1. SURGICAL RESEARCH
IMPACT OF BLOOD FLOW OCCLUSION ON LIVER NECROSIS FOLLOWING
THERMAL ABLATION
MEHRDAD NIKFARJAM, VIJAYARAGAVAN MURALIDHARAN, CATERINA MALCONTENTI-WILSON,
WENDY MCLAREN AND CHRISTOPHER CHRISTOPHI
Department of Surgery, University of Melbourne, Austin Hospital, Melbourne, Victoria, Australia
Background: Laser, radiofrequency and microwave are common techniques for local destruction of liver tumours by thermal
ablation. The main limitation of thermal ablation treatment is the volume of necrosis that can be achieved. Blood flow occlusion is
commonly advocated as an adjunct to thermal ablation to increase the volume of tissue necrosis based on macroscopic and
histological assessment of immediate or direct thermal injury. This study examines the impact of blood flow occlusion on direct
and indirect laser induced thermal liver injury in a murine model using histochemical methods to assess tissue vitality.
Methods: Thermal ablation produced by neodymium yttrium-aluminium-garnet laser (wavelength 1064 nm) was applied to the
liver of inbred male CBA strain mice at 2 W for 50 s (100 J). Treatment was performed with and without temporary portal vein and
hepatic artery blood flow occlusion. Animals were killed upon completion of the procedure to assess direct thermal injury or at 24, 48
and 72 h to assess the progression of tissue damage. The maximum diameter of necrosis was assessed by vital staining for nic-
otinamide adenine dinucleotide (NADH) diaphorase. Microvascular changes were assessed by laser Doppler flowmetry, confocal
in vivo microscopy and scanning electron microscopy.
Results: The direct thermal injury (mean SE) assessed by NADH diaphorase staining was significantly greater following thermal
ablation treatment without blood flow occlusion than with blood flow occlusion (3.3 (0.4) mm vs 2.9 (0.3) mm; P = 0.005). Tissue
disruption, cracking and vacuolization was more pronounced adjacent to the fibre insertion site in the group treated with thermal
ablation combined with blood flow occlusion. There was an equivalent increase in the extent of injury following therapy in both
groups that reached a peak at 48 h. The maximum diameter of necrosis in the thermal ablation alone group at 48 h was significantly
greater than the thermal ablation combined with blood flow occlusion group (5.8 (0.4) mm vs 5.3 (0.3) mm; P = 0.011). The patterns
of microvascular injury were similar in both groups, varying in extent.
Conclusion: Temporary blood flow inflow occlusion appears to decrease the extent of initial injury measured by vital staining
techniques and does not alter the time sequence of progressive tissue injury following thermal ablation therapy.
Key words: blood flow occlusion, microvasculature, progressive injury, thermal ablation, tissue necrosis, vital staining.
Abbreviations: H&E, haematoxylin and eosin; NADH, nicotinamide adenine dinucleotide; Nd:YAG, neodymium
yttrium-aluminium-garnet; SEM, scanning electron microscopy.
INTRODUCTION
Thermal ablation produced by laser, radiofrequency and micro-
wave are thermal ablative techniques used for the treatment of
liver tumours.1 Heat energy applied to tumours can produce com-
plete destruction and in selected patients achieves long-term sur-
vival results that is comparable to surgical resection.1,2 One factor
that limits the wider application of thermal ablation is the max-
imum volume of tumour necrosis that can be reliably achieved
with a single optical fibre. Hepatic blood flow is thought to im-
pede the expansion of thermal injury through its heat dissipating
effect. A reduction in ‘heat sink’ effect is postulated to allow an
expansion of the volume of tissue that can be ablated. Studies
have shown varied results, but generally report an increase in
tissue injury with thermal ablation combined with hepatic blood
flow occlusion.3–12 The majority of positive studies assess imme-
diate or direct tissue injury by macroscopic and histological
evaluation of tissue injury. Accurate assessment of injury by
histochemical staining of tissue vitality is seldom reported.1,2
Previous studies have shown that tissue injury following ther-
mal ablation treatment occurs in two distinct phases.13 There is
an initial direct thermal effect that is followed by a progressive
increase in tissue injury. The effect of blood flow occlusion com-
bined with thermal ablation on the progression of liver injury is,
however, undefined. This study uses histochemical staining tech-
niques of tissue vitality to determine the effect of hepatic blood
flow occlusion on the extent and time sequence of tissue injury in
liver parenchyma after thermal ablation treatment by laser. It also
assesses the impact of therapy on microvascular injury.
METHODS
Animals
Male inbred CBA strain mice, 6–8 weeks of age, were used in
all experiments. Animals were housed in standard cages with
access to irradiated food and water ad libitum and exposed to
a 12 hour light–dark cycle. All procedures were performed
according to the guidelines of the Austin Hospital Animal Ethics
Committee.
M. Nikfarjam MBBS; V. Muralidharan PhD, FRACS; C. Malcontenti-
Wilson BSc; W. McLaren PhD, BSc; C. Christophi MD, FRACS.
Correspondence: Dr Mehrdad Nikfarjam, Department of Surgery, University
of Melbourne, Austin Hospital, Lance Townsend Building Level 8, Studley Road,
Heidelberg, Vic. 3084, Australia.
Email: surgery-armc@unimelb.edu.au
Accepted for publication 5 May 2005.
ANZ J. Surg. 2006; 76: 84–91 doi: 10.1111/j.1445-2197.2006.03559.x
Ó 2006 Royal Australasian College of Surgeons

2. Thermal ablation
A neodymium yttrium-aluminium-garnet laser (Nd:YAG, wave-
length 1064 nm) (Dorniermedilas fibertom 4100; Medizintechnik,
Mu¨nchen, Germany) was used as a focal heat source. Mice were
anaesthetized with a mixture consisting of ketamine 100 mg/kg
(Parke Davis, New South Wales, Australia) and xylazine 10 mg/
kg (Bayer, New South Wales, Australia) by intraperitoneal in-
jection. Carprofen was administered subcutaneously as a long-
acting analgesic 5 mg/kg (Pfizer, New South Wales, Australia).
A bilateral subcostal incision was used to fully expose the liver.
A 400 lm bare tip optical quartz fibre was used to deliver laser
energy into the substance of the left or right liver lobe parenchyma
in predetermined locations. The region of tissue coagulation could
bevisualizedon theliver surface.The laserwasactivated atapower
of 2 W for 50 s, depositing a total energy of 100 J. This chosen
setting was determined in preliminary experiments to produce an
immediate area of tissue necrosis, 2–3 mm diameter, without char-
ring or dramatic temperature fluctuations under normal conditions
or when combined with blood flow occlusion. The amount of liver
ablated at the stated settings is equivalent to 5–8% total liver
volume, based on our previous studies.14 At the completion of
the procedure the liver was removed in cases where immediate
injury was assessed or else 2 mL of 0.9% saline was instilled into
the peritoneum prior to abdominal closure and animals were killed
at subsequent time points.
Blood flow occlusion
Blood flow occlusion was achieved by placement of a titanium
aneurysm clip across the hepatoduodenal ligament. This achieved
complete occlusion of the portal vein and hepatic artery supplying
the liver. Thermal ablation by laser was delivered into the liver
substance 1 min following blood flow occlusion, with the clip
removed immediately following the procedure. Effective reduc-
tion in hepatic blood flow by inflow occlusion was verified
initially in non-treated animals (n = 5) using laser Doppler
flowmetry (Oxyflo 2000; Oxford Optronix Oxford, UK).
Experimental groups
Animal requirements were based on preliminary studies estimat-
ing that a minimum of 10 animals were required in each group
to detect a 10–15% difference in the diameter of necrosis with
thermal ablation treatment alone or combined with blood flow
occlusion achieving a power of 0.8 and P < 0.05. An additional
10 animals were required in each group for microvascular studies.
Study 1: immediate injury
The immediate effect of thermal ablation on liver tissue was
assessed by inducing laser injury alone (n = 20) or with inflow
blood occlusion (n = 20). Animals were killed immediately fol-
lowing each procedure and analysed to determine morphological
tissue changes and diameter of maximum tissue necrosis. Micro-
vascular changes were assessed using laser Doppler flowmetry,
scanning electron microscopy (SEM) and confocal in vivo micro-
scopy. Oxygen tension and tissue temperature measurements were
performed 2 mm from the site of laser application.
Study 2: progressive injury
Thermal ablation by laser was performed on liver as previously
described with and without blood flow occlusion. Animals in each
group (n = 20) were allowed to recover and were subsequently
killed at 24, 48 or 72 h. Previous experiments had confirmed that
maximum injury occurs by 48 h following thermal ablation by
laser.13 Morphologic tissue changes and maximum diameter of
necrosis was assessed at each time point. Microvascular changes
were assessed by laser Doppler flowmetry and SEM at one or
more of the timepoints.
Assessment
Tissue injury
Histological assessment of tissue necrosis. Laser treated tissue
was cut perpendicular to the fibre insertion tract. Specimens were
fixed in 10% formalin for 24 h and embedded in paraffin. They
were then sectioned at 3 lm thickness and stained with haema-
toxylin and eosin (H&E). Histological assessment of tissue mor-
phology was performed using an Olympus light microscope
(BHT; Olympus, Tokyo, Japan).
Histochemical assessment of tissue vitality. Heat-induced tissue
injury was assessed by nicotinamide adenine dinucleotide
(NADH) diaphorase staining as an early and accurate marker of
mitochondrial activity.15,16 Laser treated regions were removed,
embedded in optimum cutting temperature medium (Tissue-Tek,
New South Wales, Australia), snap frozen and stored at –80°C.
Cryostat sections (10 lm) were mounted onto glass slides and
incubated for 15 min in a humidity chamber at room temperature
with a test solution containing 1 mL of reduced NADH (2.5 mg/
mL), 2.5 mL nitroblue tetrazolium chloride, 1 mL phosphate buff-
ered saline and 0.5 mL of Ringer’s solution. Sections covered
with a permanent aqueous mountant (PMT030; Scytek Laborato-
ries, Logan, UT, USA), dried at 70°C for 20 min and mounted in
DePeX (Gurr; BDH Laboratories, Poole, UK). Vital stains with
NADH diaphorase mitochondrial activity stained blue. The max-
imum diameter of necrosis symmetrical to the fibre insertion site
was determined by light microscopy using image analysis soft-
ware (Image-Pro Plus Version 4.5; Media Cybernetics, San Diego,
CA, USA).
Microvasculature studies
Laser doppler flowmetry. The relative blood flow in each group
(n = 5) was recorded at 2 mm and 4 mm from the fibre insertion
site in animals immediately following thermal ablation treatment
by laser Doppler flowmetery. Similar experiments were per-
formed at 72 h following heat application, by which time max-
imum diameter of necrosis had occurred. The left anterior lobe of
the liver was chosen as the normal reference point for relative
blood flow measurements. Blood flow recordings were made
immediately beneath the liver capsule using a 0.5 mm diameter
probe after a period of equilibration.
Scanning electron microscopy. Microvascular injury immedi-
ately following laser therapy alone (n = 5) or combined with
blood flow occlusion (n = 5) was assessed by SEM at 72 h for
determination of maximum vascular injury by methods previously
described.17
Confocal in vivo microscopy. The acute effects of thermal abla-
tion on liver microvasculature were assessed by confocal in vivo
microscopy in animals immediately following thermal ablation
with (n = 5) and without blood flow occlusion (n = 5). This
THERMAL ABLATION WITH BLOOD FLOW OCCLUSION 85
Ó 2006 Royal Australasian College of Surgeons

3. involved administration of 0.05 mL of fluorescein isothiocyanate
dextran (MW 160 000) (Sigma, New South Wales, Australia) by
tail vein injection under general anaesthesia immediately follow-
ing the completion of thermal ablation. Animals then underwent
a laparotomy, were placed onto a viewing stage and examined
under an Olympus BH2 microscope connected to an F990e per-
sonal confocal system (OPTISCAN, Victoria, Australia). Regions of
injury were viewed at varying magnifications and continuous
recordings made over a period of 15 min. Disruption in micro-
vascular architecture, including vascular leakage, was recorded
for the treatment groups.
Oxygen tension measurement. Tissue oxygen tension measure-
ments were performed for 60 s 2 mm away from the fibre appli-
cation site immediately following laser application as an adjunct
to the other acute microvascular assessments. The measuring
probe was inserted approximately 2 mm into the liver substance
and recordings undertaken, with corrections for tissue tempera-
ture (OxyLab pO2; Oxford Optronix).
Statistical analysis
Comparative variables were expressed as mean (SE) and analysed
using SPSS (SPSS, Chicago, IL, USA) by the Mann–Whitney
U-test. Logistic regression was applied to the plot of the tissue
necrosis diameter against the time following laser thermotherapy,
with the curve of best correlation (r) determined by a statistical
software package (Microsoft CurveExpert version 1.34. A P-value
of less than 0.05 was considered statistically significant.
RESULTS
Temporary portal and hepatic inflow occlusion was successfully
performed. There was no evidence of vascular injury at the clamp
site or hepatic lobe infarction. Clear visual evidence of hepatic
ischaemia was not apparent during the short treatment period. Laser
Doppler flowmetry at 1 min following vascular occlusion showed
reductions in blood flow to 44.3 (3.8)% of preclamping values.
Study 1: immediate tissue injury
Thermal ablation alone
Coagulative tissue change was visible immediately following
laser application surrounded by a zone of hyperaemia. On H&E
histology clear morphological changes indicative of liver necrosis
were not apparent. Tissue morphology was generally well pre-
served without major tissue disruption adjacent to the fibre inser-
tion site (Fig. 1). NADH diaphorase staining was required to
clearly define the extent of injury and to differentiate vital from
necrotic tissue (Fig. 2).
Thermal ablation with blood flow occlusion
The macroscopic changes in the liver treated by thermal ablation
combined with inflow blood flow occlusion were similar to the
group treated by thermal ablation alone. NADH diaphorase
staining was utilized to accurately define the extent of injury. On
histological examination of H&E stained liver sections, major tissue
disruption, cracking and vacuolization was evident immediately
adjacent to the fibre insertion site (Fig. 1). This contrasted with
the minimal damage observed in liver treated by thermal ablation
alone.
The maximum diameter of necrosis based on NADH dia-
phorase staining was significantly greater with laser treatment
alone compared with laser treatment with blood flow occlusion
(3.3 (0.4) mm vs 2.9 (0.4) mm; P = 0.005) (Fig. 3).
Study 2: progression in tissue injury
Thermal ablation alone
The extent of injury at 24–72 h following laser treatment could be
clearly defined on macroscopic examination of the liver. Coagu-
lated tissue was pale and well demarcated from surrounding nor-
mal liver. The region of visible tissue coagulation clearly increased
compared to the initial injury. H&E stained sections showed
a region of complete tissue necrosis surrounding the fibre insertion
site with a variable leucocyte infiltration. The border between
viable and necrotic liver was well defined. NADH diaphorase
staining clearly confirmed the extent of tissue necrosis, which
correlated with the macroscopic and H&E histology findings.
Thermal ablation with blood flow occlusion
Morphological features of tissue injury at 24–72 h in animals
treated by thermal ablation and vascular occlusion was similar
to treated animals without blood flow occlusion. An increase in
extent of tissue coagulation was evident compared with the initial
treatment group (Fig. 3).
At all time points examined, the diameter of necrosis based on
NADH diaphorase measurement was significantly greater in the
thermal ablation alone treatment group, in comparison to thermal
Fig. 1. (A) Haematoxylin and
eosin (H&E) section of thermal
injury by laser alone immediately
following application adjacent to
the fibre insertion site (*) showing
early coagulative changes with mild
tissue disruption. (B) More pro-
nounced changes evident adjacent
to the fibre insertion site, with sig-
nificant cracking in when thermal
injury is caused by laser combined
with blood flow occlusion (magnifi-
cation ·310).
86 NIKFARJAM ET AL.
Ó 2006 Royal Australasian College of Surgeons

4. ablation treatment combined with blood flow occlusion (Fig. 3).
The absolute increase in tissue necrosis was equivalent between
the two groups. A quadratic function curve was established by
logistic regression to best correlate with data points in both the
thermal ablation alone (r = 0.89) and thermal ablation combined
with blood flow occlusion groups (r = 0.86), peaking at 48 h
(Fig. 3).
Microvascular changes
Microvascular injury assessment consisted of laser Doppler flow-
metry, SEM of vascular resin casts and tissue oxygen tension
measurements.
Laser Doppler flowmetry
Thermal ablation alone
Doppler flowmetery recordings 2 mm from the fibre insertion
site corresponded to the margin of coagulated liver tissue. Meas-
urement at 4 mm was in macroscopically normal tissue. There was
a dramatic relative reduction in blood flow 2 mm from the fibre
site immediately following thermal ablation (Fig. 4). The reduc-
tion in blood flow at 4 mm from the fibre site was, however, less
pronounced. At 72 h there was further reduction in blood flow at
2 mm from the fibre insertion site, in the region of fully developed
coagulative necrosis, compared to the immediate time point. The
blood flow was equivalent to normal liver at 4 mm from the fibre
insertion site at 72 h within macrocopically normal tissue, just
beyond the coagulated tissue margins.
Thermal ablation with blood flow occlusion
The relative blood flow changes following thermal ablation com-
bined with vascular inflow occlusion was similar to that observed
for treatment of liver by thermal ablation alone (Fig. 4). The
relative blood flow reduction 2 mm from the fibre insertion site
immediately after laser application was, however, significantly
greater with thermal ablation combined with blood flow occlusion
compared to thermal ablation alone (27.3 (7.6)% vs 38.5
(16.6)%); P = 0.002). Blood flow changes at later time points
were similar in the treatment groups.
Scanning electron microscopy
Thermal ablation alone
Disruption of the liver microvasculature was evident immediately
following treatment (Fig. 5). Vascular destruction was most prom-
inent adjacent to the fibre insertion site. The absolute extent of
vascular injury could not be clearly ascertained at this time point
due to the patchy nature of the microvascular damage at the mar-
gin of treated regions. Vessels greater than 60 lm were preserved
in regions of otherwise major vessel destruction. An expansion in
the area of vessel disruption was clearly evident by 72 h following
injury. The margins of injury were distinct allowing accurate meas-
urement of damage at this time point.
Thermal ablation with blood flow occlusion
The morphological changes in the extent of injury following ther-
mal ablation treatment combined with blood flow occlusion were
similar to those observed with thermal ablation therapy alone. The
Fig. 2. Nicotinamide adenine dinucleotide (NADH) diaphorase
stained cryostat section showing clear demarcation of viable, blue
staining liver, from poorly stained non-viable liver immediately
following treatment (magnification ·200).
Fig. 3. (A) Comparison of the diameter of necrosis produced by (h)
thermal ablation treatment alone and (j) thermal ablation with blood
flow occlusion at different time points. Mann–Whitney U-test. (B) A
quadratic function curve best correlates with the change in diameter
of necrosis following thermal ablation therapy alone (r = 0.89) and
thermal ablation combined with blood flow occlusion (r = 0.89). The
differences in the extent of necrosis are maintained between groups
and the peak injury occurs at 2 days following therapy.
THERMAL ABLATION WITH BLOOD FLOW OCCLUSION 87
Ó 2006 Royal Australasian College of Surgeons

5. extent of immediate injury could not be accurately assessed due to
the patchy regions of damage observed at the treatment margins.
The maximum diameter of injury at 72 h was well defined and
significantly less than the diameter of injury in the thermal abla-
tion alone treatment group (5.1 (0.091) mm vs 5.6 (0.14) mm;
P = 0.016) correlating with tissue injury findings.
Confocal in vivo microscopy
Immediate vascular injury only was assessed by confocal in vivo
microscopy.
Thermal ablation alone
A heterogeneous pattern of microvascular damage was noted
immediately following thermal ablation. A spherical zone of
avascularity was seen adjacent to the fibre insertion site
(Fig. 6). Beyond this region contrast build-up could be identified
in thrombosed vessels. Leakage of contrast material into the tissue
interstitium was evident. Regions of vascular disruption inter-
spersed among otherwise normal vessels, with gradual build-up
of contrast suggestive of vessel leakage, was identified in the
peripheral regions of injury.
Thermal ablation with blood flow occlusion
The confocal in vivo microscopy changes in the microvasculature
were morphologically similar to those in the thermal ablation
alone treatment group.
Tissue oxygenation tension
Thermal ablation alone
Oxygen tension measurements performed immediately following
treatment 2 mm from the fibre insertion site corresponded to the
margin of coagulated liver. The oxygen tension in this region was
37.6 (5.0) mmHg compared to untreated controls 67.0
(4.4) mmHg (P < 0.001).
Thermal ablation with blood flow occlusion
The tissue oxygen tension 2 mm from the fibre insertion imme-
diately following thermal ablation combined with blood flow
occlusion was 41.7 (11.1) mmHg. Statistically, this was not sig-
nificantly different to the oxygen tension of tissue in animals
treated by thermal ablation alone (P = 0.302).
DISCUSSION
Thermal ablation by laser or radiofrequency is shown to be a safe
and effective technique for the treatment of liver tumours.1,18 A
limiting factor of laser induced thermal ablation and other local
ablative techniques is the size of tissue ablation achieved by a sin-
gle fibre. Coherent light in the form of laser is absorbed by the
tissue and converted to heat. The maximum volume of tissue ne-
crosis is dictated by the balance between the radial conduction of
thermal energy and its removal by conduction and convection
effects, mostly by hepatic blood flow. Initial attempts to overcome
this ‘heat sink’ effect by increasing applied energy produced poor
results due to the occurrence of carbonization around the fibre
application site.10 Carbonization impedes the transmission of
thermal energy.10 Subsequently many studies investigated the
effects of hepatic blood flow occlusion in increasing the size of
tissue necrosis achieved.
Hepatic blood flow occlusion combined with thermal ablation
has been advocated to increase tissue injury.3,7,19–24 In most stud-
ies, these changes are based on macroscopic measurements of
coagulated tissue in animal models and lack accurate assessment
of the extent of tissue injury by vital staining techniques.
Hiesterkamp et al.4 demonstrated in a porcine model an increase
in ablation volume from 6.4 cm3 to 30.6 cm3 when laser therapy
(5 W for 6 min) was combined with hepatic blood flow occlusion.
Similar increases ranging from two- to six-fold have been dem-
onstrated with various other techniques of thermal ablation,
namely radiofrequency ablation and microwave coagulation.3,7–9
A number of clinical studies report increased ablation volumes
with hepatic blood flow occlusion at open or laparoscopic sur-
gery.6,19,25–28 These studies generally describe imaging results
without providing direct histological evidence of significant in-
creases in the volume of necrosis with blood flow occlusion.
There are also reports of thermal ablation combined with blood
flow occlusion for treatment of tumours located in the proximity
of large vessels without increased ablative effects.11,12,29 Ng
et al.12 in a porcine model showed no increased cellular destruc-
tion adjacent to major portal vessels with radiofrequency com-
bined with hepatic inflow occlusion based on histochemical
Fig. 4. (A) Bar graph of the relative reduction in blood flow at
2 mm and 4 mm from the fibre insertion site following (h) thermal
ablation alone or (j) combined with blood flow occlusion. (B) Blood
flow changes 3 days following treatment. Mann–Whitney U-test.
88 NIKFARJAM ET AL.
Ó 2006 Royal Australasian College of Surgeons

6. Fig. 5. (A) Scanning electron
microscopy of microvascular resin
casts of the liver immediately follow-
ing laser injury alone, and (B) ther-
mal ablation combined with blood
flow occlusion. The region of laser
application is demarcated by an avas-
cular zone (*). The extent of injury at
the margin (arrows) is heterogeneous
in both groups merging with normal
liver (L) sinusoids. Large vessels are
preserved within the treatment zone
(V). (C) The vascular injury is more
clearly demarcated 3 days following
injury in both laser therapy alone, and
(D) thermal ablation by laser com-
bined with blood flow occlusion
groups, varying in extent. A clear dis-
tinction is evident between the area of
injury and normal liver sinusoids
(arrowheads). Preservation of large
vessels (V) within the treatment zone
is demonstrated.
Fig. 6. (A) Confocal in vivo
microscopy following injection of
a fluorescein labelled dextran (MW
160 000). Normal liver sinusoids
can be demonstrated in untreated
liver with large draining interlobular
veins (V). Treatment by (B) thermal
ablation therapy alone, and (C) laser
with blood flow occlusion results in
an area of avascularity surrounding
the region of fibre application (*).
Immediately beyond this region
marked increased vascular permeabil-
ity is noted with leakage of contrast
material into the liver interstitium
(arrows). Heterogeneous regions of
increased contrast leakage can be
seen more peripherally (x). (D) High
magnification imaging reveals re-
gions of contrast build-up indicative
of vessel thrombosis and leakage.
Red blood cells can be seen (arrow-
heads) interspersed within the area
of leakage. Hepatocyte damage (H)
with influx of contrast material is
also evident.
THERMAL ABLATION WITH BLOOD FLOW OCCLUSION 89
Ó 2006 Royal Australasian College of Surgeons

7. staining assessment of liver vitality for adenosine 59-triphosphate.
Increased vascular and biliary complications have been associated
with thermal ablation combined with blood flow occlusion.11,12,29
The theoretical mechanisms involved in the augmentation of
injury by hepatic blood flow occlusion are generally based on
concepts of the ‘heat sink’ effect and, to a lesser extent, tissue
ischaemia. Several studies have clearly demonstrated that increased
injury with hepatic blood flow occlusion is directly related to
increased tissue temperatures attained.3 Therefore, as long as the
critical temperatures are achieved, hepatic blood flow occlusion
should not influence the extent of tissue necrosis. This has been
confirmed in some studies using radiofrequency ablation.11
The main mechanism of tissue injury in thermal ablation is
thought to be the production of heat, although other mechanisms
may play a part.30 Thermal energy that is applied to tissue is
absorbed by the target tissue to cause direct injury. It has been
demonstrated the direct injury is then followed by a progressive
expansion of the damage that is unrelated to the direct thermal
effects.13 Peak injury in normal liver occurs 24–72 h following
therapy, as demonstrated in this study. Other studies also demon-
strate an expansion of tissue injury following thermal abla-
tion.3,22,31,32 However, the impact of blood flow inflow occlusion
on this progressive tissue injury has not been previously described.
The treatment protocol in this study was designed to emulate
the type of injury that is expected to occur clinically at the outer
margins of treated tumours, where blood flow characteristics are
similar to normal liver.11,12,29 Temporary blood flow occlusion at
1 min reduced blood flow to 44% of preclamping levels based on
laser Doppler flowmetry. Clamping of the hepatic artery and por-
tal vein in other studies reduces blood flow to 10–20% of pre-
clamping values.33,34 Similar reductions occur in our model if
measurements are made 10–15 min following clamping, once
there is system equilibration. There is also macroscopic evidence
of hepatic ischaemia and appreciable mortality with more pro-
longed clamping. Treatment was commenced 1 min following
clamping in our study, prior to system equilibration to minimize
ischaemic injury. Blood flow reduction was expected to reduce
heat dissipation at the periphery of treated areas and consequently
produce greater necrosis. However, the extent of injury based
NADH diaphorase staining was less when blood flow inflow
occlusion was implemented, although the extent of tissue disrup-
tion adjacent to the fibre insertion site was greater in the group
treated with laser thermal ablation combined with blood flow
occlusion. Studies demonstrate that tissue optical properties can
significantly influence laser induced tissue injury.35,36 It is pos-
sible that blood flow occlusion under the conditions described may
have altered the optical and thermal conduction properties of the
liver parenchyma and limited the penetration and dissipation of
laser induced thermal energy and consequently the extent of
injury. This would explain the greater tissue disruption adjacent
to the fibre insertion site in the blood flow occlusion group, with
greater energy deposition in the proximity of the activated fibre.
The patterns of initial microvasculature injury based on
Doppler flowmetry, SEM, confocal in vivo microscopy and oxy-
gen tension measures were similar in both treatment groups. The
relative reduction in blood flow immediately following therapy on
laser Doppler flowmetry was, however, greater at the margin of
coagulated tissue following thermal ablation combined with
blood flow occlusion than with thermal ablation alone. This
may be the result of initially greater focal tissue damage adjacent
to the fibre insertion site. These differences resolved by 3 days
following therapy.
Progression in injury was observed in both treatment groups
following the direct thermal effects. This evolved over 24–72 h
and was characterized by progressive tissue necrosis on assess-
ment of NADH diaphorase stained sections and progressive
microvascular damage based on SEM of vascular resin casts.
Blood flow occlusion did not appear to alter this progressive
injury, with the time course and absolute increase in the progres-
sion of injury equivalent in both groups. This finding suggests that
progression of tissue damage occurs independently of blood flow
occlusion following the initial injury.
In conclusion, this study demonstrates the quantitative effects
of thermal ablation and blood flow occlusion on the extent and
progression of injury and its effects on the microvasculature.
Unlike other studies, temporary blood flow occlusion resulted
in a reduction in the extent of initial injury based on vital staining
techniques, possibly due to greater energy deposition adjacent to
the laser fibre. Overall, there was no apparent effect on the pro-
gression of injury with blood flow inflow occlusion with equiva-
lent increases in tissue damage following therapy. A greater
understanding of the mechanisms involved in progressive tissue
injury and manipulation of these processes may be useful in maxi-
mizing injury. The use of blood flow occlusion in a clinical set-
ting may not provide additional increases in the volume of
necrosis in all circumstances. It may significantly reduce the
extent of the direct injury at the periphery. Until the mechanisms
of injury are further investigated, the routine use of hepatic blood
flow occlusion to complement thermal ablation must be tempered
with caution.
REFERENCES
1. Nikfarjam M, Christophi C. Interstitial laser thermotherapy for
liver tumours. Br. J. Surg. 2003; 90: 1033–47.
2. Vogl TJ, Straub R, Eichler K, Sollner O, Mack MG. Colorectal
carcinoma metastases in liver: laser-induced interstitial thermo-
therapy – local tumor control rate and survival data. Radiology
2004; 230: 450–8.
3. Wiersinga WJ, Jansen MC, Straatsburg IH et al. Lesion progres-
sion with time and the effect of vascular occlusion following
radiofrequency ablation of the liver. Br. J. Surg. 2003; 90: 306–12.
4. Heisterkamp J, van Hillegersberg R, Mulder PG, Sinofsky EL, Jn
IJ. Importance of eliminating portal flow to produce large intra-
hepatic lesions with interstitial laser coagulation. Br. J. Surg.
1997; 84: 1245–8.
5. Moller PH, Hannesson PH, Ivarsson K, Olsrud J, Stenram U,
Tranberg KG. Interstitial laser thermotherapy in pig liver: effect
of inflow occlusion on extent of necrosis and ultrasound image.
Hepatogastroenterology 1997; 44: 1302–11.
6. Murakami T, Shibata T, Ishida T et al. Percutaneous microwave
hepatic tumor coagulation with segmental hepatic blood flow
occlusion in seven patients. AJR Am. J. Roentgenol. 1999; 172:
637–40.
7. Patterson EJ, Scudamore CH, Owen DA, Nagy AG, Buczkowski
AK. Radiofrequency ablation of porcine liver in vivo: effects of
blood flow and treatment time on lesion size. Ann. Surg. 1998;
227: 559–65.
8. Rossi S, Garbagnati F, De Francesco I et al. Relationship between
the shape and size of radiofrequency induced thermal lesions and
hepatic vascularization. Tumori 1999; 85: 128–32.
9. Takamura M, Murakami T, Shibata T et al. Microwave coagula-
tion therapy with interruption of hepatic blood in- or outflow: an
experimental study. J. Vasc. Interv. Radiol. 2001; 12: 619–22.
90 NIKFARJAM ET AL.
Ó 2006 Royal Australasian College of Surgeons

8. 10. Muralidharan V, Malcontenti-Wilson C, Christophi C. Effect of
blood flow occlusion on laser hyperthermia for liver metastases.
J. Surg. Res. 2002; 103: 165–74.
11. Shen P, Fleming S, Westcott C, Challa V. Laparoscopic radio-
frequency ablation of the liver in proximity to major vasculature:
effect of the Pringle maneuver. J. Surg. Oncol. 2003; 83: 36–41.
12. Ng KK, Lam CM, Poon RT, Shek TW, Fan ST, Wong J. Delayed
portal vein thrombosis after experimental radiofrequency abla-
tion near the main portal vein. Br. J. Surg. 2004; 91: 632–9.
13. Nikfarjam M, Malcontenti-Wilson C, Christophi C. Focal hyper-
thermia produces progressive tumor necrosis independent of the
initial thermal effects. J. Gastrointest. Surg. 2005; 9: 410–17.
14. MuralidharanV,NikfarjamM,Malcontenti-WilsonC,ChristophiC.
Interstitial laser hyperthermia and the biological characteristics
of tumor: study in a murine model of colorectal liver metastases.
J. Clin. Laser Med. Surg. 2003; 21: 75–83.
15. Neumann RA, Knobler RM, Leonhartsberger H, Bohler-
Sommeregger K, Gebhart W. Histochemical evaluation of the
coagulation depth after argon laser impact on a port-wine stain.
Lasers Surg. Med. 1991; 11: 606–15.
16. Heisterkamp J, van Hillegersberg R, Zondervan PE, IJzermans J.
Metabolic activity and DNA integrity in human hepatic meta-
stases after interstitial laser coagulation (ILC). Lasers Surg. Med.
2001; 28: 80–6.
17. Kuruppu D, Christophi C, O’Brien PE. Microvascular architec-
ture of hepatic metastases in a mouse model. HPB Surg. 1997;
10: 149–57; discussion 58.
18. Curley SA, Cusack JC, Tanabe KK, Ellis LM. Advances in the
treatment of liver tumors. Curr. Probl. Surg. 2002; 39: 449–571.
19. Goldberg SN, Hahn PF, Tanabe KK et al. Percutaneous radio-
frequency tissue ablation: does perfusion-mediated tissue cool-
ing limit coagulation necrosis? J. Vasc. Interv. Radiol. 1998; 9:
101–11.
20. Chinn SB, Lee FT, Kennedy GD et al. Effect of vascular occlu-
sion on radiofrequency ablation of the liver: results in a porcine
model. AJR Am. J. Roentgenol. 2001; 176: 789–95.
21. Chang CK, Hendy MP, Smith JM, Recht MH, Welling RE.
Radiofrequency ablation of the porcine liver with complete
hepatic vascular occlusion. Ann. Surg. Oncol. 2002; 9: 594–8.
22. Matsumoto R, Selig AM, Colucci VM, Jolesz FA. Interstitial
Nd:YAG laser ablation in normal rabbit liver: trial to maximize
the size of laser-induced lesions. Lasers Surg. Med. 1992; 12:
650–8.
23. Fujitomi Y, Kashima K, Ueda S, Mori H, Uchida Y. Histopatho-
logical features of liver damage induced by laser ablation in
rabbits. Lasers Surg. Med. 1999; 24: 14–23.
24. Aschoff AJ, Merkle EM, Wong V et al. How does alteration of
hepatic blood flow affect liver perfusion and radiofrequency-
induced thermal lesion size in rabbit liver? J. Magn. Reson.
Imaging 2001; 13: 57–63.
25. Ishida T, Murakami T, Shibata T et al. Percutaneous microwave
tumor coagulation for hepatocellular carcinomas with interrup-
tion of segmental hepatic blood flow. J. Vasc. Interv. Radiol.
2002; 13: 185–91.
26. de Baere T, Bessoud B, Dromain C et al. Percutaneous radio-
frequency ablation of hepatic tumors during temporary venous
occlusion. AJR Am. J. Roentgenol. 2002; 178: 53–9.
27. Washburn WK, Dodd GD, Kohlmeier RE et al. Radiofrequency
tissue ablation: effect of hepatic blood flow occlusion on thermal
injuries produced in cirrhotic livers. Ann. Surg. Oncol. 2003; 10:
773–7.
28. Verhoef C, Kuiper JW, Heisterkamp J, de Man RA, Pattynama
PM, Jn IJ. Interstitial laser coagulation with temporary hepatic
artery occlusion for patients with cirrhosis and irresectable
hepatoma. Br. J. Surg. 2003; 90: 950–5.
29. Denys AL, De Baere T, Mahe C et al. Radio-frequency tissue
ablation of the liver: effects of vascular occlusion on lesion diam-
eter and biliary and portal damages in a pig model. Eur. Radiol.
2001; 11: 2102–8.
30. Castren-Persons M, Schroder T, Lehtonen E. Sensitivity to
Nd:YAG induced laserthermia is a cell-type-specific feature
not directly related to tumorigenic potential or proliferation rate.
Lasers Surg. Med. 1996; 18: 420–8.
31. Matthewson K, Coleridge-Smith P, O’Sullivan JP, Northfield TC,
Bown SG. Biological effects of intrahepatic neodymium:
yttrium-aluminum-garnet laser photocoagulation in rats. Gastro-
enterology 1987; 93: 550–7.
32. Sweetland HM, Wyman A, Rogers K. Evaluation of the effect on
the normal liver of interstitial laser hyperethermia using artificial
sapphire probes. Lasers Med. Sci. 1993; 8: 99–105.
33. Wheatley AM, Almond NE, Stuart ET, Zhao D. Interpretation of
the laser Doppler flow signal from the liver of the rat. Microvasc.
Res. 1993; 45: 290–301.
34. Arvidsson D, Svensson H, Haglund U. Laser-Doppler flowmetry
for estimating liver blood flow. Am. J. Physiol. 1988; 254: G471–6.
35. Svaasand LO, Boerslid T, Oeveraasen M. Thermal and optical
properties of living tissue: application to laser-induced hyper-
thermia. Lasers Surg. Med. 1985; 5: 589–602.
36. Ritz JP, Roggan A, Germer CT, Isbert C, Muller G, Buhr HJ.
Continuous changes in the optical properties of liver tissue dur-
ing laser-induced interstitial thermotherapy. Lasers Surg. Med.
2001; 28: 307–12.
THERMAL ABLATION WITH BLOOD FLOW OCCLUSION 91
Ó 2006 Royal Australasian College of Surgeons