The Comparison of the Effects of Ellagic Acid and Diclofenac Sodium on Intra-Abdominal Adhesion: An In Vivo Study in the Rat Model

Peritoneal adhesions are fibrotic adhesions that are formed intra-abdominally on the visceral or peritoneal surface during the healing of peritoneal injury.^1,2  Adhesions often occur after laparotomy and are also a significant cause of postoperative morbidity.^3,4  Adhesions can be asymptomatic but can also lead to serious complications such as intestinal obstruction, perforation and fistula.^3–6 

Adhesion formation begins with mesothelial cell injury on the peritoneal membrane surface. An inflammatory process consisting of cellular elements of small venules and fibrin exudation, edema, and hyperemia begins in this region and a serous exudates develops.^5,7–9  Fibrous bands develop between fibrinous exudate and serosal surfaces, leading to formation of fibrinous bridges. Mesothelial cells produce plasminogen activator that dissolves fibrin clots, and the fibrinolytic system is engaged causing the fibrinous exudates to be rapidly resorbed. Adhesions are formed otherwise.^10–12 

DS (diclofenac sodium) is a nonsteroidal anti-inflammatory agent often used in clinical practice for postoperative analgesia and is known to have a fibrinolytic effect.^13,14  EA (ellagic acid) is a natural antioxidant reported to be present in walnuts, carrots, tomatoes, pomegranate, grape juice, grape wine, blueberries, blackberries, and strawberries in significant quantities.^15–18  EA is a phenolic acid derivative that inhibits lipid peroxidation in addition to its anti-inflammatory, antiproliferative, anti-angiogenic and anticarcinogenic effects.^16,17 

We aimed to evaluate the effects of the oral use of DS that has an anti-inflammatory effect and EA that has an antioxidant effect in addition to its anti-inflammatory effect on intra-abdominal adhesion development in comparison in the present study.

This study was conducted with the permission of Kafkas University's Local Ethics Committee for Animal Experiments (KAÜ-HADYEK - 2012-31).

Animals: This study was conducted on 32 Wistar Albino female rats, 12 weeks of age, with a weight of 250 to 300 g in 4 groups with 8 animals in each group.

Group I: The EA group (85 mg/kg ellagic acid was administered orally, a daily single dose after the surgery),

Group II: The DS group (50 mg/kg diclofenac sodium was administered orally, a daily single dose after the surgery),

Group III: The control group (only an adhesion model was created but no drug was administered),

Group IV: Sham group (only laparotomy was performed).

Drug administration: Orogastric gavage was used in a daily single dose and 85 mg/kg EA (ellagic acid, Sigma-Aldrich Chemical Co., St. Louis, MO, USA) was administered to Group I, and 50 mg/kg DS (Dikloron 50 mg, tablet, DEVA, Tekirdağ, Turkey) to Group II while no drug was administered in Group III and Group IV.

Surgical procedure: Rats were kept in the laboratory environment for a week before the experiment in all study groups and were fed with ad libitum standard rat chow and water. The rats were anesthetized by using intraperitoneal 10 mg/kg xylazine HCl (Rompun 2%, Bayer, Istanbul, Turkey) and 80 mg/kg ketamine HCl (Ketasol, Interhas, Turkey) combination. The ventral abdominal area was prepared for aseptic surgery and a midline incision of 3 cm was made. The cecum was revealed in Groups I, II, and III. Its anterior wall was determined and rubbed with a gauze pad held with a clamp until serous punctate hemorrhages developed. Once the adhesion model was created, the abdomen was closed in accordance with the anatomy (Fig. 1). Only laparotomy with a midline incision was performed in the Sham group (Group IV). After completion of all procedures in each group, the abdominal wall was closed in accordance with the anatomy. The animals were housed in individual cages and were postoperatively followed. A second-generation cephalosporin was administered to the gluteal muscle for a total of 3 pre- and postoperative days.

Macroscopic evaluation: The rats were sacrificed by cervical dislocation on the 14th day under general anesthesia. A “U” incision was made in the abdomen, the abdominal wall was retracted and maximum visibility was provided. The adhesions were quantitatively evaluated and scored blindly by 2 physicians using the classification described in the literature.^18–20 

Grade 0: No adhesion

Grade 1: A single adhesive band between the organs or between the organ and abdominal wall (Fig. 2a)

Grade 2: Two adhesive bands between the organs or between the organ and abdominal wall (Fig. 2b)

Grade 3: More than two adhesive bands between the organs or between the organ and abdominal wall, or adhesions of intestinal loops with or without adhesions to the abdominal wall (Fig. 2c)

Grade 4: Viscera directly attached to the abdominal wall (Fig. 2d)

Biochemical evaluation: Tissue samples were taken from the area where the suture was placed in the cecum and malondialdehyde (MDA) and glutathione (GSH) measurements were performed spectrophotometrically according to the method reported by Beutler et al and Yoshioka et al (UV-1201, Shimadzu, Japan).^21,22 

Histopathologic evaluation: Adhesive bands were histopathologically evaluated with the affected organs in rats with adhesions. The parietal peritoneum and anterior cecum were resected and evaluated histopathologically in rats without adhesions. 5-μ-thick sections obtained from the paraffin blocks were stained with hematoxylin-eosin (HE), evaluated under a light microscope (Leica DM4000B) and recorded (Leica DFC280). Grading of lesions was performed in 10 different fields and the same magnification (HE × 100) based on inflammatory and necrotic changes as below:

• (−) No lesion found

• (+) Mild lesions (<10%)

• (++) Moderate lesions (10–50%)

• (+++) Severe lesions (<50%)

Statistics: The SPSS version 18 software (SPSS Japan Inc.) was used for statistical analysis of all data obtained from the study. Mean and standard deviations that are the central distribution criteria were calculated with statistical analysis and the differences of nominal values between the groups were evaluated using Fisher's definite chi-square test in the analysis of nonparametric data. A P value <0.05 was considered significant. The significance of the difference between 2 groups regarding the parameters found statistically significant with the Kruskal-Wallis test was evaluated with the Whitney U test for the analysis of the values obtained by measurement. A P value <0.015 with Bonferroni correction was considered significant.

The results obtained from the study were macroscopically, histopathologically, and biochemically categorized (Table 1).

Two animals each in Group I and Group III and 1 animal each in Group II and Group IV died in the postoperative period and these animals were not included in the evaluation. The cause of death for the rats was thought to be individual infection.

Table 1 shows the statistically significant differences between the groups according to macroscopic adhesion scoring following evaluation after relaparotomy (P < 0.050). Group I and Group II showed a significant difference when compared with the control group (Group III) but there was no significant difference between Group I and Group II.

There was a statistically significant difference between Group I and the other groups in terms of MDA (P < 0.05), but this difference was not present between Group II and Group III. However, the MDA level in Group I was lower than in Group II and Group III. There was no statistically significant difference between Group I and the other groups in terms of GSH (P > 0.05) but the values in Group I were higher than the others.

Group I: Mild hyperemia and mild edematous changes were present in the vessels of the lamina propria in the tissues of 3 cases in the group where ellagic acid was administered (Fig. 3). Edema and hyperemia of the lamina propria was not seen in the other 2 cases. Cellular infiltration with sporadic neutrophil leukocytes was present. Fibrosis in the tunica serosa was observed in 1 case. No fibrosis was seen in the other cases. Adhesion of the liver to the Glisson capsule was seen in only 1 case in this group. None of the patients showed necrosis in the lamina propria.

Group II: Marked necrosis (Fig. 4) and accompanying neutrophil leukocyte infiltration, edematous changes, hyperemia, and capillarization were noticed in the lamina propria of 6 of 7 cases in the group where diclofenac sodium was administered. Degenerative changes in the muscles and mononuclear cells were observed in the lumen and around the vessels in particular in the submucosa. Mononuclear cell infiltrations were also found in the tunica muscularis and tunica serosa. Ovarium and adhesion were noticed in 1 case.

Group III:Large areas of necrosis spreading from the cecal lumen lamina epithelia to the tunica muscularis and intact or necrotic neutrophilic leukocyte infiltration (Fig. 5), edematous changes, hyperemia, and capillarization were noticed in 1 of the 6 cases in the control group. Thickening due to organized fibrosis was also observed in this layer in 4 cases. Mononuclear cell infiltration was found in the lamina propria, tunica muscularis (Fig. 6), and tunica serosa.

Group IV: The lesions detected in the patients in the sham group were usually limited to the lamina propria and submucosa. There was marked mononuclear cell infiltration in both layers. These cells were accompanied by a small number of neutrophilic leukocytes in the lamina propria in one case and by a granuloma in the submucosa in a different case (Fig. 7). Edematous changes were present in some places.

Intra-abdominal adhesions after surgery is a significant cause of morbidity in the short and long term and many studies have therefore focused on their prevention.^3,4  Peritoneal adhesions occurring after abdominal surgery reduce the quality of life. Although adhesions are formed in almost all abdominal operations, there is almost no effective method to prevent them. The repair process as a result of any disruption of tissue integrity consists of inflammation, fibroblast proliferation, and then the maturation phase with connective tissue and matrix formation.³  We aimed to investigate the effects of EA that has many effects but is mainly antioxidant and anti-inflammatory on intra-abdominal adhesion development in comparison with DS in this study.

DS is a nonsteroidal anti-inflammatory agent used in various ways as an adhesion inhibitor.^13,14  Some effects of EA have not yet been fully revealed although its antioxidant, anti-inflammatory, and anti-angiogenesis effects are known but studies on its anti-adhesive effect are limited. Furthermore, the anti-inflammatory effect of both substances is known to be through inhibition of COX-2.^13,14,23  We administered the oral form of both agents a daily single dose after the surgery and compared their effects in our study.

Positive effects of DS were reported on pleural adhesions by Lisete et al and after thoracic surgery by Lardinois et al^13,14  Although there are some studies on EA, we did not come across any that evaluated its effect on intraabdominal adhesion. There was a statistically significant difference between the groups regarding macroscopic classification in our study. Adhesion development was less in Group I than in Group II and Group III. Various antioxidants and substances such as hyaluronic acid have been used to prevent intra-abdominal adhesions. Another study similar to ours evaluated the effect of hyaluronic acid on adhesion with macroscopic and microscopic examination. They reported that the trauma decreased to a minimum level when the tissue was covered with hyaluronic acid before the adhesion model is created and increased hyaluronic acid viscosity led to a macroscopic and microscopic decrease of intraabdominal adhesion in rats.²⁴ 

The level of MDA, an oxidative stress parameter, has been reported to be increased in peritoneal adhesions and could be reversed to normal limits with antioxidant or anti-inflammatory agents.^3,4  A significant decrease in the MDA level with EA was reported by Böyük et al in rats where intestinal ischemia developed after lung trauma and by Özkaya et al in liver toxicity.^25,26  Similarly, the MDA levels showed a significant decrease in Group I compared to Group II and Group III in our study. A decrease was present in Group II compared to Group III but this was not as marked as in Group I. The level of GSH, an anti-oxidant in contrast to MDA, has been reported to decrease with ischemia, trauma or adhesions.^3,4,27  No statistically significant difference was found between Group I and the other groups in terms of GSH level in our study. However, the GSH level in Group I was higher than in Group II and Group III and closer to the sham group. The decrease in the MDA level and increase in the GSH level in the group where ellagic acid was administered is significant for our study. Similarly EA was reported to reduce cardiac levels of glutathione peroxidase (GPx), catalase, tumor necrosis factor alpha (TNF alpha), and malondialdehyde significantly.^17,18 

Rosillo et al reported that EA decreases inflammatory cell infiltration, edema, hyperemia, and necrosis in Crohn's disease and starts re-epithelialization during recovery, and linked this effect to the inhibition of cyclooxygenase (COX 2) and nitric oxide synthase (iNOS) activity.²³  Girish et al reported a decrease in rat liver necrosis, fat accumulation, and inflammatory reactions in rats given EA after liver toxicity was created by administering carbontetrachloride in a different study.²⁸ 

Lisete et al reported that diclofenac sodium decreased leukocyte migration, edema, hyperemia, and fibrosis.¹³  Lardinois et al reported an anti-adhesive effect through decreasing granulation tissue, collagen deposition and fibrosis.¹⁴  We found a decrease in edema, hyperemia, inflammation, fibrosis, and adhesion in Group I as well as in Group II compared to Group III. Although a decrease was seen in Group II compared to Group III in terms of necrosis, this decrease was not as marked as in Group I. There are many products containing ellagic acid in the market but not enough information is present regarding their ingredients and side effects. Problems may arise when determining when and at what dose EA should be used. Previous studies have not clearly reported the side effects of ellagic acid.²⁹  Side effects need to be determined according to the dose with clinical use in future periods. The clinical use of EA can also enable the determination of the rate of prevention for conditions such as intestinal obstruction, infertility, and chronic pain with patient follow-up and laparoscopic methods as necessary.

In conclusion, taking all our study data into account, the positive effects of the oral form of EA on the development of peritoneal adhesions are more marked than DS, a nonsteroidal anti-inflammatory drug. These positive effects could be a result of the strong antioxidant effect of EA reducing oxidative stress together with its anti-inflammatory and anti-angiogenesis effects. The clinical use of EA as for DS can be targeted with experimental studies performed by using different doses in line with our study.