Zifeng Zhang, Yingjie Zhang, Kaibo Wang, Yanbing Lv, Daqing Pan, Fei Wang, Shiping Huang, Shouren Zhang, Baocheng Yang, Xiaohong Li
Chitosan is widely employed as a hemostatic material for hemorrhage control due to its excellent biocompatibility, and intrinsic cationic property that promotes red blood cell aggregation. However, the insufficient stability under high-pressure arterial flow and the poor mechanical strength of the formed clots can lead to a high risk of secondary bleeding. Here, machine learning-guided hierarchical sized chitosan hemostatic particles (CSHP) with electrostatic-hydrophobic synergy were fabricated through multistage particle size distributions and lauric acid loading. The CSHP combined physical plugging (large, medium, and small particles for anchoring, bridging, and dense sealing) with electrostatic attraction and hydrophobic interactions to form a compact clot network that resisted arterial flow independent of systemic coagulation activation. In vitro coagulation tests showed that CSHP-3 reduced whole blood clotting time by 33.3% compared to LA-free CS. Rheological measurements showed that CSHP-3 formed blood clots exhibited a storage modulus of 1566 Pa, which was 51.3% higher than that of the commercial product CELOX™ (1035 Pa) and 6 times that of the blank group (263 Pa). Moreover, in a Bama minipig femoral artery injury model, CSHP-3 achieved successful hemostasis within 180 s. This work paves the way for more stable, more effective hemorrhage control and faster clinical translation in emergency management of deep arterial wounds.