Zhiyang Zhang‡
ab,
Rui Su‡bc,
Fei Han‡d,
Zhiqiang Zhengb,
Yuan Liub,
Xiaomeng Zhoud,
Qingsong Lid,
Xinyun Zhaie,
Jun Wuf,
Xiaohua Pang,
Haobo Panb,
Peizhi Guo
c,
Zhaoyang Li
a,
Zhiyuan Liu
*d and
Xiaoli Zhao
*b
aSchool of Materials Science and Engineering, Tianjin Key Laboratory of Composite and Functional Materials, Tianjin University, Tianjin 300350, PR China
bResearch Center for Human Tissue and Organs Degeneration, Institute of Biomedicine and Biotechnology, Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences, Shenzhen 518055, PR China. E-mail: zhaoxltju@gmail.com
cInstitute of Materials for Energy and Environment, State Key Laboratory of Bio-fibers and Eco-textiles, School of Materials Science and Engineering, Qingdao University, Qingdao 266071, PR China
dNeural Engineering Center, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, PR China. E-mail: zy.liu1@siat.ac.cn
eCenter for Rare Earth and Inorganic Functional Materials, School of Materials Science and Engineering, National Institute for Advanced Materials, Nankai University, Tianjin 300350, PR China
fShenzhen Key Laboratory for Innovative Technology in Orthopaedic Trauma, The University of Hong Kong-Shenzhen Hospital, Shenzhen 518053, PR China
gSouthern Medical University, Shenzhen Bao'an People's Hospital, Dept Orthoped & Traumatol, Shenzhen 518101, PR China
First published on 25th January 2022
Wound infection is a common clinical problem. Traditional detection methods can not provide infection early warning information in time. With the development of flexible electronics, flexible wearable devices have been widely used in the field of intelligent monitoring. Here, we describe the development of a soft wound infection monitoring system with pH sensors and temperature sensors. The measurement range of pH was 4–10, the fitting accuracy was 99.8%, and the response time was less than 6 s. The temperature sensor array showed good accuracy and short response times in the range of 30 °C to 40 °C. A series of in vitro tests and the use of a rat model of Staphylococcus aureus infection confirmed that this flexible detection system can monitor the pH and temperature changes occurring in the early stage of infection, which provides an effective reference for clinical application.
Early detection of pathogenic infections is often difficult before clinical signs and symptoms appear.5 Currently, clinicians mainly rely on conventional culture-based approaches using wound swabs to identify infective bacterial species. Final results of these tests usually take 3–5 days, so this approach does not provide timely guidance.6 Dressings are usually needed to cover the wound to protect wounds from pathogenic infections and to provide a humid environment that accelerates healing. The dressings are changed every 2–3 days, during which the wound condition was difficult to directly observe. Therefore, a pressing need exists to develop dressings that can monitor wound conditions and predict infections. The basic requirement for this smart dressing is the conformability to the wound, because a wound's surface is always cambered, and the smart dressing needs to firstly adhere well to it for detecting desired signals. Many phenomena are caused by infection, such as changes in metabolites, protein signals, pH, and temperature. Specific biomarker of bacteria has relatively high specificity, but infection is often caused by multiple bacteria and complicated factors. Previous studies have shown that pH and temperature, as common physiological parameters, are closely related to the changes of wound.7,8 For skin pH, normal skin has an acidic pH, ranging between 4 and 6. Since body's internal environment maintains a neutral, the pH will rise when a wound is formed. Once infection occurs, the proliferation of microorganisms will induce the pH to rise to even slightly alkaline. At the same time, the temperature of the wound will also rise due to the inflammatory reaction and the relaxation of its own blood vessels, and the infection will aggravate this trend.9–11 Therefore, the changes of wound pH and temperature are widely studied parameters to determine the wound state. They are the simplest and most effective parameters to design the sensor. Previous studies have used optical signals to monitor skin surface pH and temperature, but the closed environment of the wound limits the optical signal, so the monitoring method based on electrical signals is more suitable for the wound environment.12–15 A single monitoring method is easy to be disturbed by the external environment. For example, the temperature change of the environment will affect the human body temperature to a certain extent. Therefore, we monitor two physiological parameters at the same time to obtain more accurate monitoring data.
Previously, based on the developed flexible electronic, advanced dressings were developed to monitor the state of acute and chronic wounds.16–18 Some of these dressings use commercial temperature sensor chips, combined with intelligent bandages, to monitor and treat infected wounds on demand, thereby confirming the possibility of combining hard electronic devices with flexible circuits.19,20 At the same time, new chemical sensors have been developed that can be integrated with flexible circuits, and electrochemical methods are now widely used.21,22 Chemical sensors, such as pH sensors based on potentiometry and glucose sensors based on cyclic voltammetry, can output signals through electrical signals, thereby providing convenient data transmission for monitoring infection processes.23–26 In reported studies, polyethylene terephthalate (PET) and polyimide (PI) serve as flexible substrates, but lacking softness limits their adhesion to wounds.21,27 Therefore, the substrate needs a stretchable material such as polydimethylsiloxane (PDMS), polyurethane, styrene ethylene butylene styrene (SEBS). The stretchability of the material can ensure that the sensor can always have a good relatively static contact with the wound when the wound deforms. SEBS shows the best stretchability and softness, and has been used as wearable electronics,28,29 however, it has not been applied in wound monitoring.
Herein, we propose a new hybrid flexible dressing for early detection of wound infection by employing thin SEBS elastomer film as substrates combined with polyaniline (PANI) based pH sensor and commercial temperature sensor (Fig. 1A). The temperature and pH values can be recorded to grasp the physiological changes of the wound in real-time. PANI, as a good conductive polymer, shows different open circuit potential (OCP) in solutions with different pH values, which is caused by its protonation reaction in acidic solutions. And the pH sensor integrated on SEBS substrate has the characteristics of softness and stretchability. The measurement range was pH 4–10, fitting accuracy was 99.8%, and the response time was less than 6 s. The temperature sensor showed good accuracy and short response times in the range of 30 to 40 °C. Fig. 1B shows both sensors converted information into electrical signals, pH sensor is based on potential, while temperature sensor is based on Negative Temperature Coefficient (NTC). A series of in vitro tests and the use of a rat model of Staphylococcus aureus infection confirmed that this flexible detection system can monitor the pH and temperature changes occurring in the early stage of infection (Fig. 1C), which provides an effective reference for clinical application.
A cutaneous wound was created on the back of each rat, and the animals were divided into two groups: uninfected and infected. The rats were anesthetized with isoflurane, and the dorsal hair was shaved to expose the skin. After sterilization of the shaved area with 75% alcohol and iodine, a square full-thickness skin wound (2 cm × 2 cm) was made with scissors. The fascia part of the lower dermis was removed with sterilized forceps, and the fascia was destroyed with scissors. The wounds were washed with saline, blotted with sterile gauze, and then inoculated with S. aureus suspension (50 μL, 1 × 109 CFU) to induce the bacterial infection. Rats that received physiological saline instead of bacterial suspension served as controls. The flexible dressing with sensors was placed on the wound sites to monitor the wound status and then covered with transparent 3M dressing. The pH and temperature were monitored at 0, 8, 24, 32, 54, and 78 h.
FT-IR spectroscopy was used to confirm the successful synthesis of PANI, as shown in Fig. 2A. The prominent peaks of the PANI-based electrode at 1548, 1499, and 806 cm−1 corresponded to CC stretching of the quinoid ring, C
C stretching of the benzenoid ring, C–N stretching of the benzenoid unit, and C–H out-of-plane bending, indicating the successful preparation of polyaniline functional films. The SEM image of the Ag/AgCl reference electrode is shown in Fig. 2B. No cracks or holes were evident, indicating strong adhesion of the Au and Ag/AgCl electrodes. Observation of the interface between PANI and the Au electrodes by SEM revealed an obvious boundary. The flaky micro-nano structure of the PANI electrode can be the main reason for pH sensor capability because of its potential stability to facilitate redox reactions. SEBS substrate could be stretched to 100% (Fig. 2C), and microcracks were present on gold layer by SEM imaging (Fig. 2D). This morphology is formed spontaneously under specific parameters. The gold electrode can be stretched by this kind of microcrack. The microcracked gold film can share the stress under tension by crack expansion, that is, in the tensile process, although the crack area of the gold electrode expands, there is always a part to maintain the connection. In this way, the extensibility of the electrode is realized.16 The stretchability was further confirmed by the sustained conductivity when the tensile resistance exceeds 100% (Fig. 2E).
The flexible materials used in wound monitoring was compared in Table S1 and Fig. S1.† Previous studies focused on demonstrating the function of the device, however, the softness of the material itself and the adhesion of the wound are also important.24,30,31 The conformability of these materials were not satisfactory. Flexural stiffness (D) of the sensor is defined as: D = Eh3/12 (1 − v2), where E is the elastic modulus, h is the thickness of the sensor and v is Poisson's ratio. Higher E or h will lead lager D, resulting in a poor conformability for the devices. For the reported typical dressings, the flexural stiffness is too high to conform on skin, and need to be further improved to adhere better to the skin.35 The stretchable property of the SEBS substrate also facilitates its application on the skin of joints.36,37 Comparison with a non-stretchable PI substrate on the finger joint shows that the SEBS film can deform to fit maximum movements of the finger joint to 105°, which is close to the maximum angle of a finger bend. By contrast, non-stretchable PI film can only rely on the stretchability of the skin for finger joint bends of 45°, due to the ductility of human skin (Fig. 2G and H). This result suggested that stretchability is important for the application of flexible materials in wearable devices.
The stability of pH sensor was investigated under 300 cycles of tension as shown in Fig. S3,† the corresponding SEM image is shown in Fig. S4.† After 300 times of 50% stretching, the potential changes slightly, and the error of pH is less than 0.07 (Fig. 3B). The micro morphology of the sensor after stretching has not changed greatly, due to the recovery of microstructure, the performance of the sensor also changes little. This property insured its application in an actual wound. The OCP signals of the pH sensor were studied at different pH values in the range of 4–10, and the stability was explored (Fig. 3C). PANI undergoes protonation or deprotonation reaction in acidic or alkaline solution to present different OCP value.24,32 The OCP signal showed a linear relationship with the pH value, and the standard calibration line of the pH sensor was obtained by linear fitting (Fig. 3D). The sensitivity of the potentiometric pH sensor can be obtained by the slope of the linear regression, according to the Nernst equation: E = E0 − (2.303RT/F) pH = E0 − 0.05916 pH, where R is the gas constant, T is the temperature, E0 is the standard electrode potential, and F is Faraday's constant. Based on Nernst behavior, the theoretical maximum sensitivity is −59 mV pH−1 at room temperature. The resultant calibration curve is linear, with a slope of −65.9 mV pH−1 (R2 = 99.8%) in a wide pH range of pH 4–10. This sensitivity value of the pH sensor is close to the theoretical pH sensitivity based on Nernstian behavior and is similar to other previous reports based on PANI electrodes. And we compared it with other similar studies, and the results are shown in Table S2.† The pH of normal skin is weakly acidic, and the pH of the wound is neutral due to the exposure of plasma. When infection occurs, the pH will raise to alkaline, ranging between 7 and 9.25 These results indicate that the pH sensor could cover the range of pH changes in different stages of the wound, and it was well matched with the theoretical accuracy.
The response time of the pH sensor was measured by increasing the pH (Fig. 3E). The OCP signal was subsequently changed and reached 90% of its steady state value within 6 s. The improvement in the response time depends on the uniformity of the microstructure; the uniform distribution of polyaniline makes the response instant and rapid. The fast response time of the pH sensor indicates that this sensor can be applied in situations where dynamic pH changes occur.
The long-term stability of the pH sensor was further studied by measuring the OCP response of the pH sensor after immersing in PBS for 12 hours (Fig. 3F). The pH sensor maintained the OCP signals, resulting in a potential drift within 2.0 mV h−1, which corresponds to a 0.036 error in the pH value after an hour of continuous measurement. The stability of the pH sensor could still be guaranteed even after a long-term measurement-a feature that is very important in the actual application of wound monitoring, and this sensor met this requirement. Modification of the gold electrode allowed successful preparation of the pH sensor on flexible stretchable substrate, and the sensor showed good sensitivity and stability.
The accuracy, responsiveness, and durability of the PDMS-encapsulated commercial temperature sensor were checked by first placing the sensor inside a Ziplock bag before inserting it into a temperature-controlled water bath with a temperature ranging from 30 to 40 °C, which covers body temperature range. The mean displayed temperature values recorded after 1 min of monitoring indicated that the temperature sensor had excellent accuracy, with an absolute deviation of < ±0.4 °C (Fig. 4B). Fig. 4C shows that the chip has good stability, which is necessary for real-time monitoring. As shown by the temperature–time curve (Fig. 4D), the temperature sensor displayed a response time of < 30 s and long stability. The performance of the TMP112A temperature sensor could therefore satisfy the needs of real-time and in situ wounding monitoring. Coupling of this sensor with the SEBS substrate Au-PANI pH sensors in a specific geometric distribution would therefore allow simultaneous monitoring of the wound pH and temperature. In summary, a commercial temperature sensor with a small volume and high sensitivity was selected and integrated onto an FPCB by a simple circuit connection.
The flexible detection system was verified in vitro to monitor the changes in pH and temperature, and its performance in actual wound monitoring was then investigated. The rat model of infection was established by inoculating S. aureus into cutaneous wounds. S. aureus is one of the most common bacterial pathogens found in infected wounds.38,39 The rats were divided into two groups: infected and uninfected. Photographs of the wounds at 0, 8, 24, 32, 54, and 78 h are shown in Fig. 5B. Purulent discharge was clearly visible in the infected group on the third day, accompanied by wound redness and swelling, indicating that the infection had induced inflammation and tissue necrosis.39,40 H&E staining of the skin biopsies of the wound sites showed inflammation infiltration in the infected group (Fig. 5C). At 8 and 30 h after inoculation, many more immune cells had accumulated in the cutaneous wounds in the infected group than in the uninfected group. Neutrophils could usually be found in S. aureus infected wounds as innate immune cells in the early inflammatory response.39,41 These results demonstrated the establishment of an infected wound.
The image of the sensor attached to the wound shows that the flexible dressing fit well with the wound (Fig. 5D and E) and was able to monitor the changes in pH (Fig. 5F) and temperature (Fig. 5G) in the wound. The infection induced a rise in pH and temperature at the wound sites as early as 8 h post operation. When the wound was just created, it had a pH around 7–7.2 and a body surface temperature of about 32.4 °C. Eight hours later, the pH of the infected wound had increased to pH 7.7 and the temperature had risen to 34.1 °C. A comparison of infrared thermometers revealed close results (Fig. 5H). By contrast, the pH and temperature of the uninfected wound surface increased only slightly. In the following 3 days, although the pH and temperature of the two wound surfaces fluctuated, the pH and temperature were always higher for the infected wound than for the uninfected one, due to the proliferation of bacteria and the response of the rat's own immune system.28 In the animal model experiment, the customized data processing module is integrated with the sensor for data acquisition. Due to the limitations of the size and energy of the data processing module, continuous wound monitoring was not performed in this study. However, with the development of smaller back-end modules and the further optimization of sensor materials, long-term real-time monitoring of wounds will be possible.
As mentioned earlier, the pH value of the wound is an important biomarker that can provide important information about the condition of the wound. The pH of the surface of normal skin is acidic, but injury can lead to the exposure of the underlying tissue and change the acidic environment of the site. When infection occurs, the changes caused by bacterial metabolism and proliferation will make the pH of wound reach alkaline state.42 Other articles have mentioned temperature as a parameter influenced by many factors, such as blood flow, bacterial infection, and oxygenation. Inflammation, new tissue formation, and remodeling during wound healing can also raise the temperature.19,43 In this system, the temperature chip was integrated on the FPCB, since the combination of hard chip and soft substrate is difficult to realize, however, this should be worked on in the future. Comparison of with the changes in pH and temperature confirmed that the sensor fabricated here can detect these changes at the corresponding time; that is, the change can be used as an early warning that the wound is infected.44 In clinic, the wounds usually are more complex, and many factors wound influence the result, such at body fluid, ambient temperature. In addition, the combination of hard chip and soft substrate is a future attempt.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ra08375a |
‡ These authors contributed equally to this work. |
This journal is © The Royal Society of Chemistry 2022 |