Polymer Hydrogel‐Based Multifunctional Theranostics for Managing Diabetic Wounds

Chronic, non‐healing wounds pose significant challenges for public health, particularly in the context of diabetes, and carry significant economic consequences. This article introduces a new solution in the form of a wireless theranostic patch, developed to meet the critical need for real‐time monitoring and targeted treatments to facilitate optimal healing. The patch incorporates advanced materials that are both multifunctional and electro‐responsive, leveraging a sophisticated blend of smart hydrogels and wearable bioelectronics to support diabetic wound management with unparalleled efficacy. With electro‐responsive multifunctional polymer hydrogels at its core, the patch delivers a stretchable, antimicrobial, and moist environment for the wound, with added benefits such as conductivity and visibility. The materials also allow for continuous and autonomous monitoring of glucose and pH levels, providing precise and personalized treatments like insulin delivery via iontophoresis and electrical stimulation. Animal models have demonstrated that this integrated system is highly adaptable, effectively promoting wound closure and healing. Overall, the wireless theranostic system offers an exciting prospect for personalized healthcare solutions, adopting a patient‐centric approach that prioritizes real‐time, targeted care for chronic wounds. Its incorporation of advanced materials and electro‐controlled treatments paves the way for new and innovative approaches to wound management.


Introduction
The persistent challenge of chronic wounds that do not heal has been a significant burden in healthcare, particularly amongst those with diabetes.This ailment has resulted in substantial DOI: 10.1002/adfm.202315564financial and healthcare costs in the global healthcare system, ranging from $9.9 to $35.8 billion annually. [1]Such wounds, frequently found in diabetic patients, have specific characteristics, including protracted inflammation, hindered cell proliferation, obstructed angiogenesis, heightened susceptibility to infections, [2][3][4] and altered extracellular matrix, with amputation being required in approximately 20% of these cases. [5,6]Individuals with diabetes are especially vulnerable to these issues during the healing process, and conventional wound treatments like gauzes, sponges, and hydrogels are limited in their effectiveness.][15][16][17] Among various smart wound patches, those featuring stimuli-responsive drug-loaded hydrogels have achieved the most success, due to their exceptional biocompatibility, high drug load capability, moisture retention, and controlled release of drugs. [18]Despite these technological advances, these wound patches fail to offer real-time insights into the wound microenvironment or the healing process, which can hinder effective treatment or lead to inaccurate conclusions.Furthermore, the drug release rates of endogenous smart wound dressing can be significantly influenced by the encapsulating matrices, making it difficult to control the dosage at the early stages of infection. [19]To address these challenges and improve personalized wound care management, feedback therapeutic systems that combine sensing capabilities with precise drug delivery are needed.This would enable doctors to receive real-time data on the wound microenvironment and tailor treatments, accordingly, resulting in more effective and efficient wound care.
An advanced theranostic wound management system contains wearable biosensors, wireless communication technology, algorithms for data analysis and feedback control, and stimulusresponsive therapeutics systems. [12]These patches utilize in situ monitoring of physiological signals, including temperature, pH, enzymes, and glucose in wound beds, to effectively diagnose wound healing status and warn of abnormal healing events.Light, temperature, and electrical signals are the most commonly used exogenous stimuli to control drug delivery.However, a major concern of light-responsive drug delivery system (DDS) is the limited penetration of light, often resulting in uneven administration. [20,21]The temperature-responsive DDSs are prone to inaccurate release caused by the wound temperature, and the high temperature often lead to discomfort. [22]Exogenous electrical stimulation, widely employed in clinics can activate the cell proliferation and migration [23] and easily integrates to control drug delivery.A highly integrated wound theranostic system that records temperature, pH, and uric acid (UA) levels while providing on-demand electrically controlled cefazolin delivery for chronic wound healing was developed by Xu et al. [24] Its efficiency has been observed in the in vivo trials; however its drug loading rate of conducive polymer films remains limited.Detailed comparision about the wound theranostic systems is listed in Table S1 (Supporting Information).Most of these patches still require manual intervention, as they lack algorithms for data analysis and feedback control between the diagnosis and therapy units.Moreover, overuse of antibiotics is prevalent in existing theranostic wound dressings due to their single function of the carrier matrix (only responsive to drug delivery).Thus, there is a need for a wound theranostic system that not only spontaneously performs dynamic wound management but also enhances therapy efficiency through integration with stimulus-responsive multifunctional materials, providing optimal healing conditions, reducing the risk of drug resistance.
In this article, we present a wireless and wearable system that is fully integrated for the purpose of diabetic wound management.The multifunctional conductive polymer hydrogel (MFCPH) is at the core of the system's design, possessing various properties such as high drug loading rates, conductivity, and broad-spectrum antimicrobial capabilities.Meticulous development of MFCPH leads to an optimal healing environment for diabetic wounds, mitigating secondary infections and reducing reliance on antibiotics.The intricately designed system combines smart hydrogels with wearable bioelectronics, allowing in situ monitoring of wound biomarkers and autonomous electro-controlled drug delivery.This approach enables real-time monitoring of pH and glucose levels, offering insights into bacterial infections and skin regeneration progresses.The system's on-demand therapy is facilitated through the integration of insulin release by iontophoresis and electrical stimulation, enhancing treatment delivery precision.The integration of wearable electronics with a tailored smartphone application emphasizes the system's autonomous operation, eliminating the need for external intervention.The development of advanced materials plays a pivotal role in elevating the efficiency and autonomy of therapeutic interventions in diabetic wound management.

Design of the Wireless Theranostic Wound Management System
A wireless theranostic wound management system has been developed for the closed-loop administration of diabetic wounds, especially for the diabetic wound.The system primarily consists of a customized smartphone App, a wearable electronic device, and a Thera-patch (Figure 1a,b).An attractive feature of this innovative wound theranostic system lies in the integration of the smart multifunctional hydrogel with wearable bioelectronics.A dual electrochemical biosensing platform was developed for real-time and simultaneous detection of wound pH and glucose levels.The smart multifunctional hydrogel, referred to as MFCPH, is made by incorporating polydopamine-doped polypyrrole (PDA-PPy) nanofibrils into polyacrylamide (PAM) network based on an in situ formation process.This structure and the polycationic backbone of PDA-PPy nanofibrils endow the hydrogel with high drug loading rate, light transmittance, skin adhesion, conductivity, and broad-spectrum antimicrobial properties (Figure 1c).Loaded with insulin, the MFCPH, in conjunction with a pair of iontophoretic electrodes, was utilized for on-demand drug delivery and electrical stimulation.The adhesive design of Therapatch and the highly flexible nature of polyethylene terephthalate (PET) ensure excellent resilience against undesirable physical deformations (Figure 1d).
A closed-loop wound administration strategy that leverages real-time wound monitoring and on-demand treatment is proposed to promote the wound healing (Figure 1e).Detected pH and glucose levels are collected and processed by the wearable electronic device, then wirelessly transmitted to the smartphone App via Bluetooth.The customized App receives the sensing data and analyzes the status of the chronic wound.Once the glucose level exceeds its threshold, the App sends an activation command to the electronic device, which then triggers insulin delivery via iontophoresis.This process lowers the glucose level and promoting wound healing.The circuit schematic, printed circuit board (PCB) layout and hardware program block diagram were designed, and optical images of PCB were obtained (Figures S1-S4, Supporting Information).The wearable electronics, fabricated on a flexible substrate, features a compact (41 × 31 mm) and lightweight (3.2 g) electronic device.The analytical performance of the wearable electronic device was Figure 1.A wireless theranostic system for diabetic wound management.a) Photograph of the fully integrated wireless diabetic wound theranostic system.b) Schematic illustration of wireless theranostic wound management system, including a customized smartphone App, a wearable electronic device, and a Thera-patch.The Thera-patch continuously monitors the wound state and executes on-demand treatment via combining iontophoresis with MFCPH.The wearable electronic device collects the sensing data and receives App commands to trigger drug delivery.Smartphone App receives the sensing data, analyzes the diabetic wound state, and sends drug delivery command.c) The closed-loop wound theranostic administration strategy of the management system for real-time wound monitoring and on-demand treatment.d) Thera-patch application steps: (i) removal of the sealing membrane, and (ii) placement of the Thara-patch on the wound of diabetic rat.Mechanical performance of the Thera-patch, including (iii) bending, and (iv) twisting.d) Block diagram of wireless wound theranostic system.The command and data flow of the system for real-time wound monitoring and on-demand drug delivery.
compared with that of the CHI660E commercial electrochemical workstation (Figures S5 and S6, Supporting Information).Linear regression results for the pH and glucose detection showed that the wearable electronic device has similar detection performance to the CHI660E.Moreover, the iontophoresis current of the wear-able electronic device remained unaffected by load changes, ensuring reliable and consistent drug delivery (Figure S7a, Supporting Information).Crucially, the drug delivery process can be directly controlled through the analytical procedure (Figure S7b, Supporting Information).

Fabrication and Characterization of the Thera-Patch
The integrated Thera-patch, illustrated in Figure 2a, comprises a flexible electrode array layer, an impermeable ring, and a drugloaded MFCPH.The assembly process is depicted in Figure 2b.The flexible electrode array is patterned on the PET substrate with detailed fabrication processes provided in Figure S8 (Supporting Information).The prepared sensing electrodes were tested in a potassium ferricyanide solution, where an obvious redox peak observed during cyclic voltammetry (CV) scanning demonstrated good performance in electrochemical analysis (Figure S9, Supporting Information).An impermeable ring is employed to insulate the iontophoresis electrodes and ensure adhesion to skin.The drug is stored within the hydrogel, directly interfacing with the wound.A pair of MFCPH are assembled on the iontophoresis electrodes for electro-controlled drug delivery.
pH and glucose were chosen as wound healing indicators due to their essential roles in various physiological processes and the accumulation of glucose at the wound site can cause impairments in growth factor production and bacterial infection. [25]he deprotonation of H + at the polyaniline (PANI) functionalized working electrode was utilized for pH detection, with quantification achieved by the open-circuit potential. [26]A layer of PANI was deposited on the working electrode (Figure 2d), while the Ag/AgCl reference electrode was modified with an ion-exchange membrane.Glucose detection employed chronoamperometry method, using a three-electrode system including an Au working electrode, a platinum counter electrode and an Ag/AgCl reference electrode (Figure 2c).The working electrode for glucose sensing was modified with dendrimers that contain the electron mediator, ferrocenecored poly (ethylenimine) [Fc-PEI] and glucose oxidase (GOx).The amperometric electrochemical response of glucose can be observed by the redox activity of the Fc-PEI following second-generation glucose detection strategy. [27]The modification process was observed using a scanning electron microscope (SEM).The surface of working electrode was deposited with AuNPs layer by magnetron sputtering (Figure S10a, Supporting Information).After electrodeposition of PANI onto this AuNPs layer, a fibrous morphology was observed (Figure 2e).For the glucose detection, the Au working electrode was covered by a film-like substance after conjugating electron mediator (Fc-PEI) and GOx (Figure S10b, Supporting Information), and the surface exhibited a 3D pore structure after the addition of PU membrane (Figure 2f).
Traditional wound treatment methods involving nonfunctional dry patches have now been replaced with responsive moist patches. [28]To achieve this, MFCPH was developed by incorporating conductive polymer PDA-PPy nanofibrils in the PAM network (Figure 2g).PDA-PPy nanoparticles (NPs) were synthesized initially with a particle size of 900 nm (Figure 2i) and a Zeta potential of 38.5 mV, measured using a Zeta potentiometer (Figure 2h).These hydrophilic nanofibers, formed by the nanoparticles in situ, were combined with the hydrogel matrix to create nanonets, allowing for more uniformly electrical stimulation, drug delivery, and the passage of visible light (Figure S11, Supporting Information).With extended aging days, the distribution of the conductive network within the hydrogel became denser, ultimately leading to the PDA-PPy nanofibers being fused with the hydrogel matrix without any phase separation (Figure 2j).

In Vitro Evaluation of the Biosensors for Wound Monitoring
The sensitivity, selectivity, reproducibility, and long-term stability of the pH sensor were evaluated.Deprotonation of H + was detected through potentiometry using the pH sensor (Figure 3a).
Regarding sensitivity, the measured potential exhibited concentration-dependent out-step responses as the pH increasing from 9 to 3 and then back to 9, effectively covering the relevant pH range of diabetic wounds (between 7 and 9) in diabetic patients (Figure 3b). [12]The measured potential is linear to pH value with a sensitivity of 62 mV per decade and a correlation coefficient of 0.9998 (Figure 3c).The selectivity of the pH sensor was investigated via the introduction of various positively charged ions (Figure 3d).These interference ions including 1 mm NH 4 + , 1 mm Mg 2+ , 1 mm Ca 2+ , 10 mm K + , and 20 mm Na + based on physiologically concentrations were subsequently added to McIlvaine's buffer at pH 6.The pH sensor barely showed any response to these interferers, demonstrating its good specificity.The measured potential of pH sensor stabilized within 90%−110% of its initial value for 12 h at pH 7, demonstrating its long-term stability (Figure 3e; Figure S12a, Supporting Information).The relative standard deviations (RSD) of the potential detected by three pH sensors for 5 times were 2.17%, 2.01%, and 2.45%, respectively.The RSD difference measured by three pH sensors was 4.68%.It well demonstrated the pH sensors had good reproducibility (Figure 3f).
Glucose concentration was measured using chronoamperometry method at a potential of 0.4 V.The amperometric signals detected by glucose sensor increased step-wise along with the increasing of glucose concentration (Figure 3h).The responsive range of the glucose sensor is in 0-22 mm, which effectively covered the relevant glucose concentration of diabetic wound (0.6-5.9 mm for chronic wounds). [29]A slower increase of current response in 16-22 mm can be observed, possibly due to the limited amounts of enzyme and electron mediator modified on the electrode.The amperometric current of glucose sensor is linear to the glucose concentration ranging from 0 to 14 mm with a sensitivity of 0.48 μA mm −1 and a correlation coefficient of 0.997 (Figure 3i).The calculated detection limit of the glucose sensor was 0.097 mm.The presence of electroactive molecules in wound exudate (such as lactic acid (LA), UA) has potential interference to glucose detection.As shown in Figure 3j, the addition of LA, cholesterol, UA, and albumin showed a negligible effect on the amperometric current, attributed to the specificity of glucose oxidase and the cross-linked electron mediators.These mediators facilitate electron transfer between the enzyme active site and the electrode, and reduce the excitation voltage required for glu-cose detection, thereby preventing redox reactions with interfering substances. [30,31]The long-term stability of the glucose sensor to 5 mm glucose over 24 h was investigated, as shown in Figure 3k.The amperometric current initially decreased in the first 4 h, subsequently remained stable, owing to the gradual stabilization of the enzyme layer.The sensor exhibits high retention of the amperometric current signal, remaining at approximately 90% of the original intensity for 2 h measurement (Figure S12a, Supporting Information).It indicated that the glucose sensor may require slight recalibration for long-term monitoring.For the reproducibility, the glucose sensors exhibited RSDs of 3.59%, 5.13%, and 3.81% for repeated measurements on the same electrode, and an RSD of 5.08% across different electrodes (Figure 3l).

In Vitro Characterization of the Electro-Controlled Drug Delivery
Various stimuli-responsive conductive hydrogels have been proposed for wound treatment (Table S2, Supporting information).However, the majority of conductive hydrogels exhibit black coloration, thereby impeding visual wound diagnosis -a pivotal clinical diagnostic tool.There is still a pressing need for a stimuliresponsive conductive hydrogel that can accurately and harmlessly deliver treatment on demand, while considering properties such as anti-microbial, adhesion to provide an optical healing environment.In the on-demand drug delivery module, a constant current is applied to perform synergistic treatment of diabetic wound through chemotherapy via electro-controlled release of insulin and physiotherapy via electrical stimulation (Figure 4a).The drug delivery module of Thera-patch mainly consists of two Ag/AgCl electrodes for iontophoresis and corresponding MFCPH for drug loading.34] Herein, insulin was loaded into the MFCPH through swellingadsorption method, with a loading efficiency of up to 85.8 ± 3.7% (Figure 4b).Additionally, the hydrogel exhibits a swelling rate of approximately 943% at the swelling equilibrium state (Figure S13, Supporting Information).Under the iontophoretic current, insulin with negative charge is propelled from the gel toward the anode due to the electrophoretic flow. [35]Accompanied by active drug delivery, a mild iontophoresis current applied at the wound bed can promote the recovery of wound through electrical stimulation. [36]esides drug loading, the MFCPH featured with high stickiness and conductivity exhibits great potential for wound healing by creating a moist, visible and sterile wound environment, and performing synergistic therapy via steady delivery and electrical stimulation.The transparency of MFCPH enables visual monitoring of the wound healing process.The effect of PDA-PPy NPs on the transparency of MFCPH within the wavelength range of 400-800 nm was investigated.The transmittance of MFCPH at 660 nm decreased from 97.50% to 32.50% when the PDA-PPy nanoparticle ratio increased from 0.15 wt% to 1.2 wt% (Figure S14a,b, Supporting Information).Light transmittance decreased from 74.5% to 16.1% with an increase of hydrogel  thickness from 1 to 3 mm, with a nanoparticle ratio of 0.6 wt% (Figure S14c,d, Supporting Information).The freshly-prepared hydrogel was firstly gray-black, then gradually turned light brown, and finally became nearly transparent (Figure 4c).The transmittance of hydrogel at 660 nm increased from 10.6% to 70.0% over three days (Figure 4d).The transparency changes can be attributed to the transformation of PDA-PPy NPs into nanofibers within the hydrogel matrix, forming a conductive and light-passible networks.
The antibacterial performance of MFCPH and PAM against Gram-negative (E.coli) and Gram-positive (S. aureus) bacteria was investigated (Figure 4e,f).The inhibition zone was measured after 12 h incubation.The larger antibacterial zone was found at the MFCPH groups, indicating the excellent antibacterial effect of the MFCPH.Particularly, the antibacterial zone area of MFCPH groups increased with the ratio of PDA-PPy NPs.It is worth noting that the growth of the inhibition zone slowed down when the PDA-PPy NPs ratio exceeded 0.6 wt%.The antibacterial activity of MFCPH was mainly governed by the polycationic backbone of PDA-PPy nanofibrils, which attracted and killed bacteria. [37]When the ratio of PDA-PPy NPs exceeded 0.6 wt%, the electrostatic interaction between nanoparticles and bacterial cells weakened, [38] resulting in the slowing down the growth of the antibacterial zone.In following work, MFCPH with a nanoparticle ratio of 0.6 wt% and a thickness of 1.5 mm were selected for assembling Thera-patch.The cytocompatibility of MFCPH was assessed through CCK-8 assay and Live/Dead staining using HUVEC cells.Fluorescent images showed no significant difference in the red signal between the MFCPH and control groups (Figure 4g).Moreover, MFCPH displayed no adverse effects on the growth and proliferation of HUVEC cells co-cultured for 72 h, compared to positive control groups (Figure 4h), emphasizing its good cytocompatibility.
Next, the impact of drug loading on the intrinsic characteristics of hydrogel was tested.The MFCPH exhibited good elasticity with the strain limit of elongation decreased from 876% to 550% before and after swelling (Figure 4i).The adhesiveness of MFCPH to porcine muscle, porcine skin and PET slices was evaluated using pull-off adhesion tests (Figure 4j).The adhesive strengths of swelling and unswelling MFCPH to the porcine muscle were 2.5 and 3.1 kPa, to porcine skin 10.6 and 13.9 kPa, and to PET slices 27.9 and 37.2 kPa, respectively.The MFCPH exhibited excellent elasticity and adhesion, ensuring a close wound interface and robust electrical transmission.Hydrogel conductivity increased from 0.5 to 1.6 S m −1 within three days due to the formation of nanofibers within the hydrogel (Figure 4k), thereby possibly decreasing the power consumption and increasing the drug delivery efficiency during iontophoresis. [39,40]Moreover, the electrical field generated by the custom-designed Ag/AgCl electrodes was analyzed using a simplified 3D model by COMSOL Multiphysics. [41]The current density was at the wound bed within 800 μA cm −2 , with current flow along the direction of wound closure (Figure 4l), which may play a crucial role in the migration of lymphocytes, fibroblasts, macrophages and keratinocytes. [36]n vitro drug delivery performance of the Thera-patch combined with the wearable electronic device was tested using vertical Franz diffusion cells (Figure S15a,b, Supporting Information).The cumulative insulin dosage delivered by the Therapatch increased almost linearly with the intensity and durance of iontophoresis.A significant difference of cumulative insulin was observed after 1 h iontophoresis and the highest accumulative permeation was 53.6 μg mL −1 at 2 mA.In comparison, cumulative drug permeation of the control group (free diffusion), 0.5 mA group, and 1 mA group were 2.0, 20.4, and 39.1 μg mL −1 , respectively (Figure S15c, Supporting Information).It demonstrated that the iontophoresis effectively promoted the insulin delivery.The insulin permeation amount within every 10 min during 1 h drug delivery of Thera-patch is illustrated in Figure 4m.The drug delivery rates of 1 and 2 mA group gradually decrease due to the decreased of drug amount in the patch, [40,42] while the 0.5 mA group maintained a relatively consistent drug perme-ation rate, making it the optimal choice for subsequent drug delivery experiments (Figure 4n).Electrical currents ranging from 0.2 to 1 mA have been mostly employed in wound healing, as higher current intensity (>1 mA) often associated with increased discomfort.[43][44][45] A current intensity of 0.5 mA was selected for insulin iontophoresis and electrical stimulation.To verify the on-demand drug administration, periodic iontophoresis currents of 0.5 mA were applied.The insulin permeation showed a stepwise increase through application of "on-off" iontophoresis cycles (Figure 4o).As shown in Figure 4p, the permeation rates during "on-off" iontophoresis cycles were 0.32, 0.41, and 0.36 μg mL −1 •min, approximately 9 times higher than passive drug diffusion without iontophoresis (0.04 μg mL −1 •min), demonstrating excellent electro-controlled on-demand drug delivery capability of the Thera-patch.

In Vivo Closed-loop Wound Management of Diabetic Rats
To validate the capabilities and efficacy of the wound theranostic system, diabetic full-thickness wounds were created on diabetic rats.The systematically evaluation aimed at the system's performance in monitoring multiplexed wound-related biomarkers, executing closed-loop drug delivery, and the therapeutic effects (Figure 5a).Diabetic rats were produced by intraperitoneal injection of streptozotocin (STZ) and a 15 × 15 mm 2 square wound was created on the back of rat.The rats were divided into three groups.(1) Control group: Wounds was rinsed with saline and treated with sterile gauze; (2) Thera-patch w/o iontophoresis group: Wound was administrated with Thera-patch in free drug diffusion mode; (3) Thera-patch w/iontophoresis group: Wound was administrated with the wireless theranostic wound management system in closed-loop management mode where insulin delivery via iontophoresis (0.5 mA cm −2 ) was triggered based on detected wound glucose concentrations.The smaller wound contact area of the Thera-patch (approximately 1cm 2 ) induces reduced cutaneous sensations, attributed to a spatial summation effect. [46]he detailed experimental procedure is shown in Figure 5b.In vivo closed-loop wound management using the wireless theranostic wound management system was executed in a diabetic rat (Video S1, Supporting Information).
The pH values and the glucose concentration in diabetic wounds were continuously monitored during the whole healing process (Figure 5c-h).Prior to in vivo testing, both glucose and pH sensors were calibrated (Figure S16, Supporting Information).The pH levels of three groups were recorded in Figure 5c-e, respectively.The pH value of the Thera-patch w/iontophoresis group decreased from 8.3 to 7.2, the Thera-patch w/o iontophoresis group dropped from 8.3 to 7.4, while the control group slightly decreased from 8.4 to 7.7 on the 7th day.The pH level of control group was consistently higher than that of the Therapatch groups during the 7-day healing.According to previous reports, the excessive inflammatory response in diabetic wound can lead to higher pH value, [47] thus a quicker decrease in pH value indicates the better healing of the wound.The Thera-patch w/iontophoresis group administrated in a closed-loop management mode exhibited superior wound healing capabilities.
The glucose levels of three groups within 8 h were also measured, as shown in Figure 5f-h.The normal wound glucose In vivo performance of the wound theranostic system.a) Optical images of the wound theranostic system on the diabetic rat.The Thera-patch and the wearable electronics were adhered at the wound site for wound monitoring and iontophoresis drug delivery, with a smartphone for data and command transmission.b) Schematic illustration of the detailed experimental procedure for establishing and treating diabetic wound.pH of c) the control group, d) the Thera-patch w/o iontophoresis group and e) the Thera-patch w/iontophoresis group during the wound-healing process, (N = 3).Glucose concentration of f) the control group, g) the Thera-patch w/o iontophoresis group and h) the Thera-patch w/iontophoresis group during the wound-healing process.The red arrow indicates start of insulin iontophoresis.The WTS and BG refer to the wound theranostic system and blood glucometer.i) The serum insulin of diabetic rats treated by Thera-patch w/and w/o iontophoresis.j) The Clarke's error grid analysis illustrating the accuracy of the glucose sensor compared to the standard blood glucometer.k) The detection errors of Thera-patch compared with the commercial blood meters at corresponding time points.Heatmap plots of the fluctuations of l) pH and m) glucose.n) The average errors of glucose concentration detected by the control, the Thera-patch w/o iontophoresis and the Thera-patch w/iontophoresis group (N = 17).
range of Sprague-Dawley (SD) rats falls into 5.6-11.1 mm (red area). [48,49]The glucose level in the control group remained stable at 18.9±2.1 mm (Figure 5f).After 2 h administration in free diffusion mode, the wound glucose of Thera-patch w/o iontophoresis group gradually decreased from 20.4±1.8 to 10.1 mm, but then quickly rebounded to the original hyperglycemia states (Figure 5g) with normoglycemia lasting only 45 mins.In the Thera-patch w/iontophoresis group, the initial wound glucose concentration of diabetic wound was 23.4 mm in hyperglycemia, so iontophoresis-driven drug delivery (0.5 mA, 30 min) was activated by the wireless theranostic wound management system (Figure 5h).The wound glucose rapidly reduced to the normal range and stabilized for 2.5 h.Then, the glucose level gradually increased to hyperglycemia state at 4th h and the insulin delivery of Thera-patch was triggered again, resulting in the normalization of wound glucose.Furthermore, serum insulin was extracted from tail of rats (Figure 5i).The pharmacokinetic parameters of serum insulin were calculated [50] (Table S3, Supporting Information).Area under curve (AUC) refers to the total area under the serum insulin curve.Serum insulin levels treated by the Therapatch w/o iontophoresis group reached the maximum concentration (C max ) of 68.6 μIU ml −1 at T max of 3 h after administration, and the AUC was approximately 297.9 μIU•h ml −1 .Comparatively, in the Thera-patch w/iontophoresis group, the C max reached 116.3 μIU ml −1 at T max of 3 h, indicating a significant contribution of iontophoresis to insulin delivery.The second escalation of serum insulin further demonstrated the second successful delivery of insulin for the diabetic wound.The glucose fluctuation tendency detected by the wireless theranostic wound management system was highly consistent with the commercial blood glucose meter.85.4% of the detected glucose data located in region A (error <20%) and 12.2% located in region B (error <20%) (Figure 5j), demonstrating the accuracy of wireless theranostic wound management system.As shown in Figure 5k, the average errors of glucose level detected by the control, the Therapatch w/o iontophoresis and the Thera-patch w/iontophoresis group were 10.8±8.2%, 14.4±11.2%,and 11.9±13.9%,respectively, which satisfied the clinical requirement of error <15%.The fluctuations of pH value and glucose value were visualized using heatmap plots (Figure 5l,m).The color blocks of the Thera-patch group w/iontophoresis were the brightest, indicating its superior wound management capability.The enzyme-containing glucose sensor can successfully track glucose content of wound exudate in the presence of interfering substances thanks to the unique modification strategy.

Promoting Effect on Healing of Diabetic Wound
The representative pictures of wounds in the control, Therapatch w/o iontophoresis, and Thera-patch w/iontophoresis group were recorded for 10 days (Figure 6a).The wound treated with Thera-patch w/iontophoresis demonstrated the fastest recovery, followed by the group administrated with Thera-patch w/o iontophoresis, while the untreated wound exhibited the slowest healing.A large dry scab was still visible on the back of the diabetic rat in the control group on the10 th day.The wound areas of three groups were further quantified based on the wound images, as shown in Figure 6b.On day 10, the wound area of the control group remained as high as 46.4%, whereas the Therapatch w/o iontophoresis group dropped to 15.4%, and the Therapatch w/iontophoresis group undergone the most significant decrease, to just 6.2% of its initial size.Thera-patch contained the MFCPH provides a sterile and moist healing environment for diabetic wounds.Importantly, it provides on-demand drug treatment based on the monitored wound status, thereby preventing drug abuse.Thera-patch w/iontophoresis group had a higher wound closure rate than Thera-patch w/o iontophoresis group (p < 0.05).Because the insulin release rate via the passive diffusion was relatively lower compared to on-demand insulin delivery via iontophoresis in a close-loop control manner.Additionally, electrical stimulation could enhance the proliferation and migration of fibroblasts, improve local blood perfusion, and promote keratinocytes to secrete more extracellular matrix, thus accelerating wound healing. [51,52]These results demonstrate the excellent diabetic wound healing properties of the wound theranostic system.
Wound healing process can be categorized into four distinct yet interconnected stages: hemostasis, inflammation, proliferation, and remodeling. [6]To evaluate the healing effectiveness of the wound theranostic system in histological view, regenerated granulation tissues were subjected to hematoxylin-eosin (H&E) and Masson trichrome staining which were collected on day 14.As illustrated in Figure 6c, after 14 days of wound administration, the residual scabs (crusts of dried blood, serum, and exudate) were observed in the control group, while no intermediate scabs were found in either of the two Thera-patch groups.A large number of dead inflammatory cells were encapsulated in the scabs in the control group (Figure 6c, white dotted circle), indicating the occurrence of inflammation at the early phase of healing.Besides, the Thera-patch w/o iontophoresis group exhibited a more organized structure, of which hair follicles, sebaceous glands and squamous epithelium can be easily observed (Figure 6c).Furthermore, the Masson staining revealed dense and well-organized collagen fibers deposition in the Thera-patch with iontophoresis group, indicating an effective therapeutic effect (Figure 6d).The collagen deposition in Thera-patch w/iontophoresis group was significantly higher than the Thera-patch w/o iontophoresis and the control group (68.6±4% vs 39.8±1.8% and 14.2±3.4%,respectively) (Figure 6i).CD31 labeled with red fluorescence was utilized to visualize the newly formed vessels.The Thera-patch w/iontophoresis group showed the most significant increase in red fluorescence intensity (Figure 6e).The relative coverage area of CD31 in Therapatch w/iontophoresis group was 3 times greater than that of the control group (Figure 6j).Therefore, the Thera-patch w/iontophoresis group effectively facilitated the formation of new blood vessels, essential for the transportation of wound growth factors, oxygen, and nutrients.Inflammation is a crucial indicator of infection and healing progress.The expression levels of interleukin-6 (IL-6) and tumor necrosis factor- (TNF-), two typical proinflammatory factors, were analyzed via immunohistochemistry.The control group exhibited high levels of IL-6 and TNF-, indicating severe wound inflammation (Figure 6f,g).In contrast, the expression of IL-6 and TNF- was down-regulated in two Thera-patch groups, likely attributed to MFCPH's inherent antibacterial properties that effectively safeguard wound endothelial cells against bacterial infection.Above all, the observed differences in results can be attributed to the synergy of the status-based on-demand insulin release and the accelerating effect of MFCPH.

Summary and Conclusion
A wireless theranostic wound management system is developed for the closed-loop administration of chronic wounds, with a particular focus on diabetic wound healing.The system is mainly consisted of a customized smartphone APP, a wearable electronic device, and a Thera-patch.The Thera-patch is capable of continuous monitoring of diabetic wound status through integrated pH and glucose sensors, as well as on-demand delivery of insulin via iontophoresis from the MFCPH reservoir.The sensors demonstrated good analytical performance, exhibiting a sensitivity of −62 mV/pH for pH and 0.48 μAmm −1 cm −2 for glucose, as well as high stability and accuracy for continuous monitoring of pH and glucose.The MFCPH, made of PDA-PPy incorporated PAM hydrogel, features a high drug loading rate of 85.8±3.6%, a high conductivity of 1.6 S m −1 , broad-spectrum antimicrobial ability, and good visibility of 70.0% with nanoparticle ratio of 0.6 wt% after aging for 3 days.Thera-patch w/iontophoresis group administrated in a closed-loop management mode significantly outperforms other groups with the fastest wound closing rate (93.8% within 10 days) and the optimal healing effect.This system offers a precise and effective approach to accelerate the healing of diabetic wounds while inhibiting bacterial infections, thereby enabling intelligent wound management with optimized intervention.
ltd, Liaoning, China).Similarly, the working electrodes, count electrode and reference electrodes were fabricated by accordingly deposition of the Ti, Au, Pt, and Ag nanoparticles on the front side.A transparent PDMS insulator was coated on the surface.Finally, the electrodes-circuit interfaces were cut by the laser.
Modifications of the Electrodes: The working electrodes were modified using an electrochemical workstation (CHI660E, CH instrument, Shanghai, USA). 1) RE preparation: Iontophoresis electrodes (anode and cathode) and two reference electrodes were chlorinated by 0.1 m FeCl 3 .2) pH sensing functionalization: PANI was deposited on the Au working electrode using the three-electrodes system (reference electrode: commercial Ag/AgCl electrode, count electrode: Pt electrode, working electrode: fabricated working electrode) for pH sensing.0.1 m aniline dissolved in 1 m HCl was first prepared.Next, the PANI was polymerized using CV from −0.2 to 1 V for 40 cycles at 0.01 V s −1 .In addition, 50 mg NaCl and 79.4 mg PVB were mixed with 1 mL methanol solution and a volume of 2 μL solution was casted on the pH RE. 3) Glucose sensing functionalization: Glucose sensors were functionalized according to the second-generation modification strategy.The mediator fabrication and functionalization process were identical to the previously reported work. [20]The glucose sensor was dried and stored in pH 7.4 PBS buffer at 4 °C prior to use.
Preparation of the MFCPH: The MFCPH was prepared according to the previously reported work. [33]The PDA-PPy NPs were firstly synthesized.1.5 g PVA was dissolved in 20 mL deionized water at 90 °C and then transferred to an ice-water bath for polymerization.1.2 g FeCl 3 •6H 2 O, 140 μL of pyrrole and 0.05 g of dopamine were added into the solution and continuous stirring for 9 h.The solution was centrifuged and washed with the hot water for 3 times to obtain PDA-PPy NPs.Subsequently, 15.6 mg NPs were dissolved in the 8 mL deionized water and subjected to sonication for 5 min to yield a homogeneous dark dispersion.2.6 g AM monomer was added with stirring.Upon dissolution of AM, 0.104 g APS (APS/AM:4 wt%), 0.003 g BIS (BIS/AM: 0.12 wt %) and 15 μL TMEDA were mixed with stirring for 5 min.Then the prepolymer solution was cast into a mold to form a pregel.After 3 days of aging at room temperature, the transparent MFCPH was obtained.
Characterizations of MFCPH: 1) Morphology characterization: The microstructures of PDA-PPy NPs and hydrogels were observed using SEM (Quanta 400F, FEI, Hillsboro, USA).The size and zeta potential of PDA-PPy NPs were analyzed using the Zetasizer Nano ZS90 (Malvern Instruments Ltd, Malvern, UK). 2) Transparent properties: The transmission spectrum of a 1.5 mm thick hydrogel in the visible wavelength range (400-800 nm) was measured using Lambda950 (PerkinElmer, MA, USA).3) Mechanical properties: The tensile strength and adhesive strength of the MFCPH were analyzed using a universal material testing machine (Instron, 5543A, Boston, USA).The extension rate was set at 100 mm min −1 for tensile test.The adhesive properties were tested by the pork muscle, pork skin and PET slices.First, the excessive fat was shed off and the skin sheets were immersed in PBS solution to guarantee the humidity.Next, the excessive PBS solution was sucked up by filter papers.The hydrogels were applied to the porcine skin tissues, muscle tissues and glass slides, respectively, with a bonding area of 12 mm × 30 mm. 4) Swelling properties: The MFCPH was dehydrated in a desiccator three days after preparation.After dehydration overnight, the initial weight was measured (W 0 ).The dehydrated hydrogel was immersed in the PBS buffer.The weight of the hydrogel (W t ) was recorded hourly.While the hydrogel equilibrates, its swelling ratio can be determined.The swelling ratio was calculated as: swelling ratio = (W t −W 0 )/W 0 × 100%.5) Conductive properties: The conductivity of MFCPH was measured by two-probe method using a programmable DC supply (IT6322, ITECH, USA).Briefly, the hydrogels (Φ = 1 cm, L = 5 cm) were placed between two Cu electrodes connected to the DC supply.The conductivity  was calculated as:  = IL/VR 2 , where I and V are the measured currents and potentials, respectively.6) Cytocompatibility: The HUVEC cell viability and proliferation in hydrogels were evaluated with CCK8 and Live/Dead assays.10 μL of the hydrogel was sterilized overnight with UV light in a 48-well plate.5000 HUVEC cells and culture medium (containing 10% BSA, 1% antibiotics, and 89% MEM Alpha medium) were added to each well and incubated at 37 °C with 5% CO 2 for 1-3 d.The medium without hydrogel was set as the control group.The medium was replaced at each interval, followed by a 1-h incubation with CCK8 in the incubator.The cell viability was assessed by measuring the absorbance at 450 nm.The cell viability was calculated as: Cell Viability (%) = ( OD Hydrogel −OD DMEM OD Control −OD DMEM ) × 100%.After 3 d proliferation, the cells were washed twice with sterile PBS.The staining solution containing calcein-AM and propidium iodide was added to each well for 30 min in darkness, followed by observation and recording of the cells using a fluorescence microscope.(ECLIPSE Ti-E, Nikon, Japan).7) Antibacterial Properties: The antibacterial efficacy of PDA-PPy-PAM hydrogels were investigated against Gram-positive bacteria S. aureus and Gram-negative bacteria E. Coli.The antibacterial rings were captured by a digital camera and analyzed by the Image J software (NIH, USA).
In Vitro Assessment of Sensing Performance: In vitro sensing performance of pH and glucose biosensors were tested using the electrochemical workstation.1) pH sensing: Mcllvaine's buffer was prepared using 0.1 m citrate and 0.1 m Na 2 HPO 4 solutions, with a pH range of 3 to 9 measured by a conventional pH meter (PHS-25, Shanghai Leici, Shanghai, China).The pH sensor was incubated in the buffer for 30 s and then recorded for 30 s.The potential effect of ion interferences, including 1 mm NH 4 + , 1 mm Mg 2+ , 1 mm Ca 2+ , 10 mm K + , and 20 mm Na + , was evaluated by adding them to the buffer.The stability was tested at pH 7 for 12 h.2) Glucose sensing: The glucose electrodes were validated in a composite solution containing 1 mm K 3 [Fe(CN) 6 ] and 0.1 m KCl.0.1 m PBS (pH = 7.4).The amperometric responses were recorded at an applied potential of 0.4 V (vs Ag/AgCl electrodes) within the physiological glucose range, up to 22 mm with increments of 2 mm.The specificity of glucose sensors was evaluated using 3 mm glucose solution containing relevant electroactive constituents.The constituents, including 10 mm LA, 2 mm cholesterol, 0.1 mm UA and 22 gm L −1 bovine serum albumin were added to the glucose-containing solution to investigate their potential effects on glucose detection.The concentration of those interferences was determined based on the concentration observed in the wound fluid. [29]The stability of glucose sensors was tested in a 5 mm glucose solution for 24 h.
In Vitro Evaluation of Drug Release Performance: The insulin (500 μL, 1.5 mg mL −1 ) was loaded into the dehydrated MFCPH by swelling for 24 h at 4 °C.Based on the isoelectric point of insulin (pI 5.3), the hydrogel on the cathode served as the drug delivery area.In vitro drug release performance of Thera-patch was tested using the Franz diffusion cells.Each chamber filled with 10 mL PBS and the insulin of Thera-patch was delivered into Franz cell via iontophoresis with the current ranging from 0 to 2 mA.The electro-controlled drug delivery capacity of the wound theranostic system was assessed by applying an iontophoresis current of 0.5 mA at 10-min intervals.Every 10 minutes, a 100 μL solution was extracted from the Franz diffusion cell and replaced with an equal volume of PBS buffer.The insulin concentration was then determined using a protein assay (Coomassie Plus protein assay, ThermoFisher) based on a preestablished standard optical absorption-concentration curve.
Animal Experiments: Male healthy SD rats (300±30 g) were provided by the Experimental Animal Center of Sun Yat-Sen University.Adult male SD rats, aged eight weeks and weighting between 250 and 300 g, were procured from the Experimental Animal Center of Sun Yat-Sen University.The type 1 diabetic rat model was induced via intraperitoneal injection of STZ at a dose of 55 mg k −1 g.STZ was prepared using a citrate-sodium citrate buffer solution at a concentration of 10 mg mL −1 , and the experimental rats underwent a 12-h fasting period prior to injection.Blood glucose levels of rats were measured using a meter on their tail tips.When the blood glucose level of the rat remained above 16.7 mm for 4 consecutive days, the diabetic model was considered successful.The diabetic rats were anesthetized with isoflurane (4% for induction, 2% for maintenance) using an R520 gas pump (RWD, China), and the hair on both sides of the posterior spine was shaved and disinfected with 2% iodophor.A 15 mm × 15 mm square full-thickness wound was surgically created on the dorsal surface of each rat.The rats were divided into three groups (n = 3).The wounds in the control group were cleaned with saline and dressed with sterile gauze.The Thera-patch w/o iontophoresis group applied a patch loaded with 20 IU insulin onto the wound and delivered insulin via passive diffusion.In the Thera-patch w/iontophoresis group, the patch loaded with 20 IU insulin was applied to the wound and insulin was on-demand delivered via iontophoresis (0.5 mA, 30 min) based on the detected glucose levels in the wound.The wound theranostic system measured pH and glucose levels in the wound, triggering insulin delivery in a closed-loop manner.The Therapatch was replaced daily, the wound morphology was recorded using a digital camera, and the wound area was quantified using Image J.
Statistical Analysis: Before in vivo testing, the glucose and pH sensors were calibrated according to the linear fitting curve.The open circuit potential and the amperometric electrochemical response were transformed to pH value and glucose concentration, respectively.As for the Clarke's error grid analysis, the glucose concentration beyond 22.22 mm was removed.The detection error of the glucose sensors was calculated by: Detection error = |WTS -BG|/BG × 100%.All the results in this study were reported as mean ± standard deviation (SD).Sample size (n) of the repeated experiments for each statistical analysis was shown in the figure legends.Statistical differences were analyzed using one-way ANOVA followed by the t-test analysis.Differences were considered significant at *p < 0.05, **p < 0.01, and ***p < 0.001 and the p-values were calculated by Origin-Pro Software (version 2017).

Figure 2 .
Figure 2. Fabrication and characterization of the Thera-patch.a) Explosive view of the Thera-patch.b) Assembly process of the Thera-patch.i) The prepared flexible electrode array.ii) The impermeable ring was bonded on the flexible electrode array.iii) The MFCPHs were fixed within the impermeable ring.iv) The patch was sealed by a PET membrane.c) The layer-by-layer modification process of the sensing electrodes.d) Electrodeposition of PANI on the working electrode of pH sensor by cyclic voltammogram.e) SEM image of PANI deposited on the working electrode of pH sensor.f) SEM image of the modified working electrode surface of glucose sensor.g) Schematic illustration of the preparation procedures of the MFCPH.h) Size distribution of PDA-PPy NPs.i) SEM image of PDA-PPy NPs.j) SEM image of nanofibrils seamlessly fused with the hydrogel matrix•.

Figure 3 .
Figure 3.In vitro evaluation of the pH and glucose sensors.a) Illustration of the detection mechanism of the pH sensor.b) The measured potential to pH level decreasing from 9 to 3 and then back to 9. c) The linear fitting curve of the detected potential versus the pH value (N = 3).d) Selectivity of the pH sensor in PBS with the addition of NH 4 + , Mg 2+ , Ca 2+ , K + , and Na + .e) Dynamic response of the pH sensor under the pH 7 for 20 h.f) The reproducibility of three pH sensors at pH 7 (N = 5).g) Illustration of the detection mechanism of the glucose sensor (redox enzyme, PDB code 1CF3).h) The amperometric responses of glucose sensor to the glucose concentration ranging from 0 to 22 mm.i) The linear fitting curve of the detected current versus the glucose concentration (N = 3).j) Selectivity of the glucose sensor in PBS with the addition of 10 mm LA, 2 mm cholesterol, 0.1 mm UA, 22 gm L −1 albumin and 3 mm glucose.k) The stability of the glucose sensor in 5 mm glucose solution for 24 h.l) The reproducibility of three glucose sensors detected in 5 mm glucose solution (N = 5).

Figure 4 .
Figure 4.In vitro characterization of the electro-controlled drug delivery module.a) Schematic illustration of the drug delivery mechanism of iontophoresis.b) Loading efficiency of insulin in MFCPH after 0.5 to 24 h incubation (N = 5).c) The transparency changes of the hydrogel block (50 × 20 × 1.5 mm 3 ) during the nanofiber formation in 3 days.d) Transmittance spectrum of MFCPH in 3 days.e) Digital graphs of survival i) E.coil and ii) S.aureus clones after contacting with PAM hydrogel, hydrogel I (MFCPH, 0.15 wt% PDA-PPy NPs), hydrogel II (MFCPH, 0.3 wt% PDA-PPy NPs), hydrogel III (MFCPH, 0.6 wt% PDA-PPy NPs), and hydrogel IV (MFCPH, 1.2 wt% PDA-PPy NPs).The red circles refer to the inhibition zones.f) Statistical analysis of inhibition zone sizes (N = 3).g) Live/dead staining of HUVEC cells after being cocultured with MFCPH.h) Effects of MFCPH on HUVEC cell viability in 3 days (N = 4).i) Stress-strain curves of unswelled and swelled MFCPH.j) Adhesion strengths of MFCPH to porcine muscle tissues, porcine skin tissues, and PET slides (N = 5).k) The conductivity of MFCPH as the PDA-PPy NPs transformed into nanofibrils (N = 4) in 3 days.The inset shows the lightening of LED via MFCPH, demonstrating good conductivity.l) Numerical simulation of the current density generated by the custom-designedAg/AgCl electrodes.m) Cumulative amounts of insulin released from MFCPH of Thera-patch at the given iontophoresis currents of 0 (free diffusion), 0.5, 1 and 2 mA for 90 min, respectively.The free drug diffusion group is designated as the control (N = 3).n) The insulin permeation amount within every 10 min during 1 h drug delivery of Thera-patch (N = 3).o) The intermittent drug release operated in an "on-off" cycle for 50 min, with "on" indicating the application of 0.5 mA iontophoresis and "off" representing free drug diffusion.(N = 3).p) The permeation rate of each "on-off" iontophoresis cycle (N = 3).*p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5 .
Figure 5.In vivo performance of the wound theranostic system.a) Optical images of the wound theranostic system on the diabetic rat.The Thera-patch and the wearable electronics were adhered at the wound site for wound monitoring and iontophoresis drug delivery, with a smartphone for data and command transmission.b) Schematic illustration of the detailed experimental procedure for establishing and treating diabetic wound.pH of c) the control group, d) the Thera-patch w/o iontophoresis group and e) the Thera-patch w/iontophoresis group during the wound-healing process, (N = 3).Glucose concentration of f) the control group, g) the Thera-patch w/o iontophoresis group and h) the Thera-patch w/iontophoresis group during the wound-healing process.The red arrow indicates start of insulin iontophoresis.The WTS and BG refer to the wound theranostic system and blood glucometer.i) The serum insulin of diabetic rats treated by Thera-patch w/and w/o iontophoresis.j) The Clarke's error grid analysis illustrating the accuracy of the glucose sensor compared to the standard blood glucometer.k) The detection errors of Thera-patch compared with the commercial blood meters at corresponding time points.Heatmap plots of the fluctuations of l) pH and m) glucose.n) The average errors of glucose concentration detected by the control, the Thera-patch w/o iontophoresis and the Thera-patch w/iontophoresis group (N = 17).

Figure 6 .
Figure 6.Healing efficiency of the diabetic wound treated with the wound theranostic system.a) The diabetic wounds treated with sterile gauze (control), Thera-patch w/o iontophoresis and Thera-patch w/iontophoresis on 0, 2, 4, 6, 8 and 10 day.The right images were the illustration of wound healing process.b) Diabetic wound area of three groups.Representative images of c) H&E, d) Masson, e) CD31, f) TNF- and g) IL-6 immunohistochemical staining obtained from wound tissues treated with sterile gauze (control), Thera-patch w/o iontophoresis and Thera-patch w/iontophoresis at 14 days post-surgery.h) Quantitative analysis of the relative coverage area of Masson.i) Quantitative analysis of the relative coverage area of CD31.Data are shown as mean ± SD (N = 3).*p < 0.05, **p < 0.01, ***p < 0.001.Thera-patch with iontophoresis group compared with the control group by OneWayANOVA.#p < 0.05, ##p < 0.01, ###p < 0.001.Thera-patch w/iontophoresis group compared with the Thera-patch w/o iontophoresis group by OneWayANOVA.