions located on the liver surface that could be observed
if the lesions were as large as 50 mm.
In contrast to LMCT, adequate analgesia may not be
provided for patients with superficial HCC due to the pain
caused by abdominal irritation from the heat when PMCT
is carried out under local anesthesia. PMCT may not be a
good choice as a treatment for superficial HCCs. In addi-
tion, if hemorrhage occurs when the electrode is removed,
it cannot be controlled sufficiently in PMCT. In the present
study, LMCT was carried out using the same type of elec-
trode was removed and the other from the abdominal wall
at the point at which the electrode had been inserted. Im-
mediate hemostasis was possible in both cases. If hemor-
rhage occurs during PMCT, on the other hand, it is not de-
tected by the surgeons immediately. In comparison with
PMCT, therefore, LMCT is able to provide better and safer
management for patients who have larger or superficial
HCCs.
Sato et al. [11] reported that the maneuverability of the
present LMCT equipment is limited. The procedure is
therefore preferable when superficially located and small
HCC nodules are well visualized. However, Ido et al. [13]
reported complete necrosis not only in superficial but also
in deep-seated HCCs, using LMCT with LUS. The authors
also state that the HCC diameter suitable for LMCT was
30 mm or smaller. Compared with PMCT, it is certainly
not easy to follow the insertion line of the electrode when
using LUS in LMCT. It is also difficult to capture the en-
tire contour of HCC in cases in which the liver surface is
unusually uneven due to cirrhosis, as the LUS probe can-
not come into close contact with this type of liver surface.
602 Endoscopy 2000; 32
We therefore use two techniques to facilitate recognition of
the inserted electrode and the entire contour of HCC. The
first is to modify the electrode – i.e., to serrate the elec-
trode surface 10 mm from the tip. The second technique is
to soak the liver surface in a physiological saline solution
that has been warmed approximately to body temperature,
so that the electrode contact is be made through the phys-
iological saline. These methods allowed treatment of rela-
tively large HCCs (40 mm or less).
Beppu et al. [18] evaluated the efficacy of various tech-
niques of microwave coagulation therapy, including micro-
wave coagulation with laparotomy or thoracotomy, LMCT,
and PMCT in 84 patients with HCCs. They reported that
the effectiveness of microwave coagulation therapy was
comparable to that of hepatectomy. Yamanaka et al. [12]
reported that the benefits of microwave coagulation ther-
apy with laparotomy or LMCT in comparison with wedge
resection included technical feasibility, negligible intraop-
erative blood loss, shorter operating time, insignificant pa-
tient morbidity, and rapid return of patients to their pre-
operative condition and activities. The authors also men-
tion that they favor LMCT as a therapeutic modality for
HCCs if the laparoscopic approach is possible. Sato et al.
[11] used both surgical microwave coagulation therapy and
LMCT as treatments for HCC, and reported that in patients
with more severely impaired liver function, they preferred
the laparoscopic approach to open surgery, since LMCT is
less invasive than surgical methods. In the present study,
the aspartate aminotransferase and total bilirubin levels in-
creased after treatment, but the changes were transient.
Both levels returned to baseline within seven days after
LMCT. In five patients who were classified as grade C in
the Child–Pugh system, hepatic function was rapidly re-
stored to the preoperative state, even though there was al-
ready severe impairment. We were therefore able to con-
firm that LMCT is a safe method of treatment for HCC.
Ebara et al. [4] carried out percutaneous ethanol injection
(PEI) in 203 patients with HCCs of 30 mm or less, and
achieved complete necrosis in 97.6 %. In the present study,
LMCT produced complete necrosis in 100 % of HCCs of
20 mm or less and in 96.8 % of those 30 mm or smaller –
showing that LMCT has a necrotizing effect comparable to
that of PEI in HCCs of 30 mm or less. In addition, LMCT
produced complete necrosis in 83.3 % of HCCs of 31 –
40 mm. LMCT can lead to a favorable outcome in the
treatment of HCCs measuring up to 40 mm, although PEI
is indicated for small HCCs that are 30 mm or smaller. A
single application of PEI is not necessarily satisfactorily ef-
fective in treating HCCs of 30 mm or less, particularly
those with a fibrous texture, as the ethanol is not distribut-
ed uniformly. By contrast, LMCT usually has satisfactory
efficacy in treating HCCs of 40 mm or less in a single
course of treatment. The hospital stay after PEI treatment
may consequently be longer than that after LMCT, and
LMCT may also be more satisfactory in comparison with
PEI in terms of cost–benefit considerations.
Abe T et al
Matsui et al. [19] reported that the cumulative incidence of
local recurrences in 113 lesions of 40 mm or less was 30 %
two years after segmental transcatheter arterial emboliza-
tion (TAE). Similarly, Watanabe et al. [20] reported that
the cumulative incidence of local recurrences of 31 smaller
lesions (average diameter 37 mm) was 56 %, also two years
after segmental TAE. In the present study, the cumulative
incidence of local recurrences was broken down as follows:
50 % for tumors 41 mm or more, 9.2 % for those 21 –
40 mm, and 0 % for those 20 mm or less. The outcome in
lesions of 40 mm or less was favorable with LMCT in
comparison with TAE. In addition, TAE may not produce
complete necrosis in a single course of treatment, while a
single application of LMCT usually produces complete ne-
crosis. For these reasons, LMCT appears to be superior to
transcatheter arterial embolization as a treatment for HCC.
Ishii et al. [3] carried out PEI in 170 lesions (average di-
ameter 20 mm), and reported that the cumulative incidence
of local recurrences was 14.2 % after two years. Seki et al.
Downloaded by: University of Pittsburgh. Copyrighted material.
[14] evaluated PMCT applied to 18 lesions measuring
20 mm or less. No local recurrence was noted during the
observation period (11 – 33 months after surgery). The out-
come in lesions of 20 mm or less was comparable to that in
lesions treated with PEI or PMCT. This study also showed
that LMCT produced a satisfactory outcome in treating
HCCs up to 40 mm. LMCT therefore seems to be superior
to percutaneous ethanol injection and PMCT, but this has
yet to be confirmed in a controlled clinical trial.
The present study suggested that LMCT is preferable for
lesions of 40 mm or smaller. However, complete necrosis
was not achieved in three lesions that were 40 mm or less.
One was located in segment 5, adjacent to the gallbladder
(the minimum distance between the cystic wall and the
margin of the HCC was 1 mm), and two were located in
segment 8 and in contact with the diaphragm. According
to Seki et al. [14], the area coagulated by the electrode
used in microwave coagulation therapy is elliptical, with
maximum and minimum diameters of 24 ± 4 mm and
16 ± 3 mm, respectively. The outermost border of the coag-
ulated area parallel to the electrode was calculated to be
8 ± 1.5 mm from the electrode. This distance has been con-
firmed by experiments in dogs (data not shown here). The
electrode was inserted 10 mm away from and parallel to
the wall of the gallbladder. Thus, when an HCC was locat-
ed close to the gallbladder (almost in contact with it), we
were not able to achieve complete coagulation. In these in-
stances, LMCT should be carried out in combination with
laparoscopic cholecystectomy.
Pneumothorax developed in three patients, two of whom
underwent incomplete coagulation (as mentioned above).
Their clinical condition improved with conservative treat-
ment. In these three cases, the HCCs were in contact with
the diaphragm. Special attention should be given to avoid-
ing pneumothorax when carrying out LMCT to treat HCCs
located close to the diaphragm (within 10 mm). Asahara et
al. [21] carried out thoracoscopic microwave coagulation
Laparoscopic Microwave Coagulation Therapy for HCC
therapy in three HCCs that were located in segment 7 or 8,
measuring 15 – 30 mm. They concluded that thoracoscopic
microwave coagulation therapy may be a less invasive op-
tion for curative treatment of HCCs in segments 7 and 8 in
patients with advanced liver cirrhosis and severe complica-
tions. As a rule, HCCs that are located in segments 7 or 8
and adjacent to the diaphragm should be treated by thora-
coscopic microwave coagulation therapy [8, 16].
It is thought that large blood vessels do not sustain injury
due to the cooling action of blood flowing in the vessel
[22]. We observed one patient who developed a hepatic in-
farction, probably due to an artery and portal vein that had
been coagulated during LMCT treatment of an HCC in
segment 2. The patient did not suffer any clinically signif-
icant sequelae, however. The artery and the portal vein in
question measured 3 mm and 5 mm in diameter, respective-
ly, on an angiogram. In view of this complication, the elec-
trode should be inserted 10 mm away from the major Glis-
son’s branch, which is larger than the third branch.
Conclusions
LMCT appears to be an appealing and promising new
treatment option, owing to its minimal invasiveness, ap-
plicability in patients with impaired liver function, and ra-
pid postoperative recovery. We would suggest that LMCT
is strongly indicated as a therapeutic modality for HCC le-
sions of 40 mm or less and those located on the liver sur-
face and visualized at laparoscopy (even when they are
possibly 50 mm or less). However, lesions located close to
the gallbladder or in contact with the diaphragm should be
excluded from this category.
References
1 embolization. Radiology 1993; 188: 79 – 83
Nagasue N, Yukaya H, Ogawa Y, et al. Clinical experience with
118 hepatic resections for hepatocellular carcinoma. Surgery
1986; 99: 694 – 701
2 1995; 31: 239 – 249
Shiina S, Tagawa K, Niwa Y, et al. Percutaneous ethanol injec-
tion therapy for hepatocellular carcinoma: result in 146 pa-
tients. Am J Roentgenol 1993; 160: 1023 – 1028
3 ima J Med Sci 1998; 47: 125 – 131
Ishii H, Okada S, Nose H, et al. Local recurrence of hepatocel-
lular carcinoma after percutaneous ethanol injection. Cancer
1996; 77: 1792 – 1796
4 1993; 9: 318 – 322
Ebara M, Fukuda H, Yoshikawa M, et al. Percutaneous ethanol
injection (PEI) for patients with hepatocellular carcinoma 3 cm
or smaller [in Japanese with English abstract]. Endosc Dig
1997; 9: 1315 – 1320
5
Shiina S, Tagawa K, Unuma T, et al. Percutaneous ethanol in-
jection for hepatocellular carcinoma: a histopathologic study.
Cancer 1991; 68: 1524 – 1530
6
Ohnishi K, Yoshioka H, Ito S, Fujiwara K. Prospective random-
ized controlled trial comparing percutaneous acetic acid injec-
tion and ethanol injection for small hepatocellular carcinoma.
Hepatology 1998; 27: 67 – 72
7
Kuroda C, Sakurai M, Monden M. Limitation of transcatheter
arterial chemoembolization using iodized oil for small hepato-
cellular carcinoma: a study in resected cases. Cancer 1991; 67:
81 – 86
Endoscopy 2000; 32 603
8
Saitsu H, Utsunomiya T, Emi Y, et al. Microwave coagulation
therapy [in Japanese with English abstract]. Endosc Dig 1997;
9: 1349 – 1354
9
Saitsu H, Yoshida M, Taniwaki S, et al. Laparoscopic coagulo-
necrotic therapy using microtase for small hepatocellular carci-
noma [in Japanese with English abstract]. Nippon Shokakibyo
Gakkai Zasshi (Jpn J Gastroenterol) 1991; 88: 2727
10
Watanabe Y, Sato M, Abe Y. Laparoscopic microwave coagula-
tion therapy for hepatocellular carcinoma: a feasible study of an
alternative option for poor-risk patients. J Laparoscopic Surg
1995; 5: 169 – 175
11
Sato M, Watanabe Y, Ueda S, et al. Microwave coagulation
therapy for hepatocellular carcinoma. Gastroenterology 1996;
110: 1507 – 1514
12
Yamanaka N, Okamoto E, Tanaka T, et al. Laparoscopic micro-
wave coagulation therapy for hepatocellular carcinoma. Surg
Laparosc Endosc 1995; 5: 444 – 449
13
Ido K, Isoda N, Kawamoto C, et al. Laparoscopic microwave
coagulation for solitary hepatocellular carcinoma performed
under laparoscopic ultrasonography. Gastrointest Endosc 1997;
45: 415 – 420
14
Seki T, Wakabayashi M, Nakagawa T, et al. Ultrasonically
Downloaded by: University of Pittsburgh. Copyrighted material.
guided percutaneous microwave coagulation therapy for small
hepatocellular carcinoma. Cancer 1994; 74: 817 – 825
15
Murakami F, Yoshimatsu S, Yamashita Y, et al. Treatment of
hepatocellular carcinoma: value of percutaneous microwave co-
agulation. AJR Am J Roentgenol 1994; 164: 1159 – 1164
16
Saitsu H, Nakayama T, Isomura T, et al. New endoscopic surgi-
cal treatment: thoracoscopic microwave coagulo-necrotic ther-
apy for small hepatocellular carcinoma [in Japanese with Eng-
lish abstract]. J Microwave Surg 1994; 12: 1 – 8
17
Couinaud C. Les enveloppes vasculo-biliaires du foie ou cap-
sule de Glisson. Lyon Chir 1954; 49: 489 – 607
18
Beppu T, Ogawa M, Matsuda T, et al. Efficacy of microwave
coagulation therapy (MCT) in patients with liver cirrhosis [in
Japanese with English abstract]. Jpn J Cancer Chemother
1998; 25: 1358 – 1361
19
Matsui O, Kadoya M, Yoshikawa J, et al. Small hepatocellular
carcinoma treatment with subsegmental transcatheter arterial
20
Watanabe M, Ichida T, Sato T. Recurrence of hepatocellular
carcinoma after segmental LP-TAE [in Japanese]. Kan Tan Sui
21
Asahara T, Katayama K, Itamoto T, et al. Thoracoscopic micro-
wave coagulation therapy for hepatocellular carcinoma. Hirosh-
22
Zhou XD, Tang ZY, Yu YQ, et al. Microwave surgery in the
treatment of hepatocellular carcinoma. Semin Surg Oncol
Corresponding Author
H. Shinzawa
Second Dept. of Internal Medicine
Yamagata University School of Medicine
PN 990-9585 2-2-2 Iida Nishi
Yamagata City, Japan
Fax: + 81-23-628-5311
E-mail: hsinzawa@med.id.yamagata-u.ac.jp
Submitted: 28 July 1999
Accepted after Revision: 24 February 2000