Development and characterization of a smart obstetric dressing with a visual indication function

Abstract

Excessive adhesion between dressings and the wound bed often causes secondary tissue damage and delays healing during dressing removal. This study aimed to develop and evaluate a multifunctional moisture-containing nonwoven dressing composed of silk fibroin protein (SF), polycaprolactone (PCL), and ellagic acid (EA) to improve wound healing while allowing visual assessment of moisture. A composite nonwoven fabric integrating SF, PCL, and EA was fabricated and characterized using spectroscopic analysis to evaluate its fluorescence properties. In vitro cytocompatibility was assessed in the four fibroblast cell lines using cell viability assays. In vivo wound-healing efficacy was investigated in an animal model, and histological evaluation was performed using Masson’s trichrome staining and histomorphometric analysis. Spectroscopic measurements indicated that SF and EA synergistically produced bright blue-violet fluorescence under UV illumination, with the fluorescence intensity increasing proportionally to the SF concentration. Cytocompatibility assays confirmed excellent biocompatibility, with all treatment groups maintaining >80% cell viability and the indicator fabric alone exceeding 100%. In vivo studies revealed that both the SF-based indicator fabric and the EA-integrated dressing significantly accelerated re-epithelialization and dermal remodeling compared to a commercial hydrogel control. Histological analysis revealed denser collagen deposition and increased neovascularization in the SF-treated wounds. The SF/PCL/EA moisture-indicating dressing demonstrated dual functionality by serving as an effective visual indicator of exudate and promoting tissue repair. This composite dressing provides both therapeutic and diagnostic benefits for advanced wound-care applications.

Introduction

Cesarean section is the most commonly performed surgical procedure in obstetrics, with a global rate of 21.1% as of 2019, particularly exceeding 50% in several countries including the Dominican Republic and Egypt. ¹ Although generally safe, postoperative wound care is critical in preventing complications such as delayed healing, wound dehiscence, and surgical site infections (SSIs), which continue to contribute significantly to maternal morbidity. ² In routine clinical practice, dressing changes are often performed according to fixed schedules or based on subjective assessments of wound moisture, primarily through visual inspection and clinical experience rather than objective indicators. This non-individualized approach may result in premature dressing removal, potentially destroying granulation tissue, or prolonged dressing retention, leading to excessive exudate accumulation, periwound maceration, and microbial colonization.^3,4 Both scenarios are detrimental to wound healing and patient well-being, particularly in postpartum scenarios, where effective wound care is essential for early ambulation, breastfeeding initiation, and maternal-infant bonding. In addition, studies have reported that longer dressing intervals without adequate monitoring are associated with higher rates of wound-related complications, particularly in at-risk populations, such as those with obesity or diabetes. ⁵ These limitations highlight the need for advanced dressings that not only provide protection and absorbency but also provide real-time visual feedback on exudate saturation to enable timely and personalized interventions.

Conventional moist wound dressings are broadly categorized into hydrocolloid and hydrogel dressings. Hydrocolloid dressings are anhydrous but form a gel on contact with wound exudate that maintains a moist environment while minimizing water loss. Gelation reduces adhesion to the wound bed and reduces the risk of trauma during dressing changes. However, with low levels of exudate or frequent dressing changes, the residual tackiness of hydrocolloids can inadvertently damage newly formed tissue.^6,7 In contrast, hydrogel dressings consist of approximately 70% water and are effective for rehydrating dry or necrotic wounds. By releasing moisture, they promote autolytic debridement by softening the devitalized tissue. However, hydrogels lack mechanical integrity and require secondary fixation with transparent film dressings. Clinically, dressings must cover an area of at least 3 cm beyond the wound margin to ensure stability. Overuse or poor fitting may result in periwound maceration due to excessive moisture exposure. ⁸ Despite their widespread use, both hydrocolloid and hydrogel dressings lack real-time indicators of exudate saturation, forcing clinicians to rely on indirect signs such as expansion, discoloration, or tactile changes, which are often nonspecific and susceptible to misinterpretation. The absence of dynamic feedback can lead to suboptimal timing of dressing changes and impaired healing, particularly in wounds with fluctuating amounts of exudate. To overcome this limitation, the development of smart, indicator-integrated wound dressings offers a promising solution as they combine effective exudate absorption with visual cues, enabling timely and personalized interventions and improved wound care outcomes.

In this study, a cost-effective, visually responsive wound dressing was developed specifically for obstetric applications that provides real-time feedback on exudate saturation through a simple, UV-activated color change. The dressing was engineered using a composite of silk fibroin protein (SF), polycaprolactone (PCL), and ellagic acid (EA), chosen for its excellent biocompatibility, mechanical flexibility, and optical sensitivity to moisture. Silk fibroin, a fibrous protein derived from Bombyx mori, has been extensively studied for its high safety profile and is used for Food and Drug Administration-approved suturing materials and tissue scaffolds. In addition, its ability to facilitate cell adhesion, proliferation, and matrix integration makes it particularly suitable for sensitive applications such as cesarean section wound care.^9–11 Regarding the preparation ratio and the fluorescence indication mechanism between SF and EA, we have conducted related studies in our previous study. In that study, it was discovered that a wound dressing composed of sericin and ellagic acid incorporated into a silk fibroin protein (SFP) nonwoven fabric could serve as a wound moisture indicator for managing wound healing. ¹² An indicative nonwoven fabric was fabricated using silk fibroin protein (SFP) and polyvinyl alcohol (PVA) through a single-spinneret electrospinning technique. The indicative function of the nonwoven fabric was provided by ellagic acid (EA), a type of polyphenol. The foam dressing, serving as an absorbent layer, was composed of sericin and could regulate the wettability of the wound microenvironment. The indicative function of the dressing was attributed to its fluorescence emission, which was confirmed by increases in both the UV absorption spectrum and the fluorescence emission spectrum. However, previous studies have shown that PVA + SF + EA nonwoven fabrics tend to rapidly disintegrate after absorbing tissue fluid, posing a significant challenge to their development as obstetric dressings suitable for clinical application. Since this study primarily aims to utilize the indicator function of SFP to develop obstetric dressings closer to clinical use, the development process also seeks to address the shortcomings identified in earlier research and reduce the associated costs of dressing preparation. Therefore, we chose to partially replace PVA with PCL for preliminary animal experiments. To improve mechanical strength and structural stability, SF was blended with PCL, a biodegradable polyester known for its slow degradation rate and elasticity, which supports the integrity of the dressing during patient movement.^13–15 EA, a naturally derived polyphenol with fluorescent properties, was incorporated as a visual indicator that provides an exudate-responsive color change when illuminated under 365 nm UV light.^16,17 This allows clinicians to assess saturation levels without removing the dressing, thereby avoiding unnecessary intervention This composite design addresses the key limitations of conventional moist dressings by combining absorbency, visual indication, and biocompatibility, providing a practical platform to improve postpartum wound monitoring and reduce infection risk in clinical settings.

The exudate absorption capacity, indicator response speed, and cytocompatibility of the composite dressings were evaluated using standard in vitro assays. Quantitative absorption tests demonstrated that the dressing efficiently retained wound-like fluid without structural degradation, whereas the fluorescence-based indicator exhibited a rapid and discernible color shift in response to increasing moisture levels under 365 nm UV illumination. This responsiveness enables non-invasive assessment of dressing saturation, potentially reducing the frequency of unnecessary dressing changes. Biocompatibility was confirmed using multiple fibroblast cell lines, including mouse (3T3, L929) and human-derived (CG1519, CG1629) cells. No cytotoxic effects were observed at any tested concentrations of the dressing extract, and in some cases, the material even promoted mild cell proliferation, highlighting its potential for safe contact with healing tissue.

Collectively, these findings support the feasibility and safety of the proposed indicator-integrated dressing as a next-generation platform for postpartum wound care. The combination of moisture sensitivity, visual indicators, and biological compatibility offers a practical solution to current challenges in surgical wound monitoring. By enabling timely and individualized dressing changes, this system can help reduce infection risk, prevent wound maceration, and minimize material waste. Furthermore, its simple design and low-cost fabrication suggest broad applicability in both high-resource hospitals and low-resource clinics where optimizing wound care efficiency is a priority.

Materials and methods

Materials

Silkworm cocoons were obtained from the Chuan-Ming Ecological Silkworm Farm (Miaoli, Taiwan). EA dihydrate (CAS 133039-73-3) and LiBr (CAS 7550-35-8) were purchased from Tokyo Chemical Industry (Tokyo, Japan). Formic acid (CAS 64-18-6), polyvinyl alcohol (CAS 9002-89-5), dimethyl sulfoxide (DMSO, D2650), sodium bicarbonate (S8875), PCL (99%, Mn 70,000–90,000, Sigma-Aldrich, MO, USA), and MTT (Thiazolyl Blue Tetrazolium Bromide (MTT) reagent (M2128) were purchased from Sigma-Aldrich (St. Louis, MO, United States). Phosphate-buffered saline (PBS, UR-PBS001) was sourced from UniRegion Bio-Tech (Taipei, Taiwan). Dulbecco’s Modified Eagle Medium (DMEM, 1965-084) was obtained from Gibco (Thermo Fisher Scientific, MA, United States). Trypan blue (03-102-1B) and trypsin (03-051-5B) were obtained from Biological Industries (Israel).

Equipment

An electrospinning system (MODEL 205B-20R) was obtained from BERTAN (Bertan High Voltage Corp. (NY, United States). A CO₂ incubator (MCO-20AIC) and −20°C freezers (SCD-200) were sourced from SANYO Electric Co. (Osaka, Japan). An autoclave (TM-328) was purchased from TOMIN (New Taipei City, Taiwan). A biosafety cabinet was soured from Tso Shin Enterprise Co., Ltd. (New Taipei City, Taiwan). Micropipettes were purchased from CILSON (Taiwan). A liquid nitrogen tank (34HC) was obtained from Taylor–Wharton (PA, United States). A magnetic stirrer (RCT Basic) was purchased from IKA (Staufen, Germany). The microbalance (S1234) was sourced from DENVER INSTRUMENT (CO, United States). −80°C freezers (ULT1786-3-D31) were obtained from Revco Scientific Inc. (NC, United States). The syringe pump (NE-300) was purchased from Just Infusion™ (NY, United States). A high-speed centrifuge (HERAEUS MULTIFUGE X1R) was obtained from Thermo Fisher Scientific (Waltham, MA, United States). Scanning electron microscope (SEM, JSM-6500F) was obtained from JEOL Ltd. (Tokyo, Japan). An optical microscope (LEICA DM 2000) was sourced from Major Instruments Co., Ltd. (New Taipei City, Taiwan). The plate reader (SPECTROstar Nano) was soured from BMG LABTECH (Ortenberg, Germany). A photoluminescence spectrometer (FP-8500) was purchased from JASCO (Tokyo, Japan). 1 mL and 24G syringes were sourced from TERUO® (Taiwan).

EA-coated silk fibroin protein/polycaprolactone (SFP) nanofibrous dressing with fluorescent indication

This study aimed to fabricate a dual-layer smart wound dressing by integrating a fluorescent SFP nanofibrous mat functionalized with EA for real-time exudate monitoring. The composite dressing was integrated with the A⁺med hydrating gel and SFP (ASFP), which comprised an A⁺med hydrogel base layer to absorb wound exudate and maintain a moist environment, thereby promoting tissue regeneration, and a nanofibrous SFP mat that served as a visual moisture indicator. As exudate saturated the dressing, it diffused into the EA-treated cross-shaped region of the nanofiber layer, resulting in fluorescence quenching under 365 nm UV light. This visual cue enabled timely dressing changes, reduced infection risk and minimized unnecessary dressing changes.

Preparation of SFP

Stage I: Degumming and separation of fibroin and sericin

Twenty milligrams of clean silkworm cocoons were weighed and placed in a 2000 mL beaker containing 1000 mL of deionized water. The beaker was sealed with aluminum foil and autoclaved at 121°C for 1 h to remove sericin. The degummed silk fibers were rinsed thoroughly with deionized water, evenly spread on a mesh, and dried in a convection oven (Scheme 1).

Scheme 1.

Schematic diagram of the silk fibroin extraction process. The workflow includes degumming, LiBr dissolution, dialysis, freeze-drying, and reconstitution, yielding purified SF suitable for electrospinning.

Stage II: Preparation of silk fibroin sponge

As shown in Scheme 1, to prepare the silk fibroin solution, 50 mL of deionized water was heated to 65°C, and lithium bromide (LiBr) was added to reach a final concentration of 9 M. Five grams of dried silk was shredded into cotton-like fragments to increase the surface area and then dissolved in the LiBr solution. A 16 cm long dialysis membrane was folded at one end by 1.5–2 cm and secured with a dialysis clip. The membrane was soaked in deionized water until fully hydrated, filled with silk solution, sealed at the other end, and dialyzed against deionized water for 5 days. The water was changed every 30 min during days 1–2 and every 1.5 h during days 3–5. The dialyzed solution was transferred to 50 mL centrifuge tubes, balanced, and centrifuged at 8000 rpm for 10 min at 4°C. The supernatant was aliquoted into molds (10 mL per mold) and frozen at −80°C for 24 h. Finally, the samples were lyophilized for 48 h to obtain silk fibroin sponge blocks, which were stored in a dry, cool environment until further use.^18,19

Stage III: Fabrication of the fluorescent indicator nanofibrous mat

The fabrication process of the fluorescent indicator nanofibrous mat is illustrated in Scheme 2. 40 mg of SF sponge and 105 mg of PCL were weighed and placed in a clean sample vial. Subsequently, 0.6 mL of formic acid was added to the vial, and the mixture was stirred vigorously with a magnetic stirrer until a homogeneous SFP solution was obtained. To ensure stable environmental conditions, a dehumidifier was used to maintain ambient humidity between 40% and 45%. The resulting solution was loaded into a 1 mL syringe and mounted onto a syringe pump set to a flow rate of 5 μL/min. Electrospinning was performed at 17 kV with a working distance of 7.5 cm, and the nanofibers were deposited onto a square iron collector to form a nonwoven mat. EA was dissolved in ethanol at a concentration of 300 μg/mL with heating to prepare the EA ethanol solution. The electrospun nanofibrous mats were then laid flat and fixed, and the EA solution was applied to the surface in a cross-shaped pattern. The mat was placed in a vacuum chamber to evaporate ethanol, and this coating process was repeated four times to ensure sufficient EA loading, resulting in a fluorescent indicator-integrated nanofibrous dressing.

Scheme 2.

Fabrication process of the moisture-indicating nonwoven fabric. SF and PCL were dissolved in formic acid and electrospun to form a fibrous mat, followed by EA coating via ethanol solution and vacuum drying to create a cross-patterned fluorescent indicator.

Optical and structural characterization

Stage I: Fiber morphology analysis by SEM

The surface morphology of the electrospun nanofibrous mats was characterized using SEM (JEOL Ltd., Tokyo, Japan). Samples were sputter coated with a thin layer of gold and imaged at an accelerating voltage of 15 kV. Fiber diameter and uniformity were quantified using ImageJ software based on the acquired micrographs.

Stage II: Ultraviolet-visible absorption spectral analysis

Ultraviolet-visible (UV-Vis) absorbance spectra of the nanofiber mats were measured using a microplate reader (BMG LABTECH, Ortenberg, Germany) in the range of 190–780 nm. Characteristic peaks were analyzed to identify conjugated structures and confirm successful incorporation of EA.

Stage III: Photoluminescence (PL) spectroscopy

PL spectroscopy was conducted to evaluate the optical responsiveness of EA-functionalized nanofiber mats. Emission spectra were recorded using a fluorescence spectrometer (JASCO, Tokyo, Japan) with excitation at 365 nm. Fluorescence intensity and wavelength shifts were monitored before and after exposure to moisture to evaluate the performance of the optical indicator.

Fibroblast cell culture

NIH/3T3 and L929 (mouse-derived dermal fibroblasts), CG1519 (normal human female dermal fibroblasts), and CG1629 (normal human male dermal fibroblasts) were obtained from the Food Industry Research and Development Institute, Bioresource Collection and Research Center (BCRC). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, pH 7.4) supplemented with 2 g/L NaHCO₃, 1% (v/v) penicillin (10 mL/L), 10% (v/v) fetal bovine serum (FBS, 100 mL/L), and 5 µg/mL bovine insulin (5 mg). The medium was sterilized using a 0.22 µm filter, and cells were maintained in a humidified incubator at 37°C and 5% CO₂. Subculturing was performed at approximately 80% confluence by detaching the cells with trypsin, neutralizing with complete medium, and centrifuging at 1300 rpm for 3 min. The resulting cell pellet was resuspended in fresh medium, and viable cells were quantified using the trypan blue exclusion method with a hemocytometer. Cell concentration was calculated using the following formula: (average cell count/8) × 10⁴ × 10. For cryopreservation, cells were mixed with freezing medium (90% complete medium and 10% DMSO), transferred to labeled cryovials, frozen at −80°C for 24 h, and subsequently stored in liquid nitrogen (−196°C) for long-term preservation.

Evaluation of the cytotoxicity of the cells

To evaluate the cytocompatibility of the moisture-sensitive fluorescent nonwoven fabric, the material was first sterilized under UV light for 1 h, then soaked in serum-free DMEM and incubated at 37°C for 24 h to allow leachable components to diffuse into the medium. 3T3, L929, CG1519, and CG1629 cells were seeded in 24-well plates at a density of 1 × 10⁵ cells/well and cultured for 24 h to allow adhesion. The following day, the extract-containing medium was filtered through a 0.22 µm syringe filter. After the original medium was removed from the wells and the cells were rinsed twice with 1 × PBS, the filtered extract (experimental group) or fresh serum-free medium (control group) was added to each well, followed by incubation at 37°C for 24 h. On day three, the MTT stock solution (0.5 mg/mL in PBS) was freshly prepared. After removing the culture medium and washing with PBS to eliminate residual medium, 50 µL of the MTT solution was added to each well and incubated for 2 h under light-protection conditions. Subsequently, 450 µL of DMSO was added to dissolve formazan crystals. Absorbance was measured using a microplate reader, and optical density of the experimental group was compared with that of the control group to determine relative cytotoxicity of the test material.

Animal wound model

Wound healing experiments were conducted with BALB/cByJ Narl mice obtained from the National Laboratory Animal Center (Taipei, Taiwan). Mice were housed at the National Defense Medical Center in a controlled environment (23–25°C, 12 h light/dark cycle, filtered air) with free access to food and water. All procedures were approved by the Institutional Animal Care and Use Committee of the National Defense Medical Center [IACUC approval No. IACUC-22-178]. To evaluate the moisture-indicating function of the wound dressing, full-thickness excisional wounds (1 cm × 1 cm) were created on the dorsal skin of anesthetized mice (Zoletil™ 20–40 mg/kg, 40–80 µL, intraperitoneal injection). After hair removal and skin marking, wounds were surgically created with sterile scissors, covered with the dressings, and analgesics were administered subcutaneously. Dressings were changed based on disappearance of fluorescence in the cross-shaped indicator zone. If the indicator was lost, mice were anesthetized again, dressings were removed for photographic documentation under UV light, and wound size and conditions were recorded. If fluorescence persisted, dressings were replaced every 2 days until complete healing. Wound area reduction was quantified using ImageJ software (NIH, Bethesda, MD, United States), and healing efficacy was statistically analyzed using a t-test.

Histological analysis

Wounded skin tissues were harvested, fixed in 10% neutral-buffered formalin for 24 h, dehydrated using a graded ethanol series, and embedded in paraffin for histological analyses. Serial sections were prepared using a microtome and mounted on glass slides for histologic staining. All stained slides were visualized with an optical microscope (Major Instruments Co., Ltd., New Taipei City, Taiwan), and representative images were captured. Quantitative image analysis of collagen area and vascularized regions was performed using ImageJ software (NIH, United States).

Stage I: Hematoxylin and eosin (H&E) staining

H&E staining was performed to assess overall tissue morphology, epithelial regeneration, and cellular infiltration. Paraffin sections were deparaffinized with xylene and rehydrated using descending grades of ethanol. The slides were stained with hematoxylin, rinsed with tap water, differentiated with acidic alcohol, and counterstained with eosin. After dehydration and clearing, the sections were mounted with a coverslip. All sample were prepared by the Department of Clinical Pathology at Tri-service General Hospital.

Stage II: Masson’s trichrome staining

Masson’s trichrome staining was performed to evaluate collagen deposition and fibrotic remodeling in the wound bed. After deparaffinization and rehydration, the tissue sections were stained using a commercial Masson’s trichrome staining kit, according to the manufacturer’s protocol. Collagen fibers appeared blue, cytoplasm red, and nuclei black. The sections were dehydrated and mounted for imaging.

Statistical analysis

Data were expressed as the mean ± standard deviation (SD). Statistical analyses were performed using Wilcoxon statistics, one way ANOVA, and Student’s t-test with SPSS software (SPSS, Chicago, IL, United States) to assess differences between experimental groups. Statistical significance was accepted at p < 0.05 (*), p < 0.01 (**), or p < 0.001 (***).

Results

Structural modification of SFP nonwoven fabric by coating with EA

To improve moisture retention and ensure timely detection of exudate in wound dressings, it is essential to optimize the microstructure of the indicator material. Specifically, increasing fiber density and reducing porosity can slow fluid penetration into the indicator layer, thereby providing a more accurate representation of wound saturation. To investigate this, the structural differences between the indicator and non-indicator regions of the SFP nonwoven fabrics were compared. Figure 1(a) presents an SEM image of the non-indicator region, revealing randomly interwoven fibers with a porous architecture that supports breathability and moderate fluid absorption. In contrast, Figure 1(b) shows the indicator region after EA ethanol treatment, which induced partial fiber dissolution and interaction with silk fibroin, resulting in thicker, more densely packed fibers and significantly reduced porosity.

Figure 1.

SEM images of SFP moisture-indicating nonwoven fabric: (a) non-indicator region and (b) indicator region after EA coating.

Quantitative image analysis revealed that the average fiber diameter increased from 6.38 ± 1.58 µm (non-indicator) to 13.52 ± 5.73 µm (indicator). Concurrently, the pore area percentage decreased from 27.5% to 3.0%, and the average pore size from 0.274 m ² to 0.063 m ². This densification is expected to reduce water permeability in the indicator zone, thereby extend fluorescence retention, and allow fluid absorption to proceed through the hydrogel layer before reaching the indicator. Such structural modulation is essential for synchronizing fluorescence quenching with dressing saturation and for improving both detection sensitivity and temporal accuracy. These findings support the strategy of localized structural tuning to balance absorption capacity with controlled indicator activation, which is critical for the development of smart wound dressings that enable visual, time-resolved exudate monitoring.

SFP dressing demonstrates effective visual moisture indication via UV-responsive quenching

To address the clinical need for real-time monitoring of wound moisture levels, an SFP-based dressing containing a cross-shaped fluorescent indicator was developed. The objective was to visually correlate fluorescence intensity with exudate saturation over time, allowing a semi-quantitative assessment of wound condition. Time-dependent fluorescence imaging under 365 nm UV illumination was used to evaluate indicator performance in response to simulated exudate. As shown in Figure 2(a) to (d), the dressing exhibited strong blue-violet fluorescence at 0 min, which gradually diminished over time as the fluid permeated the indicator region. Partial quenching was observed after 60 min, substantial reduction at 90 min, complete fluorescence loss at 120 min, indicating that the dressing had reached full saturation. These time points correspond to known volumes of simulated exudate absorbed by the hydrogel layer, allowing fluid content estimation from the fluorescence state. The indicator system showed high sensitivity to moisture levels, producing distinguishable visual changes within clinically relevant timeframes.

Figure 2.

Analysis of composite dressing exudate absorption under 365 nm UV illumination: (a)–(d) time-lapse fluorescence images (0, 60, 90, and 120 min) of the indicator region. (e) UV–Vis absorption spectra of SF, EA, and SF + EA.

To further investigate the optical mechanism underlying this response, UV–Vis absorption spectra were obtained for SF, EA, and their combination (SF + EA) (Figure 2(e)). The SF + EA complex exhibited increased absorbance at 365 nm and a distinct shoulder at 359 nm, indicating the formation of a UV-responsive conjugated system through molecular interactions. This structure likely accounts for both strong initial fluorescence and the fluorescence quenching observed upon hydration. Collectively, these findings demonstrate that the SF + EA modified dressing provides reliable and sensitive moisture indication based on a time-dependent optical signal, making it a practical tool for non-invasive monitoring of wound exudate and optimal timing of dressing change.

Enhancement of fluorescence intensity with increased silk fibroin content

To evaluate the effect of SF content on optical performance of the indicator material, PL spectra were measured at 365 nm excitation for composite films containing 30 mg or 60 mg of SF. As shown in Figure 3(a), both samples exhibited distinct emission peaks at approximately 412 and 466 nm, corresponding to blue-violet and blue fluorescence, respectively. As the SF content increased from 30 to 60 mg, the emission intensity increased substantially (Figure 3(b)). Specifically, the 60 mg sample showed peak intensities of 1714 (×10⁴ counts) at 412 nm and 1195 at 466 nm, compared with 1361 and 983 for the 30 mg sample. This dose-dependent increase suggests that higher SF concentrations promote greater formation of fluorescent conjugates with EA, enhancing the brightness and visibility of the moisture-indicating region. These results confirm that silk fibroin concentration is a tunable parameter that directly affects fluorescence output. For practical wound-care applications, the 60 mg concentration provides a stronger initial signal while balancing material costs, rendering it suitable for real-time moisture detection under UV light.

Figure 3.

Photoluminescence (PL) spectra of SF + EA composites at 365 nm excitation: (a) emission peaks at 412 nm and 466 nm correspond to blue-violet fluorescence and (b) 30 and 60 mg SF concentration of fluorescence intensity.

Both SFP and ASFP maintain cell viability in multiple fibroblast lineages

The biocompatibility of the developed fluorescent indicator dressings was evaluated using four fibroblast cell lines: murine 3T3 and L929 and human dermal fibroblasts CG1519 and CG1629. As shown in Figure 4(a), SFP dressing extracts (25%–100%) did not exhibit cytotoxicity in any cell type, with cell viability consistently above 100% compared to the control. Notably, in 3T3 and CG1629 cells, higher concentrations of SF resulted in a mild but reproducible proliferation-promoting effect (p = 0.04 < 0.05), indicating that silk fibroin and its composite components support cell growth. In contrast, extracts from the ASFP dressing (Figure 4(b)) resulted in slightly reduced viability, particularly at 100% extract concentration. Nonetheless, all cell viability values remained within the acceptable range (80–100%, p = 0.04 < 0.05) and well above the ISO 10993-5 cytotoxicity threshold of 70%, confirming overall safety for wound-contact applications. The decrease in cell viability was likely attributed to additive components of the hydrogel matrix, which did not induce significant cytotoxic effects. Overall, these results demonstrated that the SFP indicator dressing possesses excellent in vitro cytocompatibility and can increase fibroblast activity, whereas the ASFP composite remains safe for clinical use, supporting its potential in exudate-monitoring wound care systems.

Figure 4.

Cytotoxicity test of SFP and ASFP. Cell viability was measured in four fibroblast lines (3T3, L929, CG1519, CG1629) after 24 h of exposure: (a) SFP group and (b) ASFP group.

Silk fibroin-based dressings improve the efficiency of in vivo wound closure

The therapeutic efficacy of SFP and ASFP dressings was evaluated using a full-thickness excisional wound model in BALB/cByJNarl mice (Figure 5(a)). Four groups were compared: negative control (NC), A⁺med hydrating gel dressing (AMED), ASFP, and non-woven SFP indicator alone. Representative images of wound sites over 44 days and the corresponding quantification of wound areas are summarized in Figure 5. All experimental groups showed progressive wound closure over time; however, the SFP and ASFP groups healed significantly faster than the AMED and NC groups. Particularly, wounds treated with SFP dressings exhibited a marked area reduction by day 7 and achieved complete closure by day 26. ASFP-treated wounds followed a similar trend but showed a slightly slower closure rate, with complete healing observed around day 30. In contrast, the AMED group displayed delayed wound resolution, with residual wound areas still visible after 35 days.

Figure 5.

Time-course analysis of wound healing in mice. Wound area percentages were quantified on days 0–44 across treatment groups. All groups data were test harvested from n = 6 mouse: (a) representative images of wound sites over 44 days, (b) quantitative analysis, (c) the indicator results of the ASFP groups, and (d) the indicator results of the SFP groups.

In Figure 5(b), the wound area curves for the AMED group and the NC group show a significant difference (day 2 to day 26: p = 0.01 to 0.04 < 0.05). Quantitative analysis revealed that the SFP group achieved over 90% wound area reduction by day 16, significantly outperforming the AMED and NC groups (Figure 5(b)). This accelerated healing effect may be attributed to the high porosity and fibroblast-supportive properties of the silk fibroin-based material, along with its passive moisture retention capability. Although the ASFP group healed slightly less quickly (day 7 to day 21: p = 0.01 to 0.04 < 0.05), it still demonstrated better healing kinetics than the commercial hydrogel alone, probably owing to the stabilizing structure provided by the SFP layer.

Notably, when comparing the indicator results presented by the ASFP and SFP groups in animal models, it is evident that the SFP tends to disintegrate and fragment easily within 0 to 20 days after absorbing wound exudate (Figure 5(c)). This occurs because SFP nonwoven fabrics are biodegradable. During the early stages of wound healing, a large amount of exudate is secreted; however, the SFP nonwoven fabrics cannot absorb such a high volume. Consequently, the nonwoven fabrics were fully saturated with exudate. Once dried, the exudate-saturated nonwoven fabrics become brittle, and the movement of mice causes the already fragile fabrics to break apart completely. It is only after day 20 that the SFP begins to maintain its integrity. By this time, however, the wound exudate has significantly decreased, so fluorescence quenching as an indicator could not be observed on the SFP. In contrast, the ASFP group benefits from the combination of nonwoven fabrics and hydrogel dressing (Figure 5(d)). The hydrogel dressing effectively absorbs exudate without disintegrating, and because it can hold a large amount of exudate, the nonwoven fabrics do not become fully saturated. Therefore, in the ASFP group, the fluorescence indicator function continues to operate steadily. Between days 0 and 20, the fluorescence in ASFP disappears after absorbing exudate. After day 20, as the wound exudate decreases to a level that the hydrogel dressing can fully absorb, the nonwoven fabrics no longer absorb enough moisture to cause fluorescence to disappear, resulting in stable fluorescence from days 20 to 40. These results confirm that SFP-based dressings not only provide visual moisture indications but also improve wound healing, supporting their dual-function utility in advanced wound care.

ASFP dressing promotes epidermal regeneration and neovascularization

Histologic analysis with H&E staining was performed to assess tissue regeneration across the treatment groups (Figure 6(a)–(d)). In the negative control (NC) group (Figure 6(a)), only partial epithelial coverage with minimal fibroblast infiltration and a thin wound bed was observed, indicating delayed re-epithelialization and poor dermal remodeling. The AMED group (Figure 6(b)) exhibited increased cellularity with diffuse fibroblast presence, but the regenerated epidermis remained disorganized, and the dermal structure lacked maturity. In contrast, the SFP-treated group (Figure 6(c)) displayed a continuous stratified epithelium, well-formed granulation tissue, and early signs of hair follicle regeneration, suggesting enhanced epithelial repair and extracellular matrix remodeling. The ASFP group (Figure 6(d)) demonstrated the most advanced healing morphology, with a well-organized multilayered epidermis, dense fibroblast infiltration in the dermis, and adnexal structures such as hair follicles and sebaceous glands, suggesting almost complete skin regeneration. Quantitative analysis of neovascularization (Figure 6(e)) revealed a significantly reduced vascular area in the AMED group (44%) compared to that in the untreated control group (100%), suggesting an impaired angiogenic response. In contrast, both SFP (81.2%) and ASFP (81.5%) restored the neovascular area to levels comparable to those of the control group. Perform t-tests comparing the data from the AMED, ASFP, and SFP groups separately against the NC group. Based on the p-value results, the AMED group (p = 0.03 < 0.05), used as the positive control, showed a significant difference compared to the NC group. This result is primarily attributed to the properties of AMED, which swell and disintegrate after absorbing tissue fluid. Consequently, in frequently active mice, AMED cannot provide stable wound protection, leading to slower and less effective wound healing compared to the other groups (shown in Figure 5). This results in a weakly measured neovascular area in the animal model (Supplemental Material). These differences were statistically significant and confirmed the angiogenesis-supportive effect of SF-based materials during wound healing. Collectively, these histologic findings demonstrate that both SFP and ASFP dressings facilitate epidermal recovery and vascular regeneration, with ASFP providing the most favorable structural and cellular recovery.

Figure 6.

Representative H&E-stained histological sections of wound beds across treatment groups: (a) NC (n = 6), (b) AMED (n = 6), (c) SFP (n = 4), (d) ASFP (n = 4), and (e) quantitative analysis of average neovascular area across treatment groups.

ASFP treatment reduces collagen deposition and fibrotic remodeling

Masson’s trichrome staining was used to evaluate collagen deposition and fibrotic response in the wound bed on day 12 post-treatment (Figure 7(a)–(d)). Excessive and disorganized collagen accumulation was observed in the NC group (Figure 7(a)), which accounted for 14.9% of the wound area. Similarly, the AMED group (Figure 7(b)) exhibited a 16.6% collagen-positive area, indicating persistent fibrotic activity and inadequate remodeling. In contrast, the SFP-treated group (Figure 7(c)) demonstrated significantly reduced collagen content (5.7%) and a better organized dermal architecture. The ASFP group (Figure 7(d)) showed the lowest collagen deposition (4.8%) of all groups, indicating a pronounced anti-fibrotic effect, which was further confirmed by quantitative analysis of collagen porosity (Figure 7(e)). Compared to the NC (12.4%) and AMED (14.1%) groups, both the SFP (9.4%) and ASFP (10.0%) groups exhibited a denser and more compact collagen matrix, suggesting mature, scar-minimized remodeling. These results highlight the ability of ASFP to modulate wound healing by accelerating wound closure, reducing fibrosis, and promoting organized collagen remodeling—a critical feature for functional skin regeneration.

Figure 7.

Masson’s trichrome staining of wound bed tissue: (a) NC (n = 6), (b) AMED (n = 6), (c) SFP (n = 4), (d) ASFP (n = 4), and (e) quantification of collagen porosity demonstrates denser collagen in SFP- and ASFP-treated wounds.

Discussion

This study demonstrated that the moisture-indicating nonwoven dressing composed of SF, PCL, and EA exhibited excellent in vitro biocompatibility, with all four fibroblast cell lines (mouse 3T3, L929; human CG1519, CG1629) showing consistent cell viability. In vivo, Masson’s trichrome and H&E staining revealed enhanced collagen deposition and re-epithelialization in ASFP-treated wounds, suggesting accelerated healing. Moreover, the fluorescence-based indicator responded visibly to moisture accumulation and provided a visual signal under 365 nm UV light that correlated with wound exudate levels. Collectively, these results suggest that the developed dressing not only promotes wound healing but also enables real-time visual monitoring of wound moisture, offering practical benefits for clinical wound management.

The integration of visual moisture-indicating functionality into wound dressings represents a significant advancement in personalized and responsive wound care. In current clinical practice, dressing changes are often performed according to fixed schedules or subjective assessments, which can lead to unnecessary wound disturbances or prolonged exposure to exudates. Both practices have been associated with suboptimal outcomes: extended dressing intervals (>4.5 days) significantly increase postoperative infection risk, ²⁰ while routine changes contribute to higher nursing workload without improving patient recovery. ²¹ Premature removal can disrupt granulation tissue formation, whereas delayed changes can lead to wound maceration and microbial colonization, particularly in surgical wounds such as cesarean section incisions. A Cochrane review reported that excessive exudate can damage periwound tissue, ²² and although delayed dressing removal (e.g., on postoperative day 5) was associated with a slightly higher rate of wound complications, the difference was not statistically significant. ²³ The moisture-indicating capability developed in this study provides a practical, real-time solution for monitoring exudate accumulation. By enabling timely, on-demand dressing changes, this innovation has the potential to reduce iatrogenic trauma, optimize moisture balance, and lower the risk of infection. These are key factors supported by previous evidence of moisture-responsive wound care. ²⁴ Beyond functionality, material safety and biologic integration are critical for clinical translation, particularly in postpartum and obstetric wound care, where tissue regeneration must occur under conditions of low inflammation and low cytotoxicity. The SF-PCL-EA composite combines well-characterized and clinically acceptable biomaterials with complementary structural, biological, and functional properties. This composition not only ensures safety and wound compatibility but also delivers smart responsiveness, and is therefore particularly suitable for low-risk and patient-friendly postpartum surgical wound care.

SF, derived from Bombyx mori, has emerged as a clinically versatile biomaterial owing to its robust biocompatibility, tunable mechanical strength, and ability to support cellular adhesion and proliferation. ²⁵ SF degradation products—amino acids such as glycine, alanine, and serine—are non-toxic and beneficial to surrounding tissue, further supporting its translational relevance, as evidenced by its established use in FDA-approved sutures and tissue scaffolds. ²⁶ Recent studies have shown that SF promotes angiogenesis and accelerates wound closure, particularly by regulating vascular endothelial growth factor (VEGF) expression and favorably interacting with fibroblasts and keratinocytes. ²⁷ In addition, SF has been produced in various formats such as hydrogels, films, electrospun mats, and 3D scaffolds, demonstrating controlled drug release properties and compatibility with stem cell delivery, further broadening its therapeutic versatility.^26–28 Importantly, its immunomodulatory profile is also favorable, with low macrophage activation and minimal fibrotic encapsulation, even in long-term implantation models. ²⁹ This makes silk fibroin a particularly attractive material for chronic wound dressings and tissue regeneration platforms.

In addition to its biological properties, SF possesses unique optical characteristics that are increasingly being researched for biosensing applications. Structural modification through molecular reassembly or chemical functionalization can alter its optical behavior, enabling it to serve as a structural color material or fluorescence-responsive matrix. ³⁰ SF has been processed into photonic films with an inverted opal structure and into submicrometer thin films to measure moisture, determine pH and prevent counterfeiting. ³¹ Moreover, SF and its companion protein sericin exhibit intrinsic fluorescence arising from aromatic amino acids such as tyrosine, phenylalanine, and tryptophan. ³² In the formation of hydrogels, the interactions between protein structure, functional groups, and side-chain environments modulate fluorescence emission. SF-based hydrogels exhibit red-shifted fluorescence peaks in the 300–700 nm range that enable label-free optical sensing in biological systems.^33,34 This autofluorescent behavior, enhanced by β-sheet stacking and changes in solvent polarity, supports the use of SF for non-invasive optical monitoring within tissue-engineered constructs. ³⁵ Importantly, such endogenous fluorescence reduces reliance on external fluorescent dyes, minimizes risk of cytotoxicity and enables long-term use in biological environments. ³⁴ These optical properties position silk fibroin not only as a structural scaffold but also as a functional biosensing platform for smart dressing systems, particularly suited for real-time exudate detection and wound monitoring in sensitive applications such as postpartum surgical care.

To meet the need for structural reinforcement and scalable processing, PCL was incorporated into the composites to improve their mechanical strength and handling stability. PCL is a semi-crystalline aliphatic polyester, first synthesized in the 1930s, that is now FDA-approved for biomedical use owing to its excellent biocompatibility, biodegradability, and rate slow rate of degradation rate, making it particularly suitable for applications involving long-term wound contact. ³⁶ PCL possesses a number of unique physicochemical properties that underpin its versatility. It has a melting point of approximately 60°C and a glass transition temperature of around −60°C, which gives it excellent flexibility and ductility under physiological conditions. Its high crystallinity (50%–70%) contributes to improved tensile strength and abrasion resistance. Tensile strengths of up to 23 MPa and elongation at break ranging from 300% to 500% have been reported.^37,38 These mechanical properties make PCL particularly attractive for load-bearing biomedical applications, such as absorbable sutures, orthopedic scaffolds, and soft-tissue implants. Chemically, PCL demonstrates excellent resistance to numerous common solvents and hydrolytically stable behavior. However, it is susceptible to enzymatic and microbial degradation, particularly by esterases. Its degradation byproducts, primarily ε-caprolactone monomers, are metabolized via natural pathways and excreted safely, which supports its long-term biocompatibility and low-inflammatory degradation profile. ³⁹ The excellent solubility of PCL in organic solvents such as chloroform, ethyl acetate, and dichloromethane also facilitates the fabrication of nanofibers, films, and microspheres through electrospinning and solvent-casting techniques. ⁴⁰ This ease of processing allows for tailored design of porous or multilayered structures, which is critical for moisture-responsive wound dressings. In addition to medical applications, PCL is also used in drug delivery systems, environmentally friendly packaging, and even consumer products such as toys and sports equipment due to its high crystallinity, mechanical resilience, and processing versatility.^{13–15,36,41} These collective features make PCL an ideal component of our smart dressing platform, providing structural integrity, elasticity, and processing compatibility with SF and EA for real-time moisture-responsive wound care.

EA, a naturally occurring polyphenol abundant in fruits, nuts, and medicinal plants, was selected as a moisture-responsive visual indicator for composite dressings. Its molecular structure contains four hydroxyl groups and two lactone rings, which contribute to its potent antioxidant, anti-inflammatory, and antimicrobial effects, offering potential secondary benefits in exudate-rich or infection-prone wound environments. ⁴² Mechanistically, EA can downregulate angiogenic factors such as VEGF and inhibit the activity of matrix metalloproteinases (MMPs), particularly MMP-2, which play a key role in extracellular matrix remodeling and neovascularization.^43–45 The inhibitory effect of EA on nucleoside diphosphate kinase (NDPK) and its ability to chelate metal ions such as Ca²⁺, Mg²⁺, and Zn²⁺ further contribute to its anti-angiogenic and anti-metastatic properties.^46,47 From an optical perspective, EA contains conjugated chromophores that allow it to interact with UV light and exhibit distinct absorption and fluorescence spectra. ⁴⁸ When EA binds with proteins, such as silk fibroin, changes in the electronic configuration of the resulting complex lead to altered UV-visible absorption and the appearance of fluorescence emissions, particularly within the 300–600 nm range. These spectral shifts are attributable to changes in resonance energy transfer and molecular polarity at the protein–polyphenol interface.^16,17 This renders EA not only a therapeutic co-factor but also a functional fluorophore, providing intrinsic indicator capabilities without the need for external dyes or probes.

Despite these promising results, this study has several limitations that should be considered. First, the fluorescence-based moisture indication relies on a 365 nm UV light source for visual activation. Although UV flashlights are inexpensive and widely available, they are not routinely used in clinical practice. This requirement could hinder widespread adoption unless complemented by user-friendly and clinically compatible solutions, such as smartphone-compatible UV attachments or integrated light-emitting diode (LED)-based visualization tools within wound care kits. ⁴⁹ In addition, vulnerable populations, including neonates and immunocompromised patients, may raise concerns about repeated, albeit minimal, UV exposure. Although 365 nm UVA is generally considered low-risk, cumulative or prolonged exposure to artificial UV sources has been flagged by regulatory authorities as a potential risk to vulnerable groups. ⁵⁰ Accordingly, safety validation under clinical conditions is essential prior to widespread implementation.

Second, fluorescence intensity is prone to variability caused by environmental factors such as ambient lighting, viewing angle, and camera sensitivity. These inconsistencies can lead to inter-user and inter-site variations in the interpretation of wound moisture levels. Although this colorimetric shift is readily visible to the naked eye, it remains a major challenge to achieve consistent, quantitative, and reproducible measurements. To overcome this limitation, recent studies have investigated the integration of mobile health applications incorporating calibrated image-processing algorithms to standardize assessments and minimize subjective bias. Notably, Wang et al. ⁵¹ developed a smartphone-based wound evaluation system that utilizes edge detection and tissue classification techniques to accurately quantify wound dimensions and monitor healing progression. Their platform demonstrated the feasibility of real-time user-independent wound assessment, even in outpatient and remote care settings. Incorporating such digital tools into future iterations of moisture-indicating dressings could substantially enhance clinical reliability, facilitate telemedicine applications, and improve patient self-monitoring.

Third, current findings are limited to in vitro cytocompatibility assays and small animal wound models. Although these preclinical results provide promising indications of biocompatibility and wound-healing potential, they do not fully address the complexity of human clinical scenarios. Parameters, such as wound size, depth, exudate composition, patient mobility, and variability in skin physiology, differ markedly between rodents and humans. ⁵² These differences may affect dressing performance, absorption kinetics, and user compliance. Therefore, rigorously designed clinical trials across diverse patient populations, including postoperative, diabetic, and obstetric cohorts, are critical to validate the safety, efficacy, usability, and long-term acceptability of dressings in real-world clinical situations.

Looking ahead, the clinical utility of moisture-indicating dressings can be substantially enhanced through the integration of digital imaging and sensing technologies. Smartphone-based applications incorporating calibrated image analysis algorithms have demonstrated strong potential for real-time wound assessment by providing standardized interpretation of wound characteristics while minimizing inter-user variability. These systems effectively correct for environmental lighting, viewing angle, and device-specific variations, allowing for consistent quantification of colorimetric changes associated with wound exudate accumulation. ⁵³ Building on this framework, the application of hyperspectral imaging and deep learning algorithms improves diagnostic precision in the analysis of wound and skin tissue. ⁵⁴ These advanced approaches can accurately distinguish tissue types and detect early signs of infection or necrosis, supporting their integration into mobile platforms for improved, automated wound monitoring.

In parallel, the integration of biosensor technologies represents a complementary strategy for the further development of smart wound care systems. Gao et al. ⁵⁵ have demonstrated the feasibility of such systems using a textile-based platform that incorporates pH and temperature sensors with wireless data transmission capabilities. Embedding flexible, miniaturized sensors capable of monitoring local physiological parameters, such as pH, temperature, and bacterial load, can provide real-time insights into the wound microenvironment and enable early detection of infection or delayed healing. Recent innovations in hydrogel-embedded electronics and wireless sensor networks have also shown promise for monitoring chronic wounds, underscoring the translational potential of multiplexed sensing platforms.^4,55 When combined with smartphone-based colorimetric analysis, these technologies could transform conventional wound dressings into intelligent, multimodal systems that support continuous, remote, and personalized wound monitoring in both hospital and home care settings.

Conclusion

In this study, we have developed a novel moisture-indicating wound dressing tailored to obstetric applications with a visual cue system that effectively reflects the degree of exudate absorption. The integration of this visual functionality holds significant promise in improving postpartum wound care by facilitating timely dressing changes, reducing user-dependent judgment errors, and potentially lowering the risk of wound infection. Given its favorable biocompatibility and functional performance in preclinical models, this smart dressing represents a practical advancement toward personalized and responsive maternal care. Future studies should focus on clinical validation through well-designed multicenter studies to evaluate safety, ease of use and efficacy in different postpartum patient populations.

Supplemental Material

sj-docx-1-jbf-10.1177_22808000261431501 – Supplemental material for Development and characterization of a smart obstetric dressing with a visual indication function

Supplemental material, sj-docx-1-jbf-10.1177_22808000261431501 for Development and characterization of a smart obstetric dressing with a visual indication function by Chi-Kang Lin, Ting-Teng Wang and Meng-Yi Bai in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-docx-2-jbf-10.1177_22808000261431501 – Supplemental material for Development and characterization of a smart obstetric dressing with a visual indication function

Supplemental material, sj-docx-2-jbf-10.1177_22808000261431501 for Development and characterization of a smart obstetric dressing with a visual indication function by Chi-Kang Lin, Ting-Teng Wang and Meng-Yi Bai in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-docx-3-jbf-10.1177_22808000261431501 – Supplemental material for Development and characterization of a smart obstetric dressing with a visual indication function

Supplemental material, sj-docx-3-jbf-10.1177_22808000261431501 for Development and characterization of a smart obstetric dressing with a visual indication function by Chi-Kang Lin, Ting-Teng Wang and Meng-Yi Bai in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-xlsx-4-jbf-10.1177_22808000261431501 – Supplemental material for Development and characterization of a smart obstetric dressing with a visual indication function

Supplemental material, sj-xlsx-4-jbf-10.1177_22808000261431501 for Development and characterization of a smart obstetric dressing with a visual indication function by Chi-Kang Lin, Ting-Teng Wang and Meng-Yi Bai in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-xlsx-5-jbf-10.1177_22808000261431501 – Supplemental material for Development and characterization of a smart obstetric dressing with a visual indication function

Supplemental material, sj-xlsx-5-jbf-10.1177_22808000261431501 for Development and characterization of a smart obstetric dressing with a visual indication function by Chi-Kang Lin, Ting-Teng Wang and Meng-Yi Bai in Journal of Applied Biomaterials & Functional Materials