Effect of Hyaluronic Acid-carboxymethylcellulose Adhesion Barrier on Wound Healing: An Experimental Study

Address correspondence to: Erdinc Kamer, MD Department of Surgery Izmir Ataturk Training and Research Hospital 1834 sk. No: 9/4, 35530 Karsiyaka, Izmir Turkey E-mail: erdinc.kamer@gmail.com Phone: 0090 232 244 4444


Abstract: The aim of this study was to investigate the effects of hyaluronic acid-carboxymethylcellulose (HA-CMC) membrane on the healing process of wounds in rats. One hundred animals were assigned randomly into two equal groups. Midline laparotomies were performed. In group 1, a 5-cm x 3-cm piece of HA-CMC membrane was placed under the laparotomy incision. The same procedure was performed in group 2, but without the HA-CMC membrane. Ten animals from each group were euthanized on postoperative days (POD) 4, 7, 14, 21, and 35 after wounding. Breaking strength, histologic examination, and tissue hydroxyproline levels were analyzed. The tensiometric test showed that there was no significant difference in the breaking strengths between the two groups (P > 0.05). Statistical difference was found to be significant on POD 4, 14, 21, and 35 when the groups were compared with regard to average hydroxyproline levels (P < 0.05). Significant differences were found in the results of histologic examination of the tissue specimens between the two groups in terms of acute inflammation on POD 14, chronic inflammation, and granulation tissue fibroblast maturation on POD 35, collagen deposition on POD 21, and neovascularization on POD 7, 14, 21, and 35 (P < 0.05). The results show that the HA-CMC membrane did not negatively affect the mechanical strength and healing process of the laparotomy incisions.
   Postoperative adhesions are an important, and thus far, unsolved surgical problem. The two main methods to prevent adhesion that have been recently evaluated are to minimize surgical trauma during the operation and by using special materials.¹ Most antiadhesive agents have raised concern over their poor healing.² Abdominal wall healing after laparotomy is important because insufficient incisional wound strength results in prolonged periods of disability for the patient secondary to fascial dehiscence and hernia formation.³    However, the possible adverse effects of modulating the inflammatory response (infection, delayed wound healing) and applying fibrinolytic stimulations (hemorrhage) have made separation of tissue surfaces the most favored option.¹ Therefore, by separating injured adjacent surfaces with a bioresorbable membrane during the critical period, fibrin bridge formation and adhesions can be presumably prevented. Hyaluronic acid (HA)-carboxymethylcellulose (CMC) bioresorbable membrane (Seprafilm®, Genzyme Biosurgery, Cambridge, Mass) has been introduced with this purpose.⁴ The combination of these two substances (HA and CMC) results in a transparent, thin, adherent, and absorbable membrane, which is included in the category of mechanical separation devices. In animal studies and in one randomized clinical trial, it has been shown that HA-CMC reduces the incidence, extent, and severity of postsurgical adhesions.⁵    Although it is an effective option to prevent postoperative adhesions, there are no clear data about wound healing of laporotmy incisions resulting from adhesion reduction with HA-CMC membrane.¹ Thus, the present study investigated the effects of the HA-CMC membrane on the wound healing process in rats.

Methods

   Animals. One hundred 5-month-old adult male Wistar-Albino rats at Ege University Faculty of Medicine Animal Research laboratory (Izmir,Turkey) weighing 250–300 g, were acclimated to the new environment for 48 hours; they were maintained on standard rat chow and water. They were fed standard laboratory chow until the night before the operation, and were given water ad libitum. All manipulations were undertaken in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the authors’ Animal Care Committee.    Experimental design. Animals were randomly assigned to two groups: 50 rats in the HA-CMC group (group 1) and 50 rats in the control group (group 2). Each rat was anesthetized with intramuscular injection of 60-mg/kg ketamine hydrochloride (Ketalar, Eczacıbası, Warner-Lambert Laboratories, Levent, Istanbul, Turkey) and 10-mg/kg xylazine hydrochloride (Rompun, Bayer Laboratories, Sisli, Istanbul, Turkey). The abdomen was shaved with standard animal clippers, and the skin was swabbed with 70% ethanol. All surgical procedures were performed by one investigator and under steril conditions. Afterward, a 4-cm midline laparotomy was performed. In group 1, a 5-cm x 3-cm piece of HA-CMC membrane was placed in the peritoneal cavity. The abdominal wall and skin were closed separately with running 3/0 silk sutures. The same procedures were performed in group 2, but without the HA-CMC membrane. All rats were given water and regular diet ad libitum on the day of operation.    Ten animals from each group were killed on postoperative days (POD) 4, 7, 14, 21, and 35 with an overdose of sodium pentobarbital (300 mg/kg, intraperitoneal). The skin sutures were removed, and the breaking strength of the midline incision at this segment was measured with a tensinometer. After the breaking strength analysis had been completed, the abdominal incision wounds were excised and divided into 2 pieces (4-cm x 1-cm) in all animals. One piece was put into a 10% formaldehyde solution and stored for pathologic examinations. The second piece was wrapped to determine the hydroxyproline level.    Wound breaking strength. Breaking strength analysis was conducted within 1 hour after the removal of tissues. Before the measurement of the fascia breaking strength, the uninterrupted sutures applied at the abdominal closure were cut off at the incision laterally, without damaging the wound. Fascia strength was tested in a specially constructed tensiometer, which provided a constantly increasing amount of pressure. Wound samples were fixed with a clamp at one end and another clamp at the other end to which a tensinometer was attached. To operate this tensinometer, water was poured from a height of 50 cm with a speed of 60 cc/min; tensile strength value (weight) was measured at the moment of breaking.⁶    Histologic grading. A small piece of the incision immediately superior to the sample was used to assess rupture strength, and was processed for routine histologic examination. Biopsy specimens from each fascial wound were obtained as described above on POD 4, 7, 14, 21, and 35. At each time point, the wounds were harvested, and their histologic features were assessed in parafin-embedded sections using hematoxylin and eosin and Gomori’s tri-chrome stains at 10x–40x magnification. The main histologic outcome measures included the amount of acute and chronic inflammatory infiltrates, the amount and maturation of granulation tissue, collagen deposition, reepithelialization, and neovascularization. Acute inflammation was defined as the presence of neutrophils, while chronic inflammation was defined as the presence of plasma and monocytic cells. The Abramov’s histologic scoring system (modified Greenhalgh’s scoring system) was used for this study.^7,8 While the Greenhalgh’s scoring system compiled several histological parameters simultaneously to create a single score, Abramov’s system assessed each parameter independently and gave a score of 0–3. Acute and chronic inflammatory infiltrates, the amount of granulation tissue and collagen deposition were graded as: 0 (none), 1 (scant), 2 (moderate), 3 (abundant). The maturation of granulation tissue was graded as either: 0 (immature), 1 (mild maturation), 2 (moderate maturation), 3 (fully matured). Reepithelialization was graded as either: 0 (none), 1 (partial), 2 (complete but immature or thin), 3 (complete and mature). Neovascularization was graded as either: 0 (none), 1 (up to 5 vessels per high-power field [HPF]), 2 (6–10 vessels per HPF), 3 (more than 10 vessels per HPF).⁷ Two independent pathologists performed the histological examinations and applied the scoring system in a blinded fashion.    Hydroxyproline analyzed. The samples for hydroxyproline levels were weighed, cut into small pieces, and homogenized in a phosphate buffer to yield a 20% homogenate. Aliquots of the homogenate were added to an equal volume of 6 N hydrochloric acid, and hydrolyzed in Teflon-capped vials at 102˚C for 16 hours. The hydroxyproline content of the tissue hydroxylates was determined spectrophotometrically by using the standard addition method developed by Kivirikko et al⁹ (Hypopronosticon Kit lot/ch. B:E 92401; Organon Teknika B.V., Boxtel, Holland). Results were expressed in milligrams, such as in hydroxyproline/100 mg (wet weight).    Statistical analysis. Mann-Whitney U test was used for statistical analysis. Inter- and intra-observer variabilities were calculated using Cohen’s k test. P < 0.05 was considered statistically significant. The statistical analysis was performed using SPSS for Windows 11.5 (SPSS, Chicago, Ill).

Results

   All 100 rats survived the surgical procedures with no complications. In both groups, the mean values of breaking strengths and hydroxyproline levels of the abdominal wounds on POD 4, 7, 14, 21 and 35, and statistical comparisons between the groups are shown in Table 1 and 2. Statistical difference was found to be significant on POD 4, 14, 21, and 35 when group 1 and group 2 were compared with regard to mean hydroxyproline levels (P < 0.05). The tensiometric test showed that there was no significant difference in breaking strengths between the two groups (P > 0.05). One-hundred wound specimens were available for histologic evaluation (10 from each day 4, 7, 14, 21, and 35). The interobserver (weighted k 0.68, 95% confidence interval [CI]: 0.55–0.65) and intraobserver (weighted k 0.74, 95% CI: 0.62–0.73) agreements of the scoring system were good.    Acute inflammation (Figure 1) peaked on POD 7 in group 1 and on POD 4 in group 2, whereas there was a significant difference between the two groups in terms of acute inflammation on POD 14 (P = 0.003). While chronic inflammation (Figure 2) peaked on POD 14 in both groups, a significant difference between the two groups was observed on POD 35 (P = 0.000). The amount of granulation tissue (Figure 3) peaked on PODs 21 and 35 in group 1, and on POD 14 in group 2; however, there was no significant differencebetween the two groups regarding the amount of granulation tissue formed (P > 0.05). Granulation tissue fibroblast maturation (Figure 4) peaked on POD 35 in group 1 and on PODs 14 and 35 in group 2, while a significant difference between the two groups was observed on POD 35 (P = 0.008). Collagen deposition (Figure 5) gradually increased beginning on POD 4 and peaked on POD 35 in group 1. Collagen deposition exhibited variable progress by increasing up to POD 14, suddenly decreasing on POD 21, and then continuing to increase until peaking on POD 35. A significant difference between the groups was observed on POD 21 in terms of collagen deposition (P = 0.019). Reepithelialization (Figure 6) gradually increased fdiffer significantly between groups (P > 0.05). Neovascularization (Figure 7) peaked on POD 14 in both groups; it differed significantly between the two groups on PODs 7, 14, 21, and 35 (P < 0.002, P < 0.002, P < 0.028, and P < 0.008, respectively). Histological comparisons with and without HA-CMC are shown in Figures 8 and 9.

Discussion

   Intraperitoneal adhesions occur in as many as 60% to 90% of patients undergoing abdominal surgery.rom POD 4 until it peaked on POD 35 in both groups, and did not Various methods and agents have been tested to prevent postoperative adhesion formation.¹⁰ One such method is the intraperitoneal application of mechanical barriers. HA and CMC are two such agents.¹⁰ HA is a polysaccharide of high molecular weight, which is widely present in tissue liquids and stimulates angiogenesis in the organization of extracellular binding tissue and in the injury healing process. Moreover, it has been demostrated that HA inhibits free radicals.¹¹ It is thought that when bound with fibrin HA stimulates fibroblast proliferation and prompts the formation of granulatin tissue.¹¹ CMC, a nontoxic polysaccharide made by monochloroacetate and cellulose, is used in the cosmetic, food, and drug industries.¹² HA and CMC were combined into a single agent (Seprafilm®, Genzyme< Biosurgery, Cambridge, Mass).¹² In 1996, a HA-CMC antiadhesion membrane became clinically available for use in abdominal surgery. Elsewhere, a significant reduction in postsurgical adhesions using HA-CMC antiadhesion membrane was reported in animal models and in other clinical studies in adults and pediatric patients.¹³ The HA-CMC membrane functions as a physical barrier by separating the serosal surfaces during epithelial regeneration.⁴ Another explanation is that it may decrease the activity or proliferation of fibroblasts, prevent fibrin deposition on the injured serosal surfaces and cellular elements during peritoneal repair.¹⁴    Bioresorbable membranes as a HA-CMC used for adhesion prevention have the theoretical risk of affecting the healing process of wounds.¹⁵ There are no clear data on healing of laparotomy incisions as a result of adhesion reduction using an HA-CMC membrane.¹    Two major parameters of the healing process were evaluated to examine the tissue repair. One parameter was mechanical tissue strength and the other was hydroxyproline levels reflecting tissue collagen concentration.^1,16 Hebda et al¹⁷ based their measurements of wound healing on standard histologic evaluation and functional assessment (tensiometer)—the same methods were used in the present study. In the present study, measurement of breaking strength for laparotomy incisions was not significantly different among the rats with and without HA-CMC. These findings revealed that HA-CMC application did not have a negative effect on mechanical strength of laparotomy incisions. Significant increases were detected on PODs 4, 14, 21, and 35 when tissue hydroxyproline levels of laparotomy incisions were compared between rats in the HA-CMC group and those in the non- HA-CMC group.    Wound healing consists of 3 phases: the inflammatory phase, which consists of inflamatory cell migration to the healing wound; the proliferative phase, which consists of angiogenesis, fibroplasia, epithelialization, and extracellular matrix accumulation; and the last phase, which is maturation.^18,19 The histopathologic results from the present study indicated that the wound-healing process in laparotomy incisions includes acute and chronic inflammation, fibroblast proliferation, neovascularization, and progressive re-epithelialization with collagen deposition.    During wound healing in the skin, fibroblasts have been shown to deposit increasing amounts of fibrillar collagen, which is important for the development of the wound’s strengh.^20,21 Collagen deposition by skin fibroblasts has been shown to begin within 3–5 days after injury and to continue for several weeks, depending on the size of the wound.⁷ In the present study, the levels of hydroxyproline content in the abdominal fascia at the incision site were determined as an indicator of collagen synthesis and wound healing. A progressive increase in collagen deposition (until POD 35) in the laparotomy incisions was found. However, collagen deposition differed significantly between rats in the two treatment groups, but only on POD 21.    Inflammation and local leukocyte infiltration are steps in the postoperative healing process.¹ Inflammatory cells were shown to marginate into the injured tissue, and an abundance of leukocytes and plasma proteins efflux into the wound. Neutrophils arrive initially, sterilize and debride the wound, while monocytes and tissue macrophages predominate in the inflammatory infiltrate within 2–3 days.⁷ In the histopathologic evaluation of the specimens taken from laparotomy incision, acute inflammation peaked at POD 4 and 7 in group with HA-CMC, while group without HA-CMC peaked at POD 7. Chronic inflammation peaked at POD 14 in both groups with or without HA-CMC. Statistical differences were significant on POD 14 and 35 between the two groups in acute and chronic inflammation, respectively. Increased inflammatory reaction causes more collagen production and scar formation due to the increased number of fibroblasts.²²    Granulation tissue on the suture lines represents the beginning of the proliferative phase of the healing.¹ In this study, granulation tissue amount was same in the both groups with or without HA-CMC.    Myofibroblast proliferation and angiogenesis are important steps in wound healing, as they promote the tensile strength of the wound. Fibroblasts begin entering the wound site 2–5 days after wounding as the inflammatory phase is ending, and their numbers peak at 1–2 weeks post-wounding.²³ In this study, tissue fibroblast maturation peaked on POD 35 in group 1 and on PODs 14 and 35 in group 2, while a significant difference between the two groups was observed on POD 35.    Re-epithelialization in dermal tissues was shown to occur within 24–48 hours after wounding when spurs of epithelial cells moved from the wound edges along the cut margins of the dermis, depositing basement membrane components as they moved.²⁴ In this study, we observed progressive re-epithelialization shortly after wounding in both groups, with or without HA-CMC, that was completed by POD 14.    Neovascularization is an important component of the wound-healing process that includes branching and extension of adjacent blood vessels, as well as recruitment of endothelial progenitor cells.²⁵ Abramov et al⁷ found that neovascularization in the abdomen peaked at day 7. In the present study, neovascularization peaked on POD 14 in both groups, while significant increases were detected on PODs 7, 14, 21, and 35 in group 1 compared to that in group 2. Hence, HA could contribute to the injury healing process by stimulating angiogenesis.

Conclusion

   In conclusion, HA-CMC, which is commonly used for preventing postoperative intra-abdominal adhesions, does not diminish the recovery of postoperative intra-abdominal incisions. The results presented here show that the HA-CMC membrane did not negatively affect the mechanical strength or the healing process of the laparotomy incisions. Further randomized prospective trials are needed to verify these results.

Acknowledgment

   The authors thank Hakan Baydur, MPH, Department of Public Health, Dokuz Eylul University, Izmir, Turkey for contributions to the statistical analysis.

From the Department of Surgery, Izmir Ataturk Training and Research Hospital, Izmir, Turkey; Department of Pathology, Dr. Behcet Uz Children’s Hospital, Izmir, Turkey