Document Type : Research Paper
Authors
1 Department of Medical Physics and Biomedical Engineering, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran
2 Nanotechnology Research Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
3 Department of Medical Physics, School of Medicine, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
4 Department of Clinical Oncology, School of Medicine, Golestan Hospital, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
5 Department of Medical Physics and Biomedical Engineering, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran
6 Department of Radiology, Golestan Medical Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
Abstract
Keywords
INTRODUCTION
Thermal therapy is a medical option for cancer treatment which increases the patient’s body temperature with the aim of boosting the immune system, inhibiting tumour growth, and increasing the sensitivity to other anticancer treatments [1-11]. Killing cancerous cells by increasing the temperature up to 45 °C inside the tumours and saving healthy tissue with a maximum temperature of 42.5 ˚C is a crucial subject [12-15]. Hence, monitoring the temperature distribution during thermal therapy is essential. Electrical impedance tomography (EIT) is a noninvasive type of medical imaging in which the electrical properties of a part of the body are inferred from surface electrode measurements and used to form a tomographic image of that part or differentiate between normal and suspected abnormal tissue within an organ. For medical purposes, the small difference in the electrical properties of tissues is a crucial limitation to image tissues based on signal intensity.
It is well known that temperature variation in tissues causes a change in their electrical impedance [15-21]. However, the temperature dependency of tissue impedance is a limiting factor for EIT, but it can be used to monitor the temperature of tissues [22-26]. One limitation of EIT for mapping the temperature distribution is the low-temperature resolution [27-29]. Therefore, it would be interesting to intensify the differences in the temperature or electrical impedance between different tissues to distinguish them from each other.
Loading of deionised (DI) water with gold nanoparticles (GNPs) can reduce the electrical impedance of the environment [30-32]. It was reported that increasing the temperature of thin films loaded with GNPs reduces its electrical impedance [17-18]. Unlike DI water, thin films, and other non-living media, biological tissues contain various molecules and cellular structures that can have a crucial effect on their thermal and electrical responses. Therefore, a practical biological phantom consisting of materials with low impedance differences is more suitable for studying EIT systems. Bera and Nagaraju (2011-2012) recommended that chicken tissues are suitable for investigating the electrical properties of tissues [33-35].
Recently, nanoparticles (NPs) have attracted significant interest for medical applications owing to their retention and penetration into biologically targeted tissues and cells. These properties of NPs have the potential to improve treatment outcomes or enhance the quality of medical imaging and other diagnostic techniques. Among the investigated NPs, gold NPs (GNPs) are of interest because of their nontoxicity and acceptable biocompatibility. It has been reported that the loading of biological tissues with GNPs alters their electrical properties [36-37]. Therefore, studying the electrical behaviour of GNPs in a tissue-like environment at different temperatures has the potential to be a basic step toward increasing the signal intensity and/or improving the sensitivity of EIT to obtain a better temperature resolution. To the best of our knowledge, there are no quantitative data on the electrical properties of different tissues (such as soft tissue and muscle) when loaded with GNPs and maintained at different temperatures. In this study, GNPs were synthesised and loaded into two sets of phantoms made from chicken muscle and fat tissues. Therefore, the aim of this study was to investigate the electrical properties of tissues with and without GNPs at different temperatures and low frequency ranges using the 4-electrode method.
MATERIAL AND METHODS
Gold nanoparticles (GNPs) synthesis
In this study, a seed growth method was used to synthesise GNPs. First 2.2 mM sodium citrate was solved in Milli-Q water (150 mL) and then heated in a 250 mL three-necked round-bottomed flask for 15 min on a magnetic hotplate (Heidolph, Germany) with strong stirring (300 rpm). During this process, a condenser was used to avoid the evaporation of the boiling solution. Immediately after boiling, 1 ml of HAuCl4 (25 mM) was injected into the boiling solution. The yellow colour of the solution changed from bluish-gray to soft pink after approximately 10 min. The production of seeds can be confirmed by fixation of the colour of the solution, and at this time, the produced GNPs (⁓10 nm) were well suspended in H2O because of the coating with negative citrate ions. The synthesised gold seeds in the same vessel were immediately cooled to decrease the temperature to 90 °C. This caused the nucleation of new seeds to be quenched, even in the presence of an extra number of precursors. To reach the considered size of the GNPs, 1 mL of HAuCl4 (25 mM) and 1 mL of sodium citrate (60 mM) were sequentially injected into the seed solution with a delay time of approximately 2 min. The solution containing the synthesised GNPs was allowed to cool to room temperature. To maintain the stability of the final solution until the measurement step, it was stored in the dark in a container at 4 °C.
Characterization of the synthesized GNPs
Transmission electron microscopy (TEM; Zeiss-EM10C-100 kV) was used to characterise the size distribution and shape of the synthesised GNPs. Chemical analysis was performed based on energy dispersion spectroscopy (EDS) using scanning electron microscopy to verify the purity of the produced sample. The morphology of the GNPs was examined by scanning electron microscopy (SEM). The UV-visible absorption spectrum of the GNPs was obtained using a spectrophotometre (Nanodrop oneC; Thermo Scientific, USA). For further verification of the synthesised GNPs, the chemical structure was studied using X-ray diffraction (XRD) (Kigaku Diffractometer, Japan: scanning speed= 2 °/min, sampling density= 0.02 °, U=40 kV, I= 30 mA), and the average particle size (d) was calculated using the Scherrer equation as follows:
where k is the Scherrer constant, λ is the wavelength of the streak rays used, B is equal to the entire width at full width at half maximum (FWHM) of diffraction (in radians), and θ is Bragg’s angle.
Phantom preparation
To study the impedance behaviour of biological tissues, an appropriate tissue-like phantom is essential. Phantoms made of materials that have capacitive impedance similar to that of real biological tissues [32-34, 37] are more suitable than phantoms with only resistive impedance, such as deionised (DI) water [29-31]. It was shown that the chicken tissue phantom has both capacitive and resistive impedance and is suitable for mimicking the electrical properties of human tissue [32-33]. Therefore, in this study, two types of homemade phantoms consisting of chicken muscle paste tissue (50 mm × 20 mm × 20 mm) and chicken fat block tissue (5 mm × 20 mm × 15 mm), with and without GNPs, were prepared to study their electrical properties.
The Ethics Committee of Ahvaz Jundishapur University of Medical Sciences, Ahvaz, approved this study (Ethical approval no. IR. AJUMS. REC. 1395. 205). Chickens were collected from market and cut in small pieces; all bones were removed carefully, and the separated muscle and fat tissues were cut into small pieces. Muscle and fat tissues were washed separately with DI water at least three times to ensure cleaning of any contamination, and then the external tissue water was filtered and removed using clean blotting papers. The reported water resistivity is approximately 18 × 106 Ωcm. For natural drying, the tissues were kept open for 30 min at room temperature. After adding 2% DI water, the tissues were mixed for 2 min using a kitchen mixer grinder (2500 rpm). To ensure thorough mixing, 1% DI water was added to the mixture and rotated, as in the last step. All prepared phantoms were kept in a refrigerator and removed 30 min prior to measurement. To investigate the effects of GNPs on the electrical properties, the determined volume of synthesised GNPs (50 and 650 µL for fat block tissue and muscle paste tissue phantoms, respectively) with a concentration of 15 mg/100 mL were divided into 10 parts, and each part was injected randomly into a phantom with a syringe.
Measuring the electrical properties
Considering the structure of biological tissues, each tissue cell can be a combination of three distinct media: intracellular medium (ICM), extracellular medium (ECM), and cell membrane (CM) which contribute to its complex bioelectrical impedance. The electrical impedances of all the biological tissues consist of resistive and capacitive elements. The components of ICM and ECM include a variety of charged particles and ions that make them electrically conductive. The CM component can be considered as a non-conductive lipid layer sandwiched between two conductive protein layers that acts like a capacitor and alters the current path. The bioimpedance of tissues is not the same for different tissues due of the 3-D array of cells and the consistency of different elements and structures.
A change in the current frequency led to a change in the impedance and phase angle of the tissue which resulted from a corresponding change in its real and imaginary components. In the high-frequency range, the capacitive impedance is very low; hence, the CM element behaves like a short circuit, and the electrical impedance of the biological tissue predominantly depends on the electrical impedance of the ICM and ECM. The electrical impedance of biological tissue depends on the frequency of the current, defined as:
The real part shows the resistive impedance (Z) or resistance (R) (see. Equation (4)), and its imaginary part represents the capacitive impedance (ZC) or reactance (X) (see. Equation (5)). Angle Ɵ shows phase difference between voltage and current and the amplitude of impedance is calculated from division of voltage to current:
The electrical impedance of biological tissues can be measured using various techniques. A standard and reliable gold standard technique is the 4-electrode method [39]. The setup of the 4-electrode method used in this study to measure the electrical properties of tissues is shown in Fig. 1.
An alternating voltage was applied to a set of phantoms, and a determined ohmic resistor (R=980 Ω) was in series with each other component. The current was entered into the phantom by two electrodes at both ends, and the voltage V1 was measured by two needle-like electrodes placed at the middle of the phantom and separated at L=10 mm and L=5 mm for the muscle and fat phantom, respectively. The voltage V2 at both ends of the resistor (R) was measured and divided by its resistance to calculate the electric current, I= V2/R. The voltage of phantom V1 and its current I were used to calculate the impedance amplitude of the tissue according to equation (1). Voltage and frequency were measured with a Goodwill GDS-1072B oscilloscope, resistor R was measured with a KAISE digital multimeter SK-6111, and the current of the circuit was checked with a PHYWE ampere meter. The phase difference between the voltage and current (θ) is equal to the phase difference voltages of V1 and V2, which were measured by an oscilloscope. The specific resistive impedance (R) and specific conductance (σ) of the tissue were calculated using Equations 6 and 7, respectively.
where ρ is the specific resistance, L is the distance between electrodes that measure voltage, and S is the cross-sectional area of the sample tissue: S=B × H (cm2), where S is the cross-sectional area of the sample tissue, which is 4 × 10-4 cm2 and 3 × 10-4 cm2 for the muscle and fat phantoms, respectively, when measuring the temperature dependency of impedance. The temperature of the phantom was measured using a noncontact IR laser thermometer (TES-1326S). Each phantom was exposed to an IR electric heat lamp (2000 watt.h) at 50 cm, and its electrical impedance and temperature measurements were repeated as the temperature increased. For the impedance measurements at different frequencies, the parameters were L=10 mm, A=50 mm, B=20 mm, and H=20 mm. In this case, a sinusoidal electric current with a frequency of 1 kHz passed through the phantom.
RESULTS AND DISCUSSION
GNPs were successfully synthesised by reduction of HAuCl4 with C6H7NaO7. From Fig. 2.a, TEM images of the synthesised GNPs confirmed that the size of the nanoparticles was approximately 17-23 nm. Morphological evaluation of the GNPs showed that they had a spherical shape. The morphology of GNPs (Fig. 2.b) was also examined by SEM, and the SEM image indicated that the GNPs were well below 100 nm. The UV-visible spectrum also confirmed the size distribution of the 510 nm peak in the absorption pattern (see Fig. 3.a). The XRD measurements (Fig. 3.b) resulted in a determined network constant of 4.198 Å, and the average crystallite size calculated using the Scherrer formula was 21 nm. EDS measurement (Fig. 3.c) of the synthesized GNPs confirmed the purity of the produced GNPs. The marked picks in the spectrum are related to the atoms of the used materials; the signal from silicon (Si) is produced by the silicon substrate.
For the muscle paste and fat tissues with and without GNPs, the temperature dependency of impedance amplitude (|Z|), phase angle (θ), resistive impedance (ZR=R), capacitive impedance (ZC=X), specific resistance (ρ), and specific conductance (δ) were shown in Fig. 4 and 5 respectively. Our results showed that the impedance amplitude (|Z|), phase angle (θ), resistive impedance (ZR=R), capacitive impedance (ZC=X), and specific resistance (ρ) of the muscle and fat tissues decreased because of the increase in temperature (as shown in Figs. 4.a, 4.b, 4.c, 4.d, 4.e and Figures of 5.a, 5.b, 5.c, 5.d, 5.e). Furthermore, a greater decrease was observed for these parameters by loading tissues with GNPs compared to tissues without GNPs. An increase in the temperature of the tissues increased the specific conductance (δ), which was amplified with the injection of GNPs (as shown in Figs. 4.f and 5.f). For example, at a temperature of 36 °C, the enhancement of the specific conductance (δ) due to the presence of GNPs for the muscle and fat tissues was 8.59 × 10-2 S/m (13.2%) and 3.8 × 10-3 S/m (15.1%), respectively.
Pfeiffer et al. (2014) reported that the resistive impedance of biological tissues can be attributed to their ionic content and mobility [40]. An increase in tissue temperature is accompanied by an increase in intracellular media (ICM) and extracellular media (ECM) [16, 20, 41]. Under these conditions, ions and other charge carriers can move easily in an alternating current (AC) electric field, leading to a decrease in the resistive impedance of tissues.
The capacitive impedance at a certain frequency in a particular tissue is proportional to the displacement current [42]. The membrane structure of cells, the number of polar molecules such as proteins, and their ability to easily rotate with an external electric field play significant roles in the displacement current in tissues [19]. Changes in these parameters with temperature have been reported by Esrick and McRae (1994) [20]. Consequently, variations in these properties lead to a change in the capacitive impedance of the tissues, which is consistent with our data. It has been shown that GNPs are effective agents for the conversion of infrared (IR) radiation to heat in tissues and could be a heat generation source following exposure of tissue to IR radiation [43-45]. Therefore, heat induces oedema, and other changes depend on temperature, which intensifies in the presence of GNPs. In addition, GNPs inside the tissue are covered by ions and other charged particles [40], which could be considered as electric dipole moments, such as polar molecules. The reorientation of these dipoles with an alternating current (AC) electric field becomes more difficult as the frequency increases, and their influence on the displacement current enhancement becomes restricted. The measured viscosity for all GNPs sizes by Abdelhalim et al (2011) has shown a decrease with a temperature increase [46]. Therefore, the enhancement in the mobility and rotation of the intended dipole moments is a consequence of the increased temperature. Consequently, a greater reduction in impedance amplitude (|Z|), phase angle (θ), resistive impedance (ZR=R), capacitive impedance (ZC=X), and specific resistance (ρ), in addition to the increased specific conductance (δ) for tissues followed by injection of GNPs, could be explained. Electrical impedance tomography (EIT) is an imaging technique that maps the conductance distribution of tissue inside the body [39]. Increasing the electrical conductance of tissues in the presence of GNPs and intensifying this phenomenon at higher temperatures have the potential to increase the signal intensity of EIT and thus elevate its sensitivity. Furthermore, it has the potential to improve temperature resolution in temperature mapping of tissue using the EIT method.
CONCLUSION
Our results showed that the impedance amplitude (|Z|), phase angle (θ), resistive impedance (ZR=R), capacitive impedance (ZC=X), and specific resistance (ρ) of the muscle and fat tissues decreased owing to an increase in temperature or frequency. Furthermore, a greater decrease was observed when tissues were loaded with GNPs compared to tissues without GNPs. An increase in temperature increases the specific conductance (δ), and this effect is intensified, followed by the injection of GNPs. Loading of tissues with GNPs has the potential to increase EIT signal intensity and improve the temperature resolution in temperature mapping of tissue using the EIT method.
ACKNOWLEDGEMENT
This study was extracted from a PhD thesis by Mohsen Ostevari and supported financially by the Research Affairs of Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran (Grant No. U-95376).
CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this manuscript.