The HM was manufactured by crosslinking Sodium Hyaluronate (HA, Shisheido Co., Tokyo, Japan) with 1,4 butanediol diglycidyl ether (BDDE, Sigma-Aldrich Co., St. Louis, MO, USA). It has uinform pores sized 10.0 ± 3.7 µm on scanning electron microscopy (Fig. 1).

This animal study was authorized with Institutional Animal Care and Use Committees of Genewel Co. (approved no. Bio-111010-01). Animals for experiments were maintained in compliance with all regulatory guidelines by Korea Ministry of Food and Drug Safety.

Three to four month-old female New Zealand white rabbits weighing 2.5-3 kg were purchased from DBL Co. (Eumseong, Korea). All the animals were stabilized for 1 week before the surgical manipulation. Open abdominal surgery was performed to make models for intra-abdominal wall and cecum abrasion.

To induce general anesthesia, Zoletil 50 (0.2 mL/kg, Virbac, Carros cedex, France) and Rompun (0.03 mL/kg, Bayer Korea, Seoul, Korea) were used. The animals induced with general anesthesia were placed in the supine position, shaved in the abdominal area, prepared with povidone solution, and then draped in a sterile manner. Five to six centimeters of an anterior midline incision was made through the abdominal wall and peritoneum. After identifying the cecum, the serosal surface of the cecum was abraded by bone burr until an area of 3 cm² × 3 cm² was denuded as evidenced by punctuate bleeding without hemostasis. The opposing peritoneal surface of the abdominal wall was also abraded by bone burr until the same size area deperitonized. Animals were grouped into three categories; negative control group, positive control group (Interceed) and experimental group (HM) (each, n = 10). A 4 cm × 4 cm tissue adhesion barrier was applied to cover the entire area exposed and fixed with nonabsorbable sutures with 5/0 nylon sutures (AILee Co., Busan, Korea).

The abdominal incision was closed with 4/0 Nylon suture, skin was closed with 3/0 Nylon suture and 330 mg D-pron (Handong Co., Seoul, Korea) and 8.5 mg gentamycin (Deasung Microbiological Institute, Seoul, Korea) were injected into thigh muscles. All the animals were fed with sufficient food and water, and they were checked daily for any abnormal signs and symptoms.

Four weeks after the operation, the animals were sacrificed, and the adhesion formation was evaluated according to adhesion scoring system as follows by a blinded and inde pendent staff using the data: score 0, no adhesion; score 1, filmy adhesion easily separable with blunt dissection; score 2, mild to moderate adhesion with free dissection; score 3, moderate to dense adhesions with difficult dissection or nondissection [13,14] (Fig. 2).

Additionally, the adhesion area was measured, and adhesion area was expressed as mean ± standard deviation. Statistical difference in the adhesion scores and areas from the three groups was made by analysis of variance (ANOVA) test. A P-value < 0.05 was considered statistically significant.

To evaluate acute toxicity of the applied antiadhesive agents, 5 mL blood samples were drawn from the ear vein in each ani mal on postoperative day 3 and 7. The blood samples were tested for the following categories: glutamic oxaloacetic transa minase, glutamic pyruvic transaminase, creatinine, blood urea nitrogen, red blood cell (RBC), white blood cell (WBC), hematocrit (Hct), and hemoglobin (Hb).

To investigate inflammation response and the degree of tissue healing, the injured tissues including the abdominal wall and cecum were excised and fixed in 10% formaldehyde for 24 hours. After dehydration in graded series of ethanol, the specimens were embedded in paraffin wax and cut into 5-µm transverse section in the center of the injury site. These sections were stained with hematoxylin and eosin (H&E) for observation by light microscopy. The morphometric evaluation consisted of a quantitative analysis of inflammatory response by the counting of leukocytes. The number of cells on each slide (observation area, 425 µm² × 425 µm²) was counted and the content of leukocytes was calculated as follows: Percentage of leukocyte (%) = (no. of leukocytes/no. of total cells) × 100. The granulation tissue formation around injured tissue was also blindly evaluated according to the following grading system: 0, no granulation tissue; 1, focal proliferation of granulation. Tissue less than on low-power field; 2, intermediate degree of granulation tissue between 1 and 3; 3, diffuse proliferation of granulation tissue involving more than half thickness of the abdominal muscle [15]. For the statistical treatment, ANOVA test was used and it was considered statistically significant at P < 0.05. Additionally, to evaluate organ toxicity, liver, kidney and spleen were procured from the animals, H&E staining was executed and they were observed by light microscopy [16,17].

To test the safety of the applied antiadhesive agents, blood samples obtained from each group on postoperative days 3 and 7 were analyzed for liver and kidney function. The result indicated no significant difference among the three groups (Fig. 3). RBC, WBC, Hct and Hb also showed no significant difference among the three groups (Fig. 4).

The animals were sacrificed at the 4th week after the opera tion and the adhesion severity was evaluated. Within a day, an animal from negative control group and another animal from positive control group died to make the final samples as 9 in negative control group, 9 in positive control group, and 10 in experimental group. The cause of mortality of control group was suggested as an anesthesia-related problem.

In the adhesion grade evaluation, negative control group showed 2.44 ± 0.72; positive control group (Interceed), 1.6 ± 0.51; and experimental group (HM), 0.22 ± 0.66. There was statistically significant difference between the three groups (P < 0.05) (Fig. 5A). Additionally, in adhesion area, negative control group showed 4.5 ± 0.73 cm², positive control group 2.87 ± 1.24 cm², and experimental group 0.22 ± 0.66 cm², which confirmed the significant reduction of adhesion area in experimental group (P < 0.05) (Fig. 5B). From these results, it was confirmed that HM had better adhesion-prevention effects than Interceed.

The inflammatory response was evaluated as the ratio of the number of leukocyte cells over total cells (%). All three groups such as negative control group, positive control group (Interceed), and experimental group (HM) showed similar re sults and there was no statistically significant difference (P = 0.507). It was confirmed that HM did not have any safety problems in inflammatory response results. There was no statistically significant difference in granulation tissue formation (P = 0.641). As experimental group showed similar results as negative control group in granulation tissue formation, it was confirmed that HM did not lead to any abnormal side effects during wound healing in a rat animal model.

In the histological observation of liver, kidney and spleen to test the safety and toxicity against major organs, there was no infiltration of inflammatory cell or tissue necrosis in experimental group just as in negative control group (Fig. 6).