Effect of Differing Parameters on Irreversible Electroporation in a Porcine Model

Abstract

Introduction and Objective:

Irreversible electroporation (IRE) is a new ablative technology to treat small renal masses. We evaluated differed ablation settings on lesion size and temperature changes in a porcine model.

Materials and Methods:

After Institutional Animal Care and Use Committee approval, 36 laparoscopy-guided and 16 open ablations were performed on 13 domestic female pigs. Ablation parameters studied were voltage (1000 V/cm, 1500 V/cm, or 2000 V/cm), probe exposure (1.0 or 1.5 cm), and lesion size over time (survival) (0-, 7-, or 14 day). Temperature changes were monitored during open ablations with differed settings. Gross lesion size was measured, and histologic analysis with hematoxylin and eosin and nicotinamide adenine dinucleotide staining was performed.

Results:

The 1000 V/cm ablations had no gross or histologic lesions. A factorial analysis of variance demonstrated that day (p = 0.56), exposure (p = 0.33), and voltage (p = 0.06) did not demonstrate statistical significance for affecting lesion size. For 1.0 cm probe exposure, 2000 V/cm did more closely approximate expected lesion size (p = 0.02) compared with 1500 V/cm. While significance was not seen for 1.5 cm probe exposure, 2000 V/cm often exceeded expected lesion volume. Only 1 of 4 temperature sensors, located adjacent to one of the IRE probes, noted a significant increase with increased voltage. However, all maximum temperatures remained less than 70°C.

Conclusions:

Variation in lesion volume was seen with different ablation settings in this porcine model. Maximal energy and probe exposure settings should be utilized to ensure full coverage of target volume/mass, potentially without concern for thermal injury to renal collecting system or nearby structures.

Introduction

Irreversible electroporation (IRE) uses short pulses of high-voltage electrical current across tissue, inducing nanopore formation in the cellular membrane phospholipid bilayer and subsequent apoptosis.¹ Early oncologic efficacy of renal tumor IRE appears to be suboptimal compared with more mature thermal ablation literature.² However, more promising results have been seen in liver,³ pancreatic,⁴ and prostate tissue.⁵ As with any new technology, the early learning phase requires experimentation to determine optimal techniques. While prior studies have assessed modifying parameters and their resultant outcomes,^6,7 these were performed in different animal models and organs.⁸ Therefore, less favorable early results could result from an insufficient understanding of modifiable settings for renal IRE.

Given the potential benefits of IRE in renal tumor ablation, including preservation of cellular architecture,⁹ limited damage to renal collecting system or nearby structures,^10

–13 and improved pain compared with radiofrequency ablation,¹⁴ investigations to better understand optimal IRE treatment settings are warranted.

We performed an in vivo study of porcine renal tissue to understand how modifying applied voltage, monopolar probe active electrode exposure, and survival time affects gross lesion size and histology. In addition, we assessed the thermal effects of these various ablation settings.

Materials and Methods

This protocol had full Institutional Animal Care and Use Committee approval. In total, 13 adult female Yorkshire pigs (weight 50–60 kg) underwent either laparoscopy-assisted (n = 9) or open (n = 4) IRE ablation with the NanoKnife^® System (AngioDynamics, Latham, NY). Each animal underwent a total of four ablations: right upper, right lower, left upper, and left lower pole. Pigs that underwent open ablation had real-time temperature monitoring during the ablation and were sacrificed immediately after the procedure (0-day survival). Parameters adjusted for with the various ablations were all selected before surgery and were voltage (1000 V/cm, 1500 V/cm, or 2000 V/cm), probe active electrode exposure (1.0 or 1.5 cm), or lesion size over time (survival) (0-, 7-, or 14 day) (Fig. 1). Each combination of parameters was repeated twice for consistency (36 ablations). For the remaining 16 ablations, parameters were differed among 0-day survival lesions as these had the greatest initial variability and to perform thermal testing.

FIG. 1.

Flow chart of the study design. Exp = probe exposure; V/cm = applied voltage.

Standard parameters for all IRE ablations were 18-gauge monopolar probes in a 2-probe array, spaced 1.5 cm apart, pulse length of 100 μs, and 140 total pulses (70 straight and 70 reverse polarity). Our standard IRE parameters for human renal ablation have been previously published and are similar to above.¹⁵ Each animal was continuously monitored with an electrocardiogram to gate the IRE pulses to the heart rate.

Laparoscopic surgical procedure

After induction of general anesthesia, the pig was placed in the flank position, and the abdomen was prepped and draped in a sterile manner. A Veress needle was used for access, and pneumoperitoneum was instilled to 15 mmHg. Three 12 mm laparoscopic ports were placed. Gerota's fascia was identified and incised exposing the renal capsule. A spacer with guides 1.5 cm apart for the monopolar probes was preplaced onto the lower pole of the kidney. IRE probes were placed percutaneously under visual guidance through the spacer into the lower pole of the kidney. Vecuronium (0.1–1.0 mg/kg) was administered ∼5 min before beginning the ablation for paralysis. A repeat ablation was performed at the upper pole of the ipsilateral kidney. The pig was then positioned on the contralateral flank, and the entire procedure was repeated on the contralateral kidney. For survival surgeries, the laparoscopic ports were removed under visual guidance, and skin was closed with 2–0 polyglactin suture and skin staples.

Open surgical procedure

After induction of general anesthesia, the pig was positioned supine. A midline abdominal incision was made. Colon and small intestine were reflected to expose Gerota's fascia and the kidney. After Gerota's fascia was incised, the 1.5 cm spacer was placed on the lower pole of the kidney, and the two monopolar IRE probes were placed to appropriate depth in renal parenchyma. Four Opsens OTG-MPK-8 fiber-optic temperature sensors (Opsens Solutions, Inc., Quebec, Canada) were secured to a depth of 1.0 cm in the middle of the IRE probes (#1), next to each IRE probe (#2 and #3), and 1.0 cm medial to probe #1 (#4, at periphery of ablation zone) (Fig. 2). Real-time temperature monitoring was performed with Opsens TempSens multichannel signal conditioner. Vecuronium (0.1–1.0 mg/kg) was administered ∼5 min before beginning the ablation for paralysis. After completion of the lower pole, a similar procedure was performed on the upper pole and then the contralateral kidney.

FIG. 2.

Experimental setup of open IRE ablation. (A) The 15 cm monopolar probes are bracketed 1.5 cm apart with the spacer block. (B) Temperature sensor probes are positioned around the monopolar probes for real-time temperature measurement. #1: middle of the IRE probes, #2 and #3: adjacent to each IRE probe, #4: medial to probe #1. Note that the spacer block has been slid up the probe and out of view to allow room for the sensors.

Tissue collection/histologic processing

Pigs were euthanized with a pentobarbital overdose at the predetermined time point. Fresh kidneys were removed, sectioned, and the area of ablation was grossly measured. Representative areas of 0- and 14-day lesions were fixed in 10% formalin, processed, and stained with hematoxylin and eosin (H&E). Fresh sections were frozen in optimal cutting temperature embedding medium and immediately stored at −80°C. Subsequently, the frozen tissue was sectioned to 8-μm thickness, placed on coverslips, and air-dried for 30 min at room temperature. Sections were incubated in the reaction medium consisting of 5.3 mL, 0.2 mol/L phosphate buffer (pH 7.6), 10 mL distilled water, 10 mg nicotinamide adenine dinucleotide (NADH) (Sigma-Aldrich, St. Louis, MO), and 5 mg nitro blue tetrazolium (Sigma-Aldrich, St. Louis, MO) for 30 min at 37°C. Sections were finally rinsed briefly with distilled water, fixed in 10% neutral buffered formalin for 10 min, and rinsed well in distilled water.

All H&E- and NADH-stained slides were evaluated by one pathologist. H&E sections were evaluated for the presence of morphologic alterations, necrosis, and inflammation. NADH-stained sections were evaluated for assessment of cellular viability.

Statistical analysis

Statistical analysis was performed with SPSS version 22 (Armonk, NY). Ablation volume was calculated using the equation for a prolate sphere ( = 4/3 × π × length × width × depth). Ablations are expected to extend 5 mm beyond the monopolar probe. Therefore, given 1.0 or 1.5 cm probe exposure and 1.5 cm spacing, we expect lesions of 2.5 × 2.5 × 1.0 cm or 2.5 × 2.0 × 1.0 cm, giving volumes of 3272 or 2618 mm³, respectively. Mean volumes for each ablation setting were calculated. These values were compared with the calculated experimental values for ablation volume as a percentage and compared between voltages for each probe exposure using a t-test for independent samples. Factorial analysis of variance (ANOVA) was performed for probe exposure, voltage, and pig survival time, to determine factors significant for lesion volume. Factorial ANOVA was also performed for probe exposure and voltage to determine factors significant for temperature changes. Significance was set at p < 0.05.

Results

In total, 50 open or laparoscopic IRE ablations were performed. Two ablations were not included due to device malfunctions from high current exceeding the safety threshold. No gross visible ablation zones were seen at 1000 V/cm for all days studied, and this was confirmed histologically with viable tissue on NADH staining at 14 days (slides not shown). Table 1 shows the mean calculated lesion volumes for the remaining voltages at both probe exposures based on gross measurements after tissue harvest, while Figure 3 depicts this information graphically. A factorial ANOVA demonstrated that all three variables tested, day (p = 0.56), exposure (p = 0.33), and voltage (p = 0.06), did not demonstrate statistical significance for predicting lesion size. However, some obvious trends are noted with regards to these findings. First, 0-day lesions tended to have significant variability, as demonstrated by the larger standard deviations despite more animals undergoing immediate sacrifice (n = 7). Second, 14-day lesions were smaller than 0- and 7 day for every probe exposure and voltage. Finally, 2000 V/cm lesions appear to be bigger than 1500 V/cm. For 1.0 cm probe exposure, 2000 V/cm more closely approximated the predicted lesion size compared with 1500 V/cm (39% ± 27% vs. 82% ± 35%, p = 0.02), although statistical significance was not reached for 1.5 cm (48% ± 36% vs. 112% ± 98%, p = 0.08).

FIG. 3.

Graphical representation of gross lesion volumes for the specified voltages and probe exposure. Black horizontal bars represent the estimated lesion volume.

Table 1.

Calculated Mean Gross Volume and Standard Deviation Based on Length, Width, and Depth Measurements, Using the Formula for a Prolate Sphere (= 4/3 × π × L × W × D)

Tissue harvest	Exposure (cm)	Voltage (V/cm)	Volume (mm³)	SD
0-Day	1.0	1500	1372	74
	1.0	2000	2167	1092
	1.5	1500	2160	1292
	1.5	2000	3803	3870
7-Day	1.0	1500	1276	1182
	1.0	2000	2412	523
	1.5	1500	1137	674
	1.5	2000	4197	434
14-Day	1.0	1500	415	53
	1.0	2000	1754	851
	1.5	1500	574	344
	1.5	2000	2585	823

SD = standard deviation; L = length; W = width; D = depth.

Table 2 shows mean maximum temperatures reached for each temperature probe at the various probe exposures and voltages during open IRE ablation. On factorial ANOVA, only temperature probe 3 showed a statistically significant increase based on the voltage applied, while all other temperature measurements did not have a significant increase with changes in voltage or probe exposure. Regardless of the temperature changes, all mean maximum temperatures were less than 70°C.

Table 2.

Mean Maximum Temperature Measurements for Each Temperature Sensory During Open Irreversible Electroporation Ablation

Exposure (cm)	Voltage (V/cm)	Sensor #1 (mean max. temp. ± SD,° C)	Sensor #2 (mean max. temp. ± SD,° C)	Sensor #3 (mean max. temp. ± SD,° C)	Sensor #4 (mean max. temp. ± SD,° C)
1.0	1500	41 ± 0	44 ± 0	35 ± 0	37 ± 0
1.0	2000	46 ± 11	55 ± 15	52 ± 9	40 ± 4
1.5	1500	44 ± 4	50 ± 5	46 ± 7	41 ± 5
1.5	2000	52 ± 13	53 ± 10	53 ± 6	48 ± 8
F-ANOVA voltage p-value		0.4	0.3	0.02	0.3
F-ANOVA exposure p-value		0.5	0.8	0.17	0.2

Temperature sensor locations are as follows: (1) middle of IRE probes, (2 and 3) adjacent to IRE probes, (4) medial to probe #1.

F-ANOVA = factorial analysis of variance; IRE = irreversible electroporation.

Histologically, all 0-day kidneys showed early tubular injury and congested glomeruli with preservation of cell viability by NADH staining (Fig. 4a, b). By 14-days, kidneys treated at both 1500 and 2000 V/cm showed a central zone of coagulative necrosis involving all kidney compartments and the expected absence of NADH staining (Figure 5a, b). Grossly, the 2000 V/cm lesions appeared to have larger zones of central necrosis compared with the 1500 V/cm zones, but too few samples precluded meaningful statistical analysis.

FIG. 4.

(A) H&E stain of 0-day, 2000 V/cm lesion demonstrated early tubular injury with congested glomeruli (H&E, original magnification × 200). (B) NADH staining of the same lesion showing preserved cellular viability (NADH, original magnification × 200). H&E = hematoxylin and eosin; NADH = nicotinamide adenine dinucleotide.

FIG. 5.

(A) H&E staining of 14-day, 1500 V/cm lesion demonstrating coagulative necrosis. There is an absence of nuclei basophilia in the tubules and glomeruli with preserved tissue architecture (H&E = original magnification × 400). (B) NADH staining of same lesion showing an absence of cellular viability (NADH = original magnification × 200).

Discussion

In this in vivo porcine study of differing parameters with renal IRE ablation, we found that 1000 V/cm was insufficient to produce cell death. Although comparing 1500–2000 V/cm saw no significant differences in lesion size, the greater voltage did appear to be larger both grossly and histologically. In addition, 2000 V/cm more closely and significantly approximated the expected lesion volume for 1.0 cm probe exposure, and although not significant, often exceeded the expected lesion volume at 1.5 cm. Increased probe exposure appeared to have larger lesions, but this was not statistically significant. Finally, while one of four temperature probes did have a significant increase with increasing voltage, maximum temperature for all temperature probes was less than 70°C, confirming the relatively athermal method of action.

Taking into account these findings, we recommend the maximum allowable voltage for a given ablation. It should be noted that high voltages can cause exceedingly high currents, which stop the NanoKnife generator from proceeding with IRE. In this scenario, lowering the voltage to the highest tolerable is advised. For probe exposure, which is often determined by the tumor size, it is best to overexpose the probe to provide sufficient coverage. And finally, increased probe exposure or the higher voltage should not affect treating tumors that abut renal collecting system or nearby structures, as any rise in temperature still remains in the athermal range.

Prior literature has looked at minimum lethal voltages for IRE. Neal and coworkers performed 14 ablations in 8 canine kidneys with 100 pulses (50 straight and 50 reverse polarity), each 100 μs, and voltages of 1250, 1750, and 2250 V.⁶ Euthanasia was performed at 6 h, and their modeling demonstrated a lethal electrical field at 575 V/cm. Similarly, Miklavcic and colleagues mathematically modeled IRE ablation zones on rabbit liver tissue and defined 637 V/cm as the cutoff between reversible and IRE.⁷ This is significantly smaller than our finding that lesions treated at 1000 V/cm maintained NADH staining at 14 days after ablation and likely would preserve cellular activity in the long term. Such discrepancies could be due to variation in ablation protocols, but might also represent differences in tissue properties as modeling was performed in canine kidneys and rabbit liver, respectively.

Other literature has emphasized how variability in tissue conductivity between organs and patients can affect IRE outcomes. Ben-David and colleagues performed 90 IRE ablations on porcine muscle, kidney, and liver (18 total pigs).⁸ Similar to our study they used a two-electrode monopolar array, but their independent variables included pulse exposure, voltage, pulse number, pulse length, and electrode spacing. Animals were euthanized immediately and demonstrated significant variation in lesion size based on tissue type. However, one potential issue is immediate tissue reactivity. In our study, we noted significant lesion variability on 0-day specimens, and prior work has shown that early lesions can overestimate true lesion size due to tissue reactivity and hemorrhagic conversion.⁹

The accuracy of predictive renal modeling for IRE has been previously questioned. Wimmer and colleagues performed 35 computed tomography (CT)-guided IRE ablations in porcine kidneys, with immediate and 24-h postprocedure CT scans.¹⁶ They used different IRE parameters and both monopolar and bipolar electrodes. For monopolar probes they noted that ablation zones on CT measured 81% and 115% of simulated size at 0- and 1 day, respectively, again emphasizing immediate variations. In addition, histopathology showed smaller ablation zones compared with simulated modeling (71%) and CT scans (47%). Although our study did not include imaging findings, this is similar in that most of our gross lesion sizes were smaller than predicted models. In addition, although our numbers were too few to determine significance, we also found that the area of coagulative necrosis appeared smaller than the measured gross lesion.

Consistent with our finding that ablation lesions became smaller over time, tissue involution after IRE has been previously demonstrated.⁹ This has also been confirmed radiologically during follow-up of both an in vivo porcine model¹⁰ and human renal IRE ablation.¹⁵

While early work touted the athermal action of IRE,⁹ some studies have demonstrated measurable temperature changes during ablation.¹⁷ Wagstaff and colleagues noted a peak temperature rise to 79°C in the core of a 4-probe configuration in porcine kidneys after 70 pulses with ablation settings of 1500 V/cm and 1.5 cm spacing. A lower peak temperature of 57°C was seen with a three-needle configuration. Such findings are in opposition to our results, but could potentially reflect the effect of increased overall number of pulses applied between probe pairs. Although we did not find similar results, our simpler probe array might have limited the number of pulses applied and the potential increased temperature. Therefore, this area requires further study to determine the potential thermal risk to nearby structures.

This study has multiple limitations that deserve mention. Most notably, although this was performed in a well-respected animal model of the human kidney, this might not be reflective of a true human response to IRE ablation. In addition, this study was performed on benign renal tissue that might not be applicable to malignant tissue. Although we expect malignancies to be more vascularized, IRE is likely not affected by heat sink. However, the conductivity of renal cell carcinoma is probably different than benign tissue. Therefore, results from benign tissue must be carefully extrapolated to clinical practice. Finally, in this surgical experiment every effort was made to keep probe placement, spacing, and depth constant to truly determine the effect of our independent variables. However, subtle differences in probe depth and spacing, as well as between animal differences in kidney shape and size, could affect our gross outcomes. We felt that 13 animals and a total of 52 ablations would resolve such discrepancies, but inconsistencies could still exist.

Conclusions

In a porcine model of renal IRE ablation, probe exposure, voltage, and survival time did not significantly affect gross lesion size. However, voltages of at least 1500 V/cm are required to induce IRE, and higher voltage appeared to have larger gross lesions. In addition, concerns about thermal effects on collecting system at high voltages and probe exposure are potentially unfounded given that small temperature rises during ablation still cause a maximal temperature less than 70°C in a simple two-probe array.

Footnotes

Acknowledgment

The authors are grateful to Stephanie Shaffer, CVT for her assistance during this project.

Funding

Extra-institutional funding: No; Institutional Animal Care and Use Committee approval: Yes.

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Davalos

, Mir

ILM

, Rubinsky

. Tissue ablation with irreversible electroporation. Ann Biomed Eng, 2005; 33:223–231.

Canvasser

, Sorokin

, Lay

, et al. Irreversible electroporation of small renal masses: Suboptimal oncologic efficacy in an early series. World J Urol, 2017; 116:3135.

Kingham

, Karkar

, D'Angelica

, et al. Ablation of perivascular hepatic malignant tumors with irreversible electroporation. J Am Coll Surgeons, 2012; 215:379–387.

Martin

RCG

, Kwon

, Chalikonda

, et al. Treatment of 200 locally advanced (stage III) pancreatic adenocarcinoma patients with irreversible electroporation: Safety and efficacy. Ann Surg, 2015; 262:486–494.

Ting

, Tran

, Böhm

, et al. Focal irreversible electroporation for prostate cancer: Functional outcomes and short-term oncological control. Prostate Cancer P D, 2016; 19:46–52.

Neal

, Garcia

, Kavnoudias

, et al. In vivo irreversible electroporation kidney ablation: Experimentally correlated numerical models. IEEE T Bio-Med Eng, 2015; 62:561–569.

Miklavcic

, Semrov

, Mekid

, et al. A validated model of in vivo electric field distribution in tissues for electrochemotherapy and for DNA electrotransfer for gene therapy. Biochim Biophys Acta, 2000; 1523:73–83.

Ben-David

, Ahmed

, Faroja

, et al. Irreversible electroporation: Treatment effect is susceptible to local environment and tissue properties. Radiology, 2013; 269:738–747.

Tracy

, Kabbani

, Cadeddu

. Irreversible electroporation (IRE): A novel method for renal tissue ablation. BJU Int, 2011; 107:1982–1987.

10.

Deodhar

, Monette

, Single

, et al. Renal tissue ablation with irreversible electroporation: Preliminary results in a porcine model. Urology, 2011; 77:754–760.

11.

Wendler

, Pech

, Blaschke

, et al. Angiography in the isolated perfused kidney: Radiological evaluation of vascular protection in tissue ablation by nonthermal irreversible electroporation. Cardiovasc Inter Rad, 2012; 35:383–390.

12.

Olweny

, Kapur

, Tan

, et al. Irreversible electroporation: Evaluation of nonthermal and thermal ablative capabilities in the porcine kidney. Urology, 2013; 81:679–684.

13.

Sommer

, Fritz

, Wachter

, et al. Irreversible electroporation of the pig kidney with involvement of the renal pelvis: Technical aspects, clinical outcome, and three-dimensional CT rendering for assessment of the treatment zone. J Vasc Interv Radiol, 2013; 24:1888–1897.

14.

Sorokin

, Lay

, Reddy

, et al. Pain after percutaneous irreversible electroporation of renal tumors is not dependent on tumor location. J Endourol, 2017; 31:751–755.

15.

Trimmer

, Khosla

, Morgan

, et al. Minimally invasive percutaneous treatment of small renal tumors with irreversible electroporation: A single-center experience. J Vasc Interv Radiol, 2015; 26:1465–1471.

16.

Wimmer

, Srimathveeravalli

, Gutta

, et al. Planning irreversible electroporation in the porcine kidney: Are numerical simulations reliable for predicting empiric ablation outcomes?. Cardiovasc Inter Rad, 2015; 38:182–190.

17.

Wagstaff

PGK

, de Bruin

, van den Bos

, et al. Irreversible electroporation of the porcine kidney: Temperature development and distribution. Urol Oncol, 2015; 33:168.e1–e7.