Contents:

 »

1. Article Information, Introduction

2. HYALURONAN-BASED ANTIADHESIVE AGENTS

3. HERNIA REPAIR AND HYALURONAN CONTAINING MESHES

4. THE USE OF HYALURONAN IN INFECTIOUS CONDITIONS

5. MECHANISMS OF ACTION OF HYALURONAN

6. Conclusions, References

HYALURONAN-BASED ANTIADHESIVE AGENTS

1. Sodium Hyaluronan Carboxymethylcellulose Membrane
The hyaluronan-carboxymethylcellulose (CMC) antiadhesive barrier, known as Seprafilm® (Genzyme Biosurgery, Inc., Cambridge, MA, USA), is a sterile, bioresorbable membrane (Fig. 3).

The US Food and Drug Administration (FDA) approval of the agent was obtained in 1996. The membrane is applied to injured peritoneal spots, thereby acting as a physical barrier to separate traumatized peritoneal layers during the first phase of peritoneal wound healing. It is composed of two polysaccharides — sodium HA and CMC — and turns into a gel within 24 hours to 48 hours, remaining at the site of placement in this gel form for up to 7 days.
Experimental studies concerning the use of this HA-CMC membrane are numerous. The report from Burns et al. in 1997 first describes its adhesion-reducing capacity in a rat cecal abrasion and sidewall injury model.12 Hellebrekers et al. studied various antiadhesive barriers in a rat model, with HA-CMC showing superior results.9 In another study by Kutlay et al., the HA-CMC membrane was more effective than heparin and aprotinin in a cecal abrasion model in rats.13 Any presumed fibrinolytic activity of the HA-CMC membrane could not be supported by experimental studies from Reijnen et al. and Tarhan et al.14,15
In a randomized clinical trial, Becker et al. studied the effect of the HA-CMC membrane in patients with ulcerative colitis or familial polyposis who were scheduled for ileal pouch-anal anastomosis with diverting-loop ileostomy.16 The included patients were randomized into two groups: in one group, the HA-CMC membrane was placed under the midline incision prior to closure; and in the other group, no antiadhesive agent was applied. Adhesions were evaluated laparoscopically at the time of ileostomy closure 8 weeks to 12 weeks later. The data of 175 patients were analyzed. Only 5 (6%) of 90 control patients were free of adhesions versus 43 (51%) of 85 patients treated with the HA-CMC membrane (p < 0.001). The mean percent of incision length involved with adhesions was also significantly greater in the control group (63% versus 23%, p < 0.001). Furthermore, the percentage of patients with dense adhesions was significantly higher in the control group (58% versus 15%, p < 0.001). The use of the HA-CMC membrane was not related to an increased incidence of adverse events.
In another randomized trial, Diamond studied the effect of the HA-CMC membrane in women undergoing uterine myomectomy.17 The 127 patients included in the trial were randomized to treatment with the HA-CMC membrane or to no adjunctive antiadhesive treatment at the end of the procedure. Adhesions were assessed during second-look laparoscopy. The mean number of sites adherent to the uterine surface was significantly lower in the HA-CMC membrane treated group (4.98 sites ± 0.52 sites) compared with the no-treatment group (7.88 sites ± 0.48 sites), as were the mean uterine adhesion severity scores (1.94 ± 0.14 versus 2.43 ± 0.10; all values treatment versus no treatment, respectively), mean extent scores (1.23 ± 0.12 versus 1.68 ± 0.10), and the mean area of adhesions (13.2 cm2 ± 1.67 cm2 versus 18.7 cm2 ± 1.66 cm2). Again, the use of the HA-CMC membrane was not associated with an increase in postoperative complications.
Vrijland et al. assessed the applicability of the HA-CMC membrane in patients requiring a Hartmann’s procedure for sigmoid diverticulitis or obstructed rectosigmoid.18 Patients were randomized to either HA-CMC membrane placement in the pelvis and under the midline incision or no additional antiadhesive treatment. Adhesions were evaluated laparoscopically at second-stage surgery for restoration of bowel continuity. In that study of 42 patients, although the incidence of adhesions did not differ significantly between groups, the severity of the adhesions was significantly reduced in the group treated with the HA-CMC membrane. There was no correlation between the use of the HA-CMC membrane and the incidence of postoperative complications.
In a worldwide trial focusing on the incidence of bowel obstruction, Fazio et al. recently reported on the efficacy of the HA-CMC membrane in patients who underwent intestinal resection.19 Patients were randomized to be treated with the HA-CMC membrane, which was applied on adhesiogenic areas, or no treatment. The mean follow-up time was 3.5 years. Although the overall rate of bowel obstruction was similar in both groups, the incidence of adhesive small-bowel obstruction requiring reoperation was significantly lower in the group treated with the HA-CMC membrane: 1.8% versus 3.4% (p < 0.05). Two other randomized studies have shown the value of the HA-CMC membrane in reducing peristomal adhesions, facilitating early closure, and in reducing postoperative adhesions in pediatric patients.20,21
In a safety study by Beck et al., the use of this membrane was not associated with an increased incidence of abdominal or pelvic abscess, pulmonary embolism, or foreign-body reaction.22 However, wrapping the suture or staple line of a bowel anastomosis with the HA-CMC membrane should be avoided, as this might increase the risk of detrimental sequelae associated with anastomotic leaks.

2. Glycerol Sodium Hyaluronan Carboxymethylcellulose Membrane
The HA-CMC membrane has been modified by adding glycerol (G). G-HA-CMC membrane, also known as Seprafilm II® (Genzyme Biosurgery, Inc.) was developed to improve the handling characteristics of the HA-CMC membrane. The HA-CMC membrane is known to be brittle and sometimes it can be technically difficult to apply; contact of the membrane with an area other than the desired one can lead to a sticky mass, which is hard to replace. Moreover, the HA-CMC membrane is difficult to use in laparoscopic procedures due to its characteristics. The addition of glycerol made the membrane easier to apply. However, only a few studies evaluated its efficacy and safety.
Kayaoglu et al. studied the use of the G-HA-CMC membrane in a rat adhesion model under clean conditions and during peritonitis.23 Surprisingly, the use of the membrane did not reduce adhesion formation under clean circumstances and even caused increased adhesion formation in bacterial peritonitis.
Recently, a clinical trial by Cohen et al. reported on the use of the G-HA-CMC membrane in patients undergoing restorative proctocolectomy and ileal pouch-anal anastomosis with diverting loop ileostomy.24 Indications for surgery were ulcerative colitis and familial polyposis. The evaluation of adhesions was performed during ileostomy closure by means of laparoscopy. A significant reduction in incidence, extent, and severity of adhesions was observed in patients treated with the G-HA-CMC membrane. However, an increased incidence of infectious complications was noted, possibly due to the glycerol application. Further production and marketing was then stopped by the manufacturers.

3. Hyaluronan Solution
The use of solutions of HA to prevent adhesions was already described in the early 1990s. In this group of agents, Sepracoat® (Genzyme Biosurgery, Inc.) is probably one of the best known products. It is a sterile-filtered, nonpyrogenic 0.4% solution of sodium HA in phosphate buffered saline.
Solutions containing HA have been studied extensively for their adhesion-reducing capacity. Burns et al. performed one of the most extensive experimental studies on this topic.25 They described the adhesion-reducing capacity of different concentrations of HA solution in a rat cecal abrasion model. The use of the HA solution resulted in a reduced severity of adhesions, with 0.4% HA solution being most successful.
Only one clinical prospective randomized trial has been published to our knowledge describing the use of 0.4% HA solution for the reduction of postsurgical adhesion formation. Diamond assessed the use of 0.4% HA solution in women who underwent various gynecologic procedures via laparotomy.26 During the procedure, the peritoneum was repeatedly coated with the 0.4% HA solution or with phosphate-buffered saline, acting as a placebo. After 40 days, adhesions were assessed during second-look laparoscopy. The data of 277 women were available for safety evaluations and of 245 women for efficacy studies. The group treated with the 0.4% HA solution had a significantly lower incidence of adhesions when compared with the placebo group. The proportion of sites involved was lower (0.23 ± 0.02 versus 0.30 ± 0.02, respectively) and the percentage of patients without de novo adhesions was higher (13.1% versus 4.6%, respectively). Furthermore, the use of the 0.4% HA solution resulted in a significantly reduced adhesion extent and severity. In that study, no obvious adverse effects of 0.4% HA solution were described.
Nevertheless, in 1997 the FDA panel voted that Sepracoat® was not yet approvable due to the lack of proven clinical effectiveness. Further studies and evaluation were suggested. However, based upon market dynamics at the time the manufacturers chose not to pursue US market approval. This led to the complete withdrawal of the product from all markets due to business economics (personal communication).

4. 0.5% Ferric Hyaluronan Gel
In the abdominal cavity, an HA solution has a very short residence time and may disappear within hours after placement. Therefore, modifications of the HA solution have been brought to the market to increase its viscosity and prolong its residence time. Crosslinking of HA with ferric ions has resulted in the production of 0.5% ferric HA gel (Gynecare Intergel®, Ethicon, Somerville, NJ, USA), a sterile, nonpyrogenic, viscous solution. Initially, this gel was brought to the market under the name of Lubricoat®.
Few experimental studies on the use of 0.5% ferric HA gel have been published. In 1997, Johns et al. reported on an experimental study on various test formulations applied as peritoneal instillates in a sidewall and uterine horn model in rabbits.27 Adhesion formation was assessed at 7 days and at 14 days. HA that was not ionically crosslinked was ineffective in reducing adhesions in these models even when applied at high viscosity, whereas the ionically crosslinked formulations of HA with trivalent iron were highly effective. Efficacy improved with increased levels of ionic crosslinking. In contrast, in more recent experimental studies the use of 0.5% ferric HA gel has shown no convincing adhesion-reducing effects in a laparoscopic adhesion model in rabbits and in rats.28,29
Thornton et al. assessed the safety and efficacy of 0.5% ferric HA gel in reducing adhesions in patients undergoing abdominal surgery by laparotomy with a planned second-look laparos-copy.30 Women desirous of fertility, aged from 24 to 41, were randomized to receive either 300 ml of 0.5% ferric HA gel or lactated Ringer’s solution at the completion of the procedure. Treated patients had significantly fewer intra-abdominal adhesions and the adhesions that did form in this group were significantly less extensive and less severe. Lundorff et al., who randomized patients undergoing a laparotomy for various gynecological procedures including myomectomy, ovariectomy, salpingostomy and adhesiolysis, performed a similar study.31 Women were aged 18 to 46 and adhesions were evaluated at 24 sites at second-look laparoscopy 6 weeks to 12 weeks later. The ferric HA group had significantly fewer adhesions and again the adhesions that formed despite the use of ferric HA were significantly less extensive and less severe compared with controls. The American Fertility Society score for adnexal adhesions was reduced by 69% in the treatment group compared to controls. In a larger clinical trial by Johns et al., 281 women were randomized with 143 women in the treatment arm and 138 in the control arm.32 Results were similar to the results in the other two studies. The American Fertility Society score for adnexal adhesions was now reduced by 59%. In all three studies, no obvious adverse effects of the use of the ferric HA solution were noted. However, a recently published trial conducted by Tang et al. examining the efficacy and safety of 0.5% ferric HA gel in colorectal resections had to be suspended because of high morbidity in the treatment group, mainly due to anastomotic dehiscence and prolonged postoperative ileus.33
In 2003, the manufacturers withdrew Intergel® from the market. The product was intended to be used in conventional gynecological surgery to reduce postsurgical adhesions as an adjunct to good surgical technique. The voluntary withdrawal was conducted to complete an assessment of information obtained during post-marketing experience with the device, including adverse events associated with off-label use in laparoscopy and laparoscopic procedures such as hysterectomies. The post-market reports included late-onset postoperative pain and repeat surgeries following the onset of pain, noninfectious foreign body reactions, and tissue adherence (information provided by the FDA and manufacturers).

5. Hyaluronan Carboxymethylcellulose Gel
As described above, the HA-CMC membrane proved to be effective in adhesion reduction. However, as mentioned its handling characteristics may not be optimal, especially in laparoscopic surgery, where application of the membrane can be difficult. Laparoscopic use of a solution or gel is much easier. HA-CMC gel (Sepragel®, Genzyme Biosurgery, Inc.) was developed as an alternative to the HA-CMC membrane.
In 1996, Burns et al. described the use of HA-CMC gel for the prevention of adhesions in adhesion models in rats and rabbits.34 Treatment with the gel resulted in an increased number of animals without adhesions by 70% in a rat cecal abrasion model and over 90% in a rabbit sidewall defect-bowel abrasion model. Leach et al. showed a reduction of adhesions using the HA-CMC gel in a rabbit uterine horn model.35 The uterine horn model was shown to be adhesiogenic, with 29 (70%) of 42 untreated uterine horns found to have adhesions. After treatment with HA-CMC gel, 22 (55%) of 40 uterine horns were free of adhesions compared with 12 (30%) of 42 uterine horns in controls. Postsurgical adhesion formation was significantly reduced in animals treated with HA-CMC gel when compared with controls (p < 0.05). Separate analysis for extent, severity, and density showed significant differences for adhesion severity and density (p < 0.05). Clinical trials studying the safety and efficacy of HA-CMC gel have not yet been performed to our knowledge. HA-CMC gel has not been submitted for regulatory review in the United States or the European Union and is therefore not available for clinical use. According to the manufacturers, this is a clinical development project that is still under investigation (personal communication).

6. Auto-Crosslinked Hyaluronan Gel
Another novelty in the field of HA-based instillates was the development of auto-crosslinked HA gel (Hyalobarrier® gel, Fidia Advanced Biopolymers, Abano Terme, Italy). This highly viscous gel was obtained by means of an internal crosslinking reaction of pure HA in the absence of any chemical substance foreign to the native HA structure. The commercially offered auto-crosslinked HA gel had a concentration of 4% and a prolonged residence time in the abdominal cavity was claimed by the manufacturers.
De Iaco et al. described the use of auto-crosslinked HA gel in a laparoscopic adhesion model in rabbits.36 Use of auto-crosslinked HA gel resulted in a significant reduction of adhesion formation: 35% of the treated animals had severe adhesions versus 66% in the control group. Furthermore the mean adhesion score was significant lower in the treated group. In a subsequent study, auto-crosslinked HA gel appeared to be effective in reducing adhesions in the presence of inadequate hemostasis as well.37 In a rat model of laparotomy and uterine horn injury, Koçak et al. managed to reduce adhesions using auto-crosslinked HA gel.38 Belluco et al. used an adhesion model in rabbits with different concentrations of the gel.39 Again, the adhesion-reducing effect was obvious, with the 4% concentration being the most efficient.
The only available clinical data to our knowledge come from studies in women undergoing abdominal surgery for gynecological indications; Acunzo et al. used the auto-crosslinked HA gel in women following hysteroscopic intrauterine adhesiolysis.40 After 3 months, a significantly lower rate of intrauterine adhesions was observed in the auto-crosslinked HA treated group (n = 43) compared with controls (n = 41; 14% versus 32%, p < 0.05). Patients in the treated group showed significantly lower adhesion scores at follow-up in comparison with those in the control group (2.0 ± 0.0 versus 5.3 ± 0.2, p < 0.001). When an intrauterine adhesion staging was performed according to the American Fertility Society, all auto-crosslinked HA gel treated patients had mild (Stage I) adhesions. Among control patients, only 25% had mild adhesions and 75% had moderate (Stage II) adhesions. The same research group has performed a comparable study in women with a single surgically remediable intrauterine lesion.41 Again, a significantly lower rate of intrauterine adhesions was observed in the auto-crosslinked HA gel treated group (n = 67) compared with controls (n = 65; 10% versus 26%, p < 0.05). The mean adhesion score was significantly lower as well (2.42 ± 0.78 versus 3.83 ± 0.98, p < 0.05). The adhesion severity according to the American Fertility Society demonstrated a significantly decreased adhesion severity in the treated group (86% mild adhesions, 14% moderate adhesions) compared with controls (24% mild, 76% moderate). No adverse gel-related effects were detected in the auto-crosslinked HA gel treated group.
In a recent study by Pellicano et al., the gel proved useful for reducing the incidence of postsurgical adhesions after laparoscopic myomectomy.42 Auto-crosslinked HA gel-treated patients (n = 18) had a significantly lower rate of postsurgical adhesions in comparison with controls (n = 18; 28% versus 78%, p < 0.01). The rate of adhesions was significantly higher (p < 0.05) in patients treated with interrupted “figure-eight” sutures than in subjects treated with subserous sutures. Four out of nine patients (44%) in the auto-crosslinked HA gel-treated group and eight out of nine patients (89%) in the control group who were treated with so-called interrupted “figure-eight” sutures developed postoperative adhesions, whereas one out of nine (11%) auto-crosslinked HA gel-treated patients and six out of nine (67%) control patients who were treated with subserous sutures developed postoperative adhesions. No side effects of the applied gel were reported.
Currently, the auto-crosslinked HA gel is also no longer available for clinical use. It was withdrawn from the market after it failed to show its effectiveness in adhesion reduction on a large scale after clinical research (personal communication).

7. Other Hyaluronan-Based Products
A few other crosslinked HA products have been described in recent years. Haney et al. studied a barrier composed of chemically crosslinked HA (Incert®, Anika Therapeutics, Woburn, MA, USA).43 The product was used in a murine uterine horn model. Fewer adhesions were present when excision injuries were separated by the barrier (43% versus 88%), whereas the number of adhesions was unchanged after electrocautery injuries (54% versus 65%, N.S.). Despite human pilot trials to test the safety and effectiveness of this product for prevention of adhesions after spinal surgery, no trials studying the abdominal use are available to our knowledge. Until now, the barrier has not been approved by the FDA for use in the United States.
Jackson et al. focused on the use of a paclitaxel-loaded crosslinked HA film — HA crosslinked with water-soluble carbodiimide and containing 10% glycerol and 1% or 5% paclitaxel.44 In a rat cecal abrasion model, both 1% and 5% paclitaxel film effectively reduced the formation of adhesions, with the 5% paclitaxel film being the most effective. However, this product led to excess fluid in the abdominal cavity at necropsy. No further studies have been reported at this time.
Li et al. described the development of a crosslinked HA hydrogel that contained a covalently bound derivative of the antiproliferative drug mitomycin C (MMC).45 The hydrogel was tested with 0.5% MMC and 2% MMC in vitro and in vivo. In vitro HA film loaded with 0.5% MMC inhibited proliferation when incubated with human T31 tracheal scar fibroblasts, whereas the 2% MMC films were cytotoxic. In vivo, the MMC films were implanted intra-abdominally in rats; the HA-MMC films reduced the thickness of fibrous tissue formed surrounding it. In a second study by the same group, HA-MMC films and gels were evaluated in a uterine horn model in rats.46 Both films and gels were tested in several concentrations and were highly efficient in reducing adhesions. The results of the HA-MMC films were dose-dependent. The efficacy of the gels was highly correlated to the concentration, with the HA 0.625% MMC gel being the most effective. The use of this novelty is not yet described inclinical studies to our knowledge.