Electrochemical Lipolysis Induces Adipocyte Death and Fat Necrosis: In Vivo Pilot Study in Pigs

Background: Current minimally invasive fat reduction modalities use equipment that can cost thousands of U.S. dollars. Electrochemical lipolysis (ECLL), using low-cost battery and electrodes (approximately $10), creates acid/base within fat (width, approximately 3 mm), damaging adipocytes. Longitudinal effects of ECLL have not been studied. In this pilot study, the authors hypothesize that in vivo ECLL induces fat necrosis, decreases adipocyte number/viability, and forms lipid droplets. Methods: Two female Yorkshire pigs (50 to 60 kg) received ECLL. In pig 1, 10 sites received ECLL, and 10 sites were untreated. In pig 2, 12 sites received ECLL and 12 sites were untreated. For ECLL, two electrodes were inserted into dorsal subcutaneous fat and direct current was applied for 5 minutes. Adverse effects of excessive pain, bleeding, infection, and agitation were monitored. Histology, live-dead (calcein, Hoechst, ethidium homodimer-1), and morphology (Bodipy and Hoechst) assays were performed on day 0 and postprocedure days 1, 2, 7, 14 (pig 1 and pig 2), and 28 (pig 2). Average particle area, fluorescence signal areas, and adipocytes and lipid droplet numbers were compared. Results: No adverse effects occurred. Live-dead assays showed adipocyte death on the anode on days 0 to 7 and the cathode on days 1 to 2 (not significant). Bodipy showed significant adipocyte loss at all sites (P < 0.001) and lipid droplet formation at the cathode site on day 2 (P = 0.0046). Histology revealed fat necrosis with significant increases in average particle area at the anode and cathode sites by day 14 (+277.3% change compared with untreated, P < 0.0001; +143.4%, P < 0.0001) and day 28 (+498.6%, P < 0.0001; +354.5%, P < 0.0001). Conclusions: In vivo ECLL induces fat necrosis in pigs. Further studies are needed to evaluate volumetric fat reduction. Clinical Relevance Statement: In vivo ECLL induces adipocyte death and fat necrosis. ECLL has the potential to be utilized in body fat contouring.

Volume 153, Number 2 • In Vivo Electrochemical Lipolysis 335e new, low-cost, nonsurgical fat reduction modality involving the use of in situ water hydrolysis to modify the biochemical and physical tissue properties.We have demonstrated that in situ water electrolysis to generate acid (at the anode) and base (at the cathode) can modify cartilage pliability, [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32] cause localized collagen injury to skin, 33- 37 and induce adipocyte necrosis and lipolysis when applied to fat. 38,39In this latter application, referred to as electrochemical lipolysis (ECLL), saline is first injected into fat for tumescence and to increase tissue electrical conductivity.Subsequently, platinum electrodes are inserted through the fat.A potential is then applied from a direct current power supply, through which pH gradients (width, approximately 3 mm) are formed by means of water oxidation.In effect, ECLL induces acute adipocyte necrosis through membrane lysis, nuclear degradation, and lipolysis-the hydrolysis of triglycerides (TGs) into fatty acids and glycerol, as evidenced in our prior ex vivo studies using porcine tissue (Fig. 1). 39In our prior experiments, approximately 40% of the TG mass liberated during cathodic ECLL was hydrolyzed into free fatty acids and glycerol, whereas anodic ECLL resulted in hydrolysis of approximately 48% of liberated TGs. 3935]39 However, these numeric models do not easily evaluate how live biological tissues heal and respond to injury. 40In this pilot study, we examine the in vivo longitudinal effects of ECLL in pigs, using histology and fluorescence microscopy assays for adipocyte viability and morphology.Given what is known about wound healing after chemical injury, 41,42 we hypothesize that ECLL decreases adipocyte viability and number, forms lipid droplets, and induces fat necrosis.

Study Design
Two female Yorkshire pigs (50 to 60 kg) received ECLL on their dorsum with multiple treatment sites to determine the longitudinal effect of ECLL (Fig. 2).Two sites on the pig dorsum were allocated for untreated controls and two sites were allocated for ECLL treatment for each evaluation time point: day 0 (2 hours after the procedure) and days 1, 2, 7, and 14 for the Plastic and Reconstructive Surgery • February 2024 336e first pig; and day 0 (2 hours after the procedure) and days 1, 2, 7, 14, and 28 for the second pig.For the first pig, a total of 10 sites were treated with ECLL and 10 sites were untreated, whereas for the second pig, there was a total of 12 ECLL treated sites with 12 untreated sites.Each site was 3.5 × 2.5 cm from the adjacent site.Tissue biopsy specimens (20 biopsy specimens for pig 1 and 24 biopsy specimens for pig 2) were acquired at each time point for histology and fluorescence assays to assess adipocyte viability and morphology.In investigating ECLL's ability to induce fat necrosis, the primary outcome of the study was to show at least a 30% statistically significant increase in average particle area of lipid droplets and adipocytes in histology in adipose tissue at sites treated with ECLL compared with the untreated sites at day 28.Secondary outcomes included at least a statistically significant 30% decrease in the number of adipocytes at day 28, and a statistically significant 30% increase in lipid droplets at day 0 at ECLL-treated sites compared with untreated sites.Furthermore, another secondary outcome of the study was to show 30% decreased calcein-AM (calcein) fluorescent signal area and 30% increased ethidium homodimer-1 (EthD-1) signal area to show decreased cell viability at ECLL-treated sites compared with untreated sites at day 0. Pigs were killed humanely with intravenous pentobarbital sodium/phenytoin sodium (0.3 mL/kg).The protocol was performed in accordance with the institutional animal care and use committee of the University of California, Irvine (AUP-17-164).

ECLL Procedure
Following shaving, normal saline (1 mL) was injected into subcutaneous fat at each treatment site.Following cannulation by means of a 25-gauge hypodermic needle, 30-gauge platinum needle electrodes (Natus, Pleasanton, CA) were inserted 3 mm apart and 1 cm deep into the skin and subcutaneous tissues.Using a direct current power supply, ECLL was administered at 5 V (transferred charge average 1.7 ± 0.18 coulomb) for 5 minutes while monitoring current.The untreated control followed an identical procedure as previously mentioned without current application.Six 2-mm tattoos indicating x and y planes of electrode sites were placed for colocalization.Ten-millimeter punch biopsy specimens of skin and subcutaneous tissue at control and treatment sites were attained and cross-sectioned; each half was used for the experiments below.Biopsy sites were sutured and covered with Telfa pads (Covidien, Minneapolis, MN); Tegaderm film (Covidien, Minneapolis, MN); Ioban drapes Fig. 2. Illustration of ECLL treatment grid applied to the dorsal surface of pigs.Each experimental area was treated with saline tumescence and ECLL at 5 V for 5 minutes.Anode (red) and cathode (black) sites are denoted by colored pinpoint circles.On the opposite side, matched control sites were placed where sham treatment (saline and electrode insertion only without ECLL treatment) was performed.Ten-millimeter punch biopsy specimens, shown by a dotted circle, were obtained on each experimental day for further analysis by histology, and morphologic and viability fluorescent microscopy assays.
Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.

Tissue Staining Using Fluorescent Dyes
Cross-sectioned biopsy specimens of untreated sites were sliced into thirds and submerged for 2 hours in 1 N hydrochloric acid (HCl) or 10 N sodium hydroxide (NaOH) (Thermo Fisher, Waltham, MA) for positive controls, or wrapped in Hank's Balanced Salt Solution-soaked gauze for untreated controls.

Live-Dead Assay
Specimens were stained with 800 μL of calcein (10 μg/mL), EthD-1 (0.75 μg/mL) (Invitrogen E1169, Carlsbad, CA), and Hoechst 33342 (2.5 μg/mL) at 37°C for 45 minutes.Calcein is a nonfluorescent cell-permeable dye that is cleaved by intracellular cytoplasmic esterases of live cells to calcein.In cells with intact membranes, calcein-which emits a green fluorescence signal-is retained.Ethidium homodimer, a nonpermeable DNA stain, was used to stain nuclei of dead cells with damaged membranes, emitting red fluorescence.Hoechst was used to stain nuclei of all cells.With this combination, loss of cytoplasmic green calcein signal and nuclear blue Hoechst signal, and the increased nuclear red EthD-1 signal, indicates the loss of cell viability and cell death.Samples were washed before imaging.

Quantification of Fluorescence Assays
For the morphology assay, adipocytes and lipid droplets were counted manually with Fiji (ImageJ) software by three trained counters blinded to treatment conditions.Adipocytes included well-formed Bodipy staining with eccentric nuclei Hoechst staining.Lipid droplets contained Bodipy staining without Hoechst staining.
To quantify the live-dead assay, a custom Fiji macromeasured the fluorescence signal area for each channel.Intensity thresholding was first performed for each channel to remove background noise.Second, the intensity threshold was converted into a binary mask.The areas of individual discrete elements of the binary mask were measured for each condition.

Histology with Quantitative Analysis
Tissues were fixed in 10% buffered formalin, processed, and paraffin embedded. 43ight-micron-thick slides were stained with hematoxylin and eosin.At least three slides per treatment condition were evaluated by two observers.Representative images of untreated and ECLLtreated sites were acquired using a light microscope (Olympus).Average particle area was calculated using Fiji. 44Histologic images at 50× were thresholded by intensity to create a binary mask of lipids present (lipid droplets and adipocytes) and subsequently underwent binary watershed segmentation.The Fiji particle analysis algorithm calculated each particle area.

Statistical Analysis
Statistical analysis was performed using PRISM (v9.3.0;GraphPad, San Diego, CA).The Shapiro-Wilk test was used for normality testing.The morphology and histology data were found to be distributed normally.One-way analyses of variance with post hoc Tukey t tests were performed for the morphology data.Two-way analyses of variance with post hoc Tukey t tests were performed for histology.As the live-dead fluorescence data were not distributed normally, the Kruskal-Wallis nonparametric test was performed.Statistical significance was determined at values of P ≤ 0.05,

RESULTS
All animals survived the study's duration.No adverse effects of excessive pain, excessive bleeding, infection, or animal agitation was noted.

ECLL Induces Tissue Inflammation and Fat Necrosis
Histology (10×) showed that ECLL induced tissue inflammation and necrosis (Fig. 3).The untreated control showed normal fat tissue architecture.On day 0 at the cathode, adipocytes without clear nuclei were shown.On day 7 at both electrode sites, anucleated adipocytes with mixed inflammatory and fibroblast infiltrate along the periphery of the injured area were shown.On day 28, fat necrosis at both electrode sites was shown.
Further microscopic histologic characteristics were visualized at 50× magnification (Fig. 4).Untreated control showed regular adipocyte spherical morphology and eccentric nuclei.At the cathode on days 0 to 7, adipocyte nuclei were lost.On days 1, 2, and 7 at the anode, nuclei were lost and adipocyte morphology was significantly altered.On day 7 at the anode and day 14 at both electrode sites, lymphocytes and macrophages (with some plasma cells and neutrophils) were present on the periphery of treated areas, in addition to fibroblasts and fibrosis surrounding lipid droplets.On day 14, inflammatory infiltrates traveled to the central treatment area.Foamy macrophages filled with lipids were present at both electrode sites.Fat necrosis (with macrocystic and microcystic fat changes with associated fibrosis) occurred on days 14 and 28.

ECLL Induces Lipid Droplet Formation with Loss of Fat Staining
Adipocyte and lipid droplet morphology findings are shown in Figure 5. Untreated control Bodipy fluorescence showed circular adipocytes containing eccentric nuclei, organized into lobules, with minimal lipid droplet formation.Acutely, HCl-treated tissues showed statistically significantly fewer adipocytes (160 versus 363.3;P = 0.001).Damage was greater in NaOH-treated tissues with complete adipocyte destruction (0 versus 363; P < 0.0001) and decreased staining and no nuclear Hoechst signals.In ECLL specimens over 28 days, there was a substantial loss in Bodipy staining of lipids after ECLL at both anode and cathode electrodes, particularly on days 14 and 28.Compared with the untreated site, the mean numbers of adipocytes in all conditions and time points were statistically significantly decreased (Table 2).Compared with untreated, there was a statistically increased mean number of lipid droplets in anode on day 1 (504.0versus 183.0;P = 0.0260) and cathode on day 2 (560.7 versus 183.0;P = 0.0046) (Table 3).Notably, nuclear Hoechst staining of adipocytes near the anode was present on day 0, and diminished through day 28.The cathodes revealed diminished nuclear Hoechst adipocyte staining on day 1 and continued to diminish through day 28.Hoechst stain surrounding remnant adipocytes and lipid droplets on days 7, 14, and 28 at the cathode sites and days 14 and 28 at the anode sites was visualized.Descriptive morphology assay findings are summarized.(See Table, Supplemental Digital Content 1, which shows a descriptive summary of morphology assay findings, http://links.lww.com/PRS/G415.)

ECLL Degrades Adipocyte Nuclei and Induces Cell Death
Findings of the live-dead assay are shown in Figure 6.Comparisons are made between mean EthD-1 (Table 4), Hoechst (Table 5), and calcein signal areas of untreated and ECLL-treated sites; however, there were no statistically significant differences (Table 6).In the untreated control, intact live adipocytes were represented by cytoplasmic calcein signal with nuclear Hoechst signal.Conversely, HCL-treated tissues showed Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.

DISCUSSION
Current minimally invasive modalities for fat contouring may require expensive equipment. 4,6,9,10,12,46ECLL has been shown to damage adipocytes and saponify TGs, 38,39 a technique that can potentially be used for low-cost fat contouring.To move toward clinical evaluation, understanding how tissue injury following ECLL evolves over time is imperative.Using histology and fluorescence assays to investigate adipocyte viability and morphology, this pilot study demonstrates that in vivo ECLL lyses adipocytes and induces fat necrosis in pigs.
Controlled production of acid/base following ECLL 38,39 induces localized fat necrosis, as confirmed with histology.Acutely, ECLL induced  Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.
Plastic and Reconstructive Surgery • February 2024 342e adipocyte membrane lysis, emptying of adipowith irregular membranes, and the formation of lipid droplets.In addition, adipocytes were enucleated at cathode and anode sites indicating nuclear degradation and fat necrosis. 47Days 14 and 28 showed and microcystic fat changes, occurring from irreversible adipocyte injury or acute inflammation. 48Following ECLL, inflammatory cells were recruited, including predominantly lymphocytes and foamy macrophages (laden with lipid products released from dead adipocytes).These findings suggest the immediate effects of initiating cell death and the indirect response of inflammatory chemotaxis to the area.These inflammatory cells can also directly disrupt or indirectly destroy adipocyte membranes by means of cytokine or lytic enzyme release. 49][51][52] Unlike histology which removes fat during processing, the use of Bodipy lipid staining with fresh tissue is effective in visualizing the presence of fat within adipocytes and lipid droplets.Bodipy staining illustrated adipocyte loss, lipid   53 The Hoechst staining around adipocytes most evident on days 14 and 28 again suggests inflammatory and fibroblast infiltration, with contribution of foamy macrophages in clearing the affected area of lipid products.
The lack of adipocyte calcein signal, with increased EthD-1 signal and diminished Hoechst signal, indicate adipocyte death, which is evident following ECLL.Combinations of these stains have been previously validated to assess adipocyte viability and death, 52,54 particularly with other minimally invasive adipose contouring techniques, including deoxycholate, phosphatidylcholine, and photobiomodulation. 49,55,56ECLL also destroys adipocyte nuclei contributing to cell death, indicated by the loss of both nuclear EthD-1 and Hoechst signals, which begins earlier at the cathode site (day 1) than the anode site (day 2).Of note, histology showed clear enucleation earlier at the cathode site on day 0. The timing Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.
Plastic and Reconstructive Surgery • February 2024 344e difference may be related to increased of biochemical dyes and functional imaging compared with hematoxylin and eosin visualization under light microscopy.
Through this study, we have investigated the longitudinal effects of ECLL treatment on subcutaneous porcine fat.Shown by means of histologic and biochemical fluorescence analysis, ECLL leads to adipocyte lysis, lipid droplet formation, and nuclear degradation; followed by inflammation, fibrosis, and fat necrosis and resorption.This method forms acid and base in situ, which can potentially be targeted to specific tissue locations for adipocyte death and subsequent body fat contouring.Insulating electrodes may easily obviate the skin effect.It is important to consider that ECLL's use may be limited in the patient population with implanted electrical devices, such pacemakers or cardioverter defibrillators.
Because of the costly and labor-intensive nature of survival studies, this pilot study has a small sample size of two animals; thus, power analysis was not performed and statistical comparisons were limited.At this current dosimetry, ECLL was shown to be safe and effective in inducing fat necrosis up to 28 days.At day 28, lipid deposits remain in the tissues and prevent the clinical observation of fat loss, which could potentially be observed after 3 months, as seen in other adipose contouring studies. 46,57Thus, further studies with

CONCLUSIONS
In this pilot study, we have demonstrated, by means of histology and fluorescence assays to assess adipocyte viability and morphology, that in vivo ECLL induces adipocyte death and fat necrosis in pigs.Further larger and longitudinal studies are needed to further confirm these findings.ECLL may eventually be adapted for procedures that alter adipose tissue morphology, such as in body fat contouring.

DISCLOSURE
The authors have no financial disclosures to report.

DISCLAIMER
The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Fig. 1 .
Fig. 1.Biomechanisms of ECLL.(Above) In ECLL, two platinum electrodes are inserted into fat tissues after saline injection.A direct current is then applied to the fat with current drawn from a power supply, creating in vivo H+ ions at the anode electrode and OH− ions at the cathode electrode.ECLL destroys adipocytes by means of (below, left) membrane lysis, (below, center) nuclear degradation, and (below, right) saponification of TGs.

Fig. 3 .
Fig. 3. Hematoxylin and eosin histology at 10× magnification of adipose tissue illustrating the longitudinal effects of ECLL.(Left) Untreated control, (Right, above panels) anode region, and (Right, below panels) cathode region.(Below, second from right) An electrode insertion site is depicted by an outline of the electrode.(Below, right) Spatial selectivity of ECLL-induced adipocyte necrosis at the cathode region adjacent to unaffected tissue.Black arrow, electrode insertion site; Blue arrow, demarcation between cathode and untreated areas.Scale bars = 400 µm.

Fig. 4 .
Fig. 4. Hematoxylin and eosin histology at 50× magnification of adipose tissue illustrating the longitudinal effects of ECLL.(Left) Untreated control, (right, above panels) longitudinal representation of adipose tissue surrounding the anode electrode of a 5-V, 5-minute ECLL treatment, and (right, center panels) longitudinal representation of adipose tissue surrounding the cathode electrode of a 5-V, 5-minute ECLL treatment.Green circle, foamy macrophages; blue asterisk, macrocyst; black arrow, fibroblast.Scale bars = 200 µm.(Below) Particle area of lipid droplets and adipocytes are shown at the untreated, anode, and cathode sites, with average and SEM bars.Statistical significance was determined at *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001.

P
Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.≤0.01, and P ≤ 0.001.All parametric data are expressed as mean, SEM, and percentage change.Nonparametric live-dead fluorescence data are expressed as median and interquartile range.

Table, Supplemental Digital Content 2, which
shows a descriptive summary of live-dead assay findings,

Table 1 . Comparisons of Average Particle Area of Lipid Droplets and Adipocytes in Histology
a P values of ≤ 0.01 were statistically significant.bP values of ≤ 0.001 were statistically significant.Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.Volume 153, Number 2 • In Vivo Lipolysis 341e

Table 3 . Comparisons of Lipid Droplets Quantity in Morphology Assay
Volume 153, Number 2 • In Vivo Electrochemical 343e droplet formation, inflammation, and fat necrosis with ECLL.Loss of Bodipy staining was most significant on days 14 and 28, suggesting a loss of fat.However, adipocyte necrosis was clearly evident early at day 0, shown by diminished nuclear Hoechst staining.Similar findings of absent nuclear Hoechst staining have been observed to distinguish nonviable adipocytes from viable cells by Suga et al., who used various biochemical cell stains to evaluate liposuctioned fat samples.
Copyright © 2023 American Society of Plastic Surgeons.Unauthorized reproduction of this article is prohibited.

Table 5 . Comparisons of Hoechst Signal Area in Live-Dead Assay
larger sample sizes and longer times will assess macroscopic effects and evaluate the reabsorption of the saponified TGs induced by ECLL.Longer term studies will also elucidate any oil cyst formation (not seen in this current study), long-term fibrosis, skin dimpling, or skin laxity as potential complications.Future studies are also needed to investigate the effects of variations in geometric and temporal electrode distributions, which may alter the amount of adipose change or spatial effect and accuracy.If multiple needles or treatments are required for clinical fat volume changes, this technology may be limited by patient discomfort and inflammation.Although no excessive pain of the animal was observed in our study, future investigations of procedure-related pain without sedation are needed.