A Mechanically Resilient and Tissue-Conformable Hydrogel with Hemostatic and Antibacterial Capabilities for Wound Care

Abstract

Hydrogels are used in wound dressings because of their tissue-like softness and biocompatibility. However, the clinical translation of hydrogels remains challenging because of their long-term stability, water swellability, and poor tissue adhesiveness. Here, tannic acid (TA) is introduced into a double network (DN) hydrogel consisting of poly(vinyl alcohol) (PVA) and poly(acrylic acid) (PAA) to realize a tough, self-healable, nonswellable, conformally tissue-adhesive, hemostatic, and antibacterial hydrogel. The TA within the DN hydrogel forms a dynamic network, enabling rapid self-healing (within 5 min) and offering effective energy dissipation for toughness and viscoelasticity. Furthermore, the hydrophobic moieties of TA provide a water-shielding effect, rendering the hydrogel nonswellable. A simple chemical modification to the hydrogel further strengthens its interfacial adhesion with tissues (shear strength of ≈31 kPa). Interestingly, the TA also can serve as an effective hemostatic (blood-clotting index of 58.40 ± 1.5) and antibacterial component, which are required for a successful wound dressing. The antibacterial effects of the hydrogel are tested against Escherichia coli and Staphylococcus aureus. Finally, the hydrogel is prepared in patch form and applied to a mouse model to test in vivo biocompatibility and hemostatic performances.

Keywords: antibacterial; hemostasis; hydrogels; tissue adhesives; wound dressing.

Figure 1

Multifunctional hydrogel for wound dressing patch. A) Schematic illustrations of a DN hydrogel wound dressing patch and its multifunctionality. B) Fabrication of the developed DN hydrogel. C) FT‐IR spectra of hydrogels. D) Thermogravimetric analyses for hydrogels. E) Gel fraction measurements for each hydrogel sample (n = 3: n is the sample size for each group).

Figure 2

Mechanical durability of the hydrogel. A) Tensile test for the TA/PVA/PAA hydrogels with varying concentrations of TA. B) Tensile test comparing PVA, PVA/PAA, and TA/PVA/PAA. C) Young's moduli of hydrogel samples (n = 3: n is the sample size for each group). D) Cyclic loading‐unloading tests for the TA/PVA/PAA hydrogel (n = 3: n is the sample size for each group). E) Schematic illustration describing the notch‐insensitive property of the hydrogel. F) Strain at break values of the hydrogels with varying notch lengths. G) Schematic illustration and optical photographs showing the self‐healing behavior of the hydrogel. H) Self‐healing efficiencies of the hydrogel as a function of time (n = 3: n is the sample size for each group). Statistical analysis was carried out with an unpaired t‐test (^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001; ns: no significant difference).

Figure 3

Interfacial stability of the hydrogel. A) Schematic illustration depicting nonswellable property of the hydrogel. B) Sequential photographs of the PVA and TA/PVA/PAA hydrogels immersed in water. C) Swelling ratio measurement of each hydrogel sample as a function of time (n = 3: n is the sample size for each group). D) Schematic illustration of the adhesion process of the hydrogel. E) Rheological analysis of the TA/PVA/PAA hydrogel. F) Stress‐relaxation test for each hydrogel sample. G) Lap shear test characterizing the adhesion strength of the hydrogel against porcine skins. H) Shear strength of the hydrogel without‐ and with NHS ester (n = 3: n is the sample size for each group). I) Photographs of the hydrogel conformally attached to a porcine skin (scale bars, 5 mm). J) SEM image of the hydrogel adhered to a porcine skin surface (scale bar, 100 µm).

Figure 4

Biocompatibility and hemostatic effects of the hydrogel. A) Schematic illustration describing the in vitro biocompatibility test. B) Fluorescence microscopy images of live/dead stainings (green: live, red: dead) on NIH 3T3 cells cultured with hydrogels (scale bars, 200 µm). C) Cell viability calculated from the live/dead assay (n = 3: n is the sample size for each group). D) Optical image of hemolysis test on hydrogels (n = 4: n is the sample size for each group). E) Hemolysis ratio of each group. F) Schematic illustration describing the hemostatic effects of the hydrogel. G) Blood clotting index value for each group (n = 4: n is the sample size for each group). H) Photographs of blood clotting on each hydrogel (scale bars, 5 mm). Statistical analyses were conducted with an unpaired t‐test (^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001; ns: no significant difference).

Figure 5

Antibacterial performances of the hydrogel. A) Representative photographs of the hydrogels and bacteria colonies of E. coli and S. Aureus (scale bars, 1 cm). Red dashed lines indicate inhibition zones. B) Diameter of inhibition zones formed near the hydrogels after 12 h of incubation (n = 3: n is the sample size for each group). C) Fluorescent images of live/dead bacteria (green: live, red: dead) of E. coli and S. aureus incubated with the hydrogels (scale bars, 200 µm). D) Quantitative analysis of adherent live and dead bacteria on the samples after 12 h of incubation in a growth medium (10⁵ bacterial mL⁻¹) (n = 3: n is the sample size for each group).

Figure 6

In vivo biocompatibility and hemostatic performances of the TA/PVA/PAA patch. A) Schematics of subcutaneous implantation of TA/PVA/PAA hydrogel into the backs of ICR mice. B) Histological analyses after 7 days of implantation: H&E, nuclei were stained blue, and the cell cytoplasm were stained pink; MT, collagen was stained in blue, nuclei were stained in black, and the cytoplasm was stained in red; F4/80, macrophage marker stained in brown (scale bars, 100 µm). C) Hepatotoxicity evaluation with measurement of serum level of ALT and AST (n = 3: n is the sample size for each group). D) Schematic describing a hemostatic experiment in a mouse model. E) Optical photographs of the N.T., PVA, and TA/PVA/PAA groups in a liver hemorrhage model (scale bars, 5 mm). F) SEM images of PVA and TA/PVA/PAA hydrogel surface after hemostasis experiment (scale bars, 5 µm). G) The accumulated amount of bleeding mass in the filter papers for N.T., PVA, and TA/PVA/PAA groups (n = 4: n is the sample size for each group). H) Optical photographs of the filter papers that absorbed bleeding blood from the damaged liver at the time points of 10 and 120 s (scale bar, 10 mm). I) H&E staining of the liver tissue where hemostasis was induced by the TA/PVA/PAA patch (scale bars, 200 µm). Statistical analysis was performed with an unpaired t‐test (^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001; ns: no significant difference).