In Vivo Evaluation of Collagen Hemostats: Biocompatibility and Resorption

In Vivo Evaluation of Collagen Hemostats: Biocompatibility and Resorption (Department of Cardiovascular Surgery, The Heart Institute of Japan and Department of Anesthesiology*, Tokyo Women's Medical University, Tokyo, Japan) Yasuko Tomizawa Makiko Komori* Katsumi Takada* Hiroshi Nishida Masahiro Endo Hiromi Kurosawa After bleeding is controlled with hemostats during surgery, the residual material should be resorbed without adverse effects such as inflammation, infection or scar formation. To evaluate the biocompatibility of hemostats, three kinds of commercially available collagen hemostats, cotton type (Integran®), microfibrillar type (Avitene®), and sheet type (TachoComb®), were examined. A rabbit ear chamber (REC), a system for viewing materials in vivo, was applied to the auricle of male Japanese white rabbits. The REC was designed to leave a 50-µm-thick and 6.4mm-diameter chamber, and 0.5mg of each specimen (Integran; n=8, Avitene; n=6, TachoComb; n=6) was placed in the chamber. Macroscopic and microscopic observations were performed every week up to 5 weeks without anesthetizing or stressing the animal. In the Integran group, capillaries infiltrated between the collagen fibers, and the vasculature in the REC field was complete in 6 out of 8 animals at 5 weeks. Cotton type collagen fibers of Integran became thinner every week without effusion. In the TachoComb group, capillaries were directed toward the effusion at 2 weeks, while in the Avitene group, a similar phenomenon was not observed. The vasculature was incomplete, with either effusion or infection at 5 weeks in the Avitene and TachoComb groups. Material was recognized up to 4 weeks in the TachoComb group, whereas the space occupied by material remained vacant without vasculature in the Avitene group. Our results suggest that cotton type configuration is excellent as a collagen hemostat, with smooth capillary infiltration, rapid resorption of material and promotion of the healing process. Jpn. J. Cardiovasc. Surg. 32: 17-22 (2003)