Sage Journals: Discover world-class research

Abstract

Unlike necrosis by thermal ablation, irreversible electroporation (IRE) is known to induce apoptosis by disrupting plasma membrane integrity with electric pulses while preserving the structure of blood vessels and bile ducts in liver tissue without a heat sink effect. This study aimed to investigate thermal damage and histopathological effects in the porcine liver by high-frequency electric pulses (5 kHz) which is much higher than the widely used 1 Hz. The electric field and thermal distributions of 5 kHz electric pulses were compared with those of 1 Hz in numerical simulations. 5 kHz-IRE was applied on pigs under ultrasound imaging to guide the electrode placement. The animals underwent computed tomography (CT) examination immediately and 1 day after IRE. After CT, IRE-treated tissues were taken and analyzed histologically. CT revealed that hepatic veins were intact for 1-day post-IRE. Histopathologically, the structure of the portal vein was intact, but endothelial cells were partially removed. In addition, the hepatic artery structure from which endothelial cells were removed were not damaged, while the bile duct structure and cholangiocytes were intact. The thermal injury was observed only in the vicinity of the electrodes as simulated in silico. 5 kHz-IRE generated high heat due to its short pulse interval, but the thermal damage was limited to the tissue around the electrodes. The histopathological damage caused by 5 kHz-IRE was close to that caused by 1 Hz-IRE. If a short-time treatment is required for reasons such as anesthesia, high-frequency IRE treatment is worth considering. Our observations will contribute to a better understanding of the IRE phenomena and search for advanced therapeutic conditions.

Keywords

irreversible electroporation 5 kHz computed tomography histopathology apoptosis

Introduction

Radiofrequency ablation (RFA) has been used to ablate tumors over the last decades. This modality is based on an electromagnetic energy of alternating electric current with a frequency of less than 900 kHz. The electric current through a needle applicator generates heat that induces cell death,^1,2 specifically for mitochondrial and cytosolic enzymes and nucleic acid-histone protein complexes.³ The tissue then undergoes central necrosis around the electrode inserted into the tissue. However, the coagulative necrosis is not complete in the vicinity of a large blood vessel because of heat loss due to convection caused by tissue blood circulation.⁴ This incomplete ablation of these tumor tissues can lead to recurrence at the ablated region.⁵ Therefore, in tumors near large vessels, RFA is avoided due to its heat sink effect.

Unlike RFA, irreversible electroporation (IRE) uses a square wave voltage to ablate tumor tissues. Typically, an electric pulse with an electric field strength of 1 to 3 kV/cm and a pulse width of 50 to 100 µs is applied between 2 electrodes. The field affects intact lipid bilayers, which are subjected to thermodynamic fluctuations at the molecular level. Therefore, while continuing a hydrophobic membrane, the random thermodynamic motion of the lipid molecules causes the space among the charged lipid heads to waves.⁶ That movement would create defects in the membrane with an intramolecular space enough for the molecule to penetrate the hydrophobic pores. Pore formation leads to cell death by disrupting cell homeostasis by increasing conductivity.⁷ Furthermore, IRE allows for preserving collagenous and other protein- and lipid-based structures, including vasculature⁸ and ductal networks.⁹ IRE can remove tumors near large blood vessels or vessel-encased tumors without the effect of a heat sink. Therefore, IRE has been mainly applied to liver tumors involving hepatic and portal veins.^10–17

As the clinical use of IRE increased, there were concerns about side effects such as muscle contraction near the electrode. To reduce the possible induced contraction in the heart muscle, synchronization of the electric pulses with heartbeat was suggested.¹⁸ For that reason, a frequency of 1 Hz is becoming widely used nowadays. European Standard in Electro-chemotherapy recommends the use of multiple pulses of 1 Hz or 5 kHz 1200 V/cm with a 100 µs pulse width for chemotherapy.¹⁹ The paper informed that treatment at a frequency of 5 kHz would reduce the number of contractions to one—however this contraction would be more powerful than at a frequency of 1 Hz. Angiodynamics Inc. proposed a series of 90 pulses at 1 Hz 1500 V/cm with a pulse width of 90 µs for the successful nonthermal IRE.^20–22 However, there have been no reports of clinical cardiac problems due to electroporation so far, and adversely it was analyzed that high frequency could reduce pain and muscle contractions in patients who undergo electroporation treatments.²³ Another concern with IRE is the heat generation in the tissue with electric field application. Davalos et al simulated that the temperature reached 50 °C near the cylindrical electrode with a single pulse of 1000 V/cm.²⁴ Wagstaff et al measured the temperature at the core of the ablation zone when 70 pulses of 1 Hz 1500 V/cm with 90 µs pulse width were applied in the kidney.²⁵ A peak temperature of 57 °C was measured with a 3-needle configuration and 79 °C was measured with a 4-needle configuration. They recommended temperature monitoring during repetitive IRE processes near vital structures. Thermal damage from 5 kHz electric pulses is expected to be greater than that of 1 Hz electric pulses due to shorter intervals between repetitive pulses. However, no one has yet conducted a study on the temperature distribution for 5 kHz electric pulses. It would generate less heat than RFA, but more heat than a 1 Hz electric pulse. In this study, we performed a simulation on the electric field and temperature distributions around the electrodes with 20 electric pulses of 5 kHz, 1000 V/cm with 100 µs pulse width. Then, IRE was executed on liver tissues in pigs in vivo with computed tomography (CT), and histological analyses were employed.

Materials and Methods

Numerical Analysis for Electric Field Distribution

For the IRE application, the applied electric fields were numerically simulated using a commercial finite element package (COMSOL Multiphysics, Version 5.4; COMSOL, Stockholm, Sweden) to identify the distribution of the electric field strength of 2000 V surrounding 1 mm-diameter needle-type electrodes made of stainless-steel cylinders (SUS304, electrical conductivity 1.37 × 10⁶ S/m, thermal conductivity 15.11 W/mK).²⁶ The configuration of electrodes was an exposure length of 20 mm and a separation of 20 mm between electrodes (Figure 1A). The electric field distribution was retained from solving the Laplace equation under the electrostatic approximation,²⁷

\nabla^{2} \emptyset = 0

(1)

where

\emptyset

denotes the electric potential. The liver tissue was modeled for the simulation as a block geometry of the dimensions 60 × 60 × 40 mm with liver tissue properties (electrical conductivity 0.33 S/m, thermal conductivity 0.52 W/mK, heat capacity 3540 J/kgK). We assumed that the electrical properties of tissue were not changed during the 20-pulse application for simplification of the simulation. The number of elements was 1,831,516. The exposed solid electric surface was treated with an electric boundary condition:

\emptyset = V

(source) and

\emptyset = 0

(sink); the other surfaces were defined as electric insulation. The electric field distribution was simulated as shown in Figure 1B, and the ablation area where the electric field was over threshold field strength 650 V/cm for liver tissue was marked with arrows.

Figure 1.

Experimental conditions and numerical simulations. (A) Schemes of experimental conditions. (B) The electric field distribution. Areas corresponding to the size of the IRE-induced tissue are indicated by arrows. (C) The heat distribution after 20 pulses at both 5 kHz and 1 Hz in horizontal and vertical sections. The maximum temperatures at the side and tip of the electrodes were marked by numbers and arrows.

Numerical Analysis for Temperature Distribution

Temperature distribution in tissue was modeled using Pennes’ bioheat equation.²⁸

\begin{aligned} ρ C_{p} \frac{\partial T}{\partial t} + \bar{\nabla} \cdot \bar{q} = Q + Q_{b i o}, \\ \bar{q} = - κ \nabla T, Q_{b i o} = ρ_{b} C_{b} ω_{b} (T_{b} - T) \end{aligned}

(2)

where

ρ

is the tissue density (1079 kg/m³), C_p is the specific heat capacity of the tissue (3540 J/kgK), T is the temperature (K),

\bar{q}

is the heat flux by conduction (W/m²), Q is the heat source (W/m³),

ρ_{b}

is the blood density (1000 kg/m³),

C_{b}

is the specific heat capacity of blood (4180 J/kgK),

ω_{b}

is the blood perfusion rate (0.00715/s),

T_{b}

is the arterial blood temperature (310.15 K), k is the thermal conductivity (0.52 W/mK). Here, the initial temperature of the tissue and electrode was set at 37 °C. The potentials were set to 2000 V with electrode distances of 20 mm. We compared 2 conditions of 20 electric pulses: 5 kHz and 1 Hz. Temperatures of the tip or the side of the electrode were calculated and plotted according to time (s).

Animals

A protocol for both animals was reviewed and approved by the Institutional Animal Care and Use Committee of Seoul National University Hospital (IACUC No.10-0106). The protocol was followed by the U.S. National Institute of Health guidelines for dealing with animals. The research complied with Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. Following this guideline, two 4-month-old female pigs weighing 40 to 45 kg for the histological investigation were purchased from Cronex (Cheongju, Korea). We reduced the number of animals for the experiments by application of IRE at 4 sites of the liver in the same animal and by using multiple CT scanning that allowed longitudinal measurements in the same animal to be taken. Animals that were tested and quarantined postprocurement based on the welfare report offered by the provider were healthy and fit for testing after observing their general symptoms for 7 days. Afterward, the animals were raised in the following conditions: temperature 23 ± 2 °C, relative humidity 50 ± 10%, ventilation frequency 10 to 15 times/h, lighting time 12 h (8 am to 8 pm off), and illuminance of 150 to 300 Lux in cages of more than 0.74 m² (AAALAC guideline) per animal during the experiments. Solid feed sterilized by irradiation was supplied to the test animals (238075 pig chow, Cargill, and Griprina). Water was sterilized via reverse osmosis and an ultraviolet device and then provided through an automatic water supply nozzle.

IRE Procedure

To perform the experimental procedure, we intramuscularly injected tiletamine hydrochloride, zolazepam hydrochloride (Zoletil 5 mg/kg, Virbac, France), and xylazine hydrochloride (Rompun^®, Bayer, Germany, 2 mg/kg) into each pig. After anesthesia, isoflurane (1.5%-2%) was maintained during the IRE procedure. The liver was exposed for easy insertion of IRE electrodes after a median incision along the ring of the abdomen after disinfecting with alcohol and povidone. In consideration of the simulation results (Figure 1B), the two electrodes were inserted into the liver (left/right lateral lobe, left/right medial lobe) one at a time, monitored by ultrasonography (US) (Accuvix Xg, Samsung Medison, Co. Ltd, Korea) to keep the electrodes parallel 20 mm apart that have an exposure length of 20 mm. To be transmitted electrical energy, electrodes were linked to a square pulsed voltage generator (EPO1, The Standard, Co. Ltd, Korea), which can output a maximum amplitude of 3000 V of a pulse width ranging from 100 to 1000 μs. The electrical energy applied to 2000 V of a pulse width of 100 μs and equal interpulse time pulsing 20 pulses. In addition, the ablation sites were performed at 4 sites per animal having the same electrical power for histological investigation.¹¹ After CT, the animals were euthanized for histological assessment. Temperature measurements were employed for heat assessment with a thermometer (Fluke 53II, Everett, Washington, USA), collecting one data per second. We manually manufactured the temperature probe by inserting a type-K thermocouple into one of the electrodes. Therefore, it was possible to measure the electrode temperature directly during the experiment.

Follow-up With US Imaging and CT

Immediately after IRE, the US was performed to confirm the IRE ablation site. In addition, the animals underwent precontrast and triple-phase postcontrast CT (Somatom Sensation 16, Siemens, Forchheim, Germany) examinations immediately and 1 day after IRE. Iohexol 1 mL/kg (Omnipaque 350, Nycomed, Princeton, NJ) was injected at a rate of 3 mL/s. Besides, images of the arterial phase at 20 s after injection, portal venous (PV) phase at 60 s after injection, and delayed venous phase at 180 s after that injection was acquired. CT parameters were 250 mA, 120 kVp, 3-mm collimation, and a pitch of 2:1. All CT images were analyzed using 3D software (Vitrea 2, Vital Images, Cincinnati, Ohio) for matching to the plane of pathological sections.

Histology

Following the CT follow-up, each part of the liver tissue was cut and photographed with a camera (Canon Power Shot A620, Canon, Tokyo, Japan). Histopathologically examined the ablated area using hematoxylin and eosin (H&E) staining. The tissues under IRE were fixed in a 10% neutral formalin solution, dehydrated in an ethanol gradient, embedded in paraffin, cut into 4 µm slices, mounted on slides, and subjected to H&E staining. To confirm apoptosis by IRE, different slicing was stained from the same sample with an apoptosis detection kit (S7100, ApopTag Peroxidase) with terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) according to the manufacturer's instructions. Slides were scanned automatically (Pannoramic 250 Flash III; 3D HITECH Ltd, Hungary).

Results

Simulation of Temperature Distribution

Simulation results showed that the maximum temperature at the midpoint of the cylindrical electrode surface was 78.7 °C, and that of the round tip was 299 °C when 5 kHz electric pulse trains were applied (Figure 1C). These values were well above 1 Hz, which were 40.1 °C and 46.2 °C, respectively. The real-time temperature change of the tip was shown in Figure 2A. The extremely high temperature of the tip reached 45 °C after 10 s of post-IRE (Figure 2B). For 1 Hz, the temperature slowly decreased, then the tip temperature reached 41.2 °C after 10 s. Figure 2C shows the temperature distribution along the line between the midpoints of 2 electrodes under both 5 kHz and 1 Hz electric pulse conditions. It shows that the generated heat was not transmitted far and was confined near the electrodes. An IRE of 5 kHz was applied to this experiment.

Figure 2.

Temperature changes during and after electric pulses at 5 kHz and 1 Hz in silico. (A) The tip-end temperature changes during pulses at both 5 kHz and 1 Hz. (B) The tip-end temperature changes for 10 s after pulses at both 5 kHz and 1 Hz. (C) The lateral temperature distribution between electrodes during pulses at both 5 kHz and 1 Hz.

US and CT-Imaging Findings

The IRE procedure was successfully performed by the protocol. The electrodes were placed into the liver tissue under US image guidance to keep the electrodes parallel, as indicated with arrows in Figure 3A(1). The pulsing (20 pulses of 2000 V with 100 μs pulse width and 100 μs pulse interval) was delivered to the liver tissues without sparks. The tissue was observed by the US to identify the RE area around the electrodes immediately after 5 kHz-IRE treatment (Figure 3A(2)). The IRE-treated area was a heterogeneous echogenicity with clear borders indicated by a white arrow. Hyperechoic small dots around one of the electrodes might be from micro-bubbles due to the IRE thermal effect, which was marked by a black arrow in the image. In addition, the tissue was examined for post-IRE by CT (Figure 3B). In the transverse and coronal images, the ablated zone was recognized as a well-demarcated nonenhancing area (Figure 3B(1-2)). The areas were indicated by white arrows. The hepatic vein within the ablated zone was preserved and well-enhanced. A sustained but diminished effect was observed on both transverse and coronal CT images 1 day after IRE (Figure 3B(3-4)). Three small hyperdense spots within the treated area after IRE suggest intact hepatic veins.

Figure 3.

US and CT images of the porcine liver treated with 5 kHz-IRE. (A) US images before (1) and after (2) 5 kHz-IRE. (1) Arrows indicate the placement of electrodes inserted into the liver. (2) White arrow indicates the darkened IRE-treated area from the different surrounding parenchyma, and the black arrow indicates the microbubbles around a positive electrode. (B) CT images immediately (1-2) and one day after (3-4) IRE. (1-2) Contrast-enhanced CT images show a hypodense zone as 5 kHz-IRE ablations with an enhancing hepatic vein within the ablation in transverse (1) and coronal (2) images. Arrows indicate the ablation zone. (3-4) Hypodensity of the ablation zone was slightly diminished but still clearly observed with hyperdense spots of preserving veins after one day in transverse (3) and coronal (4) images. Scale bars are inserted separately in each image.

Histological Findings

One day later, the 5 kHz-IRE-treated sites were distinguished from the CT image. The size was measured to be about 20 to 30 mm in the enlarged picture inserted in Figure 3B(3). It was also distinguishable from resected tissue. Visually, we demarcated the IRE-treated areas in the sliced tissue masses in Figure 4A. The border of the IRE-treated area had a white border, and the inside was a dark red due to red blood cells out of the blood vessels. The IRE-treated site was oval and might vary depending on the slice angle of the tissue, but the measured size was 12.6 to 26.6 mm. Vessels visible on CT might be large vessels (1.2 to 4.5 mm in diameter) visible to the naked eye in tissue. The vessel boundaries seemed to be clear and intact.

Figure 4.

Pictures of 5 kHz-IRE-treated liver tissues and their histology images. (A)(1-2). Pictures of 5 kHz-IRE treated-liver tissues. IRE-treated areas were clearly distinguished by the naked eye in the sliced tissue masses. The positions corresponding to the diameters of the elliptical long and short sides are marked with arrows, and the lengths are marked. The diameters of blood vessels inside the IRE-treated area were marked with dotted arrows. The electrode positions were marked with dotted circles. (B)(1-2). H&E staining of sections from each tissue of (A)(1) and (A)(2). 5 kHz-IRE-treated areas were clearly distinguished by bright pink middle and dark purple border areas. (C)(1-2). Enlarged images from (B)(1). (1) The border of the IRE treatment, which looked white to the naked eye, was dyed strongly with hematoxylin. (2) Electrode position is distinguished by pale pink color and unclear cell boundaries. The expected electrode position was marked with a dotted line. (D)(1-2). Enlarged images from (B)(1). (1) The border of the IRE treatment has a wider space between portal lobules. (2) The IRE-treated region has a narrower space between portal lobules. A scale bar is inserted for each picture.

Even when the tissue was sliced and H&E staining was performed, the region was visually distinguishable (Figure 4B(1-2)). The bright pink medium and dark purple border areas delimited IRE-treated areas. Vascular borders were also clearly observed. The bright pink color in the middle came from the Eosin in exposed red blood cells. The part stained in deep purple was magnified in Figure 4C(1), but no specific structure was observed. It might be negative molecules that were selectively attached to hematoxylin. However, what causes them to gather at the border and why the tissue looks white could not be distinguished by their shape. The electrode position was also magnified and displayed in Figure 4C(2). The tissue surrounding the electrode was clearly distinguished in H&E staining. The cell boundaries became a blur, and the cytoplasmic material might be expelled. The heat from the electrode affected an area of about 2 mm. Interestingly, we also observed another specific characteristic of liver tissue at the boundary region. Interconnected lobules of connective tissue were found to be widened in the border region of IRE ablation. The differences became apparent when we magnified part of each region in Figure 4D(1-2). The spaces between portal lobules became broader in the boundary region. The spaces between hepatocytes were hard to compare due to hemorrhage.

We further investigated portal tract blood vessels and bile ducts with H&E staining and TUNEL assay (Figure 5). The portal vein, the hepatic artery, and the bile duct in the untreated region (NON-IRE (1, 3, 5, 7)) and the IRE-treated region (IRE (2, 4, 6, 8)) were displayed in parallel. All structures in the portal tract of IRE remained intact with the NON-IRE. However, the endothelial cells of portal veins were partially separated from the blood wall in the IRE region (Figure 5A(3-4)). TUNEL assay confirmed that those cells are under apoptosis (Figure 5B(3-4)). The endothelial cells of the hepatic artery in the IRE region were also destroyed, so intact endothelial cells could not be identified (Figure 5A and B(5-6)). However, unlike endothelial cells, cholangiocytes of the bile duct were not damaged in the IRE region as in the NON-IRE (Figure 5A(7-8)). As shown in the TUNEL assay, the cells were not in progress of apoptosis (Figure 5B(7-8)).

Figure 5.

H&E staining and TUNEL assay images of portal tracts for the NON-IRE and IRE regions. (A). H&E images. (1-2) Portal tract, (3-4) portal vein, (5-6) hepatic artery, (7-8) bile duct. Images (3, 5, 7) and images (4, 6) were magnified from images (1) and (2), respectively. Image (8) was from another site in the same tissue. (B). TUNEL assay. Images (1-8) were for the same part of portal tracts as A. Arrows indicate the endothelial cells of blood vessels or cholangiocytes of the bile duct. A scale bar is inserted for each picture.

We investigated portal lobules for NON-IRE and IRE regions as well (Figure 6). In the NON-IRE area, portal lobules were observed in the composition of the central vein, the portal canal, and the connective tissue septa surrounding the lobule (Figure 6A(1, 3, 5, 7)). In the IRE region, we could distinguish the area for the central vein, portal canal, and connective tissue septa (Figure 6A(2, 4, 6, 8)). However, the hemorrhage over the treated region was so severe that the tissue structure was not observed. Red blood cells were stained crimson with Eosin, and arrows indicated some isolated ones in the figure. In addition, endothelial cells were not observed in the central vein (Figure 6A(4) and B(4)). The portal canal vein was not damaged in the NON-IRE site (Figure 6A(5)). Portal endothelial cells were also found in the IRE region (Figure 6A(6)). However, cells were in part alive and in part apoptosis, which was confirmed by the TUNEL assay (Figure 6B(6)). Furthermore, the fibers were degraded in the connective tissue septa in the IRE region, whereas fibers were intact in the NON-IRE region (Figure 6A(7-8)). Cells in the connective tissue were not found in the IRE region (Figure 6A(8) and B(8)).

Figure 6.

H&E staining and TUNEL assay images of portal lobules for the NON-IRE and IRE regions. (A). H&E images. (1-2) Portal lobule, (3-4) central vein, (5-6) portal canal, (7-8) connective tissue septa. Images (3, 5, 7) and images (4, 6, 8) were magnified from images (1) and (2), respectively. (B). TUNEL assay. Images (1-8) were for the same part of portal lobules as A. Arrows indicate red blood cells, and arrowheads indicate endothelial cells from the portal canal. A scale bar is inserted for each picture.

Discussion

The results of the study show that although the mechanism of action of IRE is a disruption of cell membranes by electric energy, thermal damage to tissues must also be mentioned because electric current induces heat generation. In the simulation of temperature distribution, the use of 1Hz-IRE dissipated the heat around the electrode into surrounding tissues during long pulse intervals. However, in the case of 5 kHz-IRE, the 100 μs pulse interval was insufficient to cool the electrode. The electrode tip temperature reached 299 °C and the tip-side temperature became 78.7 °C after 20 electric pulses at 5 kHz (Figure 1C). This extremely high temperature of the tissue around the electrode should induce necrosis.^29,30 However, the simulation of the temperature distribution along the tissue between the electrodes shows that only the tissue around the electrode has a high temperature. Even at just 2 mm away from the electrode, the temperature is below 40 °C (Figure 2C). We could confirm this in animal experiments. Only the tissue adjacent to the electrode suffered a slight thermal injury, which was observed in the H&E staining image as a lightly stained region with cell structure destruction (Figure 4C(2)). The area was demarcated and identified as having a diameter of approximately 2 mm. This confirmed that the thermal damage was highly localized. How long the high temperature lasts after the pulse is also a very important part of the thermal damage analysis. The temperature at the tip of the electrode, which was close to 300 °C, drops below 45 °C after 10 s, and the temperature at the side of the electrode, which was close to 79 °C, drops below 40 °C after 5 s. In the case of the temperature next to the electrode, the total time when the tissue temperature exceeds 40 °C does not exceed 5 s. Interestingly, it is shorter than the 1 Hz case. This was reflected in the temperature measurement of the animal experiment. After 6 s of the pulse application, it could be confirmed that the average temperature of the electrode drops 45 °C (Figure 7). This is because the time when the temperature of the tissue exceeds 40 °C is 20 s plus 5 s for the pulse application and for cooling, respectively. Therefore, if 20 pulses are designated as one group, 100 pulses can be applied if 5 groups are shot, so this 5 kHz-IRE can be a pulsed method.

Figure 7.

The temperature of an anodic electrode for 10 s after 5 kHz 20 pulse application.

In addition, our histological data showed that the IRE-induced area can be explained only by the electric field simulation results. Figure 4A shows the distinct white border of the IRE-treated area in sliced tissues, which can be misunderstood as being due to thermal damage. The border did not contain necrotic cells and was stained dark purple in H&E staining, which implies a strong interaction with hematoxylin (Figure 4B). Hematoxylin reacts strongly with negatively charged molecules such as DNA. According to the electric properties, this molecular separation implies that the border originates from the electrical impact of IRE. In addition, the length of tissue affected by IRE was measured as 23.3 to 26.6 mm in the electrode direction (Figure 4A(1-2)). Compared to the electric field simulation, the threshold at which the IRE occurred seemed to be approximately 650 V/cm (Figure 1B). That value was similar to the known value of IRE threshold field strength for liver tissue.³¹ We expected that the joule heating of the electrode did not affect the macroscopic IRE phenomenon.

We investigated how much the IRE condition induced pathological effects. Our IRE condition transfers lower electric energy than the generally used IRE protocol (1 Hz-IRE) for complete ablation of hepatic cancer; 90 pulses of 1 Hz 1500 V/cm with 90 µs pulse width. Electrical current increases by about 10 to 15 A during pulse application, resulting in greater electric energy.^32,33 The entire electrical energy of our conditions can be calculated at 2000 V × ∼20 A × 0.0001 s × 20∼80 J, and the energy of the 1Hz-IRE treatment at 3000 V × ∼28 (20∼35) A × 0.00009 s × 90∼680.4 J. However, the power for our condition is 80 J/4 ms = 20 kW, and the power for the previous IRE condition is 680 J/20 s = 34 W. Though our condition has lower electric energy, it was transferred to tissue in a very short time. With 1 Hz-IRE treatment, the preservation of venous structures near the IRE ablation region was not significantly affected.^12,34 In animal studies, vascular narrowing was found in a few veins, but returned to normal size.¹² Clinical trials have shown no postoperative complications of IRE for tumors near a major hepatic vein.^13,35 The 1 Hz-IRE showed no considerable vessel damage or thrombus formation, while there was mild endothelial damage in some small vessels,³⁶ even when electrodes were placed near the hepatic artery. Additionally, the connective matrix of the surrounding blood vessels and bile ducts within the ablated zones remained intact. Meanwhile, vascular occlusions were in the small vessels within the ablation zone.^10,35–40 In our IRE condition, we also observed the preservation of large blood vessels by using CT, but we did not continue postoperative observation. Other clinical reports showed that bile ducts were preserved but underwent narrowing.^41,42 Unfortunately, we could not distinguish bile ducts inside hepatic tissues in CT images.

We compared the histopathological results of our IRE condition with previous research in the 1 Hz-IRE condition. Histological investigation of the portal tract showed that the portal vein, the hepatic artery, and the bile duct were structurally intact under IRE. However, the endothelial cells in the portal vein were partially intact and partially destroyed, which was confirmed by H&E and TUNEL assay. In addition, the TUNEL assay showed that the cells in the hepatic artery were utterly destroyed. However, cholangiocytes of the bile duct were intact in the IRE, which was confirmed by the TUNEL assay. Investigation on portal lobules showed that they were structurally preserved. However, endothelial cells were not found in the central vein and were partially found in the portal vein in the portal canal. Also, cells were not observed in connective tissue. In comparison, the 1Hz-IRE destroyed endothelial cells of blood vessels wholly, while our condition partially removed endothelial cells in the portal vein. Second, our IRE condition has remained cholangiocytes alive in the bile duct while the 1 Hz-IRE has completely collapsed. Third, our IRE condition has a rarely partially destroyed matrix, while the 1Hz-IRE has completely fallen. Unfortunately, we could not accurately judge our IRE condition with a more extended follow-up period. The recovery of the endothelial cells that underwent apoptosis, the thrombosis occurrence in the damaged veins, and the restoration of necrotic tissue abutting the electrode need to be answered.

Conclusion

In conclusion, 5kHz-IRE generates high heat due to its short pulse interval, but the thermal damage is limited to the tissue in the vicinity of the electrode. The histopathological damage by 20 IRE pulses at 5 kHz approached that by 90 pulses at 1 Hz. Most of the blood vessels and bile ducts remained, and partial damage to endothelial cells was observed. The size or histopathological effect of the IRE-treated tissue seemed to be affected more by the voltage itself than by the electrical energy or power. If a short-time treatment is required for reasons such as anesthesia, high-frequency (near kHz) IRE treatment is also worth considering. If higher electrical energy is required, we would like to suggest a method of processing 20 pulses several times with enough intervals. Of course, systematic studies on the immunological effects of necrotic tissue near electrodes, on the occurrence of thermal thrombosis, or on ECG issues should be followed for accurate information on safety. In addition, we observed that the ECG signal was restored after IRE on the liver with our condition (not shown data). However, more research is needed to see if it can be applied to tissues near the heart. Our observations will contribute to a better understanding of the IRE phenomena and the search for advanced treatment conditions.

Footnotes

Abbreviations

Acknowledgements

This work was supported by the New Faculty Startup Fund from Seoul National University.

Authors’ Note

These authors contributed equally to this work.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics Statement

For the animal study, the protocol was approved by Seoul National University Hospital (IACUC No.10-0106) and conformed to the U.S. National Institute of Health guidelines for handling laboratory animals. The study was conducted in compliance with ARRIVE guidelines.

ORCID iDs

Hong Bae Kim

Chang Kyu Sung

References

Gazelle

Goldberg

Solbiati

Livraghi

. Tumor ablation with radio-frequency energy. Radiology. 2000;217:633‐646.

Goldberg

. Radiofrequency tumor ablation: principles and techniques. Eur J Ultrasound. 2001;13:129‐147.

Goldberg

Gazelle

Mueller

. Thermal ablation therapy for focal malignancy: a unified approach to underlying principles, techniques, and diagnostic imaging guidance. AJR Am J Roentgenol. 2000;174:323-331.

Patterson

Scudamore

Owen

Nagy

Buczkowski

. Radiofrequency ablation of porcine liver in vivo: effects of blood flow and treatment time on lesion size. Ann Surg. 1998;227:559‐565.

Lin ZY, Li GL, Chen ZW, Chen YP, Lin SZ. Effect of heat sink on the recurrence of small malignant hepatic tumors after radiofrequency ablation. J Cancer Res Ther. 2016;12:C153‐C158.

Kashchiev

Exerowa

. Bilayer lipid membrane permeation and rupture due to hole formation. Biochim Biophys Acta. 1983;732:133‐145.

Aycock

Davalos

. Irreversible electroporation: background, theory, and review of recent developments in clinical oncology. Bioelectricity. 2019;1:214‐234.

Maor

Ivorra

Leor

Rubinsky

. The effect of irreversible electroporation on blood vessels. Technol Cancer Res Treat. 2007;6:307‐312.

Ueshima E, Schattner M, Mendelsohn R, et al. Transmural ablation of the normal porcine common bile duct with catheter-directed irreversible electroporation is feasible and does not affect duct patency. Gastrointest Endosc. 2018;87:300.

10.

Guo Y, Zhang Y, Nijm GM, et al. Irreversible electroporation in the liver: contrast-enhanced inversion-recovery MR imaging approaches to differentiate reversibly electroporated penumbra from irreversibly electroporated ablation zones. Radiology. 2011;258:461‐468.

11.

Charpentier KP, Wolf F, Noble L, Winn B, Resnick M, Dupuy DE. Irreversible electroporation of the liver and liver hilum in swine. HPB (Oxford). 2011;13:168‐173.

12.

Lee

Osuagwu

Lassman

. Irreversible electroporation in porcine liver: short- and long-term effect on the hepatic veins and adjacent tissue by CT with pathological correlation. Invest Radiol. 2012;47:671‐675.

13.

Appelbaum L, Ben-David E, Faroja M, Nissenbaum Y, Sosna J, Goldberg SN. Irreversible electroporation ablation: creation of large-volume ablation zones in in vivo porcine liver with four-electrode arrays. Radiology. 2014;270:416‐424.

14.

Dollinger M, Beyer LP, Haimerl M, et al. Adverse effects of irreversible electroporation of malignant liver tumors under CT fluoroscopic guidance: a single-center experience. Diagn Interv Radiol. 2015;21:471‐475.

15.

Lyu

Lopez-Ichikawa

Rubinsky

Chang

. Normal and fibrotic liver parenchyma respond differently to irreversible electroporation. HPB (Oxford). 2019;21:1344‐1353.

16.

O'Brien TJ, Passeri M, Lorenzo MF, et al. Experimental high-frequency irreversible electroporation using a single-needle delivery approach for nonthermal pancreatic ablation in vivo. J Vasc Interv Radiol. 2019;30:854‐862.

17.

Tameez Ud Din

Tameez-Ud-Din

Chaudhary

FMD

Chaudhary

Siddiqui

. Irreversible electroporation for liver tumors: a review of literature. Cureus. 2019;11:e4994.

18.

Mali B, Jarm T, Corovic S, et al. The effect of electroporation pulses on functioning of the heart. Med Biol Eng Comput. 2008;46(8):745‐757.

19.

Mir LM, Gehl J, Sersa G, et al. Standard operating procedures of the electrochemotherapy: instructions for the use of bleomycin or cisplatin administered either systemically or locally and electric pulses delivered by the cliniporator TM by means of invasive or non-invasive electrodes. E J C Supplements. 2006;4:14‐25.

20.

Scheltema MJ, O'Brien TJ, van den Bos W, et al. Numerical simulation modeling of the irreversible electroporation treatment zone for focal therapy of prostate cancer, correlation with whole-mount pathology and T2-weighted MRI sequences. Ther Adv Urol. 2019;11:1756287219852305.

21.

Sugimoto K, Moriyasu F, Takeuchi H, et al. Case study to assess the safety of irreversible electroporation near the heart. Springerplus. 2015;4:74.

22.

http://vitusprostate.com/en/science/treatment/ire-technology-on-detail/

23.

Fusco R, Di Bernardo E, D'Alessio V, Salati S, Cadossi M. Reduction of muscle contraction and pain in electroporation-based treatments: an overview. World J Clin Oncol. 2021;2(5):367‐381.

24.

Davalos

Rubinsky

. Temperature considerations during irreversible electroporation. Int J Heat Mass Transfer. 2008;51:5617‐5622.

25.

Peter

, et al. Irreversible electroporation of the porcine kidney: temperature development and distribution. Urol Oncol: Semin Orig Invest. 2015;33(4):168.e1‐168.e7.

26.

Chu

TK.

Electrical resistivity and thermal conductivity of nine selected AISI stainless steels, Tech. Rep. DTIC Document. 1977.

27.

Kim

Chung

. Incorporation of reversible electroporation into electrolysis accelerates apoptosis for rat liver tissue. Technol Cancer Res Treat. 2020;19:1533033820948051.

28.

Choi YS, Kim HB, Chung J, Kim HS, Yi JH, Park JK. Preclinical analysis of irreversible electroporation on rat liver tissues using a microfabricated electroporator. Tissue Eng Part C Methods. 2010;16:1245‐1253.

29.

Zhu

, et al. Temperature-feedback upconversion nanocomposite for accurate photothermal therapy at facile temperature. Nat Commun. 2016;7:10437.

30.

Song AS, Najjar AM, Diller KR. Thermally induced apoptosis, necrosis, and heat shock protein expression in 3D culture. J Biomech Eng. 2014;136(7):071006.

31.

Ritter A, Bruners P, Isfort P, et al. Electroporation of the liver: more than 2 concurrently active, curved electrodes allow new concepts for irreversible electroporation and electrochemotherapy. Technol Cancer Res Treat. 2018;17:1533033818809994.

32.

Kos

Voigt

Miklavcic

Moche

. Careful treatment planning enables safe ablation of liver tumors adjacent to major blood vessels by percutaneous irreversible electroporation (IRE). Radiol Oncol. 2015;49:234‐241.

33.

Dunki-Jacobs

Philips

Martin

II . Evaluation of resistance as a measure of successful tumor ablation during irreversible electroporation of the pancreas. J Am Coll Surg. 2014;218:179‐187.

34.

Dollinger M, Müller-Wille R, Zeman F, et al. Irreversible electroporation of malignant hepatic tumors-alterations in venous structures at subacute follow-up and evolution at mid-term follow-up. PLoS One. 2015;10:e0135773.

35.

Kingham TP, Karkar AM, D'Angelica MI, et al. Ablation of perivascular hepatic malignant tumors with irreversible electroporation. J Am Coll Surg. 2012;215:379‐387.

36.

Chen X, Ren Z, Zhu T, et al. Electric ablation with irreversible electroporation (IRE) in vital hepatic structures and follow-up investigation. Sci Rep. 2015;5:16233.

37.

Appelbaum

Ben-David

Sosna

Nissenbaum

Goldberg

. US findings after irreversible electroporation ablation: radiologic-pathologic correlation. Radiology. 2012;262:117‐125.

38.

Neal RE, Rossmeisl JH, D'alfonso V, et al. In vitro and numerical support for combinatorial irreversible electroporation and electrochemotherapy glioma treatment. Ann Biomed Eng. 2014;42:475‐487.

39.

Tamura M, Pedersoli F, Schulze-Hagen M, et al. Predictors of occlusion of hepatic blood vessels after irreversible electroporation of liver tumors. J Vasc Interv Radiol. 2020;31:2033‐2042.

40.

Rubinsky

Onik

Mikus

. Irreversible electroporation: a new ablation modality-clinical implications. Technol Cancer Res Treat. 2007;6:37‐48.

41.

Long G, Bakos G, Shires PK, et al. Histological and finite element analysis of cell death due to irreversible electroporation. Technol Cancer Res Treat. 2014;13:561-569.

42.

Lee

Loh

Kee

. Imaging guided percutaneous irreversible electroporation: ultrasound and immunohistological correlation. Technol Cancer Res Treat. 2007;6:287‐294.

Histological Response to 5 kHz Irreversible Electroporation in a Porcine Liver Model

Abstract

Keywords

Introduction

Materials and Methods

Numerical Analysis for Electric Field Distribution

Numerical Analysis for Temperature Distribution

Animals

IRE Procedure

Follow-up With US Imaging and CT

Histology

Results

Simulation of Temperature Distribution

US and CT-Imaging Findings

Histological Findings

Discussion

Conclusion

Footnotes

Abbreviations

Acknowledgements

Authors’ Note

Declaration of Conflicting Interests

Funding

Ethics Statement

ORCID iDs

References