World J Gastroenterol 2012 December 7; 18(45): 6587-6596

ISSN 1007-9327 (print) ISSN 2219-2840 (online)

Online Submissions: http://www.wjgnet.com/esps/

[email protected]
doi:10.3748/wjg.v18.i45.6587

2012 Baishideng. All rights reserved.

ORIGINAL ARTICLE

Tumour seeding after percutaneous cryoablation for

hepatocellular carcinoma
Chun-Ping Wang, Hong Wang, Jian-Hui Qu, Yin-Ying Lu, Wen-Lin Bai, Zheng Dong, Xu-Dong Gao,
Guang-Hua Rong, Zhen Zeng, Yong-Ping Yang
1-90) mo. Seeding was detected in 11 patients (0.76%)
at 1-24 (median 6.0) mo after cryoablation. Seeding
occurred along the needle tract in 10 patients and at
a distant location in 1 patient. Seeded tumours usually
showed similar imaging and histopathological features
to the primary HCCs. Univariate analyses identified
subcapsular tumour location and direct subcapsular
needle insertion as risk factors for seeding. Multivariate
analysis showed that only direct subcapsular needle
insertion was an independent risk factor for seeding (P
= 0.017; odds ratio 2.57; 95%CI: 1.47-3.65). Seeding after cryoablation occurred earlier in patients with
poorly differentiated HCC than those with well or moderately differentiated HCC [1.33 0.577 mo vs 11.12
6.896 mo; P = 0.042; 95%CI: (-19.115)-(-0.468)].

Chun-Ping Wang, Hong Wang, Jian-Hui Qu, Yin-Ying Lu,

Wen-Lin Bai, Zheng Dong, Xu-Dong Gao, Guang-Hua
Rong, Zhen Zeng, Yong-Ping Yang, Center of Therapeutic
Research for Hepatocellular Carcinoma, 302 Hospital of PLA,
Beijing 100039, China
Author contributions: Wang CP and Yang YP designed the
study and wrote the manuscript; Wang H and Qu JH performed
the data analysis and interpretation; Lu YY, Bai WL and Dong
Z performed the treatment; Gao XD, Rong GH and Zeng Z
provided the collection of all the patient material; Yang YP provided financial and administrative support for this work.
Supported by Grants from the Key Scientific and Technological Research Foundation of the National Special-Purpose
Program, No. 2008ZX10002-018; Military Special-Purpose
Program, No. BWS11J074; and the Capital Medical Research
and Development Fund, No. 2009-2041, China
Correspondence to: Dr. Yong-Ping Yang, Center of Therapeutic Research for Hepatocellular Carcinoma, 302 Hospital of
PLA, Beijing 100039, China. [email protected]
Telephone: +86-10-63879193 Fax: +86-10-63879193
Received: June 10, 2012
Revised: July 30, 2012
Accepted: August 14, 2012
Published online: December 7, 2012

CONCLUSION: The risk of seeding after cryoablation

for HCC is small. Direct puncture of subcapsular tumours
should be avoided to minimise seeding.
2012 Baishideng. All rights reserved.

Key words: Cryoablation; Hepatocellular carcinoma; Tumour seeding; Clinical feature; Risk factor

Abstract

Peer reviewers: Salvatore Gruttadauria, MD, Assistant Profes-

AIM: To assess the rate and risk factors for tumour

seeding in a large cohort of patients.

sor, Abdominal Transplant Surgery, ISMETT, Via E. Tricomi,

190127 Palermo, Italy; Zenichi Morise, MD, PhD, Professor
and Chairman, Department of Surgery, Banbuntane Houtokukai
Hospital, Fujita Health University School of Medicine, 3-6-10
Otobashi Nakagawa-ku, Nagoya, Aichi 454-8509, Japan

METHODS: Over an 8-year period, 1436 hepatocellular carcinoma (HCC) patients with 2423 tumour nodules
underwent 3015 image-guided percutaneous cryoablation sessions [1215 guided by ultrasonography and
221 by spiral computed tomography (CT)]. Follow-up
CT or magnetic resonance imaging was performed every 3 mo. The detailed clinical data were recorded to
analyse the risk factors for seeding.

Wang CP, Wang H, Qu JH, Lu YY, Bai WL, Dong Z, Gao XD,
Rong GH, Zeng Z, Yang YP. Tumour seeding after percutaneous
cryoablation for hepatocellular carcinoma. World J Gastroenterol
2012; 18(45): 6587-6596 Available from: URL: http://www.
wjgnet.com/1007-9327/full/v18/i45/6587.htm DOI: http://dx.doi.
org/10.3748/wjg.v18.i45.6587

RESULTS: The median follow-up time was 18 (range

WJG|www.wjgnet.com

6587

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

INTRODUCTION

MATERIALS AND METHODS

In patients with hepatocellular carcinoma (HCC) superimposed on cirrhosis, orthotopic liver transplantation,
surgical resection and percutaneous ablation are considered radical treatments as they provide better survival
rates compared with no treatment[1]. Because of the poor
acceptance of surgery and a severe shortage of donor
organs, image-guided percutaneous ablation therapies
play an important role in the management of HCC. Various local ablation therapies such as percutaneous ethanol
injection (PEI), radiofrequency ablation (RFA), microwave (MW) ablation and cryoablation have been developed for the treatment of unresectable HCC. All these
procedures require the insertion of long, sharp needles
into the liver parenchyma and tumours, which may cause
various complications, although the complication rates
are low. Tumour seeding is one of the most serious complications, especially in patients who are waiting for liver
transplantation[2]. The reported incidence of seeding
after other ablation procedures varies widely: 0.2%-1.4%
following PEI[3,4], 0.005%-12.5% following RFA[5,6] and
0.75% following MW ablation[7].
Certain factors have been found to increase the likelihood of needle-tract seeding, including a superficial
or subcapsular tumour[8], high number of needle insertions[9,10], large needle bore[9,11], end-cutting needle[8,9], absent or thin layer of normal liver parenchyma surrounding the needle tract[9,10], high-grade HCC (moderately or
poorly differentiated[3,8,9], high serum alpha-fetoprotein
(AFP) level[2], tumour volumes > 2 cm3 and immunosuppression[12].
Argon-helium cryoablation is a new local ablation modality. At one time, this technology caused some authors
to question its use in HCC. Most of the bias against this
percutaneous setting is based on a theoretical risk of postprocedure haemorrhage. However, the gradual downsizing of cryoprobes has fueled interest in percutaneous use,
which offers several potential advantages versus the heatbased ablation modalities[13]. First, multiple cryoprobes can
be used simultaneously to generate a large zone of ablation. Second, the size and shape of the developing ice ball
can be readily visualized using intra-procedural computed
tomography (CT), magnetic resonance imaging (MRI) or
ultrasound (US). Third, in contrast to heat-based ablation,
percutaneous cryoablation is a relatively painless procedure. Recently, many studies have reported that imagingguided percutaneous cryoablation is safe and effective for
the treatment of HCC[14-16]. In our previous study, the 1-,
2- and 3-year survival rates in patients with HCC < 5 cm
in diameter who were treated with cryoablation were 92%,
82% and 64%, respectively, and the rate of serious complications was low[15]. To our knowledge, tumour seeding
after percutaneous cryoablation for HCC has not been described to date. The present study was conducted to evaluate the incidence and possible risk factors for seeding after
percutaneous cryoablation, by reviewing the prospective
database of HCC patients treated by cryoablation in our
department.

Patients
This study included 1436 consecutive patients with HCC
who were treated at the Center of Therapeutic Research
for Hepatocellular Carcinoma, 302 Hospital of PLA,
between April 2003 and June 2011. HCC was diagnosed
based on typical findings on MRI or CT (hyperattenuation
in the arterial phase and hypoattenuation in the portalvenous phase) and serum AFP level. The diagnosis was
confirmed by histopathological examination of US- or
CT-guided biopsy specimens in 736 patients. Until 2007,
we biopsied almost all tumours before treatment, and after
2007 we only biopsied cases in which we could not make
a definite diagnosis using dynamic CT or MRI. Biopsy
specimens for histological examination were obtained
with 1-2 passes of a 19.5-gauge end-cutting needle (AutoVac; Angiomed, Karlsruhe, Germany). Histopathological
grading of tumour differentiation was performed using
the criteria described by Edmondson et al[17]. Tumour
stage was defined according to the Barcelona Clinic Liver
Cancer (BCLC) classification[18]. Performance status (PS)
was defined according to the Eastern Cooperative Oncology Group criteria (ECOG). The 1436 patients had a total
of 2423 tumours with a diameter of 1.2-15.0 cm (mean
4.5 2.3 cm). The clinical characteristics of the patients
are shown in Table 1. All cryoablation treatments were approved by the Research Ethics Committee at 302 Hospital
of PLA. Written informed consent was obtained from all
patients who met the inclusion criteria, before blood and
tumour specimens were obtained, and before data were
collected and analysed.

WJG|www.wjgnet.com

Inclusion and exclusion criteria

Inclusion criteria for cryoablation were as follows: contraindications to surgical resection or orthotopic liver transplantation, Child-Pugh class A or B liver function, total
serum bilirubin level < 51.3 mol/L, platelet count 20
109/L and performance status 2. Ascites was controlled before the procedure with diuretics. Patients with
early HCC who were reluctant to undergo hepatic resection or transplantation were included. Patients were excluded for the following reasons: BCLC stage D (ECOG
PS > 2, Child-Pugh C), tumour thrombosis at the main
branch of the portal vein and the size of the thrombosis
exceeded 50% of the diameter of the portal vein, extrahepatic metastasis, tumours which were not accessible
percutaneously, or a history of other ablation therapies.
We generally performed percutaneous cryoablation in
patients with up to three lesions, all of which were 5
cm in diameter, but we performed a combination of repeated cryoablation and transarterial chemoembolisation
(TACE) in some patients who did not meet these criteria.
Technical terms
We defined a session as a single treatment consisting of
one or more ablations performed on one or more tumours. To assess tumour depth, we categorised tumours

6588

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

After determining the most favourable percutaneous approach, we inserted the 3-mm cryoprobe into the tumour
through the sheath introducer system under US or CT
guidance, and advanced the tip to the distal margin of the
targeted lesion. The number of probes used depended
on the location and size of the lesions to be ablated.
The dual freeze-thaw cycle comprised a 20-min freeze,
a 10-min thaw and a further 15-min freeze. The dimensions of the frozen tissue were monitored by US or CT.
The cryoprobe temperatures were reduced to -135
2 within 1 min. After removal of the probes, all tracts
were packed with Surgicel (Johnson and Johnson, Inc.,
Arlington, TX, United States) through the sheath introducer to control bleeding, and the sheath introducer was
removed. We aimed to perform curative ablation of all
tumours in each session by single or multiple cryoablation, particularly for tumours < 5 cm in diameter. Dynamic CT or MRI was performed 2-3 d after treatment
to evaluate treatment efficacy. Complete ablation was defined as non-enhancement of the entire lesion on CT or
MRI with a safety margin in the surrounding liver parenchyma. Patients underwent additional ablation sessions
until complete ablation was confirmed in all nodules, to a
maximum of three sessions. If ablation was incomplete
after three sessions, we performed TACE. The cryoablation procedure was performed under conscious sedation.
Echocardiography, ventilation and oxygen saturation
levels were monitored throughout the procedure. Patients
were kept warm during cryoablation with warming mats.

Table 1 Baseline characteristics of patients n (%)

Baseline characteristics
Age (yr)1
Sex
Male
Female
Aetiology
HBs-Ag positive only
HCV-Ab positive only
Both positive
Both negative
Child-Pugh score
Class A
Class B
Tumour size (cm)
3
3-5
5
Tumour number
Single
2-3
Tumour location
Subcapsular
No subcapsular
Route of needle insertion
Direct subcapsular insertion
Deep
BCLC staging
Stage A
Stage B
Stage C
Completed ablation
Yes
No
Tumour differentiation2
Well or moderate
Poorly
Biopsy performed prior to cryoablation2
Number of sessions
Single
2
>2
AFP (ng/mL)1

Value
55.9 9.2
1176 (81.9)
260 (18.1)
1229 (85.6)
168 (11.7)
19 (1.3)
20 (1.4)
874 (60.9)
562 (39.1)
411 (28.6)
656 (45.7)
369 (25.7)
1213 (84.5)
223 (15.5)
484 (33.7)
952 (66.3)
213 (14.8)
1223 (85.2)
787 (54.8)
453 (31.5)
196 (13.7)
1168 (81.3)
268 (18.7)
490 (66.6)
246 (33.4)
736 (51.3)

Follow-up and tumour seeding

All patients underwent MRI and CT at 1 mo after cryoablation. Patients were then assessed every 2-3 mo, including measurements of liver biochemistry and AFP level,
and by CT or MRI. A newly detected tumour attached to
the peritoneum or pleura was considered to be seeded,
and the diagnosis was confirmed by biopsy and histopathological examination. Seeded tumours were treated
with repeat cryoablation, PEI or TACE when feasible.
The seeding rate was calculated based on the number of
patients.

336 (23.4)
743 (51.7)
357 (24.9)
575 2039

Values are expressed as mean SD, n = 1436; 2Of 736 cases in which
biopsy was performed. HBs-Ag: Hepatitis B surface antigen; HCV-Ab:
Hepatitis C virus antibody; BCLC: The Barcelona Clinic Liver Cancer classification; AFP: Alpha-fetoprotein.

Statistical analysis
Potential risk factors for seeding were analysed. The following variables were recorded: age, sex, viral markers, tumour size, number of tumour nodules, tumour location,
direct subcapsular needle insertion, tumour differentiation, number of cryoablation sessions, number of needle
insertions, percutaneous biopsy prior to cryoablation and
serum AFP level. Continuous variables were compared
between patients with and without seeding using the Students t test. The 2 test or Fishers exact test was used to
compare categorical variables between the groups. Variables with P < 0.1 were entered into a multivariate logistic
regression model using stepwise selection of variables.
Variables with P < 0.05 were considered statistically significant. All analysis were conducted using SPSS software
version 13 (SPSS Inc., Chicago, IL, United States).

as subcapsular or deep. Tumours were defined as subcapsular when they were located adjacent to the surface of
the liver, less than 0.5 cm of parenchyma between the tumour and the liver capsule, otherwise, they were defined
as deep. Direct subcapsular needle insertion was defined
as puncture of subcapsular tumours without traversing
a sufficient portion of normal hepatic parenchyma. The
number of needle insertions was defined as the total
number of needle positions in all sessions.
Argon-helium cryoablation procedure
Argon-helium cryoablation was performed as described
in our previous report[16]. Briefly, an argon-helium gasbased CRYOcare system (EndoCare, Irvine, CA, United
States) and cryoprobes were used to freeze the tumour
with a dual freeze-thaw cycle under US or CT guidance.

WJG|www.wjgnet.com

6589

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

Table 2 Characteristics of the 11 hepatocellular carcinoma patients who had tumour seeding after cryoablation
Case No.
Age (yr)
Sex
Child-Pugh class
BCLC Stage
AFP (ng/mL)
No. of tumours
Tumour size (cm)
No. of sessions
No. of needle insertions
Completed ablation
Biopsy
Tumour differentiation
Primary HCC
Seeding HCC
Subcapsular location
Direct subcapsular insertion
Seeding location
Seeding time (mo)
Overall survival (mo)
Prognosis

10

11

66
M
A
A
7
1
6
2
4
Yes
No

51
F
A
B
9
1
3
1
2
Yes
Yes

47
M
A
B
3550
1
5.4
4
6
No
Yes

51
M
A
B
368
1
6
3
4
No
Yes

65
F
B
B
8589
1
3.2
1
2
Yes
Yes

43
M
A
C
16
1
8
3
6
No
No

49
M
A
A
75
1
2.4
1
1
Yes
No

61
M
B
B
33
3
4.8
2
3
Yes
No

58
M
A
A
23
1
2.6
1
1
Yes
No

58
F
A
A
48
2
2
4
4
Yes
Yes

72
M
B
B
294
1
5.6
2
3
Yes
Yes

Mod
Yes
Yes
Ip
24
36
Died

Poor
Poor
No
No
Ip
2
25
Alive

Mod
Mod
Yes
Yes
Sf
5
9
Died

Mod
Mod
Yes
Yes
Ip
5
19
Died

Poor
Yes
Yes
Pleura
1
18
Died

Mod
Yes
Yes
Ip and Im
6
12
Alive

Mod
Yes
Yes
Ip
12
18
Alive

Mod
No
No
Ip
7
26
Died

Mod
Yes
Yes
Ip
12
36
Alive

Well
Well
Yes
Yes
Ip
18
60
Alive

Poor
No
No
Pc
1
5
Alive

M: Male; F: Female; Ip: Intraperitoneal; Sf: Subcutaneous fat; Im: Intercostal muscle; Pc: Peritoneal cavity; HCC: Hepatocellular carcinoma; BCLC: The Barcelona Clinic Liver Cancer classification; AFP: Alpha-fetoprotein; Mod: Moderate.

were analysed using Kaplan-Meier estimates, and were

0.49% at 1 year and 1.0% at 2 years (Figure 1). Table 2
shows the baseline characteristics of the 11 patients with
seeding. Eight of these patients were male, and the mean
patient age was 56.5 9.0 years. Ten patients were hepatitis B surface antigen positive and one patient was hepatitis
C virus antibody positive. Eight patients were Child-Pugh
class A and three were Child-Pugh class B. The mean
tumour size was 4.5 1.9 cm, and the mean number
of tumours was 1.3 0.6. The mean number of needle
insertions was 3.3 1.7. Direct subcapsular needle insertions were performed in eight patients with subcapsular
tumours. Six patients underwent biopsy prior to cryoablation, of which three had poorly differentiated HCC. Four
patients were classified as BCLC stage A, six as stage B,
and one as stage C. The tumours were completely ablated
in eight patients. The mean serum AFP level was 1182.9
2668.6 ng/mL.
Seeding occurred along the cryoablation needle tract
in 10 patients, and at a distant location in 1 patient (Figure 2). The seeding was intraperitoneal in seven patients,
intraperitoneal and in the intercostal muscles in one patient, pleural in one patient, and in the abdominal wall
(subcutaneous fat) in one patient. One patient had distant
seeding in the peritoneal cavity. In ten patients, the seeded
tumours were < 3 cm in diameter, and in one patient the
tumour was 3 cm in diameter. Nine (81.8%) patients had a
single seeded tumour, and the other two (18.2%) patients
had two and three seeded tumours, respectively, indicating
that multiple seeding was not uncommon. One patient developed treatment-related liver haemorrhage 5 mo before
the seeding was detected.
CT and MRI are the preferred imaging modalities for
detecting needle-tract seeding. The seeded tumours are
usually detected as one or a few round or oval-shaped

0.10

Cumulative tumor seeding rate

0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00

12

24
36
48
60
72
84
Seeding time from first cryoablation (mo)

96

Figure 1 Cumulative tumour seeding rate. The cumulative rate was 0.49%
at 1 year and 1.0% at 2 years.

RESULTS
Clinical features of patients with tumour seeding
A total of 1436 HCC patients underwent 3015 cryoablation sessions (1215 guided by ultrasonography and 221
guided by CT; average 2.1 sessions per patient) for 2423
nodules. When a patient underwent more than one treatment session, the data from the initial session were used.
During the follow-up period (median 18 mo; range 1-90
mo), seeding was diagnosed in 11 patients at an interval
of 1-24 (median 6.0) mo after the first cryoablation. The
seeding rate was 0.76% per patient (11/1436). The longest
interval between the first cryoablation session and detection of seeding was 2 years. The cumulative seeding rates

WJG|www.wjgnet.com

6590

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

Figure 2 Locations of hepatocellular carcinoma and of seeding after cryoablation. A: Intraperitoneal seeding (case 4); B: Intraperitoneal and intercostal muscle
seeding (case 6); C: Seeding in subcutaneous fat (case 3); D: Complete ablation of primary hepatocellular carcinoma (HCC) (case 11); E and F (case 11): Multiple,
small, intraperitoneal seeding nodules 1 mo later; G: Complete ablation of primary HCC (case 5); H: Pleural seeding nodules 1 mo later (case 5). Arrowheads indicate
primary HCCs, arrows indicate seeding.

enhancing nodules along the needle tract, with a few occurring at a distant location. Seeded tumours showed a
similar imaging pattern to primary HCCs, with arterial
phase hyperattenuation followed by portal-venous phase
hypoattenuation (Figure 3).
Tumour biopsies performed before cryoablation in
six patients who developed seeding showed that one
patient had a well differentiated tumour, two patients
had moderately differentiated tumours, and three patients had poorly differentiated tumours. The seeded
tumours in the nine patients without distant or pleural
seeding were confirmed by biopsy and histopathological
examination. The seeded tumour showed similar differentiation features to the primary HCC in four of these
patients (Figure 4).
Seeding was treated by PEI in four patients, resection
in one patient, cryoablation in two patients, cryoablation
plus sorafenib in two patients, and conservative treatment
in two patients. Of the nine patients with seeding who
were treated, recurrence of seeding after treatment occurred in five (55.6%), including three treated with PEI
and two treated with cryoablation plus sorafenib. Three

WJG|www.wjgnet.com

of the patients with recurrent seeding were treated with

RFA plus radiation and did not have further recurrence,
and the other two patients were treated with sorafenib
plus cryoablation and TACE (Figure 5).
At the end of the follow-up period, five patients with
seeding had died. The causes of death were intrahepatic
HCC progression and liver failure. No patient died of
their seeded tumour nodules. In patients with seeding,
the cumulative survival rates were 90% at 1 year, 68% at 2
years, 53% at 3 years, 32% at 4 years and 32% at 5 years.
In patients without seeding, the cumulative survival rates
were 86% at 1 year, 61% at 2 years, 51% at 3 years, 43% at
4 years and 34% at 5 years. There were no significant differences in survival rates between the two patient groups (P
= 0.942) (Figure 6).
Risk factors for tumour seeding
Table 3 shows the results of the univariate analysis to
identify risk factors for seeding. Direct subcapsular needle
insertion (P = 0.0043) and subcapsular tumour location
(P = 0.0152) were associated with seeding. There were
no significant associations between seeding and age, sex,

6591

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

Figure 3 Computed tomography showing tumour seeding in case 2 after cryoablation for hepatocellular carcinoma, with a history of transcatheter arterial chemoembolisation. A: Contrast-enhanced computed tomography (CT) image showing a 3-cm diameter hepatocellular carcinoma (HCC) in segment before
cryoablation (black arrow); B and C: Contrast-enhanced CT images during the arterial phase (B) and portal-venous phase (C) showing intraperitoneal seeding (white
arrow) at 2 mo after biopsy and percutaneous cryoablation. The tumour showed hyperattenuation during the arterial phase and hypoattenuation during the portalvenous phase, similar to the primary HCC. Histopathological examination of the seeded tumour showed a poorly differentiated HCC. Note that the intrahepatic tumour
was completely ablated.

Figure 4 Computed tomography image and histopathology of primary

and seeded tumours (case 10). The
58-year-old woman underwent biopsy
and percutaneous cryoablation for a
2-cm diameter subcapsular hepatocellular carcinoma (HCC). A: The tumour
was completely ablated (black arrow);
B: Intraperitoneal seeding at 18 mo
after treatment (white arrow); C: Histopathological examination of the biopsy
specimen from the primary tumour
showed a well differentiated HCC; D:
The seeded tumour was resected, and
histopathological examination showed
a well differentiated HCC. The patient
was alive and tumour-free at 60 mo
after cryoablation.

viral markers, Child-Pugh class, tumour size, number of

nodules, number of sessions, number of needle insertions, tumour differentiation, biopsy prior to cryoablation,
BCLC stage, incomplete ablation, or serum AFP level.
Even though there was no significant association between seeding and tumour differentiation, seeding was
detected earlier in patients with poorly differentiated
HCC than in those with well or moderately differentiated HCC [1.33 0.577 mo vs 11.12 6.896 mo; 95%CI:
(-19.115)-(-0.468); P = 0.042)]. Multivariate analysis
showed that the only significant risk factor for seeding
was direct subcapsular needle insertion (odds ratio 2.57;
95%CI: 1.47-7.65; P = 0.017).

WJG|www.wjgnet.com

DISCUSSION
The new modality of imaging-guided percutaneous argonhelium cryoablation has been widely developed in China.
Many studies have reported the safety and efficacy of this
technique in the treatment of HCC[14-16]. Although many
complications have been reported, the majority are minor
and can be treated conservatively. In carefully selected patients, the rate of serious complications is low[16]. Because
of its minimal invasiveness and resulting large ablation
zone, percutaneous cryoablation is a useful treatment modality for HCC[19].
However, occasional tumour seeding after percuta-

6592

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

Figure 5 Patient treated with sorafenib plus cryoablation and transcatheter arterial chemoembolisation for hepatocellular carcinoma seeding (case 7). A:
Complete ablation of the subcapsular hepatocellular carcinoma (arrowhead); B: Multiple, small, intraperitoneal seeded nodules (white arrows); C: Seeded nodules
were treated with cryoablation (black arrows) plus sorafenib; D: Recurrence of seeding (white arrow); E: The recurrent seeding was treated with cryoablation (black
arrows); F: The seeding was also treated with transarterial chemoembolisation (white arrow). Six months later, there was no further recurrence.

period, 580 patients had died, and these patients had a median survival time of 15 (range 1 to 65) mo. All patients
were under close observation, and no patients were lost to
follow-up. Because of the duration and quality of followup, the likelihood of having missed a seeded tumour in
this study is minimal.
Similar to other percutaneous interventions such as
biopsy, PEI and RFA, the sites of seeding after cryoablation were the thoracic wall, abdominal wall, diaphragm and
peritoneal cavity[20-22]. Seeding usually occurred along the
needle tract, but a few cases were at a distant location, with
pleural and peritoneal cavity seeding in one patient each.
Regular follow-up with contrast-enhanced CT or MRI
from the chest to the pelvis is therefore very important.
The median time to diagnosis of seeding has been
reported to be 13 (range 1-58) mo after biopsy[21], 6 (range
2-48) mo after PEI[21] and 28.5 (range 8.6-60.7) mo after
RFA[22]. In the present study, the median time to diagnosis
was 6.0 (range 1-24) mo after cryoablation. The longest
interval from the first cryoablation session to the diagnosis of seeding was 2 years. It can therefore be concluded
that it is necessary to carefully monitor patients for at least
2 years after cryoablation for HCC. The reason for the
longest interval to the diagnosis of seeding in the present
study being shorter than in previous studies on biopsy,
PEI and RFA is unknown. There are no reports of growth
rates for seeded tumours after cryoablation, but it has
been reported that the growth rate of needle-tract seeding
after biopsy varies depending on the initial number of implanted tumour cells and the doubling time of the tumour,
as well as the microenvironment surrounding the seeded
tumour. The doubling time of seeded tumours after bi-

1.0

Cumulative survival

0.8

0.6
Patients without seeding
0.4
Patients with seeding

P = 0.942

0.2

0.0

12
24
36
48
60
72
84
Overall survival from first cryoablation (mo)

96

Figure 6 Cumulative overall survival in patients with and without tumour

seeding. Survival curves are drawn according to the Kaplan-Meier method, using the log-rank test.

neous cryoablation for HCC remains unavoidable, as in

other local ablation treatments such as PEI, RFA and
MW ablation. In this study, we systematically searched for
evidence of seeding using state-of-the-art imaging over a
long follow-up period, and found seeding in 11 of 1436
patients treated with 3015 cryoablation sessions. The
seeding rate was 0.76% per patient. The median interval
between the first cryoablation session and detection of
seeding was 6 (range 1-24) mo, with a median follow-up
time of 18 (range 1-90) mo. At the end of the follow-up

WJG|www.wjgnet.com

6593

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

Table 3 Characteristics of patients with and without tumour
seeding

Variable
Age (yr)1
Gender (male/female)
HBs-Ag positive only
HCV-Ab positive only
Both positive
Both negative
Child-Pugh class (A/B)
Tumour size (cm)1
Number of tumours1
Number of sessions1
Number of needle insertions1
Direct subcapsular insertion
Subcapsular location
Biopsy performed2
Poorly differentiated tumour
BCLC stage (A/B/C)
Completed ablation (yes/no)
AFP (ng/mL)1

Seeding was
Seeding was
not identified P value
identified
(n = 1425)
(n = 11)
56.5 9.0
55.5 9.3
0.8970
8/3
1168/257
0.6895
10
1219
0.9412
1
167
1.0000
0
19
1.0000
0
20
1.0000
8/3
866/559
0.6176
4.5 1.9
4.6 3.0
0.8800
1.3 0.6
1.3 0.7
0.9520
2.2 1.2
2.3 1.1
0.8764
3.3 1.7
3.8 1.1
0.3430
8
205
0.0043
8
476
0.0152
6
730
0.8264
3
243
0.6674
4/6/1
783/447/195 0.2869
8/3
1161/264
0.7235
1182.9 2668.6 577.0 2038.1 0.3270

Figure 7 The sheath introducer system. The cryoablation needle (black arrow)
is inserted and removed through the sheath introducer system (white arrow).

the present study, univariate analyses identified subcapsular tumour location and direct subcapsular needle insertion as risk factors. There were no significant associations between seeding and age, sex, viral markers, ChildPugh class, tumour size, number of nodules, number of
sessions, number of needle insertions, tumour differentiation, biopsy prior to cryoablation, BCLC stage, incomplete ablation or serum AFP level. Multivariate analysis
showed that only direct subcapsular needle insertion was
an independent risk factor for seeding.
Several studies have reported that subcapsular tumour
location was a risk factor for seeding[2,27-29]. In a study reporting a 12.5% seeding rate after RFA, all patients with
seeding had a subcapsular tumour[2]. This is consistent
with the results of our univariate analysis. In our initial
experience, percutaneous cryoablation of subcapsular
tumours was also associated with liver haemorrhage[16].
In the present study, treatment-related liver haemorrhage
occurred in one patient with seeding. We therefore insert
our cryoprobe across a portion of normal hepatic parenchyma, and avoid direct subcapsular needle insertion for
subcapsular tumours whenever possible. This minimises
both liver haemorrhage and needle-tract seeding. This
may explain why multivariate analysis only identified direct
subcapsular needle insertion as an independent risk factor.
There is still controversy regarding whether tumour biopsy prior to treatment or a poorly differentiated tumour
increase the risk of seeding[21-23,30]. In this study, biopsy
and a poorly differentiated tumour were not associated
with a higher rate of seeding. The current study also did
not show a significant association between seeding and tumour size or incomplete cryoablation, which is consistent
with the findings of other studies[21,26]. Although the 3-mm
cryoprobe was large, the risk of seeding after cryoablation
was small. The risk of seeding may be reduced by the use
of the sheath introducer system (Figure 7), through which
cryoablation needles are inserted and removed. Similarly,
Maturen et al[31] reported that no seeding occurred when
they used a needle introducer that remained in position
during multiple passes of a coaxial cutting needle for biopsies, which may protect the tissue along the needle tract
and reduce seeding. Further studies should be conducted

Values are expressed as mean SD; 2Of 736 cases in which biopsy was
performed. HBs-Ag: Hepatitis B surface antigen; HCV-Ab: Hepatitis C virus antibody; BCLC: The Barcelona Clinic Liver Cancer classification; AFP:
Alpha-fetoprotein.

opsy is 112 (range 22-415) d[8], which is comparable to

that of primary HCC. Regarding differentiation features,
Matsukuma et al[23] reported that peritoneal seeding can infrequently proliferate aggressively with more differentiated
features. In the current study, four patients with seeding
showed similar tumour differentiation features to the primary HCC, and only three patients had poorly differentiated HCC, so it is difficult to explain the shorter interval in
terms of tumour differentiation. The available data report
a median HCC diameter of 32.5 mm in patients who underwent PEI and 30 mm in patients who underwent RFA,
and the use of 14- to 22-gauge needles for biopsies, 21- to
22-gauge needles for PEI and 14- to 17-gauge needles for
RFA[20,21]. The shorter interval to detection of seeding after cryoablation may be related to the use of larger needles
(a 3-mm cryoprobe correlates to an 11-gauge needle) and
larger nodule size (mean 45 19 mm).
HCC is particularly prone to seeding, with higher
seeding rates after biopsy (0%-5.1%)[24] than other solid
tumours such as pancreatic tumours (0.003%-0.017%)
and other abdominal tumours (0%-0.03%)[9,25].
In a recent review, several factors were suggested to
contribute to seeding after percutaneous interventional
procedures, which were listed as follows: poorly differentiated tumour, high serum AFP level, subcapsular
tumour location, biopsy prior to RFA, high number of
sessions and high number of electrode placements[19].
However, only Shirai et al[22] and Imamura et al[26] have reported multivariate analyses of these factors. Imamura et
al[26] reported that only poor tumour differentiation was
an independent risk factor for seeding. Shirai et al[22] reported that only RFA was an independent risk factor. In

WJG|www.wjgnet.com

6594

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

and knowledge of seeding and its risk factors are helpful

for minimising its occurrence.

to assess the effects of the sheath introducer system on

the risk of seeding in percutaneous cryoablation for HCC.
In addition, the low risk of cryoablation may be related to
the mechanisms of cryoablation. Cryotherapy is believed
to kill cells by several mechanisms, including intracellular
ice formation, solute-solvent shifts that cause cell dehydration and rupture, small vessel obliteration causing hypoxia
and specific anti-tumour immunoreactions that limit tumour growth[32,33]. Several studies found that cryoablation
resulted in both local tumour necrosis and necrosis and
shrinkage of the tissues adjacent to the tumour, which
was thought to indicate ectopic tumour suppression[33].
Preclinical evidence of a cryo-immunologic response as
well as some clinical data indicate that cryoablation may
generate an anti-tumour response[34,35]. Our previous study
indicated that cryoablation not only directly destroys
malignant tissues, but also has effects on the adjacent
tissues[36]. Cryotherapy resulted in reduced numbers of
peripheral Treg cells and a lowering of the CD8-FoxP3+/
CD8+FoxP3- ratio in malignant tissues[37]. We therefore
speculate that anti-tumour immunoreactions induced by
cryoablation may limit seeding. This concept deserves further study.
Although poorly differentiated tumour did not increase the risk of seeding, we found that seeding occurred
earlier in patients with poorly differentiated HCC than in
those with well or moderately differentiated HCC. It is
possible that poorly differentiated HCC lacks cohesiveness[17] and grows more rapidly, allowing the seeding to be
identified earlier.
It is not clear whether seeding affects prognosis. Shirai
et al[22] and Imamura et al[26] investigated the prognosis of
HCC patients with seeding after RFA. They reported that
the survival rate was not particularly low in patients with
seeding, and that seeding itself did not directly affect survival. The present study also did not find significant differences in the cumulative survival rates of patients with
and without seeding. By the end of the follow-up period,
five patients with seeding had died of intrahepatic HCC
progression and liver failure. No patient died due to the
growth of seeded nodules. Nevertheless, the survival rates
of patients with seeding tended to be lower from the second year onwards. The reasons for the lack of significant
differences in survival rates may be as follows: First, the
number of patients with seeding was very small compared
with the number without seeding. Second, seeded tumours
were treated radically, which may improve outcome[7,38]. It
is therefore impossible to claim that seeding does not affect prognosis, and seeded tumours should be treated with
the aim of achieving local cure. It is essential to recognise
that seeding is difficult to treat successfully. In the present
study, the recurrence rate after local radical treatment of
seeding was 55.6%.
In conclusion, the relatively low rate of tumour seeding after cryoablation for HCC is considered an acceptable clinical risk. Direct puncture of subcapsular tumours
was found to be a risk factor for seeding. Although seeding is sometimes unavoidable, strict attention to detail

WJG|www.wjgnet.com

COMMENTS
COMMENTS
Background

Imaging-guided percutaneous argon-helium cryoablation is widely used in China, and this technique has been found to be safe and effective for the treatment
of hepatocellular carcinoma (HCC). However, details of tumour seeding after
this procedure have not been reported to date, even though seeding is one of
the most important complications.

Research frontiers

This study reports the rate of tumour seeding after percutaneous cryoablation
and analyses the risk factors for seeding in a large cohort of HCC patients who
were treated with cryoablation sessions over an 8-year period.

Innovations and breakthroughs

Seeding occurred in 11 (0.76%) of 1436 patients treated with percutaneous

cryoablation in this study. Only direct puncture of a subcapsular tumour was an
independent risk factor for seeding.

Applications

This study indicates that the risk of seeding after percutaneous cryoablation
for HCC is small and is considered an acceptable clinical risk. This procedure
is minimally invasive and results in a large ablation zone, making it a useful
treatment modality for HCC. However, direct puncture of a subcapsular tumour
should be avoided. The small risk of seeding may be due to the use of an introducer sheath, or to the mechanisms of cryoablation, and further research is
warranted.

Terminology

Percutaneous cryoablation requires the insertion of needle into the liver parenchyma and tumour, which may cause tumour seeding. However, the incidence
of HCC seeding after the procedure is low.

Peer review

The authors analyzed the incidence of HCC tumour seeding after percutaneous
cryoablation. It is very interesting study and has a great scientific value for physicians who take care of patients with this pathology. The study is well designed
and data is convincing.

REFERENCES
1
2

5
6

6595

Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma.

Lancet 2003; 362: 1907-1917
Llovet JM, Vilana R, Br C, Bianchi L, Salmeron JM, Boix L,
Ganau S, Sala M, Pags M, Ayuso C, Sol M, Rods J, Bruix
J. Increased risk of tumor seeding after percutaneous radiofrequency ablation for single hepatocellular carcinoma.
Hepatology 2001; 33: 1124-1129
Ishii H, Okada S, Okusaka T, Yoshimori M, Nakasuka H,
Shimada K, Yamasaki S, Nakanishi Y, Sakamoto M. Needle
tract implantation of hepatocellular carcinoma after percutaneous ethanol injection. Cancer 1998; 82: 1638-1642
Tarantino L, Francica G, Esposito F, Pisaniello D, Parmeggiani D, Marzullo G, Sordelli IM, Sperlongano P. Seeding
from hepatocellular carcinoma after percutaneous ablation:
color Doppler ultrasound findings. Abdom Imaging 2006; 31:
69-77
Smith EH. The hazards of fine-needle aspiration biopsy. Ultrasound Med Biol 1984; 10: 629-634
Bonatti H, Bodner G, Obrist P, Bechter O, Wetscher G, Oefner D. Skin implant metastasis after percutaneous radio-frequency therapy of liver metastasis of a colorectal carcinoma.
Am Surg 2003; 69: 763-765
Yu J, Liang P, Yu XL, Cheng ZG, Han ZY, Dong BW. Needle
track seeding after percutaneous microwave ablation of malignant liver tumors under ultrasound guidance: analysis of
14-year experience with 1462 patients at a single center. Eur J
Radiol 2012; 81: 2495-2499

December 7, 2012|Volume 18|Issue 45|

Wang CP et al . HCC seeding related to cryoablation

8

9
10

11

12

13
14

15

16

17
18
19
20

21

22

23
24

Chang S, Kim SH, Lim HK, Lee WJ, Choi D, Lim JH. Needle
tract implantation after sonographically guided percutaneous
biopsy of hepatocellular carcinoma: evaluation of doubling
time, frequency, and features on CT. AJR Am J Roentgenol
2005; 185: 400-405
Smith EH. Complications of percutaneous abdominal fineneedle biopsy. Review. Radiology 1991; 178: 253-258
Kim SH, Lim HK, Lee WJ, Cho JM, Jang HJ. Needle-tract
implantation in hepatocellular carcinoma: frequency and CT
findings after biopsy with a 19.5-gauge automated biopsy
gun. Abdom Imaging 2000; 25: 246-250
Huang GT, Sheu JC, Yang PM, Lee HS, Wang TH, Chen DS.
Ultrasound-guided cutting biopsy for the diagnosis of hepatocellular carcinoma--a study based on 420 patients. J Hepatol
1996; 25: 334-338
Arriv L, Vurgait A, Monnier-Cholley L, Lewin M, Balladur P,
Poupon R, Tubiana JM. Long-term follow-up after neoplastic seeding complicating percutaneous ethanol injection for
treatment of hepatocellular carcinoma. Eur Radiol 2002; 12:
74-76
McWilliams JP, Yamamoto S, Raman SS, Loh CT, Lee EW,
Liu DM, Kee ST. Percutaneous ablation of hepatocellular carcinoma: current status. J Vasc Interv Radiol 2010; 21: S204-S213
Xu KC, Niu LZ, He WB, Guo ZQ, Hu YZ, Zuo JS. Percutaneous cryoablation in combination with ethanol injection for
unresectable hepatocellular carcinoma. World J Gastroenterol
2003; 9: 2686-2689
Wang C, Lu Y, Chen Y, Feng Y, An L, Wang X, Su S, Bai W,
Zhou L, Yang Y, Xu D. Prognostic factors and recurrence
of hepatitis B-related hepatocellular carcinoma after argonhelium cryoablation: a prospective study. Clin Exp Metastasis
2009; 26: 839-848
Yang Y,Wang C,Lu Y,Bai W,An L,Qu J,Gao X,Chen
Y,Zhou L,Wu Y,Feng Y,Zhang M,Chang X,Lv J. Outcomes
of ultrasound-guided percutaneous argon-helium cryoablation of hepatocellular carcinoma. J Hepatobiliary Pancreat Sci
2011 Dec 21; Epub ahead of print
Edmondson HA, Steiner PE. Primary carcinoma of the liver:
a study of 100 cases among 48,900 necropsies. Cancer 1954; 7:
462-503
Llovet JM, Br C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999;
19: 329-338
Hinshaw JL, Lee FT. Cryoablation for liver cancer. Tech Vasc
Interv Radiol 2007; 10: 47-57
Chang S, Kim SH, Lim HK, Kim SH, Lee WJ, Choi D, Kim
YS, Rhim H. Needle tract implantation after percutaneous
interventional procedures in hepatocellular carcinomas: lessons learned from a 10-year experience. Korean J Radiol 2008; 9:
268-274
Stigliano R, Marelli L, Yu D, Davies N, Patch D, Burroughs
AK. Seeding following percutaneous diagnostic and therapeutic approaches for hepatocellular carcinoma. What is the
risk and the outcome? Seeding risk for percutaneous approach of HCC. Cancer Treat Rev 2007; 33: 437-447
Shirai K, Tamai H, Shingaki N, Mori Y, Moribata K, Enomoto
S, Deguchi H, Ueda K, Maekita T, Inoue I, Iguchi M, Yanaoka
K, Oka M, Ichinose M. Clinical features and risk factors of extrahepatic seeding after percutaneous radiofrequency ablation
for hepatocellular carcinoma. Hepatol Res 2011; 41: 738-745
Matsukuma S, Sato K. Peritoneal seeding of hepatocellular
carcinoma: clinicopathological characteristics of 17 autopsy
cases. Pathol Int 2011; 61: 356-362
Stigliano R, Burroughs AK. Should we biopsy each liver

25

26

27

28

29

30

31

32
33

34

35
36

37

38

mass suspicious for HCC before liver transplantation?--no,

please dont. J Hepatol 2005; 43: 563-568
Fornari F, Civardi G, Cavanna L, Di Stasi M, Rossi S, Sbolli
G, Buscarini L. Complications of ultrasonically guided fineneedle abdominal biopsy. Results of a multicenter Italian
study and review of the literature. The Cooperative Italian
Study Group. Scand J Gastroenterol 1989; 24: 949-955
Imamura J, Tateishi R, Shiina S, Goto E, Sato T, Ohki T, Masuzaki R, Goto T, Yoshida H, Kanai F, Hamamura K, Obi S,
Yoshida H, Omata M. Neoplastic seeding after radiofrequency ablation for hepatocellular carcinoma. Am J Gastroenterol
2008; 103: 3057-3062
Livraghi T, Solbiati L, Meloni MF, Gazelle GS, Halpern EF,
Goldberg SN. Treatment of focal liver tumors with percutaneous radio-frequency ablation: complications encountered
in a multicenter study. Radiology 2003; 226: 441-451
Jaskolka JD, Asch MR, Kachura JR, Ho CS, Ossip M, Wong
F, Sherman M, Grant DR, Greig PD, Gallinger S. Needle tract
seeding after radiofrequency ablation of hepatic tumors. J
Vasc Interv Radiol 2005; 16: 485-491
Shiozawa K, Watanabe M, Wakui N, Ikehara T, Iida K, Sumino Y. Analysis of patients with tumor seeding after percutaneous radiofrequency ablation of hepatocellular carcinoma.
Mol Med Report 2008; 1: 851-855
Silva MA, Hegab B, Hyde C, Guo B, Buckels JA, Mirza DF.
Needle track seeding following biopsy of liver lesions in the
diagnosis of hepatocellular cancer: a systematic review and
meta-analysis. Gut 2008; 57: 1592-1596
Maturen KE, Nghiem HV, Marrero JA, Hussain HK, Higgins
EG, Fox GA, Francis IR. Lack of tumor seeding of hepatocellular carcinoma after percutaneous needle biopsy using coaxial cutting needle technique. AJR Am J Roentgenol 2006; 187:
1184-1187
Sabel MS. Cryo-immunology: a review of the literature and
proposed mechanisms for stimulatory versus suppressive
immune responses. Cryobiology 2009; 58: 1-11
Osada S, Imai H, Tomita H, Tokuyama Y, Okumura N, Matsuhashi N, Sakashita F, Nonaka K. Serum cytokine levels
in response to hepatic cryoablation. J Surg Oncol 2007; 95:
491-498
den Brok MH, Sutmuller RP, Nierkens S, Bennink EJ, Frielink C, Toonen LW, Boerman OC, Figdor CG, Ruers TJ,
Adema GJ. Efficient loading of dendritic cells following cryo
and radiofrequency ablation in combination with immune
modulation induces anti-tumour immunity. Br J Cancer 2006;
95: 896-905
Sabel MS, Su G, Griffith KA, Chang AE. Rate of freeze alters
the immunologic response after cryoablation of breast cancer.
Ann Surg Oncol 2010; 17: 1187-1193
Zhou L, Wang CP, Lu YY, Bai WL, Qu JH, Lou M, Zeng Z,
Wu Y, Chen Y, Wang H, Chang XJ, Yang YP. Decrease in size
of non-treated lesions after cryoablation for hepatocellular
carcinoma. Hepatogastroenterology 2012; 59: 252-254
Zhou L, Fu JL, Lu YY, Fu BY, Wang CP, An LJ, Wang XZ,
Zeng Z, Zhou CB, Yang YP, Wang FS. Regulatory T cells are
associated with post-cryoablation prognosis in patients with
hepatitis B virus-related hepatocellular carcinoma. J Gastroenterol 2010; 45: 968-978
Hashimoto K, Sasaki Y, Yokoyama S, Hiraki M, Matsumoto S,
Tokuoka M, Matsuyama J, Morita S, Morimoto T, Fukushima
Y, Nomura T, Yoshida S, Takeda M. A long-term survival
case after two resections of the peritoneal metastasis from
hepatocellular carcinoma. Gan To Kagaku Ryoho 2011; 38:
2466-2468
S- Editor Shi ZF

WJG|www.wjgnet.com

6596

L- Editor Webster JR

E- Editor Li JY

December 7, 2012|Volume 18|Issue 45|