Ultra-fast self-gelling self-expanding self-propelling high-adhesion procoagulant hemostatic powder for non-compressible hemorrhage hemostasis in pigs

Huang, Shengfei; Li, Meng; Xu, Huiru; Shu, Tianyu; Jia, Honglin; Li, Zhenlong; Yang, Yutong; Zhao, Xin; Guo, Baolin

doi:10.1038/s41467-026-68683-y

Download PDF

Article
Open access
Published: 29 January 2026

Ultra-fast self-gelling self-expanding self-propelling high-adhesion procoagulant hemostatic powder for non-compressible hemorrhage hemostasis in pigs

Nature Communications volume 17, Article number: 2146 (2026) Cite this article

8089 Accesses
1 Citations
Metrics details

Subjects

Abstract

Non-compressible hemorrhage remains the chief cause of battlefield mortality and civilian trauma death. Here, we propose a hemostatic strategy based on ultrafast self-gelling self-expanding powder of polyacrylic acid (PAA), polyethyleneimine (PEI) and foaming agent, which achieves non-compressible hemorrhage hemostasis through the multiple synergistic effects of expansion plugging, self-gelling adhesion and activating the coagulation factors. The hemostatic powder integrates rapid physical crosslinking with spontaneous gas foaming, exhibits a fast gelation rate, high expansion ratio, strong tissue adhesion, and activation of red blood cells, platelets, and fibrin. The optimized formulation (PP/PT5-TXA30) achieves superior hemostatic performance compared to commercial powders in rat liver volumetric defect, femoral artery and vein transection, as well as complete transection of the subclavian artery and vein in rabbits. Notably, in a lethal porcine model of complete subclavian artery and vein transection, PP/PT5-TXA30 achieves superior performance of non-compressible hemorrhage compared to gauze and XStat™. Additionally, PP/PT5-TXA30 accelerates full-thickness skin wound healing. This work demonstrates a strategy based on ultra-fast self-gelling and self-expanding mechanisms for developing hemostatic materials for non-compressible hemorrhage.

Comparison of a gelatin thrombin versus a modified absorbable polymer as a unique treatment for severe hepatic hemorrhage in swine

Article Open access 27 November 2023

Superporous sponge prepared by secondary network compaction with enhanced permeability and mechanical properties for non-compressible hemostasis in pigs

Article Open access 27 June 2024

Hemorrhagic risk prediction in coronary artery disease patients based on photoplethysmography and machine learning

Article Open access 10 November 2022

Introduction

Uncontrollable hemorrhage—defined as bleeding at amputation sites or junctional areas (e.g., groin) that cannot be managed through manual or mechanical compression due to trauma in battlefield or civilian accidents—is a leading cause of preventable death^1,2,3,4. Such non-compressible hemorrhage represents 90% of potentially survivable fatalities among military personnel and constitutes 30–40% of trauma-related deaths in civilian populations. Consequently, it stands as the leading cause of mortality in emergency care settings^5,6. Traditional hemostatic materials, such as gauze⁷, sponges^8,9, and glues¹⁰, typically require manual or physical pressure to achieve hemostasis, but their efficacy is often limited in these scenarios^11,12. Therefore, the development of hemostatic strategies for non-compressible hemorrhage, aimed at reducing both blood loss and time to hemostasis, is urgently needed to significantly lower trauma-related mortality in both battlefield and civilian emergency settings¹³.

Given that non-compressible hemorrhage typically occurs in deep wounds and internal organs, often involving profuse bleeding accompanied by irregular wound geometries and rupture of deep-seated arteries and veins, manual or physical compression is not feasible, necessitating emergency and pre-hospital hemostatic intervention¹⁴. Researchers have developed a variety of hemostatic materials for non-compressible hemorrhage based on physical, chemical, and biological mechanisms^1,15, including sponges/foaming agents^6,8, sealants/adhesives¹¹, and platelet mimetics¹⁶. Among these, expandable hemostatic materials have been widely recognized as the most promising translational approach due to their wound-adaptive capacity and immediate physical blockage effect, enabling rapid control of non-compressible hemorrhage^17,18. Existing expandable hemostatic materials can be classified based on their expansion mechanisms into the following categories: (1) Shape memory sponge-based expandable hemostatic materials (e.g., XStat™, cryogels). The U.S. military utilizes XStat™, a commercial injectable system loaded with compressed cellulose sponges. These sponges rapidly swell to occupy the cavity and generate localized pressure in deep, non-compressible wound tracts^6,14,19, thereby achieving hemostasis. However, it lacks biochemical procoagulant activity, and its relatively low blood absorption capacity and compressive strength may prolong hemostasis time and increase blood loss. To address these limitations, we previously reported a series of injectable cryogels that achieve effective hemostasis in non-compressible hemorrhage through the synergistic effects of biochemical procoagulation and expansion-induced sealing^{20,21,22,23,24,25}. Shape memory sponge-based hemostatic materials are easy to sterilize and convenient for storage. Nonetheless, they may be unable to access concealed bleeding sites or treat complex wounds with narrow grooves or voids, cannot cope with internal or diffuse bleeding in the abdominal cavity, and excessive expansion may tear the wound, resulting in secondary hemorrhage. To better adapt to wound characteristics and avoid excessive compression of the wound, a second category emerged: (2) Fluid foaming-type expandable hemostatic materials (e.g., ResQFoam, expandable hemostatic hydrogels). Upon injection into the wound site, ResQFoam generates foam through contact with blood, which exerts pressure at the bleeding site to achieve hemostasis. However, the complex administration procedure of ResQFoam, along with its operational intricacies that necessitate trained personnel, poses a significant risk of misadministration, and the risk can lead to severe complications or even death, thereby limiting its application in emergency care^1,19. To simplify the application process, our team engineered an injectable, self-propelling hydrogel adhesive with inherent procoagulant properties and fast-setting capability¹⁷. This hydrogel adhesive exhibits a fast expansion rate, high self-expanding capacity, and strong tissue adhesion, enabling effective control of complex bleeding while minimizing excessive compression on surrounding tissues. Nevertheless, its relatively low mechanical strength and liquid form—characterized by poor sterilization compatibility (low irradiation tolerance), as well as limited storage stability and portability—restrict its practical use in emergency settings. Therefore, retaining the high hemostatic efficiency of fluid foaming-type expandable materials while overcoming the challenges of sterilization and mechanical weakness associated with liquid formulations has become a major bottleneck in the development of hemostatic material design targeting non-compressible hemorrhage.

In this study, we develop an ultra-fast self-gelling, self-expanding, self-propelling, highly adhesive, procoagulant portable hemostatic powder by integrating the moisture-induced self-gelling property of powders with self-expanding and self-propelling capabilities. Dopamine-modified montmorillonite (PDA-MMT) is incorporated into polyethyleneimine/polyacrylic acid powder (PP/PT), which possesses ultra-fast self-gelling and wet adhesion properties, to enhance its adhesive performance and hemostatic efficacy. Furthermore, a gas-generating strategy based on protonated tranexamic acid (TXA-NH₃⁺) and sodium carbonate (Na₂CO₃) is introduced into the self-gelling hemostatic powder system to confer self-expanding and self-propelling behaviors, while synergistically accelerating coagulation. We systematically characterize the blood absorption and gelation capacity, mechanical properties, tissue adhesiveness, antibacterial activity, biocompatibility, and volumetric expansion performance of the hemostatic powder. The hemostatic efficacy in non-compressible hemorrhage is evaluated using rat models of liver volumetric defect, cardiac puncture, and femoral artery/vein transection, as well as lethal rabbit and porcine models involving complete transection of the subclavian artery and vein. In addition, the powder’s ability to achieve sutureless closure of full-thickness skin incisions and promote wound healing is assessed. The results demonstrate that the self-gelling expandable hemostatic powder not only possesses portability, storage stability, and ease of use, but also exhibits high wet tissue adhesion, rapid self-expanding and self-propelling properties, and superior antibacterial and procoagulant activities. It effectively controls bleeding from surface wounds, irregularly shaped deep cavity injuries, and non-visible hemorrhage sites, and further accelerates skin wound healing.

Results

Preparation of ultra-fast self-gelling self-expanding self-propelling high-adhesion hemostatic powder

We prepared a hemostatic powder with ultra-fast self-gelling, self-expanding, and self-propelling properties, which triggers rapid gelation upon contact with blood and exhibits good adhesion and coagulation-promoting properties. This hemostatic powder can be used to control irregular, non-compressible hemorrhage (Fig. 1). First, montmorillonite (MMT) was modified with hydrochloric acid to enhance its dispersion in water, allowing dopamine to self-assemble on the MMT to obtain dopamine-modified MMT (PDA-MMT). Thermogravimetric analysis (TGA) results showed that PDA was successfully coated on MMT, with a dopamine coating ratio of approximately 2.51% (Supplementary Fig. 1a). After PDA coating, the Zeta potential of MMT decreased from −14.3 ± 1.3 mV to −25.6 ± 0.2 mV (Supplementary Fig. 1b). Then, PDA-MMT was incorporated into polyethyleneimine (PEI) and polyacrylic acid (PAA), which exhibit ultra-fast self-gelling and wet adhesion, to enhance the adhesion and hemostatic properties. After freeze-drying and grinding, PEI/PAA/PDA-MMT powder (abbreviated as PP/PT) was obtained. In the preparation of PP/PT powder, the mass ratio of PEI to PAA was maintained at 1:1, and the amount of PDA-MMT in the PP/PT powder was varied to enhance the adhesion strength of PP/PT. The specific feed ratio is shown in Supplementary Table 1. The introduction of catechol groups enhances the adhesion properties of the powder^17,26. However, as the content of PDA-MMT increased, the adhesion strength of the powder initially increased and then decreased. This is due to a change in the cohesion between the polymers^12,27. The PP/PT with 5 wt% PDA-MMT exhibited the optimal adhesion strength (>40 kPa) (Supplementary Fig. 2). Therefore, in subsequent experiments, we used the self-gelling hemostatic powder with 5 wt% PDA-MMT content, named PP/PT5. To verify the successful preparation of the material, we characterized the product using SEM, EDS, and FT-IR spectroscopy. Material morphology was examined via SEM, and the successful synthesis of PDA-MMT and PP/PT was confirmed by EDS (Supplementary Fig. 3 and Supplementary Fig. 4). The molecular structure of the materials was systematically characterized using FT-IR, as shown in Supplementary Fig. 5. In the spectrum of PDA-MMT composite, a significant N–H in-plane bending vibration absorption peak was observed at 1603 cm⁻¹, which confirmed that dopamine was successfully surface-modified onto the montmorillonite matrix. Further comparison of the spectral features revealed that the characteristic absorption peak of the carboxyl group (–COOH) in pure PAA appeared at 1696 cm⁻¹, while the vibration peaks of primary/secondary amines (–NH₂/–NH) in PEI were observed in the 1585 cm⁻¹ region. Notably, when constructing the PP/PT5 composite system, the aforementioned characteristic peaks shifted to 1632 cm⁻¹ and 1541 cm⁻¹, respectively, and no new characteristic peaks were observed. This spectroscopic evidence indicates that the crosslinking interactions between PEI, PAA, and PDA-MMT in the system primarily arise from physical bonding mechanisms such as hydrogen bonding and electrostatic interactions, rather than the formation of covalent bonds^12,27. The crosslinking interactions between PEI, PAA, and PDA-MMT in the system primarily arise from physical bonding mechanisms such as hydrogen bonding and electrostatic interactions, rather than the formation of covalent bonds²⁷. This conclusion provides key spectroscopic evidence for the non-covalent crosslinking nature of the composite material.

**Fig. 1: Schematic illustration of the preparation, application, and hemostatic mechanism of hemostatic powder.**

To enhance the material’s expandable properties, a foaming agent was introduced into the system to obtain PP/PT5-TXAn; the specific feed ratios are provided in Supplementary Table 2. The protonated tranexamic acid (TXA-NH₃⁺) and sodium carbonate (Na₂CO₃) rapidly react to continuously release CO₂ gas bubbles, causing the system to expand and providing self-propelling capabilities. After the foaming reaction, TXA-NH₃⁺ remains in the system as tranexamic acid, which can further promote hemostasis by inhibiting plasminogen activation, thereby enhancing the material’s hemostatic performance^6,28. Next, the content of the foaming agent was optimized by evaluating the mechanical properties, expansion performance, and hemostatic capability of the hemostatic powder to identify the most optimal expandable self-gelling hemostatic powder, PP/PT5-TXA30. The key steps in the preparation of PP/PT5-TXA30 hemostatic powder are shown in Fig. 2a. Experimental verification demonstrated that the self-gelling, expandable, self-propelling hemostatic powder rapidly gelled upon contact with blood, while gas production as a propellant caused the system to expand, physically sealing the wound and promoting coagulation through biochemical synergistic effects. Consequently, the material achieves exceptional hemorrhage control in a lethal, non-compressible pig model.

**Fig. 2: Preparation process, blood absorption capacity, swelling behavior, gelation performance, and mechanical properties of hemostatic powders.**

Blood uptake ratio, swelling ratio, gelation process of the hemostatic powder, and mechanical properties of the hemostatic powder after self-gelling

The liquid absorption capacity of hemostatic materials was quantitatively assessed through systematic blood adsorption performance tests (Fig. 2b). As shown in Fig. 2c, comparative studies demonstrated that the commercial chitosan-based hemostatic powder (Chitosan HP), PEI/PAA (PP) powder, PP/PT5, and their TXA-NH₃⁺ and Na₂CO₃ composite system (PP/PT5-TXA30) all exhibited good blood adsorption characteristics, with adsorption ratios exceeding 240% (w/w) for all samples. Among them, the PP/PT5 group displayed the optimal adsorption performance under equal mass conditions, and a peak blood uptake ratio of 2.9 times was achieved. This extraordinary adsorption effect can be attributed to two key mechanisms: first, the spontaneous gelation effect produced by the polymer network through physical interactions such as hydrogen bonding, which effectively traps and immobilizes blood components²⁷; second, the introduction of PDA-MMT nano-fillers, with their high specific surface area due to the layered structure, generates significant capillary adsorption synergistic enhancement²⁹.

Through quantitative studies on the swelling kinetics of the hydrogel materials, it was found that excessive swelling leads to significant mechanical performance degradation, posing a critical challenge to maintaining the mechanical integrity of hemostatic materials³⁰. As shown in Fig. 2d, the systematic swelling assessment of hemostatic powders with equal mass, formed in 37 °C PBS buffer, Chitosan HP swells rapidly at first but collapses structurally after 180 min due to the absence of a stable crosslinked network, resulting in gel dispersion and loss of structural integrity. In contrast, the PP-based composite systems (including PP, PP/PT5, and PP/PT5-TXA30) exhibit long-term swelling stability, with their swelling curves remaining stable after reaching equilibrium, confirming that the material can maintain stable mechanical properties during the hemostatic process. This controllable swelling behavior and stable network structure hold significant clinical implications for achieving precise hemostasis and preventing secondary bleeding. Moreover, the swelling ratio of the powder-derived hydrogel at different time points in blood was not significantly different from that in PBS buffer (Supplementary Fig. 6). The hydrogel’s low swelling in blood ensures the stability of its mechanical properties during hemostasis and subsequent removal.

The dynamically reversible physical crosslinked network within the PP/PT5 system imparts good intrinsic self-healing properties (Supplementary Fig. 7). When exposed to anticoagulated blood or aqueous media, PP/PT5 particles exhibit unique swelling kinetics: discrete components rapidly absorb the liquid phase to form microgel units (Supplementary Movie 1). Notably, these swollen microgels spontaneously undergo topological entanglement and reorganization at the contact interface. In the absence of exogenous crosslinking agents, a complete three-dimensional hydrogel network can be formed within just 2 s through interface fusion effects. This self-healing process arises from the synergistic dynamic equilibrium of multiple hydrogen bonds and charge interactions between polymer chains. This supramolecular mechanism ensures a rapid bonding response while maintaining a moderate binding energy for structural remodeling²⁷. Through design validation experiments, the dry PP/PT5 powder was pre-shaped into the letter “X” configuration and then activated in the liquid phase, successfully reproducing a structured hydrogel. And the powder was accumulated, added with water to make it self-gelling and stretched for gelation display (Fig. 2e). Moreover, we also quantitatively measured the gelation time of the PP/PT5 powder using a TA rheometer (Supplementary Fig. 8a). As shown in Supplementary Fig. 8b, the PP/PT5 powder self-gelled within 10 s. This instant self-assembly characteristic and morphology controllability provide an essential foundation for expandable non-compressible hemostatic materials development.

Robust mechanical strength is essential for adhesive hydrogels to maintain structural integrity and resist re-rupture under external forces. The mechanical properties of the powder-derived hydrogels were evaluated through rheological analysis, uniaxial compression tests, and tensile strain tests (Fig. 2f–i). Oscillatory frequency sweep measurements confirmed the solid-like viscoelastic behavior of the hydrogels (Fig. 2f), where the storage modulus (G′) consistently exceeded the loss modulus (G″) across the tested frequency range. With increasing concentrations of TXA-NH₃⁺ and Na₂CO₃, both G′ and G″ exhibited progressive declines, with an accelerated rate of reduction observed between PP/PT5-TXA30 and PP/PT5-TXA50, where G′ dropped sharply from 87 kPa to 21 kPa. These results indicate that the hydrogel exhibits greater stiffness at lower foaming agent concentrations, while excessive foaming agent addition significantly reduces strength. Furthermore, uniaxial compression tests were performed on the gelled powders (Fig. 2g). With the gradual increase in foaming agent content, the compressive strength of hydrogels formed from PP/PT5, PP/PT5-TXA10, PP/PT5-TXA20, PP/PT5-TXA30, PP/PT5-TXA40, and PP/PT5-TXA50 progressively decreased. The maximum compressive strengths at 80% strain were 127, 124, 109, 97, 62, and 32 kPa, respectively. This trend is consistent with the rheological results discussed above, in which compressive strength decreases with increasing acid content (Fig. 2h). Subsequently, tensile properties of the hydrogels were evaluated to further probe foaming agent concentration effects on mechanical behavior. The results, including peak stress, peak strain, tensile modulus, and toughness, were visualized using radar charts¹² (Fig. 2i). The findings revealed that as the foaming agent concentration increased, the peak tensile stress and tensile modulus of the hydrogels decreased, whereas their flexibility improved, as evidenced by the increase in peak strain. The hydrogel networks are principally stabilized by physical bonds: hydrogen bonding and electrostatic forces¹². The increased content of foaming agents affects the efficiency of physical interactions, thereby reducing the crosslinking density of the hydrogel network¹⁷. Additionally, increased foaming agent concentration produces more CO₂ bubbles, generating a macroporous hydrogel architecture. These bubbles also absorb mechanical energy, allowing the expandable porous hydrogel to achieve a lowered elastic modulus and elevated fracture strain.

In hemostatic applications, the material must possess sufficient mechanical strength to withstand the impact of arterial blood flow and endure deformations caused by body movements, thereby preventing seal failure and ensuring effective hemostasis. This unique balance of properties arises from the synergistic interaction between the physically crosslinked network and the porous structure, where the former ensures rapid gelation and structural integrity, and the latter enhances fatigue resistance through energy dissipation mechanisms. The results demonstrate that this intelligent hemostatic system enables precise tailoring of mechanical properties by adjusting the foaming agent concentration, offering an innovative strategy for the development of next-generation trauma emergency care materials.

Self-expanding and self-propelling performance of the hemostatic powder

Hemostatic materials with self-expanding and self-propelling properties offer an innovative solution for managing severe hemorrhagic conditions. Upon injection of the hemostatic powder into a bleeding cavity, the material rapidly expands by absorbing blood and generating gas, exerting pressure on surrounding tissues and conforming closely to the wound surface. Simultaneously, it forms a dense barrier during the self-gelling phase. Notably, the expansion-induced driving force enables the active components to autonomously infiltrate perforated regions and deep, narrow wound tracts. Upon contact with blood, in situ gelation is triggered, establishing strong adhesion at the tissue interface. This intelligent material can adaptively fill complex wound geometries, achieving simultaneous spatial occupation and blood coagulation, ultimately enabling efficient sealing of irregular wounds with massive hemorrhage¹⁷.

The self-expanding and self-propelling properties of the injectable hemostatic powder were investigated, including the expansion process, expansion time, and volume expansion ratio. With increasing foaming agent content, the volume expansion ratios of PP/PT5-TXA10, PP/PT5-TXA20, PP/PT5-TXA30, PP/PT5-TXA40, and PP/PT5-TXA50 progressively increased (Supplementary Fig. 9). However, the rate of increase gradually slowed, which may be attributed to the decreased proportion of polymer mass, thereby limiting its ability to effectively encapsulate the generated gas bubbles. Given that non-compressible hemorrhage requires the application of high pressure for effective hemostasis, an ideal hemostatic material must strike a balance between expansion capacity and mechanical strength¹⁷. To assess its blood-triggered expansion performance, PP/PT5-TXA30 was injected into a quantified volume of heparinized whole blood (Fig. 3a). Upon contact with blood, the powder initiated ultra-fast self-gelling and expansion, achieving a 400% expansion ratio within 5 s (Fig. 3b). We also explored its self-propelling performance. Experiments show that it has good propulsion performance (Supplementary Movie 2). These results validate its ultra-fast self-gelling, self-expanding, and self-propelling properties, suggesting its strong potential for rapid hemorrhage control. Comprehensive evaluation revealed that the PP/PT5-TXA30 composite powder, owing to its optimized crosslinked network structure, achieves controlled expansion while maintaining sufficient compressive strength, making it the preferred candidate for this application.

**Fig. 3: Expansion properties of PP/PT5-TXA hemostatic powder.**

The expansion behavior of the self-expanding, self-gelling hemostatic powder within a biomimetic wound cavity was further investigated through simulation using ANSYS Workbench 2020R1. The model replicated realistic wound geometries, with soft tissue defined using hyperelastic material parameters (C10 = 0.347 MPa, C01 = 0.0352 MPa), and incorporated the simulation of carbon dioxide bubble flow (inlet flow rate: 0.05 kg/s). After applying fixed and displacement constraints, the dynamic expansion process was analyzed over a time interval of 0–8 s. The simulation results demonstrated the pressure and stress exerted by the expanding self-gelling hemostatic powder on the cavity walls (Fig. 3c, d). This stress is much lower than the breaking strength of skin (~10 MPa)³¹, blood vessels (~2 MPa), and muscles (~100 kPa)³², and is also lower than the in situ stress of surrounding tissues on nerves (~50 kPa)³³, so it will not cause compressive damage to the surrounding tissues. As shown in Fig. 3e, the changes in volume and density during the expansion process were also modeled. These results indicate that the PP/PT5-TXA30 powder exhibits rapid volumetric expansion capability and generates high extrusion stress. This self-gelling adhesive, characterized by self-expanding, self-propelling, and adaptability to extrusion pressure, is expected to deliver superior in vivo hemostatic performance and to be well-suited for complex, non-compressible hemorrhage sites with irregular geometries.

Figure 3f illustrates the expansion mechanism of the self-gelling hemostatic powder. Upon contact with blood, the PP/PT5-TXA30 powder undergoes a rapid reaction between protonated tranexamic acid (TXA-NH₃⁺) and sodium carbonate, continuously releasing CO₂ bubbles. Owing to the immediate formation of a physically crosslinked polymer network, the generated gas is effectively entrapped within the gel matrix, driving volumetric expansion and establishing a stable three-dimensional structure. In a relatively enclosed wound cavity, the bubble-containing composite hydrogel maintains a sustained expansion state due to network densification. This self-driven expansion, triggered by a chemical reaction, presents an innovative solution for hemorrhage control in internal and limb trauma scenarios.

Adhesive strength and burst pressure of hemostatic powder for soft tissues

Effective hemostatic adhesives must maintain sufficient tissue adhesion and burst resistance to seal injuries and ensure reliable hemorrhage control³⁴. As shown in Fig. 4a, b to evaluate the adhesive performance of the powder, porcine skin was used as a substrate. By varying the order of dye solution application, the powder was triggered to undergo in situ gelation directly on the skin. After gel formation, the skin was gradually rotated from 0° to 180°, followed by rinsing, stretching, bending, twisting, and a second rinsing process. In all cases, the material remained firmly adhered to the tissue surface. Moreover, PP/PT5-TXA30 demonstrated strong adhesion across a variety of organs and tissues, including the lung, intestine, stomach, liver, kidney, spleen, and heart (Fig. 4c). To quantitatively assess the adhesive strength to fresh porcine skin, lap shear tests were conducted (Fig. 4e). The results showed that the adhesive strength of the PP/PT5 powder reached 42.84 kPa, significantly higher than that of PP powder (22.21 kPa), which is primarily attributed to the wet adhesion of catechol groups on PDA at the tissue interface and the interfacial diffusion bonding between the polymer and the tissue (Fig. 4d)³⁵. After introducing the acid–base components TXA-NH₃⁺ and Na₂CO₃, the PP/PT5-TXA30 powder still retained a robust adhesion strength of approximately 34.78 kPa. Benefiting from its tissue adhesion, the sealing capability of the material was further evaluated via a burst pressure test (Fig. 4f). As shown in Fig. 4g, h, the burst pressures of PP, PP/PT5, and PP/PT5-TXA30 powders all significantly exceeded the range of normal human systolic blood pressure (90–140 mmHg). Notably, owing to enhanced mechanical properties, the PP/PT5 (439.3 mmHg) and PP/PT5-TXA30 (421.3 mmHg) powders exhibited significantly higher burst pressures than the PP powder (349.5 mmHg), indicating their ability to withstand hemodynamic impacts. These results collectively demonstrate that the PP/PT5-TXA30 powder can efficiently absorb interfacial moisture and form a physically crosslinked hydrogel in situ. This enables rapid tissue adhesion and wound sealing, highlighting its potential as an effective hemostatic material.

**Fig. 4: Adhesive properties of the hemostatic powder.**

Antibacterial capacity, hemolytic activity, cytocompatibility, and in vivo biocompatibility of the hemostatic powder

Severe wound contamination is often associated with war-related or civilian trauma, and serious infections may lead to fatal consequences^36,37. The antibacterial activity of the hemostatic powders against Escherichia coli (E. coli) and methicillin-resistant Staphylococcus aureus (MRSA) was evaluated in vitro. For the in vitro antibacterial assay, bacteria were incubated with the hemostatic powders for 2 h. The bactericidal ratio of both PP/PT5 and PP/PT5-TXA30 powders against the two bacterial strains reached 100%. No bacterial colonies of E. coli or MRSA were observed on the agar plates, whereas numerous colonies were present in the PBS control group (Supplementary Fig. 10). These results indicate that the hemostatic powders exhibit excellent antibacterial properties, making them promising antibacterial hemostatic materials.

The hemostatic material requires good biocompatibility^38,39. The biocompatibility of the hemostatic powder was tested by hemolysis test (Supplementary Fig. 11), in vitro cytocompatibility test (Supplementary Fig. 12), and in vivo subcutaneous implantation test in rats (Supplementary Fig. 13). Firstly, we evaluated the hemocompatibility of the powder by testing its hemolysis ratio. The experimental results showed that the hemolysis ratio of the PP/PT5-TXA30 hemostatic powder was less than 5%, showing good hemocompatibility. Secondly, the cytocompatibility of the PP/PT5-TXA30 hemostatic powder was evaluated using L929 fibroblasts co-cultured for 24 h with disc-shaped preformed gel powders (20 mg) under a contact mode assay, and the PP/PT5-TXA30 hemostatic powder groups exhibited cell viability above 95% compared to the TCP control group, demonstrating good cell compatibility. Finally, to further evaluate the biocompatibility of the materials, preformed PP/PT5-TXA30 hemostatic powders were subcutaneously implanted into SD rats. H&E staining revealed that PP/PT5-TXA30 hemostatic powders induced relatively mild acute and chronic inflammatory responses at both 7- and 28-day post-implantation. Collectively, these findings demonstrate that the PP/PT5-TXA30 hemostatic powder exhibits good hemocompatibility, in vitro cytocompatibility, and in vivo biocompatibility and has the potential to serve as an effective hemostatic material.

In vitro blood-clotting performance of the hemostatic powder

Through red blood cell and platelet adhesion assays, the in vitro coagulation effect of the hemostatic powder was qualitatively assessed. As shown in Fig. 5a, the number of aggregated red blood cells in the three powder groups (PP, PP/PT5, PP/PT5-TXA30) was significantly higher than that in the three control groups (Gauze, Gelatin sponge, Chitosan HP), and most of them exhibited an activated state with pseudopodia. This is primarily attributed to the efficient liquid absorption and self-expanding properties of the materials, indicating that after contact with blood, the powder can quickly wrap the blood and form a hydrogel/blood clot composite, showing potential for sealing the bleeding site and achieving hemostasis. Furthermore, the catechol groups of PDA interact with red blood cells, leading to greater red blood cell aggregation in the PP/PT5 powder group compared to the PP group^29,39. Additionally, the introduction of positively charged TXA-NH₃⁺ generates electrostatic interactions between the PP/PT5-TXA30 powder and red blood cells, more effectively promoting red blood cell adhesion and aggregation (Fig. 5b), thereby achieving rapid coagulation^6,40. In the coagulation pathway, red blood cells have been shown to induce or enhance platelet activation, leading to platelet plug formation and primary hemostasis. The adhesion and aggregation of activated platelets further induce the involvement of thrombin in fibrin formation and stabilization, thereby completing secondary hemostasis⁴¹. Therefore, we further compared the adhesion and aggregation abilities of platelets and fibrin for Gauze, PP/PT5, and PP/PT5-TXA30 groups (Fig. 5c, d). The results were consistent with the red blood cell adhesion assay¹, where the aggregation of platelets and fibrin in the powder groups was significantly higher than in the gauze group, with the PP/PT5-TXA30 group showing significantly better in vitro coagulation effects than the PP/PT5 group. This can be attributed to the following three points: First, the hemostatic powder absorbs blood and concentrates coagulation factors^{12,38,42,43,44,45,46}. Second, the catechol groups on the dopamine-modified montmorillonite surface further accelerate hemostasis²⁹. Additionally, the introduction of tranexamic acid, an antifibrinolytic drug, helps stabilize fibrin and blood clots, thus accelerating the coagulation cascade⁴⁰.

**Fig. 5: In vitro coagulation properties of the hemostatic powder.**

The coagulation performance of the powders was qualitatively evaluated using plasma recalcification time and dynamic whole-blood coagulation assays. In the plasma recalcification time assay, a shorter time indicates a faster blood coagulation rate. As shown in Fig. 5e, the coagulation time for all powder groups was within 21 s, significantly shorter than the gauze group (36 s). With the introduction of PT, the coagulation time was further reduced to 13.5 s. Moreover, as the content of TXA-NH₃⁺ increased, the coagulation time continued to decrease, demonstrating that the inclusion of foaming components not only provides expansion properties but also accelerates coagulation, leading to enhanced hemostasis.

In dynamic whole-blood coagulation testing, reduced hemoglobin solution absorbance reflects accelerated clot formation. Gelatin sponge, gauze, chitosan HP, PP, and PP/PT5 powders were selected as controls to evaluate the coagulation performance of the PP/PT5-TXA30 hemostatic powder, with an untreated group serving as a blank control. An equal volume of ACD (A: citric acid, C: trisodium citrate, D: dextrose) -treated whole blood was dropped onto each material at 37 °C. After 120 s, water was added to rinse off uncoagulated blood. As shown in Fig. 5f, the blank control group remained dark red, indicating that coagulation did not occur effectively within 120 s. From left to right, the sample color progressively lightened, and the quantitative results shown in Fig. 5g were consistent with the observed visual trends, further confirming the strong procoagulant performance of PP/PT5-TXA30.

Taken together, the hemostatic mechanism mediated by PP/PT5-TXA30 powder can be described as follows (Fig. 5h): upon deposition on a bleeding wound, the powder rapidly absorbs blood, concentrating clotting factors. Simultaneously, it forms a hydrated, adhesive hydrogel in situ, which tightly adheres to the tissue and builds a stable physical barrier capable of withstanding blood pressure. Moreover, the hydrogel derived from PP/PT5-TXA30 powder can enhance blood cell agglomeration, activate platelets, and inhibit plasminogen activation, thereby reducing fibrin degradation and enhancing the overall hemostatic effect. In summary, PP/PT5-TXA30 powder exhibits high hemostatic efficacy and good in vitro coagulation capability.

In vivo hemostatic performance of hemostatic powder on rat liver volumetric defect bleeding and rat femoral artery transection bleeding models

After confirming the good in vitro hemostatic performance and biocompatibility of the hemostatic powders, we first evaluated the ability of PP/PT5-TXA30 hemostatic powder to control non-compressible hemorrhage using a rat liver volumetric defect model (7 mm in diameter, 3 mm in depth) (Fig. 6a). Comparisons were made with commercial chitosan-based hemostatic powder, PP powder, and PP/PT5 powder, with an untreated group serving as the blank control (Fig. 6b). In the absence of any hemostatic treatment, the injured liver resulted in a blood loss of 2515 mg within 6 min, with no evident signs of hemostasis. Upon application of different hemostatic materials, both blood loss and hemostasis time were significantly reduced (p < 0.001) (Fig. 6c, d). Although the chitosan-based hemostatic powder initially absorbed blood, it lacked barrier integrity, dissolving in blood and causing 1035 mg blood loss with high rebleeding risk. The PP hemostatic powder achieved a reduced blood loss of 861 mg and a hemostasis time of 2.7 min, outperforming commercial chitosan powder through superior blood absorption, coagulation factor enrichment, and stable hemostatic barrier formation. Treatment with PP/PT5 further decreased blood loss and hemostasis time to 526 mg and 2 min, respectively, owing to the enhanced adhesive strength and blood absorption capacity imparted by the dopamine-modified montmorillonite. The PP/PT5-TXA30 hemostatic powder achieved ultra-fast hemostasis within 1 min, with a minimal blood loss of 125 mg. This was accomplished by rapidly absorbing large volumes of blood to concentrate and aggregate blood cells and platelets, forming a physical hydrogel barrier with strong adhesion at the wound site. Additionally, its self-expanding and self-propelling properties enabled rapid filling and sealing of the entire cavity wound, thereby achieving fast and effective hemostasis.

**Fig. 6: Liver volumetric defect and femoral artery complete transection non-compressible hemorrhage hemostasis in rats.**

To further verify the hemostatic efficacy of the powders in cases of fatal acute hemorrhage, we employed a rat model involving complete transection of the femoral artery to simulate uncontrollable, non-compressible hemorrhage. The hemostatic powders were accurately delivered to the bleeding site using an injectable hemostatic device (Fig. 6e). As shown in Fig. 6f and Fig. 6g, the trends in blood loss and hemostasis time were consistent with those observed in the rat liver volumetric defect model. Treatment with the commercial chitosan-based hemostatic powder resulted in an average blood loss of 5725 mg and a hemostasis time of 3 min. Compared to the untreated group, which exhibited a blood loss of 7583 mg, the improvement was not statistically significant. This is attributed to the inability of the commercial chitosan-based powder to form an effective barrier under conditions of high blood volume and pressure. The blood loss and hemostasis time for PP powder, PP/PT5 powder, and PP/PT5-TXA30 powder exhibited a progressive improvement. Among these, the PP/PT5-TXA30 hemostatic powder again demonstrated the most effective performance, reducing blood loss to 1743 mg and achieving hemostasis within 1 min. Compared to the commercial chitosan-based powder, this represented a 70% reduction in blood loss and a 67% reduction in hemostasis time, effectively mitigating the risk of excessive blood loss caused by delayed hemostasis during massive hemorrhage. Notably, we also systematically monitored the survival time and calculated the mortality ratio of rats treated with different hemostatic materials (Fig. 6h, i). Without any treatment, all rats succumbed to excessive blood loss within 12 min. In contrast, the PP/PT5-TXA30-treated group exhibited a mortality ratio of only 20%; the sole death occurred at 75 min post-injury, indicating a substantial extension in survival time following a lethal injury. Photographs in Supplementary Fig. 14 are consistent with the quantitative findings.

Next, we demonstrated the hemostatic efficacy of PP/PT5-TXA30 hemostatic powder in vivo for surface acute hemorrhage using rat models of cardiac puncture and tail transection. Cardiac puncture in rats causes excessive bleeding with blood spurting out, and the deposition of PP/PT5-TXA30 powder (200 mg) effectively stops the bleeding within approximately 5 s. No rebleeding occurred after 10 min (Fig. 6j). We also induced acute bleeding by cutting off the rats’ tails. Similarly, PP/PT5-TXA30 powder (100 mg) achieved rapid and effective hemostasis within about 3 s (Supplementary Fig. 15). The results demonstrate that PP/PT5-TXA30 hemostatic powder exhibited exceptionally effective hemostasis for both cardiac puncture and tail transection bleeding. This effectiveness is attributed to the powder’s ability to absorb blood and self-gel, resulting in high adhesion and mechanical properties. The powder maintains high burst pressure, and its integrity is not compromised by blood flow. These results from the rat hemorrhage models confirm that PP/PT5-TXA30 powder is not only effective for massive bleeding in deep, irregular cavities but also highly efficient for surface wound bleeding, without the need for additional manual compression.

In vivo hemostatic performance of hemostatic powder on rabbit subclavian artery and vein complete transection bleeding model

Furthermore, we selected a fatal rabbit subclavian artery and vein complete transection hemorrhage model to evaluate the ability of PP/PT5-TXA30 hemostatic powder to control acute massive hemorrhage (Fig. 7a). The untreated group and commercial chitosan-based hemostatic powder were used as controls. According to the experimental results (Fig. 7b–e), rabbits in the untreated group stopped bleeding around 5 min but died within 7 min, with blood loss reaching up to 55.7 g, resulting in a mortality ratio of 100%. In contrast, the PP/PT5-TXA30 group achieved 100% hemostasis within 1 min (60% of rabbits had no rebleeding after treatment, and 40% achieved hemostasis within 1 min), while the commercial chitosan-based hemostatic powder group only achieved a 20% success ratio within 1 min (p < 0.001). Regarding blood loss, the PP/PT5-TXA30 group had a blood loss of 3.1 ± 1.2 g, which was significantly lower than that of the commercial chitosan-based hemostatic powder control group (29.9 ± 9.8 g), a reduction of 90% (p < 0.001). Moreover, the hemostasis time in the PP/PT5-TXA30 group (0.9 ± 0.5 min) was much shorter than that of the gauze control group (3.9 ± 0.8 min), representing a 78% reduction in hemostasis time (p < 0.001). The photographic results in Fig. 7f were consistent with the quantitative findings. All results and discussions indicate that our developed PP/PT5-TXA30 hemostatic powder demonstrated superior hemostatic performance compared to the commercial chitosan-based hemostatic powder in the fatal rabbit subclavian artery and vein complete transection hemorrhage model under blind conditions. Its strong hemostatic ability is attributed to the synergistic effects of its good blood absorption, swelling properties, and high procoagulant activity, which not only enable rapid sealing of the bleeding site and filling of the wound cavity but also form a strong adhesive seal around the surrounding skin. Thus, PP/PT5-TXA30 hemostatic powder can effectively treat acute severe non-compressible hemorrhage and significantly improve the survival ratio of rabbits.

**Fig. 7: Hemostasis of the non-compressible rabbit subclavian artery and vein after complete transection of bleeding.**

In vivo hemostasis of ultra-fast self-gelling hemostatic powder adhesives on lethal pig non-compressible hemorrhage

This study further employed a fatal porcine subclavian arterial/venous complete transection non-compressible hemorrhage model to validate the in vivo hemostatic efficacy of the developed ultra-fast self-gelling, self-expanding, self-propelling, high-adhesion procoagulant portable hemostatic powder (PP/PT5-TXA30) in large animals. This wound model precisely simulates the pathological characteristics of fatal massive hemorrhage in combat injuries, and such penetrating injuries also have a high incidence and mortality ratio in civilian environments^47,48. As shown in Fig. 8a, the ultra-fast self-gelling, expandable hemostatic powder was applied to the bleeding site. Upon contact with blood, the powder absorbed the blood and self-gelled, while the foaming agent—protonated tranexamic acid (TXA-NH₃⁺) and sodium carbonate (Na₂CO₃)—produced carbon dioxide gas, causing the powder to expand, thus filling the entire wound cavity and further sealing the wound to achieve hemostasis. Gauze packing with manual compression was used as the control group. In a pig with complete transection of the subclavian artery and vein, significant blood loss occurred rapidly (approximately 450–750 mL within 30 s). Without intervention, the pig would die within a few minutes¹⁷. As shown in Fig. 8b, after the removal of manual compression following gauze packing, the wound continued to bleed, whereas the PP/PT5-TXA30 hemostatic powder demonstrated superior hemostatic performance. Compared to the control group, the hemostatic powder exhibited significant advantages in terms of blood loss, hemostasis time, and application time (Fig. 8c–e). Specifically, the blood loss in the hemostatic powder group was significantly reduced to 11.7 mL, a 98.0% reduction compared to the control group (591.7 mL) (p < 0.001). This difference is attributed to the self-gelling characteristic of the hemostatic powder—within 2 s of contact with blood, a dense adhesive layer is formed, and the dual mechanism of physical sealing induced by self-expanding and the coagulation cascade activation mediated by tranexamic acid immediately seals the vascular rupture. In terms of hemostasis time, the hemostatic powder group achieved effective hemostasis in just 1.2 min, a 96.4% reduction compared to the control group (32.97 min) (p < 0.001). This ultra-fast hemostatic ability is due to the material’s self-propelling characteristics: after swelling with blood, the hemostatic powder undergoes directional migration and can completely fill irregular wound cavities. In contrast, traditional gauze packing, lacking the ability for autonomous deformation, is unable to effectively seal deeper parts of the wound tract. Additionally, the application time of the hemostatic powder was significantly reduced to 10.7 s, while the control group required 31.3 s for material packing and an additional 3 min of manual compression (p < 0.01). Therefore, patients can achieve rapid and simple self-rescue, which is crucial during the golden treatment window (<1 min) in battlefield environments. Furthermore, at the end of the 1-h post-hemostasis observation period, the average arterial pressure and blood oxygen saturation in the hemostatic powder group were higher than those in the control group, with a lower fluid volume required for resuscitation (Supplementary Table 3). The survival ratio in the hemostatic powder group was 100%, while two pigs in the control group died within 3 h after the observation period. The photographs in Supplementary Fig. 16, Supplementary Fig. 17, and the movies (Supplementary Movie 3 and Movie 4) were consistent with the quantitative results. Moreover, we enhanced the postoperative removability of the hemostatic powder through the following strategies: (1) the hemostatic powder was combined into a whole through self-gelation, which improved its removal convenience; (2) we screened a high-strength self-gelling system based on high molecular weight PEI/PAA to enhance the mechanical strength of the expanded hydrogel, which further ensured the efficient removal of the hemostatic powder after hemostasis. Based on the above material design, as shown in the Supplementary Fig. 18, the PP/PT5-TXA30 hemostatic powder has demonstrated the ability to be efficiently removed after stopping bleeding in the lethal porcine model of complete subclavian artery and vein transection.

**Fig. 8: Hemostasis of the non-compressible pig subclavian artery and vein after complete transection of bleeding.**

The U.S. military has adopted a lethal porcine subclavian artery and vein transection non-compressible hemorrhage model to assess the hemostatic performance of XStat™. The study indicated that XStat™ achieved a 75% hemostasis ratio within 4 min, with a post-treatment blood loss of 118 ± 308 mL and an application time of 25 ± 5 s^17,47. Compared to XStat™, the results above demonstrate that the ultra-fast self-gelling, self-expanding, self-propelling, and high-adhesion hemostatic powder shows superior hemostatic performance in lethal non-compressible hemorrhage models. The significantly reduced blood loss and shortened hemostasis time indicate that this hemostatic powder can quickly and effectively control massive hemorrhage, thereby improving the survival ratio. This performance enhancement stems from the powder’s self-gelling and expandable properties, which allow it to rapidly fill the wound cavity and exert appropriate pressure to prevent secondary damage. Its high adhesiveness ensures effective hemostasis. Additionally, we further evaluated the stability of the hemostatic powder after electron beam irradiation sterilization and long-term storage (1 year). First, the PP/PT5-TXA30 hemostatic powder was sterilized by electron beam irradiation (Electron Beam Sterilization Device: DZ type 10 MeV/20KW electron linear accelerator; irradiation intensity: 15.0 kGy). We evaluated the self-gelling time, tissue adhesive strength, and expansion properties of the PP/PT5-TXA30 hemostatic powder before and after irradiation sterilization (Supplementary Fig. 19a–d). The results showed that irradiation sterilization had no significant effect on the key properties of the powder-derived hydrogel. Second, as shown in Supplementary Fig. 20a–d, we evaluated the self-gelling time, expansion properties, and tissue adhesive strength of the PP/PT5-TXA30 hemostatic powder before and after one-year of storage. The results showed that one-year storage had no significant effect on the key properties of the powder-derived hydrogel, demonstrating the good storage stability of the PP/PT5-TXA30 hemostatic powder. Overall, the powder’s easy long-term storage, sterilization, portability, rapid onset of action, and the lack of need for manual intervention give it a significant advantage in battlefield and emergency situations, enabling a shortened hemostasis time to secure the critical treatment window on the battlefield. This leads to tactical time gains, improved emergency response efficiency, and reduced risk of complications. Its feasibility for use by individual soldiers enhances its applicability in combat. The rapid and efficient hemostasis it provides also offers a solution for patients suffering from coagulopathies, acidosis, and hypothermia. Overall, the expandable hemostatic powder demonstrates good performance in halting non-compressible hemorrhages, providing an effective solution for battlefield first aid and pre-hospital treatment. We will continue to optimize the performance of this hemostatic powder and actively explore the large-scale production technology of hemostatic powder, striving to promote its clinical translation.

Sutureless wound closure and repair of skin incision in a rat model

After demonstrating the in vivo hemostatic properties of PP/PT5-TXA30 hemostatic powder, we set out to investigate the efficacy of this hemostatic powder for sutureless closure and repair of skin defects. A full-thickness skin incision (2 cm) was made on the dorsum of SD rats, and the incisions were treated with biomedical glue, surgical sutures, and high-adhesion hemostatic powder, respectively, with the untreated wounds serving as controls (Fig. 9a). The incisions treated with adhesive hemostatic powder had better closure effects than those treated with surgical sutures and biomedical glue. In the early stage of closure, due to the activity of SD rats, the incisions without any treatment healed slowly. Although the biomedical glue can keep the incision closed, the hair near the skin wound is reduced, indicating that the skin is damaged. Surgical sutures are effective at securing wounds, but the process is time-consuming (>2 min), and the needle sticks can cause damage to the tissue. In contrast, our high-adhesion powders close instantly in seconds, creating a strong, stable, and fully conformal seal. Throughout the 2-week follow-up period, both the high-adhesion hemostatic powder and surgical sutures healed faster than the biomedical glue and untreated groups (Fig. 9b). To quantitatively evaluate the tissue repair ability of PP/PT5-TXA30 hemostatic powder, the mechanical properties of the healed skin were characterized by tensile testing (Supplementary Fig. 21). The results showed that the skin healed with PP/PT5-TXA30 showed higher tensile strength than the other groups (Fig. 9c). All these results demonstrate the good application of high-adhesion hemostatic powder as a wound closure agent in promoting incision healing.

**Fig. 9: In vivo evaluation of the effect of PP/PT5-TXA30 adhesive powder on wound closure and repair in a full-thickness skin incision model in mice.**

To assess the wound healing process, histological analysis was performed on the skin incisions on days 7 and 14. On day 7, H&E staining results (Fig. 9d) showed that wounds sealed with the high-adhesion hemostatic powder exhibited complete re-epithelialization and closure of the epidermis. In contrast, the untreated control group, the biomedical adhesive-treated group, and the surgical suture-treated group exhibited clear gaps in the incision with incomplete epidermal healing, and showed limited inflammatory infiltration. By day 14, the wounds treated with PP/PT5-TXA30 adhesive powder displayed well-organized epithelial tissue, with more hair follicles compared to the control group. After two weeks of repair, Masson’s trichrome staining revealed that, compared to the control group, the wounds treated with PP/PT5-TXA30 adhesive powder exhibited denser and more organized collagen deposition (Fig. 9e). Inflammation regulation and angiogenesis are key factors in promoting tissue regeneration^49,50. Immunohistochemical staining for tumor necrosis factor-alpha (TNF-α, a pro-inflammatory cytokine) and CD31 (an endothelial cell marker) showed that, compared to the untreated control group, sutures, and biomedical adhesives, PP/PT5-TXA30 adhesive powder induced significantly lower inflammatory responses and higher microvascular density (Fig. 9f–i). All these histological and immunofluorescence results indicate that PP/PT5-TXA30 adhesive powder can effectively accelerate wound closure and tissue regeneration by promoting the transition from the inflammatory to the proliferative phase, enhancing collagen deposition, and improving vascularization.

Discussion

This study presents an innovative hemostatic approach by introducing a gas-foaming strategy to develop a rapid self-gelling, self-expanding, self-propelling, and coagulation-promoting hemostatic powder. The hemostatic powder utilizes an in situ gas generation expansion mechanism that allows adaptive filling of the wound cavity upon contact with the wound site, while simultaneously triggering polymer crosslinking to form a hydrogel barrier. Experimental results demonstrate that the optimized PP/PT5-TXA30 hemostatic powder exhibits ultra-fast response characteristics (forming a hydrogel within 2 s, expanding its volume threefold within 5 s), strong and tough interfacial adhesion (pig skin adhesion strength of 35 kPa), high burst pressure (400 mmHg), and synergistic coagulation effects (enriched red blood cells and platelet activation). In complex wound hemorrhage models, the powder shows significant advantages: compared to commercial chitosan-based hemostatic powders, it demonstrates superior in vivo hemostatic ability in rat femoral artery bleeding models (70% reduction in blood loss, 67% reduction in hemostasis time), rat hepatic cylindrical defect models (88% reduction in blood loss, 62% reduction in hemostasis time), and rabbit subclavian artery and vein transection non-compressible hemorrhage models (90% reduction in blood loss, 75% reduction in hemostasis time). Importantly, this hemostatic powder effectively controls bleeding in the pig subclavian artery and vein complete transection model (compared to the medical gauze and 3-min manual pressure group, blood loss is reduced by 98%, and hemostasis time is shortened by 96%). This hemostatic powder can be directly injected into deep bleeding sites, making it suitable for treating puncture wounds and complex cavity injuries where manual compression is not feasible. Its portable and easy-to-use characteristics provide a more flexible and reliable solution for emergency care. Additionally, PP/PT5-TXA30 accelerates the healing of full-thickness skin incisions. The strategy we propose not only resolves the inherent contradiction between the conformability of solid materials and the stability of liquid formulations but also provides an innovative solution for battlefield trauma care that enables individual self-rescue. Therefore, developing self-propelling, portable hemostatic materials with rapid expansion and strong procoagulant properties holds significant value in military medicine and promising potential for clinical translation.

Methods

Ethical statement

All animal experiments were conducted in compliance with all relevant ethical regulations and were approved by “The Biomedical Ethics Committee of Health Science Center” of Xi’an Jiaotong University under protocol number 2021-1720. As the hemostatic efficacy evaluated in this study is a fundamental physiological response not known to vary with the sex of the experimental animals, it was not considered as a variable in the study design or analysis. This rationale has been explicitly stated in both the “Methods” and the “Reporting summary” sections.

Materials

Polyethyleneimine (PEI, molecular weight ≈70,000 Da, 50 wt% aqueous solution) and tranexamic acid (TXA-NH₂) were purchased from Aladdin (Shanghai, China). Polyacrylic acid (PAA, molecular weight ≈240,000 Da, 25 wt% aqueous solution) was obtained from J&K Scientific (Beijing, China). Sodium carbonate (Na₂CO₃), Montmorillonite (MMT), and dopamine (DA) were purchased from Macklin. All reagents were used as received without further purification. Antibody TNF-a (ab 1793) was purchased from commercial sources (Abcam, UK), and antibody CD31 (AF6191) was purchased from commercial sources (Affinity, UK). L929 (BNCC100314) cell line was purchased from Shangcheng Beinachuanglian Biotechnology Co., LTD.

Synthesis of dopamine-modified montmorillonite (PDA-MMT, PT)

Dopamine-modified montmorillonite (PDA-MMT, abbreviated as PT) was prepared according to an established protocol²⁹. Briefly, 5 g of montmorillonite was treated with 100 mL of 3 M HCl under stirring at 45 °C for 8 h. The mixture was subsequently centrifuged and washed repeatedly (5–6 cycles) until the supernatant tested negative for chloride ions using silver nitrate. The resulting solid was collected and dried overnight at 50 °C. In parallel, a Tris buffer solution (150 mL) was adjusted to pH 8.5 with HCl, followed by the addition of 1.5 g of dopamine under stirring until complete dissolution. The acid-treated montmorillonite was then introduced into this dopamine solution and stirred at room temperature for 24 h. After reaction, the product was isolated by centrifugation, washed thoroughly with water (3–4 times) until the supernatant became clear, and finally lyophilized for further use.

Synthesis of organic acid protonated tranexamic acid (TXA-NH₃ ⁺)

The preparation of protonated tranexamic acid (TXA-NH₃⁺) using organic acids was carried out according to protocols described in earlier studies⁴⁰. Concentrated HCl was added to 0.5 M TXA-NH₂ until the pH reached 4.3, and then solid TXA-NH₃⁺ was collected after freeze-drying.

Preparation of polyethyleneimine/polyacrylic acid/dopamine-modified montmorillonite (PEI/PAA/PT, PP/PTn) powder

Dopamine-modified montmorillonite (PT) was mixed with polyethyleneimine (PEI, 10 wt%) and polyacrylic acid (PAA, 10 wt%) at various ratios, followed by lyophilization and grinding to obtain PP/PTn powders. Specifically, we added PT to an equal volume of 10 wt% PEI (Mn = 70,000) aqueous solution and 10 wt% PAA (Mn = 240,000) aqueous solution at 1%, 3%, 5%, 10%, and 20% by mass and mixed them uniformly. The mixture was immediately immersed in liquid nitrogen for about 15 min without pouring out any liquid, and then freeze-dried to remove water. Finally, the dried solid was ground with a mortar and pestle to obtain PP/PTn powder, and the detailed content of each component is shown in Supplementary Table 1. In addition, we mixed 10 wt% PEI (Mn = 70,000) and 10 wt% PAA (Mn = 240,000) in a volume ratio of 5:5. Immerse in liquid nitrogen, freeze-dry, and grind to obtain PP powder.

Preparation of self-gelling self-expanding high-adhesion procoagulant hemostatic PP/PT5-TXAn powder

Optimized Polyethyleneimine/polyacrylic acid/dopamine-modified montmorillonite (PP/PT5), protonated tranexamic acid (TXA-NH₃⁺), and sodium carbonate (Na₂CO₃) were mixed in different ratios, followed by drying and storage to obtain the PP/PT5-TXAn hemostatic powder. TXA-NH₃⁺ and Na₂CO₃ act as foaming agents in the system, generating CO₂ gas upon contact with liquids. PP/PT5, TXA-NH₃⁺, and Na₂CO₃ were mixed in the proportions shown in Supplementary Table 2 to obtain a fast self-gelling, self-expanding, high-adhesion hemostatic powder PP/PT5-TXAn.

Morphology, elemental composition, and FT-IR spectra of the powder

The morphology of freeze-dried MMT, PT, PEI/PAA (PP), and PP/PT powders was observed by using a field emission scanning electron microscope (FEI Quanta FEG 250). The surface of the MMT, PT, PEI/PAA (PP), and PP/PT powders was sprayed with gold before observation. In addition, the elemental compositions of MMT, PT, and PP/PT5 were analyzed using energy dispersive X-ray spectroscopy (EDS) equipped in the SEM by mapping measurements.

FT-IR spectra of PT, PEI, PP/PT, and PAA were recorded in the range of 4000–400 cm⁻¹ by using a Nicolet 6700 FT-IR spectrometer (Thermo Scientific Instrument).

Blood uptake ratio test

Chitosan HP, PP, PP/PT5, and PP/PT5-TXA30 powders were weighed and recorded as m₀, and then immersed in anticoagulated blood water at 37 °C for 10 s, respectively. Then, the samples were dried, and their weight were recorded as m₁. Blood uptake ratio was calculated as follows:

$${\rm{Blood}}\; {\rm{uptake}}\; {\rm{ratio}}=({{\rm{m}}}_{1}-{{\rm{m}}}_{0})/{{\rm{m}}}_{0}\times 100\%$$

(1)

Swelling ratio test

The powders were immersed in PBS at room temperature to transform into hydrogels. And then, immersed in PBS buffer (0.01 M, pH = 7.4) and placed on a shaker at 37 °C. The shaking speed was set as 100 rpm. When reaching the preset time interval, the hydrogel was taken out of the bottle, and its surface water was removed using filter paper. Then the hydrogel was weighed. Measurements were repeated three times for each group. The swelling ratio is calculated using the following formula:

$${\rm{Swelling}}\; {\rm{ratio}}\,(\%)=({\rm{M}}-{\rm{m}})/{\rm{m}}\times 100\%$$

(2)

where M and m represent the wet weight after reaching the equilibrium swelling state and the initial weight of the samples, respectively.

Mechanical properties tests

The rheological property of the powders expanding to the maximum volume was performed using a TA rheometer (DHR-2), and the testing data was collected by TRIOS Installer (version 5.1.1.46572) provided by TA Instruments. Before the test, the powder was mixed with PBS at room temperature to form the hydrogel. The expanded hydrogels were cut into discs with dimeter of 20 mm and a thickness of 1.5 mm. Before the collection of the data, the hydrogel disc was placed between 20 mm parallel plates. Oscillation-frequency test with 1% constant strain and the frequency varying from 1 to 100 rad/s at 37 °C was used to evaluate the stiffness of the expanded hydrogels.

The compression strength of the expanded hydrogels was tested by using a TA rheometer (DHR-2), and the testing data was collected by TRIOS Installer (version 5.1.1.46572) provided by TA Instruments. Before the test, the powder was mixed with PBS into a cylindrical tube with a diameter of 10 mm at room temperature to form the hydrogel. After being expanded to the maximum volume, the hydrogel was taken out and cut into a cylindrical shape with a diameter of 10 mm and a height of 10 mm, and then placed between parallel plates with a diameter of 20 mm and a gap of 10 mm. The maximum compression strain was set to 80%, and the compression speed was set to 100 μm/s.

Tensile tests were performed on the dumbbell-shaped samples with the custom clamps of the Instron Materials Test system (MTS Criterion 43, MTS Criterion) at a cross-head speed of 5 mm/min, and the testing data was collected by Testworks 4 (version 4.12 D) provided by MTS Systems Corporation (MTS). The tensile strain was defined as the change in length relative to the initial gauge length. Measurements were repeated three times for each group.

Volumetric expansion test

1 g of PP/PT5-TXAn with varying foaming agent content was injected into an equal volume of heparinized whole blood to study its blood absorption, self-gelling, and expansion properties. The volumetric expansion performance of powders was determined by calculating the volume expansion ratio and recording the volume expansion time¹⁷. In brief, 2 mL of blood was put into a container, and then 1 g of powder was mixed with the blood. The end of the mixing was regarded as the start time. The timing was stopped when the hydrogel expanded to the maximum volume. The initial volume, volume after expansion, and volume expansion time of the hydrogels were recorded. Each powder was tested in triplicate. The calculation formulas of the volume expansion ratio are as follows:

$${\rm{Expansion}}\; {\rm{ratio}}\,(\%)=V{\rm{t}}/{Vs}\times 100\%$$

(3)

Where Vt and Vs represent the preset time volume and initial volume, respectively.

Simulation calculation

A fluid-structure coupling module in ANSYS Workbench 2020R1 was used to simulate the expansion process of the hydrogel precursor⁴¹. A rotational geometry model of the wound cavity, with a shape similar to an actual wound, was established. The component outside of the wound cavity was assigned as soft tissues with the Mooney-Rivlin 2 model, with the material constant C10 = 0.347 MPa, C01 = 0.0352 MPa, the density of 1048 kg m⁻³, and the incompressibility parameter D1 = 0.00134 MPa⁻¹. A mass-flow inlet with a normal mass-flow rate of 0.05 kg/s was used at the bottom boundary of the wound cavity section to simulate the continuously generated bubbles. The laminar flow module was employed. The fluid materials were set as carbon dioxide with 1.7878 kg m⁻³ and 1.37 × 10⁻⁵kg·m⁻¹·s⁻¹. The large deflection was available to simulate the geometric nonlinear characteristics of the soft tissue. A fixed constraint was added at the inlet of the cavity, while a displacement constraint was added on the upper and lower surfaces of the soft tissue. We focused on the changes from 0 s to 8 s, with a step size of 0.5 s.

Tissue adhesion test

The tissue adhesion strength of the powders was evaluated by using porcine skin tissue via a lap shear test. The porcine skin tissue was cut into 10 mm × 30 mm rectangular shapes. Then, the powder was put on the porcine skin, and PBS buffer was added in situ at room temperature to form a hydrogel. The contact area of the two tissues was kept at 10 mm × 10 mm. Then the samples were placed at 37 °C for 1 h under the condition of maintaining high humidity. Then, the samples were lap shear-tested to failure on an Instron Materials Test system (MTS Criterion 43, MTS Criterion) equipped with a 50 N load cell by using a cross-head speed of 5 mm/min at room temperature, and the testing data was collected by Testworks 4 (version 4.12 D) provided by MTS Systems Corporation (MTS). All measurements were repeated six times for all groups.

Bursting pressure test

In order to test the performance of the powder as tissue sealant, the burst pressure test was conducted by referring to the previous work with some modifications³⁹. In brief, the fresh porcine skin tissues were cut into rectangles (20 × 20 mm), and then a 2-mm-diameter hole was punched in the center of the skin tissue. After that, the skin tissue was fixed on a hole (diameter: 5 mm) in the upper surface of a custom-made cubic box, and the powder was put onto the center of the skin tissue, and PBS was added in situ at room temperature to form the hydrogel. A digital pressure gauge (MIK-Y290, Asmik, China) and a syringe pump (20 mL) were connected to the cubic box, and the cubic box and the tubes were filled with PBS before the test. During the test, PBS was injected into the apparatus gently, and the burst pressure of the whole process was recorded. The test was repeated for 3 times.

Antibacterial activity of the powder

The in vitro antibacterial activity of hydrogels against Methicillin-Resistant Staphylococcus Aureus (MRSA) and Escherichia Coli (E. coli) was evaluated using the surface contact antibacterial test by referring to the reported method with modifications^24,49,51. 50 mg powders were placed into 48-well plates. After 60 min of UV radiation, 10 μL of bacterial suspension (10⁷ CFU/mL) was added onto the surface of the powders, and the inoculated hydrogels were incubated at 37 °C for 2 h. After the incubation, 990 µL of PBS was added to the well to resuspend any bacterial survivors. 10 μL of bacterial suspension (10⁷ CFU/mL) suspended in 990 µL of PBS served as the negative control. After incubating for 18–24 h at 37 °C, the colony-forming units on the agar plate were counted. Each sample was repeated three times. The bacterial killing ratio of the powder is calculated using the following formulas:

$${\rm{Bacterial}}\; {\rm{killing}}\; {\rm{ratio}}\,(\%)=({\rm{Nc}}-{\rm{Np}})/{\rm{Nc}}\times 100\%$$

(4)

Where Nc and Np represent the count of bacteria in the control group and the count of bacteria in the powder group, respectively.

Hemocompatibility, cytotoxicity, and in vivo host response of self-gelling hemostatic powder

The biosafety of PP/PT5-TXAn powders was evaluated by hemocompatibility, cytocompatibility, and in vivo biocompatibility tests.

The hemolytic activity of the powders on blood cells was evaluated by evaluating the hemolysis ratio of the rat blood cells, as we reported previously^21,25,48,52. The rat blood was centrifuged at 200 × g for 10 min to obtain blood cells. After removing the supernatant, the resulting blood cells were washed with PBS buffer and then diluted to a final concentration of 5% (v/v) blood cell suspension. 0.5 mL of powder after gelling was placed into a 24-well plate, and then 600 µL of PBS buffer and 600 µL of blood cell suspension were added. The 24-well plate was incubated at 37 °C with a shaking speed of 100 rpm for 1 h. After that, the blood cell suspension was transferred to a tube and centrifuged at 200 x g for 10 min. 100 μL of the supernatant was added to the 96-well plate, and the absorbance of the solution was read at 540 nm using a microplate reader. 0.1% of Triton X-100 was set as a positive control, while PBS buffer was set as the negative control. Each group was repeated 6 times. The hemolysis ratio was calculated using the following formula:

$${\rm{Hemolysis}}\; {\rm{ratio}}\,(\%)=[({{\rm{A}}}_{{\rm{c}}}-{{\rm{A}}}_{{\rm{b}}})/({{\rm{A}}}_{{\rm{t}}}-{{\rm{A}}}_{{\rm{b}}})]\times 100\%$$

(5)

Where A_c, A_t, and A_b represent the absorbance values of samples, positive control, and PBS, respectively. The cytotoxicity of the powders was evaluated by a contact test method^39,53. For the contacting test method, the freeze-dried hydrogels were first cut into disks with a diameter of 8 mm and a height of 1 mm, and then sterilized by immersing them into 75% alcohol. The complete growth medium was Dulbecco’s modified Eagle medium (DMEM) (Gibco) supplemented with 10% fetal bovine serum (Gibco), 1.0 × 10⁵U/L penicillin (Hyclone), and 100 mg/L streptomycin (Hyclone). L929 cells were seeded into a 48-well plate with a cell density of 7500 cells/well. After being cultivated for one day, the sample disks equilibrated in the culture medium were added into each well, and the culture medium level was kept slightly lower than the sample’s upper surface to ensure the fine contact between the cells and the samples. After culturing for 24 h, the cell viability under the powders was tested by Live/Dead Viability/Cytotoxicity Kit assay. After culturing for 1 day, respectively, the cell viability under the powders was tested by alamarBlue® assay. 20 µL of alamarBlue® reagent in 200 µL culture medium was added to each well after removing the sample disks and medium. After incubating at 37 °C for 4 h, 100 µL of the medium in each well was transferred to a 96-well black plate (Costar). Fluorescence was read by a microplate reader (Ex. 560 nm; Em. 600 nm). Each group was repeated 6 times. Cell adhesion and viability were observed under an inverted fluorescence microscope (IX53, Olympus). The cell viability was calculated using the following formula:

$${\rm{Cell\; viability}}\,(\%)=[({{\rm{A}}}_{{\rm{e}}}-{{\rm{A}}}_{0})/({{\rm{A}}}_{{\rm{p}}}-{{\rm{A}}}_{0})]\times 100\%$$

(6)

Where A_e is the fluorescence value of the experimental group, A_p is the fluorescence value of the positive control group, and A_o is the fluorescence value of the blank background (10 µL of alamarBlue® in 90 µL of complete growth medium).

The in vivo biocompatibility of the powder was verified by a subcutaneous implantation test¹⁷. All the samples were cut into the same shape and size (diameter of 8 mm and height of 2 mm), sterilized with 75% ethanol, and rinsed in phosphate-buffered saline (PBS) overnight. Four symmetrical small incisions were created on the back of each rat (female, Sprague Dawley, 180–210 g). The prepared samples were placed into the incisions, and then the skin was closed. After 7 days and 28 days, the rats were sacrificed, and the testing samples were excised with the adjacent tissues. The obtained material-tissue samples were embedded in paraffin, sectioned, and mounted onto slides. The acute inflammatory response and chronic inflammatory response were measured by both hematoxylin and eosin (H&E) staining. The stained slides were observed and analyzed by a microscope. Each group was repeated 3 times.

Dynamic whole-blood-clotting test

The whole-blood-clotting ability of the powders was evaluated by referring to the reference with some modifications²³. Gelatin sponge, Combat Gauze, and commercial powder were set as the control groups. The powders or control groups were placed into polypropylene tubes and then prewarmed to 37 °C. 10 μL of 0.1 M CaCl₂ solution was mixed with 190 μL of citrated rat whole blood, and then 50 μL of the recalcified whole blood was immediately added to the powder or control surface. The tubes were incubated at 37 °C for 30 s, 60 s, 90 s, and 120 s, respectively. After the incubation, the blood cells that were not trapped in the clot were hemolyzed with 10 mL of DI water, and the absorbance of the resulting hemoglobin solution was measured at 540 nm. The absorbance of 50 μL of recalcified whole blood in 10 mL DI water was used as the reference value (positive control). Each sample was repeated six times. The blood-clotting index (BCI) was calculated using the following equation:

$${\rm{BCI}}\,(\%)=({{\rm{I}}}_{{\rm{s}}}-{{\rm{I}}}_{{\rm{o}}})/({\rm{I}_{n}}-{{\rm{I}}}_{{\rm{o}}})\times 100\%$$

(7)

where I_s, I_n, and I_o represent the absorbance of samples, positive control, and DI water, respectively.

Blood cell adhesion and platelet aggregation test

The blood cell adhesion and platelet aggregation tests were conducted referring to references^17,20,23. In brief, put the powder, gelatin sponge, and Combat Gauze (same weight) into 24-well plates. For the blood cell adhesion test, the samples were immersed in PBS for 1 h at 37 °C. Following that, the whole blood with ACD anticoagulant solution (volume ratio: 6:1) was added dropwise onto the sample and then incubated for 5 min at 37 °C. For the platelet aggregation test, platelet-rich plasma (PRP) was separated from the whole blood with ACD anticoagulant solution by centrifugation of blood at 500 × g for 15 min. The PRP was then added dropwise onto the sample and incubated for 1 h at 37 °C. All samples were then washed with PBS solution three times to remove the physically adhered blood cells, and then fixed with 2.5% glutaraldehyde for another 2 h. After that, blood cells were dehydrated using 50%, 60%, 70%, 80%, 90%, and 100% ethanol solutions with a time interval of 10 min. Finally, the samples were dried and observed using SEM.

Plasma recalcification time test

Platelet-rich plasma (PRP) was separated from the whole blood with ACD anticoagulant solution by centrifugation of blood at 500 × g for 15 min. 10 mg of different powder samples were weighed in a glass bottle, then 90 μL PRP and 10 mL CaCl₂ solution were added to it, and the coagulation time was recorded as PRT. Each sample was repeated three times⁵⁴.

In vivo hemostatic performance of self-gelling hemostatic powder

Hemostatic performance in non-compressible hemorrhage was analyzed using rat liver volumetric defect and complete femoral artery laceration hemorrhage models^10,17,20,39. The hemostatic effect of the material on surface non-compressible hemorrhage was assessed in a rat cardiac puncture hemorrhage model and a tail amputation hemorrhage model^{12,17,24,38,55}.

Rat liver volumetric defect bleeding model

For the rat liver non-compressible hemorrhage model, the rats (SD rats, weight 250–300 g, male) were randomly and equally divided into 5 groups. The animals were anesthetized and fixed on a surgical corkboard. The liver of the rat was exposed by abdominal incision, and serous fluid around the liver was carefully removed to prevent inaccuracies in the estimation of the blood weight obtained by the hemostatic samples. Then, a 6 mm wide and 3 mm deep narrow wound was created. Immediately after wiping off the blood using gauze, the Chitosan HP, PP powder, PP/PT5 powder, and PP/PT5-TXA30 powder were applied onto the site of the lesion. The group without hemostatic materials was used as the blank control group. The hemostatic time and blood loss were recorded. Each group contains 5 rats.

Rat femoral artery and vein complete transection bleeding model

For the rat femoral artery complete transection bleeding model, the rats (SD rat, 250–300 g, male) were anesthetized and fixed on a surgical corkboard. A midline skin incision approximately 1–1.5 cm long was made adjacent to the femoral vessels, and the femoral artery was then exposed. The femoral artery of the rats was cut off with surgical scissors, and the bleeding blood was quickly wiped using gauze. The commercial powder (Chitosan HP), PP powder, PP/PT5 powder, and PP/PT5-TXA30 powder were immediately applied to the bleeding site, respectively. During the hemostasis, the blood loss, hemostatic time, and dead time were recorded. No treatment after bleeding served as the control group. Each group contains 5 rats.

Rat heart puncture hemostasis model

For the rat heart puncture hemostasis model, the rats (SD rat, 250–300 g, male) were anesthetized and fixed on a surgical corkboard. A midline skin incision of about 2–3 cm long was made near the chest, and then the heart was exposed. The rat heart was punctured with a needle, then the commercial powder (Chitosan HP) and PP/PT5-TXA30 powder were immediately applied to the bleeding site, respectively. During the hemostasis, the blood loss, hemostatic time, and dead time were recorded. No treatment after bleeding served as the control group.

Rat tail truncation model

For the rat tail truncation model, the rats (SD rat, 250–300 g, male) were anesthetized and fixed on a surgical corkboard. Cut off with scissors about a third of the length of the tail, then treated immediately with the commercial powder (Chitosan HP) and PP/PT5-TXA30 powder, respectively. During the hemostasis, the blood loss and hemostatic time were recorded. No treatment after bleeding served as the control group.

Hemostatic effect of self-gelling hemostatic powder on non-compressible hemorrhage from complete transection of the rabbit subclavian artery/vein

Following previously reported methods²⁵, the hemostatic effect of the powder was evaluated on non-compressible hemorrhage by performing hemostasis on fully transected rabbit subclavian arteries and veins. The hemostatic powder was syringe-loaded for injection, with an untreated group and a commercial chitosan-based hemostatic powder group as controls. New Zealand White rabbits (weight of about 2 kg, male) were anesthetized and fixed on the surgical corkboard. A cut-down was performed to expose the right femoral artery and femoral vein. Subsequent transection of the femoral artery and vein using the scissors resulted in a hemorrhage model with an invisible bleeding site. The commercial powder (Chitosan HP) and PP/PT5-TXA30 powder were injected into the bleeding site through a skin incision. No treatment after bleeding served as the control group. During the hemostatic process, the weighed gauze was used to absorb the flowing blood. The hemostatic time and blood loss were recorded accordingly. Any additional blood shed from the wound site was recorded as blood loss. If there is no blood flow from the wound site, the hemostasis time is 0 min. Each group contains 5 rabbits.

Hemostatic effect of self-gelling hemostatic powder on non-compressible hemorrhage from complete transection of pig subclavian artery/vein

The efficiency of the hemostatic powder in non-compressible hemorrhage was evaluated by complete transection of the subclavian artery and vein in pig^17,25,56. Male Cross-bred Yorkshire pigs (n = 3 per group) were fasted for ≥12 h (water permitted) prior to surgery. Anesthesia was induced via ear vein injection of 3% pentobarbital (1 mL/kg). Following intubation, the animals were positioned dorsally on a surgical table. The right femoral artery and vein were cannulated to allow continuous monitoring of mean arterial pressure (MAP) and for fluid administration, respectively. Core temperature was maintained at 37–39 °C using a rectal probe and heating pad.

The pigs were abducted, with the left front leg to allow surgical access to the subclavian vessels. A 4.5 cm incision was made parallel to the sternum, over the cranial superficial pectoralis muscle. After dividing the pectoral muscles, the subclavian artery, vein, and brachial nerve plexus were exposed, and a 5 cm section of the subclavian artery and vein was then dissected free from surrounding tissues. The potential wound cavity volume was pre-measured by saline filling. Topical 2% lidocaine was applied to induce vasodilation. A 10-min stabilization period commenced, requiring a stable MAP > 65 mmHg, core temperature of 37–39 °C, and an arterial diameter >6 mm. At the midaxillary line, the subclavian artery, vein, and nerve plexus were completely transected with surgical scissors. A 30-s free-bleeding period was allowed. Pretreatment blood loss was calculated as the sum of (1) blood aspirated during free bleeding and initial application, and (2) the pre-measured cavity volume (accounting for pooled blood). Animals were randomized to receive either the injectable hemostatic powder or standard gauze immediately after free bleeding. For the hemostatic powder group, the material was applied to fill the cavity without subsequent manual compression. In the gauze group, the cavity was packed followed by 3 min of mild manual pressure. The time to fill the cavity was recorded as the application time. MAP was recorded at application and at 15-min intervals thereafter. Resuscitation with prewarmed Lactated Ringer’s solution (37 °C) was administered to maintain MAP ≥ 65 mmHg. 1 h observation period followed the 5 min application window. Five minutes after the start of the 1 h observation, the wound site was inspected for signs of bleeding. Any additional blood shed from the wound site was collected and recorded as post-treatment blood loss. The 1 h observation phase commenced immediately following the completion of manual compression (for gauze) or upon injection of the hemostatic powder. Prospectively identified primary endpoints include hemostasis (defined as no blood exiting the wound cavity) at 4 min, hemostasis at 60 min, and survival at 60 min (survival defined by MAP > 60 mmHg). The secondary endpoints, defined a priori, comprised the mean arterial pressure (MAP) recorded at the study’s conclusion, total blood loss measured following intervention, and the time required to administer the treatment.

Wound healing effect of the self-gelling hemostatic powder on full-thickness skin incisions on the back of rats

The wound healing experiment was conducted by referring to our previous report⁵⁰. The animal experiments were approved by the institutional review board of Xi’an Jiaotong University. All rats were randomly and equally divided into 4 groups: Tegaderm^TM film, glue, suture, and powder groups. Each group contained 6 rats. The rats were anesthetized, and then the dorsal region of the mouse above the tail but below the back was shaved for further surgery. Two full-thickness wounds with a length of 20 mm were made on each side of the mouse midline. Subsequently, different groups of samples were applied to the wound site. For wound area monitoring, on the 0th, 3rd, 5th, 7th, and 14th day, the rats were anesthetized, and then the wound area was photographed. And the tensile strength of the wound was tested by the stretching machine.

To evaluate the epidermal regeneration and inflammation in the wound area, the collected samples on the 7th and 14th day were fixed with 4% paraformaldehyde for 1 h, embedded in paraffin, and then cross-sectioned into 4 μm thickness slices. The obtained slices were then stained with Hematoxylin-Eosin (Beyotime, China) and Masson. All slices were analyzed and photographed by a microscope (IX53, Olympus, Japan). Meanwhile, TNF-α immunofluorescence staining was performed to observe the inflammatory response of the wounds in each group. The angiogenesis of wounds in each group was observed by CD31 immunofluorescence staining.

Statistical analysis

Results express continuous variables as mean ± standard deviation (SD). Following ≥3 experimental replicates per condition. Statistical analyses included Two-tailed Student’s t-tests for two-group comparisons and One-way ANOVA plus Tukey’s post-hoc testing for multi-group analyses (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). Statistical significance was set at p < 0.05. OriginPro (version 2025b) was used to analyze all the statistical data; Images were analyzed by ImageJ (version 1.53 k).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data supporting the findings of this study are available within the article and its supplementary files. Any additional requests for information can be directed to and will be fulfilled by the corresponding authors. Source data are provided with this paper.

References

Dong, R. N., Zhang, H. L. & Guo, B. L. Emerging hemostatic materials for non-compressible hemorrhage control. Natl. Sci. Rev. 9, nwac162 (2022).
Article CAS PubMed PubMed Central Google Scholar
Guo, B. L., Dong, R. N., Liang, Y. P. & Li, M. Haemostatic materials for wound healing applications. Nat. Rev. Chem. 5, 773–791 (2021).
Article CAS PubMed Google Scholar
Hickman, D. A., Pawlowski, C. L., Sekhon, U. D. S., Marks, J. & Gupta, A. S. Biomaterials and advanced technologies for hemostatic management of bleeding. Adv. Mater. 30, 1700859 (2018).
Article Google Scholar
Zhang, J. et al. In situ ultrafast self-gelling coacervate powder with antibacterial, antioxidant, and robust wet adhesion properties for hemostasis and wound healing. Adv. Funct. Mater. 35, 2502577 (2025).
Article CAS Google Scholar
Feng, Y. et al. A sprayable hydrogel based on biomimetic polypeptide-modified lipid nanoparticles for treating non-compressible hemorrhaging. Adv. Mater. 37, 2500908 (2025).
Article CAS Google Scholar
Li, Q. et al. Self-propelling janus particles for hemostasis in perforating and irregular wounds with massive hemorrhage. Adv. Funct. Mater. 30, 2004153 (2020).
Article CAS Google Scholar
He, H. Y. et al. Efficient, biosafe and tissue adhesive hemostatic cotton gauze with controlled balance of hydrophilicity and hydrophobicity. Nat. Commun. 13, 552 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Du, X. C. et al. Microchannelled alkylated chitosan sponge to treat noncompressible hemorrhages and facilitate wound healing. Nat. Commun. 12, 4733 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Jiang, T. S. et al. Superporous sponge prepared by secondary network compaction with enhanced permeability and mechanical properties for non-compressible hemostasis in pigs. Nat. Commun. 15, 5460 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Yu, L. et al. Sequential-crosslinking fibrin glue for rapid and reinforced hemostasis. Adv. Sci. 11, 2308171 (2024).
Article CAS Google Scholar
Bao, G. et al. Liquid-infused microstructured bioadhesives halt non-compressible hemorrhage. Nat. Commun. 13, 5035 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Peng, X. et al. Ultrafast self-gelling and wet adhesive powder for acute hemostasis and wound healing. Adv. Funct. Mater. 31, 2102583 (2021).
Article CAS Google Scholar
Chen, Q. et al. Nanocomposite reinforced powder with fast self-gelling and robust wet adhesive performances for acute hemostasis and wound repair. Small 21, 2501830 (2025).
Article CAS Google Scholar
Chen, S., Carlson, M. A., Zhang, Y. S., Hu, Y. & Xie, J. Fabrication of injectable and superelastic nanofiber rectangle matrices (“peanuts”) and their potential applications in hemostasis. Biomaterials 179, 46–59 (2018).
Article CAS PubMed PubMed Central Google Scholar
Jiang, S., Liu, S., Lau, S. & Li, J. Hemostatic biomaterials to halt non-compressible hemorrhage. J. Mater. Chem. B 10, 7239–7259 (2022).
Article CAS PubMed Google Scholar
Zeng, Y. et al. A platelet-substitute-releasing supramolecular material for cellular assembly mediated on-demand hemostasis. Adv. Funct. Mater. 35, 2422686 (2025).
Article CAS Google Scholar
Zhao, X. et al. Injectable self-expanding/self-propelling hydrogel adhesive with procoagulant activity and rapid gelation for lethal massive hemorrhage management. Adv. Mater. 36, 2308701 (2024).
Article CAS Google Scholar
Wang, Y. et al. Cellulose fibers-reinforced self-expanding porous composite with multiple hemostatic efficacy and shape adaptability for uncontrollable massive hemorrhage treatment. Bioact. Mater. 6, 2089–2104 (2021).
MathSciNet CAS PubMed PubMed Central Google Scholar
van Oostendorp, S. E., Tan, E. C. T. H. & Geeraedts, L. M. G. Prehospital control of life-threatening truncal and junctional haemorrhage is the ultimate challenge in optimizing trauma care; a review of treatment options and their applicability in the civilian trauma setting. Scand. J. Trauma Resusc. Emerg. Med. 24, 110 (2016).
Article PubMed PubMed Central Google Scholar
Zhao, X., Guo, B. L., Wu, H., Liang, Y. & Ma, P. X. Injectable antibacterial conductive nanocomposite cryogels with rapid shape recovery for noncompressible hemorrhage and wound healing. Nat. Commun. 9, 2784 (2018).
Article ADS PubMed PubMed Central Google Scholar
Zhao, X. et al. Injectable dry cryogels with excellent blood-sucking expansion and blood clotting to cease hemorrhage for lethal deep-wounds, coagulopathy and tissue regeneration. Chem. Eng. J. 403, 126329 (2021).
Article CAS Google Scholar
Huang, Y., Bai, L., Yang, Y. T., Yin, Z. & Guo, B. Biodegradable gelatin/silver nanoparticle composite cryogel with excellent antibacterial and antibiofilm activity and hemostasis for Pseudomonas aeruginosa-infected burn wound healing. J. Colloid Interface Sci. 608, 2278–2289 (2022).
Article ADS CAS PubMed Google Scholar
Wang, J. X. et al. Hemostatic, antibacterial, conductive and vascular regenerative integrated cryogel for accelerating the whole wound healing process. Chem. Eng. J. 479, 147577 (2024).
Article CAS Google Scholar
Wang, J. et al. All-in-one: a multifunctional composite biomimetic cryogel for coagulation disorder hemostasis and infected diabetic wound healing. NanoMicro Lett. 17, 171 (2025).
ADS PubMed PubMed Central Google Scholar
Huang, Y. et al. Degradable gelatin-based IPN cryogel hemostat for rapidly stopping deep noncompressible hemorrhage and simultaneously improving wound healing. Chem. Mater. 32, 6595–6610 (2020).
Article CAS Google Scholar
Dreyer, D. R., Miller, D. J., Freeman, B. D., Paul, D. R. & Bielawski, C. W. Perspectives on poly(dopamine). Chem. Sci. 4, 3796–3802 (2013).
Article CAS Google Scholar
Peng, X. et al. Ultrafast self-gelling powder mediates robust wet adhesion to promote healing of gastrointestinal perforations. Sci. Adv. 7, eabe8739 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Cau, M. F. et al. Percutaneous delivery of self-propelling hemostatic powder for managing non-compressible abdominal hemorrhage: A proof-of-concept study in swine. Injury 53, 1603–1609 (2022).
Article PubMed Google Scholar
Cao, S. J., Yang, Y. F., Zhang, S. K., Liu, K. & Chen, J. Multifunctional dopamine modification of green antibacterial hemostatic sponge. Mater. Sci. Eng. C. 127, 112227 (2021).
Article CAS Google Scholar
Gong, Y. et al. Exudate absorbing and antimicrobial hydrogel integrated with multifunctional curcumin-loaded magnesium polyphenol network for facilitating burn wound healing. ACS Nano 17, 22355–22370 (2023).
Article CAS PubMed Google Scholar
Yang, W. et al. On the tear resistance of skin. Nat. Commun. 6, 6649 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Nakao, G. et al. Breaking points of human hamstring muscle–tendon complex: A cadaveric study. J. Mech. Behav. Biomed. Mater. 172, 107180 (2025).
Article PubMed Google Scholar
Chen, L.-E. et al. The functional recovery of peripheral nerves following defined acute crush injuries. J. Orthop. Res. 10, 657–664 (1992).
Article CAS PubMed Google Scholar
Yang, Y. et al. An injectable hydrogel with ultrahigh burst pressure and innate antibacterial activity for emergency hemostasis and wound repair. Adv. Mater. 36, 2404811 (2024).
Article CAS Google Scholar
Lynge, M. E., van der Westen, R., Postma, A. & Städler, B. Polydopamine—a nature-inspired polymer coating for biomedical science. Nanoscale 3, 4916–4928 (2011).
Article ADS CAS PubMed Google Scholar
Zhang, Y. et al. An injectable antibacterial chitosan-based cryogel with high absorbency and rapid shape recovery for noncompressible hemorrhage and wound healing. Biomaterials 285, 121546 (2022).
Article CAS PubMed Google Scholar
Blyth, D. M., Yun, H. C., Tribble, D. R. & Murray, C. K. Lessons of war: Combat-related injury infections during the Vietnam War and Operation Iraqi and enduring freedom. J. Trauma Acute Care Surg. 79, S227–S235 (2015).
Article PubMed PubMed Central Google Scholar
Lei, X.-X. et al. A self-assembly pro-coagulant powder capable of rapid gelling transformation and wet adhesion for the efficient control of non-compressible hemorrhage. Adv. Sci. 11, 2306289 (2024).
Article CAS Google Scholar
Shan, Y. L. et al. Procoagulant, antibacterial and antioxidant high-strength porous hydrogel adhesives in situ formed via self-gelling hemostatic microsheets for emergency hemostasis and wound repair. Biomaterials 315, 122936 (2025).
Article CAS PubMed Google Scholar
Baylis, J. R. et al. Self-propelled particles that transport cargo through flowing blood and halt hemorrhage. Sci. Adv. 1, e1500379 (2015).
Article ADS PubMed PubMed Central Google Scholar
Shu, T. et al. Nanoscaled titanium oxide layer provokes quick osseointegration on 3D-printed dental implants: a domino effect induced by hydrophilic surface. ACS Nano 18, 783–797 (2024).
Article CAS PubMed Google Scholar
Shang, L. L. et al. Hydro-sensitive, in situ ultrafast physical self-gelatinizing, and red blood cells strengthened hemostatic adhesive powder with antibiosis and immunoregulation for wound repair. Adv. Sci. 11, 2306528 (2024).
Article CAS Google Scholar
Zhou, H. et al. Ultrafast self-gelling, adhesive, anti-bacterial coacervate-based powders for enhanced hemostasis and wound healing. Small 21, 2409164 (2025).
Article CAS Google Scholar
Cui, C. Y. et al. Water-triggered hyperbranched polymer universal adhesives: From strong underwater adhesion to rapid sealing hemostasis. Adv. Mater. 31, 1905761 (2019).
Article CAS Google Scholar
Zhao, Q. et al. Underwater contact adhesion and microarchitecture in polyelectrolyte complexes actuated by solvent exchange. Nat. Mater. 15, 407–412 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhang, K. W. et al. Gelable and adhesive powder for lethal non-compressible hemorrhage control. Adv. Funct. Mater. 33, 2305222 (2023).
Article CAS Google Scholar
Mueller, G. R. et al. A novel sponge-based wound stasis dressing to treat lethal noncompressible hemorrhage. J. Trauma Acute Care Surg. 73, S134–S139 (2012).
Article CAS PubMed Google Scholar
Huang, Y. et al. High-strength anti-bacterial composite cryogel for lethal noncompressible hemorrhage hemostasis: Synergistic physical hemostasis and chemical hemostasis. Chem. Eng. J. 427, 131977 (2022).
Article CAS Google Scholar
Yang, Y., Li, M., Pan, G., Chen, J. & Guo, B. Multiple stimuli-responsive nanozyme-based cryogels with controlled NO release as self-adaptive wound dressing for infected wound healing. Adv. Funct. Mater. 33, 2214089 (2023).
Article CAS Google Scholar
Liang, Y., Xu, H., Li, Z., Zhangji, A. & Guo, B. Bioinspired injectable self-healing hydrogel sealant with fault-tolerant and repeated thermo-responsive adhesion for sutureless post-wound-closure and wound healing. NanoMicro Lett. 14, 185 (2022).
ADS CAS PubMed PubMed Central Google Scholar
Yang, Y. et al. Glucose-activated programmed hydrogel with self-switchable enzyme-like activity for infected diabetic wound self-adaptive treatment. Adv. Mater. 37, 2419158 (2025).
Article CAS Google Scholar
Huang, Y., Mu, L., Zhao, X., Han, Y. & Guo, B. Bacterial growth-induced tobramycin smart release self-healing hydrogel for Pseudomonas aeruginosa-infected burn wound healing. ACS Nano 16, 13022–13036 (2022).
Article CAS PubMed Google Scholar
Yang, Y. et al. Bacteria-responsive programmed self-activating antibacterial hydrogel to remodel regeneration microenvironment for infected wound healing. Natl. Sci. Rev. 11, nwae044 (2024).
Article CAS PubMed PubMed Central Google Scholar
Yao, K. et al. Superwettable calcium ion exchanged carboxymethyl cellulose powder with self-gelation, tissue adhesion and bioabsorption for effective hemorrhage control. Chem. Eng. J. 481, 148770 (2024).
Article CAS Google Scholar
Gong, Y. et al. A self-gelling hemostatic powder boosting radiotherapy-elicited NK cell immunity to combat postoperative hepatocellular carcinoma relapse. Biomaterials 317, 123068 (2025).
Article CAS PubMed Google Scholar
Fang, Y. et al. Blood-triggered self-sealing and tissue adhesive hemostatic nanofabric. Nat. Commun. 16, 4910 (2025).
Article ADS CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China: No. 52403203 (M.L.), No.52273149 (B.L.G.) and No.52573176 (B.L.G.), the Youth Project of Shaanxi Provincial Science and Technology Plan: No. 2024JC-YBQN-0407 (M.L.), the Natural Science Basic Research Program of Shaanxi: No. 2025JC-QYXQ-016 (B.L.G.), General Project of China Postdoctoral Fund: No. 2023M732753 (M.L.), the World-Class Universities (Disciplines) and Characteristic Development Guidance Funds for the Central Universities, and Fundamental Research Funds for the Central Universities.

Author information

These authors contributed equally: Shengfei Huang, Meng Li.

Authors and Affiliations

State Key Laboratory for Mechanical Behavior of Materials, and Frontier Institute of Science and Technology, Xi’an Jiaotong University, Xi’an, China
Shengfei Huang, Huiru Xu, Zhenlong Li, Yutong Yang, Xin Zhao & Baolin Guo
Interdisciplinary Research Center of Frontier science and technology, Xi’an Jiaotong University, Xi’an, China
Shengfei Huang & Baolin Guo
Key Laboratory of Shaanxi Province for Craniofacial Precision Medicine Research, College of Stomatology, Xi’an Jiaotong University, Xi’an, China
Meng Li, Tianyu Shu & Honglin Jia

Authors

Shengfei Huang
View author publications
Search author on:PubMed Google Scholar
Meng Li
View author publications
Search author on:PubMed Google Scholar
Huiru Xu
View author publications
Search author on:PubMed Google Scholar
Tianyu Shu
View author publications
Search author on:PubMed Google Scholar
Honglin Jia
View author publications
Search author on:PubMed Google Scholar
Zhenlong Li
View author publications
Search author on:PubMed Google Scholar
Yutong Yang
View author publications
Search author on:PubMed Google Scholar
Xin Zhao
View author publications
Search author on:PubMed Google Scholar
Baolin Guo
View author publications
Search author on:PubMed Google Scholar

Contributions

M.L. and B.L.G. conceived the idea. S.F.H., M.L., and B.L.G. designed experiments and wrote the manuscript. S.F.H. and M.L. synthesized the hemostatic powder, completed polymer characterization, and in vitro evaluations with the help of H.L.J., Z.L.L., and Y.T.Y. S.F.H., H.R.X., T.Y.S., and X.Z. completed the in vivo testing of materials. S.F.H., M.L., and B.L.G. analyzed the results.

Corresponding author

Correspondence to Baolin Guo.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Lianyong Wang, Liming Bian, and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supporting information (download PDF )

Description of Additional Supplementary Files (download PDF )

Supplementary Movie 1 (download MP4 )

Supplementary Movie 2 (download MP4 )

Supplementary Movie 3 (download MP4 )

Supplementary Movie 4 (download MP4 )

Reporting Summary (download PDF )

Transparent Peer Review file (download PDF )

Source data

Source data (download XLSX )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Huang, S., Li, M., Xu, H. et al. Ultra-fast self-gelling self-expanding self-propelling high-adhesion procoagulant hemostatic powder for non-compressible hemorrhage hemostasis in pigs. Nat Commun 17, 2146 (2026). https://doi.org/10.1038/s41467-026-68683-y

Download citation

Received: 05 September 2025
Accepted: 12 January 2026
Published: 29 January 2026
Version of record: 03 March 2026
DOI: https://doi.org/10.1038/s41467-026-68683-y