GI bleeding is a major cause of morbidity and mortality worldwide resulting in a hospitalization rate of 21 per 100,000 adults, and a mortality rate of 2% to 15% [23]. There are various causes of GI bleeding, including hemorrhoids, peptic ulcers, tears, inflammation, colonic polyps, or cancer in the colon, stomach, or esophagus [24]. Endoscopic intervention is the gold standard for GI hemostasis and wound healing. Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are well-established interventions to remove early and middle-stage tumors from the gastrointestinal tract and their use reduces the incidence and mortality of GI cancer [25]. However, adverse events arise from endoscopic resection including intraprocedural bleeding, delayed bleeding, delayed perforation, hemorrhaging, and sepsis [25,26,27], with delayed bleeding being the most common and affecting 3–12% of patients [28]. Clinically, electrocoagulation, argon plasma coagulation, and mechanical clip placement are used to control bleeding and close the mucosal defect post EMR or ESD [29]. Although these techniques are effective in achieving acute hemostasis in about 90% of cases [30], there is limited evidence supporting their efficacy in preventing delayed bleeding and delayed perforation [26]. Additionally, electrocoagulation carries the risk of thermal injury, perfusion, and post-polypectomy coagulation syndrome [27]. Complete clip closure of the wound can be technically challenging depending on the bleeding site and endoscopist experience [31], or even impossible in 40% of cases due to the large wound size or poor accessibility [25].

In the last decade, several topical hemostats have been developed for surgical use, including formulations based on oxidized cellulose, gelatin, collagen, fibrin and thrombin, hyaluronic acid, and cyanoacrylates. All of these hemostats possess unique strengths, but none are ideal due to their weak mechanical properties, cytotoxicity, long procedural time, quick degradation, and/or risk of pathogen infection [30,32]. Recently, topical hemostatic powders such as Hemospray (TC-325), EndoClot, and Ankaferd Blood Stopper have been used to control bleeding [33]. The immediate control rate is in the range of 88–100%, with a rebleeding rate of 3–13% [33]. These powders do not precisely cover the target legion and lack bio-adhesion to the GI mucosa, thereby giving rise to the risk of re-bleeding. Additionally, existing powders are opaque, which can block the surgeon’s view of the operative field and hinder the surgical procedure [31].

An ideal device for GI hemostasis requires: (1) rapid and long-term hemostatic effects; (2) easy use with minimal endoscopist skills; (3) sufficient adhesion and mechanical strength in a wet tissue environment; (4) anti-bacterial and anti-infection properties; and (5) biocompatibility. Hydrogel-based GI hemostats adhere and provide a mechanical barrier to promote rapid hemostasis. Due to the nature of endoscopic intervention, one prefers sprayable and injectable in situ forming hydrogels for ease of operation. Fast gelation of the hydrogel (gelation time < 5 min) is the number one design criterion for rapid hemostasis. Additionally, the hydrogel should withstand the burst pressure of arterial bleeding (burst pressure > 120 mmHg) and the mechanical motility of the GI tract with strong and stable wet adhesion capabilities that surpass that of the conventional surgical glues and match the GI epithelial turnover rate (adhesive strength > 10 kPa on wet tissue for more than 48 h). These requirements compel a multi-functional design of the hydrogel network and its adhesion mechanisms.

Most hydrogels exhibit weak mechanical properties compared to GI tissues which possess a storage modulus (G′) ranging from 100 Pa to 10 kPa, with the proximal colon being the strongest [34]. PEG-based hydrogels containing crosslinked symmetrical tetra-arm macromonomers, of the same size, form homogenous structures with substantial mechanical strength comparable to natural cartilage [35]. Therefore, many studies report the use of multi-arm PEG as the base material due to its biocompatibility, ease of manufacturing, and high mechanical strength [27,36,37,38,39]. For example, a tetra-arm PEG-amine (NH₂) gels with a NHS functionalized PEG (NHS-PEG-NHS) in 100 s [27]; a tetra-arm PEG with end-capped thiols (SH) gels with a tetra-arm PEG-maleimide through a thiol-ene reaction in 4 min [36]; a tetra-arm PEG-NH₂ and a tetra-arm PEG-aldehyde (CHO) gel through Schiff base reactions in 25 s [37]. All three hydrogel formulations exhibit high storage moduli of over 5 kPa, good adhesive strengths of around 10 kPa, comparable to commercially available adhesives, and short in situ gelation time within 5 min, enabling rapid hemostasis [40]. However, their crosslinking mechanisms may not be optimal for the gastric environment (pH = 1.0–2.5). At this low pH, the amino groups will protonate and be inactive, and the Schiff base quickly hydrolyzes [37], leading to the fast disintegration of the hydrogel. In contrast, the pH ranges from 6.6–7.5 in the lower GI organs including the small intestine and colon [41]. Therefore, these reaction chemistries and formulations are appropriate for lower GI hemostasis and wound healing. However, a study using tetra-arm PEG hydrogels in a rat gastric wound model shows that only 16% of hydrogel remains at 48 h after application, which is deemed inadequate [36]. The weak adhesion mechanism, which relies on hydrogen bonding between the polymer and the hydrophilic groups on the tissue, is likely the cause. GI motility and peristalsis as well as epithelial cell turnover create a challenging environment for hydrogel adhesion.

Anchoring chemical adhesion moieties such as NHS and CHO that react with tissue amines will enhance adhesive strength. Adopting other crosslinking mechanisms may also improve the integrity and acidic stability of the hydrogel. Hydrogels prepared by the free-radical polymerization of acryloyl-6-aminocaproic acid (AA) and AA-NHS crosslinked by methylene bisacrylamide (BIS) exhibit comparable storage modulus and adhesive strength as the PEG gels. Notably, they also show good injectability, autonomous self-healing capacity, and stable adhesive behavior in gastric conditions, which facilitates good hemostatic performance by stopping acute bleeding and preventing delayed bleeding in a swine gastric hemorrhage model [42]. The hydrogels accelerate gastric wound healing by controlling inflammation, suppressing fibrosis, and promoting ECM remodeling and angiogenesis [42]. However, the complete gelation procedure requires 9 min, with an initial 3-min pre-polymerization step outside the body, adding complexity to clinical procedures and thus limiting its translational potential [42].

To enhance wet adhesion and hydrogel integrity in the GI environment, researchers install additional functional groups within hydrogels, including those inspired by nature. Mussels adhere strongly to diverse wet substrates via excreted proteins. The key component of mussel adhesion protein is 3,4-dihydroxyphenylalanine (L-DOPA). The mechanism of adhesion encompasses both physical interactions, such as hydrogen bonding, metal chelation, and π–π bonding, as well as chemical interactions, such as oxidation of polyphenols, Schiff base, or Michael addition reactions with the wet tissue surface [8]. A variety of polyphenols, including dopamine, tannic acid, and catechol (Cat), are under investigation to enhance wet adhesion. Xia et al. report a hybrid hydrogel comprised of hyaluronic acid (HA) modified with catechol (Cat) or thiourea (NCSN) and tetra-arm PEG with end-capped thiols (SH) for the treatment of acute upper GI hemorrhage [30]. The dual intertwined networks of HA and PEG endow the hydrogel with a G′ of over 8 kPa and shortens the in situ gelation time to less than 5 s after oxidation is induced for catechol-thiourea coupling and disulfide bond crosslinking. Due to the anchorage of Cat moieties on the tissue, the adhesion strength reaches 14 kPa and the burst pressure is 140 mmHg, higher than the normal maximum arterial pressure of 120 mmHg. Complete hemostasis occurs within 2 min of application and the hydrogel remains adherent for more than 48 h in a pig hemorrhage model, demonstrating rapid and long-term hemostatic effects of the hydrogel [30].

Another barrier to wet adhesion and sealing hemostasis is interfacial water, which is present as a film on tissue surfaces. To address this challenge, researchers incorporate hydrophobic components into the hydrogel to displace the interfacial water layer to facilitate subsequent chemical and physical bonding with the tissue [28,44,45,46]. Han et al. describe a novel dual adhesive hydrogel formed via an ammonium persulfate (APS) radical polymerization of chitosan grafted with methacrylate (CS-MA), dopamine, and N-hydroxymethyl acrylamide. When in contact with water, the abundant hydrophobic residues of CS aggregate to repel interfacial water and the catechol and amine groups chemically and physically bond with the tissue surface. The hydrogel achieves an adhesive strength of over 34 kPa and a burst pressure up to 168 mmHg, and facilitates the hemostasis of a rabbit’s heart in vivo [43]. This formulation has not been evaluated for preventing GI bleeding. However, due to the cytotoxicity of the catalysts and the viscosity of the hydrogel solution, this hydrogel is pre-made and administered as a patch. Alternatively, microparticle-based injectable wound dressings with wet-adhesion stability are being developed based on hydrophobically-modified Alaska pollock gelatin via thermal crosslinking for GI hemostasis and wound healing [44]. Hydrophobic modification drastically enhances the mechanical properties and the underwater stability of the microparticles to up to 4 days. The microparticles also suppress fibrosis and inflammation in a rat skin wound healing model. Additionally, self-assembling peptide (SAP) hydrogels spontaneously form nanofibers and interact with the tissue surface through both hydrophobic and hydrophilic interactions [25,31,36,47,48]. The SAP hydrogel, also known as RADA16, consists of a fully synthetic 16-amino-acid polypeptide with a repeating sequence of R (positively charged arginine), A (hydrophobic alanine), and D (negatively charged aspartic acid). The monomer building blocks form crosslinked β-sheet structures via non-covalent interactions and mimic a natural extracellular matrix scaffold that adheres to the tissue [47]. SAP hydrogels are effective in controlling bleeding after EMR for gastric cancer, with an average time-to-hemostasis of 105 s [39,48].

In addition to hemorrhage control and wet adhesion, bacteria mitigation and infection prevention are also crucial to an advantageous wound healing outcome. Exposure of GI wound tissues to pathogenic bacteria, such as Pseudomonas aeruginosa, results in enhanced collagen-degradation [6]. Traditional protein-based sealants such as fibrin glue and collagen patches are particularly susceptible to bacterial degradation—limiting their utility [6]. One approach to inhibit bacteria migration to the wound site is to create hydrogels with pore sizes smaller than the bacteria size and thus restrict bacterial mobility and migration to the tissue [27]. Alternatively, antibiotics such as tetracycline and vancomycin loaded into hydrogels exert antibacterial activity against gram-negative and gram-positive bacteria, respectively [37,49]. Chitosan is a nontoxic natural antimicrobial polymer and kills microbes by destabilizing the negatively charged membrane of the bacteria [50]. Hydrogels consisting of modified chitosan show anti-bacterial and anti-infection capabilities against Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus), respectively, while maintaining biocompatibility [38,43].

Several hydrogel designs show promise to treat acute GI bleeding and facilitate wound healing. Due to the nature of endoscopic intervention, the design of hydrogel adhesives for GI hemostasis focuses on injectability for easy delivery and quick gelation for rapid hemostasis. However, a challenge remains to identify a hydrogel that balances facile application with prolonged hemostatic and adhesive capabilities without over-complicating the formulation. For example, strong wet adhesion may require longer gelation time and patch delivery [42,43]. On the other hand, forming more acid-tolerant hydrogel networks may entail the risk of cytotoxicity from required catalysts during in situ polymerization [42]. Excitingly, many of the formulations are still in their very early stages and significant research opportunities are present. Further design explorations and in vivo animal studies are needed to identify reliable hydrogel candidates for different location-specific GI hemorrhages followed by the first in-human trials to advance these materials to the clinic—as none currently exist.