# Fileset

[R3_hMPs_Bleeding_MANUSCRIPIT_20260128_Clean.pdf](https://mdr.nims.go.jp/filesets/650d455d-e413-44c5-949a-356f321d2746/download)

## Creator

Shohei Uehara, Fumisato Sasaki, Hidehito Maeda, Makoto Hinokuchi, Akihito Tanaka, Shiho Arima, Shinichi Hashimoto, Shuji Kanmura, Hisashi Sahara, Yasuko Kobayashi, [Akihiro Nishiguchi](https://orcid.org/0000-0002-3160-6385), [Tetsushi Taguchi](https://orcid.org/0000-0003-2541-2530)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Sprayable gelatin microparticles prevent delayed gastric bleeding in an anticoagulated swine model](https://mdr.nims.go.jp/datasets/399246a3-5d2e-4b39-91e7-c0967ea07886)

## Fulltext

1  Title 1 Sprayable gelatin microparticles prevent delayed gastric bleeding in an anticoagulated 2 swine model 3  4 Shohei Uehara1, Fumisato Sasaki1, Hidehito Maeda1, Makoto Hinokuchi1, Akihito Tanaka1, 5 Shiho Arima1, Shinichi Hashimoto1, Shuji Kanmura1, Hisashi Sahara2, Yasuko Kobayashi3, 6 Akihiro Nishiguchi3, Tetsushi Taguchi3  7  8 1Department of Digestive and Life-style related Diseases, Kagoshima University Graduate 9 School of Medical and Dental Sciences, Kagoshima, Japan 10 2Division of Experimental Large Animal Research, Life Science and Laboratory Animal 11 Research Unit, Center for Advanced Science Research and Promotion, Kagoshima University, 12 Kagoshima, Japan 13 3Research Center for Macromolecules and Biomaterials, National Institute for Materials 14 Science, Tsukuba, Japan 15 †Shohei Uehara and Fumisato Sasaki contributed equally to this study and share first 16 authorship. 17   18 2  Corresponding author: 19 Fumisato Sasaki, MD, PhD 20 Digestive and Lifestyle Diseases, Kagoshima University Graduate School of Medical and 21 Dental Sciences 22 8-35-1, Sakuragaoka, Kagoshima, 890-8520 Japan 23 Tel: +81-99-275-5326 24 Fax: +81-99-264-3504 25 Email: bungohs@m2.kufm.kagoshima-u.ac.jp 26  27 Funding information 28 This work was supported by the Translational Research Program TR-SPRINT (Strategic 29 Promotion for Practical Application of Innovative Medical Technology) of the Japan Agency 30 of Medical Research and Development (grant number: 20 lm0203114h0001), and the Japan 31 Society for the Promotion of Science KAKENHI (grant numbers: 19K17467 and 20H0247).  32  33 Author contributions 34 Conception and design: Fumisato Sasaki; data analysis and interpretation: Shohei Uehara and 35 Fumisato Sasaki; drafting of the article: Shohei Uehara and Fumisato Sasaki; critical revision 36 of the article for important intellectual content: Hidehito Maeda, Makoto Hinokuchi, Akihito 37 Tanaka, Shiho Arima, Shinichi Hashimoto, Hisashi Sahara, and Shuji Kanmura; final approval 38 of the article: all authors.  39 3  Shohei Uehara and Fumisato Sasaki contributed equally to this study and share first authorship. 40   41 4  Abstract 42 Background: Delayed bleeding following endoscopic mucosal resection or endoscopic 43 submucosal dissection is a serious concern in patients receiving anticoagulant therapy. We 44 previously established an experimental animal model of delayed gastric bleeding following 45 endoscopic resection under controlled anticoagulation. In this study, we aimed to evaluate the 46 potential preventive effect of hydrophobized microparticles (hMPs), a novel sprayable wound-47 covering material derived from Alaska pollock gelatin, in preventing delayed bleeding. 48 Methods: Twelve gastric mucosal defects were created by endoscopic mucosal resection with 49 ligation in three CLAWN miniature swine. To induce delayed bleeding, systemic heparin was 50 administered after confirming an activated clotting time greater than 220 s. hMPs were 51 endoscopically sprayed into each ulcer base (200 mg/ulcer). Endoscopic examination was 52 performed 24 h later. The primary endpoint was the presence or absence of delayed bleeding. 53 Secondary endpoints included hMP retention, ulcer coverage rate, presence of exposed vessels, 54 and vascular coverage by hMPs. 55 Results: During the 24-hour observation period, no delayed bleeding was observed in any of 56 the three CLAWN miniature swine treated with hMPs. Hemoglobin levels on Day 1 remained 57 stable in all swine. hMPs were present in 100% of the ulcers. Complete ulcer base coverage 58 (100%) was achieved in 55.6% (20/36) of the sections, with ≥50% coverage observed in all 59 sections. Exposed vessels were identified in 36.1% (13/36) of ulcers, and all were covered with 60 5  hMPs. 61 Conclusion: Endoscopic spray application of hMPs demonstrated strong adhesion to gastric 62 ulcers and was associated with the absence of delayed bleeding in this short-term experimental 63 model. These findings suggest that hMPs may represent a promising strategy for reducing the 64 risk of delayed bleeding after gastrointestinal endoscopic surgery. 65  66 Keywords: endoscopic submucosal dissection; endoscopic mucosal resection; delayed gastric 67 bleeding; anticoagulation therapy; hydrophobized microparticles, animal model  68 6   69 Introduction 70 Endoscopic submucosal dissection (ESD) and endoscopic mucosal resection (EMR) are well-71 established treatments for early gastric tumors [1]. Delayed bleeding occurs in approximately 72 1.8–15.6% of patients after ESD [2,3], with antithrombotic therapy recognized as a significant 73 risk factor. According to a recently proposed predictive model from Japan, the incidence of 74 delayed bleeding ranges from 2.8% in the low-risk group to 29.7% in the very-high-risk group 75 [4]. With an aging population, the number of patients receiving antithrombotic therapy 76 continues to increase [5], making delayed bleeding after ESD an increasingly important clinical 77 concern.  78 Despite its clinical importance, no definitive strategy has been established for preventing 79 delayed bleeding after ESD [6]. A major limitation in developing such interventions is the lack 80 of a reliable animal model for delayed post-ESD bleeding. We recently established such a 81 model using gastric mucosal resection with systemic anticoagulation via a single bolus of 82 heparin followed by continuous infusion [7]. We developed a sprayable wound dressing 83 comprising multifunctional hydrophobized microparticles (hMPs) derived from swine gelatin 84 [7]. These particles demonstrate strong tissue adhesion, even in wet environments, and 85 effectively suppress submucosal fibrosis in the post-ESD setting [7,8]. More recently, we 86 developed hMPs using fine particles derived from Alaska pollock gelatin, which maintained 87 7  strong adhesive properties [9,10]. These hMPs have shown efficacy in closing gastrointestinal 88 perforations, reducing inflammation in duodenal ESD ulcers [11], and preventing esophageal 89 strictures in animal models [12]. 90 In this study, we aimed to evaluate the efficacy of hMPs, a novel sprayable wound-covering 91 material derived from Alaska pollock gelatin, in preventing delayed bleeding. 92  93 Materials and Methods 94 Experimental animals 95 Three CLAWN miniature swine (age: 6 months; weight: 14–17 kg; procured from Kagoshima 96 Miniature Swine Research Center, Kagoshima, Japan) were used in this study. As 97 premedication, the animals were intramuscularly injected with ketamine (15 mg/kg; Daiichi 98 Sankyo Propharma Co., Ltd., Tokyo, Japan) or xylazine (2 mg/kg; Bayer Yakuhin, Ltd., Osaka, 99 Japan). An endotracheal tube (Smith Medical Japan, Tokyo, Japan) was inserted, and anesthesia 100 was maintained with inhaled isoflurane (1.5–3.0%) (DS Pharma Animal Health Co., Ltd., 101 Osaka, Japan), delivered via a ventilator, with continuous monitoring of heart rate, respiratory 102 status, and oxygen saturation throughout the procedure. 103 Endoscopic mucosal resection with ligation (EMR-L) procedure and post-procedure follow-104 up 105 An upper gastrointestinal endoscope was orally inserted into each swine to create gastric 106 8  mucosal defects using the EMR-L technique [13]. Twelve artificial gastric ulcers were created 107 by an endoscopist by performing the EMR-L technique using an upper gastrointestinal 108 endoscope (GIF-Q260J; Olympus, Tokyo, Japan), and a videoscope system (EVIS LUCERA 109 CV-260SL; Olympus) was used to ensure consistent collection of data on delayed bleeding. 110 Markings were made at 12 locations: lesser curvature, anterior and posterior walls of the upper 111 and middle gastric bodies, gastric angle, and lower gastric body. A polypectomy snare 112 (Captivator™ II 15 mm; Boston Scientific, USA) in coagulation mode was used for marking. 113 One milliliter of 0.9% saline containing indigo carmine (Otsuka Pharmaceutical Co. Ltd., 114 Tokyo, Japan) was injected into the submucosa using an injection needle to achieve adequate 115 lifting. Endoscopic variceal ligation (EVL) was performed using a ligation device (Pneumo 116 Activate EVL; Sumitomo Bakelite, Tokyo, Japan), followed by snaring and resection with a 117 high-frequency device (Pulse-Cut Fast mode, 120 W, ESG-100; Olympus) [13,14]. All EMR-118 L procedures were performed by S.U. Solid food was withheld from the day before the 119 procedure until the day after; however, water was provided ad libitum. No proton pump 120 inhibitor was administered following EMR-L [7]. Feeding was managed by a specialized 121 animal technician at the Kagoshima University Animal Experimental Facility, and the animals' 122 conditions were monitored daily. Animal care, housing, and surgical procedures were 123 conducted in compliance with the Kagoshima University Animal Experiment Committee 124 guidelines. Necropsies were performed by several researchers in accordance with the ethical 125 9  guidelines of the Kagoshima University Animal Experimentation Facility. This study was 126 approved by the Animal Experiment Committee of Kagoshima University (approval number: 127 MD23056). 128 This study is reported in accordance with the ARRIVE guidelines. 129  130 hMPs 131 Hydrophobically modified Alaska pollock gelatin was synthesized via the reaction between a 132 primary amine and a decanoic anhydrate [15]. The hMPs were prepared using a coacervation 133 method in a water/ethanol mixed solvent [16]. Optimization of the alkyl chain length (decanoyl 134 groups, C10) and the degree of substitution (50 mol% of amino groups in Alaska pollock 135 gelatin) enhanced the mechanical strength of the hydrogel formed by the hydration and fusion 136 of the microparticles. Scanning electron microscopy confirmed that the resulting hMPs were 137 microparticles (Figure 1). 138 Delivery of hMPs 139 The hMPs (200 mg) were packed into small vials. A battery-powered endoscopic injector (Alto 140 Shooter®, Kaigen, Tokyo, Japan), designed for the application of powdered agents, was used 141 to spray the hMPs. The vial was directly attached to the Alto Shooter®, and the powder was 142 sprayed onto the injured mucosa through the endoscope channel after removing the nozzle. 143 Overall, 200 mg hMPs (one vial) were applied to each ulcer (Figure 2). 144 10  Heparin administration, activated clotting time (ACT) measurement, and swine dissection 145 All animals received heparin via a catheter inserted into the external jugular vein. Catheters 146 (Argyle Fukuroi; CV catheter, 14 cm × 30 cm; Cardinal Health, USA) were inserted for blood 147 sampling and heparin administration [7]. The experimental protocol is illustrated in Figure 3. 148 After EMR-L, 50 U/kg unfractionated heparin (heparin sodium; Mochida Pharmaceutical Co., 149 Ltd., Tokyo, Japan) was administered intravenously. The ACT was measured 10 min later using 150 a coagulation analyzer (Hemochron Jr Signature+, Accriva Diagnostics, Inc., USA). Additional 151 doses of 50 U/kg were administered every 10 min until the ACT exceeded 220 s, following a 152 previously established protocol [7]. Continuous heparin infusion (50 U/kg/h) was then initiated 153 using a portable disposable infusion pump (SUREFUSER® A, SFS-1002D, flow rate 2.1 mL/h; 154 NIPRO, Osaka, Japan). ACT was monitored at 0.5, 1, 2, and 4 h after starting the continuous 155 infusion. Endoscopic observation was performed 24 h after EMR-L to assess delayed bleeding. 156 The animals were then euthanized with an intravenous injection of thiamylal sodium 157 (ISOZOL®, Nichi-Iko Pharmaceutical Co., Ltd., Toyama, Japan; 90 mg/kg) and potassium 158 chloride (Terumo Corporation, Tokyo, Japan; 20 mEq), followed by abdominal dissection and 159 removal of the stomach.  160 Humane endpoints and animal welfare monitoring 161 All animals were monitored at least twice daily by trained animal technicians for activity level, 162 respiratory pattern, posture, and signs of bleeding. Humane endpoints were predefined as 163 11  follows: (1) persistent inactivity despite stimulation, (2) signs of respiratory distress, and (3) 164 marked deterioration in physiological status. Animals meeting any of these criteria were 165 humanely euthanized immediately to minimize suffering. No unplanned humane euthanasia 166 was required during the study. 167  168 Histological analyses 169 Tissue specimens were fixed in 10% neutral-buffered formalin (Kenei Pharmaceutical, Osaka, 170 Japan) for 48 h, sectioned into 20-mm squares, embedded in paraffin, sliced into 2-µm-thick 171 sections, and stained using hematoxylin and eosin or Masson’s trichrome. The retention and 172 coverage of hMPs, the presence of exposed vessels at the ulcer base, and the extent of hMPs 173 adherence to these vessels were assessed histologically. 174 Outcome measures 175 The primary endpoint was the presence or absence of delayed bleeding [7]. Hematemesis was 176 assessed on the day after the procedure, and melena was evaluated during necropsy. Delayed 177 bleeding was diagnosed if at least one of the following criteria was met: (1) hematemesis or 178 vomiting of blood within 24 h after the procedure, or (2) presence of blood clots or retained 179 blood in the stomach observed via endoscopy within 24 h [7]. Hemoglobin (Hb) levels were 180 measured immediately after catheter insertion and again between anesthesia induction and 181 endoscopic observation the next day, and the values were compared.  182 12  Secondary endpoints included the retention and coverage rates of hMPs at ulcer sites (n = 36), 183 the presence of exposed vessels at the ulcer base, and the proportion of hMPs adherent to those 184 vessels. The coverage rate was defined as the percentage of the ulcer base length covered by 185 hMPs, as assessed on the bisecting plane of the ulcer. 186  187  188 Results 189 The characteristics and findings for each swine are presented in Table 1. All 12 ulcers had an 190 approximate diameter of 10 mm. The size of the ulcer was measured using a major forceps. 191 The submucosal layer remained intact at all sites, and the muscularis propria was identifiable. 192 No perforations or active bleeding were observed. 193 The time course of ACT values following heparin administration is shown in Figure 4. Heparin 194 (50 U/kg) was repeatedly administered until the ACT exceeded 220 s in all the swine. The ACT 195 values before continuous infusion were 265, 281, and 236 s in the first, second, and third swine, 196 respectively. The maximum recorded ACT values were 310, 347, and 240 s, respectively. 197 No delayed bleeding was observed in any swine treated with hMPs (Figure 5). Hb levels were 198 10.4 ± 1.0 g/dL at baseline and 11.5 ± 1.2 g/dL after treatment. The mean change was 1.1 ± 1.8 199 g/dL. The Hb levels remained stable in all swine（Before EMR-L: 10.4 ± 1.0 g/dL, After EMR-200 L: 11.5 ± 1.2 g/dL). hMPs were detected in all 36 ulcer sections, with a 100% retention rate. 201 13  Complete coverage (100%) of the ulcer base was observed in 55.6% (20/36); 75–99%, in 202 19.4% (7/36); and 50–74%, in 25.0% (9/36); all ulcers exhibited ≥ 50% coverage (Figure 6). 203 Exposed vessels were identified in 36.1% (13/36) of the ulcer bases, and all were covered by 204 hMPs. Representative histological images are shown in Figure 7. hMPs were firmly attached 205 to the ulcer base and provided additional coverage over the exposed vessels. 206  207 Discussion 208 In this study, we evaluated the preventive efficacy of hMPs derived from Alaska pollock gelatin 209 using a newly established animal model of delayed bleeding after endoscopic treatment. Under 210 the conditions of this study, no delayed bleeding was observed, and relatively high ulcer 211 coverage rates as well as stable adherence to exposed vessels were confirmed, suggesting that 212 hMPs may contribute to the prevention of delayed bleeding in an anticoagulated swine model. 213 Delayed bleeding following endoscopic procedures is a serious adverse event in patients 214 receiving antithrombotic therapy [2,3]. Various preventive strategies, including mechanical 215 closure and the use of fibrin sealants, have been investigated [17-20].The hMPs used in this 216 study demonstrated strong adhesive properties even in wet environments, which is consistent 217 with the previous success of gelatin-based derivatives in ulcer protection and stricture 218 prevention, supporting their potential applicability as a wound-covering material in the 219 gastrointestinal tract.  220 14  A notable strength of this study is that it is the first report of a preventive intervention against 221 delayed bleeding in our newly developed [7], reproducible post-endoscopic bleeding animal 222 model. This model provides a valuable platform for the preclinical evaluation of various 223 hemostatic and wound-covering technologies. 224 We hypothesized that hMPs prevent delayed bleeding by firmly adhering to the ulcer base and 225 shielding exposed vessels from chemical and mechanical insults such as gastric acid and food 226 residue. Histological evaluation confirmed that the hMPs adhered to and covered the exposed 227 vessels. Thus, the endoscopic application of hMPs may effectively prevent delayed bleeding 228 following gastric mucosal resection (Figure 8). 229 The clinical significance of our findings lies in their potential to offer a novel prophylactic 230 strategy for high-risk patients after gastric ESD, including those receiving anticoagulant 231 therapy. If implemented in clinical practice, this technology could not only reduce the risk of 232 bleeding but also decrease the procedural workload for physicians and offer economic benefits 233 by shortening hospital stays and avoiding readmissions. Given the aging population and the 234 increasing prevalence of antithrombotic medication use, hMPs represent a promising and 235 practical option for clinical implementation. This study has some limitations. First, the 236 observation period was limited to 24 hours after EMR-L. Although this duration is relatively 237 short for fully evaluating delayed bleeding, previous clinical studies have reported that a 238 substantial proportion of post-ESD or post-EMR bleeding events occur within the first 24 hours, 239 15  particularly in patients receiving antithrombotic therapy [21,22].Therefore, evaluation during 240 this early phase remains clinically relevant; however, longer-term observation will be required 241 in future studies. Second, this was a preclinical animal study involving only three CLAWN 242 miniature swine, which limited the statistical power and generalizability of the findings. 243 Therefore, these results should be interpreted with caution. Third, no concurrent control group 244 was included. In a previously established swine model using a similar anticoagulation protocol, 245 delayed bleeding occurred in 75% (3 of 4) of swine when continuous heparin infusion was 246 initiated at an ACT threshold of ≥200 seconds, and in all swine when initiated at an ACT 247 threshold of ≥220 seconds [7]. The present study employed the latter, stricter anticoagulation 248 condition, suggesting that this model represents a high-risk setting for delayed bleeding. 249 Although these findings support the feasibility of the observed preventive effect, direct 250 comparison with a concurrent control group would further strengthen the evidence. Fourth, the 251 in vivo retention time of hMPs could not be fully evaluated because all animals were euthanized 252 24 h after EMR-L. Although hMPs were present at that time, their long-term persistence 253 remains unclear. In addition, the effects of human gastric peristalsis and the digestive 254 environment on hMPs retention require further investigation. Furthermore, mucosal resection 255 was performed using the EMR-L method rather than the ESD; therefore, its usefulness in 256 preventing post-ESD bleeding remains unknown. However, as EMR-L removes almost the 257 entire submucosal layer, similar to ESD, and multiple areas, it may also be useful for preventing 258 16  post-ESD bleeding. Finally, the gelatin microparticles used in this study are not yet 259 commercially available, and precise cost evaluation is therefore difficult at this stage. 260 Manufacturing costs and cost-effectiveness compared with existing hemostatic strategies will 261 be important considerations for future clinical translation. 262 Collectively, these limitations indicate that our work represents an initial proof-of-concept 263 study demonstrating the feasibility and potential efficacy of sprayable hMPs in preventing 264 delayed bleeding. Future studies should ideally include larger sample sizes and, ultimately, 265 multicenter clinical trials to assess safety and efficacy, along with investigations exploring its 266 applicability to other gastrointestinal sites and broader high-risk populations. 267 Endoscopic application of hMPs on mucosal defects resulted in strong tissue adherence and 268 was associated with the absence of delayed bleeding in this high-risk animal model. These 269 findings suggest that hMPs may represent a promising prophylactic strategy for managing post-270 endoscopic bleeding, particularly in patients receiving anticoagulant therapy. 271  272  273   274 17  Acknowledgements 275 We would like to thank Editage (www.editage.jp) for English language editing.  276  277 Author contributions 278 Conception and design: Fumisato Sasaki; data analysis and interpretation: Shohei Uehara and 279 Fumisato Sasaki; drafting of the article: Shohei Uehara and Fumisato Sasaki; critical revision 280 of the article for important intellectual content: Hidehito Maeda, Makoto Hinokuchi, Akihito 281 Tanaka, Shiho Arima, Shinichi Hashimoto, Hisashi Sahara, and Shuji Kanmura; final approval 282 of the article: all authors.  283 Shohei Uehara and Fumisato Sasaki contributed equally to this study and share first authorship. 284  285  286 Conflict of interest 287 Shohei Uehara, Fumisato Sasaki, Hidehito Maeda, Makoto Hinokuchi, Akihito Tanaka, Shiho  288 Arima, Shinichi Hashimoto, Shuji Kanmura, Hisashi Sahara, Yasuko Kobayashi, Akihito 289 Nishiguchi and Tetsushi Taguchi declare no conflict of interest for this article. 290  291  Informed Consent: N/A  292  Registry and the Registration No. of the study/trial: N/A. 293 18   Animal Studies: Animal care, housing, and surgical procedures were conducted in 294 compliance with the Kagoshima University Animal Experiment Committee guidelines. 295 Necropsies were performed by several researchers in accordance with the ethical 296 guidelines of the Kagoshima University Animal Experimentation Facility. 297  298 Data availability 299 All processed data supporting the findings of this study are included in this published article 300 and its supplementary information files. The raw image data are not publicly available due to 301 file size limitations, but are available from the corresponding author upon reasonable request. 302  303   304 19  References 305 1. Pimentel-Nunes P, Dinis-Ribeiro M, Ponchon T, Repici A, Vieth M, De Ceglie A, 306 Amato A, Berr F, Bhandari P, Bialek A, Conio M, Haringsma J, Langner C, Meisner S, 307 Messmann H, Morino M, Neuhaus H, Piessevaux H, Rugge M, Saunders BP, 308 Robaszkiewicz M, Seewald S, Kashin S, Dumonceau JM, Hassan C, Deprez PH (2015) 309 Endoscopic submucosal dissection: European Society of Gastrointestinal Endoscopy 310 (ESGE) Guideline. Endoscopy 47:829-854. https://doi.org/10.1055/s-0034-1392882 311 2. Isomoto H, Shikuwa S, Yamaguchi N, Fukuda E, Ikeda K, Nishiyama H, Ohnita K, 312 Mizuta Y, Shiozawa J, Kohno S (2009) Endoscopic submucosal dissection for early 313 gastric cancer: a large-scale feasibility study. Gut 58:331-336. 314 https://doi.org/10.1136/gut.2008.165381 315 3. Park YM, Cho E, Kang HY, Kim JM (2011) The effectiveness and safety of endoscopic 316 submucosal dissection compared with endoscopic mucosal resection for early gastric 317 cancer: a systematic review and metaanalysis. Surg Endosc 25:2666-2677. 318 https://doi.org/10.1007/s00464-011-1627-z 319 4. Hatta W, Tsuji Y, Yoshio T, Kakushima N, Hoteya S, Doyama H, Nagami Y, Hikichi T, 320 Kobayashi M, Morita Y, Sumiyoshi T, Iguchi M, Tomida H, Inoue T, Koike T, Mikami 321 T, Hasatani K, Nishikawa J, Matsumura T, Nebiki H, Nakamatsu D, Ohnita K, Suzuki 322 H, Ueyama H, Hayashi Y, Sugimoto M, Yamaguchi S, Michida T, Yada T, Asahina Y, 323 20  Narasaka T, Kuribasyashi S, Kiyotoki S, Mabe K, Nakamura T, Nakaya N, Fujishiro M, 324 Masamune A (2021) Prediction model of bleeding after endoscopic submucosal 325 dissection for early gastric cancer: BEST-J score. Gut 70:476-484. 326 https://doi.org/10.1136/gutjnl-2019-319926 327 5. Ohman EM, Bhatt DL, Steg PG, Goto S, Hirsch AT, Liau CS, Mas JL, Richard AJ, 328 Röther J, Wilson PW (2006) The REduction of Atherothrombosis for Continued Health 329 (REACH) Registry: an international, prospective, observational investigation in 330 subjects at risk for atherothrombotic events-study design. Am Heart J 151:786.e1-10. 331 https://doi.org/10.1016/j.ahj.2005.11.004 332 6. Kataoka Y, Tsuji Y, Sakaguchi Y, Minatsuki C, Asada-Hirayama I, Niimi K, Ono S, 333 Kodashima S, Yamamichi N, Fujishiro M, Koike K (2016) Bleeding after endoscopic 334 submucosal dissection: Risk factors and preventive methods. World J Gastroenterol 335 22:5927-5935. https://doi.org/10.3748/wjg.v22.i26.5927 336 7. Uehara S, Sasaki F, Sahara H, Tanaka A, Hinokuchi M, Maeda H, Arima S, Hashimoto 337 S, Kanmura S, Ido A (2025) Establishment of a swine model of delayed bleeding after 338 endoscopic procedure. DEN Open 5:e411. https://doi.org/10.1002/deo2.411 339 8. Maeda H, Sasaki F, Morinaga Y, Kabayama M, Iwaya H, Komaki Y, Arima S, Nasu Y, 340 Tanoue S, Hashimoto S, Kanmura S, Nishiguchi A, Taguchi T, Ido A (2021) Covering 341 Post-Endoscopic Submucosal Dissection Ulcers in Miniature Swine with Hexanoyl 342 https://doi.org/10.1002/deo2.41121  (Hx:C6) Group-Modified Alkaline-Treated Gelatin Porous Film (HAG) Induces Proper 343 Healing by Decreasing Inflammation and Fibrosis. Digestion 102:415-427. 344 https://doi.org/10.1159/000509056 345 9. Nishiguchi A, Sasaki F, Maeda H, Kabayama M, Ido A, Taguchi T (2019) 346 Multifunctional Hydrophobized Microparticles for Accelerated Wound Healing after 347 Endoscopic Submucosal Dissection. Small 15:e1901566. 348 https://doi.org/10.1002/smll.201901566 349 10. Ito S, Nishiguchi A, Sasaki F, Maeda H, Kabayama M, Ido A, Taguchi T (2021) Robust 350 closure of post-endoscopic submucosal dissection perforation by microparticle-based 351 wound dressing. Mater Sci Eng C Mater Biol Appl 123:111993. 352 https://doi.org/10.1016/j.msec.2021.111993 353 11. Kabayama M, Sasaki F, Hidehito M, Fujino Y, Yano H, Tanaka A, Arima S, Tanoue S, 354 Hashimoto S, Kanmura S, Hirade H, Nishiguchi A, Taguchi T, Ido A (2023) Effects of 355 Sprayable, Highly Adhesive Hydrophobized Gelatin Microparticles on Endoscopic 356 Submucosal Dissection: A Swine Model. Digestion 104:137-147. 357 https://doi.org/10.1159/000526650 358 12. Yano H, Sasaki F, Maeda H, Uehara S, Kabayama M, Fujino Y, Tanaka A, Hinokuchi 359 M, Arima S, Hashimoto S, Kanmura S, Ito S, Nishiguchi A, Taguchi T, Ido A (2025) 360 Effect of sprayable, highly adhesive hydrophobized gelatin microparticles on 361 22  esophageal stenosis after endoscopic submucosal dissection: an experimental study in 362 a swine model. Esophagus 22:95-104. https://doi.org/10.1007/s10388-024-01090-8 363 13. Suzuki H (2001) Endoscopic mucosal resection using ligating device for early gastric 364 cancer. Gastrointest Endosc Clin N Am 11:511-518. https://doi.org/10.1016/S1052-365 5157(18)30058-8 366 14. Ito S, Nagasaka K, Komatsu H, Palai D, Nishiguchi A, Taguchi T (2024) Improved 367 hydration property of tissue adhesive/hemostatic microparticle based on 368 hydrophobically-modified Alaska pollock gelatin. Biomater Adv 159:213834. 369 https://doi.org/10.1016/j.bioadv.2024.213834 370 15. Yahagi N, Kato M (2020) “Endoscopic Treatment: EMR and ESD. In: Ando N, editor. 371 Esophageal Squamous Cell Carcinoma. Singapore:” Springer Singapore 149-160. 372 https://doi.org/10.1007/978-981-15-4190-2_9 373 16. Nishiguchi A, Kurihara Y, Taguchi T (2019) Underwater-adhesive microparticle 374 dressing composed of hydrophobically-modified Alaska pollock gelatin for 375 gastrointestinal tract wound healing. Acta Biomater 99:387-396. 376 https://doi.org/10.1016/j.actbio.2019.08.040 377 17. Higuchi K, Goto O, Koizumi E, Nakagome S, Habu T, Ishikawa Y, Kirita K, Noda H, 378 Onda T, Omori J, Akimoto N, Iwakiri K (2024) Potential for expanded application of 379 https://doi.org/10.1016/j.bioadv.2024.213834https://doi.org/10.1007/978-981-15-4190-2_923  endoscopic hand suturing: A pilot study of 15 cases. Endosc Int Open 12:E507-E512. 380 https://doi.org/10.1055/a-2284-9492 381 18. Nomura T, Sugimoto S, Temma T, Oyamada J, Kamei A (2022) Clip-line closure with 382 the reopenable clip over line method for a large mucosal defect after gastric endoscopic 383 submucosal dissection. Endoscopy 54:E1-E2. https://doi.org/10.1055/a-1346-8991 384 19. Nishiyama N, Kobara H, Kobayashi N, Chiyo T, Tada N, Kozuka K, Matsui T, Yachida 385 T, Fujihara S, Shi T, Masaki T (2022) Efficacy of endoscopic ligation with O-ring 386 closure for prevention of bleeding after gastric endoscopic submucosal dissection under 387 antithrombotic therapy: a prospective observational study. Endoscopy 54:1078-1084. 388 https://doi.org/10.1055/a-1782-3448 389 20. Kikuchi D, Iizuka T, Makino S, Hayasaka J, Odagiri H, Ochiai Y, Suzuki Y, Nomura K, 390 Ohkura Y, Okamoto Y, Tanaka M, Matsui A, Mitani T, Hoteya S (2019) Utility of 391 autologous fibrin glue and polyglycolic acid sheet for preventing delayed bleeding 392 associated with antithrombotic therapy after gastric ESD. Endosc Int Open 7:E1542-393 E1548. https://doi.org/10.1055/a-1007-1694 394 21. Tsuji Y, Ohata K, Ito T, Chiba H, Gunji T, Saito N, Matsuhashi N (2010) Risk factors 395 for bleeding after endoscopic submucosal dissection for gastric cancer. Gastrointest 396 Endosc 71: 1005–1012. https://doi.org/10.1016/j.gie.2009.12.033 397 https://doi.org/10.1055/a-2284-9492https://doi.org/10.1055/a-1346-8991https://doi.org/10.1055/a-1782-3448https://doi.org/10.1055/a-1007-1694https://doi.org/10.1016/j.gie.2009.12.03324  22. Nam HS, Lee DH, Kim JW, Jeon HK, Kim GH, Park DY, Kim DU, Song GA (2019) 398 Risk factors for delayed bleeding by onset time after endoscopic submucosal dissection 399 for gastric neoplasm. Scientific Reports 9: 2674. https://doi.org/10.1038/s41598-019-400 39381-1 401   402 25  Figure legends 403 Figure 1. Morphology and appearance of the hMPs 404 Left: SEM image of hMPs shows spherical microparticles with a uniform size distribution 405 (magnification ×2,000) 406 Right: A vial containing dried hMP powder 407 Abbreviations: hMPs, hydrophobized microparticles; SEM, scanning electron microscopy 408  409 Figure 2. Endoscopic views of hMP application 410 (a) Artificial gastric ulcer created via EMR-L 411 (b) Appearance immediately after spraying hMPs onto the ulcer base 412 (c) Formation of a hydrogel layer as hMPs rapidly gel upon contact with the moist ulcer surface 413 Abbreviations: hMPs, hydrophobized microparticles; EMR-L, endoscopic mucosal resection 414 with ligation 415  416 Figure 3. Experimental design of the delayed bleeding model 417 Artificial gastric ulcers are created using EMR-L (step 1), followed by the endoscopic spraying 418 of hMPs on the ulcer base (step 2). A bolus of heparin (50 U/kg) is administered until the ACT 419 exceeds 220 s (step 3). Continuous heparin infusion (50 U/kg/h) is initiated, and ACT is 420 monitored at 0.5, 1, 2, and 4 h (step 4). The animals are sacrificed 24 h after EMR-L, for 421 26  endoscopic and histological evaluations. 422 Abbreviations: hMPs, hydrophobized microparticles; EMR-L, endoscopic mucosal resection 423 with ligation; ACT, activated clotting time 424  425 Figure 4. Time course of ACT following intravenous heparin administration 426 Changes in ACT over time in three individual swine (numbers 1–3) are shown. After bolus 427 administration of heparin (50 U/kg), ACT is measured repeatedly, and continuous infusion (50 428 U/kg/h) is initiated when ACT exceeds 220 s. ACT values are subsequently monitored at 429 multiple time points during the infusion period. 430 Abbreviation: ACT, activated clotting time 431  432 Figure 5. Endoscopic and gross anatomical images of the stomach 24 h after the procedure 433 in three swine 434 Upper panels: Endoscopic views from swine numbers 1–3 showing ulcer bases covered with 435 hMPs 436 Lower panels: Gross anatomical views of resected stomachs showing no evidence of delayed 437 bleeding (no active bleeding or adherent clots) 438 Abbreviation: hMPs, hydrophobized microparticles 439  440 27  Figure 6. Ulcer base coverage by hMPs 441 Distribution of ulcer base coverage by hMPs in 36 ulcers. Complete coverage (100%) was 442 achieved in 20 ulcers (55.6%), 75–99% coverage in 7 ulcers (19.4%), and 50–74% coverage 443 in 9 ulcers (25.0%). None of the ulcers showed a coverage below 50%. 444 Abbreviation: hMPs, hydrophobized microparticles 445  446 Figure 7. Histological evaluation of ulcer bases after EMR-L and hMP application 447 The hMPs are retained on the ulcer base following EMR-L and are observed to cover the 448 exposed blood vessels within the submucosa (a, c). Higher-magnification images (b, d) clearly 449 show hMPs adhering to the surfaces of the exposed vessel-like structures. 450 Abbreviations: hMPs, hydrophobized microparticles; EMR-L, endoscopic mucosal resection 451 with ligation 452  453 Figure 8. Proposed mechanism of delayed bleeding prevention by hMPs 454 Post-ESD ulcers are susceptible to damage and delayed bleeding under standard conditions. 455 Spraying hMPs on the ulcer base creates a protective barrier, suppresses local inflammation, 456 and prevents bleeding. hMPs can be easily delivered through an endoscope. 457 Abbreviations: hMPs, hydrophobized microparticles; ESD, endoscopic submucosal dissection458 28   459