Microwave Ablation Antenna for Functional Adenomas in the Adrenal Gland

Microwave thermal ablation has been thoroughly investigated in the last decades. However, new challenges for this technology are nowadays represented by specific applications. The present work investigates a promising ablative solution for the treatment of functional adenomas in the adrenal gland. In this paper, a numerical study conducted to build and test an optimized internally cooled triaxial antenna is presented. The antenna has been designed to minimize its transversal dimension merging radiating and cooling elements. The proposed antenna has been then built in our laboratories and experimentally tested on ex vivo biological tissues. Numerical and experimental results will be reported and discussed.


INTRODUCTION
Microwave thermal ablation (MWA) is a widespread minimally-invasive technique used for the treatment of different solid tumors [1]. MWA exploits the interaction between the electromagnetic (EM) field radiated by an antenna at microwave frequencies and the target biological tissue to induce a cytotoxic temperature increase [1]. The technique is nowadays well investigated and the researchers in the field focus their work on the development of accurate treatment planning and tools for the physicians [2]. Lately, the application of MWA to specific and novel targets and areas has been considered [3]. In this scenario, microwave thermal ablation has been recently taken into consideration for the treatment of medical conditions correlated or caused by anomalies in the adrenal gland.
Functional adenomas in the adrenal gland are the cause of several hypertensive clinical conditions. The most common form of secondary hypertension is represented by Primary aldosteronism (PA) (or Conns syndrome). Primary Aldosteronism (PA) represents up to 12 % of all cases of hypertension within the population [4][5][6]. In patients with resistant hypertension, the prevalence of PA approaches 20% [6]. The most commonly performed treatment for PA is adrenalectomy: the complete removal of the adrenal gland. Unfortunately, PA often reoccurs in the remaining contralateral adrenal gland. In such cases, and in the case of bilateral diseases (60% of the cases) [6], the disease is managed with pharmaceuticals, which are poorly tolerated due to their side effects. Thermal ablation treatments have been primarily used to remove the entire adrenal gland (i.e., thermal adrenalectomy) showing detrimental effect [7]. However, recently thermal ablation has been proposed to locally eliminate adenomas in the adrenal gland [8]. MWA has been mainly adopted to eliminate adrenal metastasis due to its ability to induce larger lesions; but its safety and efficacy have been investigated also in the treatment of functional adenomas [9][10][11]. The advantage of MWA is its ability to efficaciously and safely treat small target, as well as large lesion. Moreover, MWA does not require long treatment time (typically less than 20 min). This short treatment time not only reduces the whole time of the intervention, with obvious effects on time-cost efficacy and patient-stress, but it reduces the time of stimulation of the adrenal gland and the subsequent occurrence of hypertensive crisis [11].
Microwave ablation represents a valuable solution to locally eliminate adenomas preserving the healthy functional adrenal tissue surrounding the adenoma and proximal sensitive structures (e.g., blood vessels, nerves). The benign shallow hormone-eluding adenomas are usually small ( < 20 mm in diameter) and located in the gland cortex periphery. The adrenal cortex protects the medulla that regulates the blood pressure through the adrenaline production, and it is surrounded by a fat layer with embedded arteries. A well-focused and controlled ablation is required to selectively treat these adenomas, sparing the healthy adrenal tissue and the surrounding sensitive structures such as nerves or blood vessels. Since puncturing of the gland is undesirable, the optimal positioning of the ablation antenna would be parallel to the adenoma at the interface between the gland and the fat layer. Additionally, the applicator would be required to be directional and radiate only in a preferred direction. To achieve this objective, several studies in the literature have presented directive applicators [12][13][14]. There are two major drawbacks of these directive applicators. Firstly, the increase of the transversal dimension of the antenna increases its invasiveness. In a target as small as the adrenal gland (about 15 cm 3 [15]), an increase of even less than 1 mm in diameter can be significant. Secondly, directive antennas must be precisely oriented accordingly with the preferred direction. This requirement represents a complication of the clinical procedure and discourages clinicians from adopting this solution in the treatment of functional adenomas of the adrenal gland. Other studies have suggested the exploitation of the natural presence of the fat layer surrounding the adrenal gland to obtain directive pattern of ablation [15][16][17]. These studies illustrated how the interface between two tissue with high contrast in dielectric properties (e.g., fat/adrenal cortex, or fat/muscle) can affect the radiation pattern of a simple omnidirectional antenna (e.g., monopole) to obtain an ablation in a preferred direction.
Due to heterogeneity and reduced dimensions of the target area (adrenal gland), the ablative performances of the antenna can vary in correlation with the antenna detuning. To overcome this issue, we present a triaxial antenna structure in this study. The triaxial structure was obtained by modifying a monopole antenna equipped with a cooling system based on polyimide tubes. The inner tube of the cooling system was replaced with a stainless-steel tube to achieve the proposed triaxial antenna structure. The distance of this tube from the feed point of the monopole antenna was tuned to optimize the electromagnetic energy transfer without reducing the flux cooling water efficiency. The proposed antenna, operating at 2.45 GHz, guarantees minimal-invasiveness (16gauge) and robustness, allowing the clinicians to operate in optimal conditions independently from the antenna insertion depth and orientation.
The proposed antenna design was firstly evaluated numerically using the adrenal gland model presented in [15]. Next, the antenna was fabricated locally and experimentally evaluated using ex vivo liver and muscle tissues in terms of radiation performance and induced ablation zone. The rest of the paper is organized as follows: Section 2 describes numerical and experimental methodology; Section 3 presents results of numerical and ex vivo experimental evaluation; and finally conclusions are presented in Section 4.

Numerical Study
A 3D commercial full-wave electromagnetic software (CST MWS Suite 2018, Darmstadt, Germany) was used to conduct the numerical study. Firstly, the MWA applicator was modelled as a simple 23gauge monopole antenna with the characteristics of a UT-020 coaxial cable as specified in Table 1. A parametric study for different monopole tip lengths, i.e., from 12.0 mm to 4.0 mm, was conducted. The optimal length of the exposed tip was selected to obtain the minimum reflected power and the maximum resonance bandwidth of the reflection coefficient (S 11 ) at the operating frequency of 2.45 GHz. Then, the optimized monopole antenna was inserted in a third concentric 18-gauge stainless-steel tube; the optimal distance of the end of this third tube from the antenna feed point was investigated through a parametric simulation. The parametric sweep was conducted varying the positioning of the stainless-steel tube with respect to the base of the monopole tip from −5.5 mm to 2.5 mm. The reflection coefficient for the different proposed configurations was tested in water. Finally, an external 16-gauge polyimide tube was used to complete the flowing paths of the integrated cooling system. The stainless-steel tube provides the onward path of water flow, whereas the outer polyimide tube provides the return flow path.
The performances of the two designed ablation applicators (23-guage coaxial and 18-guage triaxial), equipped with the refrigerating system (16-gauge), were tested in liver in terms of the magnitude of the reflection coefficient and in terms of the resonance bandwidth within the 0-3 GHz frequency range. The antennas were placed at the center of a block mimicking the mechanical, electric and thermal properties of the chosen biological tissue. Two different antennas' insertion depth (10 and 30 mm from the feed point) were also tested to verify the improved robustness of the triaxial antenna with respect to the coaxial one. Finally, the triaxial antenna was placed in the adrenal numerical model presented in [15], parallel to the interface between adrenal cortex and fat. The dielectric properties of liver and fat tissues, as well as the values related to heat capacity, thermal conductivity and density were manually loaded in the material settings of the CST MWS software from the ITIS database (ITIS Foundation, Zurich, CH). The thermal properties of the adrenal gland were obtained from the same source, while the dielectric properties were obtained from [18]. A tetrahedrons-based meshing was used, which was made denser for the material with higher relative permittivity. The water temperature of the cooling system was fixed at 18 • C during all simulations. Due to the ex vivo experimental scenario proposed, the starting temperature was considered at 25 • C for the biological tissues and the blood perfusion parameters were not included. Coupled electromagnetic and thermal simulations were executed for a single power level of 60 W for a treatment duration of 60 s.

Experimental study
The designed triaxial antenna was prototyped in our laboratory. A monopole antenna with an exposed tip of 6.5 mm was made with a UT-020 coaxial cable (Micro-Coax Inc., Pottstown, PA, US); it was inserted in an 18-gauge stainless-steel tube, positioned 0.5 mm away from the monopole base, and in a 16-gauge polyimide casing to implement efficient water cooling. The stainless-steel tube was simultaneously used for the antenna cooling system and to improve the antenna performance. An SMA connector was placed at 165.0 mm from the antenna feed-point. The fabricated antennas' performances were evaluated in terms of reflection coefficient (S 11 ) by connecting the antenna to an antenna analyser (Rohde & Schwarz ZVH8 100 kHz-8 GHz) through a low-phase coaxial cable. A full one-port calibration procedure was performed, considering three different calibration standards: open circuit, short circuit and 50 Ω load. Measurements were conducted in water at room temperature. The S 11 values were recorded at 201 linearly spaced frequency points over 0.5-3 GHz. The in-house triaxial antenna was then used to conduct ex vivo experimental studies to test its performances. The experiments were conducted on ex vivo lamb liver and porcine muscle. The temperature of the material under test was measured using an infrared thermometer (Fluxe 62 Max IR Thermometer, −30 • C-500 • C temperature range, accuracy of 1.5 • C of reading at temperature ≥ 0 • C). A single power level of 60 W was set at the microwave generator (Sairem, SAS, France) at 2.45 GHz. Ablations were performed for 30, 60 and 90 s in liver, and for 60 s in muscle, for a total of 6 experiments. A peristaltic dispensing pump (DP2000, Thermo-Fisher Scientific Inc., Waltham, Massachusetts, US) was connected to the inflow channel of the ablation applicator, operating at 50 ml/min. After each experiment, radial and longitudinal ablation extents were measured by a millimeter ruler. The longitudinal dimension was evaluated parallel to the antenna axis and the radial dimension was measured perpendicularly to the antenna axis at the feed-point.

Numerical study
The MWA antenna performances were evaluated in terms of the magnitude of the reflection coefficient and in terms of the resonance bandwidth within the 0-3 GHz frequency range. From the parametric simulations conducted in water, the optimal tip length of the MWA 23-gauge monopole antenna was found to be 6.5 mm; whereas the optimal distance of the third concentric 18-gauge stainless-steel tube from the feed point resulted −0.5 mm. The performances of the optimized monopole and triaxial antennas were tested in liver for two different insertion depths: 10 mm and 30 mm from the feed point. The reflection coefficients (S 11 ) simulated are reported and compared in Figure 1. It can be observed from Figure 1 that the third metallic element guarantees the antenna matching independently from the antenna insertion depth, ensuring less than 10% of reflected power.
The triaxial antenna was adopted to conduct coupled EM-thermal simulation to investigate the effect of the target area heterogeneity on the resulting ablation zone. A single setting of 60 W-60 s was tested placing the antenna both in liver and in an adrenal numerical model [15]. The adrenal model is proposed as generic representative of a heterogeneous scenario. A simulated reflection coefficient (S 11 ) of −10 dB is guaranteed also in the adrenal gland model. Figure 2 illustrates the thermal profiles obtained from the simulations. The edge of the thermally ablated region is considered where the tissue reaches a temperature of approximately 55-60 • C [19], thus in the figure, the estimated ablation zone is represented by the red area. Figure 2 confirms a possible shielding effect of a tissue with low dielectric properties, resulting in a directive ablation pattern, as desired, independently from the antenna orientation [16,17], for controlled settings of power and time.

Experimental study
The triaxial MWA antenna used in the experimental study showed a reflection coefficient of −18 dB at the operating frequency 2.45 GHz. This value resulted as an improvement of the reflection coefficient measured for the monopole antenna (−14 dB) before adding the stainless-steel tube, during the antenna fabrication.
The final antenna prototype was used to conduct different ex vivo tests (N = 6). In liver, a 3% of reflected power was observed for all the experiments, whereas in muscle, a 7% of reflected power was observed. Table 2 summarizes the experimental settings and the results obtained and Figure 3 illustrates the ablation areas experimentally obtained with the different settings. It can be observed that the results obtained in the simulation are confirmed by the experimental practice. A good sphericity in the ablation zones obtained in liver for 60 and 90 s is noticeable (≥ 0.85), while for shorter time of ablation a poor sphericity is observed. The ablation zone in muscle results 30% bigger than in liver, given the same power and time settings, and a sufficient sphericity index is guaranteed (> 0.75). A structural difference in the two tissue (i.e., the directionality of the muscle fibres) and a difference in the dielectric properties can determine a different absorption of the EM field [20].

CONCLUSION
In this work, an optimized minimally-invasive triaxial antenna is presented. The performance improvements linked to the introduction of a third metallic element to a monopole antenna are numerically illustrated. The triaxial antenna introduces robustness to the antenna performances, guaranteeing a good impedance matching even for reduced insertion depth (10 mm). The prototyped antenna was successfully tested in ex vivo biological tissues.
Exploiting the stainless-steel tube as element of the cooling system, the antenna radial dimension is not enlarged and the cooling efficacy is preserved. The possibility of using the proposed antenna in heterogeneous scenario has been numerically proved. This ablative solution would permit the clinicians to efficiently and safely use an ablation device for the treatment of functional adenomas of the adrenal gland, without caring of the antenna orientation and insertion depth.
Further study will be conducted to validate the proposed antenna and its application in adrenal gland models ex vivo and in vivo.