Non-Contact Irreversible Electroporation in the Esophagus With a Wet Electrode Approach

Abstract

Our objective was to develop a technique for performing irreversible electroporation (IRE) of esophageal tumors while mitigating thermal damage to the healthy lumen wall. We investigated noncontact IRE using a wet electrode approach for tumor ablation in a human esophagus with finite element models for electric field distribution, joule heating, thermal flux, and metabolic heat generation. Simulation results indicated the feasibility of tumor ablation in the esophagus using an catheter mounted electrode immersed in diluted saline. The ablation size was clinically relevant, with substantially lesser thermal damage to the healthy esophageal wall when compared to IRE performed by placing a monopolar electrode directly into the tumor. Additional simulations were used to estimate ablation size and penetration during noncontact wet-electrode IRE (wIRE) in the healthy swine esophagus. A novel catheter electrode was manufactured and wIRE evaluated in seven pigs. wIRE was performed by securing the device in the esophagus and using diluted saline to isolate the electrode from the esophageal wall while providing electric contact. Computed tomography and fluoroscopy were performed post-treatment to document acute lumen patency. Animals were sacrificed within four hours following treatment for histologic analysis of the treated esophagus. The procedure was safely completed in all animals; post-treatment imaging revealed intact esophageal lumen. The ablations were visually distinct on gross pathology, demonstrating full thickness, circumferential regions of cell death (3.52 ± 0.89 mm depth). Acute histologic changes were not evident in nerves or extracellular matrix architecture within the treatment site. Catheter directed noncontact IRE is feasible for performing penetrative ablations in the esophagus while avoiding thermal damage.

Keywords: Arrhenius thermal damage; animal experiments; bioheat; finite element model; in vivo validation; irreversible electroporation; joule heating; medical device design; tissue ablation.

Fig. 1

Catheter based wIRE in swine esophagus. (a) The prototype catheter ablation device featuring an electrical connection (triangle) to the pulse generator, a 16/40 mm PET blocking balloon (circle) and a 2.5/10 mm monopolar coil electrode (square). The Luer lock connection to the balloon and a syringe can also be seen. (b) Anterior-posterior fluoroscopy projection showing the ablation device in situ in the swine esophagus. The esophagus lumen (arrowheads) proximal to the blocking balloon (circle) contains the coil electrode (square) and is filled with contrast-enhanced diluted saline. (c) Immediate post-treatment sagittal CT reconstruction showing the coil electrode (square) surrounded by intraluminal contrast media (arrowheads), proximal to the blocking balloon (circle).

Fig. 2

Finite element models of IRE in the esophagus. A catheter with an electrode at the tip is placed within an idealized esophagus (gray arrow), surrounded by muscle (white arrow). (a)Computational model of an esophagus with a tumor obstruction and an electrode centered in the tumor. This model was used for two simulations to compare theeffects of the tumor being surrounded by moist air versus diluted chilled-saline. (b)Acomputational model of an esophagus with a tumor obstruction and an electrode centered in the esophageal lumen adjacent to the tumor. (c) Computational mesh of an idealized healthy swine esophagus with an electrode centered in the lumen. (d) A close-up view of the esophageal tumor with dimensions.

Fig. 3

Schematic of wIRE-Swine study. The catheter body constructed with nylon (top arrow) on which a stainless steel electrode has been mounted (second from top arrow, orange coil), surrounded by diluted chilled saline (blue) and swine esophageal lining (third from top arrow), with a PET nonconductive balloon (bottom arrow) sealing the lumen.

Fig. 4

Electric field strength computation in the esophagus during IRE. In each case 3000 V was applied with an electrode (arrow) in the esophagus, and the EFS was computed in V/cm. The results were collected from the following configurations: (a) and (b) conIRE, the electrode is placed in the center of the esophageal tumor (σ) surrounded by air (Δ) within a human-sized esophagus (μ), (c) and (d) an electrode placed in the center of a healthy swine esophagus (π) surrounded by chilled diluted saline (θ), (e) and (f) conIRE-Sal, an electrode placed in the center of an esophageal tumor surrounded by chilled diluted saline, and (g)–(h) wIRE, an electrode placed adjacent to the tumor in the esophagus while surrounded by chilled diluted saline.

Fig. 5

Estimation of thermal damage percentage during IRE. Anticipated cell death from thermal injury was assessed within the tumor (σ) and the adjacent human and swine esophagus (μ and π, respectively) as a result of IRE pulse application using a modified Arrhenius Equation. For each simulation, a midcross-sectional view that is perpendicular to the electrode (arrow) and midcross-sectional view that is parallel to electrode is presented. (a) and (b) conIRE, the lumen is filled with air (Δ) for this condition. (c) and (d) conIRE-Sal. (e)–(f) wIRE. (g)–(h) wIRE-Swine. The lumen is filled with chilled saline for all other conditions (θ).

Fig. 6

wIRE simulation at varying electrode distance from tumor surface and applied voltage. Each simulation has an electrode (white arrow) placed within an esophagus (=μ), adjacent to a tumor (σ), and surrounded by chilled diluted saline (θ). A range of voltages (1000 V–3000 V) was simulated with a range of distances between the electrode and tumor (1 mm – 7 mm) to determine optimal placement that maximimzes tumor coverage above criticial electrical field strength.

Fig. 7

Comparison of tissue heating during conIRE and wIRE. In each simulation, temperature measurements were taken at 30 s (a), 60 s (b), 90 s (c), and 120 s (d). (1) Idealized healthy swine esophagus (π) with an electrode (arrow) centered in the lumen and surrounded by chilled diluted saline (θ). (2) Esophagus with tumor and (μ) an electrode centered in the tumor (σ) and surrounded by air (circle); conIRE condition. (3) Esophagus with tumor and an electrode centered in the tumor, where the esophageal lumen is filled with chilled diluted saline; conIRE-Sal condition. (4) Esophagus with tumor and the electrode placed adjacent to the tumor while surrounded by chilled diluted saline wIRE condition.

Fig. 8

Estimated temperature graphs during conIRE and wIRE. In each simulation, temperature measurements were taken at 30 s, 60 s, 90 s, and 120 s. The temperature measurements were taken at the line depicted for (a) conIRE and conIRE-Sal, (b) wIRE-Swine, and (c) wIRE. The results were collected from all configurations where the measurements between the dashed lines indicate the lumen: (d) conIRE, (e) conIRE-Sal, (f)wIRE-Swine, and (g) noncontact wIRE.

Fig. 9

Esophageal resections post-IRE-treatment. Photographs of unfixed wIRE treated swine esophagus in (a) axial, treated section are on the left and an untreated control esophagus is on the right, and (b) transverse cross section immediately following euthanasia. IRE treatment manifests as submucosal edema and transmural discoloration of the esophageal lumen.

Fig. 10

H&E stained sections of wIRE treated swine esophagus. (a) Ablated area displaying necrosis of all layers: mucosa (m), submucosa (sm), including glands (g), muscularis propria (mp), and adventitia (a). Approximate location from where images were acquired for subfigures *-(b), -(c) and ^(d). There is submucosal edema (arrow). ((b)–(d) Images ((b), mucosa), ((c), submucosa with glands), and ((d), muscularis propria) are high magnification micrographs showing coagulative necrosis, characterized by nuclear pyknosis and cytoplasmic hypereosinophilia, with retention of architecture and stromal integrity. There is vascular hyperemia (arrowhead). (e) Nerve adjacent to the esophagus, without evidence of injury. Hematoxylin and eosin stain. Original magnification and scale bars: A, 20× and 1 mm; (b)–(e), 600× and 20 μm.