Comparison of Ex-vivo Perfused and Non-perfused Porcine Liver

Ablations using Uncooled Microwave Applicators

Mattia Dimitri

1,*

, Fabio Staderini

, Sara Aquino

, Lucrezia Mazzantini

, Andrea Corvi

and Guido Biffi Gentili

Department of Industrial Engineering, University of Florence, via di Santa Marta 3, Florence, Italy

Department of Experimental and Clinical Medicine, University of Florence, S. Marco Square 4, Florence, Italy

Department of Information Engineering, University of Florence, via di Santa Marta 3, Florence, Italy

Keywords: Perfusion, Ex-vivo, Liver, Thermal Ablation, Microwave.

Abstract: The paper compares thermal ablation on perfused and not perfused porcine liver produced by 17 and 14

Gauge non internally cooled (NIC) microwave (MW) applicators. To the knowledge of the authors this

comparison, already made using cooled applicators has not so far been carried out with uncooled applicators

that have a very different thermo-kinetic behavior when they operate inside a biological tissue. The purpose

of this preliminary study is to explore the possibility of using ex vivo perfused liver in order to define an

optimal protocol to allow more reliable translation of the experiments to the clinical practice and compare

the obtained results with those of previous studies that used similar energy delivery.

1 INTRODUCTION

A typical MW system consists of a generator and

almost one minimally invasive applicator with its

cable, however there are no two equal MW systems,

in particular as regards the geometry of the

applicator.

There are several technological approaches that

every doctor should adopt to plan and perform a

microwave thermal ablation ensuring that the tumor

is completely treated with sufficient ablative

margins (> 0.5 cm).

Thanks to continuous advances in microwave

technology, manufacturers of MW systems are

proposing increasingly advanced minimally invasive

technologies (Ruiter 2019, Yung 2017, Meloni

2017) with the valuable support of experienced

clinical interventionists.

To minimize costs and the need for multiple

insertions, the current tendency of most

manufacturers is to use a single high powered (up to

140W) water or gas cooled applicator to create a

pseudo-spherical ablation with up to 5 cm diameter

in less than 10 minutes (Kodama 2018).

 Corresponding Author: Mattia Dimitri Address: Via di

Santa Marta 3, Florence (FI), Italy. Phone Number:

+39 3407024067. e-Mail: mattia.dimitri@unifi.it

Moreover, this high-energy approach is not

without risks, especially when the treatment involves

tumors that are very close to delicate organs, which

must be preserved from thermal damage.

A possible alternative to this method is to

subdivide the energy needed to destroy the tumor

among multiple applicators (Biffi Gentili 2014). In

this case the input power of each applicator can be

reduced to the level that make the use of cooling

unnecessary.

Non Internally Cooled applicators are

structurally simpler and robust than the Internally

Cooled ones (IC), and also more reliable because

they operate at limited power (40 W maximum) and

are free from cooling system failures that can cause

serious damage to healthy tissues.

Shape and volume of the ablation zone after MW

treatment are depending on (De Cobelli 2017):

biophysical parameters as thermal conductivity and

perfusion rate of the liver parenchyma that can be

different in human liver tissue due to fibrosis,

cirrhosis or steatosis; effective MW power and

hyperthermic treatment duration; structure, gauge

and cooling mechanism (if present) of the applicator

(MW needle).

Planning for ablation is essentially based on

manufacturer algorithms or ablation charts in

combination with personal experience of the

118

Dimitri, M., Staderini, F., Aquino, S., Mazzantini, L., Corvi, A. and Gentili, G.

Comparison of Ex-vivo Perfused and Non-perfused Porcine Liver Ablations using Uncooled Microwave Applicators.

DOI: 10.5220/0008947501180123

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 1: BIODEVICES, pages 118-123

ISBN: 978-989-758-398-8; ISSN: 2184-4305

operator. These algorithms and charts, which try to

predict the three-dimensional diameter of the

ablation zone in relation to the amount of applied

energy, are often based on experiments which have

serious shortcomings, preventing a reliable

translation to daily clinical practice.

These shortcomings are the result of studies

performed in non-perfused ex vivo bovine or porcine

livers (as opposed to perfused in vivo human livers

with variable arterial and portal blood flow).

These differences affect the way in which the

applied energy is transferred into heat, resulting in

highly unpredictable ablation zone dimensions and

volumes.

Despite several individual papers reporting on

these shortcomings, a systematic review on this

topic is lacking.

A higher reproducibility of the experiments with

better predictability of the clinical results could be

achieved by defining a standardized procedure to be

shared among all manufactures, allowing to extract

the treatment parameters through ablation tests on

normothermic (body temperature) perfused ex-vivo

organs.

2 MATERIALS AND METHOD

Ablation procedures where performed using the

TATO thermal ablation system with 14 and 17

gauge NIC applicators.

TATO is a multi-applicator system that has been

developed by a small multidisciplinary team at the

University of Florence, in the framework of a

collaborative academic-industry agreement with

Biomedical Srl, Florence.

The experiment was performed in ex vivo fresh

porcine liver retrieved from animals in the food

chain. In this way important ethical implications can

be overcome and the economic burden is lower than

more common in vivo studies.

The Experimental platform for hepatic flow

simulation developed at the Industrial Engineering

Department (DIEF) of the University of Florence

(patent pending) was employed to maintain the

explanted porcine liver in a normothermic

physiological perfusion state.

Freshly taken porcine liver from adult animal

with intact in-and outflow vessels where obtained

from an abattoir and immediately connected to the

perfusion platform. Hepatic inflow was simulated

trough flexible plastic tubes sutured to the veins and

connected to a perfusion pump system. The hepatic

flow was established to emulate the average human

cardiac output. To acquire information on the

maximum temperature reached by the MW

applicators during thermo ablation procedures and

the divergences between active perfusion and

blocked perfusion, the test process was video

recorded with a Thermal-CAM A320 thermal

imaging camera (FLIR Systems, Inc., Wilsonville,

OR).

3 PERFUSION PLATFORM

The experimental platform for the simulation of the

hepatic flow is composed of 3 distinct sections, with

the possibility of independent activation /

management, which perform the following

functions:

1. pre-heating the blood and filtering any clots;

2. perfusion of the venous tree;

3. perfusion of the arterial tree, equipped with a

high efficiency oxygenator.

Each section is controlled by a dedicated processor

which, following the automatic learning approach,

processes the sensor signals (pressure, temperature,

flow) placed in strategic areas and performs the

actions necessary to maintain the normal

physiological conditions of the organ during the

extracorporeal perfusion process.

In the literature are described analogous systems

which perform similar functions but not in an

integrated and adaptive manner such as the current

system that was designed and built at the prototype

level at the DIEF of UNIFI (R. Ravikumar 2016).

and for which a patent application was filed.

4 THE ABLATION PROCEDURE

The ablation procedure was performed in

collaboration by a surgeon (FS) and an engineer

(MD).

As a first step, the perfusor was replenished by

filling the main tank with the perfusion fluid which

has the same characteristics as normal blood. In

order to evaluate the functioning of the system

avoiding the use of blood, a fluid was identified that

was able to simulate its behavior. The blood has a

dynamic viscosity of about 3 cP. To make a fluid

capable of reproducing the rheological properties of

blood in terms of viscosity and density, it is possible

to use distilled water, glycerol and cornstarch. Since

the blood has a physiological temperature of about

Comparison of Ex-vivo Perfused and Non-perfused Porcine Liver Ablations using Uncooled Microwave Applicators

119

37 °C, the percentages of the constituent elements of

the fluid must therefore be determined. With a

capillary viscometer, it is possible to measure the

kinematic viscosity [m2⁄s], from which it is possible

to calculate the dynamic viscosity [cP]. It was

estimated that at 37 °C the fluid with a viscosity as

close as possible to the blood is composed of:

 70% distilled water; 30% vegetable glycerol.

Dynamic viscosity μ = 3.52 ± 0.01 cP

 70% distilled water; 30% vegetable glycerol; 1%

cornstarch. Dynamic viscosity μ = 3.52 ± 0.15

cP.

Therefore, since about 3.5 L of fluid is circulated in

the system in question, it is possible to simulate

blood using 2450 mL of distilled water and 1050 mL

of glycerol (and 35 mL of cornstarch).

The main pump was then activated and the

heating system was switched on to gradually

increase the fluid temperature up to 38 °. Gradual

heating of the fluid is necessary to avoid thermal

shock and clot formation. The temperature was then

monitored through the RP thermocouple and the RA

thermocouple and kept constant until the liver

cannulation was completed.

The liver was removed immediately after

slaughtering the pig. The hepatic hylum was

carefully dissected. The main hepatic artery was

isolated and cannulated with a flexible 8Fr Nelaton

catheter. The choledocic duct was isolated and

sutured. The main portal vein was isolated and

cannulated with a 16 fr Nelaton catheter. The lower

cave vein has been identified and sutured. The upper

cave vein was cannulated with a 45fr plastic tube.

The suprahepatic accessory vein was isolated,

cannulated and connected to the upper cave vein

tube with a Y-shaped joint.

Before the connection to the perfusor, the hepatic

vessels were washed with a heparinized

physiological solution to prevent the formation of

clots.

The hepatic artery and the portal vein cannulas

were connected to the respective pumps and the

upper vena cava tube was connected to the main

reservoir. At the end of this procedure the perfusor

was turned on and the correct functionality was

checked.

Fig. 1 shows the final installation immediately

before the thermal ablation tests.



Figure 1: Final set up immediately before the thermal

ablation tests.

4.1 Thermal Ablation Tests

For the microwave thermal ablation tests two NIC

applicators of 14 and 17 G were used respectively.

In order not to influence the results the two needles

were inserted in two distant hepatic lobes. The

TATO system microwave generator was then set to

supply 40 W to the 14 G needle and 30 W to the 17

one. The TATO microwave generator was switched

on for 10 minutes immediately after the perfusor

reached the normothermia state.

At the end of this first procedure the perfusion

was stopped and the needles were repositioned to

repeat the same procedure but in the absence of

perfusion.

At the end of both procedures the parenchyma

was sectioned longitudinally to photograph the

obtained ablations and to measure their diameters

along the two main axes (Fig. 2, 3).



Figure 2: Post ablation sectioning.

5 PRELIMINARY RESULTS

The results of the ablation test are summarized in

Table 1.

Table 1: Preliminary results wit perfusion blocked.

Applicator

Gauge

Length

(cm)

Diameter

(cm)

Ablation

volume

(cm

)

Aspect

ratio

14 4.2 3.3 191 0.66

17 4 2.2 81 0.55

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

120

Table 2: Preliminary results with perfusion active.

Applicator

Gauge

Length

(cm)

Diameter

(cm)

Ablation

volume

(cm

)

Aspect

ratio

14 3.7 2.2 74 0.59

17 3.3 1.5 31 0.45

Without hepatic perfusion, the 14G needle

generated a 4.2x3.3 cm ablation (Fig. 3a and 3b)

while in perfused mode the ablation was smaller due

to the heat sink phenomenon, with dimensions of

3.7x2.2 cm (Fig. 3c and 3d).



Figure 3: Comparison between 14 Gauge ablation in non-

perfused (Fig. a and b) and in perfused (Fig c and d) liver.

The 17G needle produced an ablation of 4x2.2

cm in non-perfused mode (Fig. 4a and 4b) and

3.3x1.5 in perfused mode (Fig. 4c and 4d).



Figura 4: Comparison between 17 Gauge ablation in non-

perfused (Fig. a and b) and in perfused (Fig. c and d) liver.

The ablation volumes for the 14G needle were 24

cm3 and 9.4 cm

, in non-perfused and perfused

mode, with a volumetric ratio of 0.66 and 0.59

respectively.

The ablation volumes for the 17G needle were 10

cm3 and 3.4 cm

in non-perfused and perfused

mode, with a volumetric ratio of 0.55 and 0.45

respectively.

The data acquired with thermal cameras allowed

to measure with great accuracy the temperature

distribution on the entire liver and therefore to verify

in real time the correct perfusion state of the

parenchyma during the two ablation procedures with

active perfusion. Within the limits of the small

number of ablations performed, the preliminary

results indicate that an isolated spherical tumor with

a diameter up to 1.5 cm can be thermally ablated

with an adequate margin using a single NIC 14 G.

applicator.

If the tumor diameter exceeds 1.5 cm for its

complete thermal ablation it will be necessary to use

multiple NIC applicators or a single adequately

cooled high power IC device.

Moreover, as shown in figure 4, the thermal

behavior highlights the differences produced by the

perfusion in terms of maximum temperature reached

by the metallic shaft and consequently by the tissue

in direct contact with the MW tools.

Figure 5: Maximum Temperature comparison between

ablation in non perfused and in perfused liver at the end

of the termoablation (10 min).

6 DISCUSSION

With the exception of special situations such as a

neoplasia localized in unresectable positions, a high

number of lesions, inadequate hepatic reserve or

multiple comorbidities that contraindicate

anesthesia, percutaneous thermal ablation techniques

(TPA) are not yet considered the gold standard in the

treatment of liver neoplasms. This is mainly due to

the fact that liver transplantation and hepatic

resection have shown their superiority in the

Comparison of Ex-vivo Perfused and Non-perfused Porcine Liver Ablations using Uncooled Microwave Applicators

121

treatment of primary and secondary liver neoplasms

demonstrating higher long-term disease-free survival

rates than TPA (F. Romano 2012). In the

oncological TPA one of the main problems facing

doctors is the uncertainty about the actual size of the

ablations, mainly as a direct consequence of the

well-known "heat sink effect".

The prediction of this effect and of the 3D

dimensions of the ablation zone in relation to the

amount of applied microwave energy is a crucial

aspect since an incorrect prediction can lead to an

insufficient volume of ablation and a relapse of the

neoplasm. Since it is very hard to predict in practice

the exact size of the ablation for each combination of

time, power and size of the needle before the

execution of the procedure, these data are often

supplied to the doctor by the manufacturers of

thermal ablation systems in the form of ablation

charts or algorithms, but in our opinion these

information, together with the operator's personal

experience does not always provide the expected

results because they are often based on experiments

on ex-vivo organs at ambient temperature that

present serious deficiencies, preventing reliable

translation into the clinical practice.

In order to overcome these limitations, the

Department of Industrial Engineering (DIEF) of the

University of Florence has developed an

experimental platform (patent pending) to keep the

explanted liver in a state of normal physiological

thermal perfusion, capable to simulate the actual

heat sink effect during a TPA procedure. We need to

underline that the platform can simulate

physiological liver perfusion but is not able to keep

the liver cells alive and this aspect has been

evaluated by the authors before performing

experimental tests. Recent studies have underlined

that a warm ischemia time up to 60min does not

generate any irreversible cellular change and is

acceptable even for hepatic transplant

(Kalisvaart,

2018).As a consequence of that, if the ablation

experiment is performed in 60min beginning from

the liver explant, the effect of liver warm ischemia

on ablation shape and dimensions is negligible or

even absent. The aim of the study was therefore to

extract preliminary data to be compared with the

literature data in vivo. In our opinion, the ex-vivo

pig perfused liver test should represent the gold

standard for the definition of truly reliable

algorithms and ablation charts for the following

reasons:

1. The test is easily reproducible and allows a

definitive evaluation of the ablation volume in

the presence of the heat sink effect;

2. The proposed approach could allow the

standardization of the experimental procedure to

extract reliable algorithms and ablation charts;

3. The economic burden is lower than the costly in

vivo animal procedures.

4. Important ethical implications can be overcome.

In order to confirm experimentally this belief, the

ablation experiments where made in two different

perfusion inflow states: active and blocked using the

UniFi Hepatic Flow platform.

Ablation results of the two inflow state are

depicted in Tab1.

Preliminary results obtained by comparing

ablation performed in blocked and active perfusion

states show a reduction of about 30% in the ablation

radius from one to the other state, regardless of the

size of the applicator.

The ablation zones in this study where

commensurate with those of previous studies (M.

Dimitri 2018) obtaining results that confirm the

substantial equivalence between in-vivo and ex-vivo

normothermic perfused liver for the same energy

delivery.

It is important to note that the electromagnetic

and thermal properties of the ex vivo liver at

ambient temperature normally used by

manufacturers to extract the ablation chart, are very

different from those of the ex vivo normothermic

perfused liver.

7 CONCLUSION

The preliminary results obtained with this study are

too limited to have a statistical relevance but if this

result will be confirmed even in future tests this

model could constitute the best procedure to

evaluate the effectiveness of TPA without the use of

in vivo animal models. The availability of a

standardized ablation model based on ex-vivo

perfused liver opens the way to a more in-depth

investigation of the heat sink effect at peripheral and

central vessel locations. The UNIFI Perfusion

Platform is very versatile and it allows to easily

change the composition and flow rate of the

perfusion solution of the liver parenchyma.

Moreover this model could be used to emulate an

open surgical ablation, allowing the surgeon to

rotate the liver within its anatomical surrounding,

manually protect heat-sensitive organs (bowel) and

easily insert clustered applicators to treat large non

spherical tumors.

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

122

The preliminary results obtained with this study

are too limited to have a statistical relevance,

therefore the final validation of the proposed

approach will require a further and more in-depth

experimental activity.

8 FUTURE WORK

Future work will have the main focus trying to

statistically validate the actual results. If the new

tests will confirm them, a compact and autonomous

platform for the normothermic perfusion of porcine

livers will be engineered.

With this new device in the future we would like

to carry out:

 the comparisons between TPA in a swine in-vivo

model and TPA in a swine ex-vivo perfused

model.

 the comparison between ablations performed by

high power (120W) cooled applicator and low

power single or multiple uncooled applicator in

term of energetic efficiency and procedural

security;

 the optimization of multi-applicator ablation

procedures and the editing of more realistic and

affordable ablation charts;

 the study and optimization of a pulsed

microwave ablation technology (PMWA)

(

M.Bedoya 2012) in order to obtain an increase of

the ablation volume with the same input average

power.

REFERENCES

Ruiter S.J.S., Heerink W.J., De Jong K.P., (2019). Liver

microwave ablation: a systematic review of various

FDA-approved systems. Eur Radiology 29(8):4026-

4035

Meloni M.F. et alii, (2017). Microwave ablation in

primary and secondary liver tumours: technical and

clinical approaches. International Journal of

Hyperthermia 33(1): 15–24.

Yung J., Liang P., (2017). Status and advancement of

microwave ablation in China. International Journal of

Hyperthermia Vol. 33, 2017 - Issue 3

Kodama H., (2018). High power microwave ablation of

normal swine lung: impact of duration of energy

delivery on adverse event and heat sink effects.

International Journal of Hyperthermia 34(8):1186-

1193.

Biffi Gentili G, Ignesti C, Tesi V. (2014). Development of

a novel switched-mode 2.45 GHz microwave multi-

applicator ablation system. International Journal of

Microwave Science and Technology, Article ID

973736

De Cobelli F et alii, (2017). Microwave ablation of liver

malignancies: comparison of effects and early

outcomes of percutaneous and intraoperative

approaches with different liver conditions. Medical

Oncology 34(4)

R. Ravikumar, et alii, (2016). Liver transplantation after

ex-vivo normothermic machine preservation: a phase 1

(first-in-man) clinical trial. American Journal of

Transplantation, pp. 1779-1787.

F. Romano et alii, (2012). Bleeding in hepatic surgery:

sorting through methods to prevent it. HPB Surgery,

vol. 2012, 12pg, Article ID 169351.

Kalisvaart M, et alii, (2018).The impact of combined

warm ischemia time on development of acute kidney

injury in donation after circulatory death liver

transplantation: stay within the golden hour. P.

Transplantation. 102(5):783-793.

M. Dimitri et alii, (2018). A new microwave applicator for

laparoscopic and robotic liver resection. Internal

Journal of Hyperthermia, vol 36: 75-86.

M.Bedoya et alii (2012). Microwave ablation energy

delivery: Influence of power pulsing on ablation

results in an ex vivo and in vivo liver model. Medical

Physics 41 (12).

Comparison of Ex-vivo Perfused and Non-perfused Porcine Liver Ablations using Uncooled Microwave Applicators

123