Prevention of de novo adhesion by ferric hyaluronate gel after laparoscopic surgery in an animal model.

BACKGROUND AND
OBJECTIVE: Adhesions remain a major cause of severe long-term complications. Attempts have been made to prevent adhesion formation by using endogenous or exogenous materials with controversial results. Our aim was to evaluate the efficacy of 0.5% ferric hyaluronate gel in the prevention of adhesion formation after laparoscopic surgery.
METHODS: This was a prospective, randomized, experimental study (animal model). The study population comprised 75 female rabbits (Fauve de Bourgogne) weighing over 3 kg. The rabbits were randomized into 3 groups of 25 (hyaluronate, saline, and control) by using a predetermined computer-generated randomization code. All rabbits underwent a peritoneal laparoscopic resection, and the main outcome measure was the adhesion formation after laparoscopic surgery.
RESULTS: The laparoscopic operating time and the mean interval before second-look surgery were not different among the 3 groups. The number of rabbits with adhesions did not differ among the 3 groups. The bowel adhesion rate was higher at the 10-mm trocar site than at the 5-mm trocar site (P=0.01). The adhesion scores did not differ among the 3 groups. A strong correlation was found between the values of the different adhesion scoring systems used.
CONCLUSION: These results obtained in a rabbit model suggest that routine intraperitoneal application of hyaluronate gel does not prevent adhesion formation after laparoscopic surgery.

INTRODUCTION

Pelvic adhesions occur in 60% to 90% of patients after laparotomy and are a consequence of wound healing.[1, 2] Adhesions recur in 45% to 80% of patients after corrective surgery.[3, 4] Relative to laparotomy, laparoscopic surgery was associated with fewer postoperative adhesions in a rabbit model.[5] In humans, despite a decrease in adhesion formation from 50% after laparotomy to 12% after laparoscopy,[6] adhesions remain a major cause of severe, long-term complications including infertility, chronic pelvic pain, and bowel obstruction, often necessitating further surgery. The annual cost of adhesions has been estimated to be at least $1.2 billion dollars in the United States.[7]

Attempts have been made to prevent adhesion formation by using fibrinolytic agents,[8] anticoagulants,[3] antiinflammatory drugs,[9] antibiotics,[10] and mechanical barriers created by instilling products such as Dextran[11] and endogenous or exogenous materials.[12-15] Recently, 0.5% ferric hyaluronate gel was reported to reduce the number, severity, and extent of adhesions in experimental studies.[2, 16, 17] A pilot study of laparotomic surgery followed by second-look laparoscopy confirmed the ability of this gel to decrease the adhesion rate.[18] Recently, in a multicenter study, Johns et al[19] reported that after laparotomic procedures a decrease occurs in both reformed and de novo adhesions in the hyaluronate gel group relative to the control group. However, except for women with myomas, the indications for surgery in Johns' study population[19] were largely compatible with the laparoscopic treatment. In contrast to laparotomy, few data are available on the ability of hyaluronate gel to prevent de novo adhesion formation after laparoscopic surgery. We therefore used an experimental animal model to evaluate 0.5% ferric hyaluronate gel in adhesion prevention after laparoscopic surgery.

METHODS

We conducted a prospective, randomized study between January 2000 and April 2000, using 75 female Fauve de Bourgogne rabbits weighing over 3 kg. The rabbits were randomized into 3 groups of 25 (hyaluronate, saline, and control) by using a predetermined computer-generated randomization code.

All laparoscopic procedures were performed with the animals in the supine position. General anesthesia was induced by intramuscular injection of ketamine (1.6 mg/kg) (Ketalar, Parke-Davis, Courbevoie, France) plus flunitrazepam (1 mL/kg) (Narcozep, Roche, Neuilly sur Seine, France). The abdomen was shaved then disinfected with 4% iodine polyvidone solution. A Veress needle was inserted through a supraumbilicus incision, and the abdomen was insufflated with CO2. After pneumoperitoneum induction and insertion of the video-laparoscope via a 10-mm trocar, a right iliac fossa trocar of 5 mm was introduced for ancillary instrumentation (Karl Stortz, Tuttlingen, Germany). The laparoscopic procedure included resection of 2 peritoneal tissue samples of 1 cm2 each in the pelvis and the anterior iliac fossa. The location of these samples was chosen such that the bowel or genital organs came into contact with the deperitonealized areas. Coagulation was not performed after resection. The first group of 25 animals received 30 mL of 0.5% ferric hyaluronate in the peritoneal cavity; the second group received 30 mL of physiological saline solution; and the third group received no product. The surface area covered with gel or saline solution was approximately 25 cm2. Antimicrobial chemoprophylaxis was not used. The peritoneal incision for the laparoscope was sutured with Vicryl 3-0 caliber (Polyglactine, Ethicon, Neuilly, France). The skin was sutured with Ethicrin 3/0 caliber (Ethicon, Neuilly, France). No postoperative analgesia was given. Two operators performed all the laparoscopic procedures.

After the tenth postsurgical day, all the animals were euthanized by intracardiac injection of 1 g of Thiopemytal (Nesdonal, Rhone Merieux) in 10 mL of normal saline. Adhesion formation was evaluated by means of median laparotomy, by an assessor blinded to the animal's group assignment.

Adhesions were scored using the following 4 classifications: Mage et al,[20], the American Fertility Society (AFS),[21] Boyers et al,[22] and Corson et al.[23] The Mage and AFS scores evaluate the impact of adhesions on the ovaries and fallopian tubes, ie, fertility. The Boyers and Corson scores evaluate adhesions in the pelvis.

The scoring system developed by Mage et al[20] allocates points for the sites and extent of adhesions by proportion and for the types of adhesion. The total score classifies adhesions as absent (0 points), mild (1 to 9 points), moderate (10 to 20 points), or severe (> 20 points).

The American Fertility Society[21] score is a prognostic classification of adnexal adhesions according to their site (ovary and tube), extent (proportion), and type. Adhesions are scored as minimal (0 to 5 points), mild (6 to 10 points), moderate (11 to 20 points) or severe (21 to 32 points).

The Boyers scoring system[22] is based on the extent, type, and tenacity of adhesions. The maximal total score is 11.

The Corson system[23] is more precise, noting separately 15 anatomic sites of adhesions, as well as their severity (mild or severe), extent (localized < 50%, extensive > 50%), and lysis.

Statistical Analysis

We used the Student t test and the Wilcoxon rank test for parametric and non-normally distributed continuous variables, respectively, and used the chi-square test or Fisher's exact test for categorical variables. P values of <0.05 were considered to denote statistically significant differences. The calculation of the required population size assumed that the incidence of adhesions in rabbits undergoing laparoscopic surgery with peritoneal resection would be 50% without treatment and 25% with the study treatment, with a type I error of 0.05 and a type II error of 0.2. It was therefore necessary to include at least 20 rabbits in each group.

RESULTS

Fifteen of the 75 rabbits included in the study died prematurely. One rabbit died during anesthetic induction and 6 during the laparoscopic procedure (due to bowel injury in 2, vessel injury in 1, and unknown causes in 3). Eight rabbits died before the second-look procedure, due to infection in 2 cases and unknown causes in 6 cases. Therefore, respectively 19, 22, and 19 rabbits were assessable for adhesions in the hyaluronate, saline, and control groups. The number of rabbits excluded from each group did not differ significantly.

The mean (±SD) laparoscopic operating time was 15±5 minutes (range, 10 to 25) (. The mean interval before second-look laparotomy was 18.2±4.1 days (range, 10 to 23) (. This interval did not differ among the 3 groups (.

Adhesions

The number of rabbits with adhesions in hyaluronate, saline, and control groups was 4, 5, and 3, respectively. The proportion of rabbits with adhesions in the control group was 15.6%. The proportion of animals with adhesions did not differ significantly among the 3 groups (21%, 22.7%, 15.6%). The pattern of adhesions in the abdominal cavity is shown in . The rates of pelvic and abdominal adhesions in the control group were 5.2% and 10.5%, respectively. The rates of pelvic and abdominal adhesions did not differ among the 3 groups.

Despite the absence of peritoneal closure, no granulomas were observed at the 5-mm trocar site. In contrast, despite systematic peritoneal and aponeurotic closure, 4 granulomas were found at the 10-mm trocar site, with none among the study groups. Among these 4 rabbits, 1 had bowel incarceration and 2 had bowel adhesions. The bowel adhesion rate was higher at the-10-mm trocar site than at the 5-mm trocar site (P=0.01).

Correlations Among the Different Adhesion Scores

No significant difference in the adhesion scores was noted among the 3 groups. A correlation was found between the Mage and AFS scores (P<0.0001, Rho=1). Similarly, a strong correlation was found between the Boyer and Cosson adhesion scores (P<0.0001, Rho=0.97) (.

DISCUSSION

The adhesion rate after laparoscopic surgery was low in this experimental model. Despite resection of 1 cm2 of the peritoneal surface, an area equivalent to approximately 20 cm2 to 40 cm2 in humans,[24] the rate of postoperative adhesions in the control group was only 15.6%. This contrasts with rates reported after laparotomy in both rat and rabbit models, which ranged from 80% to 100%.[2,16,25] However, our data are in keeping with those of Fiedler et al,[26] who observed a lower adhesion rate after laparoscopic than laparotomic surgery in rabbits.

We found no difference in the proportion of rabbits with adhesions among the 3 groups. The rate of adhesions depends on several factors including the extent of deperitonealized surface,[27] fibrinogen deposition associated with persistent bleeding,[28] tissue desiccation (partly related to the operating time and the impact of the gas used for pneumoperitoneum creation),[29] inflammatory reactions,[30] and, potentially, the use of antiadhesion barriers[31]. In our study, the surface area and location of the deperitonealized area were standardized, and hemostasis was not used. Furthermore, the quantity and type of gas and the mean operating time were similar in the 3 groups. In addition to the use of the laparoscopic approach, the low incidence of adhesions in this study may have been due to experimental factors such as the choice of a rabbit model, the absence of multiple peritoneal resections, the absence of deperitonealized areas associated with mechanical abrasion of adjacent peritoneum, and non use of talc. Despite its higher cost and fragility, we preferred the rabbit model because of similarities in healing parameters with those of humans.[32] In contrast to previous experimental studies using the laparotomic approach,[2,26] no foreign agents were left in the peritoneal cavity after the surgical procedure, apart from the antiadhesion gel and normal saline. In previous studies with rabbits,[16,25,28] the use of large abrasion or resection of peritoneal surfaces was far removed from routine human surgical conditions and made it difficult to extrapolate the results to the clinical setting.

No difference in the adhesion scores was observed among the 3 groups of rabbits. More than 20 scoring systems have been developed to evaluate the characteristics of adhesions, such as their location, type (filmy or vascular), density, and extent.[33] The use of different scoring systems makes it difficult to compare the results of experimental and clinical studies of adhesion-preventing substances. However, we found good correlations between the AFS and Mage classifications and between the Corson and Boyers classifications.

The low rate of adhesions in the untreated control group (15.6%) may explain the lack of significant differences among the 3 groups. In contrast to previous animal studies,[34-37] we chose to use limited deperitonization to mimic classical clinical laparoscopic conditions. To detect a 50% reduction in the adhesion rate with the antiadhesion gel, with a type I error of 0.05 and a type II error of 0.2, at least 197 animals would have had to be included in each arm.

Several factors affect the risk of adhesions, including a previous history of adhesions, underlying inflammatory disease, and the type and route of surgery. Indeed, the risk of adhesion recurrence reaches 86% to 97% after initial adhesiolysis.[38, 39] Moreover, the incidence of de novo adhesions is higher in patients with inflammatory disease,[40,41] after myomectomy[42, 43] and laparotomy than after laparoscopy.[3, 44] This may explain why ferric hyaluronate gel was found effective in patients who underwent myomectomy or laparotomic surgery.[18, 19]

CONCLUSION

Our results, obtained in a rabbit model, suggest that routine use of hyaluronate gel is not warranted to prevent adhesion formation after laparoscopic surgery. Prospective clinical trials are necessary to determine the clinical situations in which an antiadhesion barrier may be beneficial.

Table 1.

Surgical Parameters in the Hyaluronate, Saline, and Control Groups*

Characteristic	Hyaluronate (n=19)	Saline (n=22)	Control (n=19)
Operating time (min, range)	17 (15-25)	19 (14-27)	16 (15-22)
Interval before second look (days, range)	18.4 (11-23)	17.1 (10-23)	19.3 (10-23)
Rabbit weight (kg) (range)	3.3 (2.8-3.9)	3.5 (3.0-4.0)	3.1 (2.9-3.3)

P values were not significant.

Table 2.

Number of Rabbits With Adhesions at Second-Look Laparotomy According to the Group

Group	Rabbits With Adhesions	Rabbits With Pelvic Adhesions	Rabbits With Abdominal Adhesions	Rabbits With Abdominal and Pelvic Adhesions
Hyaluronate (n=19)	4	2	2	0
Saline (n=22)	5	2	2	1
Control (n=19)	3	1	2	0

Table 3.

Adhesions According to the Scoring System*

Score (Mean±SD)	Hyaluronate (n=19)	Saline (n=22)	Control (n=19)
Mage et al20	0.2±0.6	0	0
AFS21	0.3±1.1	0	0
Boyers et al22	0.9±2.1	0.8±1.8	0.8±2.1
Corson et al23	1.4±3.3	0.9±1.8	0.8±2.0

P values were not significant.