Eur Radiol (2008) 18: 28482853

DOI 10.1007/s00330-008-1096-5 HEPATOBILI ARY-PANCREAS

Jie Tang
Faqin Lv
Wenxiu Li
Huiqin Zhang
Yukun Luo
Lichun An
Tanshi Li
Received: 16 December 2007
Revised: 24 April 2008
Accepted: 25 April 2008
Published online: 11 July 2008
# European Society of Radiology 2008
Percutaneous injection of hemostatic agents
for severe blunt hepatic trauma:
an experimental study
Abstract This study was designed to
evaluate whether percutaneous injec-
tion of hemostatic agents under the
guidance of contrast-enhanced ultra-
sound (CEUS) can stop hemorrhage
from severe hepatic trauma. Eighteen
dogs were impacted by a miniature
impactor to create blunt hepatic trau-
ma. Fourteen with appropriate liver
lesions were divided into two groups:
the treatment group (n=7) and the
control group (n=7). In the treatment
group, hemocoagulase atrox and
-cyanoacrylate were respectively
injected into the injury sites and
transected micro-vessels under the
guidance of CEUS. In the control
group, normal saline was injected into
the injury sites. CEUS and CT were
performed at 3, 7, 14, and 21 days
after the focal injection. Surviving
animals were killed on the 21st day for
pathologic examination. All animals
of the treatment group survived. Three
dogs of the control group died in the
first 24 h. In the treatment group,
CEUS and CT demonstrated that he-
patic lesions became smaller gradually
from the 3rd to the 21st day after
injection. The focal injection of he-
mostatic agents under the guidance of
CEUS can stop hemorrhage from
hepatic trauma of grade III~IVor IV.
During the period of 3 weeks, no side
effect was found.
Keywords Hepatic trauma
.
Contrast-enhanced ultrasound
.
Interventional treatment
.
Hemostatic agent
Introduction
Hepatic trauma management has evolved from operation
upon injury recognition to non-operative management in the
majority of patients [1]. Literature [2] describing hepatic and
splenic injury treatments indicates up to 67% of exploratory
operations for blunt trauma are non-therapeutic. Non-
operative management can be adopted in more than 80%
of blunt hepatic trauma cases, while immediate surgical
management is necessary in about 20% [3]. More than two-
thirds of patients with grade IV and V injuries (American
Association for Surgery of Trauma, AAST, organ injury scale
[4]) can be treated non-operatively; however, 50% of these
patients typically require interventional treatment, including
but not limited to laparotomy [5].
Because non-operative hepatic trauma management has
been extended to severe trauma patients previously
considered for surgery, the incidence of delayed, uncom-
mon and serious complications, such as delayed rupture
hemorrhage, and biliary leak [68], has increased. Thus, it
is necessary to develop minimally invasive techniques for
controlling hepatic trauma hemorrhage.
Intravenous enhancement agents, such as SonoVue
TM
(Bracco, Milan, Italy), improve conventional ultrasound
(US) sensitivity to detect and characterize focal liver lesions
[9]. Owing to the possibilities of low-mechanical index, real-
time, contrast-specific systems, US can detect contrast
medium leakage [10]. Literature [11] indicates contrast-
enhanced sonography (CEUS) is effective in blunt hepatic
trauma evaluation, as it is more sensitive and correlates better
than baseline sonography with computed tomography (CT).
Hemocoagulase atrox (Solco Basle Ltd., Birsfelden,
Switzerland), fibrin sealants, and -cyanoacrylate have
successfully stopped focal blood loss [12, 13]. For example,
J. Tang (*)
.
F. Lv
.
W. Li
.
H. Zhang
.
Y. Luo
.
L. An
Department of Ultrasound,
Chinese Peoples Liberation
Army General Hospital,
28 Fuxing Road,
Beijing, 100853, China
e-mail: tangxier@163.com
Tel.: +86-10-68282055
e-mail: lvjin8912@163.com
T. Li
Department of Intensive Care Unit,
Chinese Peoples Liberation Army
General Hospital,
Beijing, China
endoscopic injection of cyanoacrylate (Guangzhou Baiyun
Medical Adhesive Glue Co. Ltd., Guangzhou, China) can
control active bleeding from gastric varices [14].
The aim of this animal experiment was to investigate
whether focal injection of hemocoagulase atrox and -
cyanoacrylate under CEUS guidance can reduce blood loss
from liver trauma.
Materials and methods
Experiments were performed according to our institutional
animal care and use committee protocols and National
Institutes of Health guidelines for the care of laboratory
animals.
Animals and anesthesia
Eighteen 1923-kg mixed-breed dogs were used in this study.
General anesthesia was induced by an intramuscular injection
of 30 mg/kg pentobarbital sodiumperformed by a veterinarian.
Modeling blunt hepatic trauma and grouping animals
Femoral arterial and femoral venous catheters were placed:
the former was for detecting arterial pressure; the latter was
for administration of the ultrasound contrast medium.
Aminiature impactor (Fig. 1) consisting of a supporter, an
impacting handle, a piston handle, a power bullet, and a
power-actuated fastening device was employed to create
blunt hepatic trauma [15]. A power bullet was loaded first.
The power-actuated fastening device was fixed into a hole of
the crossbeam of the supporter. After the impacting handle
was inserted into the piston barrel, the impact was performed
by moving the trigger. The force from the power bullet
bombing pushed the piston handle and the impacting handle.
As a result, the designated target organ was impacted.
CEUS and CTwere performed immediately after impact
to determine hepatic trauma and detect the size and extent
of the lesion site, graded according to the AASTs organ
injury scale[4].
Equipment and methods of direct injection
under CEUS guidance
The ultrasound contrast agent used in this study was
SonoVue (Bracco, Milan, Italy). Conventional ultra-
sound and CEUS were performed by an ACUSON Sequoia
512 ultrasound system (Siemens Medical Solutions,
Mountain View, CA). Plain liver scanning was performed
using the SOMATOM Sensation 64 CT scanner (Siemens
Medical Solutions) followed by spiral enhancement scan-
ning with 3-mm layer thickness.
At least 20 min usually elapse after trauma before the
patient arrives at the hospital [16]. Therefore, focal
injection treatment was performed 20 min after impact.
Once CEUS confirmed the hepatic injury site, transected
micro-vessels, active hemorrhage, and the pathway of
transcutaneous puncture (using an 18-gauge needle) in the
treatment group, hemocoagulase atrox (1 KU in 2 ml of
saline) was injected into the injury sites under contrast
pulse sequencing (CPS) guidance. Then, -cyanoacrylate
(1 ml) was injected into the transected micro-vessels under
micro-vascular display (MVD) guidance. In the control
group, normal saline (3 ml) was injected.
Hemostatic agents
Hemocoagulase atrox is obtained from Brazils spearhead
agkistrodon, a genus of venomous pit vipers, after
segregation and purification. It can make blood platelets
aggregate and stop wound bleeding. -cyanoacrylate, a
medical adhesive with cohering, sealing, and consolidating
properties, undergoes polymerization upon contact with an
organic protein, producing an exothermic reaction and
forming a glue membrane in the lesion site.
CEUS and CTwere performed immediately and on days
3, 7, 14, and 21 after focal injection. On day 21, survivors
were killed and underwent laparotomy. Hepatic tissue was
collected for gross and histopathological examinations.
Statistical analysis
All measurements are presented as means SD. Differ-
ences between group means were determined with analysis
of variance (Version 13 SPSS, Inc., Chicago, IL) with
Fig. 1 Construction plan of the impactor, consisting of a supporter,
an impacting handle, a piston handle, a power bullet, and a power
actuated fastening tool. The piston handle height was regulated from
2050 cm depending on the different impacting force from 300 N-
5.7 KN
2849
Students t test or nonparametric test, where applicable.
The difference was considered statistically significant at
p<0.05.
Results
General materials in the two groups
Four dogs were excluded from the experiment: two had
hepatic lesions no more than 3 cm deep and two died
immediately after impact of hemorrhagic shock (one from
vena cava injury; one from main portal vein injury).
Fourteen dogs remained for study with appropriate liver
injuries of grade III-IV or IV. They were divided into the
treatment group (n=7) or the control group (n=7). There
was no difference in body weight between the groups
(21.751.77 kg and 22.181.65 kg, P=0.65).
Efficacy of focal injection under CEUS guidance
Pre-impact, pre-focal injection, and post-focal injection
mean arterial pressures were compared between the
treatment group and the control group (Table 1). Pre-
impact and pre-focal injection mean arterial pressures did
not vary between groups (P=0.8453, P=0.8174). Post-
focal injection mean arterial pressure in the control group
was significantly lower than the treatment group (P=
0.000). The mean arterial pressure of post-focal injection
was not different in the treatment group compared with pre-
impact (P=0.0596) and pre-focal injection (P=0.2744). In
the control group, the mean arterial pressure turned low
gradually (P=0.00010.0165).
CEUS demonstrated anechoic perfusion defect regions
and transected micro-vessels of the liver lesions before
percutaneous injection in all 14 dogs (Fig. 2a). Hepatic
lesion sizes in both showed no difference (P=0.810.98)
(Table 2). Active hemorrhage, appearing as contrast
material pooling or flowing out of the liver capsule,
occurred in four dogs in the treatment group and in three
dogs in the control group. In the injection procedure, the
puncture needle was inserted precisely into the lesion site,
active hemorrhage region, or the transected micro-vessel
under real-time CEUS guidance (Fig. 2b,c).
After the percutaneous injection, three dogs in the
control group died within 24 h. Four dogs in the control and
all dogs in the treatment group survived for 21 days. Lesion
sizes were recorded by CEUS, demonstrated by CT during
the 21-day follow-up (Fig. 3a,b). Hepatic lesions in the
Table 1 Comparison of systolic arterial pressures between the treatment group and the control before and after percutaneous focal injection
(mmHg, x s)
Pre-impact Pre-focal injection Post-focal injection
Treatment group 188.8612.5 169.579.7 175.8610.82
Control group 187.5711.7 170.8610.73 135.8611.32
Note: In the treatment group, the systolic arterial pressure before focal injection was lower than pre-impact (P=0.0073). The systolic artery
pressure after focal injection was not different from those before impact and before focal injection (P=0.0596, 0.2744). In the control, the
systolic artery pressure turned low gradually from pre-impact to post-focal injection (P=0.0001~0.0165)
Fig. 2 The liver lesion on CEUS and the focal injection under the
guidance of CEUS. a Liver lesion appeared as anechoic perfusion
defect regions (arrows) before focal injection. b In the injection
procedure, the puncture needle (fine arrow) was inserted into the
injury region (thick arrows) under the guidance of CPS. c The
puncture needle (fine arrow) was inserted into the transected micro-
vessel (thick arrow) under guidance of MVD. The puncture needle
was hyperechoic on CEUS
2850
treatment group gradually became smaller and disappeared
(Fig. 4a-d) (Table 3).
During follow-up, no side effects were observed in the
treatment group. On day 21, the body weight of the control
group was (17.832.32) kg, compared to (22.971.69) kg
of the treatment group (P=0.0021).
Pathological findings
Laparotomy showed the great omentum had moved
superiorly and partially covered the liver lesion. The
intestine was not affected. Scars on the liver surface were
visualized in the treatment group. Similar to the control
group, histopathology demonstrated normal hepatic lobule
texture, mild congestion of the hepatic sinusoids, fibrous
tissue with mild hyperplasia, and some inflammatory cells
infiltrating under the liver capsule.
Discussion
The key to management is to control wound bleeding. With
non-operative therapy gaining acceptance, some other
imaging techniques facilitate decreasing bleeding, primary
surgery obviation, and treatment of complications [1719].
Second-generation contrast agents make it is possible to
evaluate abdominal parenchymal organ trauma with
CEUS. Moreover, the puncture needle can be percuta-
neously inserted exactly into hepatic lesions under sono-
graphic guidance [20]. Therefore, it is feasible to perform
percutaneous injection therapy of hepatic trauma under
CEUS guidance.
Hemocoagulase atrox has been successful in patients
with low-flow hemorrhage, while -cyanoacrylate has
been used for adhesion and hemostasis in various
operations, including the nervous system, trachea, esoph-
agus, and vessels [21, 22], and as a focal embolism agent in
conventional sclerotherapy of variceal bleeding [23]. At
present, -cyanoacrylate has been extended to operation
incisions of abdominal organs previously considered for
suture closure [24].
In our experiment, hemocoagulase atrox was injected
first into the hepatic lesion region. Then, -cyanoacrylate
was injected into the damaged micro-vessel or bleeding
site. While hemocoagulase atrox can form soft blood clots
and temporarily stop bleeding, it has no adhesive or seal
effect on hepatic lacerations. The site may re-bleed,
requiring re-treatment. -cyanoacrylate worked effec-
tively when injected directly into the bleeding site where
hemocoagulase atrox had slowed or stopped bleeding.
When -cyanoacrylate is used alone, with blood overflow
it cannot uniformly adhere to the wounds surface. But
together, they can decrease bleeding from severe hepatic
Table 2 Before the percutaneous injection the sizes of hepatic lesions measured by CEUS in both groups were compared (x s, cm, [range])
n Largest diameter Orthogonal diameter
Treatment group 7 5.281.36 [3.46.5] 4.581.23 [2.55.6]
Control group 7 5.311.33 [3.56.3] 4.331.56 [2.46.2]
Fig. 3 CT images of liver lesions before focal injection and on day
21 after focal injection in the treatment group. a Arrows indicate the
liver lesion. b At 21 days after focal injection, CT showed the liver
lesion (arrow) had become obviously smaller than before manage-
ment. The lesion border line was not clear
2851
trauma, as our preliminary experiments under laparotomy
demonstrate.
Percutaneous injection of hemocoagulase and -cya-
noacrylate is effective in controlling hemorrhage from
grade III-IV or IV blunt hepatic trauma. All dogs in the
treatment group recovered without side effects during the
21-day observation.
Gross and histopathologic examinations demonstrated
hepatic wounds in the treatment group healed without
abnormal effect on surrounding organs and tissues.
There were several limitations of this experiment. First,
blunt hepatic injury models were created by a miniature
impactor, but while this instrument was simple, convenient,
and effective for modeling blunt abdominal parenchymal
organ trauma, its reproducibility was not satisfying and
would cause waste of animals. Second, the room time for
CEUS in this experiment was 6 min at most. Generally, the
contrast agent was first used for hepatic lesion detection
and set puncture pathway, followed by focal injection of
hemocoagulase atrox, -cyanoacrylate, and normal saline.
Fig. 4 Liver lesion in the treatment group gradually turned small
and finally disappeared after therapy by CEUS. a The liver lesion
appeared as anechoic perfusion defect regions (arrows) before the
focal injection, and its size was 5.4 cm3.2 cm. The contrast
material pooling that presented irregular hyperechoic in the lesion
site was considered as active hemorrhage. b and c At day 3 and day
7 after therapy, the lesions turned small gradually from 5.3 cm
3.1 cm-4.3 cm2.6 cm with an unclear border line. d At 21 days
after therapy, the liver lesion disappeared. The arrow indicates that
the hyperechoic region was formed by the injected -cyanoacrylate
Table 3 CEUS showed hepatic lesions in treatment group became smaller during the period of 21 days (x s, cm)
Immediate Day 3 Day 7 Day 14 Day 21
Largest diameter 5.361.27 5.431.18 5.030.79 4.230.62 1.250.23
Orthogonal diameter 4.521.41 4.380.94 3.780.52 2.860.48 1.140.85
2852
Therefore, this therapy required adequate operator exper-
tise. Otherwise, the CEUS imaging cost would increase
because of contrast medium consumption. (The price for
the 5.0 ml dose is currently 700 RMB.)
In conclusion, CEUS-guided injection of hemocoagu-
lase atrox and -cyanoacrylate may be a fast, feasible,
effective modality for controlling hemorrhage from severe
hepatic trauma.
Acknowledgement The financial support of the Military Medical
Science Program (no. 06G108) is gratefully acknowledged.
References
1. Mooney DP (2002) Multiple trauma:
liver and spleen injury. Curr Opin
Pediatr 14:482485
2. Ochsner MG (2001) Factors of failure
for nonoperative management of blunt
liver and splenic injuries. World J Surg
25:13931396
3. Letoublon C, Arvieux C (2002)
Nonoperative management of blunt
hepatic trauma. Minerva Anestesiol
68:132137
4. Moore EE, Cogbill TH, Jurkovich GJ et
al (1995) Organ injury scaling: spleen
and liver (1994 revision). J Trauma
38:323324
5. Carrillo EH, Wohltmann C, Richardson
JD et al (2001) Evolution in the treat-
ment of complex blunt liver injuries.
Curr Probl Surg 38:160
6. Trunkey DD (2004) Hepatic trauma:
contemporary management. Surg Clin
North Am 84:437450
7. Galvan DA, Peitzman AB (2006) Fail-
ure of nonoperative management of
abdominal solid organ injuries. Curr
Opin Crit Care 12:590594
8. Zyromski NJ, Lillemoe KD (2006)
Current management of biliary leaks.
Adv Surg 40:2146
9. Thorelius L (2004) Contrast-enhanced
ultrasound for extrahepatic lesions:
preliminary experience. Eur J Radiol 51
(Suppl):S31S38
10. Catalano O, Cusati B, Nunziata A et al
(2006) Active abdominal bleeding:
contrast-enhanced sonography. Abdom
Imaging 31:916
11. Catalano O, Lobianco R, Raso MM et
al (2005) Blunt hepatic trauma:
evaluation with contrast-enhanced
sonography: sonographic findings and
clinical application. J Ultrasound Med
24:299310
12. Bahar I, Kaiserman I, Slomovic A et al
(2007) Fibrin glue for opposing wound
edges in top hat penetrating kerato-
plasty: a laboratory study. Cornea
26:12351238
13. Maluf Filho F, Sakai P, Ishioka S et al
(2001) Endoscopic sclerosis versus
cyanoacrylate endoscopic injection for
the first episode of variceal bleeding: a
prospective, controlled, and rando-
mized study in Child-Pugh class C
patients. Endoscopy 33:421427
14. Greenwald BD, Caldwell SH,
Hespenheide EE et al (2003) N-2-butyl-
cyanoacrylate for bleeding gastric
varices: a United States pilot study and
cost analysis. Am J Gastroenterol
98:19821988
15. Li Z, Dong XZ, You FS et al (2006)
Establishment of animal model of
intraperitoneal bleeding in rabbits.
Chinese Journal of Comparative Med-
icine 16:217219
16. Holcomb JB, McClain JM, Pusateri AE
et al (2000) Fibrin sealant foam sprayed
directly on liver injuries decreases
blood loss in resuscitated rats. J Trauma
49:246250
17. Carrillo EH, Richardson JD (2001) The
current management of hepatic trauma.
Adv Surg 35:3959
18. Goffette PP, Laterre PF (2002) Trau-
matic injuries: imaging and interven-
tion in post-traumatic complications
(delayed intervention). Eur Radiol
12:9941021
19. Rogers CG, Knight V, MacUra KJ et al
(2004) High-grade renal injuries in
childrenis conservative management
possible? Urology 64:574579
20. Raza SA, Walser E, Hernandez A et al
(2006) Transhepatic puncture of portal
and hepatic veins for TIPS using a
single-needle pass under sonographic
guidance. AJR Am J Roentgenol 187:
W87W91
21. Elgazzar RF, Addulmajeed I,
Mutabbakani M (2007) Cyanoacrylate
glue versus suture in peripheral nerve
reanastomosis. Oral Surg Oral Med
Oral Pathol Oral Radiol Endo
104:465472
22. Canonico S, Campitiello F, Santoriello
A et al (2001) Sutureless skin closure in
varicose vein surgery: preliminary re-
sults. Dermatol Surg 27:306308
23. Thabut D, Bernard Chabert B (2007)
Management of acute bleeding from
portal hypertension. Best Pract Res
Clin Gastroenterol 21:1929
24. Fotiadis C, Leventis I, Adamis S et al
(2005) The use of isobutyl-cyanoacry-
late as a tissue adhesive in abdominal
surgery. Acta Chir Belg 105:392396
2853